bird_classification.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "_qke6-kBfhaz",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "d65409fc-ec2a-4530-df5e-587d1b581b80"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: kaggle in /usr/local/lib/python3.10/dist-packages (1.5.16)\n",
            "Requirement already satisfied: six>=1.10 in /usr/local/lib/python3.10/dist-packages (from kaggle) (1.16.0)\n",
            "Requirement already satisfied: certifi in /usr/local/lib/python3.10/dist-packages (from kaggle) (2023.5.7)\n",
            "Requirement already satisfied: python-dateutil in /usr/local/lib/python3.10/dist-packages (from kaggle) (2.8.2)\n",
            "Requirement already satisfied: requests in /usr/local/lib/python3.10/dist-packages (from kaggle) (2.27.1)\n",
            "Requirement already satisfied: tqdm in /usr/local/lib/python3.10/dist-packages (from kaggle) (4.65.0)\n",
            "Requirement already satisfied: python-slugify in /usr/local/lib/python3.10/dist-packages (from kaggle) (8.0.1)\n",
            "Requirement already satisfied: urllib3 in /usr/local/lib/python3.10/dist-packages (from kaggle) (1.26.16)\n",
            "Requirement already satisfied: bleach in /usr/local/lib/python3.10/dist-packages (from kaggle) (6.0.0)\n",
            "Requirement already satisfied: webencodings in /usr/local/lib/python3.10/dist-packages (from bleach->kaggle) (0.5.1)\n",
            "Requirement already satisfied: text-unidecode>=1.3 in /usr/local/lib/python3.10/dist-packages (from python-slugify->kaggle) (1.3)\n",
            "Requirement already satisfied: charset-normalizer~=2.0.0 in /usr/local/lib/python3.10/dist-packages (from requests->kaggle) (2.0.12)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests->kaggle) (3.4)\n"
          ]
        }
      ],
      "source": [
        "!pip install kaggle"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "zn--Q1ksgMCX"
      },
      "outputs": [],
      "source": [
        "!mkdir -p ~/.kaggle\n",
        "!cp kaggle.json ~/.kaggle/\n",
        "!chmod 600 ~/.kaggle/kaggle.json"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QBjOFy1Ofw4j",
        "outputId": "a36bec5c-ef41-45ea-b954-4bce0642f8cc"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading birds-20-species-image-classification.zip to /content\n",
            " 82% 57.0M/69.3M [00:01<00:00, 44.1MB/s]\n",
            "100% 69.3M/69.3M [00:01<00:00, 45.5MB/s]\n"
          ]
        }
      ],
      "source": [
        "!kaggle datasets download -d umairshahpirzada/birds-20-species-image-classification"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "gumcsOa4gt2g",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "80345a02-64ce-4610-c3a9-2b9d0d737034"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Archive:  birds-20-species-image-classification.zip\n",
            "  inflating: images to predict/1.jpg  \n",
            "  inflating: images to predict/2.jpg  \n",
            "  inflating: images to predict/3.jpg  \n",
            "  inflating: images to predict/4.jpg  \n",
            "  inflating: images to predict/5.jpg  \n",
            "  inflating: images to predict/6.jpg  \n",
            "  inflating: test/ABBOTTS BABBLER/1.jpg  \n",
            "  inflating: test/ABBOTTS BABBLER/2.jpg  \n",
            "  inflating: test/ABBOTTS BABBLER/3.jpg  \n",
            "  inflating: test/ABBOTTS BABBLER/4.jpg  \n",
            "  inflating: test/ABBOTTS BABBLER/5.jpg  \n",
            "  inflating: test/ABBOTTS BOOBY/1.jpg  \n",
            "  inflating: test/ABBOTTS BOOBY/2.jpg  \n",
            "  inflating: test/ABBOTTS BOOBY/3.jpg  \n",
            "  inflating: test/ABBOTTS BOOBY/4.jpg  \n",
            "  inflating: test/ABBOTTS BOOBY/5.jpg  \n",
            "  inflating: test/ABYSSINIAN GROUND HORNBILL/1.jpg  \n",
            "  inflating: test/ABYSSINIAN GROUND HORNBILL/2.jpg  \n",
            "  inflating: test/ABYSSINIAN GROUND HORNBILL/3.jpg  \n",
            "  inflating: test/ABYSSINIAN GROUND HORNBILL/4.jpg  \n",
            "  inflating: test/ABYSSINIAN GROUND HORNBILL/5.jpg  \n",
            "  inflating: test/AFRICAN CROWNED CRANE/1.jpg  \n",
            "  inflating: test/AFRICAN CROWNED CRANE/2.jpg  \n",
            "  inflating: test/AFRICAN CROWNED CRANE/3.jpg  \n",
            "  inflating: test/AFRICAN CROWNED CRANE/4.jpg  \n",
            "  inflating: test/AFRICAN CROWNED CRANE/5.jpg  \n",
            "  inflating: test/AFRICAN EMERALD CUCKOO/1.jpg  \n",
            "  inflating: test/AFRICAN EMERALD CUCKOO/2.jpg  \n",
            "  inflating: test/AFRICAN EMERALD CUCKOO/3.jpg  \n",
            "  inflating: test/AFRICAN EMERALD CUCKOO/4.jpg  \n",
            "  inflating: test/AFRICAN EMERALD CUCKOO/5.jpg  \n",
            "  inflating: test/AFRICAN FIREFINCH/1.jpg  \n",
            "  inflating: test/AFRICAN FIREFINCH/2.jpg  \n",
            "  inflating: test/AFRICAN FIREFINCH/3.jpg  \n",
            "  inflating: test/AFRICAN FIREFINCH/4.jpg  \n",
            "  inflating: test/AFRICAN FIREFINCH/5.jpg  \n",
            "  inflating: test/AFRICAN OYSTER CATCHER/1.jpg  \n",
            "  inflating: test/AFRICAN OYSTER CATCHER/2.jpg  \n",
            "  inflating: test/AFRICAN OYSTER CATCHER/3.jpg  \n",
            "  inflating: test/AFRICAN OYSTER CATCHER/4.jpg  \n",
            "  inflating: test/AFRICAN OYSTER CATCHER/5.jpg  \n",
            "  inflating: test/AFRICAN PIED HORNBILL/1.jpg  \n",
            "  inflating: test/AFRICAN PIED HORNBILL/2.jpg  \n",
            "  inflating: test/AFRICAN PIED HORNBILL/3.jpg  \n",
            "  inflating: test/AFRICAN PIED HORNBILL/4.jpg  \n",
            "  inflating: test/AFRICAN PIED HORNBILL/5.jpg  \n",
            "  inflating: test/AFRICAN PYGMY GOOSE/1.jpg  \n",
            "  inflating: test/AFRICAN PYGMY GOOSE/2.jpg  \n",
            "  inflating: test/AFRICAN PYGMY GOOSE/3.jpg  \n",
            "  inflating: test/AFRICAN PYGMY GOOSE/4.jpg  \n",
            "  inflating: test/AFRICAN PYGMY GOOSE/5.jpg  \n",
            "  inflating: test/ALBATROSS/1.jpg    \n",
            "  inflating: test/ALBATROSS/2.jpg    \n",
            "  inflating: test/ALBATROSS/3.jpg    \n",
            "  inflating: test/ALBATROSS/4.jpg    \n",
            "  inflating: test/ALBATROSS/5.jpg    \n",
            "  inflating: test/ALBERTS TOWHEE/1.jpg  \n",
            "  inflating: test/ALBERTS TOWHEE/2.jpg  \n",
            "  inflating: test/ALBERTS TOWHEE/3.jpg  \n",
            "  inflating: test/ALBERTS TOWHEE/4.jpg  \n",
            "  inflating: test/ALBERTS TOWHEE/5.jpg  \n",
            "  inflating: test/ALEXANDRINE PARAKEET/1.jpg  \n",
            "  inflating: test/ALEXANDRINE PARAKEET/2.jpg  \n",
            "  inflating: test/ALEXANDRINE PARAKEET/3.jpg  \n",
            "  inflating: test/ALEXANDRINE PARAKEET/4.jpg  \n",
            "  inflating: test/ALEXANDRINE PARAKEET/5.jpg  \n",
            "  inflating: test/ALPINE CHOUGH/1.jpg  \n",
            "  inflating: test/ALPINE CHOUGH/2.jpg  \n",
            "  inflating: test/ALPINE CHOUGH/3.jpg  \n",
            "  inflating: test/ALPINE CHOUGH/4.jpg  \n",
            "  inflating: test/ALPINE CHOUGH/5.jpg  \n",
            "  inflating: test/ALTAMIRA YELLOWTHROAT/1.jpg  \n",
            "  inflating: test/ALTAMIRA YELLOWTHROAT/2.jpg  \n",
            "  inflating: test/ALTAMIRA YELLOWTHROAT/3.jpg  \n",
            "  inflating: test/ALTAMIRA YELLOWTHROAT/4.jpg  \n",
            "  inflating: test/ALTAMIRA YELLOWTHROAT/5.jpg  \n",
            "  inflating: test/AMERICAN AVOCET/1.jpg  \n",
            "  inflating: test/AMERICAN AVOCET/2.jpg  \n",
            "  inflating: test/AMERICAN AVOCET/3.jpg  \n",
            "  inflating: test/AMERICAN AVOCET/4.jpg  \n",
            "  inflating: test/AMERICAN AVOCET/5.jpg  \n",
            "  inflating: test/AMERICAN BITTERN/1.jpg  \n",
            "  inflating: test/AMERICAN BITTERN/2.jpg  \n",
            "  inflating: test/AMERICAN BITTERN/3.jpg  \n",
            "  inflating: test/AMERICAN BITTERN/4.jpg  \n",
            "  inflating: test/AMERICAN BITTERN/5.jpg  \n",
            "  inflating: test/AMERICAN COOT/1.jpg  \n",
            "  inflating: test/AMERICAN COOT/2.jpg  \n",
            "  inflating: test/AMERICAN COOT/3.jpg  \n",
            "  inflating: test/AMERICAN COOT/4.jpg  \n",
            "  inflating: test/AMERICAN COOT/5.jpg  \n",
            "  inflating: test/AMERICAN FLAMINGO/1.jpg  \n",
            "  inflating: test/AMERICAN FLAMINGO/2.jpg  \n",
            "  inflating: test/AMERICAN FLAMINGO/3.jpg  \n",
            "  inflating: test/AMERICAN FLAMINGO/4.jpg  \n",
            "  inflating: test/AMERICAN FLAMINGO/5.jpg  \n",
            "  inflating: test/AMERICAN GOLDFINCH/1.jpg  \n",
            "  inflating: test/AMERICAN GOLDFINCH/2.jpg  \n",
            "  inflating: test/AMERICAN GOLDFINCH/3.jpg  \n",
            "  inflating: test/AMERICAN GOLDFINCH/4.jpg  \n",
            "  inflating: test/AMERICAN GOLDFINCH/5.jpg  \n",
            "  inflating: test/AMERICAN KESTREL/1.jpg  \n",
            "  inflating: test/AMERICAN KESTREL/2.jpg  \n",
            "  inflating: test/AMERICAN KESTREL/3.jpg  \n",
            "  inflating: test/AMERICAN KESTREL/4.jpg  \n",
            "  inflating: test/AMERICAN KESTREL/5.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/001.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/002.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/003.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/004.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/005.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/006.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/007.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/008.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/009.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/010.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/011.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/012.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/013.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/014.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/015.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/016.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/017.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/018.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/019.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/020.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/021.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/022.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/023.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/024.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/025.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/026.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/027.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/028.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/029.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/030.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/031.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/032.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/033.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/034.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/035.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/036.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/037.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/038.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/039.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/040.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/041.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/042.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/043.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/044.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/045.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/046.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/047.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/048.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/049.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/050.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/051.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/052.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/053.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/054.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/055.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/056.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/057.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/058.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/059.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/060.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/061.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/062.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/063.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/064.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/065.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/066.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/067.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/068.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/069.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/070.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/071.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/072.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/073.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/074.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/075.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/076.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/077.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/078.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/079.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/080.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/081.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/082.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/083.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/084.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/085.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/086.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/087.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/088.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/089.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/090.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/091.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/092.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/093.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/094.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/095.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/096.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/097.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/098.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/099.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/100.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/101.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/102.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/103.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/104.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/105.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/106.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/107.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/108.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/109.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/110.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/111.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/112.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/113.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/114.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/115.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/116.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/117.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/118.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/119.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/120.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/121.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/122.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/123.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/124.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/125.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/126.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/127.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/128.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/129.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/130.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/131.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/132.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/133.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/134.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/135.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/136.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/137.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/138.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/139.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/140.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/141.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/142.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/143.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/144.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/145.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/146.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/147.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/148.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/149.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/150.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/151.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/152.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/153.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/154.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/155.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/156.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/157.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/158.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/159.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/160.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/161.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/162.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/163.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/164.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/165.jpg  \n",
            "  inflating: train/ABBOTTS BABBLER/166.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/001.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/002.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/003.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/004.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/005.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/006.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/007.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/008.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/009.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/010.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/011.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/012.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/013.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/014.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/015.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/016.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/017.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/018.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/019.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/020.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/021.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/022.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/023.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/024.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/025.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/026.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/027.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/028.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/029.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/030.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/031.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/032.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/033.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/034.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/035.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/036.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/037.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/038.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/039.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/040.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/041.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/042.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/043.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/044.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/045.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/046.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/047.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/048.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/049.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/050.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/051.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/052.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/053.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/054.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/055.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/056.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/057.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/058.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/059.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/060.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/061.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/062.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/063.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/064.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/065.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/066.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/067.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/068.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/069.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/070.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/071.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/072.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/073.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/074.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/075.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/076.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/077.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/078.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/079.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/080.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/081.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/082.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/083.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/084.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/085.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/086.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/087.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/088.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/089.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/090.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/091.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/092.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/093.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/094.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/095.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/096.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/097.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/098.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/099.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/100.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/101.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/102.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/103.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/104.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/105.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/106.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/107.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/108.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/109.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/110.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/111.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/112.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/113.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/114.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/115.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/116.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/117.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/118.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/119.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/120.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/121.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/122.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/123.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/124.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/125.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/126.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/127.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/128.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/129.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/130.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/131.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/132.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/133.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/134.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/135.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/136.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/137.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/138.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/139.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/140.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/141.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/142.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/143.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/144.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/145.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/146.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/147.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/148.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/149.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/150.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/151.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/152.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/153.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/154.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/155.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/156.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/157.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/158.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/159.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/160.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/161.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/162.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/163.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/164.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/165.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/166.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/167.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/168.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/169.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/170.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/171.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/172.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/173.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/174.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/175.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/176.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/177.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/178.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/179.jpg  \n",
            "  inflating: train/ABBOTTS BOOBY/180.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/001.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/002.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/003.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/004.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/005.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/006.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/007.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/008.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/009.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/010.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/011.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/012.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/013.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/014.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/015.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/016.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/017.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/018.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/019.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/020.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/021.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/022.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/023.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/024.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/025.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/026.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/027.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/028.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/029.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/030.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/031.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/032.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/033.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/034.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/035.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/036.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/037.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/038.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/039.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/040.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/041.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/042.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/043.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/044.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/045.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/046.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/047.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/048.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/049.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/050.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/051.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/052.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/053.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/054.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/055.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/056.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/057.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/058.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/059.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/060.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/061.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/062.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/063.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/064.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/065.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/066.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/067.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/068.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/069.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/070.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/071.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/072.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/073.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/074.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/075.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/076.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/077.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/078.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/079.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/080.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/081.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/082.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/083.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/084.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/085.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/086.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/087.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/088.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/089.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/090.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/091.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/092.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/093.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/094.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/095.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/096.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/097.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/098.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/099.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/100.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/101.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/102.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/103.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/104.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/105.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/106.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/107.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/108.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/109.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/110.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/111.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/112.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/113.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/114.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/115.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/116.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/117.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/118.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/119.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/120.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/121.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/122.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/123.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/124.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/125.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/126.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/127.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/128.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/129.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/130.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/131.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/132.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/133.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/134.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/135.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/136.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/137.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/138.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/139.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/140.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/141.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/142.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/143.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/144.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/145.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/146.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/147.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/148.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/149.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/150.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/151.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/152.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/153.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/154.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/155.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/156.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/157.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/158.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/159.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/160.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/161.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/162.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/163.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/164.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/165.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/166.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/167.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/168.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/169.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/170.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/171.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/172.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/173.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/174.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/175.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/176.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/177.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/178.jpg  \n",
            "  inflating: train/ABYSSINIAN GROUND HORNBILL/179.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/001.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/002.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/003.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/004.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/005.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/006.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/007.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/008.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/009.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/010.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/011.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/012.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/013.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/014.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/015.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/016.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/017.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/018.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/019.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/020.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/021.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/022.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/023.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/024.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/025.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/026.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/027.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/028.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/029.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/030.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/031.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/032.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/033.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/034.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/035.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/036.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/037.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/038.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/039.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/040.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/041.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/042.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/043.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/044.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/045.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/046.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/047.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/048.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/049.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/050.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/051.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/052.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/053.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/054.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/055.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/056.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/057.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/058.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/059.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/061.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/062.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/063.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/064.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/065.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/066.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/067.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/068.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/069.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/070.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/071.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/072.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/073.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/074.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/075.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/076.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/077.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/078.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/079.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/080.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/081.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/082.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/083.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/084.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/085.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/086.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/087.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/088.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/089.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/090.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/091.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/092.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/093.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/094.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/095.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/096.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/097.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/098.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/099.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/100.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/101.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/102.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/103.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/104.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/105.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/106.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/107.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/108.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/109.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/110.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/112.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/113.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/114.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/115.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/116.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/118.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/119.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/120.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/121.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/122.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/123.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/124.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/125.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/126.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/127.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/128.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/129.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/130.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/131.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/132.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/133.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/134.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/135.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/136.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/137.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/138.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/139.jpg  \n",
            "  inflating: train/AFRICAN CROWNED CRANE/140.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/001.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/002.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/003.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/004.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/005.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/006.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/007.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/008.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/009.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/010.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/011.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/012.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/013.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/014.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/015.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/016.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/017.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/018.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/019.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/020.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/021.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/022.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/023.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/024.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/025.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/026.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/027.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/028.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/029.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/030.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/031.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/032.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/033.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/034.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/035.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/036.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/037.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/038.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/039.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/040.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/041.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/042.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/043.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/044.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/045.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/046.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/047.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/048.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/049.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/050.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/051.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/052.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/053.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/054.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/055.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/056.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/057.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/058.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/059.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/060.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/061.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/062.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/063.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/064.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/065.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/066.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/067.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/068.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/069.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/070.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/071.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/072.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/073.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/074.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/075.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/076.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/077.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/078.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/079.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/080.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/081.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/082.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/083.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/084.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/085.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/086.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/087.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/088.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/089.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/090.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/091.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/092.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/093.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/094.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/095.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/096.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/097.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/098.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/099.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/100.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/101.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/102.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/103.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/104.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/105.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/106.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/107.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/108.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/109.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/110.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/111.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/112.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/113.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/114.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/115.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/116.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/117.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/118.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/119.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/120.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/121.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/122.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/123.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/124.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/125.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/126.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/127.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/128.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/129.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/130.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/131.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/132.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/133.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/134.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/135.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/136.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/137.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/138.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/139.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/140.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/141.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/142.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/143.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/144.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/145.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/146.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/147.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/148.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/149.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/150.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/151.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/152.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/153.jpg  \n",
            "  inflating: train/AFRICAN EMERALD CUCKOO/154.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/001.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/002.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/003.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/004.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/005.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/006.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/007.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/008.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/009.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/010.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/011.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/012.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/013.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/014.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/015.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/016.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/017.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/018.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/019.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/020.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/021.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/022.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/023.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/024.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/025.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/026.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/027.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/028.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/029.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/030.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/031.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/032.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/033.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/034.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/035.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/036.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/037.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/038.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/039.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/040.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/041.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/042.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/043.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/044.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/045.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/046.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/047.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/048.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/049.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/050.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/051.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/052.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/053.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/054.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/055.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/056.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/057.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/058.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/059.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/060.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/061.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/062.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/063.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/064.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/065.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/066.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/067.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/068.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/069.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/070.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/071.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/072.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/073.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/074.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/075.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/076.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/077.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/078.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/079.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/080.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/081.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/082.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/083.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/084.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/085.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/086.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/087.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/088.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/089.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/090.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/091.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/092.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/093.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/094.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/095.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/096.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/097.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/098.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/099.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/100.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/101.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/102.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/103.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/104.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/105.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/106.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/107.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/108.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/109.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/110.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/111.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/112.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/113.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/114.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/115.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/116.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/117.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/118.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/119.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/120.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/121.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/122.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/123.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/124.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/125.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/126.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/127.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/128.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/129.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/130.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/131.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/132.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/133.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/134.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/135.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/136.jpg  \n",
            "  inflating: train/AFRICAN FIREFINCH/137.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/001.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/002.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/003.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/004.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/005.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/006.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/007.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/008.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/009.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/010.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/011.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/012.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/013.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/014.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/015.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/016.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/017.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/018.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/019.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/020.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/021.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/022.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/023.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/024.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/025.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/026.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/027.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/028.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/029.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/030.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/031.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/032.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/033.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/034.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/035.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/036.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/037.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/038.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/039.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/040.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/041.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/042.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/043.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/044.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/045.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/046.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/047.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/048.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/049.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/050.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/051.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/052.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/053.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/054.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/055.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/056.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/057.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/058.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/059.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/060.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/061.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/062.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/063.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/064.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/065.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/066.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/067.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/068.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/069.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/070.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/071.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/072.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/073.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/074.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/075.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/076.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/077.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/078.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/079.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/080.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/081.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/082.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/083.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/084.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/085.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/086.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/087.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/088.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/089.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/090.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/091.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/092.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/093.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/094.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/095.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/096.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/097.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/098.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/099.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/100.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/101.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/102.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/103.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/104.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/105.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/106.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/107.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/108.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/109.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/110.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/111.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/112.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/113.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/114.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/115.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/116.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/117.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/118.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/119.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/120.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/121.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/122.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/123.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/124.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/125.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/126.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/127.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/128.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/129.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/130.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/131.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/132.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/133.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/134.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/135.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/136.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/137.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/138.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/139.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/140.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/141.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/142.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/143.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/144.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/145.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/146.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/147.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/148.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/149.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/150.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/151.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/152.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/153.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/154.jpg  \n",
            "  inflating: train/AFRICAN OYSTER CATCHER/155.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/001.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/002.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/003.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/004.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/005.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/006.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/007.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/008.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/009.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/010.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/011.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/012.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/013.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/014.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/015.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/016.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/017.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/018.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/019.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/020.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/021.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/022.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/023.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/024.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/025.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/026.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/027.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/028.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/029.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/030.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/031.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/032.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/033.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/034.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/035.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/036.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/037.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/038.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/039.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/040.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/041.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/042.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/043.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/044.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/045.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/046.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/047.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/048.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/049.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/050.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/051.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/052.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/053.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/054.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/055.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/056.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/057.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/058.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/059.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/060.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/061.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/062.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/063.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/064.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/065.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/066.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/067.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/068.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/069.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/070.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/071.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/072.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/073.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/074.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/075.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/076.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/077.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/078.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/079.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/080.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/081.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/082.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/083.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/084.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/085.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/086.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/087.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/088.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/089.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/090.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/091.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/092.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/093.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/094.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/095.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/096.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/097.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/098.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/099.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/100.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/101.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/102.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/103.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/104.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/105.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/106.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/107.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/108.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/109.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/110.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/111.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/112.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/113.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/114.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/115.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/116.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/117.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/118.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/119.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/120.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/121.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/122.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/123.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/124.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/125.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/126.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/127.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/128.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/129.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/130.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/131.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/132.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/133.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/134.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/135.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/136.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/137.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/138.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/139.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/140.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/141.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/142.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/143.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/144.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/145.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/146.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/147.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/148.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/149.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/150.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/151.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/152.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/153.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/154.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/155.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/156.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/157.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/158.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/159.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/160.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/161.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/162.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/163.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/164.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/165.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/166.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/167.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/168.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/169.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/170.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/171.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/172.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/173.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/174.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/175.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/176.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/177.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/178.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/179.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/180.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/181.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/182.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/183.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/184.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/185.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/186.jpg  \n",
            "  inflating: train/AFRICAN PIED HORNBILL/187.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/001.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/002.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/003.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/004.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/005.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/006.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/007.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/008.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/009.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/010.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/011.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/012.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/013.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/014.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/015.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/016.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/017.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/018.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/019.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/020.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/021.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/022.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/023.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/024.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/025.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/026.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/027.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/028.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/029.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/030.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/031.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/032.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/033.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/034.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/035.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/036.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/037.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/038.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/039.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/040.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/041.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/042.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/043.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/044.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/045.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/046.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/047.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/048.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/049.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/050.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/051.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/052.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/053.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/054.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/055.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/056.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/057.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/058.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/059.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/060.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/061.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/062.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/063.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/064.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/065.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/066.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/067.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/068.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/069.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/070.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/071.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/072.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/073.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/074.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/075.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/076.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/077.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/078.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/079.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/080.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/081.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/082.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/083.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/084.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/085.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/086.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/087.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/088.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/089.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/090.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/091.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/092.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/093.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/094.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/095.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/096.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/097.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/098.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/099.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/100.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/101.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/102.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/103.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/104.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/105.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/106.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/107.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/108.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/109.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/110.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/111.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/112.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/113.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/114.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/115.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/116.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/117.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/118.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/119.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/120.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/121.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/122.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/123.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/124.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/125.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/126.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/127.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/128.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/129.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/130.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/131.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/132.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/133.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/134.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/135.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/136.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/137.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/138.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/139.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/140.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/141.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/142.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/143.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/144.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/145.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/146.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/147.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/148.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/149.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/150.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/151.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/152.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/153.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/154.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/155.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/156.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/157.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/158.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/159.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/160.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/161.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/162.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/163.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/164.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/165.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/166.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/167.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/168.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/169.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/170.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/171.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/172.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/173.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/174.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/175.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/176.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/177.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/178.jpg  \n",
            "  inflating: train/AFRICAN PYGMY GOOSE/179.jpg  \n",
            "  inflating: train/ALBATROSS/001.jpg  \n",
            "  inflating: train/ALBATROSS/002.jpg  \n",
            "  inflating: train/ALBATROSS/003.jpg  \n",
            "  inflating: train/ALBATROSS/004.jpg  \n",
            "  inflating: train/ALBATROSS/005.jpg  \n",
            "  inflating: train/ALBATROSS/006.jpg  \n",
            "  inflating: train/ALBATROSS/007.jpg  \n",
            "  inflating: train/ALBATROSS/008.jpg  \n",
            "  inflating: train/ALBATROSS/009.jpg  \n",
            "  inflating: train/ALBATROSS/010.jpg  \n",
            "  inflating: train/ALBATROSS/011.jpg  \n",
            "  inflating: train/ALBATROSS/012.jpg  \n",
            "  inflating: train/ALBATROSS/013.jpg  \n",
            "  inflating: train/ALBATROSS/014.jpg  \n",
            "  inflating: train/ALBATROSS/015.jpg  \n",
            "  inflating: train/ALBATROSS/016.jpg  \n",
            "  inflating: train/ALBATROSS/017.jpg  \n",
            "  inflating: train/ALBATROSS/018.jpg  \n",
            "  inflating: train/ALBATROSS/019.jpg  \n",
            "  inflating: train/ALBATROSS/020.jpg  \n",
            "  inflating: train/ALBATROSS/021.jpg  \n",
            "  inflating: train/ALBATROSS/022.jpg  \n",
            "  inflating: train/ALBATROSS/023.jpg  \n",
            "  inflating: train/ALBATROSS/024.jpg  \n",
            "  inflating: train/ALBATROSS/025.jpg  \n",
            "  inflating: train/ALBATROSS/026.jpg  \n",
            "  inflating: train/ALBATROSS/027.jpg  \n",
            "  inflating: train/ALBATROSS/028.jpg  \n",
            "  inflating: train/ALBATROSS/029.jpg  \n",
            "  inflating: train/ALBATROSS/030.jpg  \n",
            "  inflating: train/ALBATROSS/031.jpg  \n",
            "  inflating: train/ALBATROSS/032.jpg  \n",
            "  inflating: train/ALBATROSS/033.jpg  \n",
            "  inflating: train/ALBATROSS/034.jpg  \n",
            "  inflating: train/ALBATROSS/035.jpg  \n",
            "  inflating: train/ALBATROSS/036.jpg  \n",
            "  inflating: train/ALBATROSS/037.jpg  \n",
            "  inflating: train/ALBATROSS/038.jpg  \n",
            "  inflating: train/ALBATROSS/039.jpg  \n",
            "  inflating: train/ALBATROSS/040.jpg  \n",
            "  inflating: train/ALBATROSS/041.jpg  \n",
            "  inflating: train/ALBATROSS/042.jpg  \n",
            "  inflating: train/ALBATROSS/043.jpg  \n",
            "  inflating: train/ALBATROSS/044.jpg  \n",
            "  inflating: train/ALBATROSS/045.jpg  \n",
            "  inflating: train/ALBATROSS/046.jpg  \n",
            "  inflating: train/ALBATROSS/047.jpg  \n",
            "  inflating: train/ALBATROSS/048.jpg  \n",
            "  inflating: train/ALBATROSS/049.jpg  \n",
            "  inflating: train/ALBATROSS/050.jpg  \n",
            "  inflating: train/ALBATROSS/051.jpg  \n",
            "  inflating: train/ALBATROSS/052.jpg  \n",
            "  inflating: train/ALBATROSS/053.jpg  \n",
            "  inflating: train/ALBATROSS/054.jpg  \n",
            "  inflating: train/ALBATROSS/055.jpg  \n",
            "  inflating: train/ALBATROSS/056.jpg  \n",
            "  inflating: train/ALBATROSS/057.jpg  \n",
            "  inflating: train/ALBATROSS/058.jpg  \n",
            "  inflating: train/ALBATROSS/059.jpg  \n",
            "  inflating: train/ALBATROSS/060.jpg  \n",
            "  inflating: train/ALBATROSS/061.jpg  \n",
            "  inflating: train/ALBATROSS/062.jpg  \n",
            "  inflating: train/ALBATROSS/063.jpg  \n",
            "  inflating: train/ALBATROSS/064.jpg  \n",
            "  inflating: train/ALBATROSS/065.jpg  \n",
            "  inflating: train/ALBATROSS/066.jpg  \n",
            "  inflating: train/ALBATROSS/067.jpg  \n",
            "  inflating: train/ALBATROSS/068.jpg  \n",
            "  inflating: train/ALBATROSS/069.jpg  \n",
            "  inflating: train/ALBATROSS/070.jpg  \n",
            "  inflating: train/ALBATROSS/071.jpg  \n",
            "  inflating: train/ALBATROSS/072.jpg  \n",
            "  inflating: train/ALBATROSS/073.jpg  \n",
            "  inflating: train/ALBATROSS/074.jpg  \n",
            "  inflating: train/ALBATROSS/075.jpg  \n",
            "  inflating: train/ALBATROSS/076.jpg  \n",
            "  inflating: train/ALBATROSS/077.jpg  \n",
            "  inflating: train/ALBATROSS/078.jpg  \n",
            "  inflating: train/ALBATROSS/079.jpg  \n",
            "  inflating: train/ALBATROSS/080.jpg  \n",
            "  inflating: train/ALBATROSS/081.jpg  \n",
            "  inflating: train/ALBATROSS/082.jpg  \n",
            "  inflating: train/ALBATROSS/083.jpg  \n",
            "  inflating: train/ALBATROSS/084.jpg  \n",
            "  inflating: train/ALBATROSS/085.jpg  \n",
            "  inflating: train/ALBATROSS/086.jpg  \n",
            "  inflating: train/ALBATROSS/087.jpg  \n",
            "  inflating: train/ALBATROSS/088.jpg  \n",
            "  inflating: train/ALBATROSS/089.jpg  \n",
            "  inflating: train/ALBATROSS/090.jpg  \n",
            "  inflating: train/ALBATROSS/091.jpg  \n",
            "  inflating: train/ALBATROSS/092.jpg  \n",
            "  inflating: train/ALBATROSS/093.jpg  \n",
            "  inflating: train/ALBATROSS/094.jpg  \n",
            "  inflating: train/ALBATROSS/095.jpg  \n",
            "  inflating: train/ALBATROSS/096.jpg  \n",
            "  inflating: train/ALBATROSS/097.jpg  \n",
            "  inflating: train/ALBATROSS/098.jpg  \n",
            "  inflating: train/ALBATROSS/099.jpg  \n",
            "  inflating: train/ALBATROSS/100.jpg  \n",
            "  inflating: train/ALBATROSS/101.jpg  \n",
            "  inflating: train/ALBATROSS/102.jpg  \n",
            "  inflating: train/ALBATROSS/103.jpg  \n",
            "  inflating: train/ALBATROSS/104.jpg  \n",
            "  inflating: train/ALBATROSS/105.jpg  \n",
            "  inflating: train/ALBATROSS/106.jpg  \n",
            "  inflating: train/ALBATROSS/107.jpg  \n",
            "  inflating: train/ALBATROSS/108.jpg  \n",
            "  inflating: train/ALBATROSS/109.jpg  \n",
            "  inflating: train/ALBATROSS/110.jpg  \n",
            "  inflating: train/ALBATROSS/111.jpg  \n",
            "  inflating: train/ALBATROSS/112.jpg  \n",
            "  inflating: train/ALBATROSS/113.jpg  \n",
            "  inflating: train/ALBATROSS/114.jpg  \n",
            "  inflating: train/ALBATROSS/115.jpg  \n",
            "  inflating: train/ALBATROSS/116.jpg  \n",
            "  inflating: train/ALBATROSS/117.jpg  \n",
            "  inflating: train/ALBATROSS/118.jpg  \n",
            "  inflating: train/ALBATROSS/119.jpg  \n",
            "  inflating: train/ALBATROSS/120.jpg  \n",
            "  inflating: train/ALBATROSS/121.jpg  \n",
            "  inflating: train/ALBATROSS/122.jpg  \n",
            "  inflating: train/ALBATROSS/123.jpg  \n",
            "  inflating: train/ALBATROSS/124.jpg  \n",
            "  inflating: train/ALBATROSS/125.jpg  \n",
            "  inflating: train/ALBATROSS/126.jpg  \n",
            "  inflating: train/ALBATROSS/127.jpg  \n",
            "  inflating: train/ALBATROSS/128.jpg  \n",
            "  inflating: train/ALBATROSS/129.jpg  \n",
            "  inflating: train/ALBATROSS/130.jpg  \n",
            "  inflating: train/ALBATROSS/131.jpg  \n",
            "  inflating: train/ALBATROSS/132.jpg  \n",
            "  inflating: train/ALBATROSS/133.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/001.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/002.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/003.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/004.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/005.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/006.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/007.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/008.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/009.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/010.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/011.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/012.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/013.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/014.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/015.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/016.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/017.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/018.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/019.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/020.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/021.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/022.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/023.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/024.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/025.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/026.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/027.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/028.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/029.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/030.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/031.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/032.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/033.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/034.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/035.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/036.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/037.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/038.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/039.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/040.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/041.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/042.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/043.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/044.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/045.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/046.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/047.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/048.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/049.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/050.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/051.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/052.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/053.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/054.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/055.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/056.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/057.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/058.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/059.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/060.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/061.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/062.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/063.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/064.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/065.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/066.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/067.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/068.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/069.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/070.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/071.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/072.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/073.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/074.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/075.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/076.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/077.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/078.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/079.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/080.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/081.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/082.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/083.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/084.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/085.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/086.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/087.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/088.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/089.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/090.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/091.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/092.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/093.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/094.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/095.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/096.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/097.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/098.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/099.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/100.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/101.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/102.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/103.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/104.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/105.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/106.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/107.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/108.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/109.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/110.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/111.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/112.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/113.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/114.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/115.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/116.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/117.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/118.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/119.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/120.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/121.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/122.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/123.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/124.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/125.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/126.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/127.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/128.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/129.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/130.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/131.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/132.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/133.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/134.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/135.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/136.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/137.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/138.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/139.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/140.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/141.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/142.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/143.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/144.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/145.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/146.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/147.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/148.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/149.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/150.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/151.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/152.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/153.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/154.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/155.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/156.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/157.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/158.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/159.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/160.jpg  \n",
            "  inflating: train/ALBERTS TOWHEE/161.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/001.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/002.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/003.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/004.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/005.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/006.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/007.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/008.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/009.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/010.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/011.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/012.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/013.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/014.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/015.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/016.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/017.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/018.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/019.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/020.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/021.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/022.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/023.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/024.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/025.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/026.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/027.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/028.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/029.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/030.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/031.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/032.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/033.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/034.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/035.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/036.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/037.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/038.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/039.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/040.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/041.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/042.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/043.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/044.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/045.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/046.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/047.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/048.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/049.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/050.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/051.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/052.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/053.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/054.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/055.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/056.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/057.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/058.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/059.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/060.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/061.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/062.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/063.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/064.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/065.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/066.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/067.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/068.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/069.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/070.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/071.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/072.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/073.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/074.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/075.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/076.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/077.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/078.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/079.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/080.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/081.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/082.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/083.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/085.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/086.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/087.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/088.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/089.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/090.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/091.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/092.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/093.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/094.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/095.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/096.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/097.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/098.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/099.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/100.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/101.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/102.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/103.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/104.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/105.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/106.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/107.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/108.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/109.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/110.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/111.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/112.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/113.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/114.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/115.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/116.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/117.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/118.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/119.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/120.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/121.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/122.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/123.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/124.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/125.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/126.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/127.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/128.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/129.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/130.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/131.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/132.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/133.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/134.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/135.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/136.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/137.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/138.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/139.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/140.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/141.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/142.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/143.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/144.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/145.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/146.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/147.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/148.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/149.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/150.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/151.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/152.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/153.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/154.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/155.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/156.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/157.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/158.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/159.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/160.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/161.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/162.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/163.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/164.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/165.jpg  \n",
            "  inflating: train/ALEXANDRINE PARAKEET/166.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/001.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/002.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/003.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/004.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/005.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/006.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/007.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/008.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/009.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/010.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/011.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/012.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/013.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/014.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/015.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/016.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/017.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/018.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/019.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/020.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/021.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/022.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/023.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/024.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/025.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/026.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/027.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/028.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/029.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/030.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/031.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/032.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/033.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/034.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/035.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/036.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/037.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/038.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/039.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/040.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/041.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/042.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/043.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/044.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/045.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/046.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/047.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/048.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/049.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/050.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/051.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/052.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/053.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/054.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/055.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/056.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/057.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/058.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/059.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/060.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/061.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/062.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/063.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/064.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/065.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/066.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/067.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/068.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/069.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/070.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/071.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/072.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/073.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/074.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/075.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/076.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/077.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/078.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/079.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/080.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/081.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/082.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/083.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/084.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/085.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/086.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/087.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/088.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/089.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/090.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/091.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/092.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/093.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/094.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/095.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/096.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/097.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/098.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/099.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/100.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/101.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/102.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/103.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/104.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/105.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/106.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/107.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/108.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/109.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/110.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/111.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/112.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/113.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/114.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/115.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/116.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/117.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/118.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/119.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/120.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/121.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/122.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/123.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/124.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/125.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/126.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/127.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/128.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/129.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/130.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/131.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/132.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/133.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/134.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/135.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/136.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/137.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/138.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/139.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/140.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/141.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/142.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/143.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/144.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/145.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/146.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/147.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/148.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/149.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/150.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/151.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/152.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/153.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/154.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/155.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/156.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/157.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/158.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/159.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/160.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/161.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/162.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/163.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/164.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/165.jpg  \n",
            "  inflating: train/ALPINE CHOUGH/166.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/001.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/002.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/003.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/004.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/005.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/006.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/007.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/008.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/009.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/010.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/011.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/012.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/013.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/014.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/015.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/016.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/017.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/018.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/019.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/020.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/021.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/022.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/023.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/024.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/025.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/026.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/027.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/028.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/029.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/030.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/031.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/032.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/033.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/034.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/035.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/036.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/037.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/038.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/039.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/040.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/041.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/042.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/043.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/044.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/045.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/046.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/047.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/048.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/049.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/050.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/051.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/052.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/053.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/054.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/055.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/056.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/057.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/058.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/059.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/060.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/061.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/062.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/063.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/064.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/065.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/066.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/067.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/068.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/069.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/070.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/071.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/072.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/073.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/074.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/075.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/076.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/077.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/078.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/079.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/080.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/081.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/082.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/083.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/084.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/085.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/086.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/087.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/088.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/089.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/090.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/091.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/092.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/093.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/094.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/095.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/096.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/097.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/098.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/099.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/100.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/101.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/102.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/103.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/104.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/105.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/106.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/107.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/108.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/109.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/110.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/111.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/112.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/113.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/114.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/115.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/116.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/117.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/118.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/119.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/120.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/121.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/122.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/123.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/124.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/125.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/126.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/127.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/128.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/129.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/130.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/131.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/132.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/133.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/134.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/135.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/136.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/137.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/138.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/139.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/140.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/141.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/142.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/143.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/144.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/145.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/146.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/147.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/148.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/149.jpg  \n",
            "  inflating: train/ALTAMIRA YELLOWTHROAT/150.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/001.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/002.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/003.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/004.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/005.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/006.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/007.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/008.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/009.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/010.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/011.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/012.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/013.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/014.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/015.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/016.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/017.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/018.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/019.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/020.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/021.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/022.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/023.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/024.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/025.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/026.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/027.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/028.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/029.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/030.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/031.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/032.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/033.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/034.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/035.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/036.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/037.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/038.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/039.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/040.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/041.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/042.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/043.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/044.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/045.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/046.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/047.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/048.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/049.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/050.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/051.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/052.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/053.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/054.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/055.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/056.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/057.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/058.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/059.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/060.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/061.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/062.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/063.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/064.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/065.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/066.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/067.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/068.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/069.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/070.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/071.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/072.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/073.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/074.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/075.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/076.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/077.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/078.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/079.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/080.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/081.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/082.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/083.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/084.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/085.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/086.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/087.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/088.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/089.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/090.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/091.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/092.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/093.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/094.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/095.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/096.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/097.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/098.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/099.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/100.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/101.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/102.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/103.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/104.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/105.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/106.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/107.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/108.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/109.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/110.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/111.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/112.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/113.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/114.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/115.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/116.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/117.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/118.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/119.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/120.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/121.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/122.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/123.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/124.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/125.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/126.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/127.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/128.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/129.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/130.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/131.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/132.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/133.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/134.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/135.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/136.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/137.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/138.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/139.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/140.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/141.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/142.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/143.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/144.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/145.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/146.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/147.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/148.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/149.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/150.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/151.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/152.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/153.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/154.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/155.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/156.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/157.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/158.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/159.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/160.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/161.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/162.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/163.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/164.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/165.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/166.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/168.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/169.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/170.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/171.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/172.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/173.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/174.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/175.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/176.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/177.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/178.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/179.jpg  \n",
            "  inflating: train/AMERICAN AVOCET/180.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/001.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/002.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/003.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/004.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/005.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/006.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/007.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/008.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/009.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/010.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/011.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/012.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/013.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/014.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/015.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/016.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/017.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/018.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/019.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/020.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/021.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/022.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/023.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/024.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/025.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/026.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/027.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/028.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/029.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/030.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/031.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/032.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/033.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/034.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/035.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/036.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/037.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/038.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/039.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/040.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/041.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/042.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/043.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/044.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/045.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/046.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/047.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/048.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/049.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/050.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/051.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/052.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/053.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/054.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/055.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/056.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/057.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/058.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/059.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/060.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/061.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/062.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/063.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/064.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/065.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/066.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/067.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/068.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/069.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/070.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/071.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/072.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/073.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/074.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/075.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/076.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/077.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/078.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/079.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/080.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/081.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/082.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/083.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/084.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/085.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/086.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/087.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/088.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/089.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/090.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/091.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/092.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/093.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/094.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/095.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/096.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/097.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/098.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/099.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/100.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/101.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/102.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/103.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/104.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/105.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/106.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/107.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/108.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/109.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/110.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/111.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/112.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/113.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/114.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/115.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/116.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/117.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/118.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/119.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/120.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/121.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/122.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/123.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/124.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/125.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/126.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/127.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/128.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/129.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/130.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/131.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/132.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/133.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/134.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/135.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/136.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/137.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/138.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/139.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/140.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/141.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/142.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/143.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/144.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/145.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/146.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/147.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/148.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/149.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/150.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/151.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/152.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/153.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/154.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/155.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/156.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/157.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/158.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/159.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/160.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/161.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/162.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/163.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/164.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/165.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/166.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/167.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/168.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/169.jpg  \n",
            "  inflating: train/AMERICAN BITTERN/170.jpg  \n",
            "  inflating: train/AMERICAN COOT/001.jpg  \n",
            "  inflating: train/AMERICAN COOT/002.jpg  \n",
            "  inflating: train/AMERICAN COOT/003.jpg  \n",
            "  inflating: train/AMERICAN COOT/004.jpg  \n",
            "  inflating: train/AMERICAN COOT/005.jpg  \n",
            "  inflating: train/AMERICAN COOT/006.jpg  \n",
            "  inflating: train/AMERICAN COOT/007.jpg  \n",
            "  inflating: train/AMERICAN COOT/008.jpg  \n",
            "  inflating: train/AMERICAN COOT/009.jpg  \n",
            "  inflating: train/AMERICAN COOT/010.jpg  \n",
            "  inflating: train/AMERICAN COOT/011.jpg  \n",
            "  inflating: train/AMERICAN COOT/012.jpg  \n",
            "  inflating: train/AMERICAN COOT/013.jpg  \n",
            "  inflating: train/AMERICAN COOT/014.jpg  \n",
            "  inflating: train/AMERICAN COOT/015.jpg  \n",
            "  inflating: train/AMERICAN COOT/016.jpg  \n",
            "  inflating: train/AMERICAN COOT/017.jpg  \n",
            "  inflating: train/AMERICAN COOT/018.jpg  \n",
            "  inflating: train/AMERICAN COOT/019.jpg  \n",
            "  inflating: train/AMERICAN COOT/020.jpg  \n",
            "  inflating: train/AMERICAN COOT/021.jpg  \n",
            "  inflating: train/AMERICAN COOT/022.jpg  \n",
            "  inflating: train/AMERICAN COOT/023.jpg  \n",
            "  inflating: train/AMERICAN COOT/024.jpg  \n",
            "  inflating: train/AMERICAN COOT/025.jpg  \n",
            "  inflating: train/AMERICAN COOT/026.jpg  \n",
            "  inflating: train/AMERICAN COOT/027.jpg  \n",
            "  inflating: train/AMERICAN COOT/028.jpg  \n",
            "  inflating: train/AMERICAN COOT/029.jpg  \n",
            "  inflating: train/AMERICAN COOT/030.jpg  \n",
            "  inflating: train/AMERICAN COOT/031.jpg  \n",
            "  inflating: train/AMERICAN COOT/032.jpg  \n",
            "  inflating: train/AMERICAN COOT/033.jpg  \n",
            "  inflating: train/AMERICAN COOT/034.jpg  \n",
            "  inflating: train/AMERICAN COOT/035.jpg  \n",
            "  inflating: train/AMERICAN COOT/036.jpg  \n",
            "  inflating: train/AMERICAN COOT/037.jpg  \n",
            "  inflating: train/AMERICAN COOT/038.jpg  \n",
            "  inflating: train/AMERICAN COOT/039.jpg  \n",
            "  inflating: train/AMERICAN COOT/040.jpg  \n",
            "  inflating: train/AMERICAN COOT/041.jpg  \n",
            "  inflating: train/AMERICAN COOT/042.jpg  \n",
            "  inflating: train/AMERICAN COOT/043.jpg  \n",
            "  inflating: train/AMERICAN COOT/044.jpg  \n",
            "  inflating: train/AMERICAN COOT/045.jpg  \n",
            "  inflating: train/AMERICAN COOT/046.jpg  \n",
            "  inflating: train/AMERICAN COOT/047.jpg  \n",
            "  inflating: train/AMERICAN COOT/048.jpg  \n",
            "  inflating: train/AMERICAN COOT/049.jpg  \n",
            "  inflating: train/AMERICAN COOT/050.jpg  \n",
            "  inflating: train/AMERICAN COOT/051.jpg  \n",
            "  inflating: train/AMERICAN COOT/052.jpg  \n",
            "  inflating: train/AMERICAN COOT/053.jpg  \n",
            "  inflating: train/AMERICAN COOT/054.jpg  \n",
            "  inflating: train/AMERICAN COOT/055.jpg  \n",
            "  inflating: train/AMERICAN COOT/056.jpg  \n",
            "  inflating: train/AMERICAN COOT/057.jpg  \n",
            "  inflating: train/AMERICAN COOT/058.jpg  \n",
            "  inflating: train/AMERICAN COOT/059.jpg  \n",
            "  inflating: train/AMERICAN COOT/060.jpg  \n",
            "  inflating: train/AMERICAN COOT/061.jpg  \n",
            "  inflating: train/AMERICAN COOT/062.jpg  \n",
            "  inflating: train/AMERICAN COOT/063.jpg  \n",
            "  inflating: train/AMERICAN COOT/064.jpg  \n",
            "  inflating: train/AMERICAN COOT/065.jpg  \n",
            "  inflating: train/AMERICAN COOT/066.jpg  \n",
            "  inflating: train/AMERICAN COOT/067.jpg  \n",
            "  inflating: train/AMERICAN COOT/068.jpg  \n",
            "  inflating: train/AMERICAN COOT/069.jpg  \n",
            "  inflating: train/AMERICAN COOT/070.jpg  \n",
            "  inflating: train/AMERICAN COOT/071.jpg  \n",
            "  inflating: train/AMERICAN COOT/072.jpg  \n",
            "  inflating: train/AMERICAN COOT/073.jpg  \n",
            "  inflating: train/AMERICAN COOT/074.jpg  \n",
            "  inflating: train/AMERICAN COOT/075.jpg  \n",
            "  inflating: train/AMERICAN COOT/076.jpg  \n",
            "  inflating: train/AMERICAN COOT/077.jpg  \n",
            "  inflating: train/AMERICAN COOT/078.jpg  \n",
            "  inflating: train/AMERICAN COOT/079.jpg  \n",
            "  inflating: train/AMERICAN COOT/080.jpg  \n",
            "  inflating: train/AMERICAN COOT/081.jpg  \n",
            "  inflating: train/AMERICAN COOT/082.jpg  \n",
            "  inflating: train/AMERICAN COOT/083.jpg  \n",
            "  inflating: train/AMERICAN COOT/084.jpg  \n",
            "  inflating: train/AMERICAN COOT/085.jpg  \n",
            "  inflating: train/AMERICAN COOT/086.jpg  \n",
            "  inflating: train/AMERICAN COOT/087.jpg  \n",
            "  inflating: train/AMERICAN COOT/088.jpg  \n",
            "  inflating: train/AMERICAN COOT/089.jpg  \n",
            "  inflating: train/AMERICAN COOT/090.jpg  \n",
            "  inflating: train/AMERICAN COOT/091.jpg  \n",
            "  inflating: train/AMERICAN COOT/092.jpg  \n",
            "  inflating: train/AMERICAN COOT/093.jpg  \n",
            "  inflating: train/AMERICAN COOT/094.jpg  \n",
            "  inflating: train/AMERICAN COOT/095.jpg  \n",
            "  inflating: train/AMERICAN COOT/096.jpg  \n",
            "  inflating: train/AMERICAN COOT/097.jpg  \n",
            "  inflating: train/AMERICAN COOT/098.jpg  \n",
            "  inflating: train/AMERICAN COOT/099.jpg  \n",
            "  inflating: train/AMERICAN COOT/100.jpg  \n",
            "  inflating: train/AMERICAN COOT/101.jpg  \n",
            "  inflating: train/AMERICAN COOT/102.jpg  \n",
            "  inflating: train/AMERICAN COOT/103.jpg  \n",
            "  inflating: train/AMERICAN COOT/104.jpg  \n",
            "  inflating: train/AMERICAN COOT/105.jpg  \n",
            "  inflating: train/AMERICAN COOT/106.jpg  \n",
            "  inflating: train/AMERICAN COOT/107.jpg  \n",
            "  inflating: train/AMERICAN COOT/108.jpg  \n",
            "  inflating: train/AMERICAN COOT/109.jpg  \n",
            "  inflating: train/AMERICAN COOT/110.jpg  \n",
            "  inflating: train/AMERICAN COOT/111.jpg  \n",
            "  inflating: train/AMERICAN COOT/112.jpg  \n",
            "  inflating: train/AMERICAN COOT/113.jpg  \n",
            "  inflating: train/AMERICAN COOT/114.jpg  \n",
            "  inflating: train/AMERICAN COOT/115.jpg  \n",
            "  inflating: train/AMERICAN COOT/116.jpg  \n",
            "  inflating: train/AMERICAN COOT/117.jpg  \n",
            "  inflating: train/AMERICAN COOT/118.jpg  \n",
            "  inflating: train/AMERICAN COOT/119.jpg  \n",
            "  inflating: train/AMERICAN COOT/120.jpg  \n",
            "  inflating: train/AMERICAN COOT/121.jpg  \n",
            "  inflating: train/AMERICAN COOT/122.jpg  \n",
            "  inflating: train/AMERICAN COOT/123.jpg  \n",
            "  inflating: train/AMERICAN COOT/124.jpg  \n",
            "  inflating: train/AMERICAN COOT/125.jpg  \n",
            "  inflating: train/AMERICAN COOT/126.jpg  \n",
            "  inflating: train/AMERICAN COOT/127.jpg  \n",
            "  inflating: train/AMERICAN COOT/128.jpg  \n",
            "  inflating: train/AMERICAN COOT/129.jpg  \n",
            "  inflating: train/AMERICAN COOT/130.jpg  \n",
            "  inflating: train/AMERICAN COOT/131.jpg  \n",
            "  inflating: train/AMERICAN COOT/132.jpg  \n",
            "  inflating: train/AMERICAN COOT/133.jpg  \n",
            "  inflating: train/AMERICAN COOT/134.jpg  \n",
            "  inflating: train/AMERICAN COOT/135.jpg  \n",
            "  inflating: train/AMERICAN COOT/136.jpg  \n",
            "  inflating: train/AMERICAN COOT/137.jpg  \n",
            "  inflating: train/AMERICAN COOT/138.jpg  \n",
            "  inflating: train/AMERICAN COOT/139.jpg  \n",
            "  inflating: train/AMERICAN COOT/140.jpg  \n",
            "  inflating: train/AMERICAN COOT/141.jpg  \n",
            "  inflating: train/AMERICAN COOT/142.jpg  \n",
            "  inflating: train/AMERICAN COOT/143.jpg  \n",
            "  inflating: train/AMERICAN COOT/144.jpg  \n",
            "  inflating: train/AMERICAN COOT/145.jpg  \n",
            "  inflating: train/AMERICAN COOT/146.jpg  \n",
            "  inflating: train/AMERICAN COOT/147.jpg  \n",
            "  inflating: train/AMERICAN COOT/148.jpg  \n",
            "  inflating: train/AMERICAN COOT/149.jpg  \n",
            "  inflating: train/AMERICAN COOT/150.jpg  \n",
            "  inflating: train/AMERICAN COOT/151.jpg  \n",
            "  inflating: train/AMERICAN COOT/152.jpg  \n",
            "  inflating: train/AMERICAN COOT/153.jpg  \n",
            "  inflating: train/AMERICAN COOT/154.jpg  \n",
            "  inflating: train/AMERICAN COOT/155.jpg  \n",
            "  inflating: train/AMERICAN COOT/156.jpg  \n",
            "  inflating: train/AMERICAN COOT/157.jpg  \n",
            "  inflating: train/AMERICAN COOT/158.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/001.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/002.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/003.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/004.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/005.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/006.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/007.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/008.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/009.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/010.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/011.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/012.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/013.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/014.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/015.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/016.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/017.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/018.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/019.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/020.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/021.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/022.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/023.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/024.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/025.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/026.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/027.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/028.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/029.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/030.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/031.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/032.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/033.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/034.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/035.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/036.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/037.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/038.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/039.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/040.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/041.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/042.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/043.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/044.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/045.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/046.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/047.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/048.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/049.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/050.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/051.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/052.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/053.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/054.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/055.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/056.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/057.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/058.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/059.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/060.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/061.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/062.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/063.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/064.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/065.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/066.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/067.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/068.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/069.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/070.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/071.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/072.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/073.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/074.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/075.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/076.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/077.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/078.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/079.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/080.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/081.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/082.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/083.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/084.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/085.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/086.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/087.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/088.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/089.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/090.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/091.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/092.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/093.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/094.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/095.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/096.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/097.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/098.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/099.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/100.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/101.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/102.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/103.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/104.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/105.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/106.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/107.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/108.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/109.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/110.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/111.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/112.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/113.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/114.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/115.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/116.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/117.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/118.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/119.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/120.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/121.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/122.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/123.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/124.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/125.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/126.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/127.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/128.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/129.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/130.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/131.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/132.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/133.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/134.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/135.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/136.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/137.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/138.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/139.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/140.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/141.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/142.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/143.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/144.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/145.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/146.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/147.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/148.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/149.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/150.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/151.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/152.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/153.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/154.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/155.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/156.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/157.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/158.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/159.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/160.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/161.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/162.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/163.jpg  \n",
            "  inflating: train/AMERICAN FLAMINGO/164.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/001.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/002.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/003.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/004.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/005.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/006.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/007.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/008.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/009.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/010.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/011.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/012.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/013.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/014.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/015.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/016.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/017.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/018.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/019.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/020.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/021.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/022.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/023.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/024.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/025.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/026.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/027.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/028.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/029.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/030.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/031.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/032.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/033.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/034.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/035.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/036.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/037.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/038.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/039.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/040.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/041.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/042.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/043.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/044.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/045.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/046.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/047.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/049.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/050.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/051.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/052.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/053.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/054.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/055.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/056.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/057.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/058.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/059.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/060.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/061.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/062.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/063.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/064.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/065.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/066.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/067.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/068.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/069.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/070.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/071.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/072.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/073.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/074.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/075.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/076.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/077.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/078.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/079.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/080.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/081.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/082.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/083.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/084.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/085.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/086.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/087.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/088.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/089.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/090.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/091.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/092.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/093.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/094.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/095.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/096.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/097.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/098.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/099.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/100.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/101.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/102.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/103.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/104.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/105.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/106.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/107.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/108.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/109.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/110.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/111.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/112.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/113.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/114.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/115.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/116.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/117.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/118.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/119.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/120.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/121.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/122.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/123.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/124.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/125.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/126.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/127.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/128.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/129.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/130.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/131.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/132.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/133.jpg  \n",
            "  inflating: train/AMERICAN GOLDFINCH/134.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/001.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/002.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/003.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/004.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/005.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/006.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/007.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/008.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/009.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/010.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/011.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/012.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/013.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/014.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/015.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/016.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/017.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/018.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/019.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/020.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/021.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/022.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/023.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/024.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/025.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/026.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/027.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/028.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/029.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/030.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/031.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/032.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/033.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/034.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/035.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/036.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/037.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/038.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/039.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/040.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/041.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/042.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/043.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/044.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/045.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/046.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/047.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/048.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/049.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/050.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/051.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/052.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/053.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/054.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/055.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/056.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/057.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/058.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/059.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/060.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/061.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/062.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/063.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/064.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/065.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/066.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/067.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/068.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/069.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/070.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/071.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/072.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/073.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/074.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/075.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/076.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/077.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/078.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/079.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/080.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/081.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/082.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/083.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/084.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/085.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/086.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/087.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/088.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/089.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/090.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/091.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/092.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/093.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/094.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/095.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/096.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/097.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/098.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/099.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/100.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/101.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/102.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/103.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/104.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/105.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/106.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/107.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/108.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/109.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/110.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/111.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/112.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/113.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/114.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/115.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/116.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/117.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/118.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/119.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/120.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/121.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/122.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/123.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/124.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/125.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/126.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/127.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/128.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/129.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/130.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/131.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/132.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/133.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/134.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/135.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/136.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/137.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/138.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/139.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/140.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/141.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/142.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/143.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/144.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/145.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/146.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/147.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/148.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/149.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/150.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/151.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/152.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/153.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/154.jpg  \n",
            "  inflating: train/AMERICAN KESTREL/155.jpg  \n",
            "  inflating: valid/ABBOTTS BABBLER/1.jpg  \n",
            "  inflating: valid/ABBOTTS BABBLER/2.jpg  \n",
            "  inflating: valid/ABBOTTS BABBLER/3.jpg  \n",
            "  inflating: valid/ABBOTTS BABBLER/4.jpg  \n",
            "  inflating: valid/ABBOTTS BABBLER/5.jpg  \n",
            "  inflating: valid/ABBOTTS BOOBY/1.jpg  \n",
            "  inflating: valid/ABBOTTS BOOBY/2.jpg  \n",
            "  inflating: valid/ABBOTTS BOOBY/3.jpg  \n",
            "  inflating: valid/ABBOTTS BOOBY/4.jpg  \n",
            "  inflating: valid/ABBOTTS BOOBY/5.jpg  \n",
            "  inflating: valid/ABYSSINIAN GROUND HORNBILL/1.jpg  \n",
            "  inflating: valid/ABYSSINIAN GROUND HORNBILL/2.jpg  \n",
            "  inflating: valid/ABYSSINIAN GROUND HORNBILL/3.jpg  \n",
            "  inflating: valid/ABYSSINIAN GROUND HORNBILL/4.jpg  \n",
            "  inflating: valid/ABYSSINIAN GROUND HORNBILL/5.jpg  \n",
            "  inflating: valid/AFRICAN CROWNED CRANE/1.jpg  \n",
            "  inflating: valid/AFRICAN CROWNED CRANE/2.jpg  \n",
            "  inflating: valid/AFRICAN CROWNED CRANE/3.jpg  \n",
            "  inflating: valid/AFRICAN CROWNED CRANE/4.jpg  \n",
            "  inflating: valid/AFRICAN CROWNED CRANE/5.jpg  \n",
            "  inflating: valid/AFRICAN EMERALD CUCKOO/1.jpg  \n",
            "  inflating: valid/AFRICAN EMERALD CUCKOO/2.jpg  \n",
            "  inflating: valid/AFRICAN EMERALD CUCKOO/3.jpg  \n",
            "  inflating: valid/AFRICAN EMERALD CUCKOO/4.jpg  \n",
            "  inflating: valid/AFRICAN EMERALD CUCKOO/5.jpg  \n",
            "  inflating: valid/AFRICAN FIREFINCH/1.jpg  \n",
            "  inflating: valid/AFRICAN FIREFINCH/2.jpg  \n",
            "  inflating: valid/AFRICAN FIREFINCH/3.jpg  \n",
            "  inflating: valid/AFRICAN FIREFINCH/4.jpg  \n",
            "  inflating: valid/AFRICAN FIREFINCH/5.jpg  \n",
            "  inflating: valid/AFRICAN OYSTER CATCHER/1.jpg  \n",
            "  inflating: valid/AFRICAN OYSTER CATCHER/2.jpg  \n",
            "  inflating: valid/AFRICAN OYSTER CATCHER/3.jpg  \n",
            "  inflating: valid/AFRICAN OYSTER CATCHER/4.jpg  \n",
            "  inflating: valid/AFRICAN OYSTER CATCHER/5.jpg  \n",
            "  inflating: valid/AFRICAN PIED HORNBILL/1.jpg  \n",
            "  inflating: valid/AFRICAN PIED HORNBILL/2.jpg  \n",
            "  inflating: valid/AFRICAN PIED HORNBILL/3.jpg  \n",
            "  inflating: valid/AFRICAN PIED HORNBILL/4.jpg  \n",
            "  inflating: valid/AFRICAN PIED HORNBILL/5.jpg  \n",
            "  inflating: valid/AFRICAN PYGMY GOOSE/1.jpg  \n",
            "  inflating: valid/AFRICAN PYGMY GOOSE/2.jpg  \n",
            "  inflating: valid/AFRICAN PYGMY GOOSE/3.jpg  \n",
            "  inflating: valid/AFRICAN PYGMY GOOSE/4.jpg  \n",
            "  inflating: valid/AFRICAN PYGMY GOOSE/5.jpg  \n",
            "  inflating: valid/ALBATROSS/1.jpg   \n",
            "  inflating: valid/ALBATROSS/2.jpg   \n",
            "  inflating: valid/ALBATROSS/3.jpg   \n",
            "  inflating: valid/ALBATROSS/4.jpg   \n",
            "  inflating: valid/ALBATROSS/5.jpg   \n",
            "  inflating: valid/ALBERTS TOWHEE/1.jpg  \n",
            "  inflating: valid/ALBERTS TOWHEE/2.jpg  \n",
            "  inflating: valid/ALBERTS TOWHEE/3.jpg  \n",
            "  inflating: valid/ALBERTS TOWHEE/4.jpg  \n",
            "  inflating: valid/ALBERTS TOWHEE/5.jpg  \n",
            "  inflating: valid/ALEXANDRINE PARAKEET/1.jpg  \n",
            "  inflating: valid/ALEXANDRINE PARAKEET/2.jpg  \n",
            "  inflating: valid/ALEXANDRINE PARAKEET/3.jpg  \n",
            "  inflating: valid/ALEXANDRINE PARAKEET/4.jpg  \n",
            "  inflating: valid/ALEXANDRINE PARAKEET/5.jpg  \n",
            "  inflating: valid/ALPINE CHOUGH/1.jpg  \n",
            "  inflating: valid/ALPINE CHOUGH/2.jpg  \n",
            "  inflating: valid/ALPINE CHOUGH/3.jpg  \n",
            "  inflating: valid/ALPINE CHOUGH/4.jpg  \n",
            "  inflating: valid/ALPINE CHOUGH/5.jpg  \n",
            "  inflating: valid/ALTAMIRA YELLOWTHROAT/1.jpg  \n",
            "  inflating: valid/ALTAMIRA YELLOWTHROAT/2.jpg  \n",
            "  inflating: valid/ALTAMIRA YELLOWTHROAT/3.jpg  \n",
            "  inflating: valid/ALTAMIRA YELLOWTHROAT/4.jpg  \n",
            "  inflating: valid/ALTAMIRA YELLOWTHROAT/5.jpg  \n",
            "  inflating: valid/AMERICAN AVOCET/1.jpg  \n",
            "  inflating: valid/AMERICAN AVOCET/2.jpg  \n",
            "  inflating: valid/AMERICAN AVOCET/3.jpg  \n",
            "  inflating: valid/AMERICAN AVOCET/4.jpg  \n",
            "  inflating: valid/AMERICAN AVOCET/5.jpg  \n",
            "  inflating: valid/AMERICAN BITTERN/1.jpg  \n",
            "  inflating: valid/AMERICAN BITTERN/2.jpg  \n",
            "  inflating: valid/AMERICAN BITTERN/3.jpg  \n",
            "  inflating: valid/AMERICAN BITTERN/4.jpg  \n",
            "  inflating: valid/AMERICAN BITTERN/5.jpg  \n",
            "  inflating: valid/AMERICAN COOT/1.jpg  \n",
            "  inflating: valid/AMERICAN COOT/2.jpg  \n",
            "  inflating: valid/AMERICAN COOT/3.jpg  \n",
            "  inflating: valid/AMERICAN COOT/4.jpg  \n",
            "  inflating: valid/AMERICAN COOT/5.jpg  \n",
            "  inflating: valid/AMERICAN FLAMINGO/1.jpg  \n",
            "  inflating: valid/AMERICAN FLAMINGO/2.jpg  \n",
            "  inflating: valid/AMERICAN FLAMINGO/3.jpg  \n",
            "  inflating: valid/AMERICAN FLAMINGO/4.jpg  \n",
            "  inflating: valid/AMERICAN FLAMINGO/5.jpg  \n",
            "  inflating: valid/AMERICAN GOLDFINCH/1.jpg  \n",
            "  inflating: valid/AMERICAN GOLDFINCH/2.jpg  \n",
            "  inflating: valid/AMERICAN GOLDFINCH/3.jpg  \n",
            "  inflating: valid/AMERICAN GOLDFINCH/4.jpg  \n",
            "  inflating: valid/AMERICAN GOLDFINCH/5.jpg  \n",
            "  inflating: valid/AMERICAN KESTREL/1.jpg  \n",
            "  inflating: valid/AMERICAN KESTREL/2.jpg  \n",
            "  inflating: valid/AMERICAN KESTREL/3.jpg  \n",
            "  inflating: valid/AMERICAN KESTREL/4.jpg  \n",
            "  inflating: valid/AMERICAN KESTREL/5.jpg  \n"
          ]
        }
      ],
      "source": [
        "!unzip birds-20-species-image-classification.zip"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "ghaUIx_QjNLK"
      },
      "outputs": [],
      "source": [
        "import os\n",
        "import cv2\n",
        "import time\n",
        "import numpy as np\n",
        "\n",
        "import torch\n",
        "import torch.nn as nn\n",
        "import torch.nn.functional as F\n",
        "\n",
        "from torch.utils.data import DataLoader, Dataset\n",
        "from torch.utils.data import RandomSampler\n",
        "\n",
        "import torchvision.transforms as T\n",
        "import torchvision.models as models\n",
        "from torchvision.utils import make_grid\n",
        "from torchvision.datasets import ImageFolder\n",
        "\n",
        "from matplotlib import pyplot as plt\n",
        "import torchvision.transforms as transforms\n",
        "from torch.utils.data import DataLoader, random_split\n",
        "from torchvision import datasets, transforms, models\n",
        "from torchvision.utils import make_grid"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def get_default_device():\n",
        "    if torch.cuda.is_available():\n",
        "        return torch.device(\"cuda\")\n",
        "    else:\n",
        "        return torch.device(\"cpu\")\n",
        "device = get_default_device()\n",
        "device"
      ],
      "metadata": {
        "id": "gSNNIxZLT-08",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "20af1bf8-5b18-4f00-db69-647b2336af59"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "device(type='cuda')"
            ]
          },
          "metadata": {},
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "XBhUubh5g5th",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "9582ad97-405b-44dc-9d3a-5072ecbb8adb"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['AFRICAN EMERALD CUCKOO',\n",
              " 'AFRICAN PYGMY GOOSE',\n",
              " 'ABBOTTS BABBLER',\n",
              " 'ALBERTS TOWHEE',\n",
              " 'ABYSSINIAN GROUND HORNBILL',\n",
              " 'ABBOTTS BOOBY',\n",
              " 'AMERICAN FLAMINGO',\n",
              " 'AMERICAN KESTREL',\n",
              " 'ALEXANDRINE PARAKEET',\n",
              " 'AFRICAN PIED HORNBILL',\n",
              " 'AMERICAN AVOCET',\n",
              " 'ALTAMIRA YELLOWTHROAT',\n",
              " 'ALBATROSS',\n",
              " 'AFRICAN CROWNED CRANE',\n",
              " 'AFRICAN OYSTER CATCHER',\n",
              " 'AMERICAN GOLDFINCH',\n",
              " 'ALPINE CHOUGH',\n",
              " 'AMERICAN BITTERN',\n",
              " 'AFRICAN FIREFINCH',\n",
              " 'AMERICAN COOT']"
            ]
          },
          "metadata": {},
          "execution_count": 16
        }
      ],
      "source": [
        "classes = os.listdir(\"./train\")\n",
        "classes"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "k34fXq5ljaXO",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "1919aa43-3e0f-413c-b962-2b2ae246cf84"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "20\n"
          ]
        }
      ],
      "source": [
        "print(len(classes))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "AmJjeWb3lsXF",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "18587d3e-d75b-453f-e550-e5b4b95d2bef"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(3208, 100)"
            ]
          },
          "metadata": {},
          "execution_count": 17
        }
      ],
      "source": [
        "# Define the transformations to apply to the data\n",
        "transform = transforms.Compose([\n",
        "    transforms.Resize((227, 227)),  # Resize images to a consistent size\n",
        "    transforms.Compose([\n",
        "    transforms.RandomRotation(10),      # rotate +/- 10 degrees\n",
        "    transforms.RandomHorizontalFlip(),  # reverse 50% of images\n",
        "    transforms.ToTensor(),\n",
        "    transforms.Normalize([0.485, 0.456, 0.406],\n",
        "                          [0.229, 0.224, 0.225])\n",
        "])\n",
        "])\n",
        "\n",
        "trainset = ImageFolder('./train', transform=transform)\n",
        "testset = ImageFolder('./test', transform=transform)\n",
        "len(trainset), len(testset)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "C0jOfBe3fXsa"
      },
      "outputs": [],
      "source": [
        "trainloader = torch.utils.data.DataLoader(trainset, batch_size=32,\n",
        "                                          shuffle=True)\n",
        "testloader = torch.utils.data.DataLoader(testset, batch_size=32,\n",
        "                                         shuffle=False)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "81KzxLornfwS",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "b50ac221-8cdf-4327-88d3-02760b3dddae"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Spatial size of images:  torch.Size([227, 227])\n",
            "Number of channels: 3\n"
          ]
        }
      ],
      "source": [
        "# Get the spatial size of images\n",
        "sample_image, _ = trainset[0]  # Retrieve the first image sample\n",
        "image_size = sample_image.shape[1:]   # Get the size of the image\n",
        "\n",
        "print(\"Spatial size of images: \", image_size)\n",
        "num_channels = sample_image.shape[0]\n",
        "\n",
        "print(\"Number of channels:\", num_channels)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "z5oXZBImdTbW",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 260
        },
        "outputId": "6f8b3f98-fe04-4cc6-91ca-9b13f9c703c9"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Label: [ 4  8 17 17  7 11 12  1  1  6 10 12  0  4  1 18  1  7  5  3  0 14 14 18\n",
            " 16  0 15 11 16 15 17 18]\n",
            "Class: ABBOTTS BOOBY ABYSSINIAN GROUND HORNBILL ALBERTS TOWHEE ALBERTS TOWHEE ALPINE CHOUGH ALBATROSS ALTAMIRA YELLOWTHROAT ABBOTTS BABBLER ABBOTTS BABBLER AFRICAN PIED HORNBILL AMERICAN GOLDFINCH ALTAMIRA YELLOWTHROAT AMERICAN COOT ABBOTTS BOOBY ABBOTTS BABBLER AFRICAN FIREFINCH ABBOTTS BABBLER ALPINE CHOUGH AMERICAN KESTREL ALEXANDRINE PARAKEET AMERICAN COOT AMERICAN FLAMINGO AMERICAN FLAMINGO AFRICAN FIREFINCH AMERICAN BITTERN AMERICAN COOT AFRICAN CROWNED CRANE ALBATROSS AMERICAN BITTERN AFRICAN CROWNED CRANE ALBERTS TOWHEE AFRICAN FIREFINCH\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7f8accd75540>"
            ]
          },
          "metadata": {},
          "execution_count": 11
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1500x1500 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "for images, labels in trainloader:\n",
        "    break\n",
        "#print the labels\n",
        "print('Label:', labels.numpy())\n",
        "print('Class:', *np.array([classes[i] for i in labels]))\n",
        "\n",
        "im=make_grid(images,nrow=16)\n",
        "plt.figure(figsize=(15,15))\n",
        "plt.imshow(np.transpose(im.numpy(),(1,2,0)))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "LueChchWmdtG"
      },
      "outputs": [],
      "source": [
        "class AlexNet(nn.Module):\n",
        "  def __init__(self, num_classes=20):\n",
        "    super(AlexNet, self).__init__()\n",
        "    self.layer1 = nn.Sequential(\n",
        "        nn.Conv2d(3, 96, 11, stride=4),\n",
        "        nn.BatchNorm2d(96),\n",
        "        nn.ReLU(),\n",
        "        # nn.LocalResponseNorm(5, 0.0001, 0.75, 2),\n",
        "        nn.MaxPool2d(3, 2)\n",
        "    )\n",
        "    self.layer2 = nn.Sequential(\n",
        "        nn.Conv2d(96, 256, 5, 1, padding=2),\n",
        "        nn.BatchNorm2d(256),\n",
        "        nn.ReLU(),\n",
        "        # nn.LocalResponseNorm(5, 0.0001, 0.75, 2),\n",
        "        nn.MaxPool2d(3, 2)\n",
        "    )\n",
        "    self.layer3 = nn.Sequential(\n",
        "        nn.Conv2d(256, 384, 3, 1, padding=1),\n",
        "        nn.BatchNorm2d(384),\n",
        "        nn.ReLU()\n",
        "    )\n",
        "    self.layer4 = nn.Sequential(\n",
        "        nn.Conv2d(384, 384, 3, 1, padding=1),\n",
        "        nn.BatchNorm2d(384),\n",
        "        nn.ReLU()\n",
        "    )\n",
        "    self.layer5 = nn.Sequential(\n",
        "        nn.Conv2d(384, 256, 3, 1, padding=1),\n",
        "        nn.BatchNorm2d(256),\n",
        "        nn.ReLU(),\n",
        "        nn.MaxPool2d(3, 2)\n",
        "    )\n",
        "    self.fc1 = nn.Sequential(\n",
        "        nn.Linear(9216, 1024),\n",
        "        nn.ReLU()\n",
        "    )\n",
        "    self.fc2 = nn.Sequential(\n",
        "        nn.Dropout(0.5),\n",
        "        nn.Linear(1024, num_classes),\n",
        "        nn.LogSoftmax(dim=1)\n",
        "    )\n",
        "\n",
        "  def forward(self, x):\n",
        "        out = self.layer1(x)\n",
        "        out = self.layer2(out)\n",
        "        out = self.layer3(out)\n",
        "        out = self.layer4(out)\n",
        "        out = self.layer5(out)\n",
        "        out = out.view(-1, 256*6*6)\n",
        "        out = self.fc1(out)\n",
        "        out = self.fc2(out)\n",
        "        return out\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "scPR1sS0vA8n",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "65609623-3645-413a-e2de-ab2f0e07f46a"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "AlexNet(\n",
              "  (layer1): Sequential(\n",
              "    (0): Conv2d(3, 96, kernel_size=(11, 11), stride=(4, 4))\n",
              "    (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "    (2): ReLU()\n",
              "    (3): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
              "  )\n",
              "  (layer2): Sequential(\n",
              "    (0): Conv2d(96, 256, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))\n",
              "    (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "    (2): ReLU()\n",
              "    (3): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
              "  )\n",
              "  (layer3): Sequential(\n",
              "    (0): Conv2d(256, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
              "    (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "    (2): ReLU()\n",
              "  )\n",
              "  (layer4): Sequential(\n",
              "    (0): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
              "    (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "    (2): ReLU()\n",
              "  )\n",
              "  (layer5): Sequential(\n",
              "    (0): Conv2d(384, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
              "    (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "    (2): ReLU()\n",
              "    (3): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
              "  )\n",
              "  (fc1): Sequential(\n",
              "    (0): Linear(in_features=9216, out_features=1024, bias=True)\n",
              "    (1): ReLU()\n",
              "  )\n",
              "  (fc2): Sequential(\n",
              "    (0): Dropout(p=0.5, inplace=False)\n",
              "    (1): Linear(in_features=1024, out_features=20, bias=True)\n",
              "    (2): LogSoftmax(dim=1)\n",
              "  )\n",
              ")"
            ]
          },
          "metadata": {},
          "execution_count": 58
        }
      ],
      "source": [
        "alex_net = AlexNet()\n",
        "alex_net = alex_net.to(device)\n",
        "alex_net.cuda()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "3tv1qkRsvMLw"
      },
      "outputs": [],
      "source": [
        "import torch.optim as optim\n",
        "criterion = nn.CrossEntropyLoss()\n",
        "optimizer = optim.SGD(alex_net.parameters(), lr=0.001)"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def train(model):\n",
        "  num_epochs = 10\n",
        "  total_step = len(trainloader)\n",
        "  # Training loop\n",
        "  for epoch in range(num_epochs):\n",
        "      for i, data in enumerate(trainloader, 0):\n",
        "          inputs, labels = data\n",
        "          inputs = inputs.to(device)\n",
        "          labels = labels.to(device)\n",
        "          optimizer.zero_grad()\n",
        "\n",
        "          outputs = model(inputs)\n",
        "          loss = criterion(outputs, labels)\n",
        "          loss.backward()\n",
        "          optimizer.step()\n",
        "\n",
        "          print ('Epoch [{}/{}], Step [{}/{}], Loss: {:.4f}'\n",
        "                    .format(epoch+1, num_epochs, i+1, total_step, loss.item()))\n",
        "      print('Epoch finished')\n",
        "\n",
        "  print('Training finished!')"
      ],
      "metadata": {
        "id": "aknPuKu5sONn"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "gM9NjltOvRiQ",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "e4042d3a-0380-4777-e9ca-9b4b7cf31c4b"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch [1/10], Step [1/101], Loss: 3.1737\n",
            "Epoch [1/10], Step [2/101], Loss: 3.1706\n",
            "Epoch [1/10], Step [3/101], Loss: 3.0118\n",
            "Epoch [1/10], Step [4/101], Loss: 3.0160\n",
            "Epoch [1/10], Step [5/101], Loss: 3.0336\n",
            "Epoch [1/10], Step [6/101], Loss: 3.0739\n",
            "Epoch [1/10], Step [7/101], Loss: 2.9973\n",
            "Epoch [1/10], Step [8/101], Loss: 3.0809\n",
            "Epoch [1/10], Step [9/101], Loss: 2.9053\n",
            "Epoch [1/10], Step [10/101], Loss: 2.9633\n",
            "Epoch [1/10], Step [11/101], Loss: 3.0770\n",
            "Epoch [1/10], Step [12/101], Loss: 2.8230\n",
            "Epoch [1/10], Step [13/101], Loss: 3.1395\n",
            "Epoch [1/10], Step [14/101], Loss: 3.0473\n",
            "Epoch [1/10], Step [15/101], Loss: 3.0394\n",
            "Epoch [1/10], Step [16/101], Loss: 3.0282\n",
            "Epoch [1/10], Step [17/101], Loss: 2.7763\n",
            "Epoch [1/10], Step [18/101], Loss: 2.9451\n",
            "Epoch [1/10], Step [19/101], Loss: 2.9270\n",
            "Epoch [1/10], Step [20/101], Loss: 3.0344\n",
            "Epoch [1/10], Step [21/101], Loss: 2.8728\n",
            "Epoch [1/10], Step [22/101], Loss: 2.9252\n",
            "Epoch [1/10], Step [23/101], Loss: 2.9757\n",
            "Epoch [1/10], Step [24/101], Loss: 2.8782\n",
            "Epoch [1/10], Step [25/101], Loss: 2.8696\n",
            "Epoch [1/10], Step [26/101], Loss: 2.8958\n",
            "Epoch [1/10], Step [27/101], Loss: 2.9997\n",
            "Epoch [1/10], Step [28/101], Loss: 2.8598\n",
            "Epoch [1/10], Step [29/101], Loss: 2.9854\n",
            "Epoch [1/10], Step [30/101], Loss: 2.9882\n",
            "Epoch [1/10], Step [31/101], Loss: 2.8826\n",
            "Epoch [1/10], Step [32/101], Loss: 2.8698\n",
            "Epoch [1/10], Step [33/101], Loss: 2.9256\n",
            "Epoch [1/10], Step [34/101], Loss: 2.8637\n",
            "Epoch [1/10], Step [35/101], Loss: 2.9275\n",
            "Epoch [1/10], Step [36/101], Loss: 2.8031\n",
            "Epoch [1/10], Step [37/101], Loss: 2.8677\n",
            "Epoch [1/10], Step [38/101], Loss: 2.8926\n",
            "Epoch [1/10], Step [39/101], Loss: 2.7889\n",
            "Epoch [1/10], Step [40/101], Loss: 2.8666\n",
            "Epoch [1/10], Step [41/101], Loss: 2.8065\n",
            "Epoch [1/10], Step [42/101], Loss: 2.9073\n",
            "Epoch [1/10], Step [43/101], Loss: 2.7470\n",
            "Epoch [1/10], Step [44/101], Loss: 2.8735\n",
            "Epoch [1/10], Step [45/101], Loss: 2.6468\n",
            "Epoch [1/10], Step [46/101], Loss: 2.7956\n",
            "Epoch [1/10], Step [47/101], Loss: 2.6720\n",
            "Epoch [1/10], Step [48/101], Loss: 2.8511\n",
            "Epoch [1/10], Step [49/101], Loss: 2.7505\n",
            "Epoch [1/10], Step [50/101], Loss: 2.8545\n",
            "Epoch [1/10], Step [51/101], Loss: 2.8161\n",
            "Epoch [1/10], Step [52/101], Loss: 2.6636\n",
            "Epoch [1/10], Step [53/101], Loss: 2.6357\n",
            "Epoch [1/10], Step [54/101], Loss: 2.7733\n",
            "Epoch [1/10], Step [55/101], Loss: 2.7033\n",
            "Epoch [1/10], Step [56/101], Loss: 2.7411\n",
            "Epoch [1/10], Step [57/101], Loss: 2.7932\n",
            "Epoch [1/10], Step [58/101], Loss: 2.7212\n",
            "Epoch [1/10], Step [59/101], Loss: 2.7385\n",
            "Epoch [1/10], Step [60/101], Loss: 2.6109\n",
            "Epoch [1/10], Step [61/101], Loss: 2.7160\n",
            "Epoch [1/10], Step [62/101], Loss: 2.6523\n",
            "Epoch [1/10], Step [63/101], Loss: 2.7149\n",
            "Epoch [1/10], Step [64/101], Loss: 2.5770\n",
            "Epoch [1/10], Step [65/101], Loss: 2.7504\n",
            "Epoch [1/10], Step [66/101], Loss: 2.7423\n",
            "Epoch [1/10], Step [67/101], Loss: 2.6689\n",
            "Epoch [1/10], Step [68/101], Loss: 2.5774\n",
            "Epoch [1/10], Step [69/101], Loss: 2.6050\n",
            "Epoch [1/10], Step [70/101], Loss: 2.6539\n",
            "Epoch [1/10], Step [71/101], Loss: 2.6080\n",
            "Epoch [1/10], Step [72/101], Loss: 2.6232\n",
            "Epoch [1/10], Step [73/101], Loss: 2.6930\n",
            "Epoch [1/10], Step [74/101], Loss: 2.7810\n",
            "Epoch [1/10], Step [75/101], Loss: 2.6458\n",
            "Epoch [1/10], Step [76/101], Loss: 2.6843\n",
            "Epoch [1/10], Step [77/101], Loss: 2.5791\n",
            "Epoch [1/10], Step [78/101], Loss: 2.6145\n",
            "Epoch [1/10], Step [79/101], Loss: 2.5653\n",
            "Epoch [1/10], Step [80/101], Loss: 2.6384\n",
            "Epoch [1/10], Step [81/101], Loss: 2.6017\n",
            "Epoch [1/10], Step [82/101], Loss: 2.6159\n",
            "Epoch [1/10], Step [83/101], Loss: 2.6443\n",
            "Epoch [1/10], Step [84/101], Loss: 2.6881\n",
            "Epoch [1/10], Step [85/101], Loss: 2.5280\n",
            "Epoch [1/10], Step [86/101], Loss: 2.6299\n",
            "Epoch [1/10], Step [87/101], Loss: 2.5879\n",
            "Epoch [1/10], Step [88/101], Loss: 2.3372\n",
            "Epoch [1/10], Step [89/101], Loss: 2.5112\n",
            "Epoch [1/10], Step [90/101], Loss: 2.4223\n",
            "Epoch [1/10], Step [91/101], Loss: 2.6865\n",
            "Epoch [1/10], Step [92/101], Loss: 2.5578\n",
            "Epoch [1/10], Step [93/101], Loss: 2.5922\n",
            "Epoch [1/10], Step [94/101], Loss: 2.5210\n",
            "Epoch [1/10], Step [95/101], Loss: 2.6323\n",
            "Epoch [1/10], Step [96/101], Loss: 2.3188\n",
            "Epoch [1/10], Step [97/101], Loss: 2.4209\n",
            "Epoch [1/10], Step [98/101], Loss: 2.4703\n",
            "Epoch [1/10], Step [99/101], Loss: 2.6939\n",
            "Epoch [1/10], Step [100/101], Loss: 2.6468\n",
            "Epoch [1/10], Step [101/101], Loss: 2.6244\n",
            "Epoch finished\n",
            "Epoch [2/10], Step [1/101], Loss: 2.5316\n",
            "Epoch [2/10], Step [2/101], Loss: 2.4978\n",
            "Epoch [2/10], Step [3/101], Loss: 2.4545\n",
            "Epoch [2/10], Step [4/101], Loss: 2.5486\n",
            "Epoch [2/10], Step [5/101], Loss: 2.2599\n",
            "Epoch [2/10], Step [6/101], Loss: 2.3252\n",
            "Epoch [2/10], Step [7/101], Loss: 2.5379\n",
            "Epoch [2/10], Step [8/101], Loss: 2.5900\n",
            "Epoch [2/10], Step [9/101], Loss: 2.4700\n",
            "Epoch [2/10], Step [10/101], Loss: 2.3361\n",
            "Epoch [2/10], Step [11/101], Loss: 2.2041\n",
            "Epoch [2/10], Step [12/101], Loss: 2.4507\n",
            "Epoch [2/10], Step [13/101], Loss: 2.4369\n",
            "Epoch [2/10], Step [14/101], Loss: 2.6199\n",
            "Epoch [2/10], Step [15/101], Loss: 2.5374\n",
            "Epoch [2/10], Step [16/101], Loss: 2.3509\n",
            "Epoch [2/10], Step [17/101], Loss: 2.3523\n",
            "Epoch [2/10], Step [18/101], Loss: 2.3637\n",
            "Epoch [2/10], Step [19/101], Loss: 2.2317\n",
            "Epoch [2/10], Step [20/101], Loss: 2.5358\n",
            "Epoch [2/10], Step [21/101], Loss: 2.3244\n",
            "Epoch [2/10], Step [22/101], Loss: 2.4253\n",
            "Epoch [2/10], Step [23/101], Loss: 2.4116\n",
            "Epoch [2/10], Step [24/101], Loss: 2.4987\n",
            "Epoch [2/10], Step [25/101], Loss: 2.3630\n",
            "Epoch [2/10], Step [26/101], Loss: 2.4070\n",
            "Epoch [2/10], Step [27/101], Loss: 2.2378\n",
            "Epoch [2/10], Step [28/101], Loss: 2.3119\n",
            "Epoch [2/10], Step [29/101], Loss: 2.4951\n",
            "Epoch [2/10], Step [30/101], Loss: 2.2072\n",
            "Epoch [2/10], Step [31/101], Loss: 2.2893\n",
            "Epoch [2/10], Step [32/101], Loss: 2.2310\n",
            "Epoch [2/10], Step [33/101], Loss: 2.3276\n",
            "Epoch [2/10], Step [34/101], Loss: 2.3193\n",
            "Epoch [2/10], Step [35/101], Loss: 2.2154\n",
            "Epoch [2/10], Step [36/101], Loss: 2.2368\n",
            "Epoch [2/10], Step [37/101], Loss: 2.4363\n",
            "Epoch [2/10], Step [38/101], Loss: 2.1605\n",
            "Epoch [2/10], Step [39/101], Loss: 2.1699\n",
            "Epoch [2/10], Step [40/101], Loss: 2.2066\n",
            "Epoch [2/10], Step [41/101], Loss: 2.3485\n",
            "Epoch [2/10], Step [42/101], Loss: 2.2865\n",
            "Epoch [2/10], Step [43/101], Loss: 2.4598\n",
            "Epoch [2/10], Step [44/101], Loss: 2.2866\n",
            "Epoch [2/10], Step [45/101], Loss: 2.3037\n",
            "Epoch [2/10], Step [46/101], Loss: 2.4542\n",
            "Epoch [2/10], Step [47/101], Loss: 2.2017\n",
            "Epoch [2/10], Step [48/101], Loss: 2.4222\n",
            "Epoch [2/10], Step [49/101], Loss: 2.3459\n",
            "Epoch [2/10], Step [50/101], Loss: 2.2651\n",
            "Epoch [2/10], Step [51/101], Loss: 2.1940\n",
            "Epoch [2/10], Step [52/101], Loss: 2.3507\n",
            "Epoch [2/10], Step [53/101], Loss: 2.1245\n",
            "Epoch [2/10], Step [54/101], Loss: 2.1302\n",
            "Epoch [2/10], Step [55/101], Loss: 2.3415\n",
            "Epoch [2/10], Step [56/101], Loss: 2.2522\n",
            "Epoch [2/10], Step [57/101], Loss: 2.0566\n",
            "Epoch [2/10], Step [58/101], Loss: 2.2702\n",
            "Epoch [2/10], Step [59/101], Loss: 2.2555\n",
            "Epoch [2/10], Step [60/101], Loss: 2.2239\n",
            "Epoch [2/10], Step [61/101], Loss: 2.2438\n",
            "Epoch [2/10], Step [62/101], Loss: 2.1384\n",
            "Epoch [2/10], Step [63/101], Loss: 2.1622\n",
            "Epoch [2/10], Step [64/101], Loss: 2.2458\n",
            "Epoch [2/10], Step [65/101], Loss: 2.2302\n",
            "Epoch [2/10], Step [66/101], Loss: 2.1353\n",
            "Epoch [2/10], Step [67/101], Loss: 2.3623\n",
            "Epoch [2/10], Step [68/101], Loss: 2.2369\n",
            "Epoch [2/10], Step [69/101], Loss: 2.2459\n",
            "Epoch [2/10], Step [70/101], Loss: 2.2282\n",
            "Epoch [2/10], Step [71/101], Loss: 2.3027\n",
            "Epoch [2/10], Step [72/101], Loss: 2.0952\n",
            "Epoch [2/10], Step [73/101], Loss: 2.3472\n",
            "Epoch [2/10], Step [74/101], Loss: 2.3512\n",
            "Epoch [2/10], Step [75/101], Loss: 2.3876\n",
            "Epoch [2/10], Step [76/101], Loss: 2.3218\n",
            "Epoch [2/10], Step [77/101], Loss: 2.1556\n",
            "Epoch [2/10], Step [78/101], Loss: 2.2714\n",
            "Epoch [2/10], Step [79/101], Loss: 2.4466\n",
            "Epoch [2/10], Step [80/101], Loss: 1.8917\n",
            "Epoch [2/10], Step [81/101], Loss: 2.1808\n",
            "Epoch [2/10], Step [82/101], Loss: 2.1985\n",
            "Epoch [2/10], Step [83/101], Loss: 2.2863\n",
            "Epoch [2/10], Step [84/101], Loss: 2.1309\n",
            "Epoch [2/10], Step [85/101], Loss: 2.2430\n",
            "Epoch [2/10], Step [86/101], Loss: 1.9340\n",
            "Epoch [2/10], Step [87/101], Loss: 2.1057\n",
            "Epoch [2/10], Step [88/101], Loss: 2.1061\n",
            "Epoch [2/10], Step [89/101], Loss: 2.1987\n",
            "Epoch [2/10], Step [90/101], Loss: 2.0051\n",
            "Epoch [2/10], Step [91/101], Loss: 2.1843\n",
            "Epoch [2/10], Step [92/101], Loss: 2.2901\n",
            "Epoch [2/10], Step [93/101], Loss: 2.1459\n",
            "Epoch [2/10], Step [94/101], Loss: 2.1783\n",
            "Epoch [2/10], Step [95/101], Loss: 2.0680\n",
            "Epoch [2/10], Step [96/101], Loss: 2.3062\n",
            "Epoch [2/10], Step [97/101], Loss: 2.0842\n",
            "Epoch [2/10], Step [98/101], Loss: 1.8665\n",
            "Epoch [2/10], Step [99/101], Loss: 2.1225\n",
            "Epoch [2/10], Step [100/101], Loss: 2.2364\n",
            "Epoch [2/10], Step [101/101], Loss: 2.0435\n",
            "Epoch finished\n",
            "Epoch [3/10], Step [1/101], Loss: 2.1550\n",
            "Epoch [3/10], Step [2/101], Loss: 1.8521\n",
            "Epoch [3/10], Step [3/101], Loss: 2.0811\n",
            "Epoch [3/10], Step [4/101], Loss: 2.0795\n",
            "Epoch [3/10], Step [5/101], Loss: 2.2519\n",
            "Epoch [3/10], Step [6/101], Loss: 2.0952\n",
            "Epoch [3/10], Step [7/101], Loss: 1.9922\n",
            "Epoch [3/10], Step [8/101], Loss: 1.9048\n",
            "Epoch [3/10], Step [9/101], Loss: 2.0304\n",
            "Epoch [3/10], Step [10/101], Loss: 1.9207\n",
            "Epoch [3/10], Step [11/101], Loss: 2.1744\n",
            "Epoch [3/10], Step [12/101], Loss: 1.9125\n",
            "Epoch [3/10], Step [13/101], Loss: 1.9258\n",
            "Epoch [3/10], Step [14/101], Loss: 2.1554\n",
            "Epoch [3/10], Step [15/101], Loss: 2.3172\n",
            "Epoch [3/10], Step [16/101], Loss: 2.3172\n",
            "Epoch [3/10], Step [17/101], Loss: 2.0796\n",
            "Epoch [3/10], Step [18/101], Loss: 1.8355\n",
            "Epoch [3/10], Step [19/101], Loss: 2.1128\n",
            "Epoch [3/10], Step [20/101], Loss: 1.9774\n",
            "Epoch [3/10], Step [21/101], Loss: 1.9095\n",
            "Epoch [3/10], Step [22/101], Loss: 1.9106\n",
            "Epoch [3/10], Step [23/101], Loss: 1.9427\n",
            "Epoch [3/10], Step [24/101], Loss: 1.9518\n",
            "Epoch [3/10], Step [25/101], Loss: 1.8010\n",
            "Epoch [3/10], Step [26/101], Loss: 2.2170\n",
            "Epoch [3/10], Step [27/101], Loss: 2.0951\n",
            "Epoch [3/10], Step [28/101], Loss: 2.1788\n",
            "Epoch [3/10], Step [29/101], Loss: 2.0340\n",
            "Epoch [3/10], Step [30/101], Loss: 1.7683\n",
            "Epoch [3/10], Step [31/101], Loss: 1.9000\n",
            "Epoch [3/10], Step [32/101], Loss: 1.9997\n",
            "Epoch [3/10], Step [33/101], Loss: 2.2127\n",
            "Epoch [3/10], Step [34/101], Loss: 2.1527\n",
            "Epoch [3/10], Step [35/101], Loss: 1.8651\n",
            "Epoch [3/10], Step [36/101], Loss: 2.0762\n",
            "Epoch [3/10], Step [37/101], Loss: 2.2094\n",
            "Epoch [3/10], Step [38/101], Loss: 2.0645\n",
            "Epoch [3/10], Step [39/101], Loss: 1.9117\n",
            "Epoch [3/10], Step [40/101], Loss: 2.0836\n",
            "Epoch [3/10], Step [41/101], Loss: 1.9797\n",
            "Epoch [3/10], Step [42/101], Loss: 2.0545\n",
            "Epoch [3/10], Step [43/101], Loss: 1.8081\n",
            "Epoch [3/10], Step [44/101], Loss: 1.8695\n",
            "Epoch [3/10], Step [45/101], Loss: 2.0910\n",
            "Epoch [3/10], Step [46/101], Loss: 2.0985\n",
            "Epoch [3/10], Step [47/101], Loss: 1.9857\n",
            "Epoch [3/10], Step [48/101], Loss: 2.1221\n",
            "Epoch [3/10], Step [49/101], Loss: 1.8026\n",
            "Epoch [3/10], Step [50/101], Loss: 1.9935\n",
            "Epoch [3/10], Step [51/101], Loss: 1.9159\n",
            "Epoch [3/10], Step [52/101], Loss: 1.9631\n",
            "Epoch [3/10], Step [53/101], Loss: 2.1772\n",
            "Epoch [3/10], Step [54/101], Loss: 1.8854\n",
            "Epoch [3/10], Step [55/101], Loss: 1.7789\n",
            "Epoch [3/10], Step [56/101], Loss: 2.0367\n",
            "Epoch [3/10], Step [57/101], Loss: 1.9403\n",
            "Epoch [3/10], Step [58/101], Loss: 1.7686\n",
            "Epoch [3/10], Step [59/101], Loss: 1.9329\n",
            "Epoch [3/10], Step [60/101], Loss: 1.8718\n",
            "Epoch [3/10], Step [61/101], Loss: 2.0289\n",
            "Epoch [3/10], Step [62/101], Loss: 1.7181\n",
            "Epoch [3/10], Step [63/101], Loss: 1.8542\n",
            "Epoch [3/10], Step [64/101], Loss: 1.9179\n",
            "Epoch [3/10], Step [65/101], Loss: 1.8737\n",
            "Epoch [3/10], Step [66/101], Loss: 1.8678\n",
            "Epoch [3/10], Step [67/101], Loss: 1.9788\n",
            "Epoch [3/10], Step [68/101], Loss: 1.7919\n",
            "Epoch [3/10], Step [69/101], Loss: 1.8092\n",
            "Epoch [3/10], Step [70/101], Loss: 1.7908\n",
            "Epoch [3/10], Step [71/101], Loss: 1.7909\n",
            "Epoch [3/10], Step [72/101], Loss: 1.8176\n",
            "Epoch [3/10], Step [73/101], Loss: 2.0523\n",
            "Epoch [3/10], Step [74/101], Loss: 1.6824\n",
            "Epoch [3/10], Step [75/101], Loss: 1.9872\n",
            "Epoch [3/10], Step [76/101], Loss: 2.1188\n",
            "Epoch [3/10], Step [77/101], Loss: 2.1364\n",
            "Epoch [3/10], Step [78/101], Loss: 1.7093\n",
            "Epoch [3/10], Step [79/101], Loss: 1.7545\n",
            "Epoch [3/10], Step [80/101], Loss: 1.8331\n",
            "Epoch [3/10], Step [81/101], Loss: 1.9991\n",
            "Epoch [3/10], Step [82/101], Loss: 1.7976\n",
            "Epoch [3/10], Step [83/101], Loss: 2.0296\n",
            "Epoch [3/10], Step [84/101], Loss: 1.8404\n",
            "Epoch [3/10], Step [85/101], Loss: 1.9331\n",
            "Epoch [3/10], Step [86/101], Loss: 1.5463\n",
            "Epoch [3/10], Step [87/101], Loss: 1.7442\n",
            "Epoch [3/10], Step [88/101], Loss: 1.8131\n",
            "Epoch [3/10], Step [89/101], Loss: 1.7057\n",
            "Epoch [3/10], Step [90/101], Loss: 1.7153\n",
            "Epoch [3/10], Step [91/101], Loss: 1.8199\n",
            "Epoch [3/10], Step [92/101], Loss: 1.7720\n",
            "Epoch [3/10], Step [93/101], Loss: 2.2348\n",
            "Epoch [3/10], Step [94/101], Loss: 1.8030\n",
            "Epoch [3/10], Step [95/101], Loss: 1.8235\n",
            "Epoch [3/10], Step [96/101], Loss: 1.7518\n",
            "Epoch [3/10], Step [97/101], Loss: 1.9545\n",
            "Epoch [3/10], Step [98/101], Loss: 1.9088\n",
            "Epoch [3/10], Step [99/101], Loss: 1.6804\n",
            "Epoch [3/10], Step [100/101], Loss: 2.0063\n",
            "Epoch [3/10], Step [101/101], Loss: 1.9200\n",
            "Epoch finished\n",
            "Epoch [4/10], Step [1/101], Loss: 1.7836\n",
            "Epoch [4/10], Step [2/101], Loss: 1.8903\n",
            "Epoch [4/10], Step [3/101], Loss: 1.6133\n",
            "Epoch [4/10], Step [4/101], Loss: 1.5398\n",
            "Epoch [4/10], Step [5/101], Loss: 1.8636\n",
            "Epoch [4/10], Step [6/101], Loss: 1.6064\n",
            "Epoch [4/10], Step [7/101], Loss: 1.8958\n",
            "Epoch [4/10], Step [8/101], Loss: 2.1430\n",
            "Epoch [4/10], Step [9/101], Loss: 1.9242\n",
            "Epoch [4/10], Step [10/101], Loss: 1.8477\n",
            "Epoch [4/10], Step [11/101], Loss: 1.7873\n",
            "Epoch [4/10], Step [12/101], Loss: 1.9202\n",
            "Epoch [4/10], Step [13/101], Loss: 1.8412\n",
            "Epoch [4/10], Step [14/101], Loss: 2.1070\n",
            "Epoch [4/10], Step [15/101], Loss: 1.7302\n",
            "Epoch [4/10], Step [16/101], Loss: 1.7334\n",
            "Epoch [4/10], Step [17/101], Loss: 1.8052\n",
            "Epoch [4/10], Step [18/101], Loss: 1.6795\n",
            "Epoch [4/10], Step [19/101], Loss: 1.6588\n",
            "Epoch [4/10], Step [20/101], Loss: 1.7605\n",
            "Epoch [4/10], Step [21/101], Loss: 1.6439\n",
            "Epoch [4/10], Step [22/101], Loss: 1.7865\n",
            "Epoch [4/10], Step [23/101], Loss: 1.9123\n",
            "Epoch [4/10], Step [24/101], Loss: 1.7859\n",
            "Epoch [4/10], Step [25/101], Loss: 1.8639\n",
            "Epoch [4/10], Step [26/101], Loss: 1.5348\n",
            "Epoch [4/10], Step [27/101], Loss: 1.8699\n",
            "Epoch [4/10], Step [28/101], Loss: 1.7663\n",
            "Epoch [4/10], Step [29/101], Loss: 1.6086\n",
            "Epoch [4/10], Step [30/101], Loss: 1.6146\n",
            "Epoch [4/10], Step [31/101], Loss: 1.8972\n",
            "Epoch [4/10], Step [32/101], Loss: 1.7035\n",
            "Epoch [4/10], Step [33/101], Loss: 1.7007\n",
            "Epoch [4/10], Step [34/101], Loss: 1.7447\n",
            "Epoch [4/10], Step [35/101], Loss: 1.5371\n",
            "Epoch [4/10], Step [36/101], Loss: 1.5139\n",
            "Epoch [4/10], Step [37/101], Loss: 1.2644\n",
            "Epoch [4/10], Step [38/101], Loss: 1.7380\n",
            "Epoch [4/10], Step [39/101], Loss: 1.5896\n",
            "Epoch [4/10], Step [40/101], Loss: 1.8327\n",
            "Epoch [4/10], Step [41/101], Loss: 1.7655\n",
            "Epoch [4/10], Step [42/101], Loss: 1.7000\n",
            "Epoch [4/10], Step [43/101], Loss: 1.6423\n",
            "Epoch [4/10], Step [44/101], Loss: 1.7002\n",
            "Epoch [4/10], Step [45/101], Loss: 1.6450\n",
            "Epoch [4/10], Step [46/101], Loss: 1.4765\n",
            "Epoch [4/10], Step [47/101], Loss: 1.8653\n",
            "Epoch [4/10], Step [48/101], Loss: 1.8290\n",
            "Epoch [4/10], Step [49/101], Loss: 1.9999\n",
            "Epoch [4/10], Step [50/101], Loss: 1.8026\n",
            "Epoch [4/10], Step [51/101], Loss: 1.6299\n",
            "Epoch [4/10], Step [52/101], Loss: 1.8546\n",
            "Epoch [4/10], Step [53/101], Loss: 1.9051\n",
            "Epoch [4/10], Step [54/101], Loss: 1.8169\n",
            "Epoch [4/10], Step [55/101], Loss: 1.5989\n",
            "Epoch [4/10], Step [56/101], Loss: 1.6299\n",
            "Epoch [4/10], Step [57/101], Loss: 1.6763\n",
            "Epoch [4/10], Step [58/101], Loss: 1.6050\n",
            "Epoch [4/10], Step [59/101], Loss: 1.5715\n",
            "Epoch [4/10], Step [60/101], Loss: 1.5157\n",
            "Epoch [4/10], Step [61/101], Loss: 1.7466\n",
            "Epoch [4/10], Step [62/101], Loss: 1.8092\n",
            "Epoch [4/10], Step [63/101], Loss: 1.7707\n",
            "Epoch [4/10], Step [64/101], Loss: 1.3918\n",
            "Epoch [4/10], Step [65/101], Loss: 1.7101\n",
            "Epoch [4/10], Step [66/101], Loss: 1.7613\n",
            "Epoch [4/10], Step [67/101], Loss: 1.7317\n",
            "Epoch [4/10], Step [68/101], Loss: 1.5423\n",
            "Epoch [4/10], Step [69/101], Loss: 1.9997\n",
            "Epoch [4/10], Step [70/101], Loss: 1.6870\n",
            "Epoch [4/10], Step [71/101], Loss: 1.7581\n",
            "Epoch [4/10], Step [72/101], Loss: 1.5356\n",
            "Epoch [4/10], Step [73/101], Loss: 1.3794\n",
            "Epoch [4/10], Step [74/101], Loss: 1.5825\n",
            "Epoch [4/10], Step [75/101], Loss: 1.6565\n",
            "Epoch [4/10], Step [76/101], Loss: 1.4503\n",
            "Epoch [4/10], Step [77/101], Loss: 1.7678\n",
            "Epoch [4/10], Step [78/101], Loss: 1.5708\n",
            "Epoch [4/10], Step [79/101], Loss: 1.5864\n",
            "Epoch [4/10], Step [80/101], Loss: 1.8262\n",
            "Epoch [4/10], Step [81/101], Loss: 1.7852\n",
            "Epoch [4/10], Step [82/101], Loss: 1.4740\n",
            "Epoch [4/10], Step [83/101], Loss: 1.5925\n",
            "Epoch [4/10], Step [84/101], Loss: 1.7108\n",
            "Epoch [4/10], Step [85/101], Loss: 1.7622\n",
            "Epoch [4/10], Step [86/101], Loss: 1.6795\n",
            "Epoch [4/10], Step [87/101], Loss: 1.9011\n",
            "Epoch [4/10], Step [88/101], Loss: 1.4705\n",
            "Epoch [4/10], Step [89/101], Loss: 1.5005\n",
            "Epoch [4/10], Step [90/101], Loss: 1.7975\n",
            "Epoch [4/10], Step [91/101], Loss: 1.5572\n",
            "Epoch [4/10], Step [92/101], Loss: 1.8959\n",
            "Epoch [4/10], Step [93/101], Loss: 1.5156\n",
            "Epoch [4/10], Step [94/101], Loss: 1.3789\n",
            "Epoch [4/10], Step [95/101], Loss: 1.7246\n",
            "Epoch [4/10], Step [96/101], Loss: 1.5652\n",
            "Epoch [4/10], Step [97/101], Loss: 1.4351\n",
            "Epoch [4/10], Step [98/101], Loss: 1.7670\n",
            "Epoch [4/10], Step [99/101], Loss: 1.5212\n",
            "Epoch [4/10], Step [100/101], Loss: 1.7478\n",
            "Epoch [4/10], Step [101/101], Loss: 1.5788\n",
            "Epoch finished\n",
            "Epoch [5/10], Step [1/101], Loss: 1.7184\n",
            "Epoch [5/10], Step [2/101], Loss: 2.1034\n",
            "Epoch [5/10], Step [3/101], Loss: 1.7186\n",
            "Epoch [5/10], Step [4/101], Loss: 1.8464\n",
            "Epoch [5/10], Step [5/101], Loss: 1.6263\n",
            "Epoch [5/10], Step [6/101], Loss: 1.6375\n",
            "Epoch [5/10], Step [7/101], Loss: 1.7678\n",
            "Epoch [5/10], Step [8/101], Loss: 1.7759\n",
            "Epoch [5/10], Step [9/101], Loss: 1.8166\n",
            "Epoch [5/10], Step [10/101], Loss: 1.5661\n",
            "Epoch [5/10], Step [11/101], Loss: 1.6515\n",
            "Epoch [5/10], Step [12/101], Loss: 1.6490\n",
            "Epoch [5/10], Step [13/101], Loss: 1.4542\n",
            "Epoch [5/10], Step [14/101], Loss: 1.7163\n",
            "Epoch [5/10], Step [15/101], Loss: 1.4847\n",
            "Epoch [5/10], Step [16/101], Loss: 1.6007\n",
            "Epoch [5/10], Step [17/101], Loss: 1.5873\n",
            "Epoch [5/10], Step [18/101], Loss: 1.6497\n",
            "Epoch [5/10], Step [19/101], Loss: 1.7120\n",
            "Epoch [5/10], Step [20/101], Loss: 1.5340\n",
            "Epoch [5/10], Step [21/101], Loss: 1.2851\n",
            "Epoch [5/10], Step [22/101], Loss: 1.5835\n",
            "Epoch [5/10], Step [23/101], Loss: 1.6786\n",
            "Epoch [5/10], Step [24/101], Loss: 1.4059\n",
            "Epoch [5/10], Step [25/101], Loss: 1.5615\n",
            "Epoch [5/10], Step [26/101], Loss: 1.4126\n",
            "Epoch [5/10], Step [27/101], Loss: 1.2811\n",
            "Epoch [5/10], Step [28/101], Loss: 1.3364\n",
            "Epoch [5/10], Step [29/101], Loss: 1.2716\n",
            "Epoch [5/10], Step [30/101], Loss: 1.5521\n",
            "Epoch [5/10], Step [31/101], Loss: 1.4578\n",
            "Epoch [5/10], Step [32/101], Loss: 1.5286\n",
            "Epoch [5/10], Step [33/101], Loss: 1.5717\n",
            "Epoch [5/10], Step [34/101], Loss: 1.7345\n",
            "Epoch [5/10], Step [35/101], Loss: 1.4938\n",
            "Epoch [5/10], Step [36/101], Loss: 1.5568\n",
            "Epoch [5/10], Step [37/101], Loss: 1.4279\n",
            "Epoch [5/10], Step [38/101], Loss: 1.5847\n",
            "Epoch [5/10], Step [39/101], Loss: 1.4152\n",
            "Epoch [5/10], Step [40/101], Loss: 1.8587\n",
            "Epoch [5/10], Step [41/101], Loss: 1.4028\n",
            "Epoch [5/10], Step [42/101], Loss: 1.4990\n",
            "Epoch [5/10], Step [43/101], Loss: 1.3647\n",
            "Epoch [5/10], Step [44/101], Loss: 1.7749\n",
            "Epoch [5/10], Step [45/101], Loss: 1.5187\n",
            "Epoch [5/10], Step [46/101], Loss: 1.3731\n",
            "Epoch [5/10], Step [47/101], Loss: 1.6625\n",
            "Epoch [5/10], Step [48/101], Loss: 1.5651\n",
            "Epoch [5/10], Step [49/101], Loss: 1.4962\n",
            "Epoch [5/10], Step [50/101], Loss: 1.6422\n",
            "Epoch [5/10], Step [51/101], Loss: 1.5452\n",
            "Epoch [5/10], Step [52/101], Loss: 1.3405\n",
            "Epoch [5/10], Step [53/101], Loss: 1.6905\n",
            "Epoch [5/10], Step [54/101], Loss: 1.6747\n",
            "Epoch [5/10], Step [55/101], Loss: 1.5852\n",
            "Epoch [5/10], Step [56/101], Loss: 1.6083\n",
            "Epoch [5/10], Step [57/101], Loss: 1.6703\n",
            "Epoch [5/10], Step [58/101], Loss: 1.5312\n",
            "Epoch [5/10], Step [59/101], Loss: 1.7477\n",
            "Epoch [5/10], Step [60/101], Loss: 1.5449\n",
            "Epoch [5/10], Step [61/101], Loss: 1.5701\n",
            "Epoch [5/10], Step [62/101], Loss: 1.3248\n",
            "Epoch [5/10], Step [63/101], Loss: 1.5764\n",
            "Epoch [5/10], Step [64/101], Loss: 1.4627\n",
            "Epoch [5/10], Step [65/101], Loss: 1.4340\n",
            "Epoch [5/10], Step [66/101], Loss: 1.5048\n",
            "Epoch [5/10], Step [67/101], Loss: 1.7205\n",
            "Epoch [5/10], Step [68/101], Loss: 1.2801\n",
            "Epoch [5/10], Step [69/101], Loss: 1.8024\n",
            "Epoch [5/10], Step [70/101], Loss: 1.5533\n",
            "Epoch [5/10], Step [71/101], Loss: 1.4141\n",
            "Epoch [5/10], Step [72/101], Loss: 1.7446\n",
            "Epoch [5/10], Step [73/101], Loss: 1.5409\n",
            "Epoch [5/10], Step [74/101], Loss: 1.5401\n",
            "Epoch [5/10], Step [75/101], Loss: 1.5608\n",
            "Epoch [5/10], Step [76/101], Loss: 1.4610\n",
            "Epoch [5/10], Step [77/101], Loss: 1.2849\n",
            "Epoch [5/10], Step [78/101], Loss: 1.6556\n",
            "Epoch [5/10], Step [79/101], Loss: 1.5093\n",
            "Epoch [5/10], Step [80/101], Loss: 1.9130\n",
            "Epoch [5/10], Step [81/101], Loss: 1.3127\n",
            "Epoch [5/10], Step [82/101], Loss: 1.4265\n",
            "Epoch [5/10], Step [83/101], Loss: 1.6720\n",
            "Epoch [5/10], Step [84/101], Loss: 1.3645\n",
            "Epoch [5/10], Step [85/101], Loss: 1.5925\n",
            "Epoch [5/10], Step [86/101], Loss: 1.6760\n",
            "Epoch [5/10], Step [87/101], Loss: 2.1761\n",
            "Epoch [5/10], Step [88/101], Loss: 1.6242\n",
            "Epoch [5/10], Step [89/101], Loss: 1.4417\n",
            "Epoch [5/10], Step [90/101], Loss: 1.4359\n",
            "Epoch [5/10], Step [91/101], Loss: 1.3940\n",
            "Epoch [5/10], Step [92/101], Loss: 1.7595\n",
            "Epoch [5/10], Step [93/101], Loss: 1.4029\n",
            "Epoch [5/10], Step [94/101], Loss: 1.3350\n",
            "Epoch [5/10], Step [95/101], Loss: 1.2613\n",
            "Epoch [5/10], Step [96/101], Loss: 1.1663\n",
            "Epoch [5/10], Step [97/101], Loss: 1.3336\n",
            "Epoch [5/10], Step [98/101], Loss: 1.4244\n",
            "Epoch [5/10], Step [99/101], Loss: 1.5023\n",
            "Epoch [5/10], Step [100/101], Loss: 1.5731\n",
            "Epoch [5/10], Step [101/101], Loss: 1.8763\n",
            "Epoch finished\n",
            "Epoch [6/10], Step [1/101], Loss: 1.6183\n",
            "Epoch [6/10], Step [2/101], Loss: 1.4473\n",
            "Epoch [6/10], Step [3/101], Loss: 1.6545\n",
            "Epoch [6/10], Step [4/101], Loss: 1.5014\n",
            "Epoch [6/10], Step [5/101], Loss: 1.6294\n",
            "Epoch [6/10], Step [6/101], Loss: 1.5305\n",
            "Epoch [6/10], Step [7/101], Loss: 1.2500\n",
            "Epoch [6/10], Step [8/101], Loss: 1.4883\n",
            "Epoch [6/10], Step [9/101], Loss: 1.3295\n",
            "Epoch [6/10], Step [10/101], Loss: 1.2565\n",
            "Epoch [6/10], Step [11/101], Loss: 1.5885\n",
            "Epoch [6/10], Step [12/101], Loss: 1.2843\n",
            "Epoch [6/10], Step [13/101], Loss: 1.3766\n",
            "Epoch [6/10], Step [14/101], Loss: 1.1594\n",
            "Epoch [6/10], Step [15/101], Loss: 1.2447\n",
            "Epoch [6/10], Step [16/101], Loss: 1.4557\n",
            "Epoch [6/10], Step [17/101], Loss: 1.4089\n",
            "Epoch [6/10], Step [18/101], Loss: 1.3749\n",
            "Epoch [6/10], Step [19/101], Loss: 1.6383\n",
            "Epoch [6/10], Step [20/101], Loss: 1.2607\n",
            "Epoch [6/10], Step [21/101], Loss: 1.4076\n",
            "Epoch [6/10], Step [22/101], Loss: 1.4522\n",
            "Epoch [6/10], Step [23/101], Loss: 1.4494\n",
            "Epoch [6/10], Step [24/101], Loss: 1.5588\n",
            "Epoch [6/10], Step [25/101], Loss: 1.4497\n",
            "Epoch [6/10], Step [26/101], Loss: 1.4706\n",
            "Epoch [6/10], Step [27/101], Loss: 1.6520\n",
            "Epoch [6/10], Step [28/101], Loss: 1.2118\n",
            "Epoch [6/10], Step [29/101], Loss: 1.5178\n",
            "Epoch [6/10], Step [30/101], Loss: 1.5525\n",
            "Epoch [6/10], Step [31/101], Loss: 1.1375\n",
            "Epoch [6/10], Step [32/101], Loss: 1.8809\n",
            "Epoch [6/10], Step [33/101], Loss: 1.2827\n",
            "Epoch [6/10], Step [34/101], Loss: 1.2718\n",
            "Epoch [6/10], Step [35/101], Loss: 1.4513\n",
            "Epoch [6/10], Step [36/101], Loss: 1.4844\n",
            "Epoch [6/10], Step [37/101], Loss: 1.6445\n",
            "Epoch [6/10], Step [38/101], Loss: 1.3519\n",
            "Epoch [6/10], Step [39/101], Loss: 1.8004\n",
            "Epoch [6/10], Step [40/101], Loss: 1.3573\n",
            "Epoch [6/10], Step [41/101], Loss: 1.3442\n",
            "Epoch [6/10], Step [42/101], Loss: 1.4554\n",
            "Epoch [6/10], Step [43/101], Loss: 1.1735\n",
            "Epoch [6/10], Step [44/101], Loss: 1.3401\n",
            "Epoch [6/10], Step [45/101], Loss: 1.3935\n",
            "Epoch [6/10], Step [46/101], Loss: 1.4356\n",
            "Epoch [6/10], Step [47/101], Loss: 1.4016\n",
            "Epoch [6/10], Step [48/101], Loss: 1.2476\n",
            "Epoch [6/10], Step [49/101], Loss: 1.2377\n",
            "Epoch [6/10], Step [50/101], Loss: 1.4027\n",
            "Epoch [6/10], Step [51/101], Loss: 1.5138\n",
            "Epoch [6/10], Step [52/101], Loss: 1.4502\n",
            "Epoch [6/10], Step [53/101], Loss: 1.2341\n",
            "Epoch [6/10], Step [54/101], Loss: 1.3496\n",
            "Epoch [6/10], Step [55/101], Loss: 1.3963\n",
            "Epoch [6/10], Step [56/101], Loss: 1.2224\n",
            "Epoch [6/10], Step [57/101], Loss: 1.6272\n",
            "Epoch [6/10], Step [58/101], Loss: 1.6026\n",
            "Epoch [6/10], Step [59/101], Loss: 1.3226\n",
            "Epoch [6/10], Step [60/101], Loss: 1.4821\n",
            "Epoch [6/10], Step [61/101], Loss: 1.3904\n",
            "Epoch [6/10], Step [62/101], Loss: 1.5903\n",
            "Epoch [6/10], Step [63/101], Loss: 1.3804\n",
            "Epoch [6/10], Step [64/101], Loss: 1.3058\n",
            "Epoch [6/10], Step [65/101], Loss: 1.4434\n",
            "Epoch [6/10], Step [66/101], Loss: 1.4996\n",
            "Epoch [6/10], Step [67/101], Loss: 1.5500\n",
            "Epoch [6/10], Step [68/101], Loss: 1.5836\n",
            "Epoch [6/10], Step [69/101], Loss: 1.3112\n",
            "Epoch [6/10], Step [70/101], Loss: 1.3022\n",
            "Epoch [6/10], Step [71/101], Loss: 1.2401\n",
            "Epoch [6/10], Step [72/101], Loss: 1.3558\n",
            "Epoch [6/10], Step [73/101], Loss: 1.4603\n",
            "Epoch [6/10], Step [74/101], Loss: 1.2839\n",
            "Epoch [6/10], Step [75/101], Loss: 1.2155\n",
            "Epoch [6/10], Step [76/101], Loss: 1.5824\n",
            "Epoch [6/10], Step [77/101], Loss: 1.5022\n",
            "Epoch [6/10], Step [78/101], Loss: 1.5909\n",
            "Epoch [6/10], Step [79/101], Loss: 1.5059\n",
            "Epoch [6/10], Step [80/101], Loss: 1.2505\n",
            "Epoch [6/10], Step [81/101], Loss: 1.5691\n",
            "Epoch [6/10], Step [82/101], Loss: 1.4678\n",
            "Epoch [6/10], Step [83/101], Loss: 1.3516\n",
            "Epoch [6/10], Step [84/101], Loss: 1.3536\n",
            "Epoch [6/10], Step [85/101], Loss: 1.3916\n",
            "Epoch [6/10], Step [86/101], Loss: 1.4373\n",
            "Epoch [6/10], Step [87/101], Loss: 1.3012\n",
            "Epoch [6/10], Step [88/101], Loss: 1.6921\n",
            "Epoch [6/10], Step [89/101], Loss: 1.1292\n",
            "Epoch [6/10], Step [90/101], Loss: 1.5279\n",
            "Epoch [6/10], Step [91/101], Loss: 1.3458\n",
            "Epoch [6/10], Step [92/101], Loss: 1.5132\n",
            "Epoch [6/10], Step [93/101], Loss: 1.6589\n",
            "Epoch [6/10], Step [94/101], Loss: 1.4825\n",
            "Epoch [6/10], Step [95/101], Loss: 1.4439\n",
            "Epoch [6/10], Step [96/101], Loss: 1.1549\n",
            "Epoch [6/10], Step [97/101], Loss: 1.5212\n",
            "Epoch [6/10], Step [98/101], Loss: 1.5263\n",
            "Epoch [6/10], Step [99/101], Loss: 1.4743\n",
            "Epoch [6/10], Step [100/101], Loss: 1.3058\n",
            "Epoch [6/10], Step [101/101], Loss: 1.9708\n",
            "Epoch finished\n",
            "Epoch [7/10], Step [1/101], Loss: 1.3273\n",
            "Epoch [7/10], Step [2/101], Loss: 1.1160\n",
            "Epoch [7/10], Step [3/101], Loss: 1.7514\n",
            "Epoch [7/10], Step [4/101], Loss: 1.2881\n",
            "Epoch [7/10], Step [5/101], Loss: 1.2504\n",
            "Epoch [7/10], Step [6/101], Loss: 1.4839\n",
            "Epoch [7/10], Step [7/101], Loss: 1.2792\n",
            "Epoch [7/10], Step [8/101], Loss: 1.3219\n",
            "Epoch [7/10], Step [9/101], Loss: 1.4409\n",
            "Epoch [7/10], Step [10/101], Loss: 1.5591\n",
            "Epoch [7/10], Step [11/101], Loss: 1.5610\n",
            "Epoch [7/10], Step [12/101], Loss: 1.4884\n",
            "Epoch [7/10], Step [13/101], Loss: 1.5636\n",
            "Epoch [7/10], Step [14/101], Loss: 1.2945\n",
            "Epoch [7/10], Step [15/101], Loss: 1.3018\n",
            "Epoch [7/10], Step [16/101], Loss: 1.6264\n",
            "Epoch [7/10], Step [17/101], Loss: 1.2077\n",
            "Epoch [7/10], Step [18/101], Loss: 1.2598\n",
            "Epoch [7/10], Step [19/101], Loss: 1.3345\n",
            "Epoch [7/10], Step [20/101], Loss: 1.1892\n",
            "Epoch [7/10], Step [21/101], Loss: 1.2300\n",
            "Epoch [7/10], Step [22/101], Loss: 1.2418\n",
            "Epoch [7/10], Step [23/101], Loss: 1.0685\n",
            "Epoch [7/10], Step [24/101], Loss: 1.5203\n",
            "Epoch [7/10], Step [25/101], Loss: 1.5824\n",
            "Epoch [7/10], Step [26/101], Loss: 1.3426\n",
            "Epoch [7/10], Step [27/101], Loss: 1.2978\n",
            "Epoch [7/10], Step [28/101], Loss: 1.4518\n",
            "Epoch [7/10], Step [29/101], Loss: 1.5963\n",
            "Epoch [7/10], Step [30/101], Loss: 1.2222\n",
            "Epoch [7/10], Step [31/101], Loss: 1.4183\n",
            "Epoch [7/10], Step [32/101], Loss: 1.2304\n",
            "Epoch [7/10], Step [33/101], Loss: 1.2654\n",
            "Epoch [7/10], Step [34/101], Loss: 1.3540\n",
            "Epoch [7/10], Step [35/101], Loss: 1.4464\n",
            "Epoch [7/10], Step [36/101], Loss: 1.2398\n",
            "Epoch [7/10], Step [37/101], Loss: 1.1540\n",
            "Epoch [7/10], Step [38/101], Loss: 1.3712\n",
            "Epoch [7/10], Step [39/101], Loss: 1.2074\n",
            "Epoch [7/10], Step [40/101], Loss: 1.2803\n",
            "Epoch [7/10], Step [41/101], Loss: 1.1918\n",
            "Epoch [7/10], Step [42/101], Loss: 1.1047\n",
            "Epoch [7/10], Step [43/101], Loss: 1.5553\n",
            "Epoch [7/10], Step [44/101], Loss: 1.3176\n",
            "Epoch [7/10], Step [45/101], Loss: 1.1948\n",
            "Epoch [7/10], Step [46/101], Loss: 1.0488\n",
            "Epoch [7/10], Step [47/101], Loss: 1.2134\n",
            "Epoch [7/10], Step [48/101], Loss: 1.4407\n",
            "Epoch [7/10], Step [49/101], Loss: 1.3761\n",
            "Epoch [7/10], Step [50/101], Loss: 1.1433\n",
            "Epoch [7/10], Step [51/101], Loss: 1.4027\n",
            "Epoch [7/10], Step [52/101], Loss: 1.8348\n",
            "Epoch [7/10], Step [53/101], Loss: 1.3625\n",
            "Epoch [7/10], Step [54/101], Loss: 1.0636\n",
            "Epoch [7/10], Step [55/101], Loss: 1.1485\n",
            "Epoch [7/10], Step [56/101], Loss: 1.8159\n",
            "Epoch [7/10], Step [57/101], Loss: 1.4990\n",
            "Epoch [7/10], Step [58/101], Loss: 1.3931\n",
            "Epoch [7/10], Step [59/101], Loss: 1.4025\n",
            "Epoch [7/10], Step [60/101], Loss: 1.1934\n",
            "Epoch [7/10], Step [61/101], Loss: 1.5220\n",
            "Epoch [7/10], Step [62/101], Loss: 1.2816\n",
            "Epoch [7/10], Step [63/101], Loss: 1.2801\n",
            "Epoch [7/10], Step [64/101], Loss: 1.2849\n",
            "Epoch [7/10], Step [65/101], Loss: 1.3363\n",
            "Epoch [7/10], Step [66/101], Loss: 1.2855\n",
            "Epoch [7/10], Step [67/101], Loss: 1.0780\n",
            "Epoch [7/10], Step [68/101], Loss: 1.3951\n",
            "Epoch [7/10], Step [69/101], Loss: 1.1858\n",
            "Epoch [7/10], Step [70/101], Loss: 1.4413\n",
            "Epoch [7/10], Step [71/101], Loss: 1.3278\n",
            "Epoch [7/10], Step [72/101], Loss: 1.3377\n",
            "Epoch [7/10], Step [73/101], Loss: 1.4620\n",
            "Epoch [7/10], Step [74/101], Loss: 1.4088\n",
            "Epoch [7/10], Step [75/101], Loss: 1.3874\n",
            "Epoch [7/10], Step [76/101], Loss: 1.2069\n",
            "Epoch [7/10], Step [77/101], Loss: 1.3886\n",
            "Epoch [7/10], Step [78/101], Loss: 1.0413\n",
            "Epoch [7/10], Step [79/101], Loss: 1.2573\n",
            "Epoch [7/10], Step [80/101], Loss: 1.2402\n",
            "Epoch [7/10], Step [81/101], Loss: 1.2924\n",
            "Epoch [7/10], Step [82/101], Loss: 1.2496\n",
            "Epoch [7/10], Step [83/101], Loss: 1.2182\n",
            "Epoch [7/10], Step [84/101], Loss: 1.1016\n",
            "Epoch [7/10], Step [85/101], Loss: 1.2659\n",
            "Epoch [7/10], Step [86/101], Loss: 1.4654\n",
            "Epoch [7/10], Step [87/101], Loss: 1.2765\n",
            "Epoch [7/10], Step [88/101], Loss: 1.2457\n",
            "Epoch [7/10], Step [89/101], Loss: 1.4138\n",
            "Epoch [7/10], Step [90/101], Loss: 1.1995\n",
            "Epoch [7/10], Step [91/101], Loss: 1.0804\n",
            "Epoch [7/10], Step [92/101], Loss: 1.6229\n",
            "Epoch [7/10], Step [93/101], Loss: 1.0957\n",
            "Epoch [7/10], Step [94/101], Loss: 1.0822\n",
            "Epoch [7/10], Step [95/101], Loss: 1.3843\n",
            "Epoch [7/10], Step [96/101], Loss: 1.5982\n",
            "Epoch [7/10], Step [97/101], Loss: 1.1077\n",
            "Epoch [7/10], Step [98/101], Loss: 1.0365\n",
            "Epoch [7/10], Step [99/101], Loss: 1.4379\n",
            "Epoch [7/10], Step [100/101], Loss: 0.8701\n",
            "Epoch [7/10], Step [101/101], Loss: 1.0090\n",
            "Epoch finished\n",
            "Epoch [8/10], Step [1/101], Loss: 1.2628\n",
            "Epoch [8/10], Step [2/101], Loss: 1.2075\n",
            "Epoch [8/10], Step [3/101], Loss: 1.3639\n",
            "Epoch [8/10], Step [4/101], Loss: 1.4097\n",
            "Epoch [8/10], Step [5/101], Loss: 1.2984\n",
            "Epoch [8/10], Step [6/101], Loss: 1.6887\n",
            "Epoch [8/10], Step [7/101], Loss: 1.0148\n",
            "Epoch [8/10], Step [8/101], Loss: 1.1776\n",
            "Epoch [8/10], Step [9/101], Loss: 1.3896\n",
            "Epoch [8/10], Step [10/101], Loss: 1.1871\n",
            "Epoch [8/10], Step [11/101], Loss: 0.8048\n",
            "Epoch [8/10], Step [12/101], Loss: 1.5225\n",
            "Epoch [8/10], Step [13/101], Loss: 1.0513\n",
            "Epoch [8/10], Step [14/101], Loss: 1.2108\n",
            "Epoch [8/10], Step [15/101], Loss: 0.7285\n",
            "Epoch [8/10], Step [16/101], Loss: 0.8151\n",
            "Epoch [8/10], Step [17/101], Loss: 0.9725\n",
            "Epoch [8/10], Step [18/101], Loss: 0.9256\n",
            "Epoch [8/10], Step [19/101], Loss: 0.9645\n",
            "Epoch [8/10], Step [20/101], Loss: 1.2887\n",
            "Epoch [8/10], Step [21/101], Loss: 1.4049\n",
            "Epoch [8/10], Step [22/101], Loss: 1.3931\n",
            "Epoch [8/10], Step [23/101], Loss: 0.9883\n",
            "Epoch [8/10], Step [24/101], Loss: 1.1519\n",
            "Epoch [8/10], Step [25/101], Loss: 1.1764\n",
            "Epoch [8/10], Step [26/101], Loss: 1.2813\n",
            "Epoch [8/10], Step [27/101], Loss: 1.2498\n",
            "Epoch [8/10], Step [28/101], Loss: 1.2013\n",
            "Epoch [8/10], Step [29/101], Loss: 1.0478\n",
            "Epoch [8/10], Step [30/101], Loss: 1.4793\n",
            "Epoch [8/10], Step [31/101], Loss: 1.3767\n",
            "Epoch [8/10], Step [32/101], Loss: 1.4501\n",
            "Epoch [8/10], Step [33/101], Loss: 1.3948\n",
            "Epoch [8/10], Step [34/101], Loss: 0.8368\n",
            "Epoch [8/10], Step [35/101], Loss: 1.2427\n",
            "Epoch [8/10], Step [36/101], Loss: 1.3647\n",
            "Epoch [8/10], Step [37/101], Loss: 1.2259\n",
            "Epoch [8/10], Step [38/101], Loss: 0.9802\n",
            "Epoch [8/10], Step [39/101], Loss: 1.8737\n",
            "Epoch [8/10], Step [40/101], Loss: 0.8751\n",
            "Epoch [8/10], Step [41/101], Loss: 1.3342\n",
            "Epoch [8/10], Step [42/101], Loss: 1.6755\n",
            "Epoch [8/10], Step [43/101], Loss: 1.1962\n",
            "Epoch [8/10], Step [44/101], Loss: 1.1436\n",
            "Epoch [8/10], Step [45/101], Loss: 1.3411\n",
            "Epoch [8/10], Step [46/101], Loss: 1.2801\n",
            "Epoch [8/10], Step [47/101], Loss: 1.2590\n",
            "Epoch [8/10], Step [48/101], Loss: 0.9967\n",
            "Epoch [8/10], Step [49/101], Loss: 1.4788\n",
            "Epoch [8/10], Step [50/101], Loss: 1.1553\n",
            "Epoch [8/10], Step [51/101], Loss: 1.4325\n",
            "Epoch [8/10], Step [52/101], Loss: 1.0090\n",
            "Epoch [8/10], Step [53/101], Loss: 1.3052\n",
            "Epoch [8/10], Step [54/101], Loss: 1.3506\n",
            "Epoch [8/10], Step [55/101], Loss: 1.2680\n",
            "Epoch [8/10], Step [56/101], Loss: 1.1656\n",
            "Epoch [8/10], Step [57/101], Loss: 1.1556\n",
            "Epoch [8/10], Step [58/101], Loss: 1.1207\n",
            "Epoch [8/10], Step [59/101], Loss: 1.4318\n",
            "Epoch [8/10], Step [60/101], Loss: 1.4871\n",
            "Epoch [8/10], Step [61/101], Loss: 0.8956\n",
            "Epoch [8/10], Step [62/101], Loss: 1.1883\n",
            "Epoch [8/10], Step [63/101], Loss: 1.2021\n",
            "Epoch [8/10], Step [64/101], Loss: 1.6682\n",
            "Epoch [8/10], Step [65/101], Loss: 1.2743\n",
            "Epoch [8/10], Step [66/101], Loss: 1.1122\n",
            "Epoch [8/10], Step [67/101], Loss: 1.3727\n",
            "Epoch [8/10], Step [68/101], Loss: 1.5535\n",
            "Epoch [8/10], Step [69/101], Loss: 1.3671\n",
            "Epoch [8/10], Step [70/101], Loss: 1.0481\n",
            "Epoch [8/10], Step [71/101], Loss: 1.0845\n",
            "Epoch [8/10], Step [72/101], Loss: 1.2966\n",
            "Epoch [8/10], Step [73/101], Loss: 1.4036\n",
            "Epoch [8/10], Step [74/101], Loss: 1.1775\n",
            "Epoch [8/10], Step [75/101], Loss: 1.3878\n",
            "Epoch [8/10], Step [76/101], Loss: 0.9647\n",
            "Epoch [8/10], Step [77/101], Loss: 1.0082\n",
            "Epoch [8/10], Step [78/101], Loss: 1.1981\n",
            "Epoch [8/10], Step [79/101], Loss: 1.3679\n",
            "Epoch [8/10], Step [80/101], Loss: 1.0460\n",
            "Epoch [8/10], Step [81/101], Loss: 1.1706\n",
            "Epoch [8/10], Step [82/101], Loss: 1.0309\n",
            "Epoch [8/10], Step [83/101], Loss: 1.1024\n",
            "Epoch [8/10], Step [84/101], Loss: 1.4503\n",
            "Epoch [8/10], Step [85/101], Loss: 1.3052\n",
            "Epoch [8/10], Step [86/101], Loss: 1.2896\n",
            "Epoch [8/10], Step [87/101], Loss: 1.1139\n",
            "Epoch [8/10], Step [88/101], Loss: 1.4463\n",
            "Epoch [8/10], Step [89/101], Loss: 1.4960\n",
            "Epoch [8/10], Step [90/101], Loss: 1.1471\n",
            "Epoch [8/10], Step [91/101], Loss: 1.3087\n",
            "Epoch [8/10], Step [92/101], Loss: 1.3311\n",
            "Epoch [8/10], Step [93/101], Loss: 1.2093\n",
            "Epoch [8/10], Step [94/101], Loss: 1.3967\n",
            "Epoch [8/10], Step [95/101], Loss: 1.3419\n",
            "Epoch [8/10], Step [96/101], Loss: 1.1730\n",
            "Epoch [8/10], Step [97/101], Loss: 1.5660\n",
            "Epoch [8/10], Step [98/101], Loss: 1.2495\n",
            "Epoch [8/10], Step [99/101], Loss: 1.2062\n",
            "Epoch [8/10], Step [100/101], Loss: 1.1114\n",
            "Epoch [8/10], Step [101/101], Loss: 1.3751\n",
            "Epoch finished\n",
            "Epoch [9/10], Step [1/101], Loss: 1.4602\n",
            "Epoch [9/10], Step [2/101], Loss: 1.1809\n",
            "Epoch [9/10], Step [3/101], Loss: 1.2115\n",
            "Epoch [9/10], Step [4/101], Loss: 1.3316\n",
            "Epoch [9/10], Step [5/101], Loss: 1.1020\n",
            "Epoch [9/10], Step [6/101], Loss: 1.0044\n",
            "Epoch [9/10], Step [7/101], Loss: 1.1303\n",
            "Epoch [9/10], Step [8/101], Loss: 1.2313\n",
            "Epoch [9/10], Step [9/101], Loss: 1.0648\n",
            "Epoch [9/10], Step [10/101], Loss: 1.1902\n",
            "Epoch [9/10], Step [11/101], Loss: 1.1850\n",
            "Epoch [9/10], Step [12/101], Loss: 1.1125\n",
            "Epoch [9/10], Step [13/101], Loss: 1.1816\n",
            "Epoch [9/10], Step [14/101], Loss: 1.1167\n",
            "Epoch [9/10], Step [15/101], Loss: 0.8537\n",
            "Epoch [9/10], Step [16/101], Loss: 0.8866\n",
            "Epoch [9/10], Step [17/101], Loss: 0.8892\n",
            "Epoch [9/10], Step [18/101], Loss: 0.8650\n",
            "Epoch [9/10], Step [19/101], Loss: 1.0189\n",
            "Epoch [9/10], Step [20/101], Loss: 1.1566\n",
            "Epoch [9/10], Step [21/101], Loss: 0.9227\n",
            "Epoch [9/10], Step [22/101], Loss: 1.1364\n",
            "Epoch [9/10], Step [23/101], Loss: 0.7892\n",
            "Epoch [9/10], Step [24/101], Loss: 1.1274\n",
            "Epoch [9/10], Step [25/101], Loss: 1.3059\n",
            "Epoch [9/10], Step [26/101], Loss: 0.9399\n",
            "Epoch [9/10], Step [27/101], Loss: 1.0641\n",
            "Epoch [9/10], Step [28/101], Loss: 1.2703\n",
            "Epoch [9/10], Step [29/101], Loss: 1.1055\n",
            "Epoch [9/10], Step [30/101], Loss: 0.9873\n",
            "Epoch [9/10], Step [31/101], Loss: 1.2364\n",
            "Epoch [9/10], Step [32/101], Loss: 1.4317\n",
            "Epoch [9/10], Step [33/101], Loss: 1.2184\n",
            "Epoch [9/10], Step [34/101], Loss: 1.0741\n",
            "Epoch [9/10], Step [35/101], Loss: 1.4589\n",
            "Epoch [9/10], Step [36/101], Loss: 1.2823\n",
            "Epoch [9/10], Step [37/101], Loss: 1.1979\n",
            "Epoch [9/10], Step [38/101], Loss: 1.3891\n",
            "Epoch [9/10], Step [39/101], Loss: 0.9235\n",
            "Epoch [9/10], Step [40/101], Loss: 0.9104\n",
            "Epoch [9/10], Step [41/101], Loss: 1.2444\n",
            "Epoch [9/10], Step [42/101], Loss: 1.1953\n",
            "Epoch [9/10], Step [43/101], Loss: 1.1667\n",
            "Epoch [9/10], Step [44/101], Loss: 1.1550\n",
            "Epoch [9/10], Step [45/101], Loss: 1.5802\n",
            "Epoch [9/10], Step [46/101], Loss: 1.1235\n",
            "Epoch [9/10], Step [47/101], Loss: 1.1475\n",
            "Epoch [9/10], Step [48/101], Loss: 1.6654\n",
            "Epoch [9/10], Step [49/101], Loss: 0.8556\n",
            "Epoch [9/10], Step [50/101], Loss: 0.9526\n",
            "Epoch [9/10], Step [51/101], Loss: 1.3297\n",
            "Epoch [9/10], Step [52/101], Loss: 1.0815\n",
            "Epoch [9/10], Step [53/101], Loss: 1.1096\n",
            "Epoch [9/10], Step [54/101], Loss: 1.3872\n",
            "Epoch [9/10], Step [55/101], Loss: 1.3391\n",
            "Epoch [9/10], Step [56/101], Loss: 1.1050\n",
            "Epoch [9/10], Step [57/101], Loss: 0.9355\n",
            "Epoch [9/10], Step [58/101], Loss: 1.1148\n",
            "Epoch [9/10], Step [59/101], Loss: 1.3405\n",
            "Epoch [9/10], Step [60/101], Loss: 1.2648\n",
            "Epoch [9/10], Step [61/101], Loss: 1.5361\n",
            "Epoch [9/10], Step [62/101], Loss: 1.0992\n",
            "Epoch [9/10], Step [63/101], Loss: 0.9723\n",
            "Epoch [9/10], Step [64/101], Loss: 1.3174\n",
            "Epoch [9/10], Step [65/101], Loss: 0.8149\n",
            "Epoch [9/10], Step [66/101], Loss: 1.4130\n",
            "Epoch [9/10], Step [67/101], Loss: 1.0449\n",
            "Epoch [9/10], Step [68/101], Loss: 1.3028\n",
            "Epoch [9/10], Step [69/101], Loss: 0.9219\n",
            "Epoch [9/10], Step [70/101], Loss: 1.1041\n",
            "Epoch [9/10], Step [71/101], Loss: 1.2961\n",
            "Epoch [9/10], Step [72/101], Loss: 1.1881\n",
            "Epoch [9/10], Step [73/101], Loss: 1.1536\n",
            "Epoch [9/10], Step [74/101], Loss: 1.4210\n",
            "Epoch [9/10], Step [75/101], Loss: 0.9835\n",
            "Epoch [9/10], Step [76/101], Loss: 1.4775\n",
            "Epoch [9/10], Step [77/101], Loss: 1.0481\n",
            "Epoch [9/10], Step [78/101], Loss: 1.0889\n",
            "Epoch [9/10], Step [79/101], Loss: 1.3831\n",
            "Epoch [9/10], Step [80/101], Loss: 1.1270\n",
            "Epoch [9/10], Step [81/101], Loss: 0.8777\n",
            "Epoch [9/10], Step [82/101], Loss: 1.1503\n",
            "Epoch [9/10], Step [83/101], Loss: 1.2046\n",
            "Epoch [9/10], Step [84/101], Loss: 1.1850\n",
            "Epoch [9/10], Step [85/101], Loss: 0.9362\n",
            "Epoch [9/10], Step [86/101], Loss: 1.1545\n",
            "Epoch [9/10], Step [87/101], Loss: 1.2322\n",
            "Epoch [9/10], Step [88/101], Loss: 1.0583\n",
            "Epoch [9/10], Step [89/101], Loss: 1.0455\n",
            "Epoch [9/10], Step [90/101], Loss: 0.9282\n",
            "Epoch [9/10], Step [91/101], Loss: 1.1048\n",
            "Epoch [9/10], Step [92/101], Loss: 1.2720\n",
            "Epoch [9/10], Step [93/101], Loss: 0.9789\n",
            "Epoch [9/10], Step [94/101], Loss: 1.2721\n",
            "Epoch [9/10], Step [95/101], Loss: 1.1174\n",
            "Epoch [9/10], Step [96/101], Loss: 1.4115\n",
            "Epoch [9/10], Step [97/101], Loss: 1.0179\n",
            "Epoch [9/10], Step [98/101], Loss: 1.0363\n",
            "Epoch [9/10], Step [99/101], Loss: 1.2440\n",
            "Epoch [9/10], Step [100/101], Loss: 1.0124\n",
            "Epoch [9/10], Step [101/101], Loss: 1.6793\n",
            "Epoch finished\n",
            "Epoch [10/10], Step [1/101], Loss: 1.1639\n",
            "Epoch [10/10], Step [2/101], Loss: 1.2517\n",
            "Epoch [10/10], Step [3/101], Loss: 0.9586\n",
            "Epoch [10/10], Step [4/101], Loss: 0.9565\n",
            "Epoch [10/10], Step [5/101], Loss: 1.1614\n",
            "Epoch [10/10], Step [6/101], Loss: 1.2218\n",
            "Epoch [10/10], Step [7/101], Loss: 1.3449\n",
            "Epoch [10/10], Step [8/101], Loss: 0.7331\n",
            "Epoch [10/10], Step [9/101], Loss: 1.3168\n",
            "Epoch [10/10], Step [10/101], Loss: 1.0359\n",
            "Epoch [10/10], Step [11/101], Loss: 1.1886\n",
            "Epoch [10/10], Step [12/101], Loss: 1.1738\n",
            "Epoch [10/10], Step [13/101], Loss: 1.1249\n",
            "Epoch [10/10], Step [14/101], Loss: 1.0466\n",
            "Epoch [10/10], Step [15/101], Loss: 1.3555\n",
            "Epoch [10/10], Step [16/101], Loss: 1.0134\n",
            "Epoch [10/10], Step [17/101], Loss: 0.8259\n",
            "Epoch [10/10], Step [18/101], Loss: 1.3313\n",
            "Epoch [10/10], Step [19/101], Loss: 1.2370\n",
            "Epoch [10/10], Step [20/101], Loss: 1.2709\n",
            "Epoch [10/10], Step [21/101], Loss: 0.8167\n",
            "Epoch [10/10], Step [22/101], Loss: 0.8460\n",
            "Epoch [10/10], Step [23/101], Loss: 1.2285\n",
            "Epoch [10/10], Step [24/101], Loss: 1.0749\n",
            "Epoch [10/10], Step [25/101], Loss: 1.3227\n",
            "Epoch [10/10], Step [26/101], Loss: 1.2456\n",
            "Epoch [10/10], Step [27/101], Loss: 1.0065\n",
            "Epoch [10/10], Step [28/101], Loss: 0.9369\n",
            "Epoch [10/10], Step [29/101], Loss: 1.1222\n",
            "Epoch [10/10], Step [30/101], Loss: 1.1686\n",
            "Epoch [10/10], Step [31/101], Loss: 0.9316\n",
            "Epoch [10/10], Step [32/101], Loss: 0.9850\n",
            "Epoch [10/10], Step [33/101], Loss: 1.0587\n",
            "Epoch [10/10], Step [34/101], Loss: 1.2312\n",
            "Epoch [10/10], Step [35/101], Loss: 1.3479\n",
            "Epoch [10/10], Step [36/101], Loss: 0.9783\n",
            "Epoch [10/10], Step [37/101], Loss: 1.4501\n",
            "Epoch [10/10], Step [38/101], Loss: 1.3200\n",
            "Epoch [10/10], Step [39/101], Loss: 1.0067\n",
            "Epoch [10/10], Step [40/101], Loss: 0.9693\n",
            "Epoch [10/10], Step [41/101], Loss: 1.0674\n",
            "Epoch [10/10], Step [42/101], Loss: 1.0697\n",
            "Epoch [10/10], Step [43/101], Loss: 1.3605\n",
            "Epoch [10/10], Step [44/101], Loss: 1.0663\n",
            "Epoch [10/10], Step [45/101], Loss: 1.1511\n",
            "Epoch [10/10], Step [46/101], Loss: 1.3350\n",
            "Epoch [10/10], Step [47/101], Loss: 1.1658\n",
            "Epoch [10/10], Step [48/101], Loss: 1.1275\n",
            "Epoch [10/10], Step [49/101], Loss: 1.3871\n",
            "Epoch [10/10], Step [50/101], Loss: 1.1438\n",
            "Epoch [10/10], Step [51/101], Loss: 0.9069\n",
            "Epoch [10/10], Step [52/101], Loss: 1.1270\n",
            "Epoch [10/10], Step [53/101], Loss: 0.8913\n",
            "Epoch [10/10], Step [54/101], Loss: 1.0513\n",
            "Epoch [10/10], Step [55/101], Loss: 1.0698\n",
            "Epoch [10/10], Step [56/101], Loss: 1.1629\n",
            "Epoch [10/10], Step [57/101], Loss: 1.0357\n",
            "Epoch [10/10], Step [58/101], Loss: 1.0954\n",
            "Epoch [10/10], Step [59/101], Loss: 1.0106\n",
            "Epoch [10/10], Step [60/101], Loss: 1.1456\n",
            "Epoch [10/10], Step [61/101], Loss: 0.7797\n",
            "Epoch [10/10], Step [62/101], Loss: 0.8797\n",
            "Epoch [10/10], Step [63/101], Loss: 0.9610\n",
            "Epoch [10/10], Step [64/101], Loss: 1.0982\n",
            "Epoch [10/10], Step [65/101], Loss: 1.1585\n",
            "Epoch [10/10], Step [66/101], Loss: 1.1955\n",
            "Epoch [10/10], Step [67/101], Loss: 1.0985\n",
            "Epoch [10/10], Step [68/101], Loss: 1.0002\n",
            "Epoch [10/10], Step [69/101], Loss: 0.8684\n",
            "Epoch [10/10], Step [70/101], Loss: 0.8913\n",
            "Epoch [10/10], Step [71/101], Loss: 1.2299\n",
            "Epoch [10/10], Step [72/101], Loss: 1.0140\n",
            "Epoch [10/10], Step [73/101], Loss: 1.0057\n",
            "Epoch [10/10], Step [74/101], Loss: 0.9647\n",
            "Epoch [10/10], Step [75/101], Loss: 0.8892\n",
            "Epoch [10/10], Step [76/101], Loss: 1.0827\n",
            "Epoch [10/10], Step [77/101], Loss: 1.0927\n",
            "Epoch [10/10], Step [78/101], Loss: 1.6060\n",
            "Epoch [10/10], Step [79/101], Loss: 1.2101\n",
            "Epoch [10/10], Step [80/101], Loss: 1.5315\n",
            "Epoch [10/10], Step [81/101], Loss: 1.0579\n",
            "Epoch [10/10], Step [82/101], Loss: 0.9909\n",
            "Epoch [10/10], Step [83/101], Loss: 1.2853\n",
            "Epoch [10/10], Step [84/101], Loss: 1.0794\n",
            "Epoch [10/10], Step [85/101], Loss: 1.2206\n",
            "Epoch [10/10], Step [86/101], Loss: 1.3292\n",
            "Epoch [10/10], Step [87/101], Loss: 1.0627\n",
            "Epoch [10/10], Step [88/101], Loss: 0.8529\n",
            "Epoch [10/10], Step [89/101], Loss: 1.5584\n",
            "Epoch [10/10], Step [90/101], Loss: 1.0133\n",
            "Epoch [10/10], Step [91/101], Loss: 1.1775\n",
            "Epoch [10/10], Step [92/101], Loss: 1.1423\n",
            "Epoch [10/10], Step [93/101], Loss: 1.0350\n",
            "Epoch [10/10], Step [94/101], Loss: 1.0852\n",
            "Epoch [10/10], Step [95/101], Loss: 0.8538\n",
            "Epoch [10/10], Step [96/101], Loss: 1.0490\n",
            "Epoch [10/10], Step [97/101], Loss: 0.8222\n",
            "Epoch [10/10], Step [98/101], Loss: 1.1738\n",
            "Epoch [10/10], Step [99/101], Loss: 1.0157\n",
            "Epoch [10/10], Step [100/101], Loss: 1.2771\n",
            "Epoch [10/10], Step [101/101], Loss: 1.2168\n",
            "Epoch finished\n",
            "Training finished!\n"
          ]
        }
      ],
      "source": [
        "train(alex_net)"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def test(model):\n",
        "  # Test the model\n",
        "  correct = 0\n",
        "  total = 0\n",
        "  with torch.no_grad():\n",
        "      for data in testloader:\n",
        "          images, labels = data\n",
        "          images = images.to(device)\n",
        "          labels = labels.to(device)\n",
        "          outputs = model(images)\n",
        "          _, predicted = torch.max(outputs.data, 1)\n",
        "          total += labels.size(0)\n",
        "          correct += (predicted == labels).sum().item()\n",
        "\n",
        "  print('Accuracy of the network on the test images: %d %%' % (\n",
        "      100 * correct / total))"
      ],
      "metadata": {
        "id": "Gd92VTiVxeKb"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "ck-lQM3Kv96l",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "c173d061-1d9d-4dd0-c35d-a7be2015c375"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy of the network on the test images: 67 %\n"
          ]
        }
      ],
      "source": [
        "test(alex_net)"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "alex= models.alexnet(pretrained=True)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FV2jYXfyFCaZ",
        "outputId": "69452a52-10f5-4949-9ae9-c3105a9e8dd3"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.\n",
            "  warnings.warn(\n",
            "/usr/local/lib/python3.10/dist-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=AlexNet_Weights.IMAGENET1K_V1`. You can also use `weights=AlexNet_Weights.DEFAULT` to get the most up-to-date weights.\n",
            "  warnings.warn(msg)\n",
            "Downloading: \"https://download.pytorch.org/models/alexnet-owt-7be5be79.pth\" to /root/.cache/torch/hub/checkpoints/alexnet-owt-7be5be79.pth\n",
            "100%|██████████| 233M/233M [00:05<00:00, 43.5MB/s]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "for param in alex.parameters():\n",
        "    param.requires_grad=False"
      ],
      "metadata": {
        "id": "5TTMv_tKFGRW"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "torch.manual_seed(42)\n",
        "\n",
        "alex.classifier=nn.Sequential(  nn.Linear(9216,1024),\n",
        "                                nn.ReLU(),\n",
        "                                nn.Dropout(p=0.5),\n",
        "                                nn.Linear(1024,len(classes)),\n",
        "                                nn.LogSoftmax(dim=1))"
      ],
      "metadata": {
        "id": "j-jKCr70FI-K"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "alex = alex.to(device)\n",
        "alex.cuda()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZtDRopiJRdM6",
        "outputId": "5cec89a9-b99f-4696-f735-e9fddebb652c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "AlexNet(\n",
              "  (features): Sequential(\n",
              "    (0): Conv2d(3, 64, kernel_size=(11, 11), stride=(4, 4), padding=(2, 2))\n",
              "    (1): ReLU(inplace=True)\n",
              "    (2): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
              "    (3): Conv2d(64, 192, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))\n",
              "    (4): ReLU(inplace=True)\n",
              "    (5): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
              "    (6): Conv2d(192, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
              "    (7): ReLU(inplace=True)\n",
              "    (8): Conv2d(384, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
              "    (9): ReLU(inplace=True)\n",
              "    (10): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
              "    (11): ReLU(inplace=True)\n",
              "    (12): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
              "  )\n",
              "  (avgpool): AdaptiveAvgPool2d(output_size=(6, 6))\n",
              "  (classifier): Sequential(\n",
              "    (0): Linear(in_features=9216, out_features=1024, bias=True)\n",
              "    (1): ReLU()\n",
              "    (2): Dropout(p=0.5, inplace=False)\n",
              "    (3): Linear(in_features=1024, out_features=20, bias=True)\n",
              "    (4): LogSoftmax(dim=1)\n",
              "  )\n",
              ")"
            ]
          },
          "metadata": {},
          "execution_count": 67
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "criterion=nn.CrossEntropyLoss()\n",
        "optimizer=torch.optim.Adam(alex.classifier.parameters(),lr=0.001)"
      ],
      "metadata": {
        "id": "1UriQ-LoFXZT"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train(alex)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LiZhCg5JFZtE",
        "outputId": "74d947a7-fc3b-4831-93be-e807e85b463b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch [1/10], Step [1/101], Loss: 3.0322\n",
            "Epoch [1/10], Step [2/101], Loss: 6.4914\n",
            "Epoch [1/10], Step [3/101], Loss: 8.8729\n",
            "Epoch [1/10], Step [4/101], Loss: 5.8030\n",
            "Epoch [1/10], Step [5/101], Loss: 6.9949\n",
            "Epoch [1/10], Step [6/101], Loss: 4.7696\n",
            "Epoch [1/10], Step [7/101], Loss: 7.4249\n",
            "Epoch [1/10], Step [8/101], Loss: 6.3890\n",
            "Epoch [1/10], Step [9/101], Loss: 6.4613\n",
            "Epoch [1/10], Step [10/101], Loss: 3.4606\n",
            "Epoch [1/10], Step [11/101], Loss: 4.6425\n",
            "Epoch [1/10], Step [12/101], Loss: 3.3393\n",
            "Epoch [1/10], Step [13/101], Loss: 4.0206\n",
            "Epoch [1/10], Step [14/101], Loss: 1.5981\n",
            "Epoch [1/10], Step [15/101], Loss: 3.0197\n",
            "Epoch [1/10], Step [16/101], Loss: 2.4464\n",
            "Epoch [1/10], Step [17/101], Loss: 1.9869\n",
            "Epoch [1/10], Step [18/101], Loss: 1.4736\n",
            "Epoch [1/10], Step [19/101], Loss: 2.3570\n",
            "Epoch [1/10], Step [20/101], Loss: 1.8464\n",
            "Epoch [1/10], Step [21/101], Loss: 1.4445\n",
            "Epoch [1/10], Step [22/101], Loss: 0.9376\n",
            "Epoch [1/10], Step [23/101], Loss: 1.2779\n",
            "Epoch [1/10], Step [24/101], Loss: 1.1445\n",
            "Epoch [1/10], Step [25/101], Loss: 1.0605\n",
            "Epoch [1/10], Step [26/101], Loss: 1.1401\n",
            "Epoch [1/10], Step [27/101], Loss: 0.9361\n",
            "Epoch [1/10], Step [28/101], Loss: 0.8400\n",
            "Epoch [1/10], Step [29/101], Loss: 1.2080\n",
            "Epoch [1/10], Step [30/101], Loss: 0.8499\n",
            "Epoch [1/10], Step [31/101], Loss: 0.7652\n",
            "Epoch [1/10], Step [32/101], Loss: 1.6700\n",
            "Epoch [1/10], Step [33/101], Loss: 1.1147\n",
            "Epoch [1/10], Step [34/101], Loss: 1.2587\n",
            "Epoch [1/10], Step [35/101], Loss: 0.7781\n",
            "Epoch [1/10], Step [36/101], Loss: 0.8605\n",
            "Epoch [1/10], Step [37/101], Loss: 1.1841\n",
            "Epoch [1/10], Step [38/101], Loss: 0.6167\n",
            "Epoch [1/10], Step [39/101], Loss: 0.9879\n",
            "Epoch [1/10], Step [40/101], Loss: 1.1748\n",
            "Epoch [1/10], Step [41/101], Loss: 0.5462\n",
            "Epoch [1/10], Step [42/101], Loss: 0.4129\n",
            "Epoch [1/10], Step [43/101], Loss: 0.7994\n",
            "Epoch [1/10], Step [44/101], Loss: 0.5682\n",
            "Epoch [1/10], Step [45/101], Loss: 0.7836\n",
            "Epoch [1/10], Step [46/101], Loss: 0.3491\n",
            "Epoch [1/10], Step [47/101], Loss: 0.4441\n",
            "Epoch [1/10], Step [48/101], Loss: 0.8093\n",
            "Epoch [1/10], Step [49/101], Loss: 0.3647\n",
            "Epoch [1/10], Step [50/101], Loss: 0.6392\n",
            "Epoch [1/10], Step [51/101], Loss: 0.7202\n",
            "Epoch [1/10], Step [52/101], Loss: 0.4535\n",
            "Epoch [1/10], Step [53/101], Loss: 0.5601\n",
            "Epoch [1/10], Step [54/101], Loss: 0.6482\n",
            "Epoch [1/10], Step [55/101], Loss: 0.5804\n",
            "Epoch [1/10], Step [56/101], Loss: 1.1337\n",
            "Epoch [1/10], Step [57/101], Loss: 0.5228\n",
            "Epoch [1/10], Step [58/101], Loss: 0.5226\n",
            "Epoch [1/10], Step [59/101], Loss: 0.5931\n",
            "Epoch [1/10], Step [60/101], Loss: 0.7161\n",
            "Epoch [1/10], Step [61/101], Loss: 0.6112\n",
            "Epoch [1/10], Step [62/101], Loss: 0.6567\n",
            "Epoch [1/10], Step [63/101], Loss: 0.5602\n",
            "Epoch [1/10], Step [64/101], Loss: 0.3084\n",
            "Epoch [1/10], Step [65/101], Loss: 0.5328\n",
            "Epoch [1/10], Step [66/101], Loss: 0.3464\n",
            "Epoch [1/10], Step [67/101], Loss: 0.4694\n",
            "Epoch [1/10], Step [68/101], Loss: 0.8996\n",
            "Epoch [1/10], Step [69/101], Loss: 0.6917\n",
            "Epoch [1/10], Step [70/101], Loss: 0.8193\n",
            "Epoch [1/10], Step [71/101], Loss: 0.4757\n",
            "Epoch [1/10], Step [72/101], Loss: 0.3707\n",
            "Epoch [1/10], Step [73/101], Loss: 0.4097\n",
            "Epoch [1/10], Step [74/101], Loss: 0.6408\n",
            "Epoch [1/10], Step [75/101], Loss: 0.4345\n",
            "Epoch [1/10], Step [76/101], Loss: 0.6215\n",
            "Epoch [1/10], Step [77/101], Loss: 0.2860\n",
            "Epoch [1/10], Step [78/101], Loss: 0.1929\n",
            "Epoch [1/10], Step [79/101], Loss: 0.7180\n",
            "Epoch [1/10], Step [80/101], Loss: 0.7929\n",
            "Epoch [1/10], Step [81/101], Loss: 0.8496\n",
            "Epoch [1/10], Step [82/101], Loss: 1.4096\n",
            "Epoch [1/10], Step [83/101], Loss: 0.7159\n",
            "Epoch [1/10], Step [84/101], Loss: 0.9094\n",
            "Epoch [1/10], Step [85/101], Loss: 0.4792\n",
            "Epoch [1/10], Step [86/101], Loss: 0.2766\n",
            "Epoch [1/10], Step [87/101], Loss: 0.7401\n",
            "Epoch [1/10], Step [88/101], Loss: 0.4000\n",
            "Epoch [1/10], Step [89/101], Loss: 0.5442\n",
            "Epoch [1/10], Step [90/101], Loss: 0.8422\n",
            "Epoch [1/10], Step [91/101], Loss: 0.4179\n",
            "Epoch [1/10], Step [92/101], Loss: 0.8467\n",
            "Epoch [1/10], Step [93/101], Loss: 0.7314\n",
            "Epoch [1/10], Step [94/101], Loss: 0.3169\n",
            "Epoch [1/10], Step [95/101], Loss: 0.4321\n",
            "Epoch [1/10], Step [96/101], Loss: 0.4147\n",
            "Epoch [1/10], Step [97/101], Loss: 0.9261\n",
            "Epoch [1/10], Step [98/101], Loss: 0.5776\n",
            "Epoch [1/10], Step [99/101], Loss: 0.3604\n",
            "Epoch [1/10], Step [100/101], Loss: 0.3472\n",
            "Epoch [1/10], Step [101/101], Loss: 1.4892\n",
            "Epoch finished\n",
            "Epoch [2/10], Step [1/101], Loss: 0.1734\n",
            "Epoch [2/10], Step [2/101], Loss: 0.5388\n",
            "Epoch [2/10], Step [3/101], Loss: 0.5621\n",
            "Epoch [2/10], Step [4/101], Loss: 0.4073\n",
            "Epoch [2/10], Step [5/101], Loss: 0.1497\n",
            "Epoch [2/10], Step [6/101], Loss: 0.3985\n",
            "Epoch [2/10], Step [7/101], Loss: 0.1953\n",
            "Epoch [2/10], Step [8/101], Loss: 0.3157\n",
            "Epoch [2/10], Step [9/101], Loss: 0.2280\n",
            "Epoch [2/10], Step [10/101], Loss: 0.1811\n",
            "Epoch [2/10], Step [11/101], Loss: 0.4312\n",
            "Epoch [2/10], Step [12/101], Loss: 0.6762\n",
            "Epoch [2/10], Step [13/101], Loss: 0.2931\n",
            "Epoch [2/10], Step [14/101], Loss: 0.6026\n",
            "Epoch [2/10], Step [15/101], Loss: 0.5683\n",
            "Epoch [2/10], Step [16/101], Loss: 0.9311\n",
            "Epoch [2/10], Step [17/101], Loss: 0.6069\n",
            "Epoch [2/10], Step [18/101], Loss: 0.3505\n",
            "Epoch [2/10], Step [19/101], Loss: 0.4293\n",
            "Epoch [2/10], Step [20/101], Loss: 0.5261\n",
            "Epoch [2/10], Step [21/101], Loss: 0.3757\n",
            "Epoch [2/10], Step [22/101], Loss: 0.4620\n",
            "Epoch [2/10], Step [23/101], Loss: 0.5299\n",
            "Epoch [2/10], Step [24/101], Loss: 0.4375\n",
            "Epoch [2/10], Step [25/101], Loss: 0.3705\n",
            "Epoch [2/10], Step [26/101], Loss: 0.6937\n",
            "Epoch [2/10], Step [27/101], Loss: 0.3820\n",
            "Epoch [2/10], Step [28/101], Loss: 0.2597\n",
            "Epoch [2/10], Step [29/101], Loss: 0.2465\n",
            "Epoch [2/10], Step [30/101], Loss: 0.3593\n",
            "Epoch [2/10], Step [31/101], Loss: 0.3111\n",
            "Epoch [2/10], Step [32/101], Loss: 0.6906\n",
            "Epoch [2/10], Step [33/101], Loss: 0.3008\n",
            "Epoch [2/10], Step [34/101], Loss: 0.3712\n",
            "Epoch [2/10], Step [35/101], Loss: 0.1908\n",
            "Epoch [2/10], Step [36/101], Loss: 0.1300\n",
            "Epoch [2/10], Step [37/101], Loss: 0.6328\n",
            "Epoch [2/10], Step [38/101], Loss: 0.4761\n",
            "Epoch [2/10], Step [39/101], Loss: 0.3751\n",
            "Epoch [2/10], Step [40/101], Loss: 0.3186\n",
            "Epoch [2/10], Step [41/101], Loss: 0.5110\n",
            "Epoch [2/10], Step [42/101], Loss: 0.3002\n",
            "Epoch [2/10], Step [43/101], Loss: 0.1818\n",
            "Epoch [2/10], Step [44/101], Loss: 0.5291\n",
            "Epoch [2/10], Step [45/101], Loss: 0.3027\n",
            "Epoch [2/10], Step [46/101], Loss: 0.0832\n",
            "Epoch [2/10], Step [47/101], Loss: 0.4549\n",
            "Epoch [2/10], Step [48/101], Loss: 0.2464\n",
            "Epoch [2/10], Step [49/101], Loss: 1.0540\n",
            "Epoch [2/10], Step [50/101], Loss: 0.4829\n",
            "Epoch [2/10], Step [51/101], Loss: 0.8422\n",
            "Epoch [2/10], Step [52/101], Loss: 0.7617\n",
            "Epoch [2/10], Step [53/101], Loss: 0.4111\n",
            "Epoch [2/10], Step [54/101], Loss: 0.3683\n",
            "Epoch [2/10], Step [55/101], Loss: 0.2137\n",
            "Epoch [2/10], Step [56/101], Loss: 0.2908\n",
            "Epoch [2/10], Step [57/101], Loss: 0.5680\n",
            "Epoch [2/10], Step [58/101], Loss: 0.4609\n",
            "Epoch [2/10], Step [59/101], Loss: 0.1871\n",
            "Epoch [2/10], Step [60/101], Loss: 0.4907\n",
            "Epoch [2/10], Step [61/101], Loss: 0.2863\n",
            "Epoch [2/10], Step [62/101], Loss: 0.2326\n",
            "Epoch [2/10], Step [63/101], Loss: 0.2173\n",
            "Epoch [2/10], Step [64/101], Loss: 0.2471\n",
            "Epoch [2/10], Step [65/101], Loss: 0.2602\n",
            "Epoch [2/10], Step [66/101], Loss: 0.6572\n",
            "Epoch [2/10], Step [67/101], Loss: 0.1391\n",
            "Epoch [2/10], Step [68/101], Loss: 0.4714\n",
            "Epoch [2/10], Step [69/101], Loss: 0.4569\n",
            "Epoch [2/10], Step [70/101], Loss: 0.0751\n",
            "Epoch [2/10], Step [71/101], Loss: 0.4905\n",
            "Epoch [2/10], Step [72/101], Loss: 0.3205\n",
            "Epoch [2/10], Step [73/101], Loss: 0.3327\n",
            "Epoch [2/10], Step [74/101], Loss: 0.4443\n",
            "Epoch [2/10], Step [75/101], Loss: 0.3183\n",
            "Epoch [2/10], Step [76/101], Loss: 0.4411\n",
            "Epoch [2/10], Step [77/101], Loss: 0.4269\n",
            "Epoch [2/10], Step [78/101], Loss: 0.3684\n",
            "Epoch [2/10], Step [79/101], Loss: 0.3958\n",
            "Epoch [2/10], Step [80/101], Loss: 0.3650\n",
            "Epoch [2/10], Step [81/101], Loss: 0.3846\n",
            "Epoch [2/10], Step [82/101], Loss: 0.4386\n",
            "Epoch [2/10], Step [83/101], Loss: 0.3819\n",
            "Epoch [2/10], Step [84/101], Loss: 0.7616\n",
            "Epoch [2/10], Step [85/101], Loss: 0.4947\n",
            "Epoch [2/10], Step [86/101], Loss: 0.2016\n",
            "Epoch [2/10], Step [87/101], Loss: 0.4471\n",
            "Epoch [2/10], Step [88/101], Loss: 0.3050\n",
            "Epoch [2/10], Step [89/101], Loss: 0.3409\n",
            "Epoch [2/10], Step [90/101], Loss: 0.2444\n",
            "Epoch [2/10], Step [91/101], Loss: 0.4836\n",
            "Epoch [2/10], Step [92/101], Loss: 0.2540\n",
            "Epoch [2/10], Step [93/101], Loss: 0.1299\n",
            "Epoch [2/10], Step [94/101], Loss: 0.6160\n",
            "Epoch [2/10], Step [95/101], Loss: 0.3408\n",
            "Epoch [2/10], Step [96/101], Loss: 0.6430\n",
            "Epoch [2/10], Step [97/101], Loss: 0.3422\n",
            "Epoch [2/10], Step [98/101], Loss: 0.0501\n",
            "Epoch [2/10], Step [99/101], Loss: 0.6742\n",
            "Epoch [2/10], Step [100/101], Loss: 0.6204\n",
            "Epoch [2/10], Step [101/101], Loss: 0.6722\n",
            "Epoch finished\n",
            "Epoch [3/10], Step [1/101], Loss: 0.3583\n",
            "Epoch [3/10], Step [2/101], Loss: 0.1834\n",
            "Epoch [3/10], Step [3/101], Loss: 0.3796\n",
            "Epoch [3/10], Step [4/101], Loss: 0.6216\n",
            "Epoch [3/10], Step [5/101], Loss: 0.3891\n",
            "Epoch [3/10], Step [6/101], Loss: 0.6333\n",
            "Epoch [3/10], Step [7/101], Loss: 0.4091\n",
            "Epoch [3/10], Step [8/101], Loss: 0.0948\n",
            "Epoch [3/10], Step [9/101], Loss: 0.2575\n",
            "Epoch [3/10], Step [10/101], Loss: 0.3840\n",
            "Epoch [3/10], Step [11/101], Loss: 0.2829\n",
            "Epoch [3/10], Step [12/101], Loss: 0.1821\n",
            "Epoch [3/10], Step [13/101], Loss: 0.2913\n",
            "Epoch [3/10], Step [14/101], Loss: 0.4122\n",
            "Epoch [3/10], Step [15/101], Loss: 0.4107\n",
            "Epoch [3/10], Step [16/101], Loss: 0.4027\n",
            "Epoch [3/10], Step [17/101], Loss: 0.3037\n",
            "Epoch [3/10], Step [18/101], Loss: 0.2507\n",
            "Epoch [3/10], Step [19/101], Loss: 0.3700\n",
            "Epoch [3/10], Step [20/101], Loss: 0.6599\n",
            "Epoch [3/10], Step [21/101], Loss: 0.2455\n",
            "Epoch [3/10], Step [22/101], Loss: 0.5967\n",
            "Epoch [3/10], Step [23/101], Loss: 0.4031\n",
            "Epoch [3/10], Step [24/101], Loss: 0.1520\n",
            "Epoch [3/10], Step [25/101], Loss: 0.6937\n",
            "Epoch [3/10], Step [26/101], Loss: 0.3786\n",
            "Epoch [3/10], Step [27/101], Loss: 0.1939\n",
            "Epoch [3/10], Step [28/101], Loss: 0.6572\n",
            "Epoch [3/10], Step [29/101], Loss: 0.5398\n",
            "Epoch [3/10], Step [30/101], Loss: 0.0454\n",
            "Epoch [3/10], Step [31/101], Loss: 0.2965\n",
            "Epoch [3/10], Step [32/101], Loss: 0.5125\n",
            "Epoch [3/10], Step [33/101], Loss: 0.2584\n",
            "Epoch [3/10], Step [34/101], Loss: 0.7097\n",
            "Epoch [3/10], Step [35/101], Loss: 0.2862\n",
            "Epoch [3/10], Step [36/101], Loss: 0.0783\n",
            "Epoch [3/10], Step [37/101], Loss: 0.1465\n",
            "Epoch [3/10], Step [38/101], Loss: 0.3199\n",
            "Epoch [3/10], Step [39/101], Loss: 0.2924\n",
            "Epoch [3/10], Step [40/101], Loss: 0.1396\n",
            "Epoch [3/10], Step [41/101], Loss: 0.1475\n",
            "Epoch [3/10], Step [42/101], Loss: 0.3524\n",
            "Epoch [3/10], Step [43/101], Loss: 0.1379\n",
            "Epoch [3/10], Step [44/101], Loss: 0.2029\n",
            "Epoch [3/10], Step [45/101], Loss: 0.3696\n",
            "Epoch [3/10], Step [46/101], Loss: 0.0312\n",
            "Epoch [3/10], Step [47/101], Loss: 0.3593\n",
            "Epoch [3/10], Step [48/101], Loss: 0.5864\n",
            "Epoch [3/10], Step [49/101], Loss: 0.1246\n",
            "Epoch [3/10], Step [50/101], Loss: 0.5513\n",
            "Epoch [3/10], Step [51/101], Loss: 0.0820\n",
            "Epoch [3/10], Step [52/101], Loss: 0.6375\n",
            "Epoch [3/10], Step [53/101], Loss: 0.0642\n",
            "Epoch [3/10], Step [54/101], Loss: 0.2935\n",
            "Epoch [3/10], Step [55/101], Loss: 0.6188\n",
            "Epoch [3/10], Step [56/101], Loss: 0.3143\n",
            "Epoch [3/10], Step [57/101], Loss: 0.3084\n",
            "Epoch [3/10], Step [58/101], Loss: 0.2384\n",
            "Epoch [3/10], Step [59/101], Loss: 0.5398\n",
            "Epoch [3/10], Step [60/101], Loss: 0.4516\n",
            "Epoch [3/10], Step [61/101], Loss: 0.6054\n",
            "Epoch [3/10], Step [62/101], Loss: 0.2142\n",
            "Epoch [3/10], Step [63/101], Loss: 0.3191\n",
            "Epoch [3/10], Step [64/101], Loss: 0.3761\n",
            "Epoch [3/10], Step [65/101], Loss: 0.0985\n",
            "Epoch [3/10], Step [66/101], Loss: 0.1387\n",
            "Epoch [3/10], Step [67/101], Loss: 0.2521\n",
            "Epoch [3/10], Step [68/101], Loss: 0.3561\n",
            "Epoch [3/10], Step [69/101], Loss: 0.9842\n",
            "Epoch [3/10], Step [70/101], Loss: 0.1926\n",
            "Epoch [3/10], Step [71/101], Loss: 0.6172\n",
            "Epoch [3/10], Step [72/101], Loss: 0.5618\n",
            "Epoch [3/10], Step [73/101], Loss: 0.2870\n",
            "Epoch [3/10], Step [74/101], Loss: 0.2524\n",
            "Epoch [3/10], Step [75/101], Loss: 0.2603\n",
            "Epoch [3/10], Step [76/101], Loss: 0.0892\n",
            "Epoch [3/10], Step [77/101], Loss: 0.5298\n",
            "Epoch [3/10], Step [78/101], Loss: 0.1752\n",
            "Epoch [3/10], Step [79/101], Loss: 0.4452\n",
            "Epoch [3/10], Step [80/101], Loss: 0.3006\n",
            "Epoch [3/10], Step [81/101], Loss: 0.6559\n",
            "Epoch [3/10], Step [82/101], Loss: 0.3898\n",
            "Epoch [3/10], Step [83/101], Loss: 0.4422\n",
            "Epoch [3/10], Step [84/101], Loss: 0.1898\n",
            "Epoch [3/10], Step [85/101], Loss: 0.3013\n",
            "Epoch [3/10], Step [86/101], Loss: 0.2214\n",
            "Epoch [3/10], Step [87/101], Loss: 0.3005\n",
            "Epoch [3/10], Step [88/101], Loss: 0.4317\n",
            "Epoch [3/10], Step [89/101], Loss: 0.9111\n",
            "Epoch [3/10], Step [90/101], Loss: 0.3062\n",
            "Epoch [3/10], Step [91/101], Loss: 0.3166\n",
            "Epoch [3/10], Step [92/101], Loss: 0.3147\n",
            "Epoch [3/10], Step [93/101], Loss: 0.3468\n",
            "Epoch [3/10], Step [94/101], Loss: 0.1636\n",
            "Epoch [3/10], Step [95/101], Loss: 0.3570\n",
            "Epoch [3/10], Step [96/101], Loss: 0.2849\n",
            "Epoch [3/10], Step [97/101], Loss: 0.2093\n",
            "Epoch [3/10], Step [98/101], Loss: 0.3895\n",
            "Epoch [3/10], Step [99/101], Loss: 0.4308\n",
            "Epoch [3/10], Step [100/101], Loss: 0.4616\n",
            "Epoch [3/10], Step [101/101], Loss: 0.1868\n",
            "Epoch finished\n",
            "Epoch [4/10], Step [1/101], Loss: 0.2897\n",
            "Epoch [4/10], Step [2/101], Loss: 0.3630\n",
            "Epoch [4/10], Step [3/101], Loss: 0.5867\n",
            "Epoch [4/10], Step [4/101], Loss: 0.4336\n",
            "Epoch [4/10], Step [5/101], Loss: 0.1434\n",
            "Epoch [4/10], Step [6/101], Loss: 0.1618\n",
            "Epoch [4/10], Step [7/101], Loss: 0.4076\n",
            "Epoch [4/10], Step [8/101], Loss: 0.2343\n",
            "Epoch [4/10], Step [9/101], Loss: 0.3743\n",
            "Epoch [4/10], Step [10/101], Loss: 0.2646\n",
            "Epoch [4/10], Step [11/101], Loss: 0.6839\n",
            "Epoch [4/10], Step [12/101], Loss: 0.1728\n",
            "Epoch [4/10], Step [13/101], Loss: 0.5491\n",
            "Epoch [4/10], Step [14/101], Loss: 0.5431\n",
            "Epoch [4/10], Step [15/101], Loss: 0.0535\n",
            "Epoch [4/10], Step [16/101], Loss: 0.2031\n",
            "Epoch [4/10], Step [17/101], Loss: 0.4300\n",
            "Epoch [4/10], Step [18/101], Loss: 0.2626\n",
            "Epoch [4/10], Step [19/101], Loss: 0.2273\n",
            "Epoch [4/10], Step [20/101], Loss: 0.2654\n",
            "Epoch [4/10], Step [21/101], Loss: 0.2752\n",
            "Epoch [4/10], Step [22/101], Loss: 0.6353\n",
            "Epoch [4/10], Step [23/101], Loss: 0.5205\n",
            "Epoch [4/10], Step [24/101], Loss: 0.5691\n",
            "Epoch [4/10], Step [25/101], Loss: 0.4909\n",
            "Epoch [4/10], Step [26/101], Loss: 0.3123\n",
            "Epoch [4/10], Step [27/101], Loss: 0.5475\n",
            "Epoch [4/10], Step [28/101], Loss: 0.1599\n",
            "Epoch [4/10], Step [29/101], Loss: 0.1758\n",
            "Epoch [4/10], Step [30/101], Loss: 0.3656\n",
            "Epoch [4/10], Step [31/101], Loss: 0.2878\n",
            "Epoch [4/10], Step [32/101], Loss: 0.2384\n",
            "Epoch [4/10], Step [33/101], Loss: 0.0957\n",
            "Epoch [4/10], Step [34/101], Loss: 0.0617\n",
            "Epoch [4/10], Step [35/101], Loss: 0.4344\n",
            "Epoch [4/10], Step [36/101], Loss: 0.3144\n",
            "Epoch [4/10], Step [37/101], Loss: 0.1193\n",
            "Epoch [4/10], Step [38/101], Loss: 0.6041\n",
            "Epoch [4/10], Step [39/101], Loss: 0.2969\n",
            "Epoch [4/10], Step [40/101], Loss: 0.1747\n",
            "Epoch [4/10], Step [41/101], Loss: 0.0324\n",
            "Epoch [4/10], Step [42/101], Loss: 0.1065\n",
            "Epoch [4/10], Step [43/101], Loss: 0.4523\n",
            "Epoch [4/10], Step [44/101], Loss: 0.2307\n",
            "Epoch [4/10], Step [45/101], Loss: 0.1789\n",
            "Epoch [4/10], Step [46/101], Loss: 0.2410\n",
            "Epoch [4/10], Step [47/101], Loss: 0.3427\n",
            "Epoch [4/10], Step [48/101], Loss: 0.1011\n",
            "Epoch [4/10], Step [49/101], Loss: 0.1106\n",
            "Epoch [4/10], Step [50/101], Loss: 0.1338\n",
            "Epoch [4/10], Step [51/101], Loss: 0.1908\n",
            "Epoch [4/10], Step [52/101], Loss: 0.2365\n",
            "Epoch [4/10], Step [53/101], Loss: 0.6854\n",
            "Epoch [4/10], Step [54/101], Loss: 0.1008\n",
            "Epoch [4/10], Step [55/101], Loss: 0.3735\n",
            "Epoch [4/10], Step [56/101], Loss: 0.0915\n",
            "Epoch [4/10], Step [57/101], Loss: 0.1448\n",
            "Epoch [4/10], Step [58/101], Loss: 0.1966\n",
            "Epoch [4/10], Step [59/101], Loss: 0.2984\n",
            "Epoch [4/10], Step [60/101], Loss: 0.5147\n",
            "Epoch [4/10], Step [61/101], Loss: 0.1481\n",
            "Epoch [4/10], Step [62/101], Loss: 0.0787\n",
            "Epoch [4/10], Step [63/101], Loss: 0.1286\n",
            "Epoch [4/10], Step [64/101], Loss: 0.2986\n",
            "Epoch [4/10], Step [65/101], Loss: 0.2117\n",
            "Epoch [4/10], Step [66/101], Loss: 0.6668\n",
            "Epoch [4/10], Step [67/101], Loss: 0.3452\n",
            "Epoch [4/10], Step [68/101], Loss: 0.6853\n",
            "Epoch [4/10], Step [69/101], Loss: 0.3563\n",
            "Epoch [4/10], Step [70/101], Loss: 0.2532\n",
            "Epoch [4/10], Step [71/101], Loss: 0.4720\n",
            "Epoch [4/10], Step [72/101], Loss: 0.3990\n",
            "Epoch [4/10], Step [73/101], Loss: 0.3433\n",
            "Epoch [4/10], Step [74/101], Loss: 0.4116\n",
            "Epoch [4/10], Step [75/101], Loss: 0.6837\n",
            "Epoch [4/10], Step [76/101], Loss: 0.6444\n",
            "Epoch [4/10], Step [77/101], Loss: 0.0709\n",
            "Epoch [4/10], Step [78/101], Loss: 0.2397\n",
            "Epoch [4/10], Step [79/101], Loss: 0.4321\n",
            "Epoch [4/10], Step [80/101], Loss: 0.6060\n",
            "Epoch [4/10], Step [81/101], Loss: 0.3914\n",
            "Epoch [4/10], Step [82/101], Loss: 0.4339\n",
            "Epoch [4/10], Step [83/101], Loss: 0.4082\n",
            "Epoch [4/10], Step [84/101], Loss: 0.3385\n",
            "Epoch [4/10], Step [85/101], Loss: 0.0529\n",
            "Epoch [4/10], Step [86/101], Loss: 0.3081\n",
            "Epoch [4/10], Step [87/101], Loss: 0.0249\n",
            "Epoch [4/10], Step [88/101], Loss: 0.4033\n",
            "Epoch [4/10], Step [89/101], Loss: 0.5909\n",
            "Epoch [4/10], Step [90/101], Loss: 0.2596\n",
            "Epoch [4/10], Step [91/101], Loss: 0.3824\n",
            "Epoch [4/10], Step [92/101], Loss: 0.2673\n",
            "Epoch [4/10], Step [93/101], Loss: 0.2753\n",
            "Epoch [4/10], Step [94/101], Loss: 0.2967\n",
            "Epoch [4/10], Step [95/101], Loss: 0.3704\n",
            "Epoch [4/10], Step [96/101], Loss: 0.1444\n",
            "Epoch [4/10], Step [97/101], Loss: 0.2272\n",
            "Epoch [4/10], Step [98/101], Loss: 0.3841\n",
            "Epoch [4/10], Step [99/101], Loss: 0.1318\n",
            "Epoch [4/10], Step [100/101], Loss: 0.2995\n",
            "Epoch [4/10], Step [101/101], Loss: 0.7174\n",
            "Epoch finished\n",
            "Epoch [5/10], Step [1/101], Loss: 0.3826\n",
            "Epoch [5/10], Step [2/101], Loss: 0.3596\n",
            "Epoch [5/10], Step [3/101], Loss: 0.2649\n",
            "Epoch [5/10], Step [4/101], Loss: 0.4342\n",
            "Epoch [5/10], Step [5/101], Loss: 1.1481\n",
            "Epoch [5/10], Step [6/101], Loss: 0.4687\n",
            "Epoch [5/10], Step [7/101], Loss: 0.2454\n",
            "Epoch [5/10], Step [8/101], Loss: 0.4648\n",
            "Epoch [5/10], Step [9/101], Loss: 0.2842\n",
            "Epoch [5/10], Step [10/101], Loss: 0.5369\n",
            "Epoch [5/10], Step [11/101], Loss: 0.2218\n",
            "Epoch [5/10], Step [12/101], Loss: 0.6166\n",
            "Epoch [5/10], Step [13/101], Loss: 0.2912\n",
            "Epoch [5/10], Step [14/101], Loss: 0.1443\n",
            "Epoch [5/10], Step [15/101], Loss: 0.0500\n",
            "Epoch [5/10], Step [16/101], Loss: 0.5142\n",
            "Epoch [5/10], Step [17/101], Loss: 0.3337\n",
            "Epoch [5/10], Step [18/101], Loss: 0.2587\n",
            "Epoch [5/10], Step [19/101], Loss: 0.2490\n",
            "Epoch [5/10], Step [20/101], Loss: 0.9150\n",
            "Epoch [5/10], Step [21/101], Loss: 0.3861\n",
            "Epoch [5/10], Step [22/101], Loss: 0.1466\n",
            "Epoch [5/10], Step [23/101], Loss: 0.3500\n",
            "Epoch [5/10], Step [24/101], Loss: 0.1307\n",
            "Epoch [5/10], Step [25/101], Loss: 0.0854\n",
            "Epoch [5/10], Step [26/101], Loss: 0.3453\n",
            "Epoch [5/10], Step [27/101], Loss: 0.6008\n",
            "Epoch [5/10], Step [28/101], Loss: 0.3682\n",
            "Epoch [5/10], Step [29/101], Loss: 0.1778\n",
            "Epoch [5/10], Step [30/101], Loss: 0.3510\n",
            "Epoch [5/10], Step [31/101], Loss: 0.1767\n",
            "Epoch [5/10], Step [32/101], Loss: 0.8172\n",
            "Epoch [5/10], Step [33/101], Loss: 0.2620\n",
            "Epoch [5/10], Step [34/101], Loss: 0.1633\n",
            "Epoch [5/10], Step [35/101], Loss: 0.2314\n",
            "Epoch [5/10], Step [36/101], Loss: 0.3444\n",
            "Epoch [5/10], Step [37/101], Loss: 1.0548\n",
            "Epoch [5/10], Step [38/101], Loss: 0.5559\n",
            "Epoch [5/10], Step [39/101], Loss: 0.3631\n",
            "Epoch [5/10], Step [40/101], Loss: 0.2553\n",
            "Epoch [5/10], Step [41/101], Loss: 0.1361\n",
            "Epoch [5/10], Step [42/101], Loss: 0.2523\n",
            "Epoch [5/10], Step [43/101], Loss: 0.1965\n",
            "Epoch [5/10], Step [44/101], Loss: 0.2774\n",
            "Epoch [5/10], Step [45/101], Loss: 0.1711\n",
            "Epoch [5/10], Step [46/101], Loss: 0.6062\n",
            "Epoch [5/10], Step [47/101], Loss: 0.2578\n",
            "Epoch [5/10], Step [48/101], Loss: 0.1092\n",
            "Epoch [5/10], Step [49/101], Loss: 0.0529\n",
            "Epoch [5/10], Step [50/101], Loss: 0.1578\n",
            "Epoch [5/10], Step [51/101], Loss: 0.2236\n",
            "Epoch [5/10], Step [52/101], Loss: 0.3617\n",
            "Epoch [5/10], Step [53/101], Loss: 0.0991\n",
            "Epoch [5/10], Step [54/101], Loss: 0.5832\n",
            "Epoch [5/10], Step [55/101], Loss: 0.3670\n",
            "Epoch [5/10], Step [56/101], Loss: 0.8054\n",
            "Epoch [5/10], Step [57/101], Loss: 0.3623\n",
            "Epoch [5/10], Step [58/101], Loss: 0.1602\n",
            "Epoch [5/10], Step [59/101], Loss: 0.1584\n",
            "Epoch [5/10], Step [60/101], Loss: 0.3695\n",
            "Epoch [5/10], Step [61/101], Loss: 0.0262\n",
            "Epoch [5/10], Step [62/101], Loss: 0.0237\n",
            "Epoch [5/10], Step [63/101], Loss: 0.1774\n",
            "Epoch [5/10], Step [64/101], Loss: 0.1437\n",
            "Epoch [5/10], Step [65/101], Loss: 0.3228\n",
            "Epoch [5/10], Step [66/101], Loss: 0.4230\n",
            "Epoch [5/10], Step [67/101], Loss: 0.1283\n",
            "Epoch [5/10], Step [68/101], Loss: 0.1440\n",
            "Epoch [5/10], Step [69/101], Loss: 0.3950\n",
            "Epoch [5/10], Step [70/101], Loss: 0.3694\n",
            "Epoch [5/10], Step [71/101], Loss: 0.4798\n",
            "Epoch [5/10], Step [72/101], Loss: 0.2499\n",
            "Epoch [5/10], Step [73/101], Loss: 0.2986\n",
            "Epoch [5/10], Step [74/101], Loss: 0.3419\n",
            "Epoch [5/10], Step [75/101], Loss: 0.0558\n",
            "Epoch [5/10], Step [76/101], Loss: 0.2649\n",
            "Epoch [5/10], Step [77/101], Loss: 0.3529\n",
            "Epoch [5/10], Step [78/101], Loss: 0.4910\n",
            "Epoch [5/10], Step [79/101], Loss: 0.0290\n",
            "Epoch [5/10], Step [80/101], Loss: 0.3994\n",
            "Epoch [5/10], Step [81/101], Loss: 0.7584\n",
            "Epoch [5/10], Step [82/101], Loss: 0.4431\n",
            "Epoch [5/10], Step [83/101], Loss: 0.4006\n",
            "Epoch [5/10], Step [84/101], Loss: 0.1784\n",
            "Epoch [5/10], Step [85/101], Loss: 0.1095\n",
            "Epoch [5/10], Step [86/101], Loss: 0.2076\n",
            "Epoch [5/10], Step [87/101], Loss: 0.5153\n",
            "Epoch [5/10], Step [88/101], Loss: 0.3842\n",
            "Epoch [5/10], Step [89/101], Loss: 0.0360\n",
            "Epoch [5/10], Step [90/101], Loss: 0.3075\n",
            "Epoch [5/10], Step [91/101], Loss: 0.6259\n",
            "Epoch [5/10], Step [92/101], Loss: 0.1806\n",
            "Epoch [5/10], Step [93/101], Loss: 0.2564\n",
            "Epoch [5/10], Step [94/101], Loss: 0.1008\n",
            "Epoch [5/10], Step [95/101], Loss: 0.3818\n",
            "Epoch [5/10], Step [96/101], Loss: 0.0837\n",
            "Epoch [5/10], Step [97/101], Loss: 0.0535\n",
            "Epoch [5/10], Step [98/101], Loss: 0.2062\n",
            "Epoch [5/10], Step [99/101], Loss: 0.2000\n",
            "Epoch [5/10], Step [100/101], Loss: 0.4191\n",
            "Epoch [5/10], Step [101/101], Loss: 0.5082\n",
            "Epoch finished\n",
            "Epoch [6/10], Step [1/101], Loss: 0.3955\n",
            "Epoch [6/10], Step [2/101], Loss: 0.5888\n",
            "Epoch [6/10], Step [3/101], Loss: 0.1399\n",
            "Epoch [6/10], Step [4/101], Loss: 0.5851\n",
            "Epoch [6/10], Step [5/101], Loss: 0.6157\n",
            "Epoch [6/10], Step [6/101], Loss: 0.3192\n",
            "Epoch [6/10], Step [7/101], Loss: 0.0852\n",
            "Epoch [6/10], Step [8/101], Loss: 0.2851\n",
            "Epoch [6/10], Step [9/101], Loss: 0.0987\n",
            "Epoch [6/10], Step [10/101], Loss: 0.0189\n",
            "Epoch [6/10], Step [11/101], Loss: 0.1396\n",
            "Epoch [6/10], Step [12/101], Loss: 0.6544\n",
            "Epoch [6/10], Step [13/101], Loss: 0.3057\n",
            "Epoch [6/10], Step [14/101], Loss: 0.5270\n",
            "Epoch [6/10], Step [15/101], Loss: 0.4818\n",
            "Epoch [6/10], Step [16/101], Loss: 0.3155\n",
            "Epoch [6/10], Step [17/101], Loss: 0.6191\n",
            "Epoch [6/10], Step [18/101], Loss: 0.4579\n",
            "Epoch [6/10], Step [19/101], Loss: 0.0845\n",
            "Epoch [6/10], Step [20/101], Loss: 0.3945\n",
            "Epoch [6/10], Step [21/101], Loss: 0.2553\n",
            "Epoch [6/10], Step [22/101], Loss: 0.2147\n",
            "Epoch [6/10], Step [23/101], Loss: 0.3687\n",
            "Epoch [6/10], Step [24/101], Loss: 0.7837\n",
            "Epoch [6/10], Step [25/101], Loss: 0.2652\n",
            "Epoch [6/10], Step [26/101], Loss: 0.1154\n",
            "Epoch [6/10], Step [27/101], Loss: 0.0789\n",
            "Epoch [6/10], Step [28/101], Loss: 0.2592\n",
            "Epoch [6/10], Step [29/101], Loss: 0.1267\n",
            "Epoch [6/10], Step [30/101], Loss: 0.2686\n",
            "Epoch [6/10], Step [31/101], Loss: 0.2276\n",
            "Epoch [6/10], Step [32/101], Loss: 0.1614\n",
            "Epoch [6/10], Step [33/101], Loss: 0.1938\n",
            "Epoch [6/10], Step [34/101], Loss: 0.2501\n",
            "Epoch [6/10], Step [35/101], Loss: 0.3321\n",
            "Epoch [6/10], Step [36/101], Loss: 0.2053\n",
            "Epoch [6/10], Step [37/101], Loss: 0.3026\n",
            "Epoch [6/10], Step [38/101], Loss: 0.1247\n",
            "Epoch [6/10], Step [39/101], Loss: 0.2932\n",
            "Epoch [6/10], Step [40/101], Loss: 0.2605\n",
            "Epoch [6/10], Step [41/101], Loss: 0.0544\n",
            "Epoch [6/10], Step [42/101], Loss: 0.1058\n",
            "Epoch [6/10], Step [43/101], Loss: 0.8060\n",
            "Epoch [6/10], Step [44/101], Loss: 0.4111\n",
            "Epoch [6/10], Step [45/101], Loss: 0.1324\n",
            "Epoch [6/10], Step [46/101], Loss: 0.1190\n",
            "Epoch [6/10], Step [47/101], Loss: 0.2015\n",
            "Epoch [6/10], Step [48/101], Loss: 0.2281\n",
            "Epoch [6/10], Step [49/101], Loss: 0.6460\n",
            "Epoch [6/10], Step [50/101], Loss: 0.1541\n",
            "Epoch [6/10], Step [51/101], Loss: 0.5673\n",
            "Epoch [6/10], Step [52/101], Loss: 0.5686\n",
            "Epoch [6/10], Step [53/101], Loss: 0.2054\n",
            "Epoch [6/10], Step [54/101], Loss: 0.3659\n",
            "Epoch [6/10], Step [55/101], Loss: 0.2247\n",
            "Epoch [6/10], Step [56/101], Loss: 0.1080\n",
            "Epoch [6/10], Step [57/101], Loss: 0.0391\n",
            "Epoch [6/10], Step [58/101], Loss: 0.4338\n",
            "Epoch [6/10], Step [59/101], Loss: 0.3025\n",
            "Epoch [6/10], Step [60/101], Loss: 0.2270\n",
            "Epoch [6/10], Step [61/101], Loss: 0.1192\n",
            "Epoch [6/10], Step [62/101], Loss: 0.2433\n",
            "Epoch [6/10], Step [63/101], Loss: 0.5126\n",
            "Epoch [6/10], Step [64/101], Loss: 0.1999\n",
            "Epoch [6/10], Step [65/101], Loss: 0.1830\n",
            "Epoch [6/10], Step [66/101], Loss: 0.2874\n",
            "Epoch [6/10], Step [67/101], Loss: 0.3505\n",
            "Epoch [6/10], Step [68/101], Loss: 0.3957\n",
            "Epoch [6/10], Step [69/101], Loss: 0.1825\n",
            "Epoch [6/10], Step [70/101], Loss: 0.2084\n",
            "Epoch [6/10], Step [71/101], Loss: 0.3218\n",
            "Epoch [6/10], Step [72/101], Loss: 0.2589\n",
            "Epoch [6/10], Step [73/101], Loss: 0.1238\n",
            "Epoch [6/10], Step [74/101], Loss: 0.0171\n",
            "Epoch [6/10], Step [75/101], Loss: 0.1331\n",
            "Epoch [6/10], Step [76/101], Loss: 0.1858\n",
            "Epoch [6/10], Step [77/101], Loss: 0.1804\n",
            "Epoch [6/10], Step [78/101], Loss: 0.0861\n",
            "Epoch [6/10], Step [79/101], Loss: 0.1838\n",
            "Epoch [6/10], Step [80/101], Loss: 0.3027\n",
            "Epoch [6/10], Step [81/101], Loss: 0.4260\n",
            "Epoch [6/10], Step [82/101], Loss: 0.3009\n",
            "Epoch [6/10], Step [83/101], Loss: 0.1022\n",
            "Epoch [6/10], Step [84/101], Loss: 0.7822\n",
            "Epoch [6/10], Step [85/101], Loss: 0.1842\n",
            "Epoch [6/10], Step [86/101], Loss: 0.0934\n",
            "Epoch [6/10], Step [87/101], Loss: 0.1569\n",
            "Epoch [6/10], Step [88/101], Loss: 0.1169\n",
            "Epoch [6/10], Step [89/101], Loss: 0.1903\n",
            "Epoch [6/10], Step [90/101], Loss: 0.6302\n",
            "Epoch [6/10], Step [91/101], Loss: 0.2907\n",
            "Epoch [6/10], Step [92/101], Loss: 0.4450\n",
            "Epoch [6/10], Step [93/101], Loss: 0.4423\n",
            "Epoch [6/10], Step [94/101], Loss: 0.9548\n",
            "Epoch [6/10], Step [95/101], Loss: 0.2018\n",
            "Epoch [6/10], Step [96/101], Loss: 0.1439\n",
            "Epoch [6/10], Step [97/101], Loss: 1.0812\n",
            "Epoch [6/10], Step [98/101], Loss: 0.0939\n",
            "Epoch [6/10], Step [99/101], Loss: 0.1676\n",
            "Epoch [6/10], Step [100/101], Loss: 0.2759\n",
            "Epoch [6/10], Step [101/101], Loss: 0.1471\n",
            "Epoch finished\n",
            "Epoch [7/10], Step [1/101], Loss: 0.1962\n",
            "Epoch [7/10], Step [2/101], Loss: 0.0979\n",
            "Epoch [7/10], Step [3/101], Loss: 0.2237\n",
            "Epoch [7/10], Step [4/101], Loss: 0.0314\n",
            "Epoch [7/10], Step [5/101], Loss: 0.1305\n",
            "Epoch [7/10], Step [6/101], Loss: 0.2370\n",
            "Epoch [7/10], Step [7/101], Loss: 0.5523\n",
            "Epoch [7/10], Step [8/101], Loss: 0.4938\n",
            "Epoch [7/10], Step [9/101], Loss: 0.2317\n",
            "Epoch [7/10], Step [10/101], Loss: 0.1384\n",
            "Epoch [7/10], Step [11/101], Loss: 0.0191\n",
            "Epoch [7/10], Step [12/101], Loss: 0.0532\n",
            "Epoch [7/10], Step [13/101], Loss: 0.3197\n",
            "Epoch [7/10], Step [14/101], Loss: 0.2235\n",
            "Epoch [7/10], Step [15/101], Loss: 0.4379\n",
            "Epoch [7/10], Step [16/101], Loss: 0.3781\n",
            "Epoch [7/10], Step [17/101], Loss: 0.1099\n",
            "Epoch [7/10], Step [18/101], Loss: 0.1950\n",
            "Epoch [7/10], Step [19/101], Loss: 0.2565\n",
            "Epoch [7/10], Step [20/101], Loss: 0.4629\n",
            "Epoch [7/10], Step [21/101], Loss: 0.0635\n",
            "Epoch [7/10], Step [22/101], Loss: 0.1822\n",
            "Epoch [7/10], Step [23/101], Loss: 0.1387\n",
            "Epoch [7/10], Step [24/101], Loss: 0.0466\n",
            "Epoch [7/10], Step [25/101], Loss: 0.1363\n",
            "Epoch [7/10], Step [26/101], Loss: 0.0151\n",
            "Epoch [7/10], Step [27/101], Loss: 0.1109\n",
            "Epoch [7/10], Step [28/101], Loss: 0.8754\n",
            "Epoch [7/10], Step [29/101], Loss: 0.5407\n",
            "Epoch [7/10], Step [30/101], Loss: 0.1609\n",
            "Epoch [7/10], Step [31/101], Loss: 0.1301\n",
            "Epoch [7/10], Step [32/101], Loss: 0.2950\n",
            "Epoch [7/10], Step [33/101], Loss: 0.2675\n",
            "Epoch [7/10], Step [34/101], Loss: 0.0588\n",
            "Epoch [7/10], Step [35/101], Loss: 0.4270\n",
            "Epoch [7/10], Step [36/101], Loss: 0.3167\n",
            "Epoch [7/10], Step [37/101], Loss: 0.3067\n",
            "Epoch [7/10], Step [38/101], Loss: 0.1933\n",
            "Epoch [7/10], Step [39/101], Loss: 0.4103\n",
            "Epoch [7/10], Step [40/101], Loss: 0.3238\n",
            "Epoch [7/10], Step [41/101], Loss: 0.1469\n",
            "Epoch [7/10], Step [42/101], Loss: 0.1647\n",
            "Epoch [7/10], Step [43/101], Loss: 0.3075\n",
            "Epoch [7/10], Step [44/101], Loss: 0.1892\n",
            "Epoch [7/10], Step [45/101], Loss: 0.1635\n",
            "Epoch [7/10], Step [46/101], Loss: 0.1303\n",
            "Epoch [7/10], Step [47/101], Loss: 0.3408\n",
            "Epoch [7/10], Step [48/101], Loss: 0.1140\n",
            "Epoch [7/10], Step [49/101], Loss: 0.3958\n",
            "Epoch [7/10], Step [50/101], Loss: 0.0839\n",
            "Epoch [7/10], Step [51/101], Loss: 0.5412\n",
            "Epoch [7/10], Step [52/101], Loss: 0.0860\n",
            "Epoch [7/10], Step [53/101], Loss: 0.2378\n",
            "Epoch [7/10], Step [54/101], Loss: 0.4088\n",
            "Epoch [7/10], Step [55/101], Loss: 0.2556\n",
            "Epoch [7/10], Step [56/101], Loss: 0.1626\n",
            "Epoch [7/10], Step [57/101], Loss: 0.2614\n",
            "Epoch [7/10], Step [58/101], Loss: 0.0733\n",
            "Epoch [7/10], Step [59/101], Loss: 0.0410\n",
            "Epoch [7/10], Step [60/101], Loss: 0.1735\n",
            "Epoch [7/10], Step [61/101], Loss: 0.0824\n",
            "Epoch [7/10], Step [62/101], Loss: 0.2637\n",
            "Epoch [7/10], Step [63/101], Loss: 0.2193\n",
            "Epoch [7/10], Step [64/101], Loss: 0.3642\n",
            "Epoch [7/10], Step [65/101], Loss: 0.1585\n",
            "Epoch [7/10], Step [66/101], Loss: 0.1922\n",
            "Epoch [7/10], Step [67/101], Loss: 0.0501\n",
            "Epoch [7/10], Step [68/101], Loss: 0.1095\n",
            "Epoch [7/10], Step [69/101], Loss: 0.3623\n",
            "Epoch [7/10], Step [70/101], Loss: 0.3691\n",
            "Epoch [7/10], Step [71/101], Loss: 0.1083\n",
            "Epoch [7/10], Step [72/101], Loss: 0.4825\n",
            "Epoch [7/10], Step [73/101], Loss: 0.1561\n",
            "Epoch [7/10], Step [74/101], Loss: 0.3571\n",
            "Epoch [7/10], Step [75/101], Loss: 0.2631\n",
            "Epoch [7/10], Step [76/101], Loss: 0.2282\n",
            "Epoch [7/10], Step [77/101], Loss: 0.1141\n",
            "Epoch [7/10], Step [78/101], Loss: 0.1017\n",
            "Epoch [7/10], Step [79/101], Loss: 0.1721\n",
            "Epoch [7/10], Step [80/101], Loss: 0.0946\n",
            "Epoch [7/10], Step [81/101], Loss: 0.3556\n",
            "Epoch [7/10], Step [82/101], Loss: 0.1266\n",
            "Epoch [7/10], Step [83/101], Loss: 0.1792\n",
            "Epoch [7/10], Step [84/101], Loss: 0.0294\n",
            "Epoch [7/10], Step [85/101], Loss: 0.1158\n",
            "Epoch [7/10], Step [86/101], Loss: 0.1466\n",
            "Epoch [7/10], Step [87/101], Loss: 0.0668\n",
            "Epoch [7/10], Step [88/101], Loss: 0.5396\n",
            "Epoch [7/10], Step [89/101], Loss: 0.0344\n",
            "Epoch [7/10], Step [90/101], Loss: 0.1566\n",
            "Epoch [7/10], Step [91/101], Loss: 0.0439\n",
            "Epoch [7/10], Step [92/101], Loss: 0.1347\n",
            "Epoch [7/10], Step [93/101], Loss: 0.0586\n",
            "Epoch [7/10], Step [94/101], Loss: 0.2245\n",
            "Epoch [7/10], Step [95/101], Loss: 0.1349\n",
            "Epoch [7/10], Step [96/101], Loss: 0.3860\n",
            "Epoch [7/10], Step [97/101], Loss: 0.2585\n",
            "Epoch [7/10], Step [98/101], Loss: 0.6293\n",
            "Epoch [7/10], Step [99/101], Loss: 0.4847\n",
            "Epoch [7/10], Step [100/101], Loss: 0.0087\n",
            "Epoch [7/10], Step [101/101], Loss: 0.5961\n",
            "Epoch finished\n",
            "Epoch [8/10], Step [1/101], Loss: 0.1546\n",
            "Epoch [8/10], Step [2/101], Loss: 0.2301\n",
            "Epoch [8/10], Step [3/101], Loss: 0.2453\n",
            "Epoch [8/10], Step [4/101], Loss: 0.1604\n",
            "Epoch [8/10], Step [5/101], Loss: 0.2520\n",
            "Epoch [8/10], Step [6/101], Loss: 0.4674\n",
            "Epoch [8/10], Step [7/101], Loss: 0.2682\n",
            "Epoch [8/10], Step [8/101], Loss: 0.4786\n",
            "Epoch [8/10], Step [9/101], Loss: 0.1000\n",
            "Epoch [8/10], Step [10/101], Loss: 0.3139\n",
            "Epoch [8/10], Step [11/101], Loss: 0.5320\n",
            "Epoch [8/10], Step [12/101], Loss: 0.7310\n",
            "Epoch [8/10], Step [13/101], Loss: 0.1595\n",
            "Epoch [8/10], Step [14/101], Loss: 0.1427\n",
            "Epoch [8/10], Step [15/101], Loss: 0.2323\n",
            "Epoch [8/10], Step [16/101], Loss: 0.3939\n",
            "Epoch [8/10], Step [17/101], Loss: 0.2948\n",
            "Epoch [8/10], Step [18/101], Loss: 0.2063\n",
            "Epoch [8/10], Step [19/101], Loss: 0.4206\n",
            "Epoch [8/10], Step [20/101], Loss: 0.0864\n",
            "Epoch [8/10], Step [21/101], Loss: 0.0956\n",
            "Epoch [8/10], Step [22/101], Loss: 0.1811\n",
            "Epoch [8/10], Step [23/101], Loss: 0.5126\n",
            "Epoch [8/10], Step [24/101], Loss: 0.5254\n",
            "Epoch [8/10], Step [25/101], Loss: 0.2623\n",
            "Epoch [8/10], Step [26/101], Loss: 0.0207\n",
            "Epoch [8/10], Step [27/101], Loss: 0.2302\n",
            "Epoch [8/10], Step [28/101], Loss: 0.1454\n",
            "Epoch [8/10], Step [29/101], Loss: 0.2125\n",
            "Epoch [8/10], Step [30/101], Loss: 0.5996\n",
            "Epoch [8/10], Step [31/101], Loss: 0.2192\n",
            "Epoch [8/10], Step [32/101], Loss: 0.2428\n",
            "Epoch [8/10], Step [33/101], Loss: 0.3706\n",
            "Epoch [8/10], Step [34/101], Loss: 0.2522\n",
            "Epoch [8/10], Step [35/101], Loss: 0.3558\n",
            "Epoch [8/10], Step [36/101], Loss: 0.2290\n",
            "Epoch [8/10], Step [37/101], Loss: 0.9644\n",
            "Epoch [8/10], Step [38/101], Loss: 0.5365\n",
            "Epoch [8/10], Step [39/101], Loss: 0.4457\n",
            "Epoch [8/10], Step [40/101], Loss: 0.4691\n",
            "Epoch [8/10], Step [41/101], Loss: 0.0926\n",
            "Epoch [8/10], Step [42/101], Loss: 0.5271\n",
            "Epoch [8/10], Step [43/101], Loss: 0.2647\n",
            "Epoch [8/10], Step [44/101], Loss: 0.0202\n",
            "Epoch [8/10], Step [45/101], Loss: 0.3375\n",
            "Epoch [8/10], Step [46/101], Loss: 0.1003\n",
            "Epoch [8/10], Step [47/101], Loss: 0.1859\n",
            "Epoch [8/10], Step [48/101], Loss: 0.2333\n",
            "Epoch [8/10], Step [49/101], Loss: 0.5227\n",
            "Epoch [8/10], Step [50/101], Loss: 0.4408\n",
            "Epoch [8/10], Step [51/101], Loss: 0.1909\n",
            "Epoch [8/10], Step [52/101], Loss: 0.2166\n",
            "Epoch [8/10], Step [53/101], Loss: 0.4215\n",
            "Epoch [8/10], Step [54/101], Loss: 0.4134\n",
            "Epoch [8/10], Step [55/101], Loss: 0.1808\n",
            "Epoch [8/10], Step [56/101], Loss: 0.0490\n",
            "Epoch [8/10], Step [57/101], Loss: 0.2582\n",
            "Epoch [8/10], Step [58/101], Loss: 0.0942\n",
            "Epoch [8/10], Step [59/101], Loss: 0.3960\n",
            "Epoch [8/10], Step [60/101], Loss: 0.0088\n",
            "Epoch [8/10], Step [61/101], Loss: 0.3786\n",
            "Epoch [8/10], Step [62/101], Loss: 0.2528\n",
            "Epoch [8/10], Step [63/101], Loss: 0.3901\n",
            "Epoch [8/10], Step [64/101], Loss: 0.2962\n",
            "Epoch [8/10], Step [65/101], Loss: 0.4956\n",
            "Epoch [8/10], Step [66/101], Loss: 0.1953\n",
            "Epoch [8/10], Step [67/101], Loss: 0.2743\n",
            "Epoch [8/10], Step [68/101], Loss: 0.2633\n",
            "Epoch [8/10], Step [69/101], Loss: 0.2508\n",
            "Epoch [8/10], Step [70/101], Loss: 0.0525\n",
            "Epoch [8/10], Step [71/101], Loss: 0.7126\n",
            "Epoch [8/10], Step [72/101], Loss: 0.3012\n",
            "Epoch [8/10], Step [73/101], Loss: 0.1449\n",
            "Epoch [8/10], Step [74/101], Loss: 0.2714\n",
            "Epoch [8/10], Step [75/101], Loss: 0.1070\n",
            "Epoch [8/10], Step [76/101], Loss: 0.5406\n",
            "Epoch [8/10], Step [77/101], Loss: 0.1355\n",
            "Epoch [8/10], Step [78/101], Loss: 0.2820\n",
            "Epoch [8/10], Step [79/101], Loss: 0.4570\n",
            "Epoch [8/10], Step [80/101], Loss: 0.1591\n",
            "Epoch [8/10], Step [81/101], Loss: 0.1406\n",
            "Epoch [8/10], Step [82/101], Loss: 0.0854\n",
            "Epoch [8/10], Step [83/101], Loss: 0.2918\n",
            "Epoch [8/10], Step [84/101], Loss: 0.4727\n",
            "Epoch [8/10], Step [85/101], Loss: 0.3874\n",
            "Epoch [8/10], Step [86/101], Loss: 0.4588\n",
            "Epoch [8/10], Step [87/101], Loss: 0.0084\n",
            "Epoch [8/10], Step [88/101], Loss: 0.2892\n",
            "Epoch [8/10], Step [89/101], Loss: 0.2963\n",
            "Epoch [8/10], Step [90/101], Loss: 0.4861\n",
            "Epoch [8/10], Step [91/101], Loss: 0.0777\n",
            "Epoch [8/10], Step [92/101], Loss: 0.7459\n",
            "Epoch [8/10], Step [93/101], Loss: 0.7480\n",
            "Epoch [8/10], Step [94/101], Loss: 0.0714\n",
            "Epoch [8/10], Step [95/101], Loss: 0.5725\n",
            "Epoch [8/10], Step [96/101], Loss: 0.0949\n",
            "Epoch [8/10], Step [97/101], Loss: 0.7909\n",
            "Epoch [8/10], Step [98/101], Loss: 0.5900\n",
            "Epoch [8/10], Step [99/101], Loss: 1.7617\n",
            "Epoch [8/10], Step [100/101], Loss: 0.8931\n",
            "Epoch [8/10], Step [101/101], Loss: 0.4819\n",
            "Epoch finished\n",
            "Epoch [9/10], Step [1/101], Loss: 0.3944\n",
            "Epoch [9/10], Step [2/101], Loss: 0.6238\n",
            "Epoch [9/10], Step [3/101], Loss: 0.0300\n",
            "Epoch [9/10], Step [4/101], Loss: 0.0600\n",
            "Epoch [9/10], Step [5/101], Loss: 0.1533\n",
            "Epoch [9/10], Step [6/101], Loss: 1.0039\n",
            "Epoch [9/10], Step [7/101], Loss: 0.0621\n",
            "Epoch [9/10], Step [8/101], Loss: 0.0789\n",
            "Epoch [9/10], Step [9/101], Loss: 0.2042\n",
            "Epoch [9/10], Step [10/101], Loss: 0.4573\n",
            "Epoch [9/10], Step [11/101], Loss: 0.5595\n",
            "Epoch [9/10], Step [12/101], Loss: 0.1371\n",
            "Epoch [9/10], Step [13/101], Loss: 0.4872\n",
            "Epoch [9/10], Step [14/101], Loss: 0.0147\n",
            "Epoch [9/10], Step [15/101], Loss: 0.2107\n",
            "Epoch [9/10], Step [16/101], Loss: 0.1681\n",
            "Epoch [9/10], Step [17/101], Loss: 0.0542\n",
            "Epoch [9/10], Step [18/101], Loss: 0.0940\n",
            "Epoch [9/10], Step [19/101], Loss: 0.2912\n",
            "Epoch [9/10], Step [20/101], Loss: 0.0588\n",
            "Epoch [9/10], Step [21/101], Loss: 0.0089\n",
            "Epoch [9/10], Step [22/101], Loss: 0.1239\n",
            "Epoch [9/10], Step [23/101], Loss: 0.1769\n",
            "Epoch [9/10], Step [24/101], Loss: 0.1037\n",
            "Epoch [9/10], Step [25/101], Loss: 0.0880\n",
            "Epoch [9/10], Step [26/101], Loss: 0.2901\n",
            "Epoch [9/10], Step [27/101], Loss: 0.0723\n",
            "Epoch [9/10], Step [28/101], Loss: 0.2629\n",
            "Epoch [9/10], Step [29/101], Loss: 0.1415\n",
            "Epoch [9/10], Step [30/101], Loss: 0.1372\n",
            "Epoch [9/10], Step [31/101], Loss: 0.2208\n",
            "Epoch [9/10], Step [32/101], Loss: 0.3573\n",
            "Epoch [9/10], Step [33/101], Loss: 0.0673\n",
            "Epoch [9/10], Step [34/101], Loss: 0.5251\n",
            "Epoch [9/10], Step [35/101], Loss: 0.1353\n",
            "Epoch [9/10], Step [36/101], Loss: 0.4030\n",
            "Epoch [9/10], Step [37/101], Loss: 0.1136\n",
            "Epoch [9/10], Step [38/101], Loss: 0.1084\n",
            "Epoch [9/10], Step [39/101], Loss: 0.0751\n",
            "Epoch [9/10], Step [40/101], Loss: 0.4638\n",
            "Epoch [9/10], Step [41/101], Loss: 0.4836\n",
            "Epoch [9/10], Step [42/101], Loss: 0.1541\n",
            "Epoch [9/10], Step [43/101], Loss: 0.1965\n",
            "Epoch [9/10], Step [44/101], Loss: 0.0650\n",
            "Epoch [9/10], Step [45/101], Loss: 0.2185\n",
            "Epoch [9/10], Step [46/101], Loss: 0.0834\n",
            "Epoch [9/10], Step [47/101], Loss: 0.0827\n",
            "Epoch [9/10], Step [48/101], Loss: 0.1827\n",
            "Epoch [9/10], Step [49/101], Loss: 0.3371\n",
            "Epoch [9/10], Step [50/101], Loss: 0.0483\n",
            "Epoch [9/10], Step [51/101], Loss: 0.2633\n",
            "Epoch [9/10], Step [52/101], Loss: 0.0293\n",
            "Epoch [9/10], Step [53/101], Loss: 0.1327\n",
            "Epoch [9/10], Step [54/101], Loss: 0.5538\n",
            "Epoch [9/10], Step [55/101], Loss: 0.4911\n",
            "Epoch [9/10], Step [56/101], Loss: 0.1933\n",
            "Epoch [9/10], Step [57/101], Loss: 0.0492\n",
            "Epoch [9/10], Step [58/101], Loss: 0.1595\n",
            "Epoch [9/10], Step [59/101], Loss: 0.1294\n",
            "Epoch [9/10], Step [60/101], Loss: 0.3135\n",
            "Epoch [9/10], Step [61/101], Loss: 0.2023\n",
            "Epoch [9/10], Step [62/101], Loss: 0.1200\n",
            "Epoch [9/10], Step [63/101], Loss: 0.1385\n",
            "Epoch [9/10], Step [64/101], Loss: 0.2031\n",
            "Epoch [9/10], Step [65/101], Loss: 0.0519\n",
            "Epoch [9/10], Step [66/101], Loss: 0.3256\n",
            "Epoch [9/10], Step [67/101], Loss: 0.2186\n",
            "Epoch [9/10], Step [68/101], Loss: 0.2451\n",
            "Epoch [9/10], Step [69/101], Loss: 0.2202\n",
            "Epoch [9/10], Step [70/101], Loss: 0.2051\n",
            "Epoch [9/10], Step [71/101], Loss: 0.2960\n",
            "Epoch [9/10], Step [72/101], Loss: 0.4345\n",
            "Epoch [9/10], Step [73/101], Loss: 0.3883\n",
            "Epoch [9/10], Step [74/101], Loss: 0.2333\n",
            "Epoch [9/10], Step [75/101], Loss: 0.1626\n",
            "Epoch [9/10], Step [76/101], Loss: 0.7499\n",
            "Epoch [9/10], Step [77/101], Loss: 0.3199\n",
            "Epoch [9/10], Step [78/101], Loss: 0.0469\n",
            "Epoch [9/10], Step [79/101], Loss: 0.5823\n",
            "Epoch [9/10], Step [80/101], Loss: 0.5156\n",
            "Epoch [9/10], Step [81/101], Loss: 0.2077\n",
            "Epoch [9/10], Step [82/101], Loss: 0.4296\n",
            "Epoch [9/10], Step [83/101], Loss: 0.1064\n",
            "Epoch [9/10], Step [84/101], Loss: 0.0558\n",
            "Epoch [9/10], Step [85/101], Loss: 0.5427\n",
            "Epoch [9/10], Step [86/101], Loss: 0.0220\n",
            "Epoch [9/10], Step [87/101], Loss: 0.2822\n",
            "Epoch [9/10], Step [88/101], Loss: 0.1532\n",
            "Epoch [9/10], Step [89/101], Loss: 0.1421\n",
            "Epoch [9/10], Step [90/101], Loss: 0.0900\n",
            "Epoch [9/10], Step [91/101], Loss: 0.1594\n",
            "Epoch [9/10], Step [92/101], Loss: 0.2288\n",
            "Epoch [9/10], Step [93/101], Loss: 0.0923\n",
            "Epoch [9/10], Step [94/101], Loss: 0.0576\n",
            "Epoch [9/10], Step [95/101], Loss: 0.4742\n",
            "Epoch [9/10], Step [96/101], Loss: 0.3255\n",
            "Epoch [9/10], Step [97/101], Loss: 0.5023\n",
            "Epoch [9/10], Step [98/101], Loss: 0.6741\n",
            "Epoch [9/10], Step [99/101], Loss: 0.3379\n",
            "Epoch [9/10], Step [100/101], Loss: 0.1725\n",
            "Epoch [9/10], Step [101/101], Loss: 0.0002\n",
            "Epoch finished\n",
            "Epoch [10/10], Step [1/101], Loss: 0.0676\n",
            "Epoch [10/10], Step [2/101], Loss: 0.0514\n",
            "Epoch [10/10], Step [3/101], Loss: 0.0301\n",
            "Epoch [10/10], Step [4/101], Loss: 0.0683\n",
            "Epoch [10/10], Step [5/101], Loss: 0.1567\n",
            "Epoch [10/10], Step [6/101], Loss: 0.2189\n",
            "Epoch [10/10], Step [7/101], Loss: 0.3759\n",
            "Epoch [10/10], Step [8/101], Loss: 0.5425\n",
            "Epoch [10/10], Step [9/101], Loss: 0.2518\n",
            "Epoch [10/10], Step [10/101], Loss: 0.3453\n",
            "Epoch [10/10], Step [11/101], Loss: 0.1656\n",
            "Epoch [10/10], Step [12/101], Loss: 0.4229\n",
            "Epoch [10/10], Step [13/101], Loss: 0.1313\n",
            "Epoch [10/10], Step [14/101], Loss: 0.2524\n",
            "Epoch [10/10], Step [15/101], Loss: 0.2861\n",
            "Epoch [10/10], Step [16/101], Loss: 0.1894\n",
            "Epoch [10/10], Step [17/101], Loss: 0.6676\n",
            "Epoch [10/10], Step [18/101], Loss: 0.1623\n",
            "Epoch [10/10], Step [19/101], Loss: 0.1459\n",
            "Epoch [10/10], Step [20/101], Loss: 0.0260\n",
            "Epoch [10/10], Step [21/101], Loss: 0.1484\n",
            "Epoch [10/10], Step [22/101], Loss: 0.3101\n",
            "Epoch [10/10], Step [23/101], Loss: 0.6106\n",
            "Epoch [10/10], Step [24/101], Loss: 0.2481\n",
            "Epoch [10/10], Step [25/101], Loss: 0.2356\n",
            "Epoch [10/10], Step [26/101], Loss: 0.0555\n",
            "Epoch [10/10], Step [27/101], Loss: 0.4340\n",
            "Epoch [10/10], Step [28/101], Loss: 0.4091\n",
            "Epoch [10/10], Step [29/101], Loss: 0.3695\n",
            "Epoch [10/10], Step [30/101], Loss: 0.2852\n",
            "Epoch [10/10], Step [31/101], Loss: 0.3047\n",
            "Epoch [10/10], Step [32/101], Loss: 0.0711\n",
            "Epoch [10/10], Step [33/101], Loss: 0.1690\n",
            "Epoch [10/10], Step [34/101], Loss: 0.2204\n",
            "Epoch [10/10], Step [35/101], Loss: 0.1869\n",
            "Epoch [10/10], Step [36/101], Loss: 0.2680\n",
            "Epoch [10/10], Step [37/101], Loss: 0.1202\n",
            "Epoch [10/10], Step [38/101], Loss: 0.0717\n",
            "Epoch [10/10], Step [39/101], Loss: 0.0885\n",
            "Epoch [10/10], Step [40/101], Loss: 0.2552\n",
            "Epoch [10/10], Step [41/101], Loss: 0.1087\n",
            "Epoch [10/10], Step [42/101], Loss: 0.0882\n",
            "Epoch [10/10], Step [43/101], Loss: 0.0750\n",
            "Epoch [10/10], Step [44/101], Loss: 0.3434\n",
            "Epoch [10/10], Step [45/101], Loss: 0.3184\n",
            "Epoch [10/10], Step [46/101], Loss: 0.1045\n",
            "Epoch [10/10], Step [47/101], Loss: 0.3895\n",
            "Epoch [10/10], Step [48/101], Loss: 0.2820\n",
            "Epoch [10/10], Step [49/101], Loss: 0.0796\n",
            "Epoch [10/10], Step [50/101], Loss: 0.7141\n",
            "Epoch [10/10], Step [51/101], Loss: 0.4311\n",
            "Epoch [10/10], Step [52/101], Loss: 0.0602\n",
            "Epoch [10/10], Step [53/101], Loss: 0.2422\n",
            "Epoch [10/10], Step [54/101], Loss: 0.0419\n",
            "Epoch [10/10], Step [55/101], Loss: 0.1644\n",
            "Epoch [10/10], Step [56/101], Loss: 0.1244\n",
            "Epoch [10/10], Step [57/101], Loss: 0.1058\n",
            "Epoch [10/10], Step [58/101], Loss: 0.2704\n",
            "Epoch [10/10], Step [59/101], Loss: 0.2354\n",
            "Epoch [10/10], Step [60/101], Loss: 0.0136\n",
            "Epoch [10/10], Step [61/101], Loss: 0.2667\n",
            "Epoch [10/10], Step [62/101], Loss: 0.2392\n",
            "Epoch [10/10], Step [63/101], Loss: 0.0525\n",
            "Epoch [10/10], Step [64/101], Loss: 0.3085\n",
            "Epoch [10/10], Step [65/101], Loss: 0.1236\n",
            "Epoch [10/10], Step [66/101], Loss: 0.4269\n",
            "Epoch [10/10], Step [67/101], Loss: 0.1717\n",
            "Epoch [10/10], Step [68/101], Loss: 0.1629\n",
            "Epoch [10/10], Step [69/101], Loss: 0.1887\n",
            "Epoch [10/10], Step [70/101], Loss: 0.0767\n",
            "Epoch [10/10], Step [71/101], Loss: 0.1138\n",
            "Epoch [10/10], Step [72/101], Loss: 0.4335\n",
            "Epoch [10/10], Step [73/101], Loss: 0.6642\n",
            "Epoch [10/10], Step [74/101], Loss: 0.3984\n",
            "Epoch [10/10], Step [75/101], Loss: 0.0775\n",
            "Epoch [10/10], Step [76/101], Loss: 0.2502\n",
            "Epoch [10/10], Step [77/101], Loss: 0.2362\n",
            "Epoch [10/10], Step [78/101], Loss: 0.2338\n",
            "Epoch [10/10], Step [79/101], Loss: 0.2578\n",
            "Epoch [10/10], Step [80/101], Loss: 0.0289\n",
            "Epoch [10/10], Step [81/101], Loss: 0.3799\n",
            "Epoch [10/10], Step [82/101], Loss: 0.2214\n",
            "Epoch [10/10], Step [83/101], Loss: 0.2004\n",
            "Epoch [10/10], Step [84/101], Loss: 0.1190\n",
            "Epoch [10/10], Step [85/101], Loss: 0.2775\n",
            "Epoch [10/10], Step [86/101], Loss: 0.7086\n",
            "Epoch [10/10], Step [87/101], Loss: 0.3265\n",
            "Epoch [10/10], Step [88/101], Loss: 0.2199\n",
            "Epoch [10/10], Step [89/101], Loss: 0.5896\n",
            "Epoch [10/10], Step [90/101], Loss: 0.1333\n",
            "Epoch [10/10], Step [91/101], Loss: 0.0067\n",
            "Epoch [10/10], Step [92/101], Loss: 0.0949\n",
            "Epoch [10/10], Step [93/101], Loss: 0.0854\n",
            "Epoch [10/10], Step [94/101], Loss: 0.4664\n",
            "Epoch [10/10], Step [95/101], Loss: 0.2846\n",
            "Epoch [10/10], Step [96/101], Loss: 1.0194\n",
            "Epoch [10/10], Step [97/101], Loss: 0.0322\n",
            "Epoch [10/10], Step [98/101], Loss: 0.1934\n",
            "Epoch [10/10], Step [99/101], Loss: 0.1671\n",
            "Epoch [10/10], Step [100/101], Loss: 0.2408\n",
            "Epoch [10/10], Step [101/101], Loss: 0.0129\n",
            "Epoch finished\n",
            "Training finished!\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test(alex)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "E4CO0DeEI_Ms",
        "outputId": "5f5a6670-28ee-4571-d812-d5a566bbd508"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy of the network on the test images: 94 %\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "pip install efficientnet_pytorch"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "c3blw0jBqivX",
        "outputId": "c60f455a-c1f1-4e8c-e2fc-189393a694eb"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: efficientnet_pytorch in /usr/local/lib/python3.10/dist-packages (0.7.1)\n",
            "Requirement already satisfied: torch in /usr/local/lib/python3.10/dist-packages (from efficientnet_pytorch) (2.0.1+cu118)\n",
            "Requirement already satisfied: filelock in /usr/local/lib/python3.10/dist-packages (from torch->efficientnet_pytorch) (3.12.2)\n",
            "Requirement already satisfied: typing-extensions in /usr/local/lib/python3.10/dist-packages (from torch->efficientnet_pytorch) (4.7.1)\n",
            "Requirement already satisfied: sympy in /usr/local/lib/python3.10/dist-packages (from torch->efficientnet_pytorch) (1.11.1)\n",
            "Requirement already satisfied: networkx in /usr/local/lib/python3.10/dist-packages (from torch->efficientnet_pytorch) (3.1)\n",
            "Requirement already satisfied: jinja2 in /usr/local/lib/python3.10/dist-packages (from torch->efficientnet_pytorch) (3.1.2)\n",
            "Requirement already satisfied: triton==2.0.0 in /usr/local/lib/python3.10/dist-packages (from torch->efficientnet_pytorch) (2.0.0)\n",
            "Requirement already satisfied: cmake in /usr/local/lib/python3.10/dist-packages (from triton==2.0.0->torch->efficientnet_pytorch) (3.25.2)\n",
            "Requirement already satisfied: lit in /usr/local/lib/python3.10/dist-packages (from triton==2.0.0->torch->efficientnet_pytorch) (16.0.6)\n",
            "Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.10/dist-packages (from jinja2->torch->efficientnet_pytorch) (2.1.3)\n",
            "Requirement already satisfied: mpmath>=0.19 in /usr/local/lib/python3.10/dist-packages (from sympy->torch->efficientnet_pytorch) (1.3.0)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from efficientnet_pytorch import EfficientNet\n",
        "model = EfficientNet.from_pretrained('efficientnet-b4')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8sSTOlOGqwPd",
        "outputId": "dab8fae1-e2e6-470e-839b-a95161f8c726"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Downloading: \"https://github.com/lukemelas/EfficientNet-PyTorch/releases/download/1.0/efficientnet-b4-6ed6700e.pth\" to /root/.cache/torch/hub/checkpoints/efficientnet-b4-6ed6700e.pth\n",
            "100%|██████████| 74.4M/74.4M [00:00<00:00, 109MB/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Loaded pretrained weights for efficientnet-b4\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model"
      ],
      "metadata": {
        "id": "_Fsx34TEq8Em"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "for param in model.parameters():\n",
        "    param.requires_grad=False"
      ],
      "metadata": {
        "id": "QxexvegZrK9e"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "in_features = model._fc.in_features\n",
        "model._fc = torch.nn.Linear(in_features, 20)\n",
        "\n",
        "# Now set the 'requires_grad' property to True for the parameters of the custom head\n",
        "for param in model._fc.parameters():\n",
        "    param.requires_grad = True"
      ],
      "metadata": {
        "id": "3oC12kNRrcLr"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model = model.to(device)\n",
        "model.cuda()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4mXHkowUr-D_",
        "outputId": "bb091d72-a28a-4feb-ab41-6614ffa7ebb3"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "EfficientNet(\n",
              "  (_conv_stem): Conv2dStaticSamePadding(\n",
              "    3, 48, kernel_size=(3, 3), stride=(2, 2), bias=False\n",
              "    (static_padding): ZeroPad2d((0, 1, 0, 1))\n",
              "  )\n",
              "  (_bn0): BatchNorm2d(48, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "  (_blocks): ModuleList(\n",
              "    (0): MBConvBlock(\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        48, 48, kernel_size=(3, 3), stride=[1, 1], groups=48, bias=False\n",
              "        (static_padding): ZeroPad2d((1, 1, 1, 1))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(48, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        48, 12, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        12, 48, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        48, 24, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(24, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "    (1): MBConvBlock(\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        24, 24, kernel_size=(3, 3), stride=(1, 1), groups=24, bias=False\n",
              "        (static_padding): ZeroPad2d((1, 1, 1, 1))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(24, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        24, 6, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        6, 24, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        24, 24, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(24, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "    (2): MBConvBlock(\n",
              "      (_expand_conv): Conv2dStaticSamePadding(\n",
              "        24, 144, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn0): BatchNorm2d(144, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        144, 144, kernel_size=(3, 3), stride=[2, 2], groups=144, bias=False\n",
              "        (static_padding): ZeroPad2d((0, 1, 0, 1))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(144, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        144, 6, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        6, 144, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        144, 32, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(32, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "    (3-5): 3 x MBConvBlock(\n",
              "      (_expand_conv): Conv2dStaticSamePadding(\n",
              "        32, 192, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn0): BatchNorm2d(192, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        192, 192, kernel_size=(3, 3), stride=(1, 1), groups=192, bias=False\n",
              "        (static_padding): ZeroPad2d((1, 1, 1, 1))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(192, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        192, 8, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        8, 192, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        192, 32, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(32, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "    (6): MBConvBlock(\n",
              "      (_expand_conv): Conv2dStaticSamePadding(\n",
              "        32, 192, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn0): BatchNorm2d(192, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        192, 192, kernel_size=(5, 5), stride=[2, 2], groups=192, bias=False\n",
              "        (static_padding): ZeroPad2d((2, 2, 2, 2))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(192, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        192, 8, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        8, 192, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        192, 56, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(56, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "    (7-9): 3 x MBConvBlock(\n",
              "      (_expand_conv): Conv2dStaticSamePadding(\n",
              "        56, 336, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn0): BatchNorm2d(336, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        336, 336, kernel_size=(5, 5), stride=(1, 1), groups=336, bias=False\n",
              "        (static_padding): ZeroPad2d((2, 2, 2, 2))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(336, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        336, 14, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        14, 336, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        336, 56, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(56, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "    (10): MBConvBlock(\n",
              "      (_expand_conv): Conv2dStaticSamePadding(\n",
              "        56, 336, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn0): BatchNorm2d(336, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        336, 336, kernel_size=(3, 3), stride=[2, 2], groups=336, bias=False\n",
              "        (static_padding): ZeroPad2d((0, 1, 0, 1))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(336, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        336, 14, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        14, 336, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        336, 112, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(112, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "    (11-15): 5 x MBConvBlock(\n",
              "      (_expand_conv): Conv2dStaticSamePadding(\n",
              "        112, 672, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn0): BatchNorm2d(672, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        672, 672, kernel_size=(3, 3), stride=(1, 1), groups=672, bias=False\n",
              "        (static_padding): ZeroPad2d((1, 1, 1, 1))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(672, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        672, 28, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        28, 672, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        672, 112, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(112, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "    (16): MBConvBlock(\n",
              "      (_expand_conv): Conv2dStaticSamePadding(\n",
              "        112, 672, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn0): BatchNorm2d(672, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        672, 672, kernel_size=(5, 5), stride=[1, 1], groups=672, bias=False\n",
              "        (static_padding): ZeroPad2d((2, 2, 2, 2))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(672, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        672, 28, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        28, 672, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        672, 160, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(160, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "    (17-21): 5 x MBConvBlock(\n",
              "      (_expand_conv): Conv2dStaticSamePadding(\n",
              "        160, 960, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn0): BatchNorm2d(960, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        960, 960, kernel_size=(5, 5), stride=(1, 1), groups=960, bias=False\n",
              "        (static_padding): ZeroPad2d((2, 2, 2, 2))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(960, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        960, 40, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        40, 960, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        960, 160, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(160, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "    (22): MBConvBlock(\n",
              "      (_expand_conv): Conv2dStaticSamePadding(\n",
              "        160, 960, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn0): BatchNorm2d(960, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        960, 960, kernel_size=(5, 5), stride=[2, 2], groups=960, bias=False\n",
              "        (static_padding): ZeroPad2d((1, 2, 1, 2))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(960, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        960, 40, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        40, 960, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        960, 272, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(272, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "    (23-29): 7 x MBConvBlock(\n",
              "      (_expand_conv): Conv2dStaticSamePadding(\n",
              "        272, 1632, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn0): BatchNorm2d(1632, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        1632, 1632, kernel_size=(5, 5), stride=(1, 1), groups=1632, bias=False\n",
              "        (static_padding): ZeroPad2d((2, 2, 2, 2))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(1632, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        1632, 68, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        68, 1632, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        1632, 272, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(272, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "    (30): MBConvBlock(\n",
              "      (_expand_conv): Conv2dStaticSamePadding(\n",
              "        272, 1632, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn0): BatchNorm2d(1632, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        1632, 1632, kernel_size=(3, 3), stride=[1, 1], groups=1632, bias=False\n",
              "        (static_padding): ZeroPad2d((1, 1, 1, 1))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(1632, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        1632, 68, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        68, 1632, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        1632, 448, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(448, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "    (31): MBConvBlock(\n",
              "      (_expand_conv): Conv2dStaticSamePadding(\n",
              "        448, 2688, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn0): BatchNorm2d(2688, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        2688, 2688, kernel_size=(3, 3), stride=(1, 1), groups=2688, bias=False\n",
              "        (static_padding): ZeroPad2d((1, 1, 1, 1))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(2688, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        2688, 112, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        112, 2688, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        2688, 448, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(448, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "  )\n",
              "  (_conv_head): Conv2dStaticSamePadding(\n",
              "    448, 1792, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "    (static_padding): Identity()\n",
              "  )\n",
              "  (_bn1): BatchNorm2d(1792, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "  (_avg_pooling): AdaptiveAvgPool2d(output_size=1)\n",
              "  (_dropout): Dropout(p=0.4, inplace=False)\n",
              "  (_fc): Linear(in_features=1792, out_features=20, bias=True)\n",
              "  (_swish): MemoryEfficientSwish()\n",
              ")"
            ]
          },
          "metadata": {},
          "execution_count": 42
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "criterion=nn.CrossEntropyLoss()\n",
        "optimizer=torch.optim.Adam(model.parameters(),lr=0.001)"
      ],
      "metadata": {
        "id": "Gbt6RoshsCM4"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train(model)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "YqIMw-ETzbGU",
        "outputId": "88385c52-ee7a-46f9-dc08-0a2a90e0b978"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch [1/10], Step [1/101], Loss: 0.3105\n",
            "Epoch [1/10], Step [2/101], Loss: 0.1753\n",
            "Epoch [1/10], Step [3/101], Loss: 0.1596\n",
            "Epoch [1/10], Step [4/101], Loss: 0.1821\n",
            "Epoch [1/10], Step [5/101], Loss: 0.1189\n",
            "Epoch [1/10], Step [6/101], Loss: 0.1855\n",
            "Epoch [1/10], Step [7/101], Loss: 0.1484\n",
            "Epoch [1/10], Step [8/101], Loss: 0.0961\n",
            "Epoch [1/10], Step [9/101], Loss: 0.2794\n",
            "Epoch [1/10], Step [10/101], Loss: 0.1936\n",
            "Epoch [1/10], Step [11/101], Loss: 0.2512\n",
            "Epoch [1/10], Step [12/101], Loss: 0.2251\n",
            "Epoch [1/10], Step [13/101], Loss: 0.1275\n",
            "Epoch [1/10], Step [14/101], Loss: 0.2009\n",
            "Epoch [1/10], Step [15/101], Loss: 0.2590\n",
            "Epoch [1/10], Step [16/101], Loss: 0.2039\n",
            "Epoch [1/10], Step [17/101], Loss: 0.4059\n",
            "Epoch [1/10], Step [18/101], Loss: 0.1571\n",
            "Epoch [1/10], Step [19/101], Loss: 0.1184\n",
            "Epoch [1/10], Step [20/101], Loss: 0.1122\n",
            "Epoch [1/10], Step [21/101], Loss: 0.0929\n",
            "Epoch [1/10], Step [22/101], Loss: 0.3149\n",
            "Epoch [1/10], Step [23/101], Loss: 0.1710\n",
            "Epoch [1/10], Step [24/101], Loss: 0.0961\n",
            "Epoch [1/10], Step [25/101], Loss: 0.1753\n",
            "Epoch [1/10], Step [26/101], Loss: 0.1353\n",
            "Epoch [1/10], Step [27/101], Loss: 0.1261\n",
            "Epoch [1/10], Step [28/101], Loss: 0.0569\n",
            "Epoch [1/10], Step [29/101], Loss: 0.0838\n",
            "Epoch [1/10], Step [30/101], Loss: 0.0497\n",
            "Epoch [1/10], Step [31/101], Loss: 0.1604\n",
            "Epoch [1/10], Step [32/101], Loss: 0.2317\n",
            "Epoch [1/10], Step [33/101], Loss: 0.3823\n",
            "Epoch [1/10], Step [34/101], Loss: 0.1304\n",
            "Epoch [1/10], Step [35/101], Loss: 0.2748\n",
            "Epoch [1/10], Step [36/101], Loss: 0.1008\n",
            "Epoch [1/10], Step [37/101], Loss: 0.0954\n",
            "Epoch [1/10], Step [38/101], Loss: 0.2525\n",
            "Epoch [1/10], Step [39/101], Loss: 0.1632\n",
            "Epoch [1/10], Step [40/101], Loss: 0.1336\n",
            "Epoch [1/10], Step [41/101], Loss: 0.2759\n",
            "Epoch [1/10], Step [42/101], Loss: 0.2008\n",
            "Epoch [1/10], Step [43/101], Loss: 0.3864\n",
            "Epoch [1/10], Step [44/101], Loss: 0.1282\n",
            "Epoch [1/10], Step [45/101], Loss: 0.1865\n",
            "Epoch [1/10], Step [46/101], Loss: 0.0863\n",
            "Epoch [1/10], Step [47/101], Loss: 0.2094\n",
            "Epoch [1/10], Step [48/101], Loss: 0.1670\n",
            "Epoch [1/10], Step [49/101], Loss: 0.2001\n",
            "Epoch [1/10], Step [50/101], Loss: 0.1645\n",
            "Epoch [1/10], Step [51/101], Loss: 0.1116\n",
            "Epoch [1/10], Step [52/101], Loss: 0.1563\n",
            "Epoch [1/10], Step [53/101], Loss: 0.2183\n",
            "Epoch [1/10], Step [54/101], Loss: 0.1473\n",
            "Epoch [1/10], Step [55/101], Loss: 0.1731\n",
            "Epoch [1/10], Step [56/101], Loss: 0.1192\n",
            "Epoch [1/10], Step [57/101], Loss: 0.1680\n",
            "Epoch [1/10], Step [58/101], Loss: 0.1775\n",
            "Epoch [1/10], Step [59/101], Loss: 0.1188\n",
            "Epoch [1/10], Step [60/101], Loss: 0.3805\n",
            "Epoch [1/10], Step [61/101], Loss: 0.2134\n",
            "Epoch [1/10], Step [62/101], Loss: 0.0761\n",
            "Epoch [1/10], Step [63/101], Loss: 0.2341\n",
            "Epoch [1/10], Step [64/101], Loss: 0.0806\n",
            "Epoch [1/10], Step [65/101], Loss: 0.2696\n",
            "Epoch [1/10], Step [66/101], Loss: 0.1001\n",
            "Epoch [1/10], Step [67/101], Loss: 0.3264\n",
            "Epoch [1/10], Step [68/101], Loss: 0.1991\n",
            "Epoch [1/10], Step [69/101], Loss: 0.0774\n",
            "Epoch [1/10], Step [70/101], Loss: 0.2511\n",
            "Epoch [1/10], Step [71/101], Loss: 0.2406\n",
            "Epoch [1/10], Step [72/101], Loss: 0.1932\n",
            "Epoch [1/10], Step [73/101], Loss: 0.1034\n",
            "Epoch [1/10], Step [74/101], Loss: 0.1405\n",
            "Epoch [1/10], Step [75/101], Loss: 0.1235\n",
            "Epoch [1/10], Step [76/101], Loss: 0.2225\n",
            "Epoch [1/10], Step [77/101], Loss: 0.1951\n",
            "Epoch [1/10], Step [78/101], Loss: 0.1958\n",
            "Epoch [1/10], Step [79/101], Loss: 0.1297\n",
            "Epoch [1/10], Step [80/101], Loss: 0.1037\n",
            "Epoch [1/10], Step [81/101], Loss: 0.1103\n",
            "Epoch [1/10], Step [82/101], Loss: 0.1414\n",
            "Epoch [1/10], Step [83/101], Loss: 0.1951\n",
            "Epoch [1/10], Step [84/101], Loss: 0.2218\n",
            "Epoch [1/10], Step [85/101], Loss: 0.0756\n",
            "Epoch [1/10], Step [86/101], Loss: 0.1343\n",
            "Epoch [1/10], Step [87/101], Loss: 0.1572\n",
            "Epoch [1/10], Step [88/101], Loss: 0.1468\n",
            "Epoch [1/10], Step [89/101], Loss: 0.1099\n",
            "Epoch [1/10], Step [90/101], Loss: 0.3100\n",
            "Epoch [1/10], Step [91/101], Loss: 0.2603\n",
            "Epoch [1/10], Step [92/101], Loss: 0.1022\n",
            "Epoch [1/10], Step [93/101], Loss: 0.2070\n",
            "Epoch [1/10], Step [94/101], Loss: 0.2304\n",
            "Epoch [1/10], Step [95/101], Loss: 0.1380\n",
            "Epoch [1/10], Step [96/101], Loss: 0.1116\n",
            "Epoch [1/10], Step [97/101], Loss: 0.1975\n",
            "Epoch [1/10], Step [98/101], Loss: 0.1593\n",
            "Epoch [1/10], Step [99/101], Loss: 0.2416\n",
            "Epoch [1/10], Step [100/101], Loss: 0.2693\n",
            "Epoch [1/10], Step [101/101], Loss: 0.6557\n",
            "Epoch finished\n",
            "Epoch [2/10], Step [1/101], Loss: 0.1309\n",
            "Epoch [2/10], Step [2/101], Loss: 0.1086\n",
            "Epoch [2/10], Step [3/101], Loss: 0.2417\n",
            "Epoch [2/10], Step [4/101], Loss: 0.1874\n",
            "Epoch [2/10], Step [5/101], Loss: 0.1044\n",
            "Epoch [2/10], Step [6/101], Loss: 0.1523\n",
            "Epoch [2/10], Step [7/101], Loss: 0.1872\n",
            "Epoch [2/10], Step [8/101], Loss: 0.2987\n",
            "Epoch [2/10], Step [9/101], Loss: 0.2360\n",
            "Epoch [2/10], Step [10/101], Loss: 0.1348\n",
            "Epoch [2/10], Step [11/101], Loss: 0.1217\n",
            "Epoch [2/10], Step [12/101], Loss: 0.3614\n",
            "Epoch [2/10], Step [13/101], Loss: 0.0924\n",
            "Epoch [2/10], Step [14/101], Loss: 0.1829\n",
            "Epoch [2/10], Step [15/101], Loss: 0.1895\n",
            "Epoch [2/10], Step [16/101], Loss: 0.0795\n",
            "Epoch [2/10], Step [17/101], Loss: 0.1724\n",
            "Epoch [2/10], Step [18/101], Loss: 0.1071\n",
            "Epoch [2/10], Step [19/101], Loss: 0.0927\n",
            "Epoch [2/10], Step [20/101], Loss: 0.1067\n",
            "Epoch [2/10], Step [21/101], Loss: 0.2164\n",
            "Epoch [2/10], Step [22/101], Loss: 0.1560\n",
            "Epoch [2/10], Step [23/101], Loss: 0.0938\n",
            "Epoch [2/10], Step [24/101], Loss: 0.2360\n",
            "Epoch [2/10], Step [25/101], Loss: 0.1533\n",
            "Epoch [2/10], Step [26/101], Loss: 0.1165\n",
            "Epoch [2/10], Step [27/101], Loss: 0.1645\n",
            "Epoch [2/10], Step [28/101], Loss: 0.1373\n",
            "Epoch [2/10], Step [29/101], Loss: 0.1703\n",
            "Epoch [2/10], Step [30/101], Loss: 0.0864\n",
            "Epoch [2/10], Step [31/101], Loss: 0.1182\n",
            "Epoch [2/10], Step [32/101], Loss: 0.3626\n",
            "Epoch [2/10], Step [33/101], Loss: 0.1581\n",
            "Epoch [2/10], Step [34/101], Loss: 0.1284\n",
            "Epoch [2/10], Step [35/101], Loss: 0.1074\n",
            "Epoch [2/10], Step [36/101], Loss: 0.3276\n",
            "Epoch [2/10], Step [37/101], Loss: 0.1097\n",
            "Epoch [2/10], Step [38/101], Loss: 0.0825\n",
            "Epoch [2/10], Step [39/101], Loss: 0.1936\n",
            "Epoch [2/10], Step [40/101], Loss: 0.1067\n",
            "Epoch [2/10], Step [41/101], Loss: 0.2327\n",
            "Epoch [2/10], Step [42/101], Loss: 0.1367\n",
            "Epoch [2/10], Step [43/101], Loss: 0.2331\n",
            "Epoch [2/10], Step [44/101], Loss: 0.0941\n",
            "Epoch [2/10], Step [45/101], Loss: 0.0927\n",
            "Epoch [2/10], Step [46/101], Loss: 0.1300\n",
            "Epoch [2/10], Step [47/101], Loss: 0.1393\n",
            "Epoch [2/10], Step [48/101], Loss: 0.1329\n",
            "Epoch [2/10], Step [49/101], Loss: 0.2242\n",
            "Epoch [2/10], Step [50/101], Loss: 0.0959\n",
            "Epoch [2/10], Step [51/101], Loss: 0.1759\n",
            "Epoch [2/10], Step [52/101], Loss: 0.1257\n",
            "Epoch [2/10], Step [53/101], Loss: 0.1234\n",
            "Epoch [2/10], Step [54/101], Loss: 0.2153\n",
            "Epoch [2/10], Step [55/101], Loss: 0.0996\n",
            "Epoch [2/10], Step [56/101], Loss: 0.0702\n",
            "Epoch [2/10], Step [57/101], Loss: 0.1299\n",
            "Epoch [2/10], Step [58/101], Loss: 0.2072\n",
            "Epoch [2/10], Step [59/101], Loss: 0.0872\n",
            "Epoch [2/10], Step [60/101], Loss: 0.1756\n",
            "Epoch [2/10], Step [61/101], Loss: 0.1796\n",
            "Epoch [2/10], Step [62/101], Loss: 0.0807\n",
            "Epoch [2/10], Step [63/101], Loss: 0.1142\n",
            "Epoch [2/10], Step [64/101], Loss: 0.1851\n",
            "Epoch [2/10], Step [65/101], Loss: 0.0958\n",
            "Epoch [2/10], Step [66/101], Loss: 0.0790\n",
            "Epoch [2/10], Step [67/101], Loss: 0.1411\n",
            "Epoch [2/10], Step [68/101], Loss: 0.1288\n",
            "Epoch [2/10], Step [69/101], Loss: 0.1436\n",
            "Epoch [2/10], Step [70/101], Loss: 0.1890\n",
            "Epoch [2/10], Step [71/101], Loss: 0.1259\n",
            "Epoch [2/10], Step [72/101], Loss: 0.0906\n",
            "Epoch [2/10], Step [73/101], Loss: 0.0982\n",
            "Epoch [2/10], Step [74/101], Loss: 0.1509\n",
            "Epoch [2/10], Step [75/101], Loss: 0.1835\n",
            "Epoch [2/10], Step [76/101], Loss: 0.2303\n",
            "Epoch [2/10], Step [77/101], Loss: 0.2687\n",
            "Epoch [2/10], Step [78/101], Loss: 0.1768\n",
            "Epoch [2/10], Step [79/101], Loss: 0.1230\n",
            "Epoch [2/10], Step [80/101], Loss: 0.1181\n",
            "Epoch [2/10], Step [81/101], Loss: 0.1790\n",
            "Epoch [2/10], Step [82/101], Loss: 0.1726\n",
            "Epoch [2/10], Step [83/101], Loss: 0.1229\n",
            "Epoch [2/10], Step [84/101], Loss: 0.2877\n",
            "Epoch [2/10], Step [85/101], Loss: 0.1154\n",
            "Epoch [2/10], Step [86/101], Loss: 0.2525\n",
            "Epoch [2/10], Step [87/101], Loss: 0.2331\n",
            "Epoch [2/10], Step [88/101], Loss: 0.1720\n",
            "Epoch [2/10], Step [89/101], Loss: 0.1455\n",
            "Epoch [2/10], Step [90/101], Loss: 0.2405\n",
            "Epoch [2/10], Step [91/101], Loss: 0.1408\n",
            "Epoch [2/10], Step [92/101], Loss: 0.0874\n",
            "Epoch [2/10], Step [93/101], Loss: 0.1235\n",
            "Epoch [2/10], Step [94/101], Loss: 0.2029\n",
            "Epoch [2/10], Step [95/101], Loss: 0.1449\n",
            "Epoch [2/10], Step [96/101], Loss: 0.0959\n",
            "Epoch [2/10], Step [97/101], Loss: 0.1960\n",
            "Epoch [2/10], Step [98/101], Loss: 0.2323\n",
            "Epoch [2/10], Step [99/101], Loss: 0.0928\n",
            "Epoch [2/10], Step [100/101], Loss: 0.1932\n",
            "Epoch [2/10], Step [101/101], Loss: 0.4382\n",
            "Epoch finished\n",
            "Epoch [3/10], Step [1/101], Loss: 0.2304\n",
            "Epoch [3/10], Step [2/101], Loss: 0.0719\n",
            "Epoch [3/10], Step [3/101], Loss: 0.1340\n",
            "Epoch [3/10], Step [4/101], Loss: 0.1934\n",
            "Epoch [3/10], Step [5/101], Loss: 0.0977\n",
            "Epoch [3/10], Step [6/101], Loss: 0.2273\n",
            "Epoch [3/10], Step [7/101], Loss: 0.2238\n",
            "Epoch [3/10], Step [8/101], Loss: 0.1004\n",
            "Epoch [3/10], Step [9/101], Loss: 0.2530\n",
            "Epoch [3/10], Step [10/101], Loss: 0.1653\n",
            "Epoch [3/10], Step [11/101], Loss: 0.1323\n",
            "Epoch [3/10], Step [12/101], Loss: 0.0884\n",
            "Epoch [3/10], Step [13/101], Loss: 0.1271\n",
            "Epoch [3/10], Step [14/101], Loss: 0.1026\n",
            "Epoch [3/10], Step [15/101], Loss: 0.0669\n",
            "Epoch [3/10], Step [16/101], Loss: 0.2965\n",
            "Epoch [3/10], Step [17/101], Loss: 0.1952\n",
            "Epoch [3/10], Step [18/101], Loss: 0.1478\n",
            "Epoch [3/10], Step [19/101], Loss: 0.1220\n",
            "Epoch [3/10], Step [20/101], Loss: 0.3393\n",
            "Epoch [3/10], Step [21/101], Loss: 0.1330\n",
            "Epoch [3/10], Step [22/101], Loss: 0.1976\n",
            "Epoch [3/10], Step [23/101], Loss: 0.0889\n",
            "Epoch [3/10], Step [24/101], Loss: 0.2126\n",
            "Epoch [3/10], Step [25/101], Loss: 0.1826\n",
            "Epoch [3/10], Step [26/101], Loss: 0.1550\n",
            "Epoch [3/10], Step [27/101], Loss: 0.1155\n",
            "Epoch [3/10], Step [28/101], Loss: 0.1211\n",
            "Epoch [3/10], Step [29/101], Loss: 0.0486\n",
            "Epoch [3/10], Step [30/101], Loss: 0.0378\n",
            "Epoch [3/10], Step [31/101], Loss: 0.1263\n",
            "Epoch [3/10], Step [32/101], Loss: 0.2270\n",
            "Epoch [3/10], Step [33/101], Loss: 0.0854\n",
            "Epoch [3/10], Step [34/101], Loss: 0.1655\n",
            "Epoch [3/10], Step [35/101], Loss: 0.2369\n",
            "Epoch [3/10], Step [36/101], Loss: 0.1279\n",
            "Epoch [3/10], Step [37/101], Loss: 0.0754\n",
            "Epoch [3/10], Step [38/101], Loss: 0.1902\n",
            "Epoch [3/10], Step [39/101], Loss: 0.1277\n",
            "Epoch [3/10], Step [40/101], Loss: 0.1896\n",
            "Epoch [3/10], Step [41/101], Loss: 0.2248\n",
            "Epoch [3/10], Step [42/101], Loss: 0.1495\n",
            "Epoch [3/10], Step [43/101], Loss: 0.2360\n",
            "Epoch [3/10], Step [44/101], Loss: 0.1320\n",
            "Epoch [3/10], Step [45/101], Loss: 0.1860\n",
            "Epoch [3/10], Step [46/101], Loss: 0.0737\n",
            "Epoch [3/10], Step [47/101], Loss: 0.1391\n",
            "Epoch [3/10], Step [48/101], Loss: 0.1442\n",
            "Epoch [3/10], Step [49/101], Loss: 0.0719\n",
            "Epoch [3/10], Step [50/101], Loss: 0.2392\n",
            "Epoch [3/10], Step [51/101], Loss: 0.0745\n",
            "Epoch [3/10], Step [52/101], Loss: 0.2083\n",
            "Epoch [3/10], Step [53/101], Loss: 0.2481\n",
            "Epoch [3/10], Step [54/101], Loss: 0.3809\n",
            "Epoch [3/10], Step [55/101], Loss: 0.0540\n",
            "Epoch [3/10], Step [56/101], Loss: 0.1073\n",
            "Epoch [3/10], Step [57/101], Loss: 0.1579\n",
            "Epoch [3/10], Step [58/101], Loss: 0.2272\n",
            "Epoch [3/10], Step [59/101], Loss: 0.1039\n",
            "Epoch [3/10], Step [60/101], Loss: 0.1618\n",
            "Epoch [3/10], Step [61/101], Loss: 0.1149\n",
            "Epoch [3/10], Step [62/101], Loss: 0.0775\n",
            "Epoch [3/10], Step [63/101], Loss: 0.1412\n",
            "Epoch [3/10], Step [64/101], Loss: 0.1989\n",
            "Epoch [3/10], Step [65/101], Loss: 0.2885\n",
            "Epoch [3/10], Step [66/101], Loss: 0.1240\n",
            "Epoch [3/10], Step [67/101], Loss: 0.0953\n",
            "Epoch [3/10], Step [68/101], Loss: 0.1505\n",
            "Epoch [3/10], Step [69/101], Loss: 0.0714\n",
            "Epoch [3/10], Step [70/101], Loss: 0.2095\n",
            "Epoch [3/10], Step [71/101], Loss: 0.2752\n",
            "Epoch [3/10], Step [72/101], Loss: 0.3850\n",
            "Epoch [3/10], Step [73/101], Loss: 0.1063\n",
            "Epoch [3/10], Step [74/101], Loss: 0.0637\n",
            "Epoch [3/10], Step [75/101], Loss: 0.1020\n",
            "Epoch [3/10], Step [76/101], Loss: 0.1277\n",
            "Epoch [3/10], Step [77/101], Loss: 0.3016\n",
            "Epoch [3/10], Step [78/101], Loss: 0.1190\n",
            "Epoch [3/10], Step [79/101], Loss: 0.1326\n",
            "Epoch [3/10], Step [80/101], Loss: 0.1147\n",
            "Epoch [3/10], Step [81/101], Loss: 0.1634\n",
            "Epoch [3/10], Step [82/101], Loss: 0.1497\n",
            "Epoch [3/10], Step [83/101], Loss: 0.2192\n",
            "Epoch [3/10], Step [84/101], Loss: 0.1286\n",
            "Epoch [3/10], Step [85/101], Loss: 0.2265\n",
            "Epoch [3/10], Step [86/101], Loss: 0.0735\n",
            "Epoch [3/10], Step [87/101], Loss: 0.1823\n",
            "Epoch [3/10], Step [88/101], Loss: 0.2519\n",
            "Epoch [3/10], Step [89/101], Loss: 0.1098\n",
            "Epoch [3/10], Step [90/101], Loss: 0.0810\n",
            "Epoch [3/10], Step [91/101], Loss: 0.2475\n",
            "Epoch [3/10], Step [92/101], Loss: 0.1609\n",
            "Epoch [3/10], Step [93/101], Loss: 0.1175\n",
            "Epoch [3/10], Step [94/101], Loss: 0.2156\n",
            "Epoch [3/10], Step [95/101], Loss: 0.1793\n",
            "Epoch [3/10], Step [96/101], Loss: 0.1403\n",
            "Epoch [3/10], Step [97/101], Loss: 0.1186\n",
            "Epoch [3/10], Step [98/101], Loss: 0.0968\n",
            "Epoch [3/10], Step [99/101], Loss: 0.1670\n",
            "Epoch [3/10], Step [100/101], Loss: 0.2152\n",
            "Epoch [3/10], Step [101/101], Loss: 0.9976\n",
            "Epoch finished\n",
            "Epoch [4/10], Step [1/101], Loss: 0.1624\n",
            "Epoch [4/10], Step [2/101], Loss: 0.1654\n",
            "Epoch [4/10], Step [3/101], Loss: 0.0979\n",
            "Epoch [4/10], Step [4/101], Loss: 0.1166\n",
            "Epoch [4/10], Step [5/101], Loss: 0.0814\n",
            "Epoch [4/10], Step [6/101], Loss: 0.4026\n",
            "Epoch [4/10], Step [7/101], Loss: 0.0688\n",
            "Epoch [4/10], Step [8/101], Loss: 0.0439\n",
            "Epoch [4/10], Step [9/101], Loss: 0.1396\n",
            "Epoch [4/10], Step [10/101], Loss: 0.0887\n",
            "Epoch [4/10], Step [11/101], Loss: 0.0334\n",
            "Epoch [4/10], Step [12/101], Loss: 0.0990\n",
            "Epoch [4/10], Step [13/101], Loss: 0.0935\n",
            "Epoch [4/10], Step [14/101], Loss: 0.1813\n",
            "Epoch [4/10], Step [15/101], Loss: 0.0988\n",
            "Epoch [4/10], Step [16/101], Loss: 0.2117\n",
            "Epoch [4/10], Step [17/101], Loss: 0.1802\n",
            "Epoch [4/10], Step [18/101], Loss: 0.1474\n",
            "Epoch [4/10], Step [19/101], Loss: 0.2373\n",
            "Epoch [4/10], Step [20/101], Loss: 0.1694\n",
            "Epoch [4/10], Step [21/101], Loss: 0.1006\n",
            "Epoch [4/10], Step [22/101], Loss: 0.2396\n",
            "Epoch [4/10], Step [23/101], Loss: 0.0575\n",
            "Epoch [4/10], Step [24/101], Loss: 0.0733\n",
            "Epoch [4/10], Step [25/101], Loss: 0.0839\n",
            "Epoch [4/10], Step [26/101], Loss: 0.1538\n",
            "Epoch [4/10], Step [27/101], Loss: 0.0820\n",
            "Epoch [4/10], Step [28/101], Loss: 0.2166\n",
            "Epoch [4/10], Step [29/101], Loss: 0.0331\n",
            "Epoch [4/10], Step [30/101], Loss: 0.1401\n",
            "Epoch [4/10], Step [31/101], Loss: 0.0888\n",
            "Epoch [4/10], Step [32/101], Loss: 0.1695\n",
            "Epoch [4/10], Step [33/101], Loss: 0.0712\n",
            "Epoch [4/10], Step [34/101], Loss: 0.1684\n",
            "Epoch [4/10], Step [35/101], Loss: 0.2002\n",
            "Epoch [4/10], Step [36/101], Loss: 0.1530\n",
            "Epoch [4/10], Step [37/101], Loss: 0.4797\n",
            "Epoch [4/10], Step [38/101], Loss: 0.1737\n",
            "Epoch [4/10], Step [39/101], Loss: 0.1893\n",
            "Epoch [4/10], Step [40/101], Loss: 0.2549\n",
            "Epoch [4/10], Step [41/101], Loss: 0.0929\n",
            "Epoch [4/10], Step [42/101], Loss: 0.1480\n",
            "Epoch [4/10], Step [43/101], Loss: 0.1016\n",
            "Epoch [4/10], Step [44/101], Loss: 0.2476\n",
            "Epoch [4/10], Step [45/101], Loss: 0.2072\n",
            "Epoch [4/10], Step [46/101], Loss: 0.1144\n",
            "Epoch [4/10], Step [47/101], Loss: 0.3142\n",
            "Epoch [4/10], Step [48/101], Loss: 0.0718\n",
            "Epoch [4/10], Step [49/101], Loss: 0.1942\n",
            "Epoch [4/10], Step [50/101], Loss: 0.2868\n",
            "Epoch [4/10], Step [51/101], Loss: 0.1674\n",
            "Epoch [4/10], Step [52/101], Loss: 0.0475\n",
            "Epoch [4/10], Step [53/101], Loss: 0.1761\n",
            "Epoch [4/10], Step [54/101], Loss: 0.1573\n",
            "Epoch [4/10], Step [55/101], Loss: 0.1658\n",
            "Epoch [4/10], Step [56/101], Loss: 0.0856\n",
            "Epoch [4/10], Step [57/101], Loss: 0.1265\n",
            "Epoch [4/10], Step [58/101], Loss: 0.2213\n",
            "Epoch [4/10], Step [59/101], Loss: 0.1804\n",
            "Epoch [4/10], Step [60/101], Loss: 0.1082\n",
            "Epoch [4/10], Step [61/101], Loss: 0.0613\n",
            "Epoch [4/10], Step [62/101], Loss: 0.1417\n",
            "Epoch [4/10], Step [63/101], Loss: 0.1507\n",
            "Epoch [4/10], Step [64/101], Loss: 0.1319\n",
            "Epoch [4/10], Step [65/101], Loss: 0.1779\n",
            "Epoch [4/10], Step [66/101], Loss: 0.1695\n",
            "Epoch [4/10], Step [67/101], Loss: 0.1549\n",
            "Epoch [4/10], Step [68/101], Loss: 0.1126\n",
            "Epoch [4/10], Step [69/101], Loss: 0.0500\n",
            "Epoch [4/10], Step [70/101], Loss: 0.1894\n",
            "Epoch [4/10], Step [71/101], Loss: 0.1533\n",
            "Epoch [4/10], Step [72/101], Loss: 0.0717\n",
            "Epoch [4/10], Step [73/101], Loss: 0.1317\n",
            "Epoch [4/10], Step [74/101], Loss: 0.0800\n",
            "Epoch [4/10], Step [75/101], Loss: 0.1025\n",
            "Epoch [4/10], Step [76/101], Loss: 0.1001\n",
            "Epoch [4/10], Step [77/101], Loss: 0.1103\n",
            "Epoch [4/10], Step [78/101], Loss: 0.2902\n",
            "Epoch [4/10], Step [79/101], Loss: 0.0592\n",
            "Epoch [4/10], Step [80/101], Loss: 0.1345\n",
            "Epoch [4/10], Step [81/101], Loss: 0.1486\n",
            "Epoch [4/10], Step [82/101], Loss: 0.1652\n",
            "Epoch [4/10], Step [83/101], Loss: 0.2125\n",
            "Epoch [4/10], Step [84/101], Loss: 0.0635\n",
            "Epoch [4/10], Step [85/101], Loss: 0.2215\n",
            "Epoch [4/10], Step [86/101], Loss: 0.2080\n",
            "Epoch [4/10], Step [87/101], Loss: 0.1829\n",
            "Epoch [4/10], Step [88/101], Loss: 0.1663\n",
            "Epoch [4/10], Step [89/101], Loss: 0.1816\n",
            "Epoch [4/10], Step [90/101], Loss: 0.1372\n",
            "Epoch [4/10], Step [91/101], Loss: 0.2387\n",
            "Epoch [4/10], Step [92/101], Loss: 0.1065\n",
            "Epoch [4/10], Step [93/101], Loss: 0.0818\n",
            "Epoch [4/10], Step [94/101], Loss: 0.1555\n",
            "Epoch [4/10], Step [95/101], Loss: 0.1170\n",
            "Epoch [4/10], Step [96/101], Loss: 0.1590\n",
            "Epoch [4/10], Step [97/101], Loss: 0.1464\n",
            "Epoch [4/10], Step [98/101], Loss: 0.1507\n",
            "Epoch [4/10], Step [99/101], Loss: 0.1782\n",
            "Epoch [4/10], Step [100/101], Loss: 0.1665\n",
            "Epoch [4/10], Step [101/101], Loss: 0.2557\n",
            "Epoch finished\n",
            "Epoch [5/10], Step [1/101], Loss: 0.2591\n",
            "Epoch [5/10], Step [2/101], Loss: 0.0892\n",
            "Epoch [5/10], Step [3/101], Loss: 0.2654\n",
            "Epoch [5/10], Step [4/101], Loss: 0.1064\n",
            "Epoch [5/10], Step [5/101], Loss: 0.1072\n",
            "Epoch [5/10], Step [6/101], Loss: 0.0782\n",
            "Epoch [5/10], Step [7/101], Loss: 0.1726\n",
            "Epoch [5/10], Step [8/101], Loss: 0.1942\n",
            "Epoch [5/10], Step [9/101], Loss: 0.0657\n",
            "Epoch [5/10], Step [10/101], Loss: 0.1496\n",
            "Epoch [5/10], Step [11/101], Loss: 0.0857\n",
            "Epoch [5/10], Step [12/101], Loss: 0.0579\n",
            "Epoch [5/10], Step [13/101], Loss: 0.1028\n",
            "Epoch [5/10], Step [14/101], Loss: 0.1386\n",
            "Epoch [5/10], Step [15/101], Loss: 0.0961\n",
            "Epoch [5/10], Step [16/101], Loss: 0.0658\n",
            "Epoch [5/10], Step [17/101], Loss: 0.1866\n",
            "Epoch [5/10], Step [18/101], Loss: 0.1362\n",
            "Epoch [5/10], Step [19/101], Loss: 0.2059\n",
            "Epoch [5/10], Step [20/101], Loss: 0.2175\n",
            "Epoch [5/10], Step [21/101], Loss: 0.0551\n",
            "Epoch [5/10], Step [22/101], Loss: 0.0794\n",
            "Epoch [5/10], Step [23/101], Loss: 0.0980\n",
            "Epoch [5/10], Step [24/101], Loss: 0.1715\n",
            "Epoch [5/10], Step [25/101], Loss: 0.1602\n",
            "Epoch [5/10], Step [26/101], Loss: 0.1859\n",
            "Epoch [5/10], Step [27/101], Loss: 0.1807\n",
            "Epoch [5/10], Step [28/101], Loss: 0.1325\n",
            "Epoch [5/10], Step [29/101], Loss: 0.0894\n",
            "Epoch [5/10], Step [30/101], Loss: 0.1720\n",
            "Epoch [5/10], Step [31/101], Loss: 0.1165\n",
            "Epoch [5/10], Step [32/101], Loss: 0.0908\n",
            "Epoch [5/10], Step [33/101], Loss: 0.1924\n",
            "Epoch [5/10], Step [34/101], Loss: 0.2010\n",
            "Epoch [5/10], Step [35/101], Loss: 0.1042\n",
            "Epoch [5/10], Step [36/101], Loss: 0.1519\n",
            "Epoch [5/10], Step [37/101], Loss: 0.1891\n",
            "Epoch [5/10], Step [38/101], Loss: 0.0650\n",
            "Epoch [5/10], Step [39/101], Loss: 0.1980\n",
            "Epoch [5/10], Step [40/101], Loss: 0.0773\n",
            "Epoch [5/10], Step [41/101], Loss: 0.1315\n",
            "Epoch [5/10], Step [42/101], Loss: 0.1288\n",
            "Epoch [5/10], Step [43/101], Loss: 0.2352\n",
            "Epoch [5/10], Step [44/101], Loss: 0.1059\n",
            "Epoch [5/10], Step [45/101], Loss: 0.0968\n",
            "Epoch [5/10], Step [46/101], Loss: 0.0796\n",
            "Epoch [5/10], Step [47/101], Loss: 0.1607\n",
            "Epoch [5/10], Step [48/101], Loss: 0.1676\n",
            "Epoch [5/10], Step [49/101], Loss: 0.1339\n",
            "Epoch [5/10], Step [50/101], Loss: 0.2482\n",
            "Epoch [5/10], Step [51/101], Loss: 0.1109\n",
            "Epoch [5/10], Step [52/101], Loss: 0.1589\n",
            "Epoch [5/10], Step [53/101], Loss: 0.0761\n",
            "Epoch [5/10], Step [54/101], Loss: 0.0778\n",
            "Epoch [5/10], Step [55/101], Loss: 0.1560\n",
            "Epoch [5/10], Step [56/101], Loss: 0.1301\n",
            "Epoch [5/10], Step [57/101], Loss: 0.0901\n",
            "Epoch [5/10], Step [58/101], Loss: 0.2125\n",
            "Epoch [5/10], Step [59/101], Loss: 0.1534\n",
            "Epoch [5/10], Step [60/101], Loss: 0.2155\n",
            "Epoch [5/10], Step [61/101], Loss: 0.2468\n",
            "Epoch [5/10], Step [62/101], Loss: 0.2132\n",
            "Epoch [5/10], Step [63/101], Loss: 0.0440\n",
            "Epoch [5/10], Step [64/101], Loss: 0.1840\n",
            "Epoch [5/10], Step [65/101], Loss: 0.1191\n",
            "Epoch [5/10], Step [66/101], Loss: 0.0942\n",
            "Epoch [5/10], Step [67/101], Loss: 0.2172\n",
            "Epoch [5/10], Step [68/101], Loss: 0.0757\n",
            "Epoch [5/10], Step [69/101], Loss: 0.1426\n",
            "Epoch [5/10], Step [70/101], Loss: 0.1160\n",
            "Epoch [5/10], Step [71/101], Loss: 0.1051\n",
            "Epoch [5/10], Step [72/101], Loss: 0.0745\n",
            "Epoch [5/10], Step [73/101], Loss: 0.2149\n",
            "Epoch [5/10], Step [74/101], Loss: 0.1429\n",
            "Epoch [5/10], Step [75/101], Loss: 0.2714\n",
            "Epoch [5/10], Step [76/101], Loss: 0.2199\n",
            "Epoch [5/10], Step [77/101], Loss: 0.1559\n",
            "Epoch [5/10], Step [78/101], Loss: 0.1580\n",
            "Epoch [5/10], Step [79/101], Loss: 0.0622\n",
            "Epoch [5/10], Step [80/101], Loss: 0.1017\n",
            "Epoch [5/10], Step [81/101], Loss: 0.2064\n",
            "Epoch [5/10], Step [82/101], Loss: 0.1871\n",
            "Epoch [5/10], Step [83/101], Loss: 0.1143\n",
            "Epoch [5/10], Step [84/101], Loss: 0.1085\n",
            "Epoch [5/10], Step [85/101], Loss: 0.0433\n",
            "Epoch [5/10], Step [86/101], Loss: 0.2069\n",
            "Epoch [5/10], Step [87/101], Loss: 0.1194\n",
            "Epoch [5/10], Step [88/101], Loss: 0.1808\n",
            "Epoch [5/10], Step [89/101], Loss: 0.1760\n",
            "Epoch [5/10], Step [90/101], Loss: 0.1080\n",
            "Epoch [5/10], Step [91/101], Loss: 0.0599\n",
            "Epoch [5/10], Step [92/101], Loss: 0.2281\n",
            "Epoch [5/10], Step [93/101], Loss: 0.0318\n",
            "Epoch [5/10], Step [94/101], Loss: 0.1610\n",
            "Epoch [5/10], Step [95/101], Loss: 0.0854\n",
            "Epoch [5/10], Step [96/101], Loss: 0.0760\n",
            "Epoch [5/10], Step [97/101], Loss: 0.1455\n",
            "Epoch [5/10], Step [98/101], Loss: 0.2327\n",
            "Epoch [5/10], Step [99/101], Loss: 0.2000\n",
            "Epoch [5/10], Step [100/101], Loss: 0.0685\n",
            "Epoch [5/10], Step [101/101], Loss: 0.2561\n",
            "Epoch finished\n",
            "Epoch [6/10], Step [1/101], Loss: 0.2121\n",
            "Epoch [6/10], Step [2/101], Loss: 0.2068\n",
            "Epoch [6/10], Step [3/101], Loss: 0.1869\n",
            "Epoch [6/10], Step [4/101], Loss: 0.1151\n",
            "Epoch [6/10], Step [5/101], Loss: 0.1825\n",
            "Epoch [6/10], Step [6/101], Loss: 0.1230\n",
            "Epoch [6/10], Step [7/101], Loss: 0.2021\n",
            "Epoch [6/10], Step [8/101], Loss: 0.1380\n",
            "Epoch [6/10], Step [9/101], Loss: 0.1057\n",
            "Epoch [6/10], Step [10/101], Loss: 0.1516\n",
            "Epoch [6/10], Step [11/101], Loss: 0.0848\n",
            "Epoch [6/10], Step [12/101], Loss: 0.1070\n",
            "Epoch [6/10], Step [13/101], Loss: 0.1147\n",
            "Epoch [6/10], Step [14/101], Loss: 0.0691\n",
            "Epoch [6/10], Step [15/101], Loss: 0.3557\n",
            "Epoch [6/10], Step [16/101], Loss: 0.1417\n",
            "Epoch [6/10], Step [17/101], Loss: 0.0649\n",
            "Epoch [6/10], Step [18/101], Loss: 0.0793\n",
            "Epoch [6/10], Step [19/101], Loss: 0.2124\n",
            "Epoch [6/10], Step [20/101], Loss: 0.0960\n",
            "Epoch [6/10], Step [21/101], Loss: 0.1608\n",
            "Epoch [6/10], Step [22/101], Loss: 0.1891\n",
            "Epoch [6/10], Step [23/101], Loss: 0.0250\n",
            "Epoch [6/10], Step [24/101], Loss: 0.1099\n",
            "Epoch [6/10], Step [25/101], Loss: 0.1256\n",
            "Epoch [6/10], Step [26/101], Loss: 0.1548\n",
            "Epoch [6/10], Step [27/101], Loss: 0.0820\n",
            "Epoch [6/10], Step [28/101], Loss: 0.0534\n",
            "Epoch [6/10], Step [29/101], Loss: 0.1329\n",
            "Epoch [6/10], Step [30/101], Loss: 0.0994\n",
            "Epoch [6/10], Step [31/101], Loss: 0.0690\n",
            "Epoch [6/10], Step [32/101], Loss: 0.2572\n",
            "Epoch [6/10], Step [33/101], Loss: 0.0648\n",
            "Epoch [6/10], Step [34/101], Loss: 0.1459\n",
            "Epoch [6/10], Step [35/101], Loss: 0.0853\n",
            "Epoch [6/10], Step [36/101], Loss: 0.0715\n",
            "Epoch [6/10], Step [37/101], Loss: 0.1620\n",
            "Epoch [6/10], Step [38/101], Loss: 0.0898\n",
            "Epoch [6/10], Step [39/101], Loss: 0.0965\n",
            "Epoch [6/10], Step [40/101], Loss: 0.0797\n",
            "Epoch [6/10], Step [41/101], Loss: 0.1064\n",
            "Epoch [6/10], Step [42/101], Loss: 0.1450\n",
            "Epoch [6/10], Step [43/101], Loss: 0.0467\n",
            "Epoch [6/10], Step [44/101], Loss: 0.1357\n",
            "Epoch [6/10], Step [45/101], Loss: 0.0692\n",
            "Epoch [6/10], Step [46/101], Loss: 0.1643\n",
            "Epoch [6/10], Step [47/101], Loss: 0.1529\n",
            "Epoch [6/10], Step [48/101], Loss: 0.0463\n",
            "Epoch [6/10], Step [49/101], Loss: 0.0851\n",
            "Epoch [6/10], Step [50/101], Loss: 0.1712\n",
            "Epoch [6/10], Step [51/101], Loss: 0.1190\n",
            "Epoch [6/10], Step [52/101], Loss: 0.1086\n",
            "Epoch [6/10], Step [53/101], Loss: 0.0416\n",
            "Epoch [6/10], Step [54/101], Loss: 0.1458\n",
            "Epoch [6/10], Step [55/101], Loss: 0.2048\n",
            "Epoch [6/10], Step [56/101], Loss: 0.2066\n",
            "Epoch [6/10], Step [57/101], Loss: 0.0604\n",
            "Epoch [6/10], Step [58/101], Loss: 0.0645\n",
            "Epoch [6/10], Step [59/101], Loss: 0.2959\n",
            "Epoch [6/10], Step [60/101], Loss: 0.1074\n",
            "Epoch [6/10], Step [61/101], Loss: 0.1352\n",
            "Epoch [6/10], Step [62/101], Loss: 0.2594\n",
            "Epoch [6/10], Step [63/101], Loss: 0.1217\n",
            "Epoch [6/10], Step [64/101], Loss: 0.1525\n",
            "Epoch [6/10], Step [65/101], Loss: 0.1504\n",
            "Epoch [6/10], Step [66/101], Loss: 0.1421\n",
            "Epoch [6/10], Step [67/101], Loss: 0.1071\n",
            "Epoch [6/10], Step [68/101], Loss: 0.0847\n",
            "Epoch [6/10], Step [69/101], Loss: 0.1490\n",
            "Epoch [6/10], Step [70/101], Loss: 0.1840\n",
            "Epoch [6/10], Step [71/101], Loss: 0.0879\n",
            "Epoch [6/10], Step [72/101], Loss: 0.2038\n",
            "Epoch [6/10], Step [73/101], Loss: 0.0966\n",
            "Epoch [6/10], Step [74/101], Loss: 0.1757\n",
            "Epoch [6/10], Step [75/101], Loss: 0.1451\n",
            "Epoch [6/10], Step [76/101], Loss: 0.1007\n",
            "Epoch [6/10], Step [77/101], Loss: 0.1938\n",
            "Epoch [6/10], Step [78/101], Loss: 0.0756\n",
            "Epoch [6/10], Step [79/101], Loss: 0.1207\n",
            "Epoch [6/10], Step [80/101], Loss: 0.3188\n",
            "Epoch [6/10], Step [81/101], Loss: 0.0777\n",
            "Epoch [6/10], Step [82/101], Loss: 0.1780\n",
            "Epoch [6/10], Step [83/101], Loss: 0.1260\n",
            "Epoch [6/10], Step [84/101], Loss: 0.1241\n",
            "Epoch [6/10], Step [85/101], Loss: 0.0818\n",
            "Epoch [6/10], Step [86/101], Loss: 0.0878\n",
            "Epoch [6/10], Step [87/101], Loss: 0.2462\n",
            "Epoch [6/10], Step [88/101], Loss: 0.0992\n",
            "Epoch [6/10], Step [89/101], Loss: 0.0647\n",
            "Epoch [6/10], Step [90/101], Loss: 0.1060\n",
            "Epoch [6/10], Step [91/101], Loss: 0.2269\n",
            "Epoch [6/10], Step [92/101], Loss: 0.1064\n",
            "Epoch [6/10], Step [93/101], Loss: 0.2633\n",
            "Epoch [6/10], Step [94/101], Loss: 0.1654\n",
            "Epoch [6/10], Step [95/101], Loss: 0.0864\n",
            "Epoch [6/10], Step [96/101], Loss: 0.0952\n",
            "Epoch [6/10], Step [97/101], Loss: 0.1846\n",
            "Epoch [6/10], Step [98/101], Loss: 0.0971\n",
            "Epoch [6/10], Step [99/101], Loss: 0.0315\n",
            "Epoch [6/10], Step [100/101], Loss: 0.1367\n",
            "Epoch [6/10], Step [101/101], Loss: 0.2509\n",
            "Epoch finished\n",
            "Epoch [7/10], Step [1/101], Loss: 0.2272\n",
            "Epoch [7/10], Step [2/101], Loss: 0.1036\n",
            "Epoch [7/10], Step [3/101], Loss: 0.0559\n",
            "Epoch [7/10], Step [4/101], Loss: 0.2476\n",
            "Epoch [7/10], Step [5/101], Loss: 0.1518\n",
            "Epoch [7/10], Step [6/101], Loss: 0.1664\n",
            "Epoch [7/10], Step [7/101], Loss: 0.0880\n",
            "Epoch [7/10], Step [8/101], Loss: 0.0914\n",
            "Epoch [7/10], Step [9/101], Loss: 0.1407\n",
            "Epoch [7/10], Step [10/101], Loss: 0.0855\n",
            "Epoch [7/10], Step [11/101], Loss: 0.1889\n",
            "Epoch [7/10], Step [12/101], Loss: 0.0194\n",
            "Epoch [7/10], Step [13/101], Loss: 0.0485\n",
            "Epoch [7/10], Step [14/101], Loss: 0.1193\n",
            "Epoch [7/10], Step [15/101], Loss: 0.2878\n",
            "Epoch [7/10], Step [16/101], Loss: 0.1679\n",
            "Epoch [7/10], Step [17/101], Loss: 0.0679\n",
            "Epoch [7/10], Step [18/101], Loss: 0.1250\n",
            "Epoch [7/10], Step [19/101], Loss: 0.0999\n",
            "Epoch [7/10], Step [20/101], Loss: 0.1586\n",
            "Epoch [7/10], Step [21/101], Loss: 0.1034\n",
            "Epoch [7/10], Step [22/101], Loss: 0.1381\n",
            "Epoch [7/10], Step [23/101], Loss: 0.0911\n",
            "Epoch [7/10], Step [24/101], Loss: 0.1240\n",
            "Epoch [7/10], Step [25/101], Loss: 0.0712\n",
            "Epoch [7/10], Step [26/101], Loss: 0.2703\n",
            "Epoch [7/10], Step [27/101], Loss: 0.0754\n",
            "Epoch [7/10], Step [28/101], Loss: 0.0850\n",
            "Epoch [7/10], Step [29/101], Loss: 0.0787\n",
            "Epoch [7/10], Step [30/101], Loss: 0.1613\n",
            "Epoch [7/10], Step [31/101], Loss: 0.1135\n",
            "Epoch [7/10], Step [32/101], Loss: 0.2480\n",
            "Epoch [7/10], Step [33/101], Loss: 0.0752\n",
            "Epoch [7/10], Step [34/101], Loss: 0.0639\n",
            "Epoch [7/10], Step [35/101], Loss: 0.1207\n",
            "Epoch [7/10], Step [36/101], Loss: 0.0823\n",
            "Epoch [7/10], Step [37/101], Loss: 0.0631\n",
            "Epoch [7/10], Step [38/101], Loss: 0.1047\n",
            "Epoch [7/10], Step [39/101], Loss: 0.1450\n",
            "Epoch [7/10], Step [40/101], Loss: 0.1790\n",
            "Epoch [7/10], Step [41/101], Loss: 0.0745\n",
            "Epoch [7/10], Step [42/101], Loss: 0.2161\n",
            "Epoch [7/10], Step [43/101], Loss: 0.1438\n",
            "Epoch [7/10], Step [44/101], Loss: 0.2068\n",
            "Epoch [7/10], Step [45/101], Loss: 0.1326\n",
            "Epoch [7/10], Step [46/101], Loss: 0.1001\n",
            "Epoch [7/10], Step [47/101], Loss: 0.1495\n",
            "Epoch [7/10], Step [48/101], Loss: 0.1425\n",
            "Epoch [7/10], Step [49/101], Loss: 0.2035\n",
            "Epoch [7/10], Step [50/101], Loss: 0.0230\n",
            "Epoch [7/10], Step [51/101], Loss: 0.2081\n",
            "Epoch [7/10], Step [52/101], Loss: 0.3317\n",
            "Epoch [7/10], Step [53/101], Loss: 0.1741\n",
            "Epoch [7/10], Step [54/101], Loss: 0.2154\n",
            "Epoch [7/10], Step [55/101], Loss: 0.1082\n",
            "Epoch [7/10], Step [56/101], Loss: 0.0556\n",
            "Epoch [7/10], Step [57/101], Loss: 0.1995\n",
            "Epoch [7/10], Step [58/101], Loss: 0.1926\n",
            "Epoch [7/10], Step [59/101], Loss: 0.3345\n",
            "Epoch [7/10], Step [60/101], Loss: 0.0654\n",
            "Epoch [7/10], Step [61/101], Loss: 0.1060\n",
            "Epoch [7/10], Step [62/101], Loss: 0.1074\n",
            "Epoch [7/10], Step [63/101], Loss: 0.2738\n",
            "Epoch [7/10], Step [64/101], Loss: 0.0850\n",
            "Epoch [7/10], Step [65/101], Loss: 0.1186\n",
            "Epoch [7/10], Step [66/101], Loss: 0.1448\n",
            "Epoch [7/10], Step [67/101], Loss: 0.0837\n",
            "Epoch [7/10], Step [68/101], Loss: 0.1681\n",
            "Epoch [7/10], Step [69/101], Loss: 0.0693\n",
            "Epoch [7/10], Step [70/101], Loss: 0.1042\n",
            "Epoch [7/10], Step [71/101], Loss: 0.1407\n",
            "Epoch [7/10], Step [72/101], Loss: 0.1900\n",
            "Epoch [7/10], Step [73/101], Loss: 0.1262\n",
            "Epoch [7/10], Step [74/101], Loss: 0.2404\n",
            "Epoch [7/10], Step [75/101], Loss: 0.1901\n",
            "Epoch [7/10], Step [76/101], Loss: 0.0706\n",
            "Epoch [7/10], Step [77/101], Loss: 0.1053\n",
            "Epoch [7/10], Step [78/101], Loss: 0.0919\n",
            "Epoch [7/10], Step [79/101], Loss: 0.2371\n",
            "Epoch [7/10], Step [80/101], Loss: 0.1440\n",
            "Epoch [7/10], Step [81/101], Loss: 0.1465\n",
            "Epoch [7/10], Step [82/101], Loss: 0.0899\n",
            "Epoch [7/10], Step [83/101], Loss: 0.2220\n",
            "Epoch [7/10], Step [84/101], Loss: 0.2306\n",
            "Epoch [7/10], Step [85/101], Loss: 0.1079\n",
            "Epoch [7/10], Step [86/101], Loss: 0.1242\n",
            "Epoch [7/10], Step [87/101], Loss: 0.2719\n",
            "Epoch [7/10], Step [88/101], Loss: 0.2229\n",
            "Epoch [7/10], Step [89/101], Loss: 0.3016\n",
            "Epoch [7/10], Step [90/101], Loss: 0.0818\n",
            "Epoch [7/10], Step [91/101], Loss: 0.1250\n",
            "Epoch [7/10], Step [92/101], Loss: 0.2734\n",
            "Epoch [7/10], Step [93/101], Loss: 0.1010\n",
            "Epoch [7/10], Step [94/101], Loss: 0.1050\n",
            "Epoch [7/10], Step [95/101], Loss: 0.1805\n",
            "Epoch [7/10], Step [96/101], Loss: 0.1801\n",
            "Epoch [7/10], Step [97/101], Loss: 0.1331\n",
            "Epoch [7/10], Step [98/101], Loss: 0.0878\n",
            "Epoch [7/10], Step [99/101], Loss: 0.1057\n",
            "Epoch [7/10], Step [100/101], Loss: 0.1184\n",
            "Epoch [7/10], Step [101/101], Loss: 0.4610\n",
            "Epoch finished\n",
            "Epoch [8/10], Step [1/101], Loss: 0.0529\n",
            "Epoch [8/10], Step [2/101], Loss: 0.1634\n",
            "Epoch [8/10], Step [3/101], Loss: 0.1998\n",
            "Epoch [8/10], Step [4/101], Loss: 0.0341\n",
            "Epoch [8/10], Step [5/101], Loss: 0.1230\n",
            "Epoch [8/10], Step [6/101], Loss: 0.1314\n",
            "Epoch [8/10], Step [7/101], Loss: 0.1413\n",
            "Epoch [8/10], Step [8/101], Loss: 0.1594\n",
            "Epoch [8/10], Step [9/101], Loss: 0.0502\n",
            "Epoch [8/10], Step [10/101], Loss: 0.1137\n",
            "Epoch [8/10], Step [11/101], Loss: 0.0525\n",
            "Epoch [8/10], Step [12/101], Loss: 0.0941\n",
            "Epoch [8/10], Step [13/101], Loss: 0.1201\n",
            "Epoch [8/10], Step [14/101], Loss: 0.1013\n",
            "Epoch [8/10], Step [15/101], Loss: 0.1104\n",
            "Epoch [8/10], Step [16/101], Loss: 0.0683\n",
            "Epoch [8/10], Step [17/101], Loss: 0.0374\n",
            "Epoch [8/10], Step [18/101], Loss: 0.1489\n",
            "Epoch [8/10], Step [19/101], Loss: 0.0577\n",
            "Epoch [8/10], Step [20/101], Loss: 0.0805\n",
            "Epoch [8/10], Step [21/101], Loss: 0.1653\n",
            "Epoch [8/10], Step [22/101], Loss: 0.1965\n",
            "Epoch [8/10], Step [23/101], Loss: 0.2029\n",
            "Epoch [8/10], Step [24/101], Loss: 0.0690\n",
            "Epoch [8/10], Step [25/101], Loss: 0.1287\n",
            "Epoch [8/10], Step [26/101], Loss: 0.1159\n",
            "Epoch [8/10], Step [27/101], Loss: 0.1680\n",
            "Epoch [8/10], Step [28/101], Loss: 0.0474\n",
            "Epoch [8/10], Step [29/101], Loss: 0.0917\n",
            "Epoch [8/10], Step [30/101], Loss: 0.1204\n",
            "Epoch [8/10], Step [31/101], Loss: 0.0246\n",
            "Epoch [8/10], Step [32/101], Loss: 0.0736\n",
            "Epoch [8/10], Step [33/101], Loss: 0.1693\n",
            "Epoch [8/10], Step [34/101], Loss: 0.0537\n",
            "Epoch [8/10], Step [35/101], Loss: 0.1151\n",
            "Epoch [8/10], Step [36/101], Loss: 0.2121\n",
            "Epoch [8/10], Step [37/101], Loss: 0.0270\n",
            "Epoch [8/10], Step [38/101], Loss: 0.2637\n",
            "Epoch [8/10], Step [39/101], Loss: 0.1716\n",
            "Epoch [8/10], Step [40/101], Loss: 0.0939\n",
            "Epoch [8/10], Step [41/101], Loss: 0.0794\n",
            "Epoch [8/10], Step [42/101], Loss: 0.0235\n",
            "Epoch [8/10], Step [43/101], Loss: 0.0662\n",
            "Epoch [8/10], Step [44/101], Loss: 0.0655\n",
            "Epoch [8/10], Step [45/101], Loss: 0.0883\n",
            "Epoch [8/10], Step [46/101], Loss: 0.0383\n",
            "Epoch [8/10], Step [47/101], Loss: 0.1902\n",
            "Epoch [8/10], Step [48/101], Loss: 0.3232\n",
            "Epoch [8/10], Step [49/101], Loss: 0.1646\n",
            "Epoch [8/10], Step [50/101], Loss: 0.0334\n",
            "Epoch [8/10], Step [51/101], Loss: 0.0581\n",
            "Epoch [8/10], Step [52/101], Loss: 0.0231\n",
            "Epoch [8/10], Step [53/101], Loss: 0.2263\n",
            "Epoch [8/10], Step [54/101], Loss: 0.0603\n",
            "Epoch [8/10], Step [55/101], Loss: 0.1124\n",
            "Epoch [8/10], Step [56/101], Loss: 0.0256\n",
            "Epoch [8/10], Step [57/101], Loss: 0.1463\n",
            "Epoch [8/10], Step [58/101], Loss: 0.2485\n",
            "Epoch [8/10], Step [59/101], Loss: 0.0873\n",
            "Epoch [8/10], Step [60/101], Loss: 0.1373\n",
            "Epoch [8/10], Step [61/101], Loss: 0.1246\n",
            "Epoch [8/10], Step [62/101], Loss: 0.1235\n",
            "Epoch [8/10], Step [63/101], Loss: 0.1971\n",
            "Epoch [8/10], Step [64/101], Loss: 0.0839\n",
            "Epoch [8/10], Step [65/101], Loss: 0.0397\n",
            "Epoch [8/10], Step [66/101], Loss: 0.1094\n",
            "Epoch [8/10], Step [67/101], Loss: 0.1419\n",
            "Epoch [8/10], Step [68/101], Loss: 0.1140\n",
            "Epoch [8/10], Step [69/101], Loss: 0.0584\n",
            "Epoch [8/10], Step [70/101], Loss: 0.2650\n",
            "Epoch [8/10], Step [71/101], Loss: 0.0958\n",
            "Epoch [8/10], Step [72/101], Loss: 0.1258\n",
            "Epoch [8/10], Step [73/101], Loss: 0.3867\n",
            "Epoch [8/10], Step [74/101], Loss: 0.0770\n",
            "Epoch [8/10], Step [75/101], Loss: 0.2792\n",
            "Epoch [8/10], Step [76/101], Loss: 0.0480\n",
            "Epoch [8/10], Step [77/101], Loss: 0.1536\n",
            "Epoch [8/10], Step [78/101], Loss: 0.0711\n",
            "Epoch [8/10], Step [79/101], Loss: 0.1292\n",
            "Epoch [8/10], Step [80/101], Loss: 0.0519\n",
            "Epoch [8/10], Step [81/101], Loss: 0.0577\n",
            "Epoch [8/10], Step [82/101], Loss: 0.2634\n",
            "Epoch [8/10], Step [83/101], Loss: 0.0869\n",
            "Epoch [8/10], Step [84/101], Loss: 0.1897\n",
            "Epoch [8/10], Step [85/101], Loss: 0.0736\n",
            "Epoch [8/10], Step [86/101], Loss: 0.1477\n",
            "Epoch [8/10], Step [87/101], Loss: 0.1003\n",
            "Epoch [8/10], Step [88/101], Loss: 0.2425\n",
            "Epoch [8/10], Step [89/101], Loss: 0.0736\n",
            "Epoch [8/10], Step [90/101], Loss: 0.1567\n",
            "Epoch [8/10], Step [91/101], Loss: 0.1884\n",
            "Epoch [8/10], Step [92/101], Loss: 0.1076\n",
            "Epoch [8/10], Step [93/101], Loss: 0.1780\n",
            "Epoch [8/10], Step [94/101], Loss: 0.2197\n",
            "Epoch [8/10], Step [95/101], Loss: 0.1015\n",
            "Epoch [8/10], Step [96/101], Loss: 0.1342\n",
            "Epoch [8/10], Step [97/101], Loss: 0.1567\n",
            "Epoch [8/10], Step [98/101], Loss: 0.1562\n",
            "Epoch [8/10], Step [99/101], Loss: 0.0630\n",
            "Epoch [8/10], Step [100/101], Loss: 0.0744\n",
            "Epoch [8/10], Step [101/101], Loss: 0.3128\n",
            "Epoch finished\n",
            "Epoch [9/10], Step [1/101], Loss: 0.0939\n",
            "Epoch [9/10], Step [2/101], Loss: 0.0864\n",
            "Epoch [9/10], Step [3/101], Loss: 0.0532\n",
            "Epoch [9/10], Step [4/101], Loss: 0.0949\n",
            "Epoch [9/10], Step [5/101], Loss: 0.2274\n",
            "Epoch [9/10], Step [6/101], Loss: 0.1001\n",
            "Epoch [9/10], Step [7/101], Loss: 0.0679\n",
            "Epoch [9/10], Step [8/101], Loss: 0.0538\n",
            "Epoch [9/10], Step [9/101], Loss: 0.1287\n",
            "Epoch [9/10], Step [10/101], Loss: 0.1553\n",
            "Epoch [9/10], Step [11/101], Loss: 0.0928\n",
            "Epoch [9/10], Step [12/101], Loss: 0.1188\n",
            "Epoch [9/10], Step [13/101], Loss: 0.0803\n",
            "Epoch [9/10], Step [14/101], Loss: 0.1479\n",
            "Epoch [9/10], Step [15/101], Loss: 0.1050\n",
            "Epoch [9/10], Step [16/101], Loss: 0.1186\n",
            "Epoch [9/10], Step [17/101], Loss: 0.0331\n",
            "Epoch [9/10], Step [18/101], Loss: 0.1143\n",
            "Epoch [9/10], Step [19/101], Loss: 0.0996\n",
            "Epoch [9/10], Step [20/101], Loss: 0.0371\n",
            "Epoch [9/10], Step [21/101], Loss: 0.2291\n",
            "Epoch [9/10], Step [22/101], Loss: 0.1677\n",
            "Epoch [9/10], Step [23/101], Loss: 0.1034\n",
            "Epoch [9/10], Step [24/101], Loss: 0.1720\n",
            "Epoch [9/10], Step [25/101], Loss: 0.1065\n",
            "Epoch [9/10], Step [26/101], Loss: 0.0729\n",
            "Epoch [9/10], Step [27/101], Loss: 0.1850\n",
            "Epoch [9/10], Step [28/101], Loss: 0.2244\n",
            "Epoch [9/10], Step [29/101], Loss: 0.2661\n",
            "Epoch [9/10], Step [30/101], Loss: 0.0538\n",
            "Epoch [9/10], Step [31/101], Loss: 0.2972\n",
            "Epoch [9/10], Step [32/101], Loss: 0.1305\n",
            "Epoch [9/10], Step [33/101], Loss: 0.1327\n",
            "Epoch [9/10], Step [34/101], Loss: 0.1458\n",
            "Epoch [9/10], Step [35/101], Loss: 0.1282\n",
            "Epoch [9/10], Step [36/101], Loss: 0.2074\n",
            "Epoch [9/10], Step [37/101], Loss: 0.0665\n",
            "Epoch [9/10], Step [38/101], Loss: 0.2719\n",
            "Epoch [9/10], Step [39/101], Loss: 0.0643\n",
            "Epoch [9/10], Step [40/101], Loss: 0.0370\n",
            "Epoch [9/10], Step [41/101], Loss: 0.2271\n",
            "Epoch [9/10], Step [42/101], Loss: 0.1850\n",
            "Epoch [9/10], Step [43/101], Loss: 0.1059\n",
            "Epoch [9/10], Step [44/101], Loss: 0.2043\n",
            "Epoch [9/10], Step [45/101], Loss: 0.0706\n",
            "Epoch [9/10], Step [46/101], Loss: 0.1101\n",
            "Epoch [9/10], Step [47/101], Loss: 0.1752\n",
            "Epoch [9/10], Step [48/101], Loss: 0.0993\n",
            "Epoch [9/10], Step [49/101], Loss: 0.1501\n",
            "Epoch [9/10], Step [50/101], Loss: 0.0915\n",
            "Epoch [9/10], Step [51/101], Loss: 0.0779\n",
            "Epoch [9/10], Step [52/101], Loss: 0.0699\n",
            "Epoch [9/10], Step [53/101], Loss: 0.0556\n",
            "Epoch [9/10], Step [54/101], Loss: 0.0322\n",
            "Epoch [9/10], Step [55/101], Loss: 0.0954\n",
            "Epoch [9/10], Step [56/101], Loss: 0.0836\n",
            "Epoch [9/10], Step [57/101], Loss: 0.1864\n",
            "Epoch [9/10], Step [58/101], Loss: 0.0596\n",
            "Epoch [9/10], Step [59/101], Loss: 0.0983\n",
            "Epoch [9/10], Step [60/101], Loss: 0.0878\n",
            "Epoch [9/10], Step [61/101], Loss: 0.1660\n",
            "Epoch [9/10], Step [62/101], Loss: 0.1406\n",
            "Epoch [9/10], Step [63/101], Loss: 0.0776\n",
            "Epoch [9/10], Step [64/101], Loss: 0.2047\n",
            "Epoch [9/10], Step [65/101], Loss: 0.0887\n",
            "Epoch [9/10], Step [66/101], Loss: 0.1649\n",
            "Epoch [9/10], Step [67/101], Loss: 0.2188\n",
            "Epoch [9/10], Step [68/101], Loss: 0.2049\n",
            "Epoch [9/10], Step [69/101], Loss: 0.1494\n",
            "Epoch [9/10], Step [70/101], Loss: 0.0874\n",
            "Epoch [9/10], Step [71/101], Loss: 0.0825\n",
            "Epoch [9/10], Step [72/101], Loss: 0.1217\n",
            "Epoch [9/10], Step [73/101], Loss: 0.0942\n",
            "Epoch [9/10], Step [74/101], Loss: 0.2418\n",
            "Epoch [9/10], Step [75/101], Loss: 0.1618\n",
            "Epoch [9/10], Step [76/101], Loss: 0.1051\n",
            "Epoch [9/10], Step [77/101], Loss: 0.0694\n",
            "Epoch [9/10], Step [78/101], Loss: 0.1314\n",
            "Epoch [9/10], Step [79/101], Loss: 0.0791\n",
            "Epoch [9/10], Step [80/101], Loss: 0.1205\n",
            "Epoch [9/10], Step [81/101], Loss: 0.0433\n",
            "Epoch [9/10], Step [82/101], Loss: 0.1098\n",
            "Epoch [9/10], Step [83/101], Loss: 0.0520\n",
            "Epoch [9/10], Step [84/101], Loss: 0.1542\n",
            "Epoch [9/10], Step [85/101], Loss: 0.0623\n",
            "Epoch [9/10], Step [86/101], Loss: 0.0802\n",
            "Epoch [9/10], Step [87/101], Loss: 0.0892\n",
            "Epoch [9/10], Step [88/101], Loss: 0.2262\n",
            "Epoch [9/10], Step [89/101], Loss: 0.1042\n",
            "Epoch [9/10], Step [90/101], Loss: 0.1056\n",
            "Epoch [9/10], Step [91/101], Loss: 0.0974\n",
            "Epoch [9/10], Step [92/101], Loss: 0.1720\n",
            "Epoch [9/10], Step [93/101], Loss: 0.1313\n",
            "Epoch [9/10], Step [94/101], Loss: 0.0522\n",
            "Epoch [9/10], Step [95/101], Loss: 0.1433\n",
            "Epoch [9/10], Step [96/101], Loss: 0.1230\n",
            "Epoch [9/10], Step [97/101], Loss: 0.0848\n",
            "Epoch [9/10], Step [98/101], Loss: 0.1112\n",
            "Epoch [9/10], Step [99/101], Loss: 0.0738\n",
            "Epoch [9/10], Step [100/101], Loss: 0.0930\n",
            "Epoch [9/10], Step [101/101], Loss: 0.1780\n",
            "Epoch finished\n",
            "Epoch [10/10], Step [1/101], Loss: 0.1163\n",
            "Epoch [10/10], Step [2/101], Loss: 0.0926\n",
            "Epoch [10/10], Step [3/101], Loss: 0.0352\n",
            "Epoch [10/10], Step [4/101], Loss: 0.0941\n",
            "Epoch [10/10], Step [5/101], Loss: 0.1931\n",
            "Epoch [10/10], Step [6/101], Loss: 0.0715\n",
            "Epoch [10/10], Step [7/101], Loss: 0.0667\n",
            "Epoch [10/10], Step [8/101], Loss: 0.1202\n",
            "Epoch [10/10], Step [9/101], Loss: 0.1534\n",
            "Epoch [10/10], Step [10/101], Loss: 0.0590\n",
            "Epoch [10/10], Step [11/101], Loss: 0.1018\n",
            "Epoch [10/10], Step [12/101], Loss: 0.2255\n",
            "Epoch [10/10], Step [13/101], Loss: 0.1651\n",
            "Epoch [10/10], Step [14/101], Loss: 0.2968\n",
            "Epoch [10/10], Step [15/101], Loss: 0.1147\n",
            "Epoch [10/10], Step [16/101], Loss: 0.1627\n",
            "Epoch [10/10], Step [17/101], Loss: 0.0836\n",
            "Epoch [10/10], Step [18/101], Loss: 0.1064\n",
            "Epoch [10/10], Step [19/101], Loss: 0.0468\n",
            "Epoch [10/10], Step [20/101], Loss: 0.1448\n",
            "Epoch [10/10], Step [21/101], Loss: 0.0981\n",
            "Epoch [10/10], Step [22/101], Loss: 0.0748\n",
            "Epoch [10/10], Step [23/101], Loss: 0.0925\n",
            "Epoch [10/10], Step [24/101], Loss: 0.0875\n",
            "Epoch [10/10], Step [25/101], Loss: 0.0903\n",
            "Epoch [10/10], Step [26/101], Loss: 0.1314\n",
            "Epoch [10/10], Step [27/101], Loss: 0.2075\n",
            "Epoch [10/10], Step [28/101], Loss: 0.1441\n",
            "Epoch [10/10], Step [29/101], Loss: 0.3248\n",
            "Epoch [10/10], Step [30/101], Loss: 0.0154\n",
            "Epoch [10/10], Step [31/101], Loss: 0.1368\n",
            "Epoch [10/10], Step [32/101], Loss: 0.0723\n",
            "Epoch [10/10], Step [33/101], Loss: 0.1648\n",
            "Epoch [10/10], Step [34/101], Loss: 0.1269\n",
            "Epoch [10/10], Step [35/101], Loss: 0.0778\n",
            "Epoch [10/10], Step [36/101], Loss: 0.0914\n",
            "Epoch [10/10], Step [37/101], Loss: 0.1437\n",
            "Epoch [10/10], Step [38/101], Loss: 0.1262\n",
            "Epoch [10/10], Step [39/101], Loss: 0.0746\n",
            "Epoch [10/10], Step [40/101], Loss: 0.0990\n",
            "Epoch [10/10], Step [41/101], Loss: 0.0544\n",
            "Epoch [10/10], Step [42/101], Loss: 0.1512\n",
            "Epoch [10/10], Step [43/101], Loss: 0.0699\n",
            "Epoch [10/10], Step [44/101], Loss: 0.0651\n",
            "Epoch [10/10], Step [45/101], Loss: 0.2684\n",
            "Epoch [10/10], Step [46/101], Loss: 0.1636\n",
            "Epoch [10/10], Step [47/101], Loss: 0.1001\n",
            "Epoch [10/10], Step [48/101], Loss: 0.0633\n",
            "Epoch [10/10], Step [49/101], Loss: 0.2771\n",
            "Epoch [10/10], Step [50/101], Loss: 0.2886\n",
            "Epoch [10/10], Step [51/101], Loss: 0.0860\n",
            "Epoch [10/10], Step [52/101], Loss: 0.1036\n",
            "Epoch [10/10], Step [53/101], Loss: 0.0701\n",
            "Epoch [10/10], Step [54/101], Loss: 0.0657\n",
            "Epoch [10/10], Step [55/101], Loss: 0.0795\n",
            "Epoch [10/10], Step [56/101], Loss: 0.0476\n",
            "Epoch [10/10], Step [57/101], Loss: 0.0593\n",
            "Epoch [10/10], Step [58/101], Loss: 0.2277\n",
            "Epoch [10/10], Step [59/101], Loss: 0.1298\n",
            "Epoch [10/10], Step [60/101], Loss: 0.0628\n",
            "Epoch [10/10], Step [61/101], Loss: 0.1315\n",
            "Epoch [10/10], Step [62/101], Loss: 0.2300\n",
            "Epoch [10/10], Step [63/101], Loss: 0.0751\n",
            "Epoch [10/10], Step [64/101], Loss: 0.1275\n",
            "Epoch [10/10], Step [65/101], Loss: 0.0670\n",
            "Epoch [10/10], Step [66/101], Loss: 0.1020\n",
            "Epoch [10/10], Step [67/101], Loss: 0.2415\n",
            "Epoch [10/10], Step [68/101], Loss: 0.1299\n",
            "Epoch [10/10], Step [69/101], Loss: 0.1768\n",
            "Epoch [10/10], Step [70/101], Loss: 0.2571\n",
            "Epoch [10/10], Step [71/101], Loss: 0.2703\n",
            "Epoch [10/10], Step [72/101], Loss: 0.0713\n",
            "Epoch [10/10], Step [73/101], Loss: 0.1613\n",
            "Epoch [10/10], Step [74/101], Loss: 0.1733\n",
            "Epoch [10/10], Step [75/101], Loss: 0.0643\n",
            "Epoch [10/10], Step [76/101], Loss: 0.1460\n",
            "Epoch [10/10], Step [77/101], Loss: 0.1384\n",
            "Epoch [10/10], Step [78/101], Loss: 0.1157\n",
            "Epoch [10/10], Step [79/101], Loss: 0.0671\n",
            "Epoch [10/10], Step [80/101], Loss: 0.1331\n",
            "Epoch [10/10], Step [81/101], Loss: 0.0552\n",
            "Epoch [10/10], Step [82/101], Loss: 0.2206\n",
            "Epoch [10/10], Step [83/101], Loss: 0.0714\n",
            "Epoch [10/10], Step [84/101], Loss: 0.0868\n",
            "Epoch [10/10], Step [85/101], Loss: 0.1076\n",
            "Epoch [10/10], Step [86/101], Loss: 0.2594\n",
            "Epoch [10/10], Step [87/101], Loss: 0.0768\n",
            "Epoch [10/10], Step [88/101], Loss: 0.0597\n",
            "Epoch [10/10], Step [89/101], Loss: 0.2506\n",
            "Epoch [10/10], Step [90/101], Loss: 0.0628\n",
            "Epoch [10/10], Step [91/101], Loss: 0.0693\n",
            "Epoch [10/10], Step [92/101], Loss: 0.0981\n",
            "Epoch [10/10], Step [93/101], Loss: 0.1849\n",
            "Epoch [10/10], Step [94/101], Loss: 0.1407\n",
            "Epoch [10/10], Step [95/101], Loss: 0.2924\n",
            "Epoch [10/10], Step [96/101], Loss: 0.2112\n",
            "Epoch [10/10], Step [97/101], Loss: 0.0233\n",
            "Epoch [10/10], Step [98/101], Loss: 0.1754\n",
            "Epoch [10/10], Step [99/101], Loss: 0.0801\n",
            "Epoch [10/10], Step [100/101], Loss: 0.0967\n",
            "Epoch [10/10], Step [101/101], Loss: 0.4296\n",
            "Epoch finished\n",
            "Training finished!\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test(model)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BuaqnBwwzf_u",
        "outputId": "d23100b5-3f22-466a-f03a-d41c6da2b0b9"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy of the network on the test images: 85 %\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from efficientnet_pytorch import EfficientNet\n",
        "model = EfficientNet.from_pretrained('efficientnet-b1', num_classes=20)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EIuhDbmOzhIE",
        "outputId": "e3ec6ecf-7d05-4b74-f8ef-d6e36018c5ab"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Downloading: \"https://github.com/lukemelas/EfficientNet-PyTorch/releases/download/1.0/efficientnet-b1-f1951068.pth\" to /root/.cache/torch/hub/checkpoints/efficientnet-b1-f1951068.pth\n",
            "100%|██████████| 30.1M/30.1M [00:00<00:00, 127MB/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Loaded pretrained weights for efficientnet-b1\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model = model.to(device)\n",
        "model.cuda()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wHC6e2WLzquN",
        "outputId": "ebdb498a-b634-494f-828d-44e15b1d58d0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "EfficientNet(\n",
              "  (_conv_stem): Conv2dStaticSamePadding(\n",
              "    3, 32, kernel_size=(3, 3), stride=(2, 2), bias=False\n",
              "    (static_padding): ZeroPad2d((0, 1, 0, 1))\n",
              "  )\n",
              "  (_bn0): BatchNorm2d(32, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "  (_blocks): ModuleList(\n",
              "    (0): MBConvBlock(\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        32, 32, kernel_size=(3, 3), stride=[1, 1], groups=32, bias=False\n",
              "        (static_padding): ZeroPad2d((1, 1, 1, 1))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(32, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        32, 8, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        8, 32, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        32, 16, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(16, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "    (1): MBConvBlock(\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        16, 16, kernel_size=(3, 3), stride=(1, 1), groups=16, bias=False\n",
              "        (static_padding): ZeroPad2d((1, 1, 1, 1))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(16, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        16, 4, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        4, 16, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        16, 16, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(16, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "    (2): MBConvBlock(\n",
              "      (_expand_conv): Conv2dStaticSamePadding(\n",
              "        16, 96, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn0): BatchNorm2d(96, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        96, 96, kernel_size=(3, 3), stride=[2, 2], groups=96, bias=False\n",
              "        (static_padding): ZeroPad2d((0, 1, 0, 1))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(96, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        96, 4, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        4, 96, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        96, 24, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(24, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "    (3-4): 2 x MBConvBlock(\n",
              "      (_expand_conv): Conv2dStaticSamePadding(\n",
              "        24, 144, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn0): BatchNorm2d(144, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        144, 144, kernel_size=(3, 3), stride=(1, 1), groups=144, bias=False\n",
              "        (static_padding): ZeroPad2d((1, 1, 1, 1))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(144, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        144, 6, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        6, 144, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        144, 24, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(24, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "    (5): MBConvBlock(\n",
              "      (_expand_conv): Conv2dStaticSamePadding(\n",
              "        24, 144, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn0): BatchNorm2d(144, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        144, 144, kernel_size=(5, 5), stride=[2, 2], groups=144, bias=False\n",
              "        (static_padding): ZeroPad2d((1, 2, 1, 2))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(144, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        144, 6, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        6, 144, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        144, 40, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(40, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "    (6-7): 2 x MBConvBlock(\n",
              "      (_expand_conv): Conv2dStaticSamePadding(\n",
              "        40, 240, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn0): BatchNorm2d(240, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        240, 240, kernel_size=(5, 5), stride=(1, 1), groups=240, bias=False\n",
              "        (static_padding): ZeroPad2d((2, 2, 2, 2))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(240, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        240, 10, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        10, 240, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        240, 40, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(40, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "    (8): MBConvBlock(\n",
              "      (_expand_conv): Conv2dStaticSamePadding(\n",
              "        40, 240, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn0): BatchNorm2d(240, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        240, 240, kernel_size=(3, 3), stride=[2, 2], groups=240, bias=False\n",
              "        (static_padding): ZeroPad2d((0, 1, 0, 1))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(240, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        240, 10, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        10, 240, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        240, 80, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(80, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "    (9-11): 3 x MBConvBlock(\n",
              "      (_expand_conv): Conv2dStaticSamePadding(\n",
              "        80, 480, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn0): BatchNorm2d(480, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        480, 480, kernel_size=(3, 3), stride=(1, 1), groups=480, bias=False\n",
              "        (static_padding): ZeroPad2d((1, 1, 1, 1))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(480, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        480, 20, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        20, 480, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        480, 80, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(80, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "    (12): MBConvBlock(\n",
              "      (_expand_conv): Conv2dStaticSamePadding(\n",
              "        80, 480, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn0): BatchNorm2d(480, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        480, 480, kernel_size=(5, 5), stride=[1, 1], groups=480, bias=False\n",
              "        (static_padding): ZeroPad2d((2, 2, 2, 2))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(480, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        480, 20, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        20, 480, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        480, 112, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(112, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "    (13-15): 3 x MBConvBlock(\n",
              "      (_expand_conv): Conv2dStaticSamePadding(\n",
              "        112, 672, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn0): BatchNorm2d(672, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        672, 672, kernel_size=(5, 5), stride=(1, 1), groups=672, bias=False\n",
              "        (static_padding): ZeroPad2d((2, 2, 2, 2))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(672, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        672, 28, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        28, 672, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        672, 112, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(112, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "    (16): MBConvBlock(\n",
              "      (_expand_conv): Conv2dStaticSamePadding(\n",
              "        112, 672, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn0): BatchNorm2d(672, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        672, 672, kernel_size=(5, 5), stride=[2, 2], groups=672, bias=False\n",
              "        (static_padding): ZeroPad2d((2, 2, 2, 2))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(672, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        672, 28, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        28, 672, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        672, 192, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(192, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "    (17-20): 4 x MBConvBlock(\n",
              "      (_expand_conv): Conv2dStaticSamePadding(\n",
              "        192, 1152, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn0): BatchNorm2d(1152, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        1152, 1152, kernel_size=(5, 5), stride=(1, 1), groups=1152, bias=False\n",
              "        (static_padding): ZeroPad2d((2, 2, 2, 2))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(1152, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        1152, 48, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        48, 1152, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        1152, 192, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(192, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "    (21): MBConvBlock(\n",
              "      (_expand_conv): Conv2dStaticSamePadding(\n",
              "        192, 1152, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn0): BatchNorm2d(1152, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        1152, 1152, kernel_size=(3, 3), stride=[1, 1], groups=1152, bias=False\n",
              "        (static_padding): ZeroPad2d((1, 1, 1, 1))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(1152, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        1152, 48, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        48, 1152, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        1152, 320, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(320, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "    (22): MBConvBlock(\n",
              "      (_expand_conv): Conv2dStaticSamePadding(\n",
              "        320, 1920, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn0): BatchNorm2d(1920, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_depthwise_conv): Conv2dStaticSamePadding(\n",
              "        1920, 1920, kernel_size=(3, 3), stride=(1, 1), groups=1920, bias=False\n",
              "        (static_padding): ZeroPad2d((1, 1, 1, 1))\n",
              "      )\n",
              "      (_bn1): BatchNorm2d(1920, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_se_reduce): Conv2dStaticSamePadding(\n",
              "        1920, 80, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_se_expand): Conv2dStaticSamePadding(\n",
              "        80, 1920, kernel_size=(1, 1), stride=(1, 1)\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_project_conv): Conv2dStaticSamePadding(\n",
              "        1920, 320, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "        (static_padding): Identity()\n",
              "      )\n",
              "      (_bn2): BatchNorm2d(320, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "      (_swish): MemoryEfficientSwish()\n",
              "    )\n",
              "  )\n",
              "  (_conv_head): Conv2dStaticSamePadding(\n",
              "    320, 1280, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
              "    (static_padding): Identity()\n",
              "  )\n",
              "  (_bn1): BatchNorm2d(1280, eps=0.001, momentum=0.010000000000000009, affine=True, track_running_stats=True)\n",
              "  (_avg_pooling): AdaptiveAvgPool2d(output_size=1)\n",
              "  (_dropout): Dropout(p=0.2, inplace=False)\n",
              "  (_fc): Linear(in_features=1280, out_features=20, bias=True)\n",
              "  (_swish): MemoryEfficientSwish()\n",
              ")"
            ]
          },
          "metadata": {},
          "execution_count": 52
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "criterion=nn.CrossEntropyLoss()\n",
        "optimizer=torch.optim.Adam(model.parameters(),lr=0.001)"
      ],
      "metadata": {
        "id": "zlDHw9ttzyGR"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train(model)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cADupHkUz5oX",
        "outputId": "6b7239bc-b387-4947-aced-adc12d292931"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch [1/10], Step [1/101], Loss: 3.0324\n",
            "Epoch [1/10], Step [2/101], Loss: 2.8444\n",
            "Epoch [1/10], Step [3/101], Loss: 2.6240\n",
            "Epoch [1/10], Step [4/101], Loss: 2.4498\n",
            "Epoch [1/10], Step [5/101], Loss: 2.3325\n",
            "Epoch [1/10], Step [6/101], Loss: 2.1264\n",
            "Epoch [1/10], Step [7/101], Loss: 1.9131\n",
            "Epoch [1/10], Step [8/101], Loss: 1.6169\n",
            "Epoch [1/10], Step [9/101], Loss: 1.7709\n",
            "Epoch [1/10], Step [10/101], Loss: 1.3859\n",
            "Epoch [1/10], Step [11/101], Loss: 1.3555\n",
            "Epoch [1/10], Step [12/101], Loss: 1.6052\n",
            "Epoch [1/10], Step [13/101], Loss: 1.3486\n",
            "Epoch [1/10], Step [14/101], Loss: 1.2763\n",
            "Epoch [1/10], Step [15/101], Loss: 1.0020\n",
            "Epoch [1/10], Step [16/101], Loss: 1.0399\n",
            "Epoch [1/10], Step [17/101], Loss: 0.9778\n",
            "Epoch [1/10], Step [18/101], Loss: 1.3411\n",
            "Epoch [1/10], Step [19/101], Loss: 0.7577\n",
            "Epoch [1/10], Step [20/101], Loss: 0.6756\n",
            "Epoch [1/10], Step [21/101], Loss: 0.6310\n",
            "Epoch [1/10], Step [22/101], Loss: 0.3101\n",
            "Epoch [1/10], Step [23/101], Loss: 0.4341\n",
            "Epoch [1/10], Step [24/101], Loss: 0.4667\n",
            "Epoch [1/10], Step [25/101], Loss: 0.5032\n",
            "Epoch [1/10], Step [26/101], Loss: 0.5641\n",
            "Epoch [1/10], Step [27/101], Loss: 0.3669\n",
            "Epoch [1/10], Step [28/101], Loss: 0.2253\n",
            "Epoch [1/10], Step [29/101], Loss: 0.5105\n",
            "Epoch [1/10], Step [30/101], Loss: 0.3841\n",
            "Epoch [1/10], Step [31/101], Loss: 0.3499\n",
            "Epoch [1/10], Step [32/101], Loss: 0.4184\n",
            "Epoch [1/10], Step [33/101], Loss: 0.3214\n",
            "Epoch [1/10], Step [34/101], Loss: 0.6052\n",
            "Epoch [1/10], Step [35/101], Loss: 0.2815\n",
            "Epoch [1/10], Step [36/101], Loss: 0.1823\n",
            "Epoch [1/10], Step [37/101], Loss: 0.3176\n",
            "Epoch [1/10], Step [38/101], Loss: 0.1503\n",
            "Epoch [1/10], Step [39/101], Loss: 0.2754\n",
            "Epoch [1/10], Step [40/101], Loss: 0.1119\n",
            "Epoch [1/10], Step [41/101], Loss: 0.2282\n",
            "Epoch [1/10], Step [42/101], Loss: 0.3356\n",
            "Epoch [1/10], Step [43/101], Loss: 0.1910\n",
            "Epoch [1/10], Step [44/101], Loss: 0.3368\n",
            "Epoch [1/10], Step [45/101], Loss: 0.3108\n",
            "Epoch [1/10], Step [46/101], Loss: 0.2033\n",
            "Epoch [1/10], Step [47/101], Loss: 0.0813\n",
            "Epoch [1/10], Step [48/101], Loss: 0.3518\n",
            "Epoch [1/10], Step [49/101], Loss: 0.3670\n",
            "Epoch [1/10], Step [50/101], Loss: 0.1900\n",
            "Epoch [1/10], Step [51/101], Loss: 0.2994\n",
            "Epoch [1/10], Step [52/101], Loss: 0.2407\n",
            "Epoch [1/10], Step [53/101], Loss: 0.3455\n",
            "Epoch [1/10], Step [54/101], Loss: 0.3363\n",
            "Epoch [1/10], Step [55/101], Loss: 0.3256\n",
            "Epoch [1/10], Step [56/101], Loss: 0.3523\n",
            "Epoch [1/10], Step [57/101], Loss: 0.0894\n",
            "Epoch [1/10], Step [58/101], Loss: 0.1906\n",
            "Epoch [1/10], Step [59/101], Loss: 0.2972\n",
            "Epoch [1/10], Step [60/101], Loss: 0.4127\n",
            "Epoch [1/10], Step [61/101], Loss: 0.1422\n",
            "Epoch [1/10], Step [62/101], Loss: 0.2039\n",
            "Epoch [1/10], Step [63/101], Loss: 0.4988\n",
            "Epoch [1/10], Step [64/101], Loss: 0.1645\n",
            "Epoch [1/10], Step [65/101], Loss: 0.1106\n",
            "Epoch [1/10], Step [66/101], Loss: 0.2813\n",
            "Epoch [1/10], Step [67/101], Loss: 0.3453\n",
            "Epoch [1/10], Step [68/101], Loss: 0.1171\n",
            "Epoch [1/10], Step [69/101], Loss: 0.5765\n",
            "Epoch [1/10], Step [70/101], Loss: 0.3729\n",
            "Epoch [1/10], Step [71/101], Loss: 0.1703\n",
            "Epoch [1/10], Step [72/101], Loss: 0.1022\n",
            "Epoch [1/10], Step [73/101], Loss: 0.3525\n",
            "Epoch [1/10], Step [74/101], Loss: 0.3705\n",
            "Epoch [1/10], Step [75/101], Loss: 0.3826\n",
            "Epoch [1/10], Step [76/101], Loss: 0.4371\n",
            "Epoch [1/10], Step [77/101], Loss: 0.5901\n",
            "Epoch [1/10], Step [78/101], Loss: 0.2694\n",
            "Epoch [1/10], Step [79/101], Loss: 0.1190\n",
            "Epoch [1/10], Step [80/101], Loss: 0.3689\n",
            "Epoch [1/10], Step [81/101], Loss: 0.1202\n",
            "Epoch [1/10], Step [82/101], Loss: 0.4559\n",
            "Epoch [1/10], Step [83/101], Loss: 0.2906\n",
            "Epoch [1/10], Step [84/101], Loss: 0.2514\n",
            "Epoch [1/10], Step [85/101], Loss: 0.2446\n",
            "Epoch [1/10], Step [86/101], Loss: 0.1820\n",
            "Epoch [1/10], Step [87/101], Loss: 0.4671\n",
            "Epoch [1/10], Step [88/101], Loss: 0.2192\n",
            "Epoch [1/10], Step [89/101], Loss: 0.3527\n",
            "Epoch [1/10], Step [90/101], Loss: 0.2950\n",
            "Epoch [1/10], Step [91/101], Loss: 0.1221\n",
            "Epoch [1/10], Step [92/101], Loss: 0.2578\n",
            "Epoch [1/10], Step [93/101], Loss: 0.4506\n",
            "Epoch [1/10], Step [94/101], Loss: 0.1543\n",
            "Epoch [1/10], Step [95/101], Loss: 0.4857\n",
            "Epoch [1/10], Step [96/101], Loss: 0.5101\n",
            "Epoch [1/10], Step [97/101], Loss: 0.2132\n",
            "Epoch [1/10], Step [98/101], Loss: 0.3779\n",
            "Epoch [1/10], Step [99/101], Loss: 0.1799\n",
            "Epoch [1/10], Step [100/101], Loss: 0.1348\n",
            "Epoch [1/10], Step [101/101], Loss: 1.2775\n",
            "Epoch finished\n",
            "Epoch [2/10], Step [1/101], Loss: 0.2175\n",
            "Epoch [2/10], Step [2/101], Loss: 0.1690\n",
            "Epoch [2/10], Step [3/101], Loss: 0.1310\n",
            "Epoch [2/10], Step [4/101], Loss: 0.1664\n",
            "Epoch [2/10], Step [5/101], Loss: 0.1164\n",
            "Epoch [2/10], Step [6/101], Loss: 0.3915\n",
            "Epoch [2/10], Step [7/101], Loss: 0.0983\n",
            "Epoch [2/10], Step [8/101], Loss: 0.1976\n",
            "Epoch [2/10], Step [9/101], Loss: 0.2293\n",
            "Epoch [2/10], Step [10/101], Loss: 0.1410\n",
            "Epoch [2/10], Step [11/101], Loss: 0.2608\n",
            "Epoch [2/10], Step [12/101], Loss: 0.3471\n",
            "Epoch [2/10], Step [13/101], Loss: 0.1833\n",
            "Epoch [2/10], Step [14/101], Loss: 0.2973\n",
            "Epoch [2/10], Step [15/101], Loss: 0.1375\n",
            "Epoch [2/10], Step [16/101], Loss: 0.1266\n",
            "Epoch [2/10], Step [17/101], Loss: 0.1230\n",
            "Epoch [2/10], Step [18/101], Loss: 0.0837\n",
            "Epoch [2/10], Step [19/101], Loss: 0.0897\n",
            "Epoch [2/10], Step [20/101], Loss: 0.0313\n",
            "Epoch [2/10], Step [21/101], Loss: 0.0428\n",
            "Epoch [2/10], Step [22/101], Loss: 0.2048\n",
            "Epoch [2/10], Step [23/101], Loss: 0.1340\n",
            "Epoch [2/10], Step [24/101], Loss: 0.1922\n",
            "Epoch [2/10], Step [25/101], Loss: 0.0844\n",
            "Epoch [2/10], Step [26/101], Loss: 0.1427\n",
            "Epoch [2/10], Step [27/101], Loss: 0.2950\n",
            "Epoch [2/10], Step [28/101], Loss: 0.1343\n",
            "Epoch [2/10], Step [29/101], Loss: 0.0634\n",
            "Epoch [2/10], Step [30/101], Loss: 0.0656\n",
            "Epoch [2/10], Step [31/101], Loss: 0.1306\n",
            "Epoch [2/10], Step [32/101], Loss: 0.1260\n",
            "Epoch [2/10], Step [33/101], Loss: 0.0630\n",
            "Epoch [2/10], Step [34/101], Loss: 0.1225\n",
            "Epoch [2/10], Step [35/101], Loss: 0.0773\n",
            "Epoch [2/10], Step [36/101], Loss: 0.2681\n",
            "Epoch [2/10], Step [37/101], Loss: 0.0357\n",
            "Epoch [2/10], Step [38/101], Loss: 0.1664\n",
            "Epoch [2/10], Step [39/101], Loss: 0.0661\n",
            "Epoch [2/10], Step [40/101], Loss: 0.0141\n",
            "Epoch [2/10], Step [41/101], Loss: 0.1348\n",
            "Epoch [2/10], Step [42/101], Loss: 0.0547\n",
            "Epoch [2/10], Step [43/101], Loss: 0.0538\n",
            "Epoch [2/10], Step [44/101], Loss: 0.0620\n",
            "Epoch [2/10], Step [45/101], Loss: 0.0284\n",
            "Epoch [2/10], Step [46/101], Loss: 0.0270\n",
            "Epoch [2/10], Step [47/101], Loss: 0.0471\n",
            "Epoch [2/10], Step [48/101], Loss: 0.2372\n",
            "Epoch [2/10], Step [49/101], Loss: 0.0294\n",
            "Epoch [2/10], Step [50/101], Loss: 0.2922\n",
            "Epoch [2/10], Step [51/101], Loss: 0.0959\n",
            "Epoch [2/10], Step [52/101], Loss: 0.1577\n",
            "Epoch [2/10], Step [53/101], Loss: 0.1962\n",
            "Epoch [2/10], Step [54/101], Loss: 0.1960\n",
            "Epoch [2/10], Step [55/101], Loss: 0.1119\n",
            "Epoch [2/10], Step [56/101], Loss: 0.0711\n",
            "Epoch [2/10], Step [57/101], Loss: 0.0477\n",
            "Epoch [2/10], Step [58/101], Loss: 0.1140\n",
            "Epoch [2/10], Step [59/101], Loss: 0.1372\n",
            "Epoch [2/10], Step [60/101], Loss: 0.1238\n",
            "Epoch [2/10], Step [61/101], Loss: 0.1522\n",
            "Epoch [2/10], Step [62/101], Loss: 0.3072\n",
            "Epoch [2/10], Step [63/101], Loss: 0.1016\n",
            "Epoch [2/10], Step [64/101], Loss: 0.1175\n",
            "Epoch [2/10], Step [65/101], Loss: 0.0626\n",
            "Epoch [2/10], Step [66/101], Loss: 0.1551\n",
            "Epoch [2/10], Step [67/101], Loss: 0.0979\n",
            "Epoch [2/10], Step [68/101], Loss: 0.2427\n",
            "Epoch [2/10], Step [69/101], Loss: 0.1990\n",
            "Epoch [2/10], Step [70/101], Loss: 0.1978\n",
            "Epoch [2/10], Step [71/101], Loss: 0.1599\n",
            "Epoch [2/10], Step [72/101], Loss: 0.0267\n",
            "Epoch [2/10], Step [73/101], Loss: 0.2273\n",
            "Epoch [2/10], Step [74/101], Loss: 0.4235\n",
            "Epoch [2/10], Step [75/101], Loss: 0.0727\n",
            "Epoch [2/10], Step [76/101], Loss: 0.0706\n",
            "Epoch [2/10], Step [77/101], Loss: 0.3161\n",
            "Epoch [2/10], Step [78/101], Loss: 0.3736\n",
            "Epoch [2/10], Step [79/101], Loss: 0.1343\n",
            "Epoch [2/10], Step [80/101], Loss: 0.0246\n",
            "Epoch [2/10], Step [81/101], Loss: 0.0539\n",
            "Epoch [2/10], Step [82/101], Loss: 0.1918\n",
            "Epoch [2/10], Step [83/101], Loss: 0.2597\n",
            "Epoch [2/10], Step [84/101], Loss: 0.0720\n",
            "Epoch [2/10], Step [85/101], Loss: 0.0884\n",
            "Epoch [2/10], Step [86/101], Loss: 0.2945\n",
            "Epoch [2/10], Step [87/101], Loss: 0.0885\n",
            "Epoch [2/10], Step [88/101], Loss: 0.0541\n",
            "Epoch [2/10], Step [89/101], Loss: 0.1466\n",
            "Epoch [2/10], Step [90/101], Loss: 0.1370\n",
            "Epoch [2/10], Step [91/101], Loss: 0.1769\n",
            "Epoch [2/10], Step [92/101], Loss: 0.0761\n",
            "Epoch [2/10], Step [93/101], Loss: 0.1477\n",
            "Epoch [2/10], Step [94/101], Loss: 0.1538\n",
            "Epoch [2/10], Step [95/101], Loss: 0.2364\n",
            "Epoch [2/10], Step [96/101], Loss: 0.1259\n",
            "Epoch [2/10], Step [97/101], Loss: 0.0538\n",
            "Epoch [2/10], Step [98/101], Loss: 0.1816\n",
            "Epoch [2/10], Step [99/101], Loss: 0.0216\n",
            "Epoch [2/10], Step [100/101], Loss: 0.1831\n",
            "Epoch [2/10], Step [101/101], Loss: 0.0133\n",
            "Epoch finished\n",
            "Epoch [3/10], Step [1/101], Loss: 0.0376\n",
            "Epoch [3/10], Step [2/101], Loss: 0.0176\n",
            "Epoch [3/10], Step [3/101], Loss: 0.0080\n",
            "Epoch [3/10], Step [4/101], Loss: 0.1270\n",
            "Epoch [3/10], Step [5/101], Loss: 0.1120\n",
            "Epoch [3/10], Step [6/101], Loss: 0.0623\n",
            "Epoch [3/10], Step [7/101], Loss: 0.0768\n",
            "Epoch [3/10], Step [8/101], Loss: 0.0612\n",
            "Epoch [3/10], Step [9/101], Loss: 0.0092\n",
            "Epoch [3/10], Step [10/101], Loss: 0.1536\n",
            "Epoch [3/10], Step [11/101], Loss: 0.0558\n",
            "Epoch [3/10], Step [12/101], Loss: 0.0858\n",
            "Epoch [3/10], Step [13/101], Loss: 0.4413\n",
            "Epoch [3/10], Step [14/101], Loss: 0.2373\n",
            "Epoch [3/10], Step [15/101], Loss: 0.1133\n",
            "Epoch [3/10], Step [16/101], Loss: 0.0373\n",
            "Epoch [3/10], Step [17/101], Loss: 0.0207\n",
            "Epoch [3/10], Step [18/101], Loss: 0.0896\n",
            "Epoch [3/10], Step [19/101], Loss: 0.1186\n",
            "Epoch [3/10], Step [20/101], Loss: 0.0126\n",
            "Epoch [3/10], Step [21/101], Loss: 0.0780\n",
            "Epoch [3/10], Step [22/101], Loss: 0.3098\n",
            "Epoch [3/10], Step [23/101], Loss: 0.0617\n",
            "Epoch [3/10], Step [24/101], Loss: 0.1503\n",
            "Epoch [3/10], Step [25/101], Loss: 0.0500\n",
            "Epoch [3/10], Step [26/101], Loss: 0.0751\n",
            "Epoch [3/10], Step [27/101], Loss: 0.0545\n",
            "Epoch [3/10], Step [28/101], Loss: 0.1057\n",
            "Epoch [3/10], Step [29/101], Loss: 0.0199\n",
            "Epoch [3/10], Step [30/101], Loss: 0.2737\n",
            "Epoch [3/10], Step [31/101], Loss: 0.0144\n",
            "Epoch [3/10], Step [32/101], Loss: 0.0882\n",
            "Epoch [3/10], Step [33/101], Loss: 0.1915\n",
            "Epoch [3/10], Step [34/101], Loss: 0.1425\n",
            "Epoch [3/10], Step [35/101], Loss: 0.0908\n",
            "Epoch [3/10], Step [36/101], Loss: 0.1568\n",
            "Epoch [3/10], Step [37/101], Loss: 0.1195\n",
            "Epoch [3/10], Step [38/101], Loss: 0.0654\n",
            "Epoch [3/10], Step [39/101], Loss: 0.2053\n",
            "Epoch [3/10], Step [40/101], Loss: 0.0327\n",
            "Epoch [3/10], Step [41/101], Loss: 0.2529\n",
            "Epoch [3/10], Step [42/101], Loss: 0.0207\n",
            "Epoch [3/10], Step [43/101], Loss: 0.0086\n",
            "Epoch [3/10], Step [44/101], Loss: 0.0736\n",
            "Epoch [3/10], Step [45/101], Loss: 0.0196\n",
            "Epoch [3/10], Step [46/101], Loss: 0.1768\n",
            "Epoch [3/10], Step [47/101], Loss: 0.1311\n",
            "Epoch [3/10], Step [48/101], Loss: 0.1267\n",
            "Epoch [3/10], Step [49/101], Loss: 0.0106\n",
            "Epoch [3/10], Step [50/101], Loss: 0.2405\n",
            "Epoch [3/10], Step [51/101], Loss: 0.2217\n",
            "Epoch [3/10], Step [52/101], Loss: 0.1294\n",
            "Epoch [3/10], Step [53/101], Loss: 0.0372\n",
            "Epoch [3/10], Step [54/101], Loss: 0.0470\n",
            "Epoch [3/10], Step [55/101], Loss: 0.0350\n",
            "Epoch [3/10], Step [56/101], Loss: 0.0574\n",
            "Epoch [3/10], Step [57/101], Loss: 0.0925\n",
            "Epoch [3/10], Step [58/101], Loss: 0.1416\n",
            "Epoch [3/10], Step [59/101], Loss: 0.0578\n",
            "Epoch [3/10], Step [60/101], Loss: 0.1924\n",
            "Epoch [3/10], Step [61/101], Loss: 0.2347\n",
            "Epoch [3/10], Step [62/101], Loss: 0.0388\n",
            "Epoch [3/10], Step [63/101], Loss: 0.1267\n",
            "Epoch [3/10], Step [64/101], Loss: 0.0831\n",
            "Epoch [3/10], Step [65/101], Loss: 0.1883\n",
            "Epoch [3/10], Step [66/101], Loss: 0.0594\n",
            "Epoch [3/10], Step [67/101], Loss: 0.0491\n",
            "Epoch [3/10], Step [68/101], Loss: 0.0854\n",
            "Epoch [3/10], Step [69/101], Loss: 0.0436\n",
            "Epoch [3/10], Step [70/101], Loss: 0.0788\n",
            "Epoch [3/10], Step [71/101], Loss: 0.0377\n",
            "Epoch [3/10], Step [72/101], Loss: 0.1133\n",
            "Epoch [3/10], Step [73/101], Loss: 0.0106\n",
            "Epoch [3/10], Step [74/101], Loss: 0.0177\n",
            "Epoch [3/10], Step [75/101], Loss: 0.0473\n",
            "Epoch [3/10], Step [76/101], Loss: 0.0812\n",
            "Epoch [3/10], Step [77/101], Loss: 0.1886\n",
            "Epoch [3/10], Step [78/101], Loss: 0.1092\n",
            "Epoch [3/10], Step [79/101], Loss: 0.0616\n",
            "Epoch [3/10], Step [80/101], Loss: 0.1450\n",
            "Epoch [3/10], Step [81/101], Loss: 0.2720\n",
            "Epoch [3/10], Step [82/101], Loss: 0.0147\n",
            "Epoch [3/10], Step [83/101], Loss: 0.0710\n",
            "Epoch [3/10], Step [84/101], Loss: 0.0358\n",
            "Epoch [3/10], Step [85/101], Loss: 0.0150\n",
            "Epoch [3/10], Step [86/101], Loss: 0.1853\n",
            "Epoch [3/10], Step [87/101], Loss: 0.0066\n",
            "Epoch [3/10], Step [88/101], Loss: 0.0697\n",
            "Epoch [3/10], Step [89/101], Loss: 0.1355\n",
            "Epoch [3/10], Step [90/101], Loss: 0.2557\n",
            "Epoch [3/10], Step [91/101], Loss: 0.0096\n",
            "Epoch [3/10], Step [92/101], Loss: 0.0405\n",
            "Epoch [3/10], Step [93/101], Loss: 0.2908\n",
            "Epoch [3/10], Step [94/101], Loss: 0.1532\n",
            "Epoch [3/10], Step [95/101], Loss: 0.1201\n",
            "Epoch [3/10], Step [96/101], Loss: 0.2135\n",
            "Epoch [3/10], Step [97/101], Loss: 0.0792\n",
            "Epoch [3/10], Step [98/101], Loss: 0.0396\n",
            "Epoch [3/10], Step [99/101], Loss: 0.0858\n",
            "Epoch [3/10], Step [100/101], Loss: 0.0781\n",
            "Epoch [3/10], Step [101/101], Loss: 0.0050\n",
            "Epoch finished\n",
            "Epoch [4/10], Step [1/101], Loss: 0.0558\n",
            "Epoch [4/10], Step [2/101], Loss: 0.1434\n",
            "Epoch [4/10], Step [3/101], Loss: 0.0202\n",
            "Epoch [4/10], Step [4/101], Loss: 0.0117\n",
            "Epoch [4/10], Step [5/101], Loss: 0.0128\n",
            "Epoch [4/10], Step [6/101], Loss: 0.1405\n",
            "Epoch [4/10], Step [7/101], Loss: 0.1479\n",
            "Epoch [4/10], Step [8/101], Loss: 0.0200\n",
            "Epoch [4/10], Step [9/101], Loss: 0.0561\n",
            "Epoch [4/10], Step [10/101], Loss: 0.1862\n",
            "Epoch [4/10], Step [11/101], Loss: 0.0208\n",
            "Epoch [4/10], Step [12/101], Loss: 0.1985\n",
            "Epoch [4/10], Step [13/101], Loss: 0.2117\n",
            "Epoch [4/10], Step [14/101], Loss: 0.2025\n",
            "Epoch [4/10], Step [15/101], Loss: 0.0581\n",
            "Epoch [4/10], Step [16/101], Loss: 0.1002\n",
            "Epoch [4/10], Step [17/101], Loss: 0.0081\n",
            "Epoch [4/10], Step [18/101], Loss: 0.0160\n",
            "Epoch [4/10], Step [19/101], Loss: 0.2486\n",
            "Epoch [4/10], Step [20/101], Loss: 0.1110\n",
            "Epoch [4/10], Step [21/101], Loss: 0.1359\n",
            "Epoch [4/10], Step [22/101], Loss: 0.1816\n",
            "Epoch [4/10], Step [23/101], Loss: 0.1022\n",
            "Epoch [4/10], Step [24/101], Loss: 0.0407\n",
            "Epoch [4/10], Step [25/101], Loss: 0.0480\n",
            "Epoch [4/10], Step [26/101], Loss: 0.0185\n",
            "Epoch [4/10], Step [27/101], Loss: 0.0149\n",
            "Epoch [4/10], Step [28/101], Loss: 0.0875\n",
            "Epoch [4/10], Step [29/101], Loss: 0.1778\n",
            "Epoch [4/10], Step [30/101], Loss: 0.0518\n",
            "Epoch [4/10], Step [31/101], Loss: 0.0253\n",
            "Epoch [4/10], Step [32/101], Loss: 0.0438\n",
            "Epoch [4/10], Step [33/101], Loss: 0.2108\n",
            "Epoch [4/10], Step [34/101], Loss: 0.0328\n",
            "Epoch [4/10], Step [35/101], Loss: 0.0981\n",
            "Epoch [4/10], Step [36/101], Loss: 0.0335\n",
            "Epoch [4/10], Step [37/101], Loss: 0.0287\n",
            "Epoch [4/10], Step [38/101], Loss: 0.0918\n",
            "Epoch [4/10], Step [39/101], Loss: 0.0928\n",
            "Epoch [4/10], Step [40/101], Loss: 0.0313\n",
            "Epoch [4/10], Step [41/101], Loss: 0.0995\n",
            "Epoch [4/10], Step [42/101], Loss: 0.1493\n",
            "Epoch [4/10], Step [43/101], Loss: 0.0532\n",
            "Epoch [4/10], Step [44/101], Loss: 0.0074\n",
            "Epoch [4/10], Step [45/101], Loss: 0.0349\n",
            "Epoch [4/10], Step [46/101], Loss: 0.0203\n",
            "Epoch [4/10], Step [47/101], Loss: 0.0781\n",
            "Epoch [4/10], Step [48/101], Loss: 0.0372\n",
            "Epoch [4/10], Step [49/101], Loss: 0.0761\n",
            "Epoch [4/10], Step [50/101], Loss: 0.0677\n",
            "Epoch [4/10], Step [51/101], Loss: 0.1332\n",
            "Epoch [4/10], Step [52/101], Loss: 0.0337\n",
            "Epoch [4/10], Step [53/101], Loss: 0.0743\n",
            "Epoch [4/10], Step [54/101], Loss: 0.0222\n",
            "Epoch [4/10], Step [55/101], Loss: 0.0675\n",
            "Epoch [4/10], Step [56/101], Loss: 0.0568\n",
            "Epoch [4/10], Step [57/101], Loss: 0.0367\n",
            "Epoch [4/10], Step [58/101], Loss: 0.0691\n",
            "Epoch [4/10], Step [59/101], Loss: 0.0203\n",
            "Epoch [4/10], Step [60/101], Loss: 0.0857\n",
            "Epoch [4/10], Step [61/101], Loss: 0.0452\n",
            "Epoch [4/10], Step [62/101], Loss: 0.0592\n",
            "Epoch [4/10], Step [63/101], Loss: 0.0798\n",
            "Epoch [4/10], Step [64/101], Loss: 0.0095\n",
            "Epoch [4/10], Step [65/101], Loss: 0.0946\n",
            "Epoch [4/10], Step [66/101], Loss: 0.0591\n",
            "Epoch [4/10], Step [67/101], Loss: 0.1385\n",
            "Epoch [4/10], Step [68/101], Loss: 0.0088\n",
            "Epoch [4/10], Step [69/101], Loss: 0.0050\n",
            "Epoch [4/10], Step [70/101], Loss: 0.1125\n",
            "Epoch [4/10], Step [71/101], Loss: 0.0286\n",
            "Epoch [4/10], Step [72/101], Loss: 0.0074\n",
            "Epoch [4/10], Step [73/101], Loss: 0.0570\n",
            "Epoch [4/10], Step [74/101], Loss: 0.0048\n",
            "Epoch [4/10], Step [75/101], Loss: 0.0777\n",
            "Epoch [4/10], Step [76/101], Loss: 0.1998\n",
            "Epoch [4/10], Step [77/101], Loss: 0.1132\n",
            "Epoch [4/10], Step [78/101], Loss: 0.0640\n",
            "Epoch [4/10], Step [79/101], Loss: 0.0837\n",
            "Epoch [4/10], Step [80/101], Loss: 0.3367\n",
            "Epoch [4/10], Step [81/101], Loss: 0.0244\n",
            "Epoch [4/10], Step [82/101], Loss: 0.1335\n",
            "Epoch [4/10], Step [83/101], Loss: 0.0079\n",
            "Epoch [4/10], Step [84/101], Loss: 0.0459\n",
            "Epoch [4/10], Step [85/101], Loss: 0.0585\n",
            "Epoch [4/10], Step [86/101], Loss: 0.0983\n",
            "Epoch [4/10], Step [87/101], Loss: 0.1355\n",
            "Epoch [4/10], Step [88/101], Loss: 0.1359\n",
            "Epoch [4/10], Step [89/101], Loss: 0.2333\n",
            "Epoch [4/10], Step [90/101], Loss: 0.0127\n",
            "Epoch [4/10], Step [91/101], Loss: 0.1814\n",
            "Epoch [4/10], Step [92/101], Loss: 0.0239\n",
            "Epoch [4/10], Step [93/101], Loss: 0.0723\n",
            "Epoch [4/10], Step [94/101], Loss: 0.0174\n",
            "Epoch [4/10], Step [95/101], Loss: 0.0771\n",
            "Epoch [4/10], Step [96/101], Loss: 0.1734\n",
            "Epoch [4/10], Step [97/101], Loss: 0.0093\n",
            "Epoch [4/10], Step [98/101], Loss: 0.0065\n",
            "Epoch [4/10], Step [99/101], Loss: 0.0788\n",
            "Epoch [4/10], Step [100/101], Loss: 0.2386\n",
            "Epoch [4/10], Step [101/101], Loss: 0.1992\n",
            "Epoch finished\n",
            "Epoch [5/10], Step [1/101], Loss: 0.0713\n",
            "Epoch [5/10], Step [2/101], Loss: 0.1872\n",
            "Epoch [5/10], Step [3/101], Loss: 0.0654\n",
            "Epoch [5/10], Step [4/101], Loss: 0.0588\n",
            "Epoch [5/10], Step [5/101], Loss: 0.2191\n",
            "Epoch [5/10], Step [6/101], Loss: 0.0203\n",
            "Epoch [5/10], Step [7/101], Loss: 0.1342\n",
            "Epoch [5/10], Step [8/101], Loss: 0.0429\n",
            "Epoch [5/10], Step [9/101], Loss: 0.0184\n",
            "Epoch [5/10], Step [10/101], Loss: 0.0138\n",
            "Epoch [5/10], Step [11/101], Loss: 0.2105\n",
            "Epoch [5/10], Step [12/101], Loss: 0.0480\n",
            "Epoch [5/10], Step [13/101], Loss: 0.0170\n",
            "Epoch [5/10], Step [14/101], Loss: 0.0053\n",
            "Epoch [5/10], Step [15/101], Loss: 0.2859\n",
            "Epoch [5/10], Step [16/101], Loss: 0.1699\n",
            "Epoch [5/10], Step [17/101], Loss: 0.0733\n",
            "Epoch [5/10], Step [18/101], Loss: 0.3453\n",
            "Epoch [5/10], Step [19/101], Loss: 0.0617\n",
            "Epoch [5/10], Step [20/101], Loss: 0.0907\n",
            "Epoch [5/10], Step [21/101], Loss: 0.9012\n",
            "Epoch [5/10], Step [22/101], Loss: 0.0561\n",
            "Epoch [5/10], Step [23/101], Loss: 0.0888\n",
            "Epoch [5/10], Step [24/101], Loss: 0.2513\n",
            "Epoch [5/10], Step [25/101], Loss: 0.0192\n",
            "Epoch [5/10], Step [26/101], Loss: 0.0225\n",
            "Epoch [5/10], Step [27/101], Loss: 0.0093\n",
            "Epoch [5/10], Step [28/101], Loss: 0.0368\n",
            "Epoch [5/10], Step [29/101], Loss: 0.3336\n",
            "Epoch [5/10], Step [30/101], Loss: 0.1213\n",
            "Epoch [5/10], Step [31/101], Loss: 0.2570\n",
            "Epoch [5/10], Step [32/101], Loss: 0.2100\n",
            "Epoch [5/10], Step [33/101], Loss: 0.0468\n",
            "Epoch [5/10], Step [34/101], Loss: 0.0088\n",
            "Epoch [5/10], Step [35/101], Loss: 0.0323\n",
            "Epoch [5/10], Step [36/101], Loss: 0.0463\n",
            "Epoch [5/10], Step [37/101], Loss: 0.0974\n",
            "Epoch [5/10], Step [38/101], Loss: 0.0714\n",
            "Epoch [5/10], Step [39/101], Loss: 0.1936\n",
            "Epoch [5/10], Step [40/101], Loss: 0.0867\n",
            "Epoch [5/10], Step [41/101], Loss: 0.0121\n",
            "Epoch [5/10], Step [42/101], Loss: 0.0670\n",
            "Epoch [5/10], Step [43/101], Loss: 0.0960\n",
            "Epoch [5/10], Step [44/101], Loss: 0.0540\n",
            "Epoch [5/10], Step [45/101], Loss: 0.1293\n",
            "Epoch [5/10], Step [46/101], Loss: 0.0738\n",
            "Epoch [5/10], Step [47/101], Loss: 0.1858\n",
            "Epoch [5/10], Step [48/101], Loss: 0.0824\n",
            "Epoch [5/10], Step [49/101], Loss: 0.0468\n",
            "Epoch [5/10], Step [50/101], Loss: 0.1735\n",
            "Epoch [5/10], Step [51/101], Loss: 0.0602\n",
            "Epoch [5/10], Step [52/101], Loss: 0.0504\n",
            "Epoch [5/10], Step [53/101], Loss: 0.1464\n",
            "Epoch [5/10], Step [54/101], Loss: 0.2279\n",
            "Epoch [5/10], Step [55/101], Loss: 0.2345\n",
            "Epoch [5/10], Step [56/101], Loss: 0.0797\n",
            "Epoch [5/10], Step [57/101], Loss: 0.0347\n",
            "Epoch [5/10], Step [58/101], Loss: 0.0820\n",
            "Epoch [5/10], Step [59/101], Loss: 0.0158\n",
            "Epoch [5/10], Step [60/101], Loss: 0.0128\n",
            "Epoch [5/10], Step [61/101], Loss: 0.0661\n",
            "Epoch [5/10], Step [62/101], Loss: 0.2237\n",
            "Epoch [5/10], Step [63/101], Loss: 0.1332\n",
            "Epoch [5/10], Step [64/101], Loss: 0.0061\n",
            "Epoch [5/10], Step [65/101], Loss: 0.1776\n",
            "Epoch [5/10], Step [66/101], Loss: 0.0268\n",
            "Epoch [5/10], Step [67/101], Loss: 0.0542\n",
            "Epoch [5/10], Step [68/101], Loss: 0.0765\n",
            "Epoch [5/10], Step [69/101], Loss: 0.0210\n",
            "Epoch [5/10], Step [70/101], Loss: 0.1355\n",
            "Epoch [5/10], Step [71/101], Loss: 0.0140\n",
            "Epoch [5/10], Step [72/101], Loss: 0.2511\n",
            "Epoch [5/10], Step [73/101], Loss: 0.0472\n",
            "Epoch [5/10], Step [74/101], Loss: 0.0140\n",
            "Epoch [5/10], Step [75/101], Loss: 0.0538\n",
            "Epoch [5/10], Step [76/101], Loss: 0.3841\n",
            "Epoch [5/10], Step [77/101], Loss: 0.0185\n",
            "Epoch [5/10], Step [78/101], Loss: 0.0227\n",
            "Epoch [5/10], Step [79/101], Loss: 0.0742\n",
            "Epoch [5/10], Step [80/101], Loss: 0.0256\n",
            "Epoch [5/10], Step [81/101], Loss: 0.2353\n",
            "Epoch [5/10], Step [82/101], Loss: 0.1700\n",
            "Epoch [5/10], Step [83/101], Loss: 0.1097\n",
            "Epoch [5/10], Step [84/101], Loss: 0.0327\n",
            "Epoch [5/10], Step [85/101], Loss: 0.0066\n",
            "Epoch [5/10], Step [86/101], Loss: 0.0685\n",
            "Epoch [5/10], Step [87/101], Loss: 0.2512\n",
            "Epoch [5/10], Step [88/101], Loss: 0.0238\n",
            "Epoch [5/10], Step [89/101], Loss: 0.0401\n",
            "Epoch [5/10], Step [90/101], Loss: 0.0684\n",
            "Epoch [5/10], Step [91/101], Loss: 0.0420\n",
            "Epoch [5/10], Step [92/101], Loss: 0.2109\n",
            "Epoch [5/10], Step [93/101], Loss: 0.0532\n",
            "Epoch [5/10], Step [94/101], Loss: 0.1052\n",
            "Epoch [5/10], Step [95/101], Loss: 0.1000\n",
            "Epoch [5/10], Step [96/101], Loss: 0.2613\n",
            "Epoch [5/10], Step [97/101], Loss: 0.0383\n",
            "Epoch [5/10], Step [98/101], Loss: 0.0064\n",
            "Epoch [5/10], Step [99/101], Loss: 0.0791\n",
            "Epoch [5/10], Step [100/101], Loss: 0.0042\n",
            "Epoch [5/10], Step [101/101], Loss: 0.0199\n",
            "Epoch finished\n",
            "Epoch [6/10], Step [1/101], Loss: 0.0494\n",
            "Epoch [6/10], Step [2/101], Loss: 0.0279\n",
            "Epoch [6/10], Step [3/101], Loss: 0.0244\n",
            "Epoch [6/10], Step [4/101], Loss: 0.0338\n",
            "Epoch [6/10], Step [5/101], Loss: 0.1349\n",
            "Epoch [6/10], Step [6/101], Loss: 0.1137\n",
            "Epoch [6/10], Step [7/101], Loss: 0.0584\n",
            "Epoch [6/10], Step [8/101], Loss: 0.1033\n",
            "Epoch [6/10], Step [9/101], Loss: 0.0963\n",
            "Epoch [6/10], Step [10/101], Loss: 0.0086\n",
            "Epoch [6/10], Step [11/101], Loss: 0.0931\n",
            "Epoch [6/10], Step [12/101], Loss: 0.0494\n",
            "Epoch [6/10], Step [13/101], Loss: 0.1237\n",
            "Epoch [6/10], Step [14/101], Loss: 0.0653\n",
            "Epoch [6/10], Step [15/101], Loss: 0.0614\n",
            "Epoch [6/10], Step [16/101], Loss: 0.0126\n",
            "Epoch [6/10], Step [17/101], Loss: 0.0026\n",
            "Epoch [6/10], Step [18/101], Loss: 0.0366\n",
            "Epoch [6/10], Step [19/101], Loss: 0.0520\n",
            "Epoch [6/10], Step [20/101], Loss: 0.1032\n",
            "Epoch [6/10], Step [21/101], Loss: 0.0634\n",
            "Epoch [6/10], Step [22/101], Loss: 0.0114\n",
            "Epoch [6/10], Step [23/101], Loss: 0.3623\n",
            "Epoch [6/10], Step [24/101], Loss: 0.1339\n",
            "Epoch [6/10], Step [25/101], Loss: 0.0108\n",
            "Epoch [6/10], Step [26/101], Loss: 0.0101\n",
            "Epoch [6/10], Step [27/101], Loss: 0.0173\n",
            "Epoch [6/10], Step [28/101], Loss: 0.0040\n",
            "Epoch [6/10], Step [29/101], Loss: 0.0586\n",
            "Epoch [6/10], Step [30/101], Loss: 0.0222\n",
            "Epoch [6/10], Step [31/101], Loss: 0.0176\n",
            "Epoch [6/10], Step [32/101], Loss: 0.0186\n",
            "Epoch [6/10], Step [33/101], Loss: 0.2389\n",
            "Epoch [6/10], Step [34/101], Loss: 0.2677\n",
            "Epoch [6/10], Step [35/101], Loss: 0.1035\n",
            "Epoch [6/10], Step [36/101], Loss: 0.2559\n",
            "Epoch [6/10], Step [37/101], Loss: 0.1327\n",
            "Epoch [6/10], Step [38/101], Loss: 0.0348\n",
            "Epoch [6/10], Step [39/101], Loss: 0.0540\n",
            "Epoch [6/10], Step [40/101], Loss: 0.0210\n",
            "Epoch [6/10], Step [41/101], Loss: 0.2788\n",
            "Epoch [6/10], Step [42/101], Loss: 0.0057\n",
            "Epoch [6/10], Step [43/101], Loss: 0.0570\n",
            "Epoch [6/10], Step [44/101], Loss: 0.0128\n",
            "Epoch [6/10], Step [45/101], Loss: 0.0570\n",
            "Epoch [6/10], Step [46/101], Loss: 0.0855\n",
            "Epoch [6/10], Step [47/101], Loss: 0.0643\n",
            "Epoch [6/10], Step [48/101], Loss: 0.0194\n",
            "Epoch [6/10], Step [49/101], Loss: 0.0434\n",
            "Epoch [6/10], Step [50/101], Loss: 0.0445\n",
            "Epoch [6/10], Step [51/101], Loss: 0.0755\n",
            "Epoch [6/10], Step [52/101], Loss: 0.0153\n",
            "Epoch [6/10], Step [53/101], Loss: 0.0618\n",
            "Epoch [6/10], Step [54/101], Loss: 0.1266\n",
            "Epoch [6/10], Step [55/101], Loss: 0.1119\n",
            "Epoch [6/10], Step [56/101], Loss: 0.1115\n",
            "Epoch [6/10], Step [57/101], Loss: 0.0560\n",
            "Epoch [6/10], Step [58/101], Loss: 0.1384\n",
            "Epoch [6/10], Step [59/101], Loss: 0.0789\n",
            "Epoch [6/10], Step [60/101], Loss: 0.0837\n",
            "Epoch [6/10], Step [61/101], Loss: 0.0983\n",
            "Epoch [6/10], Step [62/101], Loss: 0.1860\n",
            "Epoch [6/10], Step [63/101], Loss: 0.0944\n",
            "Epoch [6/10], Step [64/101], Loss: 0.0343\n",
            "Epoch [6/10], Step [65/101], Loss: 0.1149\n",
            "Epoch [6/10], Step [66/101], Loss: 0.0123\n",
            "Epoch [6/10], Step [67/101], Loss: 0.0273\n",
            "Epoch [6/10], Step [68/101], Loss: 0.0600\n",
            "Epoch [6/10], Step [69/101], Loss: 0.0716\n",
            "Epoch [6/10], Step [70/101], Loss: 0.0379\n",
            "Epoch [6/10], Step [71/101], Loss: 0.0341\n",
            "Epoch [6/10], Step [72/101], Loss: 0.1341\n",
            "Epoch [6/10], Step [73/101], Loss: 0.0364\n",
            "Epoch [6/10], Step [74/101], Loss: 0.2448\n",
            "Epoch [6/10], Step [75/101], Loss: 0.0141\n",
            "Epoch [6/10], Step [76/101], Loss: 0.0369\n",
            "Epoch [6/10], Step [77/101], Loss: 0.0356\n",
            "Epoch [6/10], Step [78/101], Loss: 0.0449\n",
            "Epoch [6/10], Step [79/101], Loss: 0.0632\n",
            "Epoch [6/10], Step [80/101], Loss: 0.0588\n",
            "Epoch [6/10], Step [81/101], Loss: 0.0104\n",
            "Epoch [6/10], Step [82/101], Loss: 0.0250\n",
            "Epoch [6/10], Step [83/101], Loss: 0.0107\n",
            "Epoch [6/10], Step [84/101], Loss: 0.0883\n",
            "Epoch [6/10], Step [85/101], Loss: 0.0015\n",
            "Epoch [6/10], Step [86/101], Loss: 0.0293\n",
            "Epoch [6/10], Step [87/101], Loss: 0.0377\n",
            "Epoch [6/10], Step [88/101], Loss: 0.0525\n",
            "Epoch [6/10], Step [89/101], Loss: 0.0053\n",
            "Epoch [6/10], Step [90/101], Loss: 0.0802\n",
            "Epoch [6/10], Step [91/101], Loss: 0.0643\n",
            "Epoch [6/10], Step [92/101], Loss: 0.0292\n",
            "Epoch [6/10], Step [93/101], Loss: 0.1181\n",
            "Epoch [6/10], Step [94/101], Loss: 0.0188\n",
            "Epoch [6/10], Step [95/101], Loss: 0.0685\n",
            "Epoch [6/10], Step [96/101], Loss: 0.0365\n",
            "Epoch [6/10], Step [97/101], Loss: 0.1986\n",
            "Epoch [6/10], Step [98/101], Loss: 0.0373\n",
            "Epoch [6/10], Step [99/101], Loss: 0.0625\n",
            "Epoch [6/10], Step [100/101], Loss: 0.0895\n",
            "Epoch [6/10], Step [101/101], Loss: 0.2319\n",
            "Epoch finished\n",
            "Epoch [7/10], Step [1/101], Loss: 0.3366\n",
            "Epoch [7/10], Step [2/101], Loss: 0.0207\n",
            "Epoch [7/10], Step [3/101], Loss: 0.0430\n",
            "Epoch [7/10], Step [4/101], Loss: 0.0021\n",
            "Epoch [7/10], Step [5/101], Loss: 0.0418\n",
            "Epoch [7/10], Step [6/101], Loss: 0.0075\n",
            "Epoch [7/10], Step [7/101], Loss: 0.0197\n",
            "Epoch [7/10], Step [8/101], Loss: 0.0586\n",
            "Epoch [7/10], Step [9/101], Loss: 0.0603\n",
            "Epoch [7/10], Step [10/101], Loss: 0.0180\n",
            "Epoch [7/10], Step [11/101], Loss: 0.0159\n",
            "Epoch [7/10], Step [12/101], Loss: 0.0187\n",
            "Epoch [7/10], Step [13/101], Loss: 0.0832\n",
            "Epoch [7/10], Step [14/101], Loss: 0.2201\n",
            "Epoch [7/10], Step [15/101], Loss: 0.0204\n",
            "Epoch [7/10], Step [16/101], Loss: 0.0186\n",
            "Epoch [7/10], Step [17/101], Loss: 0.2674\n",
            "Epoch [7/10], Step [18/101], Loss: 0.0084\n",
            "Epoch [7/10], Step [19/101], Loss: 0.0032\n",
            "Epoch [7/10], Step [20/101], Loss: 0.0223\n",
            "Epoch [7/10], Step [21/101], Loss: 0.0025\n",
            "Epoch [7/10], Step [22/101], Loss: 0.0615\n",
            "Epoch [7/10], Step [23/101], Loss: 0.1072\n",
            "Epoch [7/10], Step [24/101], Loss: 0.0417\n",
            "Epoch [7/10], Step [25/101], Loss: 0.0222\n",
            "Epoch [7/10], Step [26/101], Loss: 0.1371\n",
            "Epoch [7/10], Step [27/101], Loss: 0.1020\n",
            "Epoch [7/10], Step [28/101], Loss: 0.0651\n",
            "Epoch [7/10], Step [29/101], Loss: 0.0086\n",
            "Epoch [7/10], Step [30/101], Loss: 0.0280\n",
            "Epoch [7/10], Step [31/101], Loss: 0.0027\n",
            "Epoch [7/10], Step [32/101], Loss: 0.0289\n",
            "Epoch [7/10], Step [33/101], Loss: 0.0167\n",
            "Epoch [7/10], Step [34/101], Loss: 0.0225\n",
            "Epoch [7/10], Step [35/101], Loss: 0.0306\n",
            "Epoch [7/10], Step [36/101], Loss: 0.2516\n",
            "Epoch [7/10], Step [37/101], Loss: 0.1003\n",
            "Epoch [7/10], Step [38/101], Loss: 0.1721\n",
            "Epoch [7/10], Step [39/101], Loss: 0.0069\n",
            "Epoch [7/10], Step [40/101], Loss: 0.0242\n",
            "Epoch [7/10], Step [41/101], Loss: 0.0074\n",
            "Epoch [7/10], Step [42/101], Loss: 0.0042\n",
            "Epoch [7/10], Step [43/101], Loss: 0.0300\n",
            "Epoch [7/10], Step [44/101], Loss: 0.0353\n",
            "Epoch [7/10], Step [45/101], Loss: 0.2379\n",
            "Epoch [7/10], Step [46/101], Loss: 0.0302\n",
            "Epoch [7/10], Step [47/101], Loss: 0.0836\n",
            "Epoch [7/10], Step [48/101], Loss: 0.0192\n",
            "Epoch [7/10], Step [49/101], Loss: 0.0510\n",
            "Epoch [7/10], Step [50/101], Loss: 0.0244\n",
            "Epoch [7/10], Step [51/101], Loss: 0.0763\n",
            "Epoch [7/10], Step [52/101], Loss: 0.0354\n",
            "Epoch [7/10], Step [53/101], Loss: 0.2129\n",
            "Epoch [7/10], Step [54/101], Loss: 0.0076\n",
            "Epoch [7/10], Step [55/101], Loss: 0.0760\n",
            "Epoch [7/10], Step [56/101], Loss: 0.0235\n",
            "Epoch [7/10], Step [57/101], Loss: 0.0760\n",
            "Epoch [7/10], Step [58/101], Loss: 0.3393\n",
            "Epoch [7/10], Step [59/101], Loss: 0.0022\n",
            "Epoch [7/10], Step [60/101], Loss: 0.0153\n",
            "Epoch [7/10], Step [61/101], Loss: 0.2659\n",
            "Epoch [7/10], Step [62/101], Loss: 0.0074\n",
            "Epoch [7/10], Step [63/101], Loss: 0.0317\n",
            "Epoch [7/10], Step [64/101], Loss: 0.0647\n",
            "Epoch [7/10], Step [65/101], Loss: 0.0696\n",
            "Epoch [7/10], Step [66/101], Loss: 0.0970\n",
            "Epoch [7/10], Step [67/101], Loss: 0.0652\n",
            "Epoch [7/10], Step [68/101], Loss: 0.0787\n",
            "Epoch [7/10], Step [69/101], Loss: 0.0176\n",
            "Epoch [7/10], Step [70/101], Loss: 0.1227\n",
            "Epoch [7/10], Step [71/101], Loss: 0.0096\n",
            "Epoch [7/10], Step [72/101], Loss: 0.0041\n",
            "Epoch [7/10], Step [73/101], Loss: 0.0269\n",
            "Epoch [7/10], Step [74/101], Loss: 0.0228\n",
            "Epoch [7/10], Step [75/101], Loss: 0.0493\n",
            "Epoch [7/10], Step [76/101], Loss: 0.3057\n",
            "Epoch [7/10], Step [77/101], Loss: 0.0153\n",
            "Epoch [7/10], Step [78/101], Loss: 0.1264\n",
            "Epoch [7/10], Step [79/101], Loss: 0.0095\n",
            "Epoch [7/10], Step [80/101], Loss: 0.0876\n",
            "Epoch [7/10], Step [81/101], Loss: 0.0064\n",
            "Epoch [7/10], Step [82/101], Loss: 0.0156\n",
            "Epoch [7/10], Step [83/101], Loss: 0.0155\n",
            "Epoch [7/10], Step [84/101], Loss: 0.0800\n",
            "Epoch [7/10], Step [85/101], Loss: 0.0060\n",
            "Epoch [7/10], Step [86/101], Loss: 0.0159\n",
            "Epoch [7/10], Step [87/101], Loss: 0.0320\n",
            "Epoch [7/10], Step [88/101], Loss: 0.0654\n",
            "Epoch [7/10], Step [89/101], Loss: 0.0093\n",
            "Epoch [7/10], Step [90/101], Loss: 0.0051\n",
            "Epoch [7/10], Step [91/101], Loss: 0.0870\n",
            "Epoch [7/10], Step [92/101], Loss: 0.1506\n",
            "Epoch [7/10], Step [93/101], Loss: 0.0279\n",
            "Epoch [7/10], Step [94/101], Loss: 0.0191\n",
            "Epoch [7/10], Step [95/101], Loss: 0.0427\n",
            "Epoch [7/10], Step [96/101], Loss: 0.0972\n",
            "Epoch [7/10], Step [97/101], Loss: 0.0152\n",
            "Epoch [7/10], Step [98/101], Loss: 0.0078\n",
            "Epoch [7/10], Step [99/101], Loss: 0.0486\n",
            "Epoch [7/10], Step [100/101], Loss: 0.0764\n",
            "Epoch [7/10], Step [101/101], Loss: 0.0137\n",
            "Epoch finished\n",
            "Epoch [8/10], Step [1/101], Loss: 0.1185\n",
            "Epoch [8/10], Step [2/101], Loss: 0.0296\n",
            "Epoch [8/10], Step [3/101], Loss: 0.0343\n",
            "Epoch [8/10], Step [4/101], Loss: 0.0271\n",
            "Epoch [8/10], Step [5/101], Loss: 0.1049\n",
            "Epoch [8/10], Step [6/101], Loss: 0.0082\n",
            "Epoch [8/10], Step [7/101], Loss: 0.0108\n",
            "Epoch [8/10], Step [8/101], Loss: 0.0194\n",
            "Epoch [8/10], Step [9/101], Loss: 0.0129\n",
            "Epoch [8/10], Step [10/101], Loss: 0.0319\n",
            "Epoch [8/10], Step [11/101], Loss: 0.0791\n",
            "Epoch [8/10], Step [12/101], Loss: 0.0472\n",
            "Epoch [8/10], Step [13/101], Loss: 0.0151\n",
            "Epoch [8/10], Step [14/101], Loss: 0.0735\n",
            "Epoch [8/10], Step [15/101], Loss: 0.0667\n",
            "Epoch [8/10], Step [16/101], Loss: 0.1091\n",
            "Epoch [8/10], Step [17/101], Loss: 0.0295\n",
            "Epoch [8/10], Step [18/101], Loss: 0.0167\n",
            "Epoch [8/10], Step [19/101], Loss: 0.0027\n",
            "Epoch [8/10], Step [20/101], Loss: 0.1003\n",
            "Epoch [8/10], Step [21/101], Loss: 0.0041\n",
            "Epoch [8/10], Step [22/101], Loss: 0.0500\n",
            "Epoch [8/10], Step [23/101], Loss: 0.0026\n",
            "Epoch [8/10], Step [24/101], Loss: 0.0200\n",
            "Epoch [8/10], Step [25/101], Loss: 0.0031\n",
            "Epoch [8/10], Step [26/101], Loss: 0.0174\n",
            "Epoch [8/10], Step [27/101], Loss: 0.0060\n",
            "Epoch [8/10], Step [28/101], Loss: 0.0393\n",
            "Epoch [8/10], Step [29/101], Loss: 0.0430\n",
            "Epoch [8/10], Step [30/101], Loss: 0.0642\n",
            "Epoch [8/10], Step [31/101], Loss: 0.0402\n",
            "Epoch [8/10], Step [32/101], Loss: 0.0794\n",
            "Epoch [8/10], Step [33/101], Loss: 0.0317\n",
            "Epoch [8/10], Step [34/101], Loss: 0.0179\n",
            "Epoch [8/10], Step [35/101], Loss: 0.0066\n",
            "Epoch [8/10], Step [36/101], Loss: 0.0237\n",
            "Epoch [8/10], Step [37/101], Loss: 0.0047\n",
            "Epoch [8/10], Step [38/101], Loss: 0.0355\n",
            "Epoch [8/10], Step [39/101], Loss: 0.0006\n",
            "Epoch [8/10], Step [40/101], Loss: 0.0365\n",
            "Epoch [8/10], Step [41/101], Loss: 0.0176\n",
            "Epoch [8/10], Step [42/101], Loss: 0.0015\n",
            "Epoch [8/10], Step [43/101], Loss: 0.0345\n",
            "Epoch [8/10], Step [44/101], Loss: 0.0003\n",
            "Epoch [8/10], Step [45/101], Loss: 0.0340\n",
            "Epoch [8/10], Step [46/101], Loss: 0.0004\n",
            "Epoch [8/10], Step [47/101], Loss: 0.0179\n",
            "Epoch [8/10], Step [48/101], Loss: 0.0009\n",
            "Epoch [8/10], Step [49/101], Loss: 0.0053\n",
            "Epoch [8/10], Step [50/101], Loss: 0.0049\n",
            "Epoch [8/10], Step [51/101], Loss: 0.0962\n",
            "Epoch [8/10], Step [52/101], Loss: 0.0013\n",
            "Epoch [8/10], Step [53/101], Loss: 0.0069\n",
            "Epoch [8/10], Step [54/101], Loss: 0.0055\n",
            "Epoch [8/10], Step [55/101], Loss: 0.0011\n",
            "Epoch [8/10], Step [56/101], Loss: 0.1163\n",
            "Epoch [8/10], Step [57/101], Loss: 0.0236\n",
            "Epoch [8/10], Step [58/101], Loss: 0.0030\n",
            "Epoch [8/10], Step [59/101], Loss: 0.1101\n",
            "Epoch [8/10], Step [60/101], Loss: 0.0028\n",
            "Epoch [8/10], Step [61/101], Loss: 0.0091\n",
            "Epoch [8/10], Step [62/101], Loss: 0.0041\n",
            "Epoch [8/10], Step [63/101], Loss: 0.0299\n",
            "Epoch [8/10], Step [64/101], Loss: 0.0156\n",
            "Epoch [8/10], Step [65/101], Loss: 0.0432\n",
            "Epoch [8/10], Step [66/101], Loss: 0.0139\n",
            "Epoch [8/10], Step [67/101], Loss: 0.0071\n",
            "Epoch [8/10], Step [68/101], Loss: 0.0266\n",
            "Epoch [8/10], Step [69/101], Loss: 0.0317\n",
            "Epoch [8/10], Step [70/101], Loss: 0.0804\n",
            "Epoch [8/10], Step [71/101], Loss: 0.0646\n",
            "Epoch [8/10], Step [72/101], Loss: 0.0292\n",
            "Epoch [8/10], Step [73/101], Loss: 0.0239\n",
            "Epoch [8/10], Step [74/101], Loss: 0.1926\n",
            "Epoch [8/10], Step [75/101], Loss: 0.0085\n",
            "Epoch [8/10], Step [76/101], Loss: 0.0014\n",
            "Epoch [8/10], Step [77/101], Loss: 0.0015\n",
            "Epoch [8/10], Step [78/101], Loss: 0.0386\n",
            "Epoch [8/10], Step [79/101], Loss: 0.0005\n",
            "Epoch [8/10], Step [80/101], Loss: 0.0565\n",
            "Epoch [8/10], Step [81/101], Loss: 0.0059\n",
            "Epoch [8/10], Step [82/101], Loss: 0.0029\n",
            "Epoch [8/10], Step [83/101], Loss: 0.0494\n",
            "Epoch [8/10], Step [84/101], Loss: 0.0772\n",
            "Epoch [8/10], Step [85/101], Loss: 0.0080\n",
            "Epoch [8/10], Step [86/101], Loss: 0.0052\n",
            "Epoch [8/10], Step [87/101], Loss: 0.0385\n",
            "Epoch [8/10], Step [88/101], Loss: 0.0081\n",
            "Epoch [8/10], Step [89/101], Loss: 0.0861\n",
            "Epoch [8/10], Step [90/101], Loss: 0.0332\n",
            "Epoch [8/10], Step [91/101], Loss: 0.0285\n",
            "Epoch [8/10], Step [92/101], Loss: 0.0246\n",
            "Epoch [8/10], Step [93/101], Loss: 0.1018\n",
            "Epoch [8/10], Step [94/101], Loss: 0.0011\n",
            "Epoch [8/10], Step [95/101], Loss: 0.0424\n",
            "Epoch [8/10], Step [96/101], Loss: 0.0912\n",
            "Epoch [8/10], Step [97/101], Loss: 0.0248\n",
            "Epoch [8/10], Step [98/101], Loss: 0.1782\n",
            "Epoch [8/10], Step [99/101], Loss: 0.0185\n",
            "Epoch [8/10], Step [100/101], Loss: 0.0004\n",
            "Epoch [8/10], Step [101/101], Loss: 0.0011\n",
            "Epoch finished\n",
            "Epoch [9/10], Step [1/101], Loss: 0.1890\n",
            "Epoch [9/10], Step [2/101], Loss: 0.0003\n",
            "Epoch [9/10], Step [3/101], Loss: 0.0034\n",
            "Epoch [9/10], Step [4/101], Loss: 0.0121\n",
            "Epoch [9/10], Step [5/101], Loss: 0.0412\n",
            "Epoch [9/10], Step [6/101], Loss: 0.0137\n",
            "Epoch [9/10], Step [7/101], Loss: 0.0079\n",
            "Epoch [9/10], Step [8/101], Loss: 0.1096\n",
            "Epoch [9/10], Step [9/101], Loss: 0.0720\n",
            "Epoch [9/10], Step [10/101], Loss: 0.0472\n",
            "Epoch [9/10], Step [11/101], Loss: 0.0091\n",
            "Epoch [9/10], Step [12/101], Loss: 0.0203\n",
            "Epoch [9/10], Step [13/101], Loss: 0.0081\n",
            "Epoch [9/10], Step [14/101], Loss: 0.0364\n",
            "Epoch [9/10], Step [15/101], Loss: 0.0218\n",
            "Epoch [9/10], Step [16/101], Loss: 0.2876\n",
            "Epoch [9/10], Step [17/101], Loss: 0.0092\n",
            "Epoch [9/10], Step [18/101], Loss: 0.0578\n",
            "Epoch [9/10], Step [19/101], Loss: 0.0236\n",
            "Epoch [9/10], Step [20/101], Loss: 0.0033\n",
            "Epoch [9/10], Step [21/101], Loss: 0.0007\n",
            "Epoch [9/10], Step [22/101], Loss: 0.1280\n",
            "Epoch [9/10], Step [23/101], Loss: 0.0640\n",
            "Epoch [9/10], Step [24/101], Loss: 0.0015\n",
            "Epoch [9/10], Step [25/101], Loss: 0.0078\n",
            "Epoch [9/10], Step [26/101], Loss: 0.1519\n",
            "Epoch [9/10], Step [27/101], Loss: 0.0682\n",
            "Epoch [9/10], Step [28/101], Loss: 0.0066\n",
            "Epoch [9/10], Step [29/101], Loss: 0.1033\n",
            "Epoch [9/10], Step [30/101], Loss: 0.0980\n",
            "Epoch [9/10], Step [31/101], Loss: 0.0274\n",
            "Epoch [9/10], Step [32/101], Loss: 0.0194\n",
            "Epoch [9/10], Step [33/101], Loss: 0.0287\n",
            "Epoch [9/10], Step [34/101], Loss: 0.0026\n",
            "Epoch [9/10], Step [35/101], Loss: 0.0508\n",
            "Epoch [9/10], Step [36/101], Loss: 0.0363\n",
            "Epoch [9/10], Step [37/101], Loss: 0.0102\n",
            "Epoch [9/10], Step [38/101], Loss: 0.1451\n",
            "Epoch [9/10], Step [39/101], Loss: 0.0214\n",
            "Epoch [9/10], Step [40/101], Loss: 0.0515\n",
            "Epoch [9/10], Step [41/101], Loss: 0.0529\n",
            "Epoch [9/10], Step [42/101], Loss: 0.0107\n",
            "Epoch [9/10], Step [43/101], Loss: 0.0779\n",
            "Epoch [9/10], Step [44/101], Loss: 0.1345\n",
            "Epoch [9/10], Step [45/101], Loss: 0.0828\n",
            "Epoch [9/10], Step [46/101], Loss: 0.0086\n",
            "Epoch [9/10], Step [47/101], Loss: 0.1307\n",
            "Epoch [9/10], Step [48/101], Loss: 0.1369\n",
            "Epoch [9/10], Step [49/101], Loss: 0.0188\n",
            "Epoch [9/10], Step [50/101], Loss: 0.1689\n",
            "Epoch [9/10], Step [51/101], Loss: 0.0168\n",
            "Epoch [9/10], Step [52/101], Loss: 0.0176\n",
            "Epoch [9/10], Step [53/101], Loss: 0.1023\n",
            "Epoch [9/10], Step [54/101], Loss: 0.1175\n",
            "Epoch [9/10], Step [55/101], Loss: 0.0098\n",
            "Epoch [9/10], Step [56/101], Loss: 0.0281\n",
            "Epoch [9/10], Step [57/101], Loss: 0.0078\n",
            "Epoch [9/10], Step [58/101], Loss: 0.0148\n",
            "Epoch [9/10], Step [59/101], Loss: 0.0462\n",
            "Epoch [9/10], Step [60/101], Loss: 0.2709\n",
            "Epoch [9/10], Step [61/101], Loss: 0.0042\n",
            "Epoch [9/10], Step [62/101], Loss: 0.0167\n",
            "Epoch [9/10], Step [63/101], Loss: 0.1172\n",
            "Epoch [9/10], Step [64/101], Loss: 0.0414\n",
            "Epoch [9/10], Step [65/101], Loss: 0.0734\n",
            "Epoch [9/10], Step [66/101], Loss: 0.0356\n",
            "Epoch [9/10], Step [67/101], Loss: 0.0029\n",
            "Epoch [9/10], Step [68/101], Loss: 0.0586\n",
            "Epoch [9/10], Step [69/101], Loss: 0.1059\n",
            "Epoch [9/10], Step [70/101], Loss: 0.0243\n",
            "Epoch [9/10], Step [71/101], Loss: 0.1525\n",
            "Epoch [9/10], Step [72/101], Loss: 0.3004\n",
            "Epoch [9/10], Step [73/101], Loss: 0.2820\n",
            "Epoch [9/10], Step [74/101], Loss: 0.0626\n",
            "Epoch [9/10], Step [75/101], Loss: 0.0073\n",
            "Epoch [9/10], Step [76/101], Loss: 0.0081\n",
            "Epoch [9/10], Step [77/101], Loss: 0.0692\n",
            "Epoch [9/10], Step [78/101], Loss: 0.1935\n",
            "Epoch [9/10], Step [79/101], Loss: 0.0484\n",
            "Epoch [9/10], Step [80/101], Loss: 0.0092\n",
            "Epoch [9/10], Step [81/101], Loss: 0.0161\n",
            "Epoch [9/10], Step [82/101], Loss: 0.0539\n",
            "Epoch [9/10], Step [83/101], Loss: 0.0049\n",
            "Epoch [9/10], Step [84/101], Loss: 0.0783\n",
            "Epoch [9/10], Step [85/101], Loss: 0.0341\n",
            "Epoch [9/10], Step [86/101], Loss: 0.0308\n",
            "Epoch [9/10], Step [87/101], Loss: 0.0399\n",
            "Epoch [9/10], Step [88/101], Loss: 0.0121\n",
            "Epoch [9/10], Step [89/101], Loss: 0.4263\n",
            "Epoch [9/10], Step [90/101], Loss: 0.0856\n",
            "Epoch [9/10], Step [91/101], Loss: 0.0931\n",
            "Epoch [9/10], Step [92/101], Loss: 0.0044\n",
            "Epoch [9/10], Step [93/101], Loss: 0.0969\n",
            "Epoch [9/10], Step [94/101], Loss: 0.0474\n",
            "Epoch [9/10], Step [95/101], Loss: 0.0546\n",
            "Epoch [9/10], Step [96/101], Loss: 0.1104\n",
            "Epoch [9/10], Step [97/101], Loss: 0.0249\n",
            "Epoch [9/10], Step [98/101], Loss: 0.1734\n",
            "Epoch [9/10], Step [99/101], Loss: 0.1592\n",
            "Epoch [9/10], Step [100/101], Loss: 0.0146\n",
            "Epoch [9/10], Step [101/101], Loss: 0.1063\n",
            "Epoch finished\n",
            "Epoch [10/10], Step [1/101], Loss: 0.0088\n",
            "Epoch [10/10], Step [2/101], Loss: 0.1557\n",
            "Epoch [10/10], Step [3/101], Loss: 0.0117\n",
            "Epoch [10/10], Step [4/101], Loss: 0.0978\n",
            "Epoch [10/10], Step [5/101], Loss: 0.0786\n",
            "Epoch [10/10], Step [6/101], Loss: 0.0420\n",
            "Epoch [10/10], Step [7/101], Loss: 0.1103\n",
            "Epoch [10/10], Step [8/101], Loss: 0.0167\n",
            "Epoch [10/10], Step [9/101], Loss: 0.1745\n",
            "Epoch [10/10], Step [10/101], Loss: 0.0473\n",
            "Epoch [10/10], Step [11/101], Loss: 0.0088\n",
            "Epoch [10/10], Step [12/101], Loss: 0.0137\n",
            "Epoch [10/10], Step [13/101], Loss: 0.0184\n",
            "Epoch [10/10], Step [14/101], Loss: 0.0128\n",
            "Epoch [10/10], Step [15/101], Loss: 0.0124\n",
            "Epoch [10/10], Step [16/101], Loss: 0.0115\n",
            "Epoch [10/10], Step [17/101], Loss: 0.0496\n",
            "Epoch [10/10], Step [18/101], Loss: 0.0111\n",
            "Epoch [10/10], Step [19/101], Loss: 0.1281\n",
            "Epoch [10/10], Step [20/101], Loss: 0.0588\n",
            "Epoch [10/10], Step [21/101], Loss: 0.1484\n",
            "Epoch [10/10], Step [22/101], Loss: 0.1030\n",
            "Epoch [10/10], Step [23/101], Loss: 0.1154\n",
            "Epoch [10/10], Step [24/101], Loss: 0.0128\n",
            "Epoch [10/10], Step [25/101], Loss: 0.0843\n",
            "Epoch [10/10], Step [26/101], Loss: 0.0097\n",
            "Epoch [10/10], Step [27/101], Loss: 0.1266\n",
            "Epoch [10/10], Step [28/101], Loss: 0.0071\n",
            "Epoch [10/10], Step [29/101], Loss: 0.0635\n",
            "Epoch [10/10], Step [30/101], Loss: 0.0206\n",
            "Epoch [10/10], Step [31/101], Loss: 0.0241\n",
            "Epoch [10/10], Step [32/101], Loss: 0.0106\n",
            "Epoch [10/10], Step [33/101], Loss: 0.1734\n",
            "Epoch [10/10], Step [34/101], Loss: 0.0232\n",
            "Epoch [10/10], Step [35/101], Loss: 0.0272\n",
            "Epoch [10/10], Step [36/101], Loss: 0.0716\n",
            "Epoch [10/10], Step [37/101], Loss: 0.0491\n",
            "Epoch [10/10], Step [38/101], Loss: 0.0199\n",
            "Epoch [10/10], Step [39/101], Loss: 0.0386\n",
            "Epoch [10/10], Step [40/101], Loss: 0.0597\n",
            "Epoch [10/10], Step [41/101], Loss: 0.0125\n",
            "Epoch [10/10], Step [42/101], Loss: 0.0051\n",
            "Epoch [10/10], Step [43/101], Loss: 0.0096\n",
            "Epoch [10/10], Step [44/101], Loss: 0.0073\n",
            "Epoch [10/10], Step [45/101], Loss: 0.0464\n",
            "Epoch [10/10], Step [46/101], Loss: 0.0036\n",
            "Epoch [10/10], Step [47/101], Loss: 0.0034\n",
            "Epoch [10/10], Step [48/101], Loss: 0.0266\n",
            "Epoch [10/10], Step [49/101], Loss: 0.0649\n",
            "Epoch [10/10], Step [50/101], Loss: 0.0008\n",
            "Epoch [10/10], Step [51/101], Loss: 0.0011\n",
            "Epoch [10/10], Step [52/101], Loss: 0.0118\n",
            "Epoch [10/10], Step [53/101], Loss: 0.0635\n",
            "Epoch [10/10], Step [54/101], Loss: 0.0426\n",
            "Epoch [10/10], Step [55/101], Loss: 0.0161\n",
            "Epoch [10/10], Step [56/101], Loss: 0.0138\n",
            "Epoch [10/10], Step [57/101], Loss: 0.0604\n",
            "Epoch [10/10], Step [58/101], Loss: 0.0090\n",
            "Epoch [10/10], Step [59/101], Loss: 0.0140\n",
            "Epoch [10/10], Step [60/101], Loss: 0.0244\n",
            "Epoch [10/10], Step [61/101], Loss: 0.0091\n",
            "Epoch [10/10], Step [62/101], Loss: 0.1156\n",
            "Epoch [10/10], Step [63/101], Loss: 0.0025\n",
            "Epoch [10/10], Step [64/101], Loss: 0.0007\n",
            "Epoch [10/10], Step [65/101], Loss: 0.0109\n",
            "Epoch [10/10], Step [66/101], Loss: 0.0315\n",
            "Epoch [10/10], Step [67/101], Loss: 0.0054\n",
            "Epoch [10/10], Step [68/101], Loss: 0.0249\n",
            "Epoch [10/10], Step [69/101], Loss: 0.0518\n",
            "Epoch [10/10], Step [70/101], Loss: 0.0106\n",
            "Epoch [10/10], Step [71/101], Loss: 0.0037\n",
            "Epoch [10/10], Step [72/101], Loss: 0.0045\n",
            "Epoch [10/10], Step [73/101], Loss: 0.0067\n",
            "Epoch [10/10], Step [74/101], Loss: 0.0040\n",
            "Epoch [10/10], Step [75/101], Loss: 0.0762\n",
            "Epoch [10/10], Step [76/101], Loss: 0.0022\n",
            "Epoch [10/10], Step [77/101], Loss: 0.0006\n",
            "Epoch [10/10], Step [78/101], Loss: 0.0898\n",
            "Epoch [10/10], Step [79/101], Loss: 0.0024\n",
            "Epoch [10/10], Step [80/101], Loss: 0.0106\n",
            "Epoch [10/10], Step [81/101], Loss: 0.0486\n",
            "Epoch [10/10], Step [82/101], Loss: 0.0213\n",
            "Epoch [10/10], Step [83/101], Loss: 0.0108\n",
            "Epoch [10/10], Step [84/101], Loss: 0.0036\n",
            "Epoch [10/10], Step [85/101], Loss: 0.0062\n",
            "Epoch [10/10], Step [86/101], Loss: 0.0110\n",
            "Epoch [10/10], Step [87/101], Loss: 0.0016\n",
            "Epoch [10/10], Step [88/101], Loss: 0.0007\n",
            "Epoch [10/10], Step [89/101], Loss: 0.0031\n",
            "Epoch [10/10], Step [90/101], Loss: 0.0353\n",
            "Epoch [10/10], Step [91/101], Loss: 0.0056\n",
            "Epoch [10/10], Step [92/101], Loss: 0.0020\n",
            "Epoch [10/10], Step [93/101], Loss: 0.0386\n",
            "Epoch [10/10], Step [94/101], Loss: 0.1646\n",
            "Epoch [10/10], Step [95/101], Loss: 0.0117\n",
            "Epoch [10/10], Step [96/101], Loss: 0.0141\n",
            "Epoch [10/10], Step [97/101], Loss: 0.0042\n",
            "Epoch [10/10], Step [98/101], Loss: 0.0577\n",
            "Epoch [10/10], Step [99/101], Loss: 0.0012\n",
            "Epoch [10/10], Step [100/101], Loss: 0.0007\n",
            "Epoch [10/10], Step [101/101], Loss: 0.0405\n",
            "Epoch finished\n",
            "Training finished!\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test(model)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6VBDFPYkz7W9",
        "outputId": "89be9036-24cb-4490-f63b-a1f156344eac"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy of the network on the test images: 95 %\n"
          ]
        }
      ]
    }
  ],
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4"
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "nbformat": 4,
  "nbformat_minor": 0
}