detectron2_comparison.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "a6fff84d",
   "metadata": {},
   "outputs": [],
   "source": [
    "import detectron2\n",
    "from detectron2.utils.logger import setup_logger\n",
    "setup_logger()\n",
    "\n",
    "# import some common libraries\n",
    "import numpy as np\n",
    "import os, json, cv2, random\n",
    "\n",
    "# import some common detectron2 utilities\n",
    "from detectron2 import model_zoo\n",
    "from detectron2.engine import DefaultPredictor\n",
    "from detectron2.config import get_cfg\n",
    "from detectron2.utils.visualizer import Visualizer\n",
    "from detectron2.data import MetadataCatalog, DatasetCatalog\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "1d1bec6f",
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_json(filename):\n",
    "    with open(filename) as f:\n",
    "        annotation = json.load(f)\n",
    "    return annotation\n",
    "\n",
    "def load_segmentation_dataset(f_p,label_names=None):\n",
    "    \"\"\"\"\n",
    "    Returns:\n",
    "        [dict]: Dictionary of list with names \n",
    "    \"\"\"\n",
    "    data=load_json(f_p)\n",
    "    cat_map={}\n",
    "    for cat in data[\"categories\"]:\n",
    "        if cat[\"name\"] in label_names:\n",
    "            cat_map[cat['id']]=cat[\"name\"] \n",
    "    image_map={}\n",
    "    for cat in data[\"images\"]:\n",
    "        image_map[cat['id']]=cat[\"file_name\"] \n",
    "    annos={}\n",
    "    for d in data[\"annotations\"]:\n",
    "        tmp=[]\n",
    "        seg=d[\"segmentation\"][0]\n",
    "        for i in range(0,len(seg)-1,2):\n",
    "            tmp.append([seg[i],seg[i+1]]) \n",
    "        if image_map[d[\"image_id\"]] not in annos:\n",
    "            annos[image_map[d[\"image_id\"]]]=[{\"class_id\":cat_map[d[\"category_id\"]],\"annotation\":tmp}]\n",
    "        else:\n",
    "            annos[image_map[d[\"image_id\"]]].append({\"class_id\":cat_map[d[\"category_id\"]],\"annotation\":tmp})\n",
    "    return annos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "f4a78877",
   "metadata": {},
   "outputs": [],
   "source": [
    "data_dir=\"/home/asad/projs/SMN/data/cucumber\"\n",
    "json_name=\"labels_my-project-name_2021-12-17-03-00-40.json\"\n",
    "classes=[\"color_board\",\"cucumber\",\"backbone\"] \n",
    "ann=load_segmentation_dataset(os.path.join(data_dir,json_name),classes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "f6706a0b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "11"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(ann['P00000000879558637815805_1076_  _032621132951.jpg'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "ee54b793",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dict_keys(['class_id', 'annotation'])\n",
      "dict_keys(['class_id', 'annotation'])\n",
      "dict_keys(['class_id', 'annotation'])\n",
      "dict_keys(['class_id', 'annotation'])\n",
      "dict_keys(['class_id', 'annotation'])\n",
      "dict_keys(['class_id', 'annotation'])\n",
      "dict_keys(['class_id', 'annotation'])\n",
      "dict_keys(['class_id', 'annotation'])\n",
      "dict_keys(['class_id', 'annotation'])\n",
      "dict_keys(['class_id', 'annotation'])\n",
      "dict_keys(['class_id', 'annotation'])\n"
     ]
    }
   ],
   "source": [
    "for item in ann['P00000000879558637815805_1076_  _032621132951.jpg']:\n",
    "    print(item.keys())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "579f86d1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(ann)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "3c4faa4d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# if your dataset is in COCO format, this cell can be replaced by the following three lines:\n",
    "# from detectron2.data.datasets import register_coco_instances\n",
    "# register_coco_instances(\"my_dataset_train\", {}, \"json_annotation_train.json\", \"path/to/image/dir\")\n",
    "# register_coco_instances(\"my_dataset_val\", {}, \"json_annotation_val.json\", \"path/to/image/dir\")\n",
    "\n",
    "from detectron2.structures import BoxMode\n",
    "\n",
    "def get_veg_dicts(data_dir): \n",
    "    \n",
    "    dataset_dicts = []\n",
    "    for idx,k in enumerate(ann.keys()):\n",
    "        c_fname = os.path.join(data_dir, k)\n",
    "        img = cv2.imread(c_fname)\n",
    "        if img is None:\n",
    "            raise(\"Some thing is wrong image is None\")\n",
    "        record = {}\n",
    "        height, width = img.shape[:2]\n",
    "        record[\"file_name\"] = c_fname\n",
    "        record[\"image_id\"] = idx\n",
    "        record[\"height\"] = height\n",
    "        record[\"width\"] = width\n",
    "        annos = ann[k]\n",
    "        objs = []\n",
    "        \n",
    "        for anno in annos:\n",
    "            if (anno[\"class_id\"]!=\"color_board\" and anno[\"class_id\"]!=\"cucumber\"):\n",
    "                continue\n",
    "            if (anno[\"class_id\"]==\"color_board\"):\n",
    "                catid=0\n",
    "            if (anno[\"class_id\"]==\"cucumber\"):\n",
    "                catid=1\n",
    "            px = [x for x, y in anno['annotation']]\n",
    "            py = [y for x, y in anno['annotation']]\n",
    "            poly = [(x + 0.5, y + 0.5) for x, y in zip(px, py)]\n",
    "            poly = [p for x in poly for p in x]\n",
    "            obj = {\n",
    "                \"bbox\": [np.min(px), np.min(py), np.max(px), np.max(py)],\n",
    "                \"bbox_mode\": BoxMode.XYXY_ABS,\n",
    "                \"segmentation\": [poly],\n",
    "                \"category_id\": catid,\n",
    "            }\n",
    "            objs.append(obj)\n",
    "        record[\"annotations\"] = objs\n",
    "        #print(f\"Processed images {idx}\")\n",
    "        dataset_dicts.append(record)\n",
    "    \n",
    "\n",
    "    return dataset_dicts\n",
    "\n",
    "#for d in [\"train\", \"val\"]:\n",
    "#    DatasetCatalog.register(\"balloon_\" + d, lambda d=d: get_balloon_dicts(\"balloon/\" + d))\n",
    "#    MetadataCatalog.get(\"balloon_\" + d).set(thing_classes=[\"balloon\"])\n",
    "#balloon_metadata = MetadataCatalog.get(\"balloon_train\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "2d68306d",
   "metadata": {},
   "outputs": [],
   "source": [
    "for d in [\"train\", \"val\"]:\n",
    "    DatasetCatalog.register(\"cucumber_\" + d, lambda d=d: get_veg_dicts(data_dir))\n",
    "    MetadataCatalog.get(\"cucumber_\" + d).set(thing_classes=[\"color_board\",\"cucumber\"])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "17df753d",
   "metadata": {},
   "outputs": [],
   "source": [
    "cucumber_metadata = MetadataCatalog.get(\"cucumber_train\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "7181c74b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "dataset_dicts = get_veg_dicts(data_dir)\n",
    "for d in random.sample(dataset_dicts, 2):\n",
    "    img = cv2.imread(d[\"file_name\"])\n",
    "    visualizer = Visualizer(img[:, :, ::-1], metadata=cucumber_metadata, scale=0.9)\n",
    "    out = visualizer.draw_dataset_dict(d)\n",
    "    plt.imshow(out.get_image())\n",
    "    cv2.imwrite(\"test.png\",out.get_image()[:, :, ::-1])\n",
    "    #cv2.imshow(out.get_image()[:, :, ::-1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "34a72a28",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32m[05/16 13:55:14 d2.engine.defaults]: \u001b[0mModel:\n",
      "GeneralizedRCNN(\n",
      "  (backbone): FPN(\n",
      "    (fpn_lateral2): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1))\n",
      "    (fpn_output2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
      "    (fpn_lateral3): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1))\n",
      "    (fpn_output3): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
      "    (fpn_lateral4): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1))\n",
      "    (fpn_output4): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
      "    (fpn_lateral5): Conv2d(2048, 256, kernel_size=(1, 1), stride=(1, 1))\n",
      "    (fpn_output5): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
      "    (top_block): LastLevelMaxPool()\n",
      "    (bottom_up): ResNet(\n",
      "      (stem): BasicStem(\n",
      "        (conv1): Conv2d(\n",
      "          3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False\n",
      "          (norm): FrozenBatchNorm2d(num_features=64, eps=1e-05)\n",
      "        )\n",
      "      )\n",
      "      (res2): Sequential(\n",
      "        (0): BottleneckBlock(\n",
      "          (shortcut): Conv2d(\n",
      "            64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05)\n",
      "          )\n",
      "          (conv1): Conv2d(\n",
      "            64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=64, eps=1e-05)\n",
      "          )\n",
      "          (conv2): Conv2d(\n",
      "            64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=64, eps=1e-05)\n",
      "          )\n",
      "          (conv3): Conv2d(\n",
      "            64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05)\n",
      "          )\n",
      "        )\n",
      "        (1): BottleneckBlock(\n",
      "          (conv1): Conv2d(\n",
      "            256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=64, eps=1e-05)\n",
      "          )\n",
      "          (conv2): Conv2d(\n",
      "            64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=64, eps=1e-05)\n",
      "          )\n",
      "          (conv3): Conv2d(\n",
      "            64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05)\n",
      "          )\n",
      "        )\n",
      "        (2): BottleneckBlock(\n",
      "          (conv1): Conv2d(\n",
      "            256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=64, eps=1e-05)\n",
      "          )\n",
      "          (conv2): Conv2d(\n",
      "            64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=64, eps=1e-05)\n",
      "          )\n",
      "          (conv3): Conv2d(\n",
      "            64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05)\n",
      "          )\n",
      "        )\n",
      "      )\n",
      "      (res3): Sequential(\n",
      "        (0): BottleneckBlock(\n",
      "          (shortcut): Conv2d(\n",
      "            256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=512, eps=1e-05)\n",
      "          )\n",
      "          (conv1): Conv2d(\n",
      "            256, 128, kernel_size=(1, 1), stride=(2, 2), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=128, eps=1e-05)\n",
      "          )\n",
      "          (conv2): Conv2d(\n",
      "            128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=128, eps=1e-05)\n",
      "          )\n",
      "          (conv3): Conv2d(\n",
      "            128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=512, eps=1e-05)\n",
      "          )\n",
      "        )\n",
      "        (1): BottleneckBlock(\n",
      "          (conv1): Conv2d(\n",
      "            512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=128, eps=1e-05)\n",
      "          )\n",
      "          (conv2): Conv2d(\n",
      "            128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=128, eps=1e-05)\n",
      "          )\n",
      "          (conv3): Conv2d(\n",
      "            128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=512, eps=1e-05)\n",
      "          )\n",
      "        )\n",
      "        (2): BottleneckBlock(\n",
      "          (conv1): Conv2d(\n",
      "            512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=128, eps=1e-05)\n",
      "          )\n",
      "          (conv2): Conv2d(\n",
      "            128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=128, eps=1e-05)\n",
      "          )\n",
      "          (conv3): Conv2d(\n",
      "            128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=512, eps=1e-05)\n",
      "          )\n",
      "        )\n",
      "        (3): BottleneckBlock(\n",
      "          (conv1): Conv2d(\n",
      "            512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=128, eps=1e-05)\n",
      "          )\n",
      "          (conv2): Conv2d(\n",
      "            128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=128, eps=1e-05)\n",
      "          )\n",
      "          (conv3): Conv2d(\n",
      "            128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=512, eps=1e-05)\n",
      "          )\n",
      "        )\n",
      "      )\n",
      "      (res4): Sequential(\n",
      "        (0): BottleneckBlock(\n",
      "          (shortcut): Conv2d(\n",
      "            512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05)\n",
      "          )\n",
      "          (conv1): Conv2d(\n",
      "            512, 256, kernel_size=(1, 1), stride=(2, 2), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05)\n",
      "          )\n",
      "          (conv2): Conv2d(\n",
      "            256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05)\n",
      "          )\n",
      "          (conv3): Conv2d(\n",
      "            256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05)\n",
      "          )\n",
      "        )\n",
      "        (1): BottleneckBlock(\n",
      "          (conv1): Conv2d(\n",
      "            1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05)\n",
      "          )\n",
      "          (conv2): Conv2d(\n",
      "            256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05)\n",
      "          )\n",
      "          (conv3): Conv2d(\n",
      "            256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05)\n",
      "          )\n",
      "        )\n",
      "        (2): BottleneckBlock(\n",
      "          (conv1): Conv2d(\n",
      "            1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05)\n",
      "          )\n",
      "          (conv2): Conv2d(\n",
      "            256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05)\n",
      "          )\n",
      "          (conv3): Conv2d(\n",
      "            256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05)\n",
      "          )\n",
      "        )\n",
      "        (3): BottleneckBlock(\n",
      "          (conv1): Conv2d(\n",
      "            1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05)\n",
      "          )\n",
      "          (conv2): Conv2d(\n",
      "            256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05)\n",
      "          )\n",
      "          (conv3): Conv2d(\n",
      "            256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05)\n",
      "          )\n",
      "        )\n",
      "        (4): BottleneckBlock(\n",
      "          (conv1): Conv2d(\n",
      "            1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05)\n",
      "          )\n",
      "          (conv2): Conv2d(\n",
      "            256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05)\n",
      "          )\n",
      "          (conv3): Conv2d(\n",
      "            256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05)\n",
      "          )\n",
      "        )\n",
      "        (5): BottleneckBlock(\n",
      "          (conv1): Conv2d(\n",
      "            1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05)\n",
      "          )\n",
      "          (conv2): Conv2d(\n",
      "            256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=256, eps=1e-05)\n",
      "          )\n",
      "          (conv3): Conv2d(\n",
      "            256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=1024, eps=1e-05)\n",
      "          )\n",
      "        )\n",
      "      )\n",
      "      (res5): Sequential(\n",
      "        (0): BottleneckBlock(\n",
      "          (shortcut): Conv2d(\n",
      "            1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=2048, eps=1e-05)\n",
      "          )\n",
      "          (conv1): Conv2d(\n",
      "            1024, 512, kernel_size=(1, 1), stride=(2, 2), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=512, eps=1e-05)\n",
      "          )\n",
      "          (conv2): Conv2d(\n",
      "            512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=512, eps=1e-05)\n",
      "          )\n",
      "          (conv3): Conv2d(\n",
      "            512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=2048, eps=1e-05)\n",
      "          )\n",
      "        )\n",
      "        (1): BottleneckBlock(\n",
      "          (conv1): Conv2d(\n",
      "            2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=512, eps=1e-05)\n",
      "          )\n",
      "          (conv2): Conv2d(\n",
      "            512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=512, eps=1e-05)\n",
      "          )\n",
      "          (conv3): Conv2d(\n",
      "            512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=2048, eps=1e-05)\n",
      "          )\n",
      "        )\n",
      "        (2): BottleneckBlock(\n",
      "          (conv1): Conv2d(\n",
      "            2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=512, eps=1e-05)\n",
      "          )\n",
      "          (conv2): Conv2d(\n",
      "            512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=512, eps=1e-05)\n",
      "          )\n",
      "          (conv3): Conv2d(\n",
      "            512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False\n",
      "            (norm): FrozenBatchNorm2d(num_features=2048, eps=1e-05)\n",
      "          )\n",
      "        )\n",
      "      )\n",
      "    )\n",
      "  )\n",
      "  (proposal_generator): RPN(\n",
      "    (rpn_head): StandardRPNHead(\n",
      "      (conv): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
      "      (objectness_logits): Conv2d(256, 3, kernel_size=(1, 1), stride=(1, 1))\n",
      "      (anchor_deltas): Conv2d(256, 12, kernel_size=(1, 1), stride=(1, 1))\n",
      "    )\n",
      "    (anchor_generator): DefaultAnchorGenerator(\n",
      "      (cell_anchors): BufferList()\n",
      "    )\n",
      "  )\n",
      "  (roi_heads): StandardROIHeads(\n",
      "    (box_pooler): ROIPooler(\n",
      "      (level_poolers): ModuleList(\n",
      "        (0): ROIAlign(output_size=(7, 7), spatial_scale=0.25, sampling_ratio=0, aligned=True)\n",
      "        (1): ROIAlign(output_size=(7, 7), spatial_scale=0.125, sampling_ratio=0, aligned=True)\n",
      "        (2): ROIAlign(output_size=(7, 7), spatial_scale=0.0625, sampling_ratio=0, aligned=True)\n",
      "        (3): ROIAlign(output_size=(7, 7), spatial_scale=0.03125, sampling_ratio=0, aligned=True)\n",
      "      )\n",
      "    )\n",
      "    (box_head): FastRCNNConvFCHead(\n",
      "      (flatten): Flatten(start_dim=1, end_dim=-1)\n",
      "      (fc1): Linear(in_features=12544, out_features=1024, bias=True)\n",
      "      (fc_relu1): ReLU()\n",
      "      (fc2): Linear(in_features=1024, out_features=1024, bias=True)\n",
      "      (fc_relu2): ReLU()\n",
      "    )\n",
      "    (box_predictor): FastRCNNOutputLayers(\n",
      "      (cls_score): Linear(in_features=1024, out_features=3, bias=True)\n",
      "      (bbox_pred): Linear(in_features=1024, out_features=8, bias=True)\n",
      "    )\n",
      "    (mask_pooler): ROIPooler(\n",
      "      (level_poolers): ModuleList(\n",
      "        (0): ROIAlign(output_size=(14, 14), spatial_scale=0.25, sampling_ratio=0, aligned=True)\n",
      "        (1): ROIAlign(output_size=(14, 14), spatial_scale=0.125, sampling_ratio=0, aligned=True)\n",
      "        (2): ROIAlign(output_size=(14, 14), spatial_scale=0.0625, sampling_ratio=0, aligned=True)\n",
      "        (3): ROIAlign(output_size=(14, 14), spatial_scale=0.03125, sampling_ratio=0, aligned=True)\n",
      "      )\n",
      "    )\n",
      "    (mask_head): MaskRCNNConvUpsampleHead(\n",
      "      (mask_fcn1): Conv2d(\n",
      "        256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)\n",
      "        (activation): ReLU()\n",
      "      )\n",
      "      (mask_fcn2): Conv2d(\n",
      "        256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)\n",
      "        (activation): ReLU()\n",
      "      )\n",
      "      (mask_fcn3): Conv2d(\n",
      "        256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)\n",
      "        (activation): ReLU()\n",
      "      )\n",
      "      (mask_fcn4): Conv2d(\n",
      "        256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)\n",
      "        (activation): ReLU()\n",
      "      )\n",
      "      (deconv_0): ConvTranspose2d(256, 256, kernel_size=(2, 2), stride=(2, 2))\n",
      "      (deconv_relu_0): ReLU()\n",
      "      (predictor): Conv2d(256, 2, kernel_size=(1, 1), stride=(1, 1))\n",
      "    )\n",
      "  )\n",
      ")\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32m[05/16 13:55:14 d2.data.build]: \u001b[0mRemoved 0 images with no usable annotations. 2 images left.\n",
      "\u001b[32m[05/16 13:55:14 d2.data.build]: \u001b[0mDistribution of instances among all 2 categories:\n",
      "\u001b[36m|  category   | #instances   |  category  | #instances   |\n",
      "|:-----------:|:-------------|:----------:|:-------------|\n",
      "| color_board | 2            |  cucumber  | 14           |\n",
      "|             |              |            |              |\n",
      "|    total    | 16           |            |              |\u001b[0m\n",
      "\u001b[32m[05/16 13:55:14 d2.data.dataset_mapper]: \u001b[0m[DatasetMapper] Augmentations used in training: [ResizeShortestEdge(short_edge_length=(640, 672, 704, 736, 768, 800), max_size=1333, sample_style='choice'), RandomFlip(), RandomRotation(angle=[0, 0]), RandomLighting(scale=1.1), RandomSaturation(intensity_min=1, intensity_max=1), RandomBrightness(intensity_min=1, intensity_max=1), RandomContrast(intensity_min=1, intensity_max=1)]\n",
      "\u001b[32m[05/16 13:55:14 d2.data.build]: \u001b[0mUsing training sampler TrainingSampler\n",
      "\u001b[32m[05/16 13:55:14 d2.data.common]: \u001b[0mSerializing 2 elements to byte tensors and concatenating them all ...\n",
      "\u001b[32m[05/16 13:55:14 d2.data.common]: \u001b[0mSerialized dataset takes 0.01 MiB\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Skip loading parameter 'roi_heads.box_predictor.cls_score.weight' to the model due to incompatible shapes: (81, 1024) in the checkpoint but (3, 1024) in the model! You might want to double check if this is expected.\n",
      "Skip loading parameter 'roi_heads.box_predictor.cls_score.bias' to the model due to incompatible shapes: (81,) in the checkpoint but (3,) in the model! You might want to double check if this is expected.\n",
      "Skip loading parameter 'roi_heads.box_predictor.bbox_pred.weight' to the model due to incompatible shapes: (320, 1024) in the checkpoint but (8, 1024) in the model! You might want to double check if this is expected.\n",
      "Skip loading parameter 'roi_heads.box_predictor.bbox_pred.bias' to the model due to incompatible shapes: (320,) in the checkpoint but (8,) in the model! You might want to double check if this is expected.\n",
      "Skip loading parameter 'roi_heads.mask_head.predictor.weight' to the model due to incompatible shapes: (80, 256, 1, 1) in the checkpoint but (2, 256, 1, 1) in the model! You might want to double check if this is expected.\n",
      "Skip loading parameter 'roi_heads.mask_head.predictor.bias' to the model due to incompatible shapes: (80,) in the checkpoint but (2,) in the model! You might want to double check if this is expected.\n",
      "Some model parameters or buffers are not found in the checkpoint:\n",
      "\u001b[34mroi_heads.box_predictor.bbox_pred.{bias, weight}\u001b[0m\n",
      "\u001b[34mroi_heads.box_predictor.cls_score.{bias, weight}\u001b[0m\n",
      "\u001b[34mroi_heads.mask_head.deconv_0.{bias, weight}\u001b[0m\n",
      "\u001b[34mroi_heads.mask_head.predictor.{bias, weight}\u001b[0m\n",
      "The checkpoint state_dict contains keys that are not used by the model:\n",
      "  \u001b[35mroi_heads.mask_head.deconv.{bias, weight}\u001b[0m\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32m[05/16 13:55:16 d2.engine.train_loop]: \u001b[0mStarting training from iteration 0\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/asad/anaconda/envs/detfastai/lib/python3.6/site-packages/torch/_tensor.py:575: UserWarning: floor_divide is deprecated, and will be removed in a future version of pytorch. It currently rounds toward 0 (like the 'trunc' function NOT 'floor'). This results in incorrect rounding for negative values.\n",
      "To keep the current behavior, use torch.div(a, b, rounding_mode='trunc'), or for actual floor division, use torch.div(a, b, rounding_mode='floor'). (Triggered internally at  /opt/conda/conda-bld/pytorch_1623448233824/work/aten/src/ATen/native/BinaryOps.cpp:467.)\n",
      "  return torch.floor_divide(self, other)\n",
      "/home/asad/anaconda/envs/detfastai/lib/python3.6/site-packages/torch/nn/functional.py:718: UserWarning: Named tensors and all their associated APIs are an experimental feature and subject to change. Please do not use them for anything important until they are released as stable. (Triggered internally at  /opt/conda/conda-bld/pytorch_1623448233824/work/c10/core/TensorImpl.h:1156.)\n",
      "  return torch.max_pool2d(input, kernel_size, stride, padding, dilation, ceil_mode)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32m[05/16 13:55:27 d2.utils.events]: \u001b[0m eta: 0:08:25  iter: 19  total_loss: 2.702  loss_cls: 0.9979  loss_box_reg: 0.9084  loss_mask: 0.692  loss_rpn_cls: 0.01862  loss_rpn_loc: 0.07638  time: 0.5355  data_time: 0.3420  lr: 4.9953e-06  max_mem: 2345M\n",
      "\u001b[32m[05/16 13:55:37 d2.utils.events]: \u001b[0m eta: 0:08:19  iter: 39  total_loss: 2.582  loss_cls: 0.9368  loss_box_reg: 0.864  loss_mask: 0.6894  loss_rpn_cls: 0.01808  loss_rpn_loc: 0.07121  time: 0.5350  data_time: 0.3472  lr: 9.9902e-06  max_mem: 2345M\n",
      "\u001b[32m[05/16 13:55:48 d2.utils.events]: \u001b[0m eta: 0:08:09  iter: 59  total_loss: 2.44  loss_cls: 0.8304  loss_box_reg: 0.8363  loss_mask: 0.6832  loss_rpn_cls: 0.01022  loss_rpn_loc: 0.06389  time: 0.5311  data_time: 0.3332  lr: 1.4985e-05  max_mem: 2345M\n",
      "\u001b[32m[05/16 13:55:58 d2.utils.events]: \u001b[0m eta: 0:07:57  iter: 79  total_loss: 2.298  loss_cls: 0.7044  loss_box_reg: 0.8597  loss_mask: 0.6751  loss_rpn_cls: 0.009266  loss_rpn_loc: 0.05683  time: 0.5274  data_time: 0.3274  lr: 1.998e-05  max_mem: 2345M\n",
      "\u001b[32m[05/16 13:56:09 d2.utils.events]: \u001b[0m eta: 0:07:44  iter: 99  total_loss: 2.192  loss_cls: 0.5999  loss_box_reg: 0.8704  loss_mask: 0.6639  loss_rpn_cls: 0.006102  loss_rpn_loc: 0.05548  time: 0.5268  data_time: 0.3332  lr: 2.4975e-05  max_mem: 2345M\n",
      "\u001b[32m[05/16 13:56:20 d2.utils.events]: \u001b[0m eta: 0:07:35  iter: 119  total_loss: 2.085  loss_cls: 0.5188  loss_box_reg: 0.8438  loss_mask: 0.6477  loss_rpn_cls: 0.00657  loss_rpn_loc: 0.05127  time: 0.5270  data_time: 0.3409  lr: 2.997e-05  max_mem: 2345M\n",
      "\u001b[32m[05/16 13:56:30 d2.utils.events]: \u001b[0m eta: 0:07:24  iter: 139  total_loss: 2.001  loss_cls: 0.4558  loss_box_reg: 0.8558  loss_mask: 0.6321  loss_rpn_cls: 0.006886  loss_rpn_loc: 0.05381  time: 0.5279  data_time: 0.3330  lr: 3.4965e-05  max_mem: 2345M\n",
      "\u001b[32m[05/16 13:56:41 d2.utils.events]: \u001b[0m eta: 0:07:16  iter: 159  total_loss: 1.905  loss_cls: 0.3949  loss_box_reg: 0.8451  loss_mask: 0.6138  loss_rpn_cls: 0.003528  loss_rpn_loc: 0.05046  time: 0.5268  data_time: 0.3339  lr: 3.996e-05  max_mem: 2345M\n",
      "\u001b[32m[05/16 13:56:52 d2.utils.events]: \u001b[0m eta: 0:07:06  iter: 179  total_loss: 1.782  loss_cls: 0.3424  loss_box_reg: 0.7909  loss_mask: 0.5922  loss_rpn_cls: 0.00495  loss_rpn_loc: 0.04307  time: 0.5307  data_time: 0.3653  lr: 4.4955e-05  max_mem: 2345M\n",
      "\u001b[32m[05/16 13:57:03 d2.utils.events]: \u001b[0m eta: 0:06:59  iter: 199  total_loss: 1.642  loss_cls: 0.2874  loss_box_reg: 0.7277  loss_mask: 0.5677  loss_rpn_cls: 0.00254  loss_rpn_loc: 0.04325  time: 0.5337  data_time: 0.3584  lr: 4.995e-05  max_mem: 2345M\n",
      "\u001b[32m[05/16 13:57:14 d2.utils.events]: \u001b[0m eta: 0:06:49  iter: 219  total_loss: 1.49  loss_cls: 0.2365  loss_box_reg: 0.6846  loss_mask: 0.5404  loss_rpn_cls: 0.001383  loss_rpn_loc: 0.04154  time: 0.5332  data_time: 0.3427  lr: 5.4945e-05  max_mem: 2345M\n",
      "\u001b[32m[05/16 13:57:25 d2.utils.events]: \u001b[0m eta: 0:06:39  iter: 239  total_loss: 1.34  loss_cls: 0.2008  loss_box_reg: 0.576  loss_mask: 0.5135  loss_rpn_cls: 0.00345  loss_rpn_loc: 0.03841  time: 0.5340  data_time: 0.3568  lr: 5.994e-05  max_mem: 2345M\n",
      "\u001b[32m[05/16 13:57:35 d2.utils.events]: \u001b[0m eta: 0:06:28  iter: 259  total_loss: 1.181  loss_cls: 0.1637  loss_box_reg: 0.4818  loss_mask: 0.4955  loss_rpn_cls: 0.0009099  loss_rpn_loc: 0.03977  time: 0.5324  data_time: 0.3286  lr: 6.4935e-05  max_mem: 2345M\n",
      "\u001b[32m[05/16 13:57:46 d2.utils.events]: \u001b[0m eta: 0:06:18  iter: 279  total_loss: 1.1  loss_cls: 0.1381  loss_box_reg: 0.4353  loss_mask: 0.4809  loss_rpn_cls: 0.0008417  loss_rpn_loc: 0.04037  time: 0.5340  data_time: 0.3659  lr: 6.993e-05  max_mem: 2345M\n",
      "\u001b[32m[05/16 13:57:56 d2.utils.events]: \u001b[0m eta: 0:06:04  iter: 299  total_loss: 1.017  loss_cls: 0.1203  loss_box_reg: 0.3873  loss_mask: 0.47  loss_rpn_cls: 0.0003274  loss_rpn_loc: 0.04134  time: 0.5295  data_time: 0.2895  lr: 7.4925e-05  max_mem: 2345M\n",
      "\u001b[32m[05/16 13:58:07 d2.utils.events]: \u001b[0m eta: 0:05:53  iter: 319  total_loss: 0.9263  loss_cls: 0.1071  loss_box_reg: 0.3416  loss_mask: 0.4392  loss_rpn_cls: 0.0003083  loss_rpn_loc: 0.03425  time: 0.5322  data_time: 0.3826  lr: 7.992e-05  max_mem: 2345M\n",
      "\u001b[32m[05/16 13:58:17 d2.utils.events]: \u001b[0m eta: 0:05:41  iter: 339  total_loss: 0.8821  loss_cls: 0.09844  loss_box_reg: 0.2972  loss_mask: 0.4311  loss_rpn_cls: 0.001821  loss_rpn_loc: 0.03461  time: 0.5301  data_time: 0.3155  lr: 8.4915e-05  max_mem: 2345M\n",
      "\u001b[32m[05/16 13:58:29 d2.utils.events]: \u001b[0m eta: 0:05:31  iter: 359  total_loss: 0.7931  loss_cls: 0.07847  loss_box_reg: 0.2558  loss_mask: 0.4142  loss_rpn_cls: 0.0002605  loss_rpn_loc: 0.03305  time: 0.5309  data_time: 0.3522  lr: 8.991e-05  max_mem: 2345M\n",
      "\u001b[32m[05/16 13:58:39 d2.utils.events]: \u001b[0m eta: 0:05:19  iter: 379  total_loss: 0.7275  loss_cls: 0.07689  loss_box_reg: 0.2076  loss_mask: 0.3986  loss_rpn_cls: 0.0008356  loss_rpn_loc: 0.03266  time: 0.5294  data_time: 0.3179  lr: 9.4905e-05  max_mem: 2345M\n",
      "\u001b[32m[05/16 13:58:49 d2.utils.events]: \u001b[0m eta: 0:05:09  iter: 399  total_loss: 0.6859  loss_cls: 0.06624  loss_box_reg: 0.1922  loss_mask: 0.3823  loss_rpn_cls: 0.0003227  loss_rpn_loc: 0.03155  time: 0.5286  data_time: 0.3435  lr: 9.99e-05  max_mem: 2345M\n",
      "\u001b[32m[05/16 13:58:59 d2.utils.events]: \u001b[0m eta: 0:04:57  iter: 419  total_loss: 0.6637  loss_cls: 0.07113  loss_box_reg: 0.1877  loss_mask: 0.369  loss_rpn_cls: 0.0003036  loss_rpn_loc: 0.03072  time: 0.5254  data_time: 0.2842  lr: 0.0001049  max_mem: 2345M\n",
      "\u001b[32m[05/16 13:59:09 d2.utils.events]: \u001b[0m eta: 0:04:46  iter: 439  total_loss: 0.5999  loss_cls: 0.0602  loss_box_reg: 0.15  loss_mask: 0.3488  loss_rpn_cls: 0.0006343  loss_rpn_loc: 0.03169  time: 0.5242  data_time: 0.3216  lr: 0.00010989  max_mem: 2345M\n",
      "\u001b[32m[05/16 13:59:18 d2.utils.events]: \u001b[0m eta: 0:04:36  iter: 459  total_loss: 0.5592  loss_cls: 0.06414  loss_box_reg: 0.1338  loss_mask: 0.3313  loss_rpn_cls: 0.0004197  loss_rpn_loc: 0.03087  time: 0.5226  data_time: 0.3188  lr: 0.00011489  max_mem: 2345M\n",
      "\u001b[32m[05/16 13:59:28 d2.utils.events]: \u001b[0m eta: 0:04:24  iter: 479  total_loss: 0.5119  loss_cls: 0.05427  loss_box_reg: 0.131  loss_mask: 0.3052  loss_rpn_cls: 0.0001269  loss_rpn_loc: 0.0279  time: 0.5205  data_time: 0.2986  lr: 0.00011988  max_mem: 2345M\n",
      "\u001b[32m[05/16 13:59:37 d2.utils.events]: \u001b[0m eta: 0:04:12  iter: 499  total_loss: 0.486  loss_cls: 0.05392  loss_box_reg: 0.1219  loss_mask: 0.2802  loss_rpn_cls: 0.0001785  loss_rpn_loc: 0.0256  time: 0.5182  data_time: 0.2878  lr: 0.00012488  max_mem: 2345M\n",
      "\u001b[32m[05/16 13:59:46 d2.utils.events]: \u001b[0m eta: 0:04:02  iter: 519  total_loss: 0.4685  loss_cls: 0.05335  loss_box_reg: 0.1146  loss_mask: 0.2668  loss_rpn_cls: 6.497e-05  loss_rpn_loc: 0.03065  time: 0.5157  data_time: 0.2787  lr: 0.00012987  max_mem: 2345M\n",
      "\u001b[32m[05/16 13:59:56 d2.utils.events]: \u001b[0m eta: 0:03:51  iter: 539  total_loss: 0.4185  loss_cls: 0.04836  loss_box_reg: 0.1055  loss_mask: 0.2332  loss_rpn_cls: 7.637e-05  loss_rpn_loc: 0.02712  time: 0.5149  data_time: 0.3077  lr: 0.00013487  max_mem: 2345M\n",
      "\u001b[32m[05/16 14:00:06 d2.utils.events]: \u001b[0m eta: 0:03:41  iter: 559  total_loss: 0.3933  loss_cls: 0.05387  loss_box_reg: 0.09544  loss_mask: 0.2112  loss_rpn_cls: 0.0001284  loss_rpn_loc: 0.02813  time: 0.5145  data_time: 0.3206  lr: 0.00013986  max_mem: 2345M\n",
      "\u001b[32m[05/16 14:00:16 d2.utils.events]: \u001b[0m eta: 0:03:30  iter: 579  total_loss: 0.3752  loss_cls: 0.04855  loss_box_reg: 0.09834  loss_mask: 0.1996  loss_rpn_cls: 0.0002118  loss_rpn_loc: 0.02753  time: 0.5128  data_time: 0.2990  lr: 0.00014486  max_mem: 2345M\n",
      "\u001b[32m[05/16 14:00:25 d2.utils.events]: \u001b[0m eta: 0:03:19  iter: 599  total_loss: 0.3486  loss_cls: 0.04531  loss_box_reg: 0.0878  loss_mask: 0.1823  loss_rpn_cls: 0.0001821  loss_rpn_loc: 0.02574  time: 0.5111  data_time: 0.2929  lr: 0.00014985  max_mem: 2345M\n",
      "\u001b[32m[05/16 14:00:35 d2.utils.events]: \u001b[0m eta: 0:03:09  iter: 619  total_loss: 0.3334  loss_cls: 0.04714  loss_box_reg: 0.08673  loss_mask: 0.1744  loss_rpn_cls: 0.0001075  loss_rpn_loc: 0.02459  time: 0.5098  data_time: 0.2962  lr: 0.00015485  max_mem: 2345M\n",
      "\u001b[32m[05/16 14:00:45 d2.utils.events]: \u001b[0m eta: 0:02:59  iter: 639  total_loss: 0.3232  loss_cls: 0.04591  loss_box_reg: 0.0889  loss_mask: 0.1608  loss_rpn_cls: 0.0001899  loss_rpn_loc: 0.02465  time: 0.5094  data_time: 0.3237  lr: 0.00015984  max_mem: 2345M\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[32m[05/16 14:00:54 d2.utils.events]: \u001b[0m eta: 0:02:49  iter: 659  total_loss: 0.3093  loss_cls: 0.03939  loss_box_reg: 0.08758  loss_mask: 0.1605  loss_rpn_cls: 0.0001582  loss_rpn_loc: 0.02279  time: 0.5082  data_time: 0.2986  lr: 0.00016484  max_mem: 2345M\n",
      "\u001b[32m[05/16 14:01:03 d2.utils.events]: \u001b[0m eta: 0:02:39  iter: 679  total_loss: 0.3038  loss_cls: 0.04231  loss_box_reg: 0.08392  loss_mask: 0.1502  loss_rpn_cls: 0.0001123  loss_rpn_loc: 0.02706  time: 0.5067  data_time: 0.2847  lr: 0.00016983  max_mem: 2345M\n",
      "\u001b[32m[05/16 14:01:12 d2.utils.events]: \u001b[0m eta: 0:02:28  iter: 699  total_loss: 0.2757  loss_cls: 0.03815  loss_box_reg: 0.07346  loss_mask: 0.1392  loss_rpn_cls: 9.153e-05  loss_rpn_loc: 0.02399  time: 0.5050  data_time: 0.2674  lr: 0.00017483  max_mem: 2345M\n",
      "\u001b[32m[05/16 14:01:22 d2.utils.events]: \u001b[0m eta: 0:02:18  iter: 719  total_loss: 0.2666  loss_cls: 0.03785  loss_box_reg: 0.07241  loss_mask: 0.1319  loss_rpn_cls: 0.0001503  loss_rpn_loc: 0.02203  time: 0.5038  data_time: 0.2953  lr: 0.00017982  max_mem: 2345M\n",
      "\u001b[32m[05/16 14:01:31 d2.utils.events]: \u001b[0m eta: 0:02:08  iter: 739  total_loss: 0.2572  loss_cls: 0.03466  loss_box_reg: 0.07548  loss_mask: 0.1248  loss_rpn_cls: 0.0001255  loss_rpn_loc: 0.0223  time: 0.5032  data_time: 0.3003  lr: 0.00018482  max_mem: 2345M\n",
      "\u001b[32m[05/16 14:01:41 d2.utils.events]: \u001b[0m eta: 0:01:58  iter: 759  total_loss: 0.2432  loss_cls: 0.03396  loss_box_reg: 0.06866  loss_mask: 0.1171  loss_rpn_cls: 0.0001623  loss_rpn_loc: 0.0203  time: 0.5026  data_time: 0.3088  lr: 0.00018981  max_mem: 2345M\n",
      "\u001b[32m[05/16 14:01:51 d2.utils.events]: \u001b[0m eta: 0:01:48  iter: 779  total_loss: 0.2497  loss_cls: 0.03461  loss_box_reg: 0.07309  loss_mask: 0.1151  loss_rpn_cls: 0.0004958  loss_rpn_loc: 0.02496  time: 0.5022  data_time: 0.3184  lr: 0.00019481  max_mem: 2345M\n",
      "\u001b[32m[05/16 14:02:01 d2.utils.events]: \u001b[0m eta: 0:01:38  iter: 799  total_loss: 0.2466  loss_cls: 0.0342  loss_box_reg: 0.07461  loss_mask: 0.1094  loss_rpn_cls: 0.0001014  loss_rpn_loc: 0.02316  time: 0.5017  data_time: 0.3129  lr: 0.0001998  max_mem: 2345M\n",
      "\u001b[32m[05/16 14:02:10 d2.utils.events]: \u001b[0m eta: 0:01:28  iter: 819  total_loss: 0.2304  loss_cls: 0.02881  loss_box_reg: 0.07466  loss_mask: 0.1028  loss_rpn_cls: 9.562e-05  loss_rpn_loc: 0.02161  time: 0.5011  data_time: 0.3062  lr: 0.0002048  max_mem: 2345M\n",
      "\u001b[32m[05/16 14:02:20 d2.utils.events]: \u001b[0m eta: 0:01:18  iter: 839  total_loss: 0.229  loss_cls: 0.03345  loss_box_reg: 0.06969  loss_mask: 0.1027  loss_rpn_cls: 0.0001691  loss_rpn_loc: 0.01954  time: 0.5006  data_time: 0.3101  lr: 0.00020979  max_mem: 2345M\n",
      "\u001b[32m[05/16 14:02:30 d2.utils.events]: \u001b[0m eta: 0:01:08  iter: 859  total_loss: 0.2149  loss_cls: 0.02722  loss_box_reg: 0.06611  loss_mask: 0.09571  loss_rpn_cls: 0.0001673  loss_rpn_loc: 0.01984  time: 0.5003  data_time: 0.3150  lr: 0.00021479  max_mem: 2345M\n",
      "\u001b[32m[05/16 14:02:40 d2.utils.events]: \u001b[0m eta: 0:00:59  iter: 879  total_loss: 0.2168  loss_cls: 0.03057  loss_box_reg: 0.07138  loss_mask: 0.09183  loss_rpn_cls: 0.0001242  loss_rpn_loc: 0.02106  time: 0.5004  data_time: 0.3197  lr: 0.00021978  max_mem: 2345M\n",
      "\u001b[32m[05/16 14:02:50 d2.utils.events]: \u001b[0m eta: 0:00:49  iter: 899  total_loss: 0.2081  loss_cls: 0.02801  loss_box_reg: 0.06832  loss_mask: 0.09093  loss_rpn_cls: 0.0003055  loss_rpn_loc: 0.02037  time: 0.5000  data_time: 0.3030  lr: 0.00022478  max_mem: 2345M\n",
      "\u001b[32m[05/16 14:03:00 d2.utils.events]: \u001b[0m eta: 0:00:39  iter: 919  total_loss: 0.2081  loss_cls: 0.03254  loss_box_reg: 0.06916  loss_mask: 0.08651  loss_rpn_cls: 0.0001819  loss_rpn_loc: 0.01811  time: 0.5007  data_time: 0.3436  lr: 0.00022977  max_mem: 2345M\n",
      "\u001b[32m[05/16 14:03:10 d2.utils.events]: \u001b[0m eta: 0:00:29  iter: 939  total_loss: 0.1993  loss_cls: 0.03308  loss_box_reg: 0.06666  loss_mask: 0.08429  loss_rpn_cls: 0.0003567  loss_rpn_loc: 0.01657  time: 0.4999  data_time: 0.2908  lr: 0.00023477  max_mem: 2345M\n",
      "\u001b[32m[05/16 14:03:20 d2.utils.events]: \u001b[0m eta: 0:00:19  iter: 959  total_loss: 0.1956  loss_cls: 0.02761  loss_box_reg: 0.06987  loss_mask: 0.08022  loss_rpn_cls: 0.0001244  loss_rpn_loc: 0.01734  time: 0.5006  data_time: 0.3313  lr: 0.00023976  max_mem: 2345M\n",
      "\u001b[32m[05/16 14:03:31 d2.utils.events]: \u001b[0m eta: 0:00:09  iter: 979  total_loss: 0.1936  loss_cls: 0.02825  loss_box_reg: 0.06376  loss_mask: 0.08179  loss_rpn_cls: 0.0001374  loss_rpn_loc: 0.01879  time: 0.5013  data_time: 0.3485  lr: 0.00024476  max_mem: 2345M\n",
      "\u001b[32m[05/16 14:03:46 d2.utils.events]: \u001b[0m eta: 0:00:00  iter: 999  total_loss: 0.1836  loss_cls: 0.02504  loss_box_reg: 0.05801  loss_mask: 0.08085  loss_rpn_cls: 0.000185  loss_rpn_loc: 0.01675  time: 0.5014  data_time: 0.3204  lr: 0.00024975  max_mem: 2345M\n",
      "\u001b[32m[05/16 14:03:46 d2.engine.hooks]: \u001b[0mOverall training speed: 998 iterations in 0:08:20 (0.5014 s / it)\n",
      "\u001b[32m[05/16 14:03:46 d2.engine.hooks]: \u001b[0mTotal training time: 0:08:29 (0:00:09 on hooks)\n"
     ]
    }
   ],
   "source": [
    "from detectron2.engine import DefaultTrainer\n",
    "\n",
    "cfg = get_cfg()\n",
    "cfg.merge_from_file(model_zoo.get_config_file(\"COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x.yaml\"))\n",
    "cfg.DATASETS.TRAIN = (\"cucumber_train\",)\n",
    "cfg.DATASETS.TEST = ()\n",
    "cfg.DATALOADER.NUM_WORKERS = 2\n",
    "cfg.MODEL.WEIGHTS = model_zoo.get_checkpoint_url(\"COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x.yaml\")  # Let training initialize from model zoo\n",
    "cfg.SOLVER.IMS_PER_BATCH = 2\n",
    "cfg.SOLVER.BASE_LR = 0.00025  # pick a good LR\n",
    "cfg.SOLVER.MAX_ITER = 1000    # 300 iterations seems good enough for this toy dataset; you will need to train longer for a practical dataset\n",
    "cfg.SOLVER.STEPS = []        # do not decay learning rate\n",
    "cfg.MODEL.ROI_HEADS.BATCH_SIZE_PER_IMAGE = 128   # faster, and good enough for this toy dataset (default: 512)\n",
    "cfg.MODEL.ROI_HEADS.NUM_CLASSES = 2  # only has one class (ballon). (see https://detectron2.readthedocs.io/tutorials/datasets.html#update-the-config-for-new-datasets)\n",
    "# NOTE: this config means the number of classes, but a few popular unofficial tutorials incorrect uses num_classes+1 here.\n",
    "cfg.MODEL.ROI_MASK_HEAD.NUM_DECONV=1\n",
    "cfg[\"INPUT\"][\"ROTATE\"] = [0, 0]\n",
    "cfg[\"INPUT\"][\"LIGHT_SCALE\"] = 1.1\n",
    "cfg[\"INPUT\"][\"Brightness_SCALE\"] = [1, 1]\n",
    "cfg[\"INPUT\"][\"Contrast_SCALE\"] = [1, 1]\n",
    "cfg[\"INPUT\"][\"Saturation_SCALE\"] = [1, 1]\n",
    "\n",
    "os.makedirs(cfg.OUTPUT_DIR, exist_ok=True)\n",
    "trainer = DefaultTrainer(cfg) \n",
    "trainer.resume_or_load(resume=False)\n",
    "trainer.train()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "a1695724",
   "metadata": {},
   "outputs": [],
   "source": [
    "## predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "80af6a3b",
   "metadata": {},
   "outputs": [],
   "source": [
    "cfg.MODEL.WEIGHTS = os.path.join(cfg.OUTPUT_DIR, \"model_final.pth\")  # path to the model we just trained\n",
    "cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.99   # set a custom testing threshold\n",
    "predictor = DefaultPredictor(cfg)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "0639b227",
   "metadata": {},
   "outputs": [],
   "source": [
    "from detectron2.utils.visualizer import ColorMode\n",
    "dataset_dicts = get_veg_dicts(data_dir)\n",
    "for i,d in enumerate(dataset_dicts):    \n",
    "    im = cv2.imread(d[\"file_name\"])\n",
    "    outputs = predictor(im)  # format is documented at https://detectron2.readthedocs.io/tutorials/models.html#model-output-format\n",
    "    v = Visualizer(im[:, :, ::-1],\n",
    "                   metadata=cucumber_metadata, \n",
    "                   scale=0.9, \n",
    "                   instance_mode=ColorMode.IMAGE_BW   # remove the colors of unsegmented pixels. This option is only available for segmentation models\n",
    "    )\n",
    "    out = v.draw_instance_predictions(outputs[\"instances\"].to(\"cpu\"))\n",
    "    \n",
    "    cv2.imwrite(f\"{i}.png\",out.get_image()[:, :, ::-1])\n",
    "    #cv2_imshow(out.get_image()[:, :, ::-1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c25d0292",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}