4_Pre_procesamiento_de_Datos_Dataset_y_DataLoader_2024.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/DiploDatos/AprendizajeProfundo/blob/add_2024_content/4_Pre_procesamiento_de_Datos_Dataset_y_DataLoader_2024.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Pre-procesamiento de Datos\n"
      ],
      "metadata": {
        "id": "GErP_xoZOrJI"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Importando líbrerias\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "from sklearn.preprocessing import MinMaxScaler\n",
        "\n",
        "import torch\n",
        "from torch.utils.data import TensorDataset, DataLoader, random_split\n",
        "\n",
        "from torchvision import datasets\n",
        "from torchvision.transforms import ToTensor, transforms"
      ],
      "metadata": {
        "id": "Z0tNDyGKRv2z"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Configuración del dispositivo\n",
        "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
        "device"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XOt0i3WMcJyH",
        "outputId": "98a8b375-92bc-46eb-855a-d58c4fb0d0e6"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "device(type='cpu')"
            ]
          },
          "metadata": {},
          "execution_count": 2
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Datos Estructurados"
      ],
      "metadata": {
        "id": "iUGS-XEdFWEf"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Clase TensorDataset"
      ],
      "metadata": {
        "id": "dX7KFvfVpTbN"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Arreglos"
      ],
      "metadata": {
        "id": "F4IbI16tFfM7"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Supongamos arreglos de una pequeña muestra de datos (analogía al batch)\n",
        "np.random.seed(42)\n",
        "n_muestras = 7\n",
        "n_features = 5\n",
        "features_array = np.array(np.random.rand(n_muestras, n_features))\n",
        "target_array = np.array(np.random.rand(n_muestras, 1))\n",
        "\n",
        "print(f'features: {features_array}\\n\\ntarget: {target_array}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3y1P2bj2-uSq",
        "outputId": "1503e5ac-c48f-40e9-a8f1-5cd7ac8ff72c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "features: [[0.37454012 0.95071431 0.73199394 0.59865848 0.15601864]\n",
            " [0.15599452 0.05808361 0.86617615 0.60111501 0.70807258]\n",
            " [0.02058449 0.96990985 0.83244264 0.21233911 0.18182497]\n",
            " [0.18340451 0.30424224 0.52475643 0.43194502 0.29122914]\n",
            " [0.61185289 0.13949386 0.29214465 0.36636184 0.45606998]\n",
            " [0.78517596 0.19967378 0.51423444 0.59241457 0.04645041]\n",
            " [0.60754485 0.17052412 0.06505159 0.94888554 0.96563203]]\n",
            "\n",
            "target: [[0.80839735]\n",
            " [0.30461377]\n",
            " [0.09767211]\n",
            " [0.68423303]\n",
            " [0.44015249]\n",
            " [0.12203823]\n",
            " [0.49517691]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Convertir arrays a tensores PyTorch\n",
        "features_torch = torch.tensor(features_array)\n",
        "target_torch = torch.tensor(target_array)\n",
        "\n",
        "# Crear un TensorDataset a partir de los dos tensores (features, target)\n",
        "dataset = TensorDataset(features_torch, target_torch)\n",
        "\n",
        "# Tamaño del TensorDataset\n",
        "print(f'Tamaño del TensorDataset: {len(dataset)}')\n",
        "\n",
        "# Retornar el último elemento del TensorDataset\n",
        "print(f'Datos del último elemento del TensorDataset:\\n\\t{dataset[-1]}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3m5WKZA1OmjS",
        "outputId": "595fba56-fe74-4fa9-aab9-76defde88f28"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Tamaño del TensorDataset: 7\n",
            "Datos del último elemento del TensorDataset:\n",
            "\t(tensor([0.6075, 0.1705, 0.0651, 0.9489, 0.9656], dtype=torch.float64), tensor([0.4952], dtype=torch.float64))\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Iterar a través del TensorDataset\n",
        "print(f'Valores de:')\n",
        "for i, data in enumerate(dataset):\n",
        "      print(f'\\t * features - muestra {i + 1}: {data[0]}')\n",
        "      print(f'\\t * target - muestra {i + 1}: {data[1]}', end='\\n\\n')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "41FBywuVEQeV",
        "outputId": "50dbfc88-6ee6-4089-921c-602c855b844e"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Valores de:\n",
            "\t * features - muestra 1: tensor([0.3745, 0.9507, 0.7320, 0.5987, 0.1560], dtype=torch.float64)\n",
            "\t * target - muestra 1: tensor([0.8084], dtype=torch.float64)\n",
            "\n",
            "\t * features - muestra 2: tensor([0.1560, 0.0581, 0.8662, 0.6011, 0.7081], dtype=torch.float64)\n",
            "\t * target - muestra 2: tensor([0.3046], dtype=torch.float64)\n",
            "\n",
            "\t * features - muestra 3: tensor([0.0206, 0.9699, 0.8324, 0.2123, 0.1818], dtype=torch.float64)\n",
            "\t * target - muestra 3: tensor([0.0977], dtype=torch.float64)\n",
            "\n",
            "\t * features - muestra 4: tensor([0.1834, 0.3042, 0.5248, 0.4319, 0.2912], dtype=torch.float64)\n",
            "\t * target - muestra 4: tensor([0.6842], dtype=torch.float64)\n",
            "\n",
            "\t * features - muestra 5: tensor([0.6119, 0.1395, 0.2921, 0.3664, 0.4561], dtype=torch.float64)\n",
            "\t * target - muestra 5: tensor([0.4402], dtype=torch.float64)\n",
            "\n",
            "\t * features - muestra 6: tensor([0.7852, 0.1997, 0.5142, 0.5924, 0.0465], dtype=torch.float64)\n",
            "\t * target - muestra 6: tensor([0.1220], dtype=torch.float64)\n",
            "\n",
            "\t * features - muestra 7: tensor([0.6075, 0.1705, 0.0651, 0.9489, 0.9656], dtype=torch.float64)\n",
            "\t * target - muestra 7: tensor([0.4952], dtype=torch.float64)\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Dataframe"
      ],
      "metadata": {
        "id": "gdVDT-XMRZXu"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "url_dataset = 'https://opendatanepal.com/dataset/a3832b84-fadf-4bdd-b7d8-cc6953cc5625/resource/ca6bca1d-4089-4c24-a12b-312d8d7341e1/download/analysis-of-water-samples-from-three-points-of-manohara-at-kathmandu.csv'\n",
        "dataframe = pd.read_csv(url_dataset, skiprows=[1, 2], encoding='unicode_escape')\n",
        "dataframe.info()\n",
        "dataframe"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 574
        },
        "id": "dQC_P0g0RfIw",
        "outputId": "84bdcc60-c66b-45b0-ee5b-853955d7310f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 4 entries, 0 to 3\n",
            "Data columns (total 16 columns):\n",
            " #   Column      Non-Null Count  Dtype  \n",
            "---  ------      --------------  -----  \n",
            " 0   Parameters  4 non-null      object \n",
            " 1   pH          4 non-null      float64\n",
            " 2   DO          4 non-null      float64\n",
            " 3   BOD         4 non-null      float64\n",
            " 4   COD         4 non-null      float64\n",
            " 5   TDS         4 non-null      int64  \n",
            " 6   EC          4 non-null      int64  \n",
            " 7   NH3-N       4 non-null      int64  \n",
            " 8   NO3-N       4 non-null      float64\n",
            " 9   NO2-N       4 non-null      float64\n",
            " 10  TOC         4 non-null      float64\n",
            " 11  TH          4 non-null      int64  \n",
            " 12  Mg          4 non-null      float64\n",
            " 13  Fe          4 non-null      float64\n",
            " 14  TC          4 non-null      int64  \n",
            " 15  E.coli      4 non-null      int64  \n",
            "dtypes: float64(9), int64(6), object(1)\n",
            "memory usage: 640.0+ bytes\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                Parameters   pH    DO   BOD   COD  TDS    EC  NH3-N  NO3-N  \\\n",
              "0       Manohara-pepsikola  7.4   7.0  14.5  23.7  620   870     60    2.3   \n",
              "1  Manohara-Mulpani bridge  7.2   8.9  11.1  12.3  100   160     20    1.4   \n",
              "2          Manohara-Sankhu  7.0  15.7   5.4  12.4   60    78     20    1.6   \n",
              "3       Manahara-Balkumari  7.6   3.9  23.8  40.5  980  1450     60    2.0   \n",
              "\n",
              "   NO2-N   TOC   TH    Mg   Fe    TC  E.coli  \n",
              "0    0.2   4.5   60   7.8  4.9  1600     140  \n",
              "1    0.1   2.0  120  23.7  0.3   900      80  \n",
              "2    0.1   6.7  130  12.5  0.4  1600     120  \n",
              "3    0.2  12.8   80  11.8  6.1   500      40  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-b366facb-b610-4491-84a0-5bb4e51814c8\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Parameters</th>\n",
              "      <th>pH</th>\n",
              "      <th>DO</th>\n",
              "      <th>BOD</th>\n",
              "      <th>COD</th>\n",
              "      <th>TDS</th>\n",
              "      <th>EC</th>\n",
              "      <th>NH3-N</th>\n",
              "      <th>NO3-N</th>\n",
              "      <th>NO2-N</th>\n",
              "      <th>TOC</th>\n",
              "      <th>TH</th>\n",
              "      <th>Mg</th>\n",
              "      <th>Fe</th>\n",
              "      <th>TC</th>\n",
              "      <th>E.coli</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Manohara-pepsikola</td>\n",
              "      <td>7.4</td>\n",
              "      <td>7.0</td>\n",
              "      <td>14.5</td>\n",
              "      <td>23.7</td>\n",
              "      <td>620</td>\n",
              "      <td>870</td>\n",
              "      <td>60</td>\n",
              "      <td>2.3</td>\n",
              "      <td>0.2</td>\n",
              "      <td>4.5</td>\n",
              "      <td>60</td>\n",
              "      <td>7.8</td>\n",
              "      <td>4.9</td>\n",
              "      <td>1600</td>\n",
              "      <td>140</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Manohara-Mulpani bridge</td>\n",
              "      <td>7.2</td>\n",
              "      <td>8.9</td>\n",
              "      <td>11.1</td>\n",
              "      <td>12.3</td>\n",
              "      <td>100</td>\n",
              "      <td>160</td>\n",
              "      <td>20</td>\n",
              "      <td>1.4</td>\n",
              "      <td>0.1</td>\n",
              "      <td>2.0</td>\n",
              "      <td>120</td>\n",
              "      <td>23.7</td>\n",
              "      <td>0.3</td>\n",
              "      <td>900</td>\n",
              "      <td>80</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Manohara-Sankhu</td>\n",
              "      <td>7.0</td>\n",
              "      <td>15.7</td>\n",
              "      <td>5.4</td>\n",
              "      <td>12.4</td>\n",
              "      <td>60</td>\n",
              "      <td>78</td>\n",
              "      <td>20</td>\n",
              "      <td>1.6</td>\n",
              "      <td>0.1</td>\n",
              "      <td>6.7</td>\n",
              "      <td>130</td>\n",
              "      <td>12.5</td>\n",
              "      <td>0.4</td>\n",
              "      <td>1600</td>\n",
              "      <td>120</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Manahara-Balkumari</td>\n",
              "      <td>7.6</td>\n",
              "      <td>3.9</td>\n",
              "      <td>23.8</td>\n",
              "      <td>40.5</td>\n",
              "      <td>980</td>\n",
              "      <td>1450</td>\n",
              "      <td>60</td>\n",
              "      <td>2.0</td>\n",
              "      <td>0.2</td>\n",
              "      <td>12.8</td>\n",
              "      <td>80</td>\n",
              "      <td>11.8</td>\n",
              "      <td>6.1</td>\n",
              "      <td>500</td>\n",
              "      <td>40</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-b366facb-b610-4491-84a0-5bb4e51814c8')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-b366facb-b610-4491-84a0-5bb4e51814c8 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-b366facb-b610-4491-84a0-5bb4e51814c8');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-22f8cad5-a0c4-4671-b126-0c5d9752ef73\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-22f8cad5-a0c4-4671-b126-0c5d9752ef73')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-22f8cad5-a0c4-4671-b126-0c5d9752ef73 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "  <div id=\"id_ae94d41d-5401-4e49-98f3-239a5aa47822\">\n",
              "    <style>\n",
              "      .colab-df-generate {\n",
              "        background-color: #E8F0FE;\n",
              "        border: none;\n",
              "        border-radius: 50%;\n",
              "        cursor: pointer;\n",
              "        display: none;\n",
              "        fill: #1967D2;\n",
              "        height: 32px;\n",
              "        padding: 0 0 0 0;\n",
              "        width: 32px;\n",
              "      }\n",
              "\n",
              "      .colab-df-generate:hover {\n",
              "        background-color: #E2EBFA;\n",
              "        box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "        fill: #174EA6;\n",
              "      }\n",
              "\n",
              "      [theme=dark] .colab-df-generate {\n",
              "        background-color: #3B4455;\n",
              "        fill: #D2E3FC;\n",
              "      }\n",
              "\n",
              "      [theme=dark] .colab-df-generate:hover {\n",
              "        background-color: #434B5C;\n",
              "        box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "        filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "        fill: #FFFFFF;\n",
              "      }\n",
              "    </style>\n",
              "    <button class=\"colab-df-generate\" onclick=\"generateWithVariable('dataframe')\"\n",
              "            title=\"Generate code using this dataframe.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "    <script>\n",
              "      (() => {\n",
              "      const buttonEl =\n",
              "        document.querySelector('#id_ae94d41d-5401-4e49-98f3-239a5aa47822 button.colab-df-generate');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      buttonEl.onclick = () => {\n",
              "        google.colab.notebook.generateWithVariable('dataframe');\n",
              "      }\n",
              "      })();\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "dataframe",
              "summary": "{\n  \"name\": \"dataframe\",\n  \"rows\": 4,\n  \"fields\": [\n    {\n      \"column\": \"Parameters\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 4,\n        \"samples\": [\n          \"Manohara-Mulpani bridge\",\n          \"Manahara-Balkumari\",\n          \"Manohara-pepsikola\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"pH\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.25819888974716104,\n        \"min\": 7.0,\n        \"max\": 7.6,\n        \"num_unique_values\": 4,\n        \"samples\": [\n          7.2,\n          7.6,\n          7.4\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"DO\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 4.99491408000845,\n        \"min\": 3.9,\n        \"max\": 15.7,\n        \"num_unique_values\": 4,\n        \"samples\": [\n          8.9,\n          3.9,\n          7.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"BOD\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 7.709301740970666,\n        \"min\": 5.4,\n        \"max\": 23.8,\n        \"num_unique_values\": 4,\n        \"samples\": [\n          11.1,\n          23.8,\n          14.5\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"COD\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 13.306483382171264,\n        \"min\": 12.3,\n        \"max\": 40.5,\n        \"num_unique_values\": 4,\n        \"samples\": [\n          12.3,\n          40.5,\n          23.7\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"TDS\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 441,\n        \"min\": 60,\n        \"max\": 980,\n        \"num_unique_values\": 4,\n        \"samples\": [\n          100,\n          980,\n          620\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"EC\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 646,\n        \"min\": 78,\n        \"max\": 1450,\n        \"num_unique_values\": 4,\n        \"samples\": [\n          160,\n          1450,\n          870\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"NH3-N\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 23,\n        \"min\": 20,\n        \"max\": 60,\n        \"num_unique_values\": 2,\n        \"samples\": [\n          20,\n          60\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"NO3-N\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.4031128874149274,\n        \"min\": 1.4,\n        \"max\": 2.3,\n        \"num_unique_values\": 4,\n        \"samples\": [\n          1.4,\n          2.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"NO2-N\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.05773502691896258,\n        \"min\": 0.1,\n        \"max\": 0.2,\n        \"num_unique_values\": 2,\n        \"samples\": [\n          0.1,\n          0.2\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"TOC\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 4.618080409289846,\n        \"min\": 2.0,\n        \"max\": 12.8,\n        \"num_unique_values\": 4,\n        \"samples\": [\n          2.0,\n          12.8\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"TH\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 33,\n        \"min\": 60,\n        \"max\": 130,\n        \"num_unique_values\": 4,\n        \"samples\": [\n          120,\n          80\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Mg\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 6.821778849146802,\n        \"min\": 7.8,\n        \"max\": 23.7,\n        \"num_unique_values\": 4,\n        \"samples\": [\n          23.7,\n          11.8\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Fe\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 3.013718633183928,\n        \"min\": 0.3,\n        \"max\": 6.1,\n        \"num_unique_values\": 4,\n        \"samples\": [\n          0.3,\n          6.1\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"TC\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 544,\n        \"min\": 500,\n        \"max\": 1600,\n        \"num_unique_values\": 3,\n        \"samples\": [\n          1600,\n          900\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"E.coli\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 44,\n        \"min\": 40,\n        \"max\": 140,\n        \"num_unique_values\": 4,\n        \"samples\": [\n          80,\n          40\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Mostrar los datos de features que se van a utilizar\n",
        "dataframe[['Mg', 'Fe', 'pH']]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 175
        },
        "id": "FmN1A__UXzo0",
        "outputId": "93a3289e-35a0-4f95-ece6-e509c9fdcec8"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "     Mg   Fe   pH\n",
              "0   7.8  4.9  7.4\n",
              "1  23.7  0.3  7.2\n",
              "2  12.5  0.4  7.0\n",
              "3  11.8  6.1  7.6"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-28da96bb-f528-46a5-a67d-9940b241384f\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Mg</th>\n",
              "      <th>Fe</th>\n",
              "      <th>pH</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>7.8</td>\n",
              "      <td>4.9</td>\n",
              "      <td>7.4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>23.7</td>\n",
              "      <td>0.3</td>\n",
              "      <td>7.2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>12.5</td>\n",
              "      <td>0.4</td>\n",
              "      <td>7.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>11.8</td>\n",
              "      <td>6.1</td>\n",
              "      <td>7.6</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-28da96bb-f528-46a5-a67d-9940b241384f')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-28da96bb-f528-46a5-a67d-9940b241384f button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-28da96bb-f528-46a5-a67d-9940b241384f');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-02657790-1cf6-4a5d-8ae5-1e7524d776a1\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-02657790-1cf6-4a5d-8ae5-1e7524d776a1')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-02657790-1cf6-4a5d-8ae5-1e7524d776a1 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "summary": "{\n  \"name\": \"dataframe[['Mg', 'Fe', 'pH']]\",\n  \"rows\": 4,\n  \"fields\": [\n    {\n      \"column\": \"Mg\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 6.821778849146802,\n        \"min\": 7.8,\n        \"max\": 23.7,\n        \"num_unique_values\": 4,\n        \"samples\": [\n          23.7,\n          11.8,\n          7.8\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Fe\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 3.013718633183928,\n        \"min\": 0.3,\n        \"max\": 6.1,\n        \"num_unique_values\": 4,\n        \"samples\": [\n          0.3,\n          6.1,\n          4.9\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"pH\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.25819888974716104,\n        \"min\": 7.0,\n        \"max\": 7.6,\n        \"num_unique_values\": 4,\n        \"samples\": [\n          7.2,\n          7.6,\n          7.4\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Crear los tensores de features y target, escalando los datos\n",
        "scaler = MinMaxScaler()\n",
        "\n",
        "features = torch.tensor(scaler.fit_transform(np.array(dataframe[['Mg', 'Fe', 'pH']]))).float()\n",
        "target = torch.tensor(np.array(dataframe['E.coli'])).float()\n",
        "\n",
        "# Crear un TensorDataset a partir de los tensores de features y target\n",
        "dataset = TensorDataset(features, target)\n",
        "dataset\n",
        "features, target, dataset"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XtLre2vSRGgV",
        "outputId": "626714de-59bf-4742-cc79-7568a22a34ee"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(tensor([[0.0000, 0.7931, 0.6667],\n",
              "         [1.0000, 0.0000, 0.3333],\n",
              "         [0.2956, 0.0172, 0.0000],\n",
              "         [0.2516, 1.0000, 1.0000]]),\n",
              " tensor([140.,  80., 120.,  40.]),\n",
              " <torch.utils.data.dataset.TensorDataset at 0x7ce3da3f7bb0>)"
            ]
          },
          "metadata": {},
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Iterar el TensorDataset\n",
        "print(f'Valores de:')\n",
        "for i, data in enumerate(dataset):\n",
        "      print(f'\\t * features - muestra {i + 1}: {data[0]}')\n",
        "      print(f'\\t * target - muestra {i + 1}: {data[1]}', end='\\n\\n')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sp830khGXClU",
        "outputId": "cffee3cf-242d-4c13-9215-c64f43e7f180"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Valores de:\n",
            "\t * features - muestra 1: tensor([0.0000, 0.7931, 0.6667])\n",
            "\t * target - muestra 1: 140.0\n",
            "\n",
            "\t * features - muestra 2: tensor([1.0000, 0.0000, 0.3333])\n",
            "\t * target - muestra 2: 80.0\n",
            "\n",
            "\t * features - muestra 3: tensor([0.2956, 0.0172, 0.0000])\n",
            "\t * target - muestra 3: 120.0\n",
            "\n",
            "\t * features - muestra 4: tensor([0.2516, 1.0000, 1.0000])\n",
            "\t * target - muestra 4: 40.0\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Tamaño del mini-batch\n",
        "train_mini_batch_size = 2\n",
        "# Número de observaciones\n",
        "number_rows = len(features)\n",
        "# Tamaño del conjunto de train, test y validación\n",
        "test_split = int(number_rows*0.25)\n",
        "validate_split = int(number_rows*0.25)\n",
        "train_split = number_rows - test_split - validate_split\n",
        "print(f'N° de observaciones: train: {train_split}, validate: {validate_split}, test: {test_split}')\n",
        "\n",
        "# random_split: Dividir los datos para entrenar, validar y testear\n",
        "train_set, validate_set, test_set = random_split(dataset, [train_split, validate_split, test_split])\n",
        "\n",
        "# Crear un cargador de datos para leer los datos por mini-batch y manejarlos en memoria\n",
        "train_loader = DataLoader(train_set, batch_size=train_mini_batch_size, shuffle=True)\n",
        "validate_loader = DataLoader(validate_set, batch_size=1)\n",
        "test_loader = DataLoader(test_set, batch_size=1)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "WkE9eKADWzoT",
        "outputId": "4f3da6a6-3bb8-4494-b066-af74f818e3a2"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "N° de observaciones: train: 2, validate: 1, test: 1\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_loader.dataset.dataset.tensors"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "iGCRRoAe94bO",
        "outputId": "b5314646-c748-4652-e747-08a873291e59"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(tensor([[0.0000, 0.7931, 0.6667],\n",
              "         [1.0000, 0.0000, 0.3333],\n",
              "         [0.2956, 0.0172, 0.0000],\n",
              "         [0.2516, 1.0000, 1.0000]]),\n",
              " tensor([140.,  80., 120.,  40.]))"
            ]
          },
          "metadata": {},
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Iterador sobre el cargador de batch de datos de entrenamiento\n",
        "mini_batchs = enumerate(train_loader)"
      ],
      "metadata": {
        "id": "WyY4FiU9o944"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Una iteración sobre el cargador de batch de datos de entrenamiento\n",
        "mini_batch_idx, (mini_batch_inputs, mini_batch_targets) = next(mini_batchs)\n",
        "print(f'N° de mini-batch: {mini_batch_idx}')\n",
        "print(f'Dimensiones de mini-batch de entradas: {mini_batch_inputs.shape}')\n",
        "print(f'Dimensiones de mini-batch de targets: {mini_batch_targets.shape}')\n",
        "print(f'Datos de entrada del mini-batch:\\n\\t {mini_batch_inputs}')\n",
        "print(f'Datos de targets del mini-batch:\\n\\t {mini_batch_targets}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Pp3VIKYJdhGy",
        "outputId": "3a2c2050-d75c-47bc-e5ce-cb6c61d7f094"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "N° de mini-batch: 0\n",
            "Dimensiones de mini-batch de entradas: torch.Size([2, 3])\n",
            "Dimensiones de mini-batch de targets: torch.Size([2])\n",
            "Datos de entrada del mini-batch:\n",
            "\t tensor([[0.2516, 1.0000, 1.0000],\n",
            "        [0.2956, 0.0172, 0.0000]])\n",
            "Datos de targets del mini-batch:\n",
            "\t tensor([ 40., 120.])\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Iteración sobre el cargador de batch de datos de entrenamiento\n",
        "for mini_batch_idx, (mini_batch_inputs, mini_batch_targets) in enumerate(train_loader):\n",
        "    print(f'N° de mini-batch: {mini_batch_idx}')\n",
        "    print(f'Dimensiones de mini-batch de entradas: {mini_batch_inputs.shape}')\n",
        "    print(f'Dimensiones de mini-batch de targets: {mini_batch_targets.shape}')\n",
        "    print(f'Datos de entrada del mini-batch:\\n\\t {mini_batch_inputs}')\n",
        "    print(f'Datos de targets del mini-batch:\\n\\t {mini_batch_targets}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "64928683-67fb-43b3-c995-db9a37e1fd48",
        "id": "ow5kek7_VVgN"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "N° de mini-batch: 0\n",
            "Dimensiones de mini-batch de entradas: torch.Size([2, 3])\n",
            "Dimensiones de mini-batch de targets: torch.Size([2])\n",
            "Datos de entrada del mini-batch:\n",
            "\t tensor([[0.2516, 1.0000, 1.0000],\n",
            "        [0.2956, 0.0172, 0.0000]])\n",
            "Datos de targets del mini-batch:\n",
            "\t tensor([ 40., 120.])\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(f'Número de mini-batches en el cargador de datos de entrenamiento: {train_loader.__len__()}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "L4MV4sE6_Wx2",
        "outputId": "e3150ad4-500e-46a0-a225-ca6377a0af7a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Número de mini-batches en el cargador de datos de entrenamiento: 1\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Datos No Estructurados - Imágenes\n",
        "---\n",
        "\n",
        "**¿Cómo normalizar datos de imagen usando PyTorch?**\n",
        "\n",
        "PyTorch proporciona una amplia gama de herramientas para trabajar con conjuntos de datos de imágenes. Una de las formas más comunes de normalizar datos de imágenes en PyTorch es mediante la función `transforms.Normalize`. Esta función toma dos argumentos: la media y la desviación estándar del conjunto de datos.\n",
        "\n",
        "- Paso 1: cargue el conjunto de datos de imágenes: usando la función `torchvision.datasets.ImageFolder`, que carga automáticamente todas las imágenes en una carpeta y las asigna a sus respectivas clases según los nombres de sus carpetas.\n",
        "- Paso 2: Calcule la media y la desviación estándar del conjunto de datos:  usando la función `torch.utils.data.DataLoader` y una función personalizada para calcular la media y la desviación estándar del conjunto de datos.\n",
        "- Paso 3: Normalizar el conjunto de datos: usando la función `transforms.Normalize`.\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "7jlr8KfPZlBC"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### MNIST\n",
        "\n",
        "Pytorch viene con un módulo integrado que proporciona conjuntos de datos listos para usar en muchas aplicaciones de aprendizaje profundo, como *visión por computadoras*, *reconocimiento de voz* y *procesamiento del lenguaje natural*. Así que, puedes introducirte fácilmente en el mundo de las redes neuronales sin necesidad de recopilar previamente un dataset.\n",
        "\n",
        "Trabajaremos con el conjunto de datos MNIST para mostrar cómo sería su pre-procesamiento, aprovechando que el módulo de torchvision nos provee fácilmente este dataset.\n",
        "\n",
        "El [dataset de MNIST (Modified National Institute of Standards and Technology)](https://en.wikipedia.org/wiki/MNIST_database) consiste en un conjunto de imágenes de dígitos escritos a mano que contiene 60 mil ejemplos y un conjunto de prueba de 10 mil imágenes. Cada imagen se compone de 28x28 píxeles, sus valores han sido previamente suavizados resultando en imágenes de tonalidades grises.\n",
        "\n",
        "Para descargar estos datos, vamos a usar el modulo [`datasets`](https://pytorch.org/vision/stable/datasets.html) de `torchvision`.\n",
        "\n",
        "Documentación:\n",
        "- https://pytorch.org/vision/stable/index.html\n",
        "\n",
        "La función [`datasets`](https://pytorch.org/vision/stable/datasets.html) usa los siguientes paramétros:\n",
        "- `root`: El directorio donde queremos descargar los datos. `\".\"` significa que guardará los datos en el mismo directorio donde se encuentra la notebook. También es posible pasarle un string con el path del directorio deseado.\n",
        "- `train`: Es una variable booleana que permite identificar si estamos descargando el conjunto de train o de test.\n",
        "- `transform`: Permite transformar los datos al descargarlos, en nuestro caso los transformarmos a tensores."
      ],
      "metadata": {
        "id": "TD-RLTYyLDkT"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Cargar imágenes de MNIST\n",
        "base_dir = '/content'\n",
        "mnist_train_images = datasets.MNIST(root=base_dir, train=True, download=True)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ItkwFZh1aUBa",
        "outputId": "c1028d6e-5b01-4725-c01f-15339a02edb1"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz\n",
            "Downloading http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz to /content/MNIST/raw/train-images-idx3-ubyte.gz\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 9912422/9912422 [00:00<00:00, 88438638.02it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Extracting /content/MNIST/raw/train-images-idx3-ubyte.gz to /content/MNIST/raw\n",
            "\n",
            "Downloading http://yann.lecun.com/exdb/mnist/train-labels-idx1-ubyte.gz\n",
            "Downloading http://yann.lecun.com/exdb/mnist/train-labels-idx1-ubyte.gz to /content/MNIST/raw/train-labels-idx1-ubyte.gz\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 28881/28881 [00:00<00:00, 74134451.54it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Extracting /content/MNIST/raw/train-labels-idx1-ubyte.gz to /content/MNIST/raw\n",
            "\n",
            "Downloading http://yann.lecun.com/exdb/mnist/t10k-images-idx3-ubyte.gz\n",
            "Downloading http://yann.lecun.com/exdb/mnist/t10k-images-idx3-ubyte.gz to /content/MNIST/raw/t10k-images-idx3-ubyte.gz\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 1648877/1648877 [00:00<00:00, 21407186.80it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Extracting /content/MNIST/raw/t10k-images-idx3-ubyte.gz to /content/MNIST/raw\n",
            "\n",
            "Downloading http://yann.lecun.com/exdb/mnist/t10k-labels-idx1-ubyte.gz\n",
            "Downloading http://yann.lecun.com/exdb/mnist/t10k-labels-idx1-ubyte.gz to /content/MNIST/raw/t10k-labels-idx1-ubyte.gz\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 4542/4542 [00:00<00:00, 3802500.75it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Extracting /content/MNIST/raw/t10k-labels-idx1-ubyte.gz to /content/MNIST/raw\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "mnist_train_images"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "YbvwrDbQuEgz",
        "outputId": "2fefb650-ea6a-4968-f4cd-39c2cd8a8e04"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Dataset MNIST\n",
              "    Number of datapoints: 60000\n",
              "    Root location: /content\n",
              "    Split: Train"
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "type(mnist_train_images)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Du-ZixT5uMxe",
        "outputId": "092fc627-ac16-48a4-9b60-cc90fbe07756"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "torchvision.datasets.mnist.MNIST"
            ]
          },
          "metadata": {},
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "type(mnist_train_images[0])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "thUrlMIt9nta",
        "outputId": "56373ab7-b8a3-4f42-b48f-226c70686224"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "tuple"
            ]
          },
          "metadata": {},
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Imagen de MNIST\n",
        "mnist_train_images[0][0]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 45
        },
        "id": "DjgjCfHgxK0e",
        "outputId": "69e28606-6c70-41cb-c089-b087ca281bae"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<PIL.Image.Image image mode=L size=28x28>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAABwAAAAcCAAAAABXZoBIAAABAElEQVR4nGNgGMyAWUhIqK5jvdSy/9/rGRgYGFhgEnJsVjYCwQwMDAxPJgV+vniQgYGBgREqZ7iXH8r6l/SV4dn7m8gmCt3++/fv37/Htn3/iMW+gDnZf/+e5WbQnoXNNXyMs/5GoQoxwVmf/n9kSGFiwAW49/11wynJoPzx4YIcRlyygR/+/i2XxCWru+vv32nSuGQFYv/83Y3b4p9/fzpAmSyoMnohpiwM1w5h06Q+5enfv39/bcMiJVF09+/fv39P+mFKiTtd/fv3799jgZiBJLT69t+/f/8eDuDEkDJf8+jv379/v7Ryo4qzMDAwMAQGMjBc3/y35wM2V1IfAABFF16Aa0wAOwAAAABJRU5ErkJggg==\n"
          },
          "metadata": {},
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Etiqueta de la imagen\n",
        "mnist_train_images[0][1]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "C4btpKE3af03",
        "outputId": "ca624991-6ca1-4ecf-8766-58b348720b91"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "5"
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Clases del dataset\n",
        "clases = mnist_train_images.classes\n",
        "clases"
      ],
      "metadata": {
        "id": "Y21CAp80Kiue",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "30d3934b-1270-45c3-c0b9-1f9db8a054ba"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['0 - zero',\n",
              " '1 - one',\n",
              " '2 - two',\n",
              " '3 - three',\n",
              " '4 - four',\n",
              " '5 - five',\n",
              " '6 - six',\n",
              " '7 - seven',\n",
              " '8 - eight',\n",
              " '9 - nine']"
            ]
          },
          "metadata": {},
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "'''Escalar el conjunto de entrenamiento y el conjunto de pruebas\n",
        "ToTensor() convierte una imagen PIL o numpy.ndarray (H x W x C) en el rango [0, 255]\n",
        "a un torch.FloatTensor de forma (C x H x W) en el rango [0.0, 1.0]\n",
        "Instanciar la clase de transformación para convertir la imagen a tensor'''\n",
        "transform = ToTensor()\n",
        "mnist_train_scaled_data = datasets.MNIST(root=\".\", train=True, download=False, transform=transform)\n",
        "\n",
        "mnist_test_scaled_data = datasets.MNIST(root=\".\", train=False, download=False, transform=transform)"
      ],
      "metadata": {
        "id": "9V2UFF6SP3mz"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "mnist_train_scaled_data"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FYOvWzi5Q2Bo",
        "outputId": "9693b5c2-3f5d-435d-f94c-b7ea24a672cd"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Dataset MNIST\n",
              "    Number of datapoints: 60000\n",
              "    Root location: .\n",
              "    Split: Train\n",
              "    StandardTransform\n",
              "Transform: ToTensor()"
            ]
          },
          "metadata": {},
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Primera observación (input y clase) - Dimensiones de la imagen: 1 canal - 28x28 pixeles\n",
        "print(f'Dimensiones de la entrada: {mnist_train_scaled_data[0][0].shape}')\n",
        "print(f'Clase: {mnist_train_scaled_data[0][1]}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rRCgtpR5-3PT",
        "outputId": "2e5a01ce-ce97-40fd-dc6c-107f5f22082d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Dimensiones de la entrada: torch.Size([1, 28, 28])\n",
            "Clase: 5\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Observación n° 60000 (input y clase) - Dimensiones de la imagen: 1 canal - 28x28 pixeles\n",
        "print(f'Dimensiones de la entrada: {mnist_train_scaled_data[59999][0].shape}')\n",
        "print(f'Clase: {mnist_train_scaled_data[59999][1]}')\n",
        "print(f'Valores:\\n\\t {mnist_train_scaled_data[0][0][0][:8]}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "L5-mWg8XwcPN",
        "outputId": "612055b1-f3c9-45c0-e645-4b6e8ba4f2fe"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Dimensiones de la entrada: torch.Size([1, 28, 28])\n",
            "Clase: 8\n",
            "Valores:\n",
            "\t tensor([[0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000,\n",
            "         0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000,\n",
            "         0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000,\n",
            "         0.0000],\n",
            "        [0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000,\n",
            "         0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000,\n",
            "         0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000,\n",
            "         0.0000],\n",
            "        [0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000,\n",
            "         0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000,\n",
            "         0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000,\n",
            "         0.0000],\n",
            "        [0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000,\n",
            "         0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000,\n",
            "         0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000,\n",
            "         0.0000],\n",
            "        [0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000,\n",
            "         0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000,\n",
            "         0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000,\n",
            "         0.0000],\n",
            "        [0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000,\n",
            "         0.0000, 0.0000, 0.0000, 0.0118, 0.0706, 0.0706, 0.0706, 0.4941, 0.5333,\n",
            "         0.6863, 0.1020, 0.6510, 1.0000, 0.9686, 0.4980, 0.0000, 0.0000, 0.0000,\n",
            "         0.0000],\n",
            "        [0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.1176,\n",
            "         0.1412, 0.3686, 0.6039, 0.6667, 0.9922, 0.9922, 0.9922, 0.9922, 0.9922,\n",
            "         0.8824, 0.6745, 0.9922, 0.9490, 0.7647, 0.2510, 0.0000, 0.0000, 0.0000,\n",
            "         0.0000],\n",
            "        [0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.0000, 0.1922, 0.9333,\n",
            "         0.9922, 0.9922, 0.9922, 0.9922, 0.9922, 0.9922, 0.9922, 0.9922, 0.9843,\n",
            "         0.3647, 0.3216, 0.3216, 0.2196, 0.1529, 0.0000, 0.0000, 0.0000, 0.0000,\n",
            "         0.0000]])\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "fig, ax = plt.subplots(figsize=(3, 3))\n",
        "ax.imshow(mnist_train_scaled_data[0][0][0], cmap=\"gray\")\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 293
        },
        "id": "saFWooeEizOZ",
        "outputId": "084ed0ae-84ad-4d13-e6b6-d331fe52ac28"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 300x300 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAARYAAAEUCAYAAADuhRlEAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAVSUlEQVR4nO3df3AUd/3H8VcC5PjR5GIK5IiQNFQFLQIOlpiCFGxKGkcGKM5Y1Bk6MmDrpSOg1IljQaptWlB0sNjyhybFEVKZKSD8kYqBJONI4iQFEZEMICPB5FKL5i6kJNDc5/uH9r69EnZz5BPuDp6Pmc9Mb9+f231323vNZnfvNsUYYwQAFqXGuwEAtx+CBYB1BAsA6wgWANYRLACsI1gAWEewALCOYAFgHcECwLrh8W7gg8LhsNra2pSenq6UlJR4twPgf4wx6urqUk5OjlJTXY5JzBB56aWXTF5envF4PGb27NmmsbFxQO9rbW01khgMRoKO1tZW18/xkARLVVWVSUtLM7/85S/NX//6V7Nq1SqTmZlpOjo6XN/b2dkZ9x3HYDBuPDo7O10/x0MSLLNnzzZ+vz/yuq+vz+Tk5Jjy8nLX9waDwbjvOAaDceMRDAZdP8fWT95evXpVzc3NKioqiixLTU1VUVGRjh49et383t5ehUKhqAEguVkPlrffflt9fX3Kzs6OWp6dna1AIHDd/PLycnm93siYNGmS7ZYA3GJxv9xcVlamYDAYGa2trfFuCcAgWb/cPHbsWA0bNkwdHR1Ryzs6OuTz+a6b7/F45PF4bLcBII6sH7GkpaVp1qxZqqmpiSwLh8OqqalRYWGh7c0BSESDuvxzA1VVVcbj8ZjKykpz6tQps3r1apOZmWkCgYDre7kqxGAk9hjIVaEhufP2S1/6kv71r39pw4YNCgQCmjlzpqqrq687oQvg9pRiTGL9mHYoFJLX6413GwBuIBgMKiMjw3FO3K8KAbj9ECwArCNYAFhHsACwjmABYB3BAsA6ggWAdQQLAOsIFgDWESwArCNYAFhHsACwjmABYB3BAsA6ggWAdQQLAOsIFgDWESwArCNYAFhHsACwjmABYB3BAsA6ggWAdQQLAOsIFgDWESwArCNYAFhHsACwjmABYB3BAsC64bZX+P3vf1+bNm2KWjZlyhSdPn3a9qZwCwwbNsyx7vV6h7yH0tJSx/ro0aMd61OmTHHdht/vd6z/6Ec/cqwvX77csd7T0+PawwsvvOBY/+DnKpFZDxZJuu+++/T73//+/zcyfEg2AyBBDcknfvjw4fL5fEOxagBJYEjOsZw5c0Y5OTmaPHmyvvKVr+jChQtDsRkACcr6EUtBQYEqKys1ZcoUtbe3a9OmTfrsZz+rkydPKj09/br5vb296u3tjbwOhUK2WwJwi1kPlpKSksg/T58+XQUFBcrLy9NvfvMbrVy58rr55eXlSXVSCoC7Ib/cnJmZqY997GM6e/Zsv/WysjIFg8HIaG1tHeqWAAyxIQ+Wy5cv69y5c5owYUK/dY/Ho4yMjKgBILlZ/1Po29/+thYtWqS8vDy1tbVp48aNGjZsmOt1flwvNzfXsZ6WluZYf+CBB1y3MXfuXMd6ZmamY33ZsmWu24i3ixcvus7Ztm2bY33p0qWO9a6uLsf6n//8Z9ce6urqXOckC+vBcvHiRS1fvlyXLl3SuHHjNHfuXDU0NGjcuHG2NwUgQVkPlqqqKturBJBk+K4QAOsIFgDWESwArCNYAFhHsACwjmABYF2KMcbEu4n3C4VCt+THgxLBzJkzHeuHDx92rN8p+8lNOBx2rH/ta19zXcfly5cH1UN7e7tj/T//+Y/rOlpaWgbVw60SDAZd75DniAWAdQQLAOsIFgDWESwArCNYAFhHsACwjmABYB0P/Ikjt6cXXLp0ybGeDPexNDY2us7p7Ox0rC9YsMCxfvXqVcf6r371K9ceYBdHLACsI1gAWEewALCOYAFgHcECwDqCBYB1BAsA67iPJY7+/e9/O9bXr1/vWP/CF77gWD927JhrD24P6nJz/Phxx/rDDz/suo7u7m7H+n333edY/+Y3v+m6DdxaHLEAsI5gAWAdwQLAOoIFgHUECwDrCBYA1hEsAKzjuUJJzO3ZLl1dXa7r2LFjh2N95cqVjvWvfvWrjvXdu3e79oDkMiTPFaqvr9eiRYuUk5OjlJQU7du3L6pujNGGDRs0YcIEjRo1SkVFRTpz5kysmwGQxGIOlu7ubs2YMUPbt2/vt75582Zt27ZNr7zyihobGzVmzBgVFxerp6dn0M0CSA4x39JfUlKikpKSfmvGGP30pz/V9773PS1evFiStHPnTmVnZ2vfvn167LHHBtctgKRg9eTt+fPnFQgEVFRUFFnm9XpVUFCgo0eP9vue3t5ehUKhqAEguVkNlkAgIEnKzs6OWp6dnR2pfVB5ebm8Xm9kTJo0yWZLAOIg7peby8rKFAwGI6O1tTXeLQEYJKvB4vP5JEkdHR1Ryzs6OiK1D/J4PMrIyIgaAJKb1WDJz8+Xz+dTTU1NZFkoFFJjY6MKCwttbgpAAov5qtDly5d19uzZyOvz58/r+PHjysrKUm5urtasWaMf/vCH+uhHP6r8/Hw988wzysnJ0ZIlS2z2DcnKie5gMDio969atcqx/tprr7muIxwOD6oHJJ6Yg6WpqSnqyXTr1q2TJK1YsUKVlZV6+umn1d3drdWrV6uzs1Nz585VdXW1Ro4caa9rAAkt5mCZP3++nL4FkJKSomeffVbPPvvsoBoDkLziflUIwO2HYAFgHcECwDqCBYB1BAsA6/ihpzvcmDFjHOsHDhxwrD/44IOO9Rt9E/79fve737nOQeIYkh96AgA3BAsA6wgWANYRLACsI1gAWEewALCOYAFgHfexwNG9997rWH/zzTcd652dna7bOHLkiGO9qanJsX6jR9G8J8H+F0963McCIC4IFgDWESwArCNYAFhHsACwjmABYB3BAsA67mPBoCxdutSxXlFR4bqO9PT0QfXw3e9+17G+c+dO13W0t7cPqoc7CfexAIgLggWAdQQLAOsIFgDWESwArCNYAFhHsACwjmABYF3MN8jV19dry5Ytam5uVnt7u/bu3aslS5ZE6o8//rheffXVqPcUFxerurp6QOvnBrnby7Rp01znbN261bH+0EMPDaqHHTt2uM557rnnHOv//Oc/B9XD7WRIbpDr7u7WjBkzHH+165FHHlF7e3tk7N69O9bNAEhiw2N9Q0lJietjMz0ej3w+3003BSC5Dck5ltraWo0fP15TpkzRk08+qUuXLt1wbm9vr0KhUNQAkNysB8sjjzyinTt3qqamRi+++KLq6upUUlKivr6+fueXl5fL6/VGxqRJk2y3BOAWi/lPITePPfZY5J8/+clPavr06br33ntVW1vb70m4srIyrVu3LvI6FAoRLkCSG/LLzZMnT9bYsWN19uzZfusej0cZGRlRA0ByG/JguXjxoi5duqQJEyYM9aYAJIiY72O5fPly5OjjU5/6lLZu3aoFCxYoKytLWVlZ2rRpk5YtWyafz6dz587p6aefVldXl/7yl7/I4/G4rp/7WO48mZmZjvVFixY51t1+TColJcW1h8OHDzvWH374Ydd13CkGch9LzOdYmpqatGDBgsjr986PrFixQi+//LJOnDihV199VZ2dncrJydHChQv1gx/8YEChAuD2EHOwzJ8/3/GRlW+88cagGgKQ/PiuEADrCBYA1hEsAKwjWABYR7AAsI4HliHp9fb2OtaHD3e/+Pnuu+861ouLix3rtbW1rtu4XfDAMgBxQbAAsI5gAWAdwQLAOoIFgHUECwDrCBYA1ln/aUrg/aZPn+4654tf/KJj/f7773esD+Q+FTenTp1yrNfX1w96G3cSjlgAWEewALCOYAFgHcECwDqCBYB1BAsA6wgWANYRLACs4wY5OJoyZYpjvbS01LH+6KOPum7D5/PF1FOs+vr6XOe0t7c71sPhsK127ggcsQCwjmABYB3BAsA6ggWAdQQLAOsIFgDWESwArIvpPpby8nK9/vrrOn36tEaNGqUHHnhAL774YtS9Dj09PfrWt76lqqoq9fb2qri4WD//+c+VnZ1tvXk4G8j9IcuXL3esu92ncs8998TS0pBoampyrD/33HOu6/jtb39rqx0oxiOWuro6+f1+NTQ06NChQ7p27ZoWLlyo7u7uyJy1a9fqwIED2rNnj+rq6tTW1jagm6QA3D5iOmKprq6Oel1ZWanx48erublZ8+bNUzAY1C9+8Qvt2rVLn/vc5yRJFRUV+vjHP66GhgZ95jOfsdc5gIQ1qHMswWBQkpSVlSVJam5u1rVr11RUVBSZM3XqVOXm5uro0aP9rqO3t1ehUChqAEhuNx0s4XBYa9as0Zw5czRt2jRJUiAQUFpamjIzM6PmZmdnKxAI9Lue8vJyeb3eyJg0adLNtgQgQdx0sPj9fp08eVJVVVWDaqCsrEzBYDAyWltbB7U+APF3U99uLi0t1cGDB1VfX6+JEydGlvt8Pl29elWdnZ1RRy0dHR03vELh8Xjk8Xhupg0ACSqmIxZjjEpLS7V3714dPnxY+fn5UfVZs2ZpxIgRqqmpiSxraWnRhQsXVFhYaKdjAAkvpiMWv9+vXbt2af/+/UpPT4+cN/F6vRo1apS8Xq9WrlypdevWKSsrSxkZGXrqqadUWFjIFaGb4Hbvzyc+8QnH+ksvveS6jalTp8bU01BobGx0rG/ZssWxvn//fsc6v6Vy68UULC+//LIkaf78+VHLKyoq9Pjjj0uSfvKTnyg1NVXLli2LukEOwJ0jpmAxxrjOGTlypLZv367t27ffdFMAkhvfFQJgHcECwDqCBYB1BAsA6wgWANbxXKEh8t4XM53s2LHDsT5z5kzH+uTJk2NpaUj88Y9/dKz/+Mc/dl3HG2+84Vi/cuVKTD0h/jhiAWAdwQLAOoIFgHUECwDrCBYA1hEsAKwjWABYR7AAsI4b5G6goKDAsb5+/XrH+uzZs1238eEPfzimnobCO++841jftm2bY/355593rL//mVO4c3DEAsA6ggWAdQQLAOsIFgDWESwArCNYAFhHsACwjvtYbmDp0qWDqttw6tQpx/rBgwcd6++++67rNtx+iKmzs9N1HcAHccQCwDqCBYB1BAsA6wgWANYRLACsI1gAWEewALDPxOD55583n/70p81dd91lxo0bZxYvXmxOnz4dNefBBx80kqLG17/+9QFvIxgMXvd+BoOROCMYDLp+jmM6Yqmrq5Pf71dDQ4MOHTqka9euaeHChdf9mM+qVavU3t4eGZs3b45lMwCSXEx33lZXV0e9rqys1Pjx49Xc3Kx58+ZFlo8ePVo+n89OhwCSzqDOsQSDQUnXP6f417/+tcaOHatp06aprKzM9ecPAdxebvq7QuFwWGvWrNGcOXM0bdq0yPIvf/nLysvLU05Ojk6cOKHvfOc7amlp0euvv97venp7e9Xb2xt5HQqFbrYlAIkilpO37/fEE0+YvLw809ra6jivpqbGSDJnz57tt75x48a4n4xiMBgDHwM5eXtTweL3+83EiRPN3//+d9e5ly9fNpJMdXV1v/Wenh4TDAYjo7W1Ne47jsFg3HgMJFhi+lPIGKOnnnpKe/fuVW1trfLz813fc/z4cUnShAkT+q17PB55PJ5Y2gCQ4GIKFr/fr127dmn//v1KT09XIBCQJHm9Xo0aNUrnzp3Trl279PnPf1533323Tpw4obVr12revHmaPn36kPwLAEhAsfwJpBscGlVUVBhjjLlw4YKZN2+eycrKMh6Px3zkIx8x69evH9Ch03u4QY7BSOwxkM9zyv8CI2GEQiF5vd54twHgBoLBoDIyMhzn8F0hANYRLACsI1gAWEewALCOYAFgHcECwDqCBYB1BAsA6wgWANYRLACsI1gAWEewALCOYAFgXcIFS4J92RrABwzkM5pwwdLV1RXvFgA4GMhnNOF+jyUcDqutrU3p6elKSUlRKBTSpEmT1Nra6vobEHDGvrTjTt2Pxhh1dXUpJydHqanOxyQ3/fiPoZKamqqJEydetzwjI+OO+o84lNiXdtyJ+3GgP8KWcH8KAUh+BAsA6xI+WDwejzZu3MgjQixgX9rBfnSXcCdvASS/hD9iAZB8CBYA1hEsAKwjWABYl/DBsn37dt1zzz0aOXKkCgoK9Kc//SneLSW8+vp6LVq0SDk5OUpJSdG+ffui6sYYbdiwQRMmTNCoUaNUVFSkM2fOxKfZBFZeXq77779f6enpGj9+vJYsWaKWlpaoOT09PfL7/br77rt11113admyZero6IhTx4kjoYPltdde07p167Rx40a9+eabmjFjhoqLi/XWW2/Fu7WE1t3drRkzZmj79u391jdv3qxt27bplVdeUWNjo8aMGaPi4mL19PTc4k4TW11dnfx+vxoaGnTo0CFdu3ZNCxcuVHd3d2TO2rVrdeDAAe3Zs0d1dXVqa2vTo48+GseuE0QsD4W/1WbPnm38fn/kdV9fn8nJyTHl5eVx7Cq5SDJ79+6NvA6Hw8bn85ktW7ZElnV2dhqPx2N2794dhw6Tx1tvvWUkmbq6OmPMf/fbiBEjzJ49eyJz/va3vxlJ5ujRo/FqMyEk7BHL1atX1dzcrKKiosiy1NRUFRUV6ejRo3HsLLmdP39egUAgar96vV4VFBSwX10Eg0FJUlZWliSpublZ165di9qXU6dOVW5u7h2/LxM2WN5++2319fUpOzs7anl2drYCgUCcukp+7+079mtswuGw1qxZozlz5mjatGmS/rsv09LSlJmZGTWXfZmA324GEpHf79fJkyf1hz/8Id6tJIWEPWIZO3ashg0bdt0Z9o6ODvl8vjh1lfze23fs14ErLS3VwYMHdeTIkaif9PD5fLp69ao6Ozuj5rMvEzhY0tLSNGvWLNXU1ESWhcNh1dTUqLCwMI6dJbf8/Hz5fL6o/RoKhdTY2Mh+/QBjjEpLS7V3714dPnxY+fn5UfVZs2ZpxIgRUfuypaVFFy5cYF/G++yxk6qqKuPxeExlZaU5deqUWb16tcnMzDSBQCDerSW0rq4uc+zYMXPs2DEjyWzdutUcO3bM/OMf/zDGGPPCCy+YzMxMs3//fnPixAmzePFik5+fb65cuRLnzhPLk08+abxer6mtrTXt7e2R8c4770TmPPHEEyY3N9ccPnzYNDU1mcLCQlNYWBjHrhNDQgeLMcb87Gc/M7m5uSYtLc3Mnj3bNDQ0xLulhHfkyBEj6bqxYsUKY8x/Lzk/88wzJjs723g8HvPQQw+ZlpaW+DadgPrbh5JMRUVFZM6VK1fMN77xDfOhD33IjB492ixdutS0t7fHr+kEwc8mALAuYc+xAEheBAsA6wgWANYRLACsI1gAWEewALCOYAFgHcECwDqCBYB1BAsA6wgWANYRLACs+z/cCStkNimn9wAAAABJRU5ErkJggg==\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Calular la media y la desviación para todo el conjunto de datos\n",
        "mini_batch_size = 10000\n",
        "trainloader = torch.utils.data.DataLoader(mnist_train_scaled_data, batch_size=mini_batch_size, shuffle=False)\n",
        "testloader = torch.utils.data.DataLoader(mnist_test_scaled_data, batch_size=mini_batch_size, shuffle=False)\n",
        "\n",
        "# Inicialización de variables\n",
        "suma_de_inputs = torch.tensor([0.0])\n",
        "suma_cuadrada_de_inputs = torch.tensor([0.0])\n",
        "contador_de_inputs = torch.tensor([0.0])\n",
        "\n",
        "# TrainLoader: Cálculo de la suma de los inputs y la suma de los cuadrados de los inputs\n",
        "print('TrainLoader:')\n",
        "for idx, (inputs, target) in enumerate(trainloader):\n",
        "  print(f'\\tIteración {idx + 1}')\n",
        "  n_observ, n_channels, image_size_weight, image_size_height = inputs.shape\n",
        "  suma_de_inputs += inputs.sum(axis=[0, 2, 3])\n",
        "  suma_cuadrada_de_inputs += (inputs ** 2).sum(axis=[0, 2, 3])\n",
        "  contador_de_inputs += n_observ\n",
        "\n",
        "# TestLoader: Cálculo de la suma de los inputs y la suma de los cuadrados de los inputs\n",
        "print('TestLoader')\n",
        "for idx, (inputs, target) in enumerate(testloader):\n",
        "  n_observ, n_channels, image_size_weight, image_size_height = inputs.shape\n",
        "  print(f'\\tIteración {idx + 1}')\n",
        "  suma_de_inputs += inputs.sum(axis=[0, 2, 3])\n",
        "  suma_cuadrada_de_inputs += (inputs ** 2).sum(axis=[0, 2, 3])\n",
        "  contador_de_inputs += n_observ\n",
        "\n",
        "  # Total de pixeles\n",
        "total_pixeles = contador_de_inputs * image_size_weight * image_size_height\n",
        "\n",
        "# Media, Varianza y Desviación\n",
        "media_gral = suma_de_inputs / total_pixeles\n",
        "varianza_gral  = (suma_cuadrada_de_inputs / total_pixeles) - (media_gral ** 2)\n",
        "desviacion_std_gral  = torch.sqrt(varianza_gral)\n",
        "print(f'Media: {media_gral} - Desviación: {desviacion_std_gral}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-1WH2m76y6cV",
        "outputId": "3d0a5018-69bf-4984-bfea-7ba13d1617ce"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "TrainLoader:\n",
            "\tIteración 1\n",
            "\tIteración 2\n",
            "\tIteración 3\n",
            "\tIteración 4\n",
            "\tIteración 5\n",
            "\tIteración 6\n",
            "TestLoader\n",
            "\tIteración 1\n",
            "Media: tensor([0.1309]) - Desviación: tensor([0.3084])\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Seteo para escalar y normalizar los datos y transformar en tensores\n",
        "transform = transforms.Compose([transforms.ToTensor(),\n",
        "                                transforms.Normalize((0.1307), (0.3081))\n",
        "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t])\n",
        "\n",
        "# Escalar y Normalizar el conjunto de entrenamiento y el conjunto de pruebas\n",
        "mnist_training_norm_data = datasets.MNIST('.',\n",
        "                                        train=True,\n",
        "                                        download=False,\n",
        "                                        transform=transform)\n",
        "mnist_test_norm_data = datasets.MNIST('.',\n",
        "                                    train=False,\n",
        "                                    download=False,\n",
        "                                    transform=transform)\n",
        "\n",
        "# Preparar el cargador de entrenamiento y el cargador de pruebas\n",
        "trainloader = torch.utils.data.DataLoader(mnist_training_norm_data,\n",
        "                                        batch_size=32,\n",
        "                                        shuffle=True,\n",
        "                                        num_workers=0)\n",
        "testloader = torch.utils.data.DataLoader(mnist_test_norm_data,\n",
        "                                        batch_size=32,\n",
        "                                        shuffle=False,\n",
        "                                        num_workers=0)\n",
        "\n",
        "# Iterador sobre el cargador de batch de datos de entrenamiento\n",
        "mini_batchs = enumerate(trainloader)"
      ],
      "metadata": {
        "id": "e_C3-6OKZq9F"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# 1ra iteración sobre el cargador de batch de datos de entrenamiento\n",
        "mini_batch_idx, (mini_batch_inputs, mini_batch_targets) = next(mini_batchs)\n",
        "print(f'N° de mini-batch: {mini_batch_idx}')\n",
        "print(f'Dimensiones de mini-batch de entradas: {mini_batch_inputs.shape}')\n",
        "print(f'Dimensiones de mini-batch de targets: {mini_batch_targets.shape}')\n",
        "print(f'Datos de entrada del mini-batch:\\n\\t {mini_batch_inputs[0][0][:8]}')\n",
        "print(f'Datos de target del mini-batch:\\n\\t {mini_batch_targets[:8]}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "e91a907d-0dd9-4d7e-8aba-904c67d9fc51",
        "id": "oID2TSCsq7zB"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "N° de mini-batch: 0\n",
            "Dimensiones de mini-batch de entradas: torch.Size([32, 1, 28, 28])\n",
            "Dimensiones de mini-batch de targets: torch.Size([32])\n",
            "Datos de entrada del mini-batch:\n",
            "\t tensor([[-0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242],\n",
            "        [-0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242],\n",
            "        [-0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242],\n",
            "        [-0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242],\n",
            "        [-0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242],\n",
            "        [-0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242],\n",
            "        [-0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.3478,  0.7977,  1.3450, -0.2842, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.3351,  2.4142,  1.1159, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242],\n",
            "        [-0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242,  1.5487,  2.8088,  2.2105,  1.0523, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242,  0.7722,  2.8088,  0.9632, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242]])\n",
            "Datos de targets del mini-batch:\n",
            "\t tensor([4, 1, 9, 7, 2, 8, 2, 1])\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# 2da iteración sobre el cargador de batch de datos de entrenamiento\n",
        "mini_batch_idx, (mini_batch_inputs, mini_batch_targets) = next(mini_batchs)\n",
        "print(f'N° de mini-batch: {mini_batch_idx}')\n",
        "print(f'Dimensiones de mini-batch de entradas: {mini_batch_inputs.shape}')\n",
        "print(f'Dimensiones de mini-batch de targets: {mini_batch_targets.shape}')\n",
        "print(f'Datos de entrada del mini-batch:\\n\\t {mini_batch_inputs[0][0][:8]}')\n",
        "print(f'Datos de target del mini-batch:\\n\\t {mini_batch_targets[:8]}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1Yj09rLI_7ln",
        "outputId": "352af178-a8a3-45f2-893e-41ba0cd55d5c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "N° de mini-batch: 1\n",
            "Dimensiones de mini-batch de entradas: torch.Size([32, 1, 28, 28])\n",
            "Dimensiones de mini-batch de targets: torch.Size([32])\n",
            "Datos de entrada del mini-batch:\n",
            "\t tensor([[-0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242],\n",
            "        [-0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242],\n",
            "        [-0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242],\n",
            "        [-0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242],\n",
            "        [-0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242],\n",
            "        [-0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242],\n",
            "        [-0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242],\n",
            "        [-0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242, -0.4242,  0.4922,  1.5105,  2.8088,\n",
            "          2.7960,  2.8088,  2.7960,  2.8088,  1.7650, -0.1569, -0.4242, -0.4242,\n",
            "         -0.4242, -0.4242, -0.4242, -0.4242]])\n",
            "Datos de target del mini-batch:\n",
            "\t tensor([9, 0, 4, 6, 6, 3, 1, 6])\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Clase ImageFolder\n",
        "\n",
        "Esta clase nos ayuda a crear fácilmente conjuntos de datos de validación y entrenamiento de PyTorch sin escribir clases personalizadas. Luego podemos usar estos conjuntos de datos para crear nuestros cargadores de datos iterables.\n",
        "\n",
        "Utilizaremos el conjunto de datos de clasificación de imágenes de [mariposas](https://www.kaggle.com/datasets/gpiosenka/butterfly-images40-species/) de Kaggle contiene 4955 imágenes para entrenamiento, 250 imágenes para validación y 250 imágenes para prueba. Y todas las imágenes son imágenes RGB de 224 × 224 dimensiones (con 3 canales de color).\n",
        "\n",
        "---\n",
        "\n",
        "**Estructura del directorio que contiene las imágenes**\n",
        "\n",
        "La carpeta raíz de los inputs debe contener una carpeta  `train`, `validation` y una `test`, cada una con carpetas para cada clase. En cada carpeta se almacenarán las imágenes correspondientes a cada clase y a cada set train/validation/test .\n",
        "\n",
        "    ├── train\n",
        "    │   ├── class1\n",
        "    |      ├── 1.jpg\n",
        "    │      ├── 2.jpg\n",
        "    │   ├── class2\n",
        "    |      ├── 1.jpg\n",
        "    │      ├── 2.jpg\n",
        "    ├── validation\n",
        "    │   ├── class1\n",
        "    |      ├── 1.jpg\n",
        "    │      ├── 2.jpg\n",
        "    │   ├── class2\n",
        "    |      ├── 1.jpg\n",
        "    │      ├── 2.jpg\n",
        "    ├── test\n",
        "    │   ├── class1\n",
        "    |      ├── 1.jpg\n",
        "    │      ├── 2.jpg\n",
        "    │   ├── class2\n",
        "    |      ├── 1.jpg\n",
        "    │      ├── 2.jpg\n"
      ],
      "metadata": {
        "id": "5n2CkZUFrxQu"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import os\n",
        "from google.colab import files\n",
        "# Cargar archivos externos desde directorio local\n",
        "base_dir = '/content/butterflies_classifier/'\n",
        "\n",
        "for dataset in ['train', 'validation']:\n",
        "    for clase in ['adonis', 'appollo']:\n",
        "      # Construir path del directorio\n",
        "      directory = os.path.join(base_dir, dataset, clase)\n",
        "      # Crear directorio\n",
        "      os.makedirs(directory, exist_ok=True)\n",
        "      # Cambiar de directorio\n",
        "      os.chdir(directory)\n",
        "      print(f'Cargar imágenes de {dataset.upper()} clase {clase.upper()} en {directory}')\n",
        "      uploaded = files.upload()"
      ],
      "metadata": {
        "id": "x7D5fRjdKgRO",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 598
        },
        "outputId": "1a7255f0-6a15-45cb-a6b0-b481f48b00ee"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Cargar imágenes de TRAIN clase ADONIS en /content/butterflies_classifier/train/adonis\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "     <input type=\"file\" id=\"files-2cc32ea9-6879-40d0-a420-4bcac34f5c06\" name=\"files[]\" multiple disabled\n",
              "        style=\"border:none\" />\n",
              "     <output id=\"result-2cc32ea9-6879-40d0-a420-4bcac34f5c06\">\n",
              "      Upload widget is only available when the cell has been executed in the\n",
              "      current browser session. Please rerun this cell to enable.\n",
              "      </output>\n",
              "      <script>// Copyright 2017 Google LLC\n",
              "//\n",
              "// Licensed under the Apache License, Version 2.0 (the \"License\");\n",
              "// you may not use this file except in compliance with the License.\n",
              "// You may obtain a copy of the License at\n",
              "//\n",
              "//      http://www.apache.org/licenses/LICENSE-2.0\n",
              "//\n",
              "// Unless required by applicable law or agreed to in writing, software\n",
              "// distributed under the License is distributed on an \"AS IS\" BASIS,\n",
              "// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
              "// See the License for the specific language governing permissions and\n",
              "// limitations under the License.\n",
              "\n",
              "/**\n",
              " * @fileoverview Helpers for google.colab Python module.\n",
              " */\n",
              "(function(scope) {\n",
              "function span(text, styleAttributes = {}) {\n",
              "  const element = document.createElement('span');\n",
              "  element.textContent = text;\n",
              "  for (const key of Object.keys(styleAttributes)) {\n",
              "    element.style[key] = styleAttributes[key];\n",
              "  }\n",
              "  return element;\n",
              "}\n",
              "\n",
              "// Max number of bytes which will be uploaded at a time.\n",
              "const MAX_PAYLOAD_SIZE = 100 * 1024;\n",
              "\n",
              "function _uploadFiles(inputId, outputId) {\n",
              "  const steps = uploadFilesStep(inputId, outputId);\n",
              "  const outputElement = document.getElementById(outputId);\n",
              "  // Cache steps on the outputElement to make it available for the next call\n",
              "  // to uploadFilesContinue from Python.\n",
              "  outputElement.steps = steps;\n",
              "\n",
              "  return _uploadFilesContinue(outputId);\n",
              "}\n",
              "\n",
              "// This is roughly an async generator (not supported in the browser yet),\n",
              "// where there are multiple asynchronous steps and the Python side is going\n",
              "// to poll for completion of each step.\n",
              "// This uses a Promise to block the python side on completion of each step,\n",
              "// then passes the result of the previous step as the input to the next step.\n",
              "function _uploadFilesContinue(outputId) {\n",
              "  const outputElement = document.getElementById(outputId);\n",
              "  const steps = outputElement.steps;\n",
              "\n",
              "  const next = steps.next(outputElement.lastPromiseValue);\n",
              "  return Promise.resolve(next.value.promise).then((value) => {\n",
              "    // Cache the last promise value to make it available to the next\n",
              "    // step of the generator.\n",
              "    outputElement.lastPromiseValue = value;\n",
              "    return next.value.response;\n",
              "  });\n",
              "}\n",
              "\n",
              "/**\n",
              " * Generator function which is called between each async step of the upload\n",
              " * process.\n",
              " * @param {string} inputId Element ID of the input file picker element.\n",
              " * @param {string} outputId Element ID of the output display.\n",
              " * @return {!Iterable<!Object>} Iterable of next steps.\n",
              " */\n",
              "function* uploadFilesStep(inputId, outputId) {\n",
              "  const inputElement = document.getElementById(inputId);\n",
              "  inputElement.disabled = false;\n",
              "\n",
              "  const outputElement = document.getElementById(outputId);\n",
              "  outputElement.innerHTML = '';\n",
              "\n",
              "  const pickedPromise = new Promise((resolve) => {\n",
              "    inputElement.addEventListener('change', (e) => {\n",
              "      resolve(e.target.files);\n",
              "    });\n",
              "  });\n",
              "\n",
              "  const cancel = document.createElement('button');\n",
              "  inputElement.parentElement.appendChild(cancel);\n",
              "  cancel.textContent = 'Cancel upload';\n",
              "  const cancelPromise = new Promise((resolve) => {\n",
              "    cancel.onclick = () => {\n",
              "      resolve(null);\n",
              "    };\n",
              "  });\n",
              "\n",
              "  // Wait for the user to pick the files.\n",
              "  const files = yield {\n",
              "    promise: Promise.race([pickedPromise, cancelPromise]),\n",
              "    response: {\n",
              "      action: 'starting',\n",
              "    }\n",
              "  };\n",
              "\n",
              "  cancel.remove();\n",
              "\n",
              "  // Disable the input element since further picks are not allowed.\n",
              "  inputElement.disabled = true;\n",
              "\n",
              "  if (!files) {\n",
              "    return {\n",
              "      response: {\n",
              "        action: 'complete',\n",
              "      }\n",
              "    };\n",
              "  }\n",
              "\n",
              "  for (const file of files) {\n",
              "    const li = document.createElement('li');\n",
              "    li.append(span(file.name, {fontWeight: 'bold'}));\n",
              "    li.append(span(\n",
              "        `(${file.type || 'n/a'}) - ${file.size} bytes, ` +\n",
              "        `last modified: ${\n",
              "            file.lastModifiedDate ? file.lastModifiedDate.toLocaleDateString() :\n",
              "                                    'n/a'} - `));\n",
              "    const percent = span('0% done');\n",
              "    li.appendChild(percent);\n",
              "\n",
              "    outputElement.appendChild(li);\n",
              "\n",
              "    const fileDataPromise = new Promise((resolve) => {\n",
              "      const reader = new FileReader();\n",
              "      reader.onload = (e) => {\n",
              "        resolve(e.target.result);\n",
              "      };\n",
              "      reader.readAsArrayBuffer(file);\n",
              "    });\n",
              "    // Wait for the data to be ready.\n",
              "    let fileData = yield {\n",
              "      promise: fileDataPromise,\n",
              "      response: {\n",
              "        action: 'continue',\n",
              "      }\n",
              "    };\n",
              "\n",
              "    // Use a chunked sending to avoid message size limits. See b/62115660.\n",
              "    let position = 0;\n",
              "    do {\n",
              "      const length = Math.min(fileData.byteLength - position, MAX_PAYLOAD_SIZE);\n",
              "      const chunk = new Uint8Array(fileData, position, length);\n",
              "      position += length;\n",
              "\n",
              "      const base64 = btoa(String.fromCharCode.apply(null, chunk));\n",
              "      yield {\n",
              "        response: {\n",
              "          action: 'append',\n",
              "          file: file.name,\n",
              "          data: base64,\n",
              "        },\n",
              "      };\n",
              "\n",
              "      let percentDone = fileData.byteLength === 0 ?\n",
              "          100 :\n",
              "          Math.round((position / fileData.byteLength) * 100);\n",
              "      percent.textContent = `${percentDone}% done`;\n",
              "\n",
              "    } while (position < fileData.byteLength);\n",
              "  }\n",
              "\n",
              "  // All done.\n",
              "  yield {\n",
              "    response: {\n",
              "      action: 'complete',\n",
              "    }\n",
              "  };\n",
              "}\n",
              "\n",
              "scope.google = scope.google || {};\n",
              "scope.google.colab = scope.google.colab || {};\n",
              "scope.google.colab._files = {\n",
              "  _uploadFiles,\n",
              "  _uploadFilesContinue,\n",
              "};\n",
              "})(self);\n",
              "</script> "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Saving 001.jpg to 001.jpg\n",
            "Saving 002.jpg to 002.jpg\n",
            "Saving 003.jpg to 003.jpg\n",
            "Saving 004.jpg to 004.jpg\n",
            "Cargar imágenes de TRAIN clase APPOLLO en /content/butterflies_classifier/train/appollo\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "     <input type=\"file\" id=\"files-0f1fba9c-4662-42b9-957d-c59488cc9fd5\" name=\"files[]\" multiple disabled\n",
              "        style=\"border:none\" />\n",
              "     <output id=\"result-0f1fba9c-4662-42b9-957d-c59488cc9fd5\">\n",
              "      Upload widget is only available when the cell has been executed in the\n",
              "      current browser session. Please rerun this cell to enable.\n",
              "      </output>\n",
              "      <script>// Copyright 2017 Google LLC\n",
              "//\n",
              "// Licensed under the Apache License, Version 2.0 (the \"License\");\n",
              "// you may not use this file except in compliance with the License.\n",
              "// You may obtain a copy of the License at\n",
              "//\n",
              "//      http://www.apache.org/licenses/LICENSE-2.0\n",
              "//\n",
              "// Unless required by applicable law or agreed to in writing, software\n",
              "// distributed under the License is distributed on an \"AS IS\" BASIS,\n",
              "// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
              "// See the License for the specific language governing permissions and\n",
              "// limitations under the License.\n",
              "\n",
              "/**\n",
              " * @fileoverview Helpers for google.colab Python module.\n",
              " */\n",
              "(function(scope) {\n",
              "function span(text, styleAttributes = {}) {\n",
              "  const element = document.createElement('span');\n",
              "  element.textContent = text;\n",
              "  for (const key of Object.keys(styleAttributes)) {\n",
              "    element.style[key] = styleAttributes[key];\n",
              "  }\n",
              "  return element;\n",
              "}\n",
              "\n",
              "// Max number of bytes which will be uploaded at a time.\n",
              "const MAX_PAYLOAD_SIZE = 100 * 1024;\n",
              "\n",
              "function _uploadFiles(inputId, outputId) {\n",
              "  const steps = uploadFilesStep(inputId, outputId);\n",
              "  const outputElement = document.getElementById(outputId);\n",
              "  // Cache steps on the outputElement to make it available for the next call\n",
              "  // to uploadFilesContinue from Python.\n",
              "  outputElement.steps = steps;\n",
              "\n",
              "  return _uploadFilesContinue(outputId);\n",
              "}\n",
              "\n",
              "// This is roughly an async generator (not supported in the browser yet),\n",
              "// where there are multiple asynchronous steps and the Python side is going\n",
              "// to poll for completion of each step.\n",
              "// This uses a Promise to block the python side on completion of each step,\n",
              "// then passes the result of the previous step as the input to the next step.\n",
              "function _uploadFilesContinue(outputId) {\n",
              "  const outputElement = document.getElementById(outputId);\n",
              "  const steps = outputElement.steps;\n",
              "\n",
              "  const next = steps.next(outputElement.lastPromiseValue);\n",
              "  return Promise.resolve(next.value.promise).then((value) => {\n",
              "    // Cache the last promise value to make it available to the next\n",
              "    // step of the generator.\n",
              "    outputElement.lastPromiseValue = value;\n",
              "    return next.value.response;\n",
              "  });\n",
              "}\n",
              "\n",
              "/**\n",
              " * Generator function which is called between each async step of the upload\n",
              " * process.\n",
              " * @param {string} inputId Element ID of the input file picker element.\n",
              " * @param {string} outputId Element ID of the output display.\n",
              " * @return {!Iterable<!Object>} Iterable of next steps.\n",
              " */\n",
              "function* uploadFilesStep(inputId, outputId) {\n",
              "  const inputElement = document.getElementById(inputId);\n",
              "  inputElement.disabled = false;\n",
              "\n",
              "  const outputElement = document.getElementById(outputId);\n",
              "  outputElement.innerHTML = '';\n",
              "\n",
              "  const pickedPromise = new Promise((resolve) => {\n",
              "    inputElement.addEventListener('change', (e) => {\n",
              "      resolve(e.target.files);\n",
              "    });\n",
              "  });\n",
              "\n",
              "  const cancel = document.createElement('button');\n",
              "  inputElement.parentElement.appendChild(cancel);\n",
              "  cancel.textContent = 'Cancel upload';\n",
              "  const cancelPromise = new Promise((resolve) => {\n",
              "    cancel.onclick = () => {\n",
              "      resolve(null);\n",
              "    };\n",
              "  });\n",
              "\n",
              "  // Wait for the user to pick the files.\n",
              "  const files = yield {\n",
              "    promise: Promise.race([pickedPromise, cancelPromise]),\n",
              "    response: {\n",
              "      action: 'starting',\n",
              "    }\n",
              "  };\n",
              "\n",
              "  cancel.remove();\n",
              "\n",
              "  // Disable the input element since further picks are not allowed.\n",
              "  inputElement.disabled = true;\n",
              "\n",
              "  if (!files) {\n",
              "    return {\n",
              "      response: {\n",
              "        action: 'complete',\n",
              "      }\n",
              "    };\n",
              "  }\n",
              "\n",
              "  for (const file of files) {\n",
              "    const li = document.createElement('li');\n",
              "    li.append(span(file.name, {fontWeight: 'bold'}));\n",
              "    li.append(span(\n",
              "        `(${file.type || 'n/a'}) - ${file.size} bytes, ` +\n",
              "        `last modified: ${\n",
              "            file.lastModifiedDate ? file.lastModifiedDate.toLocaleDateString() :\n",
              "                                    'n/a'} - `));\n",
              "    const percent = span('0% done');\n",
              "    li.appendChild(percent);\n",
              "\n",
              "    outputElement.appendChild(li);\n",
              "\n",
              "    const fileDataPromise = new Promise((resolve) => {\n",
              "      const reader = new FileReader();\n",
              "      reader.onload = (e) => {\n",
              "        resolve(e.target.result);\n",
              "      };\n",
              "      reader.readAsArrayBuffer(file);\n",
              "    });\n",
              "    // Wait for the data to be ready.\n",
              "    let fileData = yield {\n",
              "      promise: fileDataPromise,\n",
              "      response: {\n",
              "        action: 'continue',\n",
              "      }\n",
              "    };\n",
              "\n",
              "    // Use a chunked sending to avoid message size limits. See b/62115660.\n",
              "    let position = 0;\n",
              "    do {\n",
              "      const length = Math.min(fileData.byteLength - position, MAX_PAYLOAD_SIZE);\n",
              "      const chunk = new Uint8Array(fileData, position, length);\n",
              "      position += length;\n",
              "\n",
              "      const base64 = btoa(String.fromCharCode.apply(null, chunk));\n",
              "      yield {\n",
              "        response: {\n",
              "          action: 'append',\n",
              "          file: file.name,\n",
              "          data: base64,\n",
              "        },\n",
              "      };\n",
              "\n",
              "      let percentDone = fileData.byteLength === 0 ?\n",
              "          100 :\n",
              "          Math.round((position / fileData.byteLength) * 100);\n",
              "      percent.textContent = `${percentDone}% done`;\n",
              "\n",
              "    } while (position < fileData.byteLength);\n",
              "  }\n",
              "\n",
              "  // All done.\n",
              "  yield {\n",
              "    response: {\n",
              "      action: 'complete',\n",
              "    }\n",
              "  };\n",
              "}\n",
              "\n",
              "scope.google = scope.google || {};\n",
              "scope.google.colab = scope.google.colab || {};\n",
              "scope.google.colab._files = {\n",
              "  _uploadFiles,\n",
              "  _uploadFilesContinue,\n",
              "};\n",
              "})(self);\n",
              "</script> "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Saving 001.jpg to 001.jpg\n",
            "Saving 002.jpg to 002.jpg\n",
            "Saving 003.jpg to 003.jpg\n",
            "Saving 004.jpg to 004.jpg\n",
            "Cargar imágenes de VALIDATION clase ADONIS en /content/butterflies_classifier/validation/adonis\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "     <input type=\"file\" id=\"files-a03fe48b-74c5-4cb2-8ce9-b84a88a39b19\" name=\"files[]\" multiple disabled\n",
              "        style=\"border:none\" />\n",
              "     <output id=\"result-a03fe48b-74c5-4cb2-8ce9-b84a88a39b19\">\n",
              "      Upload widget is only available when the cell has been executed in the\n",
              "      current browser session. Please rerun this cell to enable.\n",
              "      </output>\n",
              "      <script>// Copyright 2017 Google LLC\n",
              "//\n",
              "// Licensed under the Apache License, Version 2.0 (the \"License\");\n",
              "// you may not use this file except in compliance with the License.\n",
              "// You may obtain a copy of the License at\n",
              "//\n",
              "//      http://www.apache.org/licenses/LICENSE-2.0\n",
              "//\n",
              "// Unless required by applicable law or agreed to in writing, software\n",
              "// distributed under the License is distributed on an \"AS IS\" BASIS,\n",
              "// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
              "// See the License for the specific language governing permissions and\n",
              "// limitations under the License.\n",
              "\n",
              "/**\n",
              " * @fileoverview Helpers for google.colab Python module.\n",
              " */\n",
              "(function(scope) {\n",
              "function span(text, styleAttributes = {}) {\n",
              "  const element = document.createElement('span');\n",
              "  element.textContent = text;\n",
              "  for (const key of Object.keys(styleAttributes)) {\n",
              "    element.style[key] = styleAttributes[key];\n",
              "  }\n",
              "  return element;\n",
              "}\n",
              "\n",
              "// Max number of bytes which will be uploaded at a time.\n",
              "const MAX_PAYLOAD_SIZE = 100 * 1024;\n",
              "\n",
              "function _uploadFiles(inputId, outputId) {\n",
              "  const steps = uploadFilesStep(inputId, outputId);\n",
              "  const outputElement = document.getElementById(outputId);\n",
              "  // Cache steps on the outputElement to make it available for the next call\n",
              "  // to uploadFilesContinue from Python.\n",
              "  outputElement.steps = steps;\n",
              "\n",
              "  return _uploadFilesContinue(outputId);\n",
              "}\n",
              "\n",
              "// This is roughly an async generator (not supported in the browser yet),\n",
              "// where there are multiple asynchronous steps and the Python side is going\n",
              "// to poll for completion of each step.\n",
              "// This uses a Promise to block the python side on completion of each step,\n",
              "// then passes the result of the previous step as the input to the next step.\n",
              "function _uploadFilesContinue(outputId) {\n",
              "  const outputElement = document.getElementById(outputId);\n",
              "  const steps = outputElement.steps;\n",
              "\n",
              "  const next = steps.next(outputElement.lastPromiseValue);\n",
              "  return Promise.resolve(next.value.promise).then((value) => {\n",
              "    // Cache the last promise value to make it available to the next\n",
              "    // step of the generator.\n",
              "    outputElement.lastPromiseValue = value;\n",
              "    return next.value.response;\n",
              "  });\n",
              "}\n",
              "\n",
              "/**\n",
              " * Generator function which is called between each async step of the upload\n",
              " * process.\n",
              " * @param {string} inputId Element ID of the input file picker element.\n",
              " * @param {string} outputId Element ID of the output display.\n",
              " * @return {!Iterable<!Object>} Iterable of next steps.\n",
              " */\n",
              "function* uploadFilesStep(inputId, outputId) {\n",
              "  const inputElement = document.getElementById(inputId);\n",
              "  inputElement.disabled = false;\n",
              "\n",
              "  const outputElement = document.getElementById(outputId);\n",
              "  outputElement.innerHTML = '';\n",
              "\n",
              "  const pickedPromise = new Promise((resolve) => {\n",
              "    inputElement.addEventListener('change', (e) => {\n",
              "      resolve(e.target.files);\n",
              "    });\n",
              "  });\n",
              "\n",
              "  const cancel = document.createElement('button');\n",
              "  inputElement.parentElement.appendChild(cancel);\n",
              "  cancel.textContent = 'Cancel upload';\n",
              "  const cancelPromise = new Promise((resolve) => {\n",
              "    cancel.onclick = () => {\n",
              "      resolve(null);\n",
              "    };\n",
              "  });\n",
              "\n",
              "  // Wait for the user to pick the files.\n",
              "  const files = yield {\n",
              "    promise: Promise.race([pickedPromise, cancelPromise]),\n",
              "    response: {\n",
              "      action: 'starting',\n",
              "    }\n",
              "  };\n",
              "\n",
              "  cancel.remove();\n",
              "\n",
              "  // Disable the input element since further picks are not allowed.\n",
              "  inputElement.disabled = true;\n",
              "\n",
              "  if (!files) {\n",
              "    return {\n",
              "      response: {\n",
              "        action: 'complete',\n",
              "      }\n",
              "    };\n",
              "  }\n",
              "\n",
              "  for (const file of files) {\n",
              "    const li = document.createElement('li');\n",
              "    li.append(span(file.name, {fontWeight: 'bold'}));\n",
              "    li.append(span(\n",
              "        `(${file.type || 'n/a'}) - ${file.size} bytes, ` +\n",
              "        `last modified: ${\n",
              "            file.lastModifiedDate ? file.lastModifiedDate.toLocaleDateString() :\n",
              "                                    'n/a'} - `));\n",
              "    const percent = span('0% done');\n",
              "    li.appendChild(percent);\n",
              "\n",
              "    outputElement.appendChild(li);\n",
              "\n",
              "    const fileDataPromise = new Promise((resolve) => {\n",
              "      const reader = new FileReader();\n",
              "      reader.onload = (e) => {\n",
              "        resolve(e.target.result);\n",
              "      };\n",
              "      reader.readAsArrayBuffer(file);\n",
              "    });\n",
              "    // Wait for the data to be ready.\n",
              "    let fileData = yield {\n",
              "      promise: fileDataPromise,\n",
              "      response: {\n",
              "        action: 'continue',\n",
              "      }\n",
              "    };\n",
              "\n",
              "    // Use a chunked sending to avoid message size limits. See b/62115660.\n",
              "    let position = 0;\n",
              "    do {\n",
              "      const length = Math.min(fileData.byteLength - position, MAX_PAYLOAD_SIZE);\n",
              "      const chunk = new Uint8Array(fileData, position, length);\n",
              "      position += length;\n",
              "\n",
              "      const base64 = btoa(String.fromCharCode.apply(null, chunk));\n",
              "      yield {\n",
              "        response: {\n",
              "          action: 'append',\n",
              "          file: file.name,\n",
              "          data: base64,\n",
              "        },\n",
              "      };\n",
              "\n",
              "      let percentDone = fileData.byteLength === 0 ?\n",
              "          100 :\n",
              "          Math.round((position / fileData.byteLength) * 100);\n",
              "      percent.textContent = `${percentDone}% done`;\n",
              "\n",
              "    } while (position < fileData.byteLength);\n",
              "  }\n",
              "\n",
              "  // All done.\n",
              "  yield {\n",
              "    response: {\n",
              "      action: 'complete',\n",
              "    }\n",
              "  };\n",
              "}\n",
              "\n",
              "scope.google = scope.google || {};\n",
              "scope.google.colab = scope.google.colab || {};\n",
              "scope.google.colab._files = {\n",
              "  _uploadFiles,\n",
              "  _uploadFilesContinue,\n",
              "};\n",
              "})(self);\n",
              "</script> "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Saving 1.jpg to 1.jpg\n",
            "Saving 2.jpg to 2.jpg\n",
            "Cargar imágenes de VALIDATION clase APPOLLO en /content/butterflies_classifier/validation/appollo\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "     <input type=\"file\" id=\"files-1c85bfc6-7e81-4bab-87b3-06ac9ddd3027\" name=\"files[]\" multiple disabled\n",
              "        style=\"border:none\" />\n",
              "     <output id=\"result-1c85bfc6-7e81-4bab-87b3-06ac9ddd3027\">\n",
              "      Upload widget is only available when the cell has been executed in the\n",
              "      current browser session. Please rerun this cell to enable.\n",
              "      </output>\n",
              "      <script>// Copyright 2017 Google LLC\n",
              "//\n",
              "// Licensed under the Apache License, Version 2.0 (the \"License\");\n",
              "// you may not use this file except in compliance with the License.\n",
              "// You may obtain a copy of the License at\n",
              "//\n",
              "//      http://www.apache.org/licenses/LICENSE-2.0\n",
              "//\n",
              "// Unless required by applicable law or agreed to in writing, software\n",
              "// distributed under the License is distributed on an \"AS IS\" BASIS,\n",
              "// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
              "// See the License for the specific language governing permissions and\n",
              "// limitations under the License.\n",
              "\n",
              "/**\n",
              " * @fileoverview Helpers for google.colab Python module.\n",
              " */\n",
              "(function(scope) {\n",
              "function span(text, styleAttributes = {}) {\n",
              "  const element = document.createElement('span');\n",
              "  element.textContent = text;\n",
              "  for (const key of Object.keys(styleAttributes)) {\n",
              "    element.style[key] = styleAttributes[key];\n",
              "  }\n",
              "  return element;\n",
              "}\n",
              "\n",
              "// Max number of bytes which will be uploaded at a time.\n",
              "const MAX_PAYLOAD_SIZE = 100 * 1024;\n",
              "\n",
              "function _uploadFiles(inputId, outputId) {\n",
              "  const steps = uploadFilesStep(inputId, outputId);\n",
              "  const outputElement = document.getElementById(outputId);\n",
              "  // Cache steps on the outputElement to make it available for the next call\n",
              "  // to uploadFilesContinue from Python.\n",
              "  outputElement.steps = steps;\n",
              "\n",
              "  return _uploadFilesContinue(outputId);\n",
              "}\n",
              "\n",
              "// This is roughly an async generator (not supported in the browser yet),\n",
              "// where there are multiple asynchronous steps and the Python side is going\n",
              "// to poll for completion of each step.\n",
              "// This uses a Promise to block the python side on completion of each step,\n",
              "// then passes the result of the previous step as the input to the next step.\n",
              "function _uploadFilesContinue(outputId) {\n",
              "  const outputElement = document.getElementById(outputId);\n",
              "  const steps = outputElement.steps;\n",
              "\n",
              "  const next = steps.next(outputElement.lastPromiseValue);\n",
              "  return Promise.resolve(next.value.promise).then((value) => {\n",
              "    // Cache the last promise value to make it available to the next\n",
              "    // step of the generator.\n",
              "    outputElement.lastPromiseValue = value;\n",
              "    return next.value.response;\n",
              "  });\n",
              "}\n",
              "\n",
              "/**\n",
              " * Generator function which is called between each async step of the upload\n",
              " * process.\n",
              " * @param {string} inputId Element ID of the input file picker element.\n",
              " * @param {string} outputId Element ID of the output display.\n",
              " * @return {!Iterable<!Object>} Iterable of next steps.\n",
              " */\n",
              "function* uploadFilesStep(inputId, outputId) {\n",
              "  const inputElement = document.getElementById(inputId);\n",
              "  inputElement.disabled = false;\n",
              "\n",
              "  const outputElement = document.getElementById(outputId);\n",
              "  outputElement.innerHTML = '';\n",
              "\n",
              "  const pickedPromise = new Promise((resolve) => {\n",
              "    inputElement.addEventListener('change', (e) => {\n",
              "      resolve(e.target.files);\n",
              "    });\n",
              "  });\n",
              "\n",
              "  const cancel = document.createElement('button');\n",
              "  inputElement.parentElement.appendChild(cancel);\n",
              "  cancel.textContent = 'Cancel upload';\n",
              "  const cancelPromise = new Promise((resolve) => {\n",
              "    cancel.onclick = () => {\n",
              "      resolve(null);\n",
              "    };\n",
              "  });\n",
              "\n",
              "  // Wait for the user to pick the files.\n",
              "  const files = yield {\n",
              "    promise: Promise.race([pickedPromise, cancelPromise]),\n",
              "    response: {\n",
              "      action: 'starting',\n",
              "    }\n",
              "  };\n",
              "\n",
              "  cancel.remove();\n",
              "\n",
              "  // Disable the input element since further picks are not allowed.\n",
              "  inputElement.disabled = true;\n",
              "\n",
              "  if (!files) {\n",
              "    return {\n",
              "      response: {\n",
              "        action: 'complete',\n",
              "      }\n",
              "    };\n",
              "  }\n",
              "\n",
              "  for (const file of files) {\n",
              "    const li = document.createElement('li');\n",
              "    li.append(span(file.name, {fontWeight: 'bold'}));\n",
              "    li.append(span(\n",
              "        `(${file.type || 'n/a'}) - ${file.size} bytes, ` +\n",
              "        `last modified: ${\n",
              "            file.lastModifiedDate ? file.lastModifiedDate.toLocaleDateString() :\n",
              "                                    'n/a'} - `));\n",
              "    const percent = span('0% done');\n",
              "    li.appendChild(percent);\n",
              "\n",
              "    outputElement.appendChild(li);\n",
              "\n",
              "    const fileDataPromise = new Promise((resolve) => {\n",
              "      const reader = new FileReader();\n",
              "      reader.onload = (e) => {\n",
              "        resolve(e.target.result);\n",
              "      };\n",
              "      reader.readAsArrayBuffer(file);\n",
              "    });\n",
              "    // Wait for the data to be ready.\n",
              "    let fileData = yield {\n",
              "      promise: fileDataPromise,\n",
              "      response: {\n",
              "        action: 'continue',\n",
              "      }\n",
              "    };\n",
              "\n",
              "    // Use a chunked sending to avoid message size limits. See b/62115660.\n",
              "    let position = 0;\n",
              "    do {\n",
              "      const length = Math.min(fileData.byteLength - position, MAX_PAYLOAD_SIZE);\n",
              "      const chunk = new Uint8Array(fileData, position, length);\n",
              "      position += length;\n",
              "\n",
              "      const base64 = btoa(String.fromCharCode.apply(null, chunk));\n",
              "      yield {\n",
              "        response: {\n",
              "          action: 'append',\n",
              "          file: file.name,\n",
              "          data: base64,\n",
              "        },\n",
              "      };\n",
              "\n",
              "      let percentDone = fileData.byteLength === 0 ?\n",
              "          100 :\n",
              "          Math.round((position / fileData.byteLength) * 100);\n",
              "      percent.textContent = `${percentDone}% done`;\n",
              "\n",
              "    } while (position < fileData.byteLength);\n",
              "  }\n",
              "\n",
              "  // All done.\n",
              "  yield {\n",
              "    response: {\n",
              "      action: 'complete',\n",
              "    }\n",
              "  };\n",
              "}\n",
              "\n",
              "scope.google = scope.google || {};\n",
              "scope.google.colab = scope.google.colab || {};\n",
              "scope.google.colab._files = {\n",
              "  _uploadFiles,\n",
              "  _uploadFilesContinue,\n",
              "};\n",
              "})(self);\n",
              "</script> "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Saving 1.jpg to 1.jpg\n",
            "Saving 2.jpg to 2.jpg\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Transformaciones de Imágenes**\n",
        "\n",
        "- `transforms.Resize((255, 255))`: redimensiona la imagen a un tamaño de píxel de 255x255. Esto se hace para asegurarse de que todas las imágenes tengan el mismo tamaño antes de aplicar las transformaciones adicionales.\n",
        "\n",
        "- `transforms.RandomCrop((224, 224))`: corta aleatoriamente la imagen en un rango de 224x224. Esto ayuda a introducir variabilidad en la ubicación de las imágenes.\n",
        "\n",
        "La transformación RandomCrop se utiliza comúnmente como una técnica de aumento de datos para introducir variabilidad en las imágenes de entrenamiento. Al recortar aleatoriamente una región de la imagen, se pueden capturar diferentes partes o características de la imagen, lo que ayuda a mejorar la capacidad de generalización del modelo y lo hace más robusto a la variación en la ubicación de los objetos en las imágenes.\n",
        "\n",
        "Es importante tener en cuenta que al aplicar esta transformación, la imagen se recorta de forma aleatoria y se pierde información fuera de la región recortada. Por lo tanto, es recomendable tener suficiente margen de espacio en las imágenes de entrada para asegurar que las regiones de interés no se corten por completo.\n",
        "\n",
        "- `transforms.RandomRotation(30)`: Esta transformación aplica una rotación aleatoria a la imagen de entrada en un rango de ±30 grados. Esto significa que la imagen puede rotarse en cualquier dirección dentro de ese rango.\n",
        "\n",
        "- `transforms.RandomHorizontalFlip(p=0.5)`: voltea aleatoriamente horizontalmente la imagen con una probabilidad de 0.5. Esto ayuda a introducir variabilidad en la orientación horizontal de las imágenes.\n",
        "\n",
        "- `transforms.RandomVerticalFlip(p=0.5)`: voltea aleatoriamente verticalmente la imagen con una probabilidad de 0.5. Esto ayuda a introducir variabilidad en la orientación vertical de las imágenes.\n",
        "\n",
        "Cuando se establece p=0.5, la transformación de volteo vertical se aplicará con una probabilidad del 50%. Esto significa que para cada imagen, hay un 50% de probabilidad de que se aplique la transformación y un 50% de probabilidad de que no se aplique.\n",
        "Si se desea aplicar la transformación de volteo vertical a todas las imágenes, se puede establecer p=1.0. Por otro lado, si se desea evitar completamente la transformación, se puede establecer p=0.0.\n",
        "\n",
        "- `transforms.GaussianBlur(kernel_size=(3, 7), sigma=(0.2, 3))`: aplica un desenfoque gaussiano aleatorio a la imagen. Con parámetros tamaño del kernel y desviación estándar, que se eligen aleatoriamente dentro de los rangos especificados. Esto puede ayudar a suavizar y difuminar la imagen.\n",
        "\n",
        "- `transforms.RandomRotation(degrees=(30, 70))`: rota aleatoriamente la imagen en un rango de grados especificado. Esto ayuda a introducir variabilidad en la orientación y la posición de las imágenes.\n",
        "\n",
        "- `transforms.Normalize(mean=[0.5, 0.5, 0.5], std=[0.5, 0.5, 0.5])`: normaliza los valores de los píxeles de la imagen. Resta la media especificada y divide por la desviación estándar especificada en cada canal de color. La normalización ayuda a que los datos tengan una escala adecuada para el procesamiento del modelo. Esta normalización escala los valores de píxeles para que queden dentro del rango [-1, 1].\n"
      ],
      "metadata": {
        "id": "AAzTF-gkOGOv"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "BATCH_SIZE = 2\n",
        "\n",
        "# Configuración para la transformación de los datos\n",
        "train_transform = transforms.Compose([\n",
        "    transforms.Resize((255, 255)),\n",
        "    # Ejemplos de data augmentation\n",
        "    # transforms.RandomHorizontalFlip(p=0.5),\n",
        "    # transforms.RandomVerticalFlip(p=0.5),\n",
        "    # transforms.GaussianBlur(kernel_size=(5, 9), sigma=(0.1, 5)),\n",
        "    # transforms.RandomRotation(degrees=(30, 70)),\n",
        "    transforms.ToTensor(),\n",
        "    transforms.Normalize(\n",
        "        mean=[0.5, 0.5, 0.5],\n",
        "        std=[0.5, 0.5, 0.5]\n",
        "    )\n",
        "])\n",
        "\n",
        "# Cargar dataset de entrenamiento\n",
        "train_dataset = datasets.ImageFolder(\n",
        "                          root='/content/butterflies_classifier/train',\n",
        "                          transform=train_transform\n",
        ")\n",
        "\n",
        "# Data loaders de entramiento\n",
        "train_loader = DataLoader(\n",
        "                    train_dataset, batch_size=BATCH_SIZE,\n",
        "                    shuffle=False,\n",
        "                    num_workers=2,\n",
        "                    # Indicar si los datos se deben fijar en la memoria.\n",
        "                    # Esto puede acelerar la transferencia de datos entre la CPU y la GPU,\n",
        "                    # si estás utilizando una GPU para el entrenamiento\n",
        "                    pin_memory=False\n",
        ")\n",
        "\n",
        "# Tranformar dataset de validación\n",
        "valid_transform = transforms.Compose([\n",
        "  # TODO\n",
        "])\n",
        "\n",
        "# Cargar dataset de validación\n",
        "# valid_dataset = datasets.ImageFolder(\n",
        "#                           root='/content/butterflies_classifier/valid',\n",
        "#                           transform=valid_transform\n",
        "# )\n",
        "\n",
        "# Data loaders de validación\n",
        "# valid_loader = DataLoader(\n",
        "#                     valid_dataset, batch_size=BATCH_SIZE, shuffle=False,\n",
        "#                     num_workers=2, pin_memory=False\n",
        "# )"
      ],
      "metadata": {
        "id": "xRzXFoCTvw7Y"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Clases del dataset\n",
        "train_dataset.classes"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "c-gUpumIgSm9",
        "outputId": "d666a0c0-58ba-43f8-a59e-dd07b0fbef21"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['adonis', 'appollo']"
            ]
          },
          "metadata": {},
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Iterar sobre el cargador de batch de datos de entrenamiento\n",
        "mini_batchs = enumerate(train_loader)"
      ],
      "metadata": {
        "id": "A2QQnt2mX44p"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# 1ra iteración sobre el cargador de batch de datos de entrenamiento\n",
        "mini_batch_idx, (mini_batch_inputs, mini_batch_targets) = next(mini_batchs)\n",
        "print(f'N° de mini-batch: {mini_batch_idx}')\n",
        "print(f'Dimensiones de mini-batch de entradas: {mini_batch_inputs.shape}')\n",
        "print(f'Dimensiones de mini-batch de targets: {mini_batch_targets.shape}')\n",
        "print(f'Datos de entrada del mini-batch:\\n\\t {mini_batch_inputs[:2]}')\n",
        "print(f'Datos de target del mini-batch:\\n\\t {mini_batch_targets}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cGA8xS00XwNJ",
        "outputId": "24bee9df-b9f5-4a5c-909d-38db5caa76ec"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "N° de mini-batch: 0\n",
            "Dimensiones de mini-batch de entradas: torch.Size([2, 3, 255, 255])\n",
            "Dimensiones de mini-batch de targets: torch.Size([2])\n",
            "Datos de entrada del mini-batch:\n",
            "\t tensor([[[[-1.0000, -1.0000, -0.9686,  ..., -1.0000, -1.0000, -1.0000],\n",
            "          [-1.0000, -0.9843, -0.9451,  ..., -1.0000, -1.0000, -1.0000],\n",
            "          [-1.0000, -0.9765, -0.9216,  ..., -1.0000, -1.0000, -1.0000],\n",
            "          ...,\n",
            "          [-0.3176, -0.3490, -0.4353,  ..., -0.9922, -0.9922, -0.9843],\n",
            "          [-0.3098, -0.3569, -0.4431,  ..., -1.0000, -1.0000, -0.9922],\n",
            "          [-0.3176, -0.3725, -0.4510,  ..., -0.9922, -1.0000, -1.0000]],\n",
            "\n",
            "         [[-0.9451, -0.8980, -0.8353,  ..., -0.9686, -0.9686, -0.9608],\n",
            "          [-0.9294, -0.8824, -0.8118,  ..., -0.9765, -0.9686, -0.9608],\n",
            "          [-0.9137, -0.8510, -0.7804,  ..., -0.9765, -0.9686, -0.9608],\n",
            "          ...,\n",
            "          [-0.1216, -0.1529, -0.2471,  ..., -0.9529, -0.9451, -0.9373],\n",
            "          [-0.1294, -0.1765, -0.2706,  ..., -0.9608, -0.9529, -0.9451],\n",
            "          [-0.1373, -0.1922, -0.2863,  ..., -0.9608, -0.9529, -0.9529]],\n",
            "\n",
            "         [[-0.9922, -0.9922, -1.0000,  ..., -1.0000, -1.0000, -0.9922],\n",
            "          [-0.9922, -0.9765, -0.9843,  ..., -1.0000, -1.0000, -0.9922],\n",
            "          [-0.9843, -0.9686, -0.9686,  ..., -1.0000, -1.0000, -0.9922],\n",
            "          ...,\n",
            "          [-0.4902, -0.5059, -0.5686,  ..., -1.0000, -0.9922, -0.9843],\n",
            "          [-0.4902, -0.5294, -0.5843,  ..., -0.9922, -0.9843, -0.9765],\n",
            "          [-0.4980, -0.5451, -0.6000,  ..., -0.9765, -0.9843, -0.9843]]],\n",
            "\n",
            "\n",
            "        [[[-0.3333, -0.3647, -0.3412,  ..., -0.2784, -0.2627, -0.2706],\n",
            "          [-0.3176, -0.2784, -0.2784,  ..., -0.2627, -0.2706, -0.2784],\n",
            "          [-0.2000, -0.2392, -0.2784,  ..., -0.2706, -0.2706, -0.2706],\n",
            "          ...,\n",
            "          [-0.0745, -0.0902, -0.0745,  ..., -0.2314, -0.2471, -0.1529],\n",
            "          [-0.1137, -0.1216, -0.0745,  ..., -0.2549, -0.2549, -0.2000],\n",
            "          [-0.0980, -0.1137, -0.0980,  ..., -0.2392, -0.2392, -0.2000]],\n",
            "\n",
            "         [[-0.4039, -0.4353, -0.4196,  ...,  0.0275,  0.0353,  0.0275],\n",
            "          [-0.3882, -0.3490, -0.3490,  ...,  0.0353,  0.0275,  0.0196],\n",
            "          [-0.2784, -0.3176, -0.3647,  ...,  0.0196,  0.0196,  0.0118],\n",
            "          ...,\n",
            "          [ 0.1373,  0.1216,  0.1373,  ..., -0.0118, -0.0431,  0.0431],\n",
            "          [ 0.0980,  0.0902,  0.1373,  ..., -0.0353, -0.0510, -0.0039],\n",
            "          [ 0.1137,  0.0980,  0.1137,  ..., -0.0196, -0.0353, -0.0039]],\n",
            "\n",
            "         [[-0.7412, -0.7569, -0.7098,  ..., -0.8667, -0.8353, -0.8431],\n",
            "          [-0.7255, -0.6784, -0.6627,  ..., -0.8588, -0.8510, -0.8510],\n",
            "          [-0.6235, -0.6706, -0.6941,  ..., -0.8667, -0.8667, -0.8510],\n",
            "          ...,\n",
            "          [-0.9686, -0.9765, -0.9686,  ..., -0.5843, -0.5922, -0.4902],\n",
            "          [-1.0000, -1.0000, -0.9765,  ..., -0.6078, -0.6000, -0.5373],\n",
            "          [-1.0000, -1.0000, -1.0000,  ..., -0.5922, -0.5922, -0.5373]]]])\n",
            "Datos de target del mini-batch:\n",
            "\t tensor([0, 0])\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Visualicemos la imágenes del batch 0: 1ra y 2da imagen\n",
        "fig = plt.figure(2, figsize=(6, 2))\n",
        "fig.add_subplot(1, 2, 1)\n",
        "plt.imshow(mini_batch_inputs[0].T)  # .T transpone los vectores [C x H x W] a [H x W x C]\n",
        "fig.add_subplot(1, 2, 2)\n",
        "plt.imshow(mini_batch_inputs[1].T)\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "YRMkRxJaytj4",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 254
        },
        "outputId": "ea330d0f-e2b3-48e5-b29e-320125a114b6"
      },
      "execution_count": null,
      "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",
            "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": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x200 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# 2da iteración sobre el cargador de batch de datos de entrenamiento\n",
        "mini_batch_idx, (mini_batch_inputs, mini_batch_targets) = next(mini_batchs)\n",
        "print(f'N° de mini-batch: {mini_batch_idx}')\n",
        "print(f'Dimensiones de mini-batch de entradas: {mini_batch_inputs.shape}')\n",
        "print(f'Dimensiones de mini-batch de targets: {mini_batch_targets.shape}')\n",
        "print(f'Datos de entrada del mini-batch:\\n\\t {mini_batch_inputs[:2]}')\n",
        "print(f'Datos de target del mini-batch:\\n\\t {mini_batch_targets}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "E5WZAnIFZ_G4",
        "outputId": "19b1bbbb-9e0b-42fd-f9e5-af1526a53813"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "N° de mini-batch: 1\n",
            "Dimensiones de mini-batch de entradas: torch.Size([2, 3, 255, 255])\n",
            "Dimensiones de mini-batch de targets: torch.Size([2])\n",
            "Datos de entrada del mini-batch:\n",
            "\t tensor([[[[ 0.0196,  0.0196,  0.0353,  ...,  0.7020,  0.7020,  0.6863],\n",
            "          [ 0.0431,  0.0353,  0.0275,  ...,  0.6078,  0.5765,  0.5608],\n",
            "          [ 0.0588,  0.0275,  0.0196,  ...,  0.4667,  0.4353,  0.4039],\n",
            "          ...,\n",
            "          [-0.4353, -0.3961, -0.3255,  ..., -0.2784, -0.2627, -0.2706],\n",
            "          [-0.4196, -0.3647, -0.2863,  ..., -0.1608, -0.1294, -0.1294],\n",
            "          [-0.3882, -0.3333, -0.2549,  ..., -0.0353, -0.0039, -0.0039]],\n",
            "\n",
            "         [[-0.0353, -0.0353, -0.0118,  ...,  0.6314,  0.6314,  0.6157],\n",
            "          [-0.0118, -0.0196, -0.0196,  ...,  0.5451,  0.5294,  0.5137],\n",
            "          [ 0.0039, -0.0196, -0.0275,  ...,  0.4275,  0.3961,  0.3725],\n",
            "          ...,\n",
            "          [-0.3176, -0.2706, -0.2157,  ..., -0.1137, -0.1137, -0.1216],\n",
            "          [-0.2863, -0.2392, -0.1686,  ..., -0.0118,  0.0039,  0.0039],\n",
            "          [-0.2549, -0.2000, -0.1294,  ...,  0.0980,  0.1216,  0.1216]],\n",
            "\n",
            "         [[-0.3961, -0.3961, -0.3882,  ..., -0.0902, -0.1059, -0.1294],\n",
            "          [-0.3725, -0.3882, -0.3961,  ..., -0.1686, -0.1922, -0.2235],\n",
            "          [-0.3647, -0.3961, -0.4039,  ..., -0.2627, -0.2941, -0.3412],\n",
            "          ...,\n",
            "          [-1.0000, -0.9922, -0.9765,  ..., -0.9843, -0.9843, -0.9843],\n",
            "          [-1.0000, -0.9843, -0.9608,  ..., -0.9216, -0.9216, -0.9373],\n",
            "          [-0.9922, -0.9686, -0.9373,  ..., -0.8353, -0.8353, -0.8510]]],\n",
            "\n",
            "\n",
            "        [[[-1.0000, -0.9765, -0.9922,  ..., -0.9686, -0.9608, -0.9529],\n",
            "          [-0.9922, -0.9686, -0.9765,  ..., -0.9765, -0.9765, -0.9686],\n",
            "          [-1.0000, -0.9922, -0.9922,  ..., -0.9686, -0.9686, -0.9686],\n",
            "          ...,\n",
            "          [-1.0000, -1.0000, -1.0000,  ..., -0.9922, -0.9922, -0.9922],\n",
            "          [-1.0000, -1.0000, -1.0000,  ..., -0.9922, -0.9922, -0.9922],\n",
            "          [-1.0000, -1.0000, -1.0000,  ..., -0.9922, -0.9843, -0.9843]],\n",
            "\n",
            "         [[-1.0000, -0.9765, -0.9922,  ..., -0.9686, -0.9608, -0.9529],\n",
            "          [-0.9922, -0.9686, -0.9765,  ..., -0.9843, -0.9765, -0.9686],\n",
            "          [-1.0000, -0.9922, -0.9922,  ..., -0.9765, -0.9686, -0.9686],\n",
            "          ...,\n",
            "          [-1.0000, -1.0000, -1.0000,  ..., -0.9922, -0.9922, -0.9922],\n",
            "          [-1.0000, -1.0000, -1.0000,  ..., -0.9922, -0.9922, -0.9922],\n",
            "          [-1.0000, -1.0000, -1.0000,  ..., -0.9922, -0.9843, -0.9843]],\n",
            "\n",
            "         [[-1.0000, -0.9765, -0.9922,  ..., -0.9373, -0.9608, -0.9686],\n",
            "          [-0.9922, -0.9686, -0.9765,  ..., -0.9529, -0.9765, -0.9843],\n",
            "          [-1.0000, -0.9922, -0.9922,  ..., -0.9451, -0.9686, -0.9843],\n",
            "          ...,\n",
            "          [-1.0000, -1.0000, -1.0000,  ..., -0.9765, -0.9765, -0.9765],\n",
            "          [-1.0000, -1.0000, -1.0000,  ..., -0.9765, -0.9765, -0.9765],\n",
            "          [-1.0000, -1.0000, -1.0000,  ..., -0.9765, -0.9686, -0.9686]]]])\n",
            "Datos de target del mini-batch:\n",
            "\t tensor([0, 1])\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Visualicemos la imágenes del 2do batch: 3ra y 4ta imagen\n",
        "fig = plt.figure(2, figsize=(6, 2))\n",
        "fig.add_subplot(1, 2, 1)\n",
        "plt.imshow(mini_batch_inputs[0].T)  # .T transpone los vectores [C x H x W] a [H x W x C]\n",
        "fig.add_subplot(1, 2, 2)\n",
        "plt.imshow(mini_batch_inputs[1].T)\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 254
        },
        "id": "tybOtcnmABo3",
        "outputId": "ab5a9f23-804c-45e3-a363-8111d65daa10"
      },
      "execution_count": null,
      "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",
            "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": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x200 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# 3ra iteración sobre el cargador de batch de datos de entrenamiento\n",
        "mini_batch_idx, (mini_batch_inputs, mini_batch_targets) = next(mini_batchs)\n",
        "print(f'N° de mini-batch: {mini_batch_idx}')\n",
        "print(f'Dimensiones de mini-batch de entradas: {mini_batch_inputs.shape}')\n",
        "print(f'Dimensiones de mini-batch de targets: {mini_batch_targets.shape}')\n",
        "print(f'Datos de entrada del mini-batch:\\n\\t {mini_batch_inputs[:2]}')\n",
        "print(f'Datos de target del mini-batch:\\n\\t {mini_batch_targets}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "osKWbHS-_LPQ",
        "outputId": "7dc59943-2f5d-446f-f3d3-9219934aaf92"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "N° de mini-batch: 2\n",
            "Dimensiones de mini-batch de entradas: torch.Size([2, 3, 255, 255])\n",
            "Dimensiones de mini-batch de targets: torch.Size([2])\n",
            "Datos de entrada del mini-batch:\n",
            "\t tensor([[[[ 0.8353,  0.7882,  0.7490,  ..., -0.0353, -0.0353, -0.0980],\n",
            "          [ 0.8039,  0.7569,  0.7176,  ..., -0.1059, -0.1137, -0.1137],\n",
            "          [ 0.7725,  0.7333,  0.7020,  ..., -0.0745, -0.1059, -0.0902],\n",
            "          ...,\n",
            "          [ 0.1686,  0.0588, -0.0745,  ..., -0.8039, -0.7725, -0.7020],\n",
            "          [ 0.1765,  0.0902, -0.0353,  ..., -0.7725, -0.7412, -0.7020],\n",
            "          [ 0.1765,  0.1059,  0.0039,  ..., -0.7412, -0.7098, -0.6863]],\n",
            "\n",
            "         [[ 0.8431,  0.7961,  0.7569,  ...,  0.0353,  0.0667,  0.0275],\n",
            "          [ 0.8118,  0.7647,  0.7255,  ..., -0.0431, -0.0118,  0.0118],\n",
            "          [ 0.7804,  0.7412,  0.7098,  ..., -0.0196, -0.0196,  0.0118],\n",
            "          ...,\n",
            "          [ 0.1608,  0.0431, -0.0902,  ..., -0.8431, -0.8353, -0.7725],\n",
            "          [ 0.1686,  0.0824, -0.0353,  ..., -0.8196, -0.8039, -0.7647],\n",
            "          [ 0.1686,  0.0980, -0.0039,  ..., -0.7804, -0.7647, -0.7490]],\n",
            "\n",
            "         [[ 0.8588,  0.8118,  0.7725,  ..., -0.6314, -0.6078, -0.6627],\n",
            "          [ 0.8275,  0.7804,  0.7412,  ..., -0.7020, -0.6706, -0.6627],\n",
            "          [ 0.7961,  0.7569,  0.7255,  ..., -0.6706, -0.6627, -0.6235],\n",
            "          ...,\n",
            "          [ 0.2000,  0.0667, -0.0667,  ..., -0.9451, -0.9373, -0.8824],\n",
            "          [ 0.2157,  0.1216, -0.0118,  ..., -0.9137, -0.9059, -0.8667],\n",
            "          [ 0.2157,  0.1373,  0.0353,  ..., -0.8745, -0.8588, -0.8510]]],\n",
            "\n",
            "\n",
            "        [[[-0.7882, -0.7961, -0.7961,  ..., -0.8039, -0.8039, -0.8039],\n",
            "          [-0.7882, -0.7961, -0.8039,  ..., -0.8039, -0.8039, -0.8039],\n",
            "          [-0.7804, -0.7882, -0.7961,  ..., -0.8039, -0.8039, -0.8039],\n",
            "          ...,\n",
            "          [-0.7020, -0.8667, -0.9137,  ..., -0.6784, -0.6863, -0.6863],\n",
            "          [-0.7725, -0.8745, -0.9216,  ..., -0.6706, -0.6784, -0.6863],\n",
            "          [-0.8039, -0.8588, -0.9216,  ..., -0.6706, -0.6784, -0.6863]],\n",
            "\n",
            "         [[-0.6157, -0.6235, -0.6235,  ..., -0.7804, -0.7804, -0.7804],\n",
            "          [-0.6157, -0.6235, -0.6314,  ..., -0.7804, -0.7804, -0.7804],\n",
            "          [-0.6078, -0.6157, -0.6235,  ..., -0.7804, -0.7804, -0.7804],\n",
            "          ...,\n",
            "          [-0.7020, -0.8667, -0.9137,  ..., -0.6314, -0.6392, -0.6392],\n",
            "          [-0.7725, -0.8824, -0.9294,  ..., -0.6235, -0.6314, -0.6392],\n",
            "          [-0.8039, -0.8667, -0.9294,  ..., -0.6235, -0.6314, -0.6392]],\n",
            "\n",
            "         [[-0.8980, -0.9059, -0.9059,  ..., -0.9137, -0.9137, -0.9137],\n",
            "          [-0.8980, -0.9059, -0.9137,  ..., -0.9137, -0.9137, -0.9137],\n",
            "          [-0.9059, -0.9137, -0.9216,  ..., -0.9137, -0.9137, -0.9137],\n",
            "          ...,\n",
            "          [-0.7647, -0.9137, -0.9529,  ..., -0.9294, -0.9373, -0.9373],\n",
            "          [-0.8353, -0.9216, -0.9608,  ..., -0.9216, -0.9294, -0.9373],\n",
            "          [-0.8667, -0.9137, -0.9686,  ..., -0.9216, -0.9294, -0.9373]]]])\n",
            "Datos de target del mini-batch:\n",
            "\t tensor([1, 1])\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Visualicemos la imágenes del 3er batch: 5ta y 6ta imagen\n",
        "fig = plt.figure(2, figsize=(6, 2))\n",
        "fig.add_subplot(1, 2, 1)\n",
        "plt.imshow(mini_batch_inputs[0].T)  # .T transpone los vectores [C x H x W] a [H x W x C]\n",
        "fig.add_subplot(1, 2, 2)\n",
        "plt.imshow(mini_batch_inputs[1].T)\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 254
        },
        "id": "E9Zb2JZUabLG",
        "outputId": "7426054b-3c2b-46f8-d467-4a68aeb8a122"
      },
      "execution_count": null,
      "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",
            "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": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x200 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Valores minimo y máximo de los datos normalizados\n",
        "mini_batch_inputs.min(), mini_batch_inputs.max()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "i7lRPnXzawqC",
        "outputId": "4e886c77-8391-4aaf-bbe4-16857d234d03"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(tensor(-1.), tensor(1.))"
            ]
          },
          "metadata": {},
          "execution_count": 19
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "_Nota_: La normalización empleando valores de media y desviación estandard de 0.5 es una forma sencilla y estandarizada de preprocesar datos de imágenes, pero puede que no sea óptima para todos los conjuntos de datos.\n"
      ],
      "metadata": {
        "id": "TVx_hFNRbYPH"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Datos No Estructurados - Texto\n",
        "---\n",
        "\n",
        "PPT: [P. III - Procesamiento de Lenguaje Natural (NLP)](https://docs.google.com/presentation/d/1hrN4z4DD-gto_nmlBHdZIaJkv7frOrTMiC_smj3Fdrg/edit#slide=id.g27cd7976a91_0_79)\n",
        "\n",
        "Notebook: [6. Procesamiento de Lenguaje Natural](https://colab.research.google.com/drive/1i6LjpdY9KWleJyySqSzr1qYONr3UqM7-?usp=drive_link])\n",
        "\n"
      ],
      "metadata": {
        "id": "Vd0YxwkMweZD"
      }
    }
  ]
}