image_classification_project_(CNN) (1).ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "in9nPVPg4gsn"
      },
      "source": [
        "#importing the library"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "82CsFlBR26WK"
      },
      "outputs": [],
      "source": [
        "import glob  # to return all file paths that match a specific pattern\n",
        "\n",
        "import tensorflow as tf\n",
        "import numpy as np\n",
        "from tensorflow import keras\n",
        "import cv2\n",
        "\n",
        "import os\n",
        "import matplotlib.pyplot as plt\n",
        "import pandas as pd\n",
        "from sklearn import preprocessing\n",
        "\n",
        "import keras\n",
        "from google.colab import files\n",
        "\n",
        "from sklearn.metrics import confusion_matrix, classification_report\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zY3jJFYVGRtS"
      },
      "source": [
        "#get the files of train and test"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "H4x6_S-15VWh",
        "outputId": "5ee7aa12-9ef4-4974-a755-bbd3e9507cb3"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Archive:  /content/drive/MyDrive/my projects/shai-level-trainingg.zip\n",
            "replace sample_submission.csv? [y]es, [n]o, [A]ll, [N]one, [r]ename: "
          ]
        }
      ],
      "source": [
        "!unzip '/content/drive/MyDrive/my projects/shai-level-trainingg.zip'"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "tZ7WeYtXBzh7"
      },
      "outputs": [],
      "source": [
        "train_dir = '/content/train'\n",
        "test_dir = '/content/test'"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UPoBIZL3HuHp"
      },
      "source": [
        "#pre-processing"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "l6KVCpIOFDiz"
      },
      "outputs": [],
      "source": [
        "image_size = 300\n",
        "Batch_size = 32\n",
        "\n",
        "train_datagen = tf.keras.preprocessing.image.ImageDataGenerator(\n",
        "    rescale = 1./255,\n",
        "    shear_range = 0.2,\n",
        "    zoom_range = 0.2,\n",
        "    horizontal_flip=True,\n",
        "    validation_split=0.1\n",
        ")\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9CVQLsePQADM"
      },
      "source": [
        "#training and validation set"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "erGBqcncKaGO",
        "outputId": "ff91c6a0-73a9-48a9-8b84-2fcadff84c17"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Found 1078 images belonging to 3 classes.\n",
            "Found 118 images belonging to 3 classes.\n"
          ]
        }
      ],
      "source": [
        "train_dataset = train_datagen.flow_from_directory(\n",
        "    train_dir,\n",
        "    target_size=(image_size, image_size),\n",
        "    batch_size=Batch_size,\n",
        "    subset='training'\n",
        ")\n",
        "\n",
        "val_dataset = train_datagen.flow_from_directory(\n",
        "    train_dir,\n",
        "    target_size=(image_size, image_size),\n",
        "    batch_size=Batch_size,\n",
        "    subset='validation'\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-Zpip9hzgGKv",
        "outputId": "9c313d27-4cfc-40a1-9209-cf83ca9d414a"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'covid': 0, 'normal': 1, 'virus': 2}"
            ]
          },
          "metadata": {},
          "execution_count": 113
        }
      ],
      "source": [
        "train_dataset.class_indices"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LGPgOpVpQFrl"
      },
      "source": [
        "#build the model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "8acYDqYLLxHU"
      },
      "outputs": [],
      "source": [
        "cnn = tf.keras.Sequential()\n",
        "\n",
        "cnn.add(tf.keras.layers.Conv2D(filters=64, padding='same', strides=2,kernel_size=3, activation='relu', input_shape=(300,300,3)))\n",
        "cnn.add(tf.keras.layers.MaxPool2D(pool_size=2, strides=2))\n",
        "\n",
        "\n",
        "cnn.add(tf.keras.layers.Conv2D(filters=32, padding='same', strides=2,kernel_size=3, activation='relu'))\n",
        "cnn.add(tf.keras.layers.MaxPool2D(pool_size=2, strides=2))\n",
        "\n",
        "# cnn.add(tf.keras.layers.Conv2D(filters=64, padding='same', strides=2,kernel_size=3, activation='relu'))\n",
        "# cnn.add(tf.keras.layers.MaxPool2D(pool_size=2, strides=2))\n",
        "\n",
        "# cnn.add(tf.keras.layers.Conv2D(filters=128, padding='same', strides=2,kernel_size=3, activation='relu'))\n",
        "# cnn.add(tf.keras.layers.MaxPool2D(pool_size=2, strides=2))\n",
        "\n",
        "cnn.add(tf.keras.layers.Flatten())\n",
        "cnn.add(tf.keras.layers.Dense(3, activation='softmax'))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "QEk-PTgML3QO"
      },
      "outputs": [],
      "source": [
        "cnn.compile(tf.keras.optimizers.Adam(clipnorm=1.0),\n",
        "            loss='categorical_crossentropy',\n",
        "            metrics=['accuracy'])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rkeJK1Z6wK56",
        "outputId": "df95321a-63e5-4b40-c3d5-3f5e33b298af"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_4\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " conv2d_8 (Conv2D)           (None, 175, 175, 64)      1792      \n",
            "                                                                 \n",
            " max_pooling2d_8 (MaxPooling  (None, 87, 87, 64)       0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " conv2d_9 (Conv2D)           (None, 44, 44, 32)        18464     \n",
            "                                                                 \n",
            " max_pooling2d_9 (MaxPooling  (None, 22, 22, 32)       0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " flatten_4 (Flatten)         (None, 15488)             0         \n",
            "                                                                 \n",
            " dense_4 (Dense)             (None, 3)                 46467     \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 66,723\n",
            "Trainable params: 66,723\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ],
      "source": [
        "cnn.summary()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7EGNf_nkASUZ"
      },
      "source": [
        "**Training**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "0ziuyD4e73on"
      },
      "outputs": [],
      "source": [
        "#Learning rate scheduling (by callback)\n",
        "#Instead of using a fixed learning rate, we will use a learning rate scheduler, which will change the learning rate after every batch of training (to enhance the model)\n",
        "def scheduler(epoch, lr):\n",
        "  if epoch < 10:\n",
        "    return lr\n",
        "  else:\n",
        "    return lr * tf.math.exp(-0.1)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "hmwHUlX07uKg"
      },
      "outputs": [],
      "source": [
        "callback = tf.keras.callbacks.LearningRateScheduler(scheduler)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4rGua4ZbO0Dm",
        "outputId": "45b1ad6b-4737-4956-8365-3955fe2a7056"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/20\n",
            "34/34 [==============================] - 85s 2s/step - loss: 0.8405 - accuracy: 0.6104 - val_loss: 0.8417 - val_accuracy: 0.6356 - lr: 0.0010\n",
            "Epoch 2/20\n",
            "34/34 [==============================] - 81s 2s/step - loss: 0.5397 - accuracy: 0.7774 - val_loss: 0.6343 - val_accuracy: 0.7203 - lr: 0.0010\n",
            "Epoch 3/20\n",
            "34/34 [==============================] - 81s 2s/step - loss: 0.4135 - accuracy: 0.8302 - val_loss: 0.4810 - val_accuracy: 0.8220 - lr: 0.0010\n",
            "Epoch 4/20\n",
            "34/34 [==============================] - 79s 2s/step - loss: 0.3735 - accuracy: 0.8571 - val_loss: 0.3610 - val_accuracy: 0.8729 - lr: 0.0010\n",
            "Epoch 5/20\n",
            "34/34 [==============================] - 83s 2s/step - loss: 0.3513 - accuracy: 0.8683 - val_loss: 0.4098 - val_accuracy: 0.8475 - lr: 0.0010\n",
            "Epoch 6/20\n",
            "34/34 [==============================] - 79s 2s/step - loss: 0.2926 - accuracy: 0.8961 - val_loss: 0.5542 - val_accuracy: 0.8220 - lr: 0.0010\n",
            "Epoch 7/20\n",
            "34/34 [==============================] - 81s 2s/step - loss: 0.2823 - accuracy: 0.8822 - val_loss: 0.3833 - val_accuracy: 0.8475 - lr: 0.0010\n",
            "Epoch 8/20\n",
            "34/34 [==============================] - 80s 2s/step - loss: 0.2942 - accuracy: 0.8850 - val_loss: 0.3835 - val_accuracy: 0.8475 - lr: 0.0010\n",
            "Epoch 9/20\n",
            "34/34 [==============================] - 79s 2s/step - loss: 0.3395 - accuracy: 0.8729 - val_loss: 0.3727 - val_accuracy: 0.8644 - lr: 0.0010\n",
            "Epoch 10/20\n",
            "34/34 [==============================] - 80s 2s/step - loss: 0.2867 - accuracy: 0.8952 - val_loss: 0.2904 - val_accuracy: 0.8983 - lr: 0.0010\n",
            "Epoch 11/20\n",
            "34/34 [==============================] - 81s 2s/step - loss: 0.2963 - accuracy: 0.8840 - val_loss: 0.3943 - val_accuracy: 0.8644 - lr: 9.0484e-04\n",
            "Epoch 12/20\n",
            "34/34 [==============================] - 79s 2s/step - loss: 0.2539 - accuracy: 0.9026 - val_loss: 0.4035 - val_accuracy: 0.8644 - lr: 8.1873e-04\n",
            "Epoch 13/20\n",
            "34/34 [==============================] - 81s 2s/step - loss: 0.2493 - accuracy: 0.9072 - val_loss: 0.3393 - val_accuracy: 0.9068 - lr: 7.4082e-04\n",
            "Epoch 14/20\n",
            "34/34 [==============================] - 81s 2s/step - loss: 0.1970 - accuracy: 0.9341 - val_loss: 0.3393 - val_accuracy: 0.8559 - lr: 6.7032e-04\n",
            "Epoch 15/20\n",
            "34/34 [==============================] - 79s 2s/step - loss: 0.1898 - accuracy: 0.9267 - val_loss: 0.4892 - val_accuracy: 0.8220 - lr: 6.0653e-04\n",
            "Epoch 16/20\n",
            "34/34 [==============================] - 81s 2s/step - loss: 0.1835 - accuracy: 0.9332 - val_loss: 0.3354 - val_accuracy: 0.8983 - lr: 5.4881e-04\n",
            "Epoch 17/20\n",
            "34/34 [==============================] - 81s 2s/step - loss: 0.1771 - accuracy: 0.9314 - val_loss: 0.3324 - val_accuracy: 0.8814 - lr: 4.9659e-04\n",
            "Epoch 18/20\n",
            "34/34 [==============================] - 79s 2s/step - loss: 0.1693 - accuracy: 0.9360 - val_loss: 0.3922 - val_accuracy: 0.8898 - lr: 4.4933e-04\n",
            "Epoch 19/20\n",
            "34/34 [==============================] - 81s 2s/step - loss: 0.1816 - accuracy: 0.9416 - val_loss: 0.3320 - val_accuracy: 0.8814 - lr: 4.0657e-04\n",
            "Epoch 20/20\n",
            "34/34 [==============================] - 81s 2s/step - loss: 0.1731 - accuracy: 0.9378 - val_loss: 0.3268 - val_accuracy: 0.8898 - lr: 3.6788e-04\n"
          ]
        }
      ],
      "source": [
        "history = cnn.fit(\n",
        "      train_dataset,\n",
        "      epochs=20,\n",
        "      validation_data=val_dataset,\n",
        "      callbacks=[callback]  \n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "anqiZ6WKUQGo",
        "outputId": "d9c04e25-67dc-4d05-d754-9e209a2989a9"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "keras.preprocessing.image.DirectoryIterator"
            ]
          },
          "metadata": {},
          "execution_count": 52
        }
      ],
      "source": [
        "type(train_dataset)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "id": "pEjjsx3mE9HN",
        "outputId": "c68ec30f-1068-4fd2-960b-500a7365dee6"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7f8039c99e10>"
            ]
          },
          "metadata": {},
          "execution_count": 120
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "plt.plot(history.history['accuracy'], label='accuracy')\n",
        "plt.plot(history.history['val_accuracy'], label = 'val_accuracy')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.ylim([0.5, 1])\n",
        "plt.legend(loc='lower right') "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vDyYbPUqojPZ"
      },
      "source": [
        "#feature extraction"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "5o2P2z5_mpzu"
      },
      "outputs": [],
      "source": [
        "feature_extractor = keras.Model(\n",
        "    inputs = cnn.inputs,\n",
        "    outputs = [layer.output for layer in cnn.layers]\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "pKryg7prn5GZ"
      },
      "outputs": [],
      "source": [
        "x = tf.ones((1,250,250,3))\n",
        "features = feature_extractor(x)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LkrdUKGK2VQT",
        "outputId": "e53f0ae9-a59d-4463-e8c6-c93d82bd6dcf"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<tf.Tensor: shape=(1, 125, 125, 64), dtype=float32, numpy=\n",
              "array([[[[0.        , 0.01121285, 0.00841435, ..., 1.023357  ,\n",
              "          0.00330275, 0.0125383 ],\n",
              "         [0.        , 0.01121285, 0.00841435, ..., 1.023357  ,\n",
              "          0.00330275, 0.0125383 ],\n",
              "         [0.        , 0.01121285, 0.00841435, ..., 1.023357  ,\n",
              "          0.00330275, 0.0125383 ],\n",
              "         ...,\n",
              "         [0.        , 0.01121285, 0.00841435, ..., 1.023357  ,\n",
              "          0.00330275, 0.0125383 ],\n",
              "         [0.        , 0.01121285, 0.00841435, ..., 1.023357  ,\n",
              "          0.00330275, 0.0125383 ],\n",
              "         [0.03244157, 0.20869194, 0.        , ..., 0.6566032 ,\n",
              "          0.        , 0.        ]],\n",
              "\n",
              "        [[0.        , 0.01121285, 0.00841435, ..., 1.023357  ,\n",
              "          0.00330275, 0.0125383 ],\n",
              "         [0.        , 0.01121285, 0.00841435, ..., 1.023357  ,\n",
              "          0.00330275, 0.0125383 ],\n",
              "         [0.        , 0.01121285, 0.00841435, ..., 1.023357  ,\n",
              "          0.00330275, 0.0125383 ],\n",
              "         ...,\n",
              "         [0.        , 0.01121285, 0.00841435, ..., 1.023357  ,\n",
              "          0.00330275, 0.0125383 ],\n",
              "         [0.        , 0.01121285, 0.00841435, ..., 1.023357  ,\n",
              "          0.00330275, 0.0125383 ],\n",
              "         [0.03244157, 0.20869194, 0.        , ..., 0.6566032 ,\n",
              "          0.        , 0.        ]],\n",
              "\n",
              "        [[0.        , 0.01121285, 0.00841435, ..., 1.023357  ,\n",
              "          0.00330275, 0.0125383 ],\n",
              "         [0.        , 0.01121285, 0.00841435, ..., 1.023357  ,\n",
              "          0.00330275, 0.0125383 ],\n",
              "         [0.        , 0.01121285, 0.00841435, ..., 1.023357  ,\n",
              "          0.00330275, 0.0125383 ],\n",
              "         ...,\n",
              "         [0.        , 0.01121285, 0.00841435, ..., 1.023357  ,\n",
              "          0.00330275, 0.0125383 ],\n",
              "         [0.        , 0.01121285, 0.00841435, ..., 1.023357  ,\n",
              "          0.00330275, 0.0125383 ],\n",
              "         [0.03244157, 0.20869194, 0.        , ..., 0.6566032 ,\n",
              "          0.        , 0.        ]],\n",
              "\n",
              "        ...,\n",
              "\n",
              "        [[0.        , 0.01121285, 0.00841435, ..., 1.023357  ,\n",
              "          0.00330275, 0.0125383 ],\n",
              "         [0.        , 0.01121285, 0.00841435, ..., 1.023357  ,\n",
              "          0.00330275, 0.0125383 ],\n",
              "         [0.        , 0.01121285, 0.00841435, ..., 1.023357  ,\n",
              "          0.00330275, 0.0125383 ],\n",
              "         ...,\n",
              "         [0.        , 0.01121285, 0.00841435, ..., 1.023357  ,\n",
              "          0.00330275, 0.0125383 ],\n",
              "         [0.        , 0.01121285, 0.00841435, ..., 1.023357  ,\n",
              "          0.00330275, 0.0125383 ],\n",
              "         [0.03244157, 0.20869194, 0.        , ..., 0.6566032 ,\n",
              "          0.        , 0.        ]],\n",
              "\n",
              "        [[0.        , 0.01121285, 0.00841435, ..., 1.023357  ,\n",
              "          0.00330275, 0.0125383 ],\n",
              "         [0.        , 0.01121285, 0.00841435, ..., 1.023357  ,\n",
              "          0.00330275, 0.0125383 ],\n",
              "         [0.        , 0.01121285, 0.00841435, ..., 1.023357  ,\n",
              "          0.00330275, 0.0125383 ],\n",
              "         ...,\n",
              "         [0.        , 0.01121285, 0.00841435, ..., 1.023357  ,\n",
              "          0.00330275, 0.0125383 ],\n",
              "         [0.        , 0.01121285, 0.00841435, ..., 1.023357  ,\n",
              "          0.00330275, 0.0125383 ],\n",
              "         [0.03244157, 0.20869194, 0.        , ..., 0.6566032 ,\n",
              "          0.        , 0.        ]],\n",
              "\n",
              "        [[0.        , 0.        , 0.4376222 , ..., 0.70446587,\n",
              "          0.27488133, 0.2423648 ],\n",
              "         [0.        , 0.        , 0.4376222 , ..., 0.70446587,\n",
              "          0.27488133, 0.2423648 ],\n",
              "         [0.        , 0.        , 0.4376222 , ..., 0.70446587,\n",
              "          0.27488133, 0.2423648 ],\n",
              "         ...,\n",
              "         [0.        , 0.        , 0.4376222 , ..., 0.70446587,\n",
              "          0.27488133, 0.2423648 ],\n",
              "         [0.        , 0.        , 0.4376222 , ..., 0.70446587,\n",
              "          0.27488133, 0.2423648 ],\n",
              "         [0.02623562, 0.0062506 , 0.35491464, ..., 0.44960365,\n",
              "          0.01002078, 0.01920793]]]], dtype=float32)>"
            ]
          },
          "execution_count": 116,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "features[0]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Fm3JOVkvoQAH",
        "outputId": "9f4e2a17-b109-4c8d-da4b-13eaa78b8ec0"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "tensorflow.python.framework.ops.EagerTensor"
            ]
          },
          "execution_count": 50,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "type(features[0])"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wtlFnqG5eHBb"
      },
      "source": [
        "#test"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "QkR01smClwqi"
      },
      "outputs": [],
      "source": [
        "test=[]\n",
        "def create_testing_data():\n",
        "        for img in os.listdir(test_dir):\n",
        "            try:\n",
        "                img_arr=cv2.imread(os.path.join(test_dir,img))\n",
        "                newarr=cv2.resize(img_arr,(300,300))\n",
        "                test.append([newarr])\n",
        "            except Exception as e :\n",
        "                pass\n",
        "        \n",
        "create_testing_data()  "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7Q5AJTC-l1nj",
        "outputId": "babe37aa-8152-4393-9c2e-6e3ac6134c9c"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "513\n"
          ]
        }
      ],
      "source": [
        "print(len(test))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "0fJaSc2ul1kq"
      },
      "outputs": [],
      "source": [
        "test_set = np.array(test).reshape(-1,300,300,3)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "aqiv_fcpxTM_"
      },
      "outputs": [],
      "source": [
        "test_set = test_set/255"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cGyMHUE-zYz-",
        "outputId": "abb11067-4743-495f-b86e-2afd027f1485"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(513, 350, 350, 3)"
            ]
          },
          "metadata": {},
          "execution_count": 134
        }
      ],
      "source": [
        "test_set.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GHSLj735sDnJ",
        "outputId": "5b4e7408-5667-4730-c7db-e1d999865f8b"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(350, 350, 3)"
            ]
          },
          "metadata": {},
          "execution_count": 135
        }
      ],
      "source": [
        "test_set[0].shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 286
        },
        "id": "H57mhNEWryFe",
        "outputId": "b2833769-e698-4dd2-d46a-297268a82d33"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7f8039937610>"
            ]
          },
          "metadata": {},
          "execution_count": 136
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "plt.imshow(test_set[0], cmap='gray')\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-Ko4x3j3AlIt"
      },
      "source": [
        "**Prediction**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "5cK1k6gv-Y6u"
      },
      "outputs": [],
      "source": [
        "predictions = np.argmax(cnn.predict(test_set), axis=1)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "VEwX8H5o3bUA",
        "outputId": "47b866ce-817b-4658-9400-f0da17dd32c7"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([2, 1, 0, 1, 2, 1, 2, 1, 1, 1, 2, 2, 0, 1, 2, 1, 2, 2, 2, 2, 0, 2,\n",
              "       0, 0, 2, 1, 1, 1, 0, 0, 1, 2, 1, 1, 2, 2, 1, 1, 1, 1, 0, 1, 2, 1,\n",
              "       0, 2, 1, 2, 1, 0, 1, 1, 0, 0, 2, 1, 1, 1, 2, 1, 0, 2, 2, 1, 2, 0,\n",
              "       2, 1, 0, 0, 1, 2, 1, 2, 0, 2, 1, 2, 1, 1, 2, 0, 2, 1, 1, 2, 1, 1,\n",
              "       2, 0, 2, 2, 1, 2, 1, 2, 2, 0, 1, 2, 1, 1, 0, 2, 0, 0, 1, 1, 1, 1,\n",
              "       2, 1, 2, 2, 1, 2, 0, 0, 1, 2, 2, 1, 1, 0, 1, 1, 1, 0, 0, 0, 1, 1,\n",
              "       1, 2, 1, 1, 2, 1, 0, 1, 2, 2, 1, 0, 1, 1, 1, 0, 0, 2, 1, 1, 2, 2,\n",
              "       0, 2, 0, 2, 2, 2, 1, 0, 1, 0, 1, 1, 1, 2, 0, 1, 0, 0, 1, 2, 2, 0,\n",
              "       2, 1, 0, 0, 0, 1, 2, 2, 0, 2, 2, 2, 0, 0, 1, 2, 1, 1, 1, 2, 1, 1,\n",
              "       1, 1, 1, 0, 0, 2, 1, 2, 1, 0, 2, 0, 1, 0, 2, 0, 1, 0, 2, 2, 1, 1,\n",
              "       2, 1, 2, 2, 2, 0, 2, 1, 2, 0, 2, 0, 0, 1, 0, 0, 1, 1, 0, 1, 1, 1,\n",
              "       0, 0, 1, 1, 2, 1, 2, 1, 0, 0, 1, 1, 1, 2, 1, 1, 2, 1, 2, 1, 0, 1,\n",
              "       0, 1, 0, 1, 1, 1, 2, 2, 0, 1, 1, 0, 0, 0, 1, 1, 0, 2, 1, 2, 0, 2,\n",
              "       1, 1, 1, 0, 0, 1, 2, 1, 1, 2, 1, 2, 0, 1, 1, 0, 1, 2, 2, 1, 2, 2,\n",
              "       1, 0, 2, 1, 2, 2, 1, 2, 0, 2, 1, 1, 1, 2, 1, 1, 1, 1, 1, 0, 2, 1,\n",
              "       2, 1, 1, 2, 1, 2, 0, 1, 0, 2, 0, 2, 0, 2, 1, 2, 1, 2, 0, 2, 2, 1,\n",
              "       2, 2, 0, 1, 2, 2, 1, 1, 2, 0, 2, 2, 1, 1, 1, 0, 1, 1, 1, 1, 0, 1,\n",
              "       2, 2, 2, 1, 0, 1, 1, 2, 2, 1, 1, 2, 1, 2, 2, 1, 2, 1, 0, 0, 1, 1,\n",
              "       2, 1, 2, 0, 2, 1, 1, 1, 2, 2, 0, 0, 2, 2, 1, 0, 0, 0, 1, 2, 2, 1,\n",
              "       1, 2, 0, 0, 0, 2, 2, 0, 1, 2, 1, 1, 2, 2, 2, 2, 0, 0, 2, 2, 1, 1,\n",
              "       1, 2, 2, 2, 2, 2, 1, 1, 1, 1, 1, 2, 2, 1, 2, 2, 2, 1, 2, 2, 0, 1,\n",
              "       2, 2, 2, 0, 2, 1, 0, 1, 1, 1, 1, 1, 0, 1, 2, 2, 1, 1, 2, 1, 1, 0,\n",
              "       1, 0, 2, 0, 2, 1, 2, 2, 1, 1, 2, 1, 1, 2, 0, 0, 0, 2, 1, 1, 1, 0,\n",
              "       1, 1, 2, 2, 0, 0, 0])"
            ]
          },
          "metadata": {},
          "execution_count": 138
        }
      ],
      "source": [
        "predictions"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0fZFH7YkslW9",
        "outputId": "5a00e553-9e5e-45d7-9acc-adff8b86bb32"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "513"
            ]
          },
          "metadata": {},
          "execution_count": 129
        }
      ],
      "source": [
        "len(predictions)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "WuOE2qtl3bR5"
      },
      "outputs": [],
      "source": [
        "#{'covid': 0, 'normal': 1, 'virus': 2}\n",
        "\n",
        "le = preprocessing.LabelEncoder()\n",
        "le.fit([\"covid\", \"normal\", \"virus\"])\n",
        "\n",
        "list(le.classes_)\n",
        "\n",
        "tr = le.transform([\"covid\", \"normal\", \"virus\"])\n",
        "\n",
        "label = list(le.inverse_transform(predictions))\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "e79JE3LiEDcD",
        "outputId": "672abd31-2f65-4796-c110-e2d05c2329c9"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0, 1, 2])"
            ]
          },
          "metadata": {},
          "execution_count": 140
        }
      ],
      "source": [
        "tr"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4Fqf9T6UBE4b",
        "outputId": "e12948c7-1c47-4063-b909-3aa64860bda2"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['virus',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'virus',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'covid',\n",
              " 'normal',\n",
              " 'normal',\n",
              " 'virus',\n",
              " 'virus',\n",
              " 'covid',\n",
              " 'covid',\n",
              " 'covid']"
            ]
          },
          "metadata": {},
          "execution_count": 141
        }
      ],
      "source": [
        "label"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "3jQ11d09FdSc"
      },
      "outputs": [],
      "source": [
        "lables_images = []\n",
        "for index, lab in enumerate(label):\n",
        "  lables_images.append({\"Image\": os.listdir(test_dir)[index], \"Label\": lab})"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "FkHLStrzLPKW"
      },
      "outputs": [],
      "source": [
        "datafram = pd.DataFrame(lables_images)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7eeX2InJ_UXp",
        "outputId": "55b6b9af-7e75-45ed-afe5-a9afe6cf9a67"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "pandas.core.frame.DataFrame"
            ]
          },
          "metadata": {},
          "execution_count": 144
        }
      ],
      "source": [
        "type(datafram)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 17
        },
        "id": "YPckGNGOMKGC",
        "outputId": "bdfb8207-9bc6-4026-fe18-9485772cc1f8"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "download(\"download_ae2b6ba8-9b6b-4655-9d34-094edd4dca1f\", \"sub21.csv\", 10742)"
            ]
          },
          "metadata": {}
        }
      ],
      "source": [
        "datafram.to_csv('sub21.csv', encoding = 'utf-8-sig', index=False,header = 1) \n",
        "files.download('sub21.csv') "
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "fI1-D_3Ka5Gr"
      },
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "colab": {
      "provenance": [],
      "collapsed_sections": [
        "vDyYbPUqojPZ"
      ]
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}