Brain_Tumor_Detection.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Brain Tumor Detection.ipynb",
      "provenance": [],
      "authorship_tag": "ABX9TyMJwEVnqCRw/aTl5TZ5WBl4",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/DawoodBhai420/Brain-Tumor-Detection/blob/main/Brain_Tumor_Detection.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "colab": {
          "resources": {
            "http://localhost:8080/nbextensions/google.colab/files.js": {
              "data": "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",
              "ok": true,
              "headers": [
                [
                  "content-type",
                  "application/javascript"
                ]
              ],
              "status": 200,
              "status_text": ""
            }
          },
          "base_uri": "https://localhost:8080/",
          "height": 493
        },
        "id": "RK5wv9JgVC0j",
        "outputId": "544f54eb-50c0-41d9-ce84-39302a3d826a"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "\n",
              "     <input type=\"file\" id=\"files-37e9e111-f1ce-41b0-a7ba-974645453b5d\" name=\"files[]\" multiple disabled\n",
              "        style=\"border:none\" />\n",
              "     <output id=\"result-37e9e111-f1ce-41b0-a7ba-974645453b5d\">\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 src=\"/nbextensions/google.colab/files.js\"></script> "
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Saving kaggle.json to kaggle.json\n",
            "Warning: Looks like you're using an outdated API Version, please consider updating (server 1.5.12 / client 1.5.4)\n",
            "ref                                                         title                                               size  lastUpdated          downloadCount  \n",
            "----------------------------------------------------------  -------------------------------------------------  -----  -------------------  -------------  \n",
            "nkitgupta/jigsaw-regression-based-data                      Jigsaw Regression Based Data                         3GB  2022-01-10 06:29:59            666  \n",
            "yamqwe/omicron-covid19-variant-daily-cases                  Omicron daily cases by country (COVID-19 variant)  432KB  2022-01-30 09:11:54          11970  \n",
            "prasertk/netflix-subscription-price-in-different-countries  Netflix subscription fee in different countries      3KB  2022-01-15 07:06:09           1672  \n",
            "yasserh/wine-quality-dataset                                Wine Quality Dataset                                21KB  2022-01-15 19:15:11           1490  \n",
            "iamsouravbanerjee/analytics-industry-salaries-2022-india    Data Professionals Salary - 2022                    57KB  2022-02-02 04:35:21           1864  \n",
            "yamqwe/netflix-showse                                       Netflix Shows                                       11KB  2022-01-23 00:03:01            970  \n",
            "sanjeetsinghnaik/top-1000-highest-grossing-movies           Top 1000 Highest Grossing Movies                   106KB  2022-01-15 16:26:14            983  \n",
            "yasserh/amazon-product-reviews-dataset                      Amazon Product Reviews Dataset                     708KB  2022-01-23 17:21:16            493  \n",
            "vishalmane10/anime-dataset-2022                             Anime DataSet 2022                                   5MB  2022-01-16 13:56:16            788  \n",
            "carlmcbrideellis/gdp-20152019-finland-norway-and-sweden     GDP 2015-2019: Finland, Norway, and Sweden          365B  2022-01-05 07:48:49            539  \n",
            "sandipdevre/petrol-prices-in-india                          Petrol Prices In India                               3KB  2022-01-27 09:07:35            305  \n",
            "dansbecker/melbourne-housing-snapshot                       Melbourne Housing Snapshot                         451KB  2018-06-05 12:52:24          79813  \n",
            "yamqwe/men-s-shoe-pricese                                   🩰 Men's Shoe Prices                                  6MB  2022-01-24 15:19:48            380  \n",
            "nenamalikah/nft-collections-by-sales-volume                 Top NFT Collections                                 47KB  2022-01-17 00:47:41            543  \n",
            "meetnagadia/netflix-stock-price-data-set-20022022           Netflix Stock Price Data set 2002-2022              94KB  2022-01-12 05:28:11            505  \n",
            "datasnaek/youtube-new                                       Trending YouTube Video Statistics                  201MB  2019-06-03 00:56:47         164752  \n",
            "zynicide/wine-reviews                                       Wine Reviews                                        51MB  2017-11-27 17:08:04         153470  \n",
            "fedesoriano/stellar-classification-dataset-sdss17           Stellar Classification Dataset - SDSS17              7MB  2022-01-15 17:11:30            485  \n",
            "rtatman/188-million-us-wildfires                            1.88 Million US Wildfires                          168MB  2020-05-12 21:03:49          18747  \n",
            "datasnaek/chess                                             Chess Game Dataset (Lichess)                         3MB  2017-09-04 03:09:09          27360  \n"
          ]
        }
      ],
      "source": [
        "! pip install -q kaggle\n",
        "from google.colab import files\n",
        "files.upload()\n",
        "! mkdir ~/.kaggle\n",
        "! cp kaggle.json ~/.kaggle/\n",
        "! chmod 600 ~/.kaggle/kaggle.json\n",
        "! kaggle datasets list"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "! kaggle datasets download -d preetviradiya/brian-tumor-dataset"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1awwRQnFVUpu",
        "outputId": "eb75e355-f01c-41b1-9229-5479023eddce"
      },
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading brian-tumor-dataset.zip to /content\n",
            " 97% 104M/107M [00:00<00:00, 103MB/s]  \n",
            "100% 107M/107M [00:00<00:00, 127MB/s]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import os\n",
        "import zipfile\n",
        "\n",
        "local_zip = '/content/brian-tumor-dataset.zip'\n",
        "zip_ref = zipfile.ZipFile(local_zip, 'r')\n",
        "zip_ref.extractall('./brian-tumor-dataset')\n",
        "\n",
        "zip_ref.close()"
      ],
      "metadata": {
        "id": "41ymJJieWopN"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "brain_tumor_images = os.path.join('/content/brian-tumor-dataset/Brain Tumor Data Set/Brain Tumor Data Set/Brain Tumor')\n",
        "healthy_images = os.path.join('/content/brian-tumor-dataset/Brain Tumor Data Set/Brain Tumor Data Set/Healthy')"
      ],
      "metadata": {
        "id": "kxZKYW0NXQS7"
      },
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(len(os.listdir(brain_tumor_images)))\n",
        "print(len(os.listdir(healthy_images)))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NI4ag-icXvKt",
        "outputId": "af12c744-c4d1-4da0-b28c-958c59908733"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "2513\n",
            "2087\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import tensorflow as tf\n",
        "from tensorflow import keras\n",
        "from keras.preprocessing.image import ImageDataGenerator"
      ],
      "metadata": {
        "id": "OHOgp6psX5NW"
      },
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "datagen = ImageDataGenerator(rescale=1./255,validation_split = 0.2,zoom_range=(0.99,0.99),dtype=tf.float32)\n",
        "train_generator = datagen.flow_from_directory('/content/brian-tumor-dataset/Brain Tumor Data Set/Brain Tumor Data Set',\n",
        "                                              target_size = (150,150),\n",
        "                                              batch_size = 256,\n",
        "                                              class_mode = \"binary\",\n",
        "                                              color_mode = \"rgb\",\n",
        "                                              shuffle = True,\n",
        "                                              seed = 123,\n",
        "                                              subset = \"training\")\n",
        "validation_generator = datagen.flow_from_directory('/content/brian-tumor-dataset/Brain Tumor Data Set/Brain Tumor Data Set',\n",
        "                                                   target_size = (150,150),\n",
        "                                                   batch_size = 8,\n",
        "                                                   class_mode = \"binary\",\n",
        "                                                   color_mode = \"rgb\",\n",
        "                                                   shuffle = True,\n",
        "                                                   seed = 123,\n",
        "                                                   subset = \"validation\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "WEDIdOiQaee5",
        "outputId": "2aad2095-3c4a-4a32-becf-3c41fe3bb524"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Found 3681 images belonging to 2 classes.\n",
            "Found 919 images belonging to 2 classes.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "classes = validation_generator.class_indices\n",
        "classes"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JapQAnoYb6dm",
        "outputId": "4b00c253-22f3-4c7a-839e-7773f4db5a67"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'Brain Tumor': 0, 'Healthy': 1}"
            ]
          },
          "metadata": {},
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt"
      ],
      "metadata": {
        "id": "k5n21TWZcIIp"
      },
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "batch = next(validation_generator)\n",
        "plt.imshow(batch[0][0])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 286
        },
        "id": "JKBOCeKLcQtT",
        "outputId": "6a417d92-9a4b-4a40-9253-b520d48a043a"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7f5205ca15d0>"
            ]
          },
          "metadata": {},
          "execution_count": 10
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model = keras.models.Sequential([\n",
        "                                 keras.layers.Conv2D(32, (3,3), activation='relu', input_shape=(150,150,3), padding='same'),\n",
        "                                 keras.layers.MaxPool2D(),\n",
        "                                 keras.layers.Conv2D(64, (3,3), activation='relu', padding='same'),\n",
        "                                 keras.layers.MaxPool2D(),\n",
        "                                 keras.layers.Flatten(),\n",
        "                                 keras.layers.Dense(128, activation='relu'),\n",
        "                                 keras.layers.BatchNormalization(),\n",
        "                                 keras.layers.Dropout(0.3),\n",
        "                                 keras.layers.Dense(64, activation='relu'),\n",
        "                                 keras.layers.BatchNormalization(),\n",
        "                                 keras.layers.Dropout(0.3),\n",
        "                                 keras.layers.Dense(1, activation='sigmoid')\n",
        "])"
      ],
      "metadata": {
        "id": "Mcg-uJqGc-9U"
      },
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model.compile(optimizer='adam',\n",
        "              loss='binary_crossentropy',\n",
        "              metrics=['accuracy'])"
      ],
      "metadata": {
        "id": "1-XGIsDye6sE"
      },
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "n4cO2ALVff7m",
        "outputId": "964ab585-19e5-4927-e1f9-0f7428e52fce"
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " conv2d (Conv2D)             (None, 150, 150, 32)      896       \n",
            "                                                                 \n",
            " max_pooling2d (MaxPooling2D  (None, 75, 75, 32)       0         \n",
            " )                                                               \n",
            "                                                                 \n",
            " conv2d_1 (Conv2D)           (None, 75, 75, 64)        18496     \n",
            "                                                                 \n",
            " max_pooling2d_1 (MaxPooling  (None, 37, 37, 64)       0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " flatten (Flatten)           (None, 87616)             0         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 128)               11214976  \n",
            "                                                                 \n",
            " batch_normalization (BatchN  (None, 128)              512       \n",
            " ormalization)                                                   \n",
            "                                                                 \n",
            " dropout (Dropout)           (None, 128)               0         \n",
            "                                                                 \n",
            " dense_1 (Dense)             (None, 64)                8256      \n",
            "                                                                 \n",
            " batch_normalization_1 (Batc  (None, 64)               256       \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " dropout_1 (Dropout)         (None, 64)                0         \n",
            "                                                                 \n",
            " dense_2 (Dense)             (None, 1)                 65        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 11,243,457\n",
            "Trainable params: 11,243,073\n",
            "Non-trainable params: 384\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "history = model.fit(train_generator,\n",
        "                    verbose=2,\n",
        "                    epochs=20,\n",
        "                    validation_data=validation_generator)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pJbIj-FRf8YV",
        "outputId": "ad4fc8be-353b-4178-dad0-a1af7c88ce8f"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/20\n",
            "15/15 - 48s - loss: 0.6623 - accuracy: 0.6762 - val_loss: 0.9265 - val_accuracy: 0.6931 - 48s/epoch - 3s/step\n",
            "Epoch 2/20\n",
            "15/15 - 36s - loss: 0.4418 - accuracy: 0.7968 - val_loss: 0.5101 - val_accuracy: 0.7682 - 36s/epoch - 2s/step\n",
            "Epoch 3/20\n",
            "15/15 - 36s - loss: 0.2647 - accuracy: 0.8941 - val_loss: 0.3353 - val_accuracy: 0.8934 - 36s/epoch - 2s/step\n",
            "Epoch 4/20\n",
            "15/15 - 35s - loss: 0.1601 - accuracy: 0.9438 - val_loss: 0.2731 - val_accuracy: 0.9543 - 35s/epoch - 2s/step\n",
            "Epoch 5/20\n",
            "15/15 - 35s - loss: 0.0919 - accuracy: 0.9756 - val_loss: 0.2704 - val_accuracy: 0.9325 - 35s/epoch - 2s/step\n",
            "Epoch 6/20\n",
            "15/15 - 35s - loss: 0.0593 - accuracy: 0.9864 - val_loss: 0.2400 - val_accuracy: 0.9695 - 35s/epoch - 2s/step\n",
            "Epoch 7/20\n",
            "15/15 - 35s - loss: 0.0393 - accuracy: 0.9938 - val_loss: 0.2243 - val_accuracy: 0.9336 - 35s/epoch - 2s/step\n",
            "Epoch 8/20\n",
            "15/15 - 35s - loss: 0.0246 - accuracy: 0.9962 - val_loss: 0.1500 - val_accuracy: 0.9739 - 35s/epoch - 2s/step\n",
            "Epoch 9/20\n",
            "15/15 - 35s - loss: 0.0151 - accuracy: 0.9995 - val_loss: 0.1370 - val_accuracy: 0.9924 - 35s/epoch - 2s/step\n",
            "Epoch 10/20\n",
            "15/15 - 35s - loss: 0.0100 - accuracy: 0.9995 - val_loss: 0.1193 - val_accuracy: 0.9967 - 35s/epoch - 2s/step\n",
            "Epoch 11/20\n",
            "15/15 - 36s - loss: 0.0074 - accuracy: 0.9997 - val_loss: 0.1144 - val_accuracy: 0.9924 - 36s/epoch - 2s/step\n",
            "Epoch 12/20\n",
            "15/15 - 35s - loss: 0.0055 - accuracy: 0.9997 - val_loss: 0.1021 - val_accuracy: 0.9967 - 35s/epoch - 2s/step\n",
            "Epoch 13/20\n",
            "15/15 - 35s - loss: 0.0045 - accuracy: 1.0000 - val_loss: 0.1047 - val_accuracy: 0.9880 - 35s/epoch - 2s/step\n",
            "Epoch 14/20\n",
            "15/15 - 34s - loss: 0.0040 - accuracy: 1.0000 - val_loss: 0.0689 - val_accuracy: 0.9978 - 34s/epoch - 2s/step\n",
            "Epoch 15/20\n",
            "15/15 - 36s - loss: 0.0038 - accuracy: 0.9997 - val_loss: 0.0644 - val_accuracy: 0.9956 - 36s/epoch - 2s/step\n",
            "Epoch 16/20\n",
            "15/15 - 36s - loss: 0.0031 - accuracy: 1.0000 - val_loss: 0.0573 - val_accuracy: 0.9978 - 36s/epoch - 2s/step\n",
            "Epoch 17/20\n",
            "15/15 - 35s - loss: 0.0029 - accuracy: 1.0000 - val_loss: 0.0446 - val_accuracy: 0.9989 - 35s/epoch - 2s/step\n",
            "Epoch 18/20\n",
            "15/15 - 34s - loss: 0.0024 - accuracy: 1.0000 - val_loss: 0.0431 - val_accuracy: 0.9989 - 34s/epoch - 2s/step\n",
            "Epoch 19/20\n",
            "15/15 - 34s - loss: 0.0018 - accuracy: 1.0000 - val_loss: 0.0364 - val_accuracy: 0.9967 - 34s/epoch - 2s/step\n",
            "Epoch 20/20\n",
            "15/15 - 34s - loss: 0.0018 - accuracy: 1.0000 - val_loss: 0.0863 - val_accuracy: 0.9902 - 34s/epoch - 2s/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "acc = history.history['accuracy']\n",
        "val_acc = history.history['val_accuracy']\n",
        "loss = history.history['loss']\n",
        "val_loss = history.history['val_loss']\n",
        "\n",
        "epochs = range(len(acc))\n",
        "\n",
        "plt.plot(epochs, acc, 'r', label='Training accuracy')\n",
        "plt.plot(epochs, val_acc, 'b', label='Validation accuracy')\n",
        "plt.title('Training and validation accuracy')\n",
        "plt.legend(loc=0)\n",
        "plt.figure()\n",
        "\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 298
        },
        "id": "lLasRinrg0W4",
        "outputId": "686a894e-800e-44a5-e345-5df5c4559de2"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 0 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.plot(epochs, loss, 'r', label='Training loss')\n",
        "plt.plot(epochs, val_loss, 'b', label='Validation loss')\n",
        "plt.title('Training and validation loss')\n",
        "plt.legend(loc=0)\n",
        "plt.figure()\n",
        "\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 298
        },
        "id": "TU1EzeBYoMpn",
        "outputId": "e6ec3a27-8c73-440b-e211-1288903433f8"
      },
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 0 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model.save(\"model.h5\")"
      ],
      "metadata": {
        "id": "ENDXE-yT44zq"
      },
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "from google.colab import files\n",
        "from keras.preprocessing import image\n",
        "\n",
        "uploaded = files.upload()\n",
        "\n",
        "for fn in uploaded.keys():\n",
        " \n",
        "  # predicting images\n",
        "  path = '/content/' + fn\n",
        "  img = image.load_img(path, target_size=(150, 150))\n",
        "  x = image.img_to_array(img)\n",
        "  x = np.expand_dims(x, axis=0)\n",
        "\n",
        "  images = np.vstack([x])\n",
        "  classes = model.predict(images, batch_size=10)\n",
        "  print(classes[0])\n",
        "  if classes[0]<0.5:\n",
        "    print(fn + \" is having Brain Tumor.\")\n",
        "  else:\n",
        "    print(fn + \" is healthy.\")\n",
        "    "
      ],
      "metadata": {
        "colab": {
          "resources": {
            "http://localhost:8080/nbextensions/google.colab/files.js": {
              "data": "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",
              "ok": true,
              "headers": [
                [
                  "content-type",
                  "application/javascript"
                ]
              ],
              "status": 200,
              "status_text": ""
            }
          },
          "base_uri": "https://localhost:8080/",
          "height": 402
        },
        "id": "2pmZc9XUoZFr",
        "outputId": "f4b8f6a7-6daf-4db5-f8d0-bdac6a79392b"
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "\n",
              "     <input type=\"file\" id=\"files-2af4c1b0-bdc9-4850-aba6-80eb26f1503c\" name=\"files[]\" multiple disabled\n",
              "        style=\"border:none\" />\n",
              "     <output id=\"result-2af4c1b0-bdc9-4850-aba6-80eb26f1503c\">\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 src=\"/nbextensions/google.colab/files.js\"></script> "
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "error",
          "ename": "KeyboardInterrupt",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-17-252dfff613e7>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0mkeras\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpreprocessing\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mimage\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 5\u001b[0;31m \u001b[0muploaded\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfiles\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mupload\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      6\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      7\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mfn\u001b[0m \u001b[0;32min\u001b[0m \u001b[0muploaded\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mkeys\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/google/colab/files.py\u001b[0m in \u001b[0;36mupload\u001b[0;34m()\u001b[0m\n\u001b[1;32m     61\u001b[0m   result = _output.eval_js(\n\u001b[1;32m     62\u001b[0m       'google.colab._files._uploadFiles(\"{input_id}\", \"{output_id}\")'.format(\n\u001b[0;32m---> 63\u001b[0;31m           input_id=input_id, output_id=output_id))\n\u001b[0m\u001b[1;32m     64\u001b[0m   \u001b[0mfiles\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_collections\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdefaultdict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0m_six\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbinary_type\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     65\u001b[0m   \u001b[0;31m# Mapping from original filename to filename as saved locally.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/google/colab/output/_js.py\u001b[0m in \u001b[0;36meval_js\u001b[0;34m(script, ignore_result, timeout_sec)\u001b[0m\n\u001b[1;32m     38\u001b[0m   \u001b[0;32mif\u001b[0m \u001b[0mignore_result\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     39\u001b[0m     \u001b[0;32mreturn\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 40\u001b[0;31m   \u001b[0;32mreturn\u001b[0m \u001b[0m_message\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread_reply_from_input\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrequest_id\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtimeout_sec\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     41\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     42\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/google/colab/_message.py\u001b[0m in \u001b[0;36mread_reply_from_input\u001b[0;34m(message_id, timeout_sec)\u001b[0m\n\u001b[1;32m     99\u001b[0m     \u001b[0mreply\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_read_next_input_message\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    100\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mreply\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0m_NOT_READY\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mreply\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdict\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 101\u001b[0;31m       \u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msleep\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m0.025\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    102\u001b[0m       \u001b[0;32mcontinue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    103\u001b[0m     if (reply.get('type') == 'colab_reply' and\n",
            "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "reconstructed_model = keras.models.load_model(\"model.h5\")"
      ],
      "metadata": {
        "id": "MtLoP6Uwot8U"
      },
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "reconstructed_model"
      ],
      "metadata": {
        "id": "6ApJ2dKZ50TH",
        "outputId": "693f0b27-d6cd-4a64-ee04-6a2fb462c826",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.engine.sequential.Sequential at 0x7f5204f02690>"
            ]
          },
          "metadata": {},
          "execution_count": 22
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "path = '/content/brian-tumor-dataset/Brain Tumor Data Set/Brain Tumor Data Set/Healthy/Not Cancer  (1001).jpg'\n",
        "img = image.load_img(path, target_size=(150, 150))\n",
        "x = image.img_to_array(img)\n",
        "x = np.expand_dims(x, axis=0)\n",
        "images = np.vstack([x])\n",
        "pred = reconstructed_model.predict(images, batch_size=10)\n",
        "print(pred)"
      ],
      "metadata": {
        "id": "xWFj_KUA56Hw",
        "outputId": "b9caebfa-9c31-42de-fe53-d6712a912000",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[[1.]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        ""
      ],
      "metadata": {
        "id": "g1G8B-P66TOw"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}