pizza_steak_sushi_imageclassify_WITH_URL.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "pizza_steak_sushi_imageclassify.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 25,
      "metadata": {
        "id": "LfEFN4YW-zHa"
      },
      "outputs": [],
      "source": [
        "import numpy as np # linear algebra\n",
        "import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import tensorflow as tf"
      ],
      "metadata": {
        "id": "Dg0jzoW_Gjek"
      },
      "execution_count": 26,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#!unzip -q /content/pizza_steak_sushi.zip"
      ],
      "metadata": {
        "id": "DxS2pbL0Gyj1"
      },
      "execution_count": 27,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "base_dir = \"/content/train\""
      ],
      "metadata": {
        "id": "ZUWAlwZ1HEs7"
      },
      "execution_count": 28,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        ""
      ],
      "metadata": {
        "id": "Q2Ev622JTEDh"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_datagen = tf.keras.preprocessing.image.ImageDataGenerator(\n",
        "    rescale = 1./255,rotation_range = 40,width_shift_range = 0.2,height_shift_range = 0.2,shear_range = 0.2,\n",
        "    zoom_range = 0.2,horizontal_flip = True,fill_mode = \"nearest\" ,validation_split = 0.2\n",
        ")\n",
        "validation_datagen = tf.keras.preprocessing.image.ImageDataGenerator(rescale = 1./255,validation_split = 0.2)\n",
        "\n",
        "train_generator = train_datagen.flow_from_directory(base_dir,target_size = (300,300),class_mode = \"categorical\",\n",
        "                                                                  batch_size = 4,subset = \"training\",seed = 42)\n",
        "validation_generator = validation_datagen.flow_from_directory(base_dir,target_size = (300,300),class_mode = \"categorical\",\n",
        "                                                         batch_size =4,subset = \"validation\",seed = 42)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "C-CyLQwwGq94",
        "outputId": "e9f2f008-3680-4803-d667-f4f3bb7179b9"
      },
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Found 181 images belonging to 3 classes.\n",
            "Found 44 images belonging to 3 classes.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow.keras.applications.inception_v3 import InceptionV3\n",
        "base_model = InceptionV3(input_shape = (300,300,3),include_top = False,weights = \"imagenet\")\n",
        "\"\"\"Freeze all layers to stop updating InceptionV3 trained weights.\"\"\"\n",
        "for layer in base_model.layers:\n",
        "    layer.trainable = False"
      ],
      "metadata": {
        "id": "vO93HsBWHn1M"
      },
      "execution_count": 30,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "\"\"\"You can change it if you want.\"\"\"\n",
        "last_layer = base_model.get_layer(\"mixed8\")\n",
        "print(last_layer.output_shape)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6hVqpIJIHsdX",
        "outputId": "6da25130-c7c1-48f6-d28d-34c1cc4e2856"
      },
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(None, 8, 8, 1280)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import os \n",
        "classes = os.listdir(base_dir)\n",
        "for i in range(len(classes)):\n",
        "    path = os.path.join(base_dir,classes[i])\n",
        "    print(f\"{classes[i]} has {len(os.listdir(path))} images\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SC01iZx6H-Vi",
        "outputId": "d178fdf1-edcd-4778-dc53-44630c7a45f8"
      },
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "sushi has 72 images\n",
            "pizza has 78 images\n",
            "steak has 75 images\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "\"\"\"Flatten layer to reduce output shape to 1dim.\"\"\"\n",
        "x = tf.keras.layers.Flatten()(last_layer.output)\n",
        "\"\"\"Add fully connected layers with 256 units.\"\"\"\n",
        "x = tf.keras.layers.Dense(256,activation = \"relu\")(x)\n",
        "\"\"\"Add Dropout layer.\"\"\"\n",
        "x = tf.keras.layers.Dropout(0.2)(x)\n",
        "\"\"\"Output layer.\"\"\"\n",
        "x = tf.keras.layers.Dense(len(classes),activation = \"softmax\")(x)\n",
        "\"\"\"Here we can connect our model end to end.\"\"\"\n",
        "model = tf.keras.models.Model(base_model.input,x)"
      ],
      "metadata": {
        "id": "1uLbWwtXH0H0"
      },
      "execution_count": 33,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EcT9d6lBIIyu",
        "outputId": "769792d3-4392-4458-e275-dcec9b14a233"
      },
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model_1\"\n",
            "__________________________________________________________________________________________________\n",
            " Layer (type)                   Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            " input_2 (InputLayer)           [(None, 300, 300, 3  0           []                               \n",
            "                                )]                                                                \n",
            "                                                                                                  \n",
            " conv2d_94 (Conv2D)             (None, 149, 149, 32  864         ['input_2[0][0]']                \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " batch_normalization_94 (BatchN  (None, 149, 149, 32  96         ['conv2d_94[0][0]']              \n",
            " ormalization)                  )                                                                 \n",
            "                                                                                                  \n",
            " activation_94 (Activation)     (None, 149, 149, 32  0           ['batch_normalization_94[0][0]'] \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv2d_95 (Conv2D)             (None, 147, 147, 32  9216        ['activation_94[0][0]']          \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " batch_normalization_95 (BatchN  (None, 147, 147, 32  96         ['conv2d_95[0][0]']              \n",
            " ormalization)                  )                                                                 \n",
            "                                                                                                  \n",
            " activation_95 (Activation)     (None, 147, 147, 32  0           ['batch_normalization_95[0][0]'] \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " conv2d_96 (Conv2D)             (None, 147, 147, 64  18432       ['activation_95[0][0]']          \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " batch_normalization_96 (BatchN  (None, 147, 147, 64  192        ['conv2d_96[0][0]']              \n",
            " ormalization)                  )                                                                 \n",
            "                                                                                                  \n",
            " activation_96 (Activation)     (None, 147, 147, 64  0           ['batch_normalization_96[0][0]'] \n",
            "                                )                                                                 \n",
            "                                                                                                  \n",
            " max_pooling2d_4 (MaxPooling2D)  (None, 73, 73, 64)  0           ['activation_96[0][0]']          \n",
            "                                                                                                  \n",
            " conv2d_97 (Conv2D)             (None, 73, 73, 80)   5120        ['max_pooling2d_4[0][0]']        \n",
            "                                                                                                  \n",
            " batch_normalization_97 (BatchN  (None, 73, 73, 80)  240         ['conv2d_97[0][0]']              \n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " activation_97 (Activation)     (None, 73, 73, 80)   0           ['batch_normalization_97[0][0]'] \n",
            "                                                                                                  \n",
            " conv2d_98 (Conv2D)             (None, 71, 71, 192)  138240      ['activation_97[0][0]']          \n",
            "                                                                                                  \n",
            " batch_normalization_98 (BatchN  (None, 71, 71, 192)  576        ['conv2d_98[0][0]']              \n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " activation_98 (Activation)     (None, 71, 71, 192)  0           ['batch_normalization_98[0][0]'] \n",
            "                                                                                                  \n",
            " max_pooling2d_5 (MaxPooling2D)  (None, 35, 35, 192)  0          ['activation_98[0][0]']          \n",
            "                                                                                                  \n",
            " conv2d_102 (Conv2D)            (None, 35, 35, 64)   12288       ['max_pooling2d_5[0][0]']        \n",
            "                                                                                                  \n",
            " batch_normalization_102 (Batch  (None, 35, 35, 64)  192         ['conv2d_102[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_102 (Activation)    (None, 35, 35, 64)   0           ['batch_normalization_102[0][0]']\n",
            "                                                                                                  \n",
            " conv2d_100 (Conv2D)            (None, 35, 35, 48)   9216        ['max_pooling2d_5[0][0]']        \n",
            "                                                                                                  \n",
            " conv2d_103 (Conv2D)            (None, 35, 35, 96)   55296       ['activation_102[0][0]']         \n",
            "                                                                                                  \n",
            " batch_normalization_100 (Batch  (None, 35, 35, 48)  144         ['conv2d_100[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_103 (Batch  (None, 35, 35, 96)  288         ['conv2d_103[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_100 (Activation)    (None, 35, 35, 48)   0           ['batch_normalization_100[0][0]']\n",
            "                                                                                                  \n",
            " activation_103 (Activation)    (None, 35, 35, 96)   0           ['batch_normalization_103[0][0]']\n",
            "                                                                                                  \n",
            " average_pooling2d_9 (AveragePo  (None, 35, 35, 192)  0          ['max_pooling2d_5[0][0]']        \n",
            " oling2D)                                                                                         \n",
            "                                                                                                  \n",
            " conv2d_99 (Conv2D)             (None, 35, 35, 64)   12288       ['max_pooling2d_5[0][0]']        \n",
            "                                                                                                  \n",
            " conv2d_101 (Conv2D)            (None, 35, 35, 64)   76800       ['activation_100[0][0]']         \n",
            "                                                                                                  \n",
            " conv2d_104 (Conv2D)            (None, 35, 35, 96)   82944       ['activation_103[0][0]']         \n",
            "                                                                                                  \n",
            " conv2d_105 (Conv2D)            (None, 35, 35, 32)   6144        ['average_pooling2d_9[0][0]']    \n",
            "                                                                                                  \n",
            " batch_normalization_99 (BatchN  (None, 35, 35, 64)  192         ['conv2d_99[0][0]']              \n",
            " ormalization)                                                                                    \n",
            "                                                                                                  \n",
            " batch_normalization_101 (Batch  (None, 35, 35, 64)  192         ['conv2d_101[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_104 (Batch  (None, 35, 35, 96)  288         ['conv2d_104[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_105 (Batch  (None, 35, 35, 32)  96          ['conv2d_105[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_99 (Activation)     (None, 35, 35, 64)   0           ['batch_normalization_99[0][0]'] \n",
            "                                                                                                  \n",
            " activation_101 (Activation)    (None, 35, 35, 64)   0           ['batch_normalization_101[0][0]']\n",
            "                                                                                                  \n",
            " activation_104 (Activation)    (None, 35, 35, 96)   0           ['batch_normalization_104[0][0]']\n",
            "                                                                                                  \n",
            " activation_105 (Activation)    (None, 35, 35, 32)   0           ['batch_normalization_105[0][0]']\n",
            "                                                                                                  \n",
            " mixed0 (Concatenate)           (None, 35, 35, 256)  0           ['activation_99[0][0]',          \n",
            "                                                                  'activation_101[0][0]',         \n",
            "                                                                  'activation_104[0][0]',         \n",
            "                                                                  'activation_105[0][0]']         \n",
            "                                                                                                  \n",
            " conv2d_109 (Conv2D)            (None, 35, 35, 64)   16384       ['mixed0[0][0]']                 \n",
            "                                                                                                  \n",
            " batch_normalization_109 (Batch  (None, 35, 35, 64)  192         ['conv2d_109[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_109 (Activation)    (None, 35, 35, 64)   0           ['batch_normalization_109[0][0]']\n",
            "                                                                                                  \n",
            " conv2d_107 (Conv2D)            (None, 35, 35, 48)   12288       ['mixed0[0][0]']                 \n",
            "                                                                                                  \n",
            " conv2d_110 (Conv2D)            (None, 35, 35, 96)   55296       ['activation_109[0][0]']         \n",
            "                                                                                                  \n",
            " batch_normalization_107 (Batch  (None, 35, 35, 48)  144         ['conv2d_107[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_110 (Batch  (None, 35, 35, 96)  288         ['conv2d_110[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_107 (Activation)    (None, 35, 35, 48)   0           ['batch_normalization_107[0][0]']\n",
            "                                                                                                  \n",
            " activation_110 (Activation)    (None, 35, 35, 96)   0           ['batch_normalization_110[0][0]']\n",
            "                                                                                                  \n",
            " average_pooling2d_10 (AverageP  (None, 35, 35, 256)  0          ['mixed0[0][0]']                 \n",
            " ooling2D)                                                                                        \n",
            "                                                                                                  \n",
            " conv2d_106 (Conv2D)            (None, 35, 35, 64)   16384       ['mixed0[0][0]']                 \n",
            "                                                                                                  \n",
            " conv2d_108 (Conv2D)            (None, 35, 35, 64)   76800       ['activation_107[0][0]']         \n",
            "                                                                                                  \n",
            " conv2d_111 (Conv2D)            (None, 35, 35, 96)   82944       ['activation_110[0][0]']         \n",
            "                                                                                                  \n",
            " conv2d_112 (Conv2D)            (None, 35, 35, 64)   16384       ['average_pooling2d_10[0][0]']   \n",
            "                                                                                                  \n",
            " batch_normalization_106 (Batch  (None, 35, 35, 64)  192         ['conv2d_106[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_108 (Batch  (None, 35, 35, 64)  192         ['conv2d_108[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_111 (Batch  (None, 35, 35, 96)  288         ['conv2d_111[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_112 (Batch  (None, 35, 35, 64)  192         ['conv2d_112[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_106 (Activation)    (None, 35, 35, 64)   0           ['batch_normalization_106[0][0]']\n",
            "                                                                                                  \n",
            " activation_108 (Activation)    (None, 35, 35, 64)   0           ['batch_normalization_108[0][0]']\n",
            "                                                                                                  \n",
            " activation_111 (Activation)    (None, 35, 35, 96)   0           ['batch_normalization_111[0][0]']\n",
            "                                                                                                  \n",
            " activation_112 (Activation)    (None, 35, 35, 64)   0           ['batch_normalization_112[0][0]']\n",
            "                                                                                                  \n",
            " mixed1 (Concatenate)           (None, 35, 35, 288)  0           ['activation_106[0][0]',         \n",
            "                                                                  'activation_108[0][0]',         \n",
            "                                                                  'activation_111[0][0]',         \n",
            "                                                                  'activation_112[0][0]']         \n",
            "                                                                                                  \n",
            " conv2d_116 (Conv2D)            (None, 35, 35, 64)   18432       ['mixed1[0][0]']                 \n",
            "                                                                                                  \n",
            " batch_normalization_116 (Batch  (None, 35, 35, 64)  192         ['conv2d_116[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_116 (Activation)    (None, 35, 35, 64)   0           ['batch_normalization_116[0][0]']\n",
            "                                                                                                  \n",
            " conv2d_114 (Conv2D)            (None, 35, 35, 48)   13824       ['mixed1[0][0]']                 \n",
            "                                                                                                  \n",
            " conv2d_117 (Conv2D)            (None, 35, 35, 96)   55296       ['activation_116[0][0]']         \n",
            "                                                                                                  \n",
            " batch_normalization_114 (Batch  (None, 35, 35, 48)  144         ['conv2d_114[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_117 (Batch  (None, 35, 35, 96)  288         ['conv2d_117[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_114 (Activation)    (None, 35, 35, 48)   0           ['batch_normalization_114[0][0]']\n",
            "                                                                                                  \n",
            " activation_117 (Activation)    (None, 35, 35, 96)   0           ['batch_normalization_117[0][0]']\n",
            "                                                                                                  \n",
            " average_pooling2d_11 (AverageP  (None, 35, 35, 288)  0          ['mixed1[0][0]']                 \n",
            " ooling2D)                                                                                        \n",
            "                                                                                                  \n",
            " conv2d_113 (Conv2D)            (None, 35, 35, 64)   18432       ['mixed1[0][0]']                 \n",
            "                                                                                                  \n",
            " conv2d_115 (Conv2D)            (None, 35, 35, 64)   76800       ['activation_114[0][0]']         \n",
            "                                                                                                  \n",
            " conv2d_118 (Conv2D)            (None, 35, 35, 96)   82944       ['activation_117[0][0]']         \n",
            "                                                                                                  \n",
            " conv2d_119 (Conv2D)            (None, 35, 35, 64)   18432       ['average_pooling2d_11[0][0]']   \n",
            "                                                                                                  \n",
            " batch_normalization_113 (Batch  (None, 35, 35, 64)  192         ['conv2d_113[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_115 (Batch  (None, 35, 35, 64)  192         ['conv2d_115[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_118 (Batch  (None, 35, 35, 96)  288         ['conv2d_118[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_119 (Batch  (None, 35, 35, 64)  192         ['conv2d_119[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_113 (Activation)    (None, 35, 35, 64)   0           ['batch_normalization_113[0][0]']\n",
            "                                                                                                  \n",
            " activation_115 (Activation)    (None, 35, 35, 64)   0           ['batch_normalization_115[0][0]']\n",
            "                                                                                                  \n",
            " activation_118 (Activation)    (None, 35, 35, 96)   0           ['batch_normalization_118[0][0]']\n",
            "                                                                                                  \n",
            " activation_119 (Activation)    (None, 35, 35, 64)   0           ['batch_normalization_119[0][0]']\n",
            "                                                                                                  \n",
            " mixed2 (Concatenate)           (None, 35, 35, 288)  0           ['activation_113[0][0]',         \n",
            "                                                                  'activation_115[0][0]',         \n",
            "                                                                  'activation_118[0][0]',         \n",
            "                                                                  'activation_119[0][0]']         \n",
            "                                                                                                  \n",
            " conv2d_121 (Conv2D)            (None, 35, 35, 64)   18432       ['mixed2[0][0]']                 \n",
            "                                                                                                  \n",
            " batch_normalization_121 (Batch  (None, 35, 35, 64)  192         ['conv2d_121[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_121 (Activation)    (None, 35, 35, 64)   0           ['batch_normalization_121[0][0]']\n",
            "                                                                                                  \n",
            " conv2d_122 (Conv2D)            (None, 35, 35, 96)   55296       ['activation_121[0][0]']         \n",
            "                                                                                                  \n",
            " batch_normalization_122 (Batch  (None, 35, 35, 96)  288         ['conv2d_122[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_122 (Activation)    (None, 35, 35, 96)   0           ['batch_normalization_122[0][0]']\n",
            "                                                                                                  \n",
            " conv2d_120 (Conv2D)            (None, 17, 17, 384)  995328      ['mixed2[0][0]']                 \n",
            "                                                                                                  \n",
            " conv2d_123 (Conv2D)            (None, 17, 17, 96)   82944       ['activation_122[0][0]']         \n",
            "                                                                                                  \n",
            " batch_normalization_120 (Batch  (None, 17, 17, 384)  1152       ['conv2d_120[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_123 (Batch  (None, 17, 17, 96)  288         ['conv2d_123[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_120 (Activation)    (None, 17, 17, 384)  0           ['batch_normalization_120[0][0]']\n",
            "                                                                                                  \n",
            " activation_123 (Activation)    (None, 17, 17, 96)   0           ['batch_normalization_123[0][0]']\n",
            "                                                                                                  \n",
            " max_pooling2d_6 (MaxPooling2D)  (None, 17, 17, 288)  0          ['mixed2[0][0]']                 \n",
            "                                                                                                  \n",
            " mixed3 (Concatenate)           (None, 17, 17, 768)  0           ['activation_120[0][0]',         \n",
            "                                                                  'activation_123[0][0]',         \n",
            "                                                                  'max_pooling2d_6[0][0]']        \n",
            "                                                                                                  \n",
            " conv2d_128 (Conv2D)            (None, 17, 17, 128)  98304       ['mixed3[0][0]']                 \n",
            "                                                                                                  \n",
            " batch_normalization_128 (Batch  (None, 17, 17, 128)  384        ['conv2d_128[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_128 (Activation)    (None, 17, 17, 128)  0           ['batch_normalization_128[0][0]']\n",
            "                                                                                                  \n",
            " conv2d_129 (Conv2D)            (None, 17, 17, 128)  114688      ['activation_128[0][0]']         \n",
            "                                                                                                  \n",
            " batch_normalization_129 (Batch  (None, 17, 17, 128)  384        ['conv2d_129[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_129 (Activation)    (None, 17, 17, 128)  0           ['batch_normalization_129[0][0]']\n",
            "                                                                                                  \n",
            " conv2d_125 (Conv2D)            (None, 17, 17, 128)  98304       ['mixed3[0][0]']                 \n",
            "                                                                                                  \n",
            " conv2d_130 (Conv2D)            (None, 17, 17, 128)  114688      ['activation_129[0][0]']         \n",
            "                                                                                                  \n",
            " batch_normalization_125 (Batch  (None, 17, 17, 128)  384        ['conv2d_125[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_130 (Batch  (None, 17, 17, 128)  384        ['conv2d_130[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_125 (Activation)    (None, 17, 17, 128)  0           ['batch_normalization_125[0][0]']\n",
            "                                                                                                  \n",
            " activation_130 (Activation)    (None, 17, 17, 128)  0           ['batch_normalization_130[0][0]']\n",
            "                                                                                                  \n",
            " conv2d_126 (Conv2D)            (None, 17, 17, 128)  114688      ['activation_125[0][0]']         \n",
            "                                                                                                  \n",
            " conv2d_131 (Conv2D)            (None, 17, 17, 128)  114688      ['activation_130[0][0]']         \n",
            "                                                                                                  \n",
            " batch_normalization_126 (Batch  (None, 17, 17, 128)  384        ['conv2d_126[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_131 (Batch  (None, 17, 17, 128)  384        ['conv2d_131[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_126 (Activation)    (None, 17, 17, 128)  0           ['batch_normalization_126[0][0]']\n",
            "                                                                                                  \n",
            " activation_131 (Activation)    (None, 17, 17, 128)  0           ['batch_normalization_131[0][0]']\n",
            "                                                                                                  \n",
            " average_pooling2d_12 (AverageP  (None, 17, 17, 768)  0          ['mixed3[0][0]']                 \n",
            " ooling2D)                                                                                        \n",
            "                                                                                                  \n",
            " conv2d_124 (Conv2D)            (None, 17, 17, 192)  147456      ['mixed3[0][0]']                 \n",
            "                                                                                                  \n",
            " conv2d_127 (Conv2D)            (None, 17, 17, 192)  172032      ['activation_126[0][0]']         \n",
            "                                                                                                  \n",
            " conv2d_132 (Conv2D)            (None, 17, 17, 192)  172032      ['activation_131[0][0]']         \n",
            "                                                                                                  \n",
            " conv2d_133 (Conv2D)            (None, 17, 17, 192)  147456      ['average_pooling2d_12[0][0]']   \n",
            "                                                                                                  \n",
            " batch_normalization_124 (Batch  (None, 17, 17, 192)  576        ['conv2d_124[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_127 (Batch  (None, 17, 17, 192)  576        ['conv2d_127[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_132 (Batch  (None, 17, 17, 192)  576        ['conv2d_132[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_133 (Batch  (None, 17, 17, 192)  576        ['conv2d_133[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_124 (Activation)    (None, 17, 17, 192)  0           ['batch_normalization_124[0][0]']\n",
            "                                                                                                  \n",
            " activation_127 (Activation)    (None, 17, 17, 192)  0           ['batch_normalization_127[0][0]']\n",
            "                                                                                                  \n",
            " activation_132 (Activation)    (None, 17, 17, 192)  0           ['batch_normalization_132[0][0]']\n",
            "                                                                                                  \n",
            " activation_133 (Activation)    (None, 17, 17, 192)  0           ['batch_normalization_133[0][0]']\n",
            "                                                                                                  \n",
            " mixed4 (Concatenate)           (None, 17, 17, 768)  0           ['activation_124[0][0]',         \n",
            "                                                                  'activation_127[0][0]',         \n",
            "                                                                  'activation_132[0][0]',         \n",
            "                                                                  'activation_133[0][0]']         \n",
            "                                                                                                  \n",
            " conv2d_138 (Conv2D)            (None, 17, 17, 160)  122880      ['mixed4[0][0]']                 \n",
            "                                                                                                  \n",
            " batch_normalization_138 (Batch  (None, 17, 17, 160)  480        ['conv2d_138[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_138 (Activation)    (None, 17, 17, 160)  0           ['batch_normalization_138[0][0]']\n",
            "                                                                                                  \n",
            " conv2d_139 (Conv2D)            (None, 17, 17, 160)  179200      ['activation_138[0][0]']         \n",
            "                                                                                                  \n",
            " batch_normalization_139 (Batch  (None, 17, 17, 160)  480        ['conv2d_139[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_139 (Activation)    (None, 17, 17, 160)  0           ['batch_normalization_139[0][0]']\n",
            "                                                                                                  \n",
            " conv2d_135 (Conv2D)            (None, 17, 17, 160)  122880      ['mixed4[0][0]']                 \n",
            "                                                                                                  \n",
            " conv2d_140 (Conv2D)            (None, 17, 17, 160)  179200      ['activation_139[0][0]']         \n",
            "                                                                                                  \n",
            " batch_normalization_135 (Batch  (None, 17, 17, 160)  480        ['conv2d_135[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_140 (Batch  (None, 17, 17, 160)  480        ['conv2d_140[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_135 (Activation)    (None, 17, 17, 160)  0           ['batch_normalization_135[0][0]']\n",
            "                                                                                                  \n",
            " activation_140 (Activation)    (None, 17, 17, 160)  0           ['batch_normalization_140[0][0]']\n",
            "                                                                                                  \n",
            " conv2d_136 (Conv2D)            (None, 17, 17, 160)  179200      ['activation_135[0][0]']         \n",
            "                                                                                                  \n",
            " conv2d_141 (Conv2D)            (None, 17, 17, 160)  179200      ['activation_140[0][0]']         \n",
            "                                                                                                  \n",
            " batch_normalization_136 (Batch  (None, 17, 17, 160)  480        ['conv2d_136[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_141 (Batch  (None, 17, 17, 160)  480        ['conv2d_141[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_136 (Activation)    (None, 17, 17, 160)  0           ['batch_normalization_136[0][0]']\n",
            "                                                                                                  \n",
            " activation_141 (Activation)    (None, 17, 17, 160)  0           ['batch_normalization_141[0][0]']\n",
            "                                                                                                  \n",
            " average_pooling2d_13 (AverageP  (None, 17, 17, 768)  0          ['mixed4[0][0]']                 \n",
            " ooling2D)                                                                                        \n",
            "                                                                                                  \n",
            " conv2d_134 (Conv2D)            (None, 17, 17, 192)  147456      ['mixed4[0][0]']                 \n",
            "                                                                                                  \n",
            " conv2d_137 (Conv2D)            (None, 17, 17, 192)  215040      ['activation_136[0][0]']         \n",
            "                                                                                                  \n",
            " conv2d_142 (Conv2D)            (None, 17, 17, 192)  215040      ['activation_141[0][0]']         \n",
            "                                                                                                  \n",
            " conv2d_143 (Conv2D)            (None, 17, 17, 192)  147456      ['average_pooling2d_13[0][0]']   \n",
            "                                                                                                  \n",
            " batch_normalization_134 (Batch  (None, 17, 17, 192)  576        ['conv2d_134[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_137 (Batch  (None, 17, 17, 192)  576        ['conv2d_137[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_142 (Batch  (None, 17, 17, 192)  576        ['conv2d_142[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_143 (Batch  (None, 17, 17, 192)  576        ['conv2d_143[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_134 (Activation)    (None, 17, 17, 192)  0           ['batch_normalization_134[0][0]']\n",
            "                                                                                                  \n",
            " activation_137 (Activation)    (None, 17, 17, 192)  0           ['batch_normalization_137[0][0]']\n",
            "                                                                                                  \n",
            " activation_142 (Activation)    (None, 17, 17, 192)  0           ['batch_normalization_142[0][0]']\n",
            "                                                                                                  \n",
            " activation_143 (Activation)    (None, 17, 17, 192)  0           ['batch_normalization_143[0][0]']\n",
            "                                                                                                  \n",
            " mixed5 (Concatenate)           (None, 17, 17, 768)  0           ['activation_134[0][0]',         \n",
            "                                                                  'activation_137[0][0]',         \n",
            "                                                                  'activation_142[0][0]',         \n",
            "                                                                  'activation_143[0][0]']         \n",
            "                                                                                                  \n",
            " conv2d_148 (Conv2D)            (None, 17, 17, 160)  122880      ['mixed5[0][0]']                 \n",
            "                                                                                                  \n",
            " batch_normalization_148 (Batch  (None, 17, 17, 160)  480        ['conv2d_148[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_148 (Activation)    (None, 17, 17, 160)  0           ['batch_normalization_148[0][0]']\n",
            "                                                                                                  \n",
            " conv2d_149 (Conv2D)            (None, 17, 17, 160)  179200      ['activation_148[0][0]']         \n",
            "                                                                                                  \n",
            " batch_normalization_149 (Batch  (None, 17, 17, 160)  480        ['conv2d_149[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_149 (Activation)    (None, 17, 17, 160)  0           ['batch_normalization_149[0][0]']\n",
            "                                                                                                  \n",
            " conv2d_145 (Conv2D)            (None, 17, 17, 160)  122880      ['mixed5[0][0]']                 \n",
            "                                                                                                  \n",
            " conv2d_150 (Conv2D)            (None, 17, 17, 160)  179200      ['activation_149[0][0]']         \n",
            "                                                                                                  \n",
            " batch_normalization_145 (Batch  (None, 17, 17, 160)  480        ['conv2d_145[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_150 (Batch  (None, 17, 17, 160)  480        ['conv2d_150[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_145 (Activation)    (None, 17, 17, 160)  0           ['batch_normalization_145[0][0]']\n",
            "                                                                                                  \n",
            " activation_150 (Activation)    (None, 17, 17, 160)  0           ['batch_normalization_150[0][0]']\n",
            "                                                                                                  \n",
            " conv2d_146 (Conv2D)            (None, 17, 17, 160)  179200      ['activation_145[0][0]']         \n",
            "                                                                                                  \n",
            " conv2d_151 (Conv2D)            (None, 17, 17, 160)  179200      ['activation_150[0][0]']         \n",
            "                                                                                                  \n",
            " batch_normalization_146 (Batch  (None, 17, 17, 160)  480        ['conv2d_146[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_151 (Batch  (None, 17, 17, 160)  480        ['conv2d_151[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_146 (Activation)    (None, 17, 17, 160)  0           ['batch_normalization_146[0][0]']\n",
            "                                                                                                  \n",
            " activation_151 (Activation)    (None, 17, 17, 160)  0           ['batch_normalization_151[0][0]']\n",
            "                                                                                                  \n",
            " average_pooling2d_14 (AverageP  (None, 17, 17, 768)  0          ['mixed5[0][0]']                 \n",
            " ooling2D)                                                                                        \n",
            "                                                                                                  \n",
            " conv2d_144 (Conv2D)            (None, 17, 17, 192)  147456      ['mixed5[0][0]']                 \n",
            "                                                                                                  \n",
            " conv2d_147 (Conv2D)            (None, 17, 17, 192)  215040      ['activation_146[0][0]']         \n",
            "                                                                                                  \n",
            " conv2d_152 (Conv2D)            (None, 17, 17, 192)  215040      ['activation_151[0][0]']         \n",
            "                                                                                                  \n",
            " conv2d_153 (Conv2D)            (None, 17, 17, 192)  147456      ['average_pooling2d_14[0][0]']   \n",
            "                                                                                                  \n",
            " batch_normalization_144 (Batch  (None, 17, 17, 192)  576        ['conv2d_144[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_147 (Batch  (None, 17, 17, 192)  576        ['conv2d_147[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_152 (Batch  (None, 17, 17, 192)  576        ['conv2d_152[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_153 (Batch  (None, 17, 17, 192)  576        ['conv2d_153[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_144 (Activation)    (None, 17, 17, 192)  0           ['batch_normalization_144[0][0]']\n",
            "                                                                                                  \n",
            " activation_147 (Activation)    (None, 17, 17, 192)  0           ['batch_normalization_147[0][0]']\n",
            "                                                                                                  \n",
            " activation_152 (Activation)    (None, 17, 17, 192)  0           ['batch_normalization_152[0][0]']\n",
            "                                                                                                  \n",
            " activation_153 (Activation)    (None, 17, 17, 192)  0           ['batch_normalization_153[0][0]']\n",
            "                                                                                                  \n",
            " mixed6 (Concatenate)           (None, 17, 17, 768)  0           ['activation_144[0][0]',         \n",
            "                                                                  'activation_147[0][0]',         \n",
            "                                                                  'activation_152[0][0]',         \n",
            "                                                                  'activation_153[0][0]']         \n",
            "                                                                                                  \n",
            " conv2d_158 (Conv2D)            (None, 17, 17, 192)  147456      ['mixed6[0][0]']                 \n",
            "                                                                                                  \n",
            " batch_normalization_158 (Batch  (None, 17, 17, 192)  576        ['conv2d_158[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_158 (Activation)    (None, 17, 17, 192)  0           ['batch_normalization_158[0][0]']\n",
            "                                                                                                  \n",
            " conv2d_159 (Conv2D)            (None, 17, 17, 192)  258048      ['activation_158[0][0]']         \n",
            "                                                                                                  \n",
            " batch_normalization_159 (Batch  (None, 17, 17, 192)  576        ['conv2d_159[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_159 (Activation)    (None, 17, 17, 192)  0           ['batch_normalization_159[0][0]']\n",
            "                                                                                                  \n",
            " conv2d_155 (Conv2D)            (None, 17, 17, 192)  147456      ['mixed6[0][0]']                 \n",
            "                                                                                                  \n",
            " conv2d_160 (Conv2D)            (None, 17, 17, 192)  258048      ['activation_159[0][0]']         \n",
            "                                                                                                  \n",
            " batch_normalization_155 (Batch  (None, 17, 17, 192)  576        ['conv2d_155[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_160 (Batch  (None, 17, 17, 192)  576        ['conv2d_160[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_155 (Activation)    (None, 17, 17, 192)  0           ['batch_normalization_155[0][0]']\n",
            "                                                                                                  \n",
            " activation_160 (Activation)    (None, 17, 17, 192)  0           ['batch_normalization_160[0][0]']\n",
            "                                                                                                  \n",
            " conv2d_156 (Conv2D)            (None, 17, 17, 192)  258048      ['activation_155[0][0]']         \n",
            "                                                                                                  \n",
            " conv2d_161 (Conv2D)            (None, 17, 17, 192)  258048      ['activation_160[0][0]']         \n",
            "                                                                                                  \n",
            " batch_normalization_156 (Batch  (None, 17, 17, 192)  576        ['conv2d_156[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_161 (Batch  (None, 17, 17, 192)  576        ['conv2d_161[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_156 (Activation)    (None, 17, 17, 192)  0           ['batch_normalization_156[0][0]']\n",
            "                                                                                                  \n",
            " activation_161 (Activation)    (None, 17, 17, 192)  0           ['batch_normalization_161[0][0]']\n",
            "                                                                                                  \n",
            " average_pooling2d_15 (AverageP  (None, 17, 17, 768)  0          ['mixed6[0][0]']                 \n",
            " ooling2D)                                                                                        \n",
            "                                                                                                  \n",
            " conv2d_154 (Conv2D)            (None, 17, 17, 192)  147456      ['mixed6[0][0]']                 \n",
            "                                                                                                  \n",
            " conv2d_157 (Conv2D)            (None, 17, 17, 192)  258048      ['activation_156[0][0]']         \n",
            "                                                                                                  \n",
            " conv2d_162 (Conv2D)            (None, 17, 17, 192)  258048      ['activation_161[0][0]']         \n",
            "                                                                                                  \n",
            " conv2d_163 (Conv2D)            (None, 17, 17, 192)  147456      ['average_pooling2d_15[0][0]']   \n",
            "                                                                                                  \n",
            " batch_normalization_154 (Batch  (None, 17, 17, 192)  576        ['conv2d_154[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_157 (Batch  (None, 17, 17, 192)  576        ['conv2d_157[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_162 (Batch  (None, 17, 17, 192)  576        ['conv2d_162[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_163 (Batch  (None, 17, 17, 192)  576        ['conv2d_163[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_154 (Activation)    (None, 17, 17, 192)  0           ['batch_normalization_154[0][0]']\n",
            "                                                                                                  \n",
            " activation_157 (Activation)    (None, 17, 17, 192)  0           ['batch_normalization_157[0][0]']\n",
            "                                                                                                  \n",
            " activation_162 (Activation)    (None, 17, 17, 192)  0           ['batch_normalization_162[0][0]']\n",
            "                                                                                                  \n",
            " activation_163 (Activation)    (None, 17, 17, 192)  0           ['batch_normalization_163[0][0]']\n",
            "                                                                                                  \n",
            " mixed7 (Concatenate)           (None, 17, 17, 768)  0           ['activation_154[0][0]',         \n",
            "                                                                  'activation_157[0][0]',         \n",
            "                                                                  'activation_162[0][0]',         \n",
            "                                                                  'activation_163[0][0]']         \n",
            "                                                                                                  \n",
            " conv2d_166 (Conv2D)            (None, 17, 17, 192)  147456      ['mixed7[0][0]']                 \n",
            "                                                                                                  \n",
            " batch_normalization_166 (Batch  (None, 17, 17, 192)  576        ['conv2d_166[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_166 (Activation)    (None, 17, 17, 192)  0           ['batch_normalization_166[0][0]']\n",
            "                                                                                                  \n",
            " conv2d_167 (Conv2D)            (None, 17, 17, 192)  258048      ['activation_166[0][0]']         \n",
            "                                                                                                  \n",
            " batch_normalization_167 (Batch  (None, 17, 17, 192)  576        ['conv2d_167[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_167 (Activation)    (None, 17, 17, 192)  0           ['batch_normalization_167[0][0]']\n",
            "                                                                                                  \n",
            " conv2d_164 (Conv2D)            (None, 17, 17, 192)  147456      ['mixed7[0][0]']                 \n",
            "                                                                                                  \n",
            " conv2d_168 (Conv2D)            (None, 17, 17, 192)  258048      ['activation_167[0][0]']         \n",
            "                                                                                                  \n",
            " batch_normalization_164 (Batch  (None, 17, 17, 192)  576        ['conv2d_164[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_168 (Batch  (None, 17, 17, 192)  576        ['conv2d_168[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_164 (Activation)    (None, 17, 17, 192)  0           ['batch_normalization_164[0][0]']\n",
            "                                                                                                  \n",
            " activation_168 (Activation)    (None, 17, 17, 192)  0           ['batch_normalization_168[0][0]']\n",
            "                                                                                                  \n",
            " conv2d_165 (Conv2D)            (None, 8, 8, 320)    552960      ['activation_164[0][0]']         \n",
            "                                                                                                  \n",
            " conv2d_169 (Conv2D)            (None, 8, 8, 192)    331776      ['activation_168[0][0]']         \n",
            "                                                                                                  \n",
            " batch_normalization_165 (Batch  (None, 8, 8, 320)   960         ['conv2d_165[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " batch_normalization_169 (Batch  (None, 8, 8, 192)   576         ['conv2d_169[0][0]']             \n",
            " Normalization)                                                                                   \n",
            "                                                                                                  \n",
            " activation_165 (Activation)    (None, 8, 8, 320)    0           ['batch_normalization_165[0][0]']\n",
            "                                                                                                  \n",
            " activation_169 (Activation)    (None, 8, 8, 192)    0           ['batch_normalization_169[0][0]']\n",
            "                                                                                                  \n",
            " max_pooling2d_7 (MaxPooling2D)  (None, 8, 8, 768)   0           ['mixed7[0][0]']                 \n",
            "                                                                                                  \n",
            " mixed8 (Concatenate)           (None, 8, 8, 1280)   0           ['activation_165[0][0]',         \n",
            "                                                                  'activation_169[0][0]',         \n",
            "                                                                  'max_pooling2d_7[0][0]']        \n",
            "                                                                                                  \n",
            " flatten_2 (Flatten)            (None, 81920)        0           ['mixed8[0][0]']                 \n",
            "                                                                                                  \n",
            " dense_3 (Dense)                (None, 256)          20971776    ['flatten_2[0][0]']              \n",
            "                                                                                                  \n",
            " dropout_2 (Dropout)            (None, 256)          0           ['dense_3[0][0]']                \n",
            "                                                                                                  \n",
            " dense_4 (Dense)                (None, 3)            771         ['dropout_2[0][0]']              \n",
            "                                                                                                  \n",
            "==================================================================================================\n",
            "Total params: 31,647,395\n",
            "Trainable params: 20,972,547\n",
            "Non-trainable params: 10,674,848\n",
            "__________________________________________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model.compile(tf.keras.optimizers.RMSprop(learning_rate = 0.005),loss = \"categorical_crossentropy\",metrics = [\"acc\"])\n",
        "class myCallback(tf.keras.callbacks.Callback):\n",
        "    def on_epoch_end(self,epoch,logs = {}):\n",
        "        if logs.get(\"acc\") > 0.95:\n",
        "            print(\"Accuracy reached %95.Stop training.\")\n",
        "            self.model.stop_training = True \n",
        "\n",
        "callback = myCallback()"
      ],
      "metadata": {
        "id": "-iCRHPY5IRff"
      },
      "execution_count": 35,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "history = model.fit(train_generator,epochs = 30,batch_size = 4,validation_data = validation_generator,verbose = 1,\n",
        "                   callbacks = [callback])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "C3TUTnm5IcM3",
        "outputId": "39a817aa-44c1-4614-e80e-5a24a977f177"
      },
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/30\n",
            "46/46 [==============================] - 63s 1s/step - loss: 81.8965 - acc: 0.4641 - val_loss: 3.7992 - val_acc: 0.8409\n",
            "Epoch 2/30\n",
            "46/46 [==============================] - 56s 1s/step - loss: 15.6730 - acc: 0.6354 - val_loss: 1.7219 - val_acc: 0.8864\n",
            "Epoch 3/30\n",
            "46/46 [==============================] - 57s 1s/step - loss: 4.2233 - acc: 0.6961 - val_loss: 0.7431 - val_acc: 0.8636\n",
            "Epoch 4/30\n",
            "46/46 [==============================] - 56s 1s/step - loss: 2.8756 - acc: 0.7901 - val_loss: 1.3197 - val_acc: 0.8409\n",
            "Epoch 5/30\n",
            "46/46 [==============================] - 56s 1s/step - loss: 3.5675 - acc: 0.7790 - val_loss: 0.8262 - val_acc: 0.8636\n",
            "Epoch 6/30\n",
            "46/46 [==============================] - 56s 1s/step - loss: 1.6727 - acc: 0.7403 - val_loss: 0.8711 - val_acc: 0.7727\n",
            "Epoch 7/30\n",
            "46/46 [==============================] - 57s 1s/step - loss: 1.3955 - acc: 0.7901 - val_loss: 0.2908 - val_acc: 0.9091\n",
            "Epoch 8/30\n",
            "46/46 [==============================] - 55s 1s/step - loss: 2.7754 - acc: 0.7956 - val_loss: 0.9050 - val_acc: 0.8636\n",
            "Epoch 9/30\n",
            "46/46 [==============================] - 55s 1s/step - loss: 2.4386 - acc: 0.8287 - val_loss: 0.7341 - val_acc: 0.8182\n",
            "Epoch 10/30\n",
            "46/46 [==============================] - 55s 1s/step - loss: 1.0973 - acc: 0.8564 - val_loss: 0.6366 - val_acc: 0.9091\n",
            "Epoch 11/30\n",
            "46/46 [==============================] - 56s 1s/step - loss: 1.5719 - acc: 0.8398 - val_loss: 0.7108 - val_acc: 0.9091\n",
            "Epoch 12/30\n",
            "46/46 [==============================] - 55s 1s/step - loss: 1.6572 - acc: 0.8785 - val_loss: 0.3317 - val_acc: 0.9091\n",
            "Epoch 13/30\n",
            "46/46 [==============================] - 56s 1s/step - loss: 1.9617 - acc: 0.8343 - val_loss: 0.2834 - val_acc: 0.9318\n",
            "Epoch 14/30\n",
            "46/46 [==============================] - 56s 1s/step - loss: 2.4536 - acc: 0.8564 - val_loss: 2.0322 - val_acc: 0.7955\n",
            "Epoch 15/30\n",
            "46/46 [==============================] - 55s 1s/step - loss: 1.6296 - acc: 0.8785 - val_loss: 0.2790 - val_acc: 0.8864\n",
            "Epoch 16/30\n",
            "46/46 [==============================] - 55s 1s/step - loss: 0.8457 - acc: 0.8895 - val_loss: 0.8317 - val_acc: 0.8636\n",
            "Epoch 17/30\n",
            "46/46 [==============================] - 55s 1s/step - loss: 1.1772 - acc: 0.8619 - val_loss: 1.2972 - val_acc: 0.8864\n",
            "Epoch 18/30\n",
            "46/46 [==============================] - 57s 1s/step - loss: 0.5334 - acc: 0.9116 - val_loss: 0.2694 - val_acc: 0.9091\n",
            "Epoch 19/30\n",
            "46/46 [==============================] - 55s 1s/step - loss: 2.1451 - acc: 0.8508 - val_loss: 0.9729 - val_acc: 0.9091\n",
            "Epoch 20/30\n",
            "46/46 [==============================] - 56s 1s/step - loss: 1.0196 - acc: 0.8895 - val_loss: 0.7426 - val_acc: 0.8864\n",
            "Epoch 21/30\n",
            "46/46 [==============================] - 55s 1s/step - loss: 0.6644 - acc: 0.9061 - val_loss: 1.2278 - val_acc: 0.8182\n",
            "Epoch 22/30\n",
            "46/46 [==============================] - 56s 1s/step - loss: 0.6716 - acc: 0.9227 - val_loss: 0.3768 - val_acc: 0.9545\n",
            "Epoch 23/30\n",
            "46/46 [==============================] - 55s 1s/step - loss: 1.0014 - acc: 0.9061 - val_loss: 3.2191 - val_acc: 0.8636\n",
            "Epoch 24/30\n",
            "46/46 [==============================] - 55s 1s/step - loss: 0.8737 - acc: 0.9061 - val_loss: 0.4100 - val_acc: 0.9773\n",
            "Epoch 25/30\n",
            "46/46 [==============================] - 55s 1s/step - loss: 0.6892 - acc: 0.9061 - val_loss: 1.8766 - val_acc: 0.8636\n",
            "Epoch 26/30\n",
            "46/46 [==============================] - 54s 1s/step - loss: 0.4655 - acc: 0.9448 - val_loss: 1.1847 - val_acc: 0.8864\n",
            "Epoch 27/30\n",
            "46/46 [==============================] - 55s 1s/step - loss: 0.6789 - acc: 0.9448 - val_loss: 1.1848 - val_acc: 0.9091\n",
            "Epoch 28/30\n",
            "46/46 [==============================] - 54s 1s/step - loss: 1.4915 - acc: 0.9116 - val_loss: 0.8701 - val_acc: 0.9091\n",
            "Epoch 29/30\n",
            "46/46 [==============================] - 55s 1s/step - loss: 0.6117 - acc: 0.9392 - val_loss: 1.3565 - val_acc: 0.9318\n",
            "Epoch 30/30\n",
            "46/46 [==============================] - 54s 1s/step - loss: 1.0765 - acc: 0.9392 - val_loss: 1.4122 - val_acc: 0.9091\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\"\"\"Plot accuracy\"\"\"\n",
        "acc = history.history[\"acc\"]\n",
        "val_acc = history.history[\"val_acc\"]\n",
        "epochs = range(30)\n",
        "plt.plot(epochs,acc,label = \"Training Accuracy\")\n",
        "plt.plot(epochs,val_acc,label = \"Validation Accuracy\")\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 268
        },
        "id": "-zihDle9TFqv",
        "outputId": "fd70fa29-485f-448f-acb7-48f5b7cf5b74"
      },
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "loss = history.history[\"loss\"]\n",
        "val_loss = history.history[\"val_loss\"]\n",
        "epochs = range(30)\n",
        "plt.plot(epochs,loss,label = \"Training Loss\")\n",
        "plt.plot(epochs,val_loss,label = \"Validation Loss\")\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "2j-ndgQkTIMd",
        "outputId": "95bca0ea-c536-4654-f7a0-b9da489886a4"
      },
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model.save(\"pizza_steak_sushitfl.h5\")\n",
        "model1 = tf.keras.models.load_model(\"pizza_steak_sushitfl.h5\")"
      ],
      "metadata": {
        "id": "N2cRq5vtTPsn"
      },
      "execution_count": 40,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from PIL import Image\n",
        "import requests\n",
        "from io import BytesIO\n",
        "import numpy as np\n",
        "\n",
        "def get_and_process(url):\n",
        "        response = requests.get(url)\n",
        "        img = Image.open(BytesIO(response.content))\n",
        "        img1 = img\n",
        "        \"\"\"Resize image for appropriate shape for model.\"\"\"\n",
        "        img = img.resize((300,300))\n",
        "        \"\"\"Convert img to numpy array,rescale it ,expand dims for model convenience. then check vertically.\"\"\"\n",
        "        x = tf.keras.preprocessing.image.img_to_array(img)\n",
        "        x = x / 255.0\n",
        "        x = np.expand_dims(x,axis = 0)\n",
        "        img_tensor = np.vstack([x])\n",
        "        return img1,img_tensor"
      ],
      "metadata": {
        "id": "OoJ4aENZTmyT"
      },
      "execution_count": 41,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "url = \"https://image.shutterstock.com/image-photo/blurred-abstract-background-pizza-260nw-460652131.jpg\"\n",
        "img1,test_img = get_and_process(url)\n",
        "pred = model1.predict(test_img)\n",
        "classes = list(train_generator.class_indices.keys())\n",
        "print(f\"Prediction is : {classes[np.argmax(pred)]}\")\n",
        "plt.imshow(img1)\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 286
        },
        "id": "wovyLBBFTo34",
        "outputId": "a1ab3295-2db5-4d48-94c7-4eedd42e8ea6"
      },
      "execution_count": 47,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Prediction is : pizza\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}