Logo_MobileNetV1.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "accelerator": "TPU",
    "colab": {
      "name": "Logo MobileNetV1.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "machine_shape": "hm"
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tAb77yZ9fzMG",
        "outputId": "ec793ffa-d2d5-4a61-cf30-704ec105c374"
      },
      "source": [
        "from google.colab import drive\r\n",
        "drive.mount('/content/drive/',force_remount=True)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Mounted at /content/drive/\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "pafL7Li0jyXW",
        "outputId": "a13811a0-4492-4bc7-fff6-724512dcdd69"
      },
      "source": [
        "from keras.layers import Input, Lambda, Dense, Flatten\r\n",
        "from keras.models import Model\r\n",
        "from keras.applications.mobilenet import MobileNet\r\n",
        "from keras.applications.mobilenet import preprocess_input\r\n",
        "from keras.preprocessing import image\r\n",
        "from keras.preprocessing.image import ImageDataGenerator\r\n",
        "from keras.models import Sequential\r\n",
        "import numpy as np\r\n",
        "from glob import glob\r\n",
        "import matplotlib.pyplot as plt\r\n",
        "from sklearn.model_selection import train_test_split\r\n",
        "from tensorflow.keras.callbacks import EarlyStopping\r\n",
        "from tensorflow.keras.callbacks import ModelCheckpoint\r\n",
        "from keras.models import load_model\r\n",
        "\r\n",
        "IMAGE_SIZE = [224, 224]\r\n",
        "train_path = '/content/drive/MyDrive/TrainData'\r\n",
        "valid_path = '/content/drive/MyDrive/Test another'\r\n",
        "\r\n",
        "mobilenet = MobileNet(input_shape=IMAGE_SIZE + [3] , weights='imagenet', include_top=False)\r\n",
        "\r\n",
        "for layer in mobilenet.layers:\r\n",
        "  layer.trainable = False\r\n",
        "  \r\n",
        "folders = glob('/content/drive/MyDrive/TrainData/*')\r\n",
        "print(\"Total number of classes are\",len(folders))  \r\n",
        "\r\n",
        "# our layers can add more if you want\r\n",
        "x = Flatten()(mobilenet.output)\r\n",
        "\r\n",
        "prediction = Dense(len(folders), activation='softmax')(x)\r\n",
        "\r\n",
        "model = Model(inputs=mobilenet.input, outputs=prediction)\r\n",
        "\r\n",
        "model.summary()\r\n",
        "\r\n",
        "model.compile(loss='categorical_crossentropy',optimizer='adam',metrics=['accuracy'])\r\n",
        "\r\n",
        "\r\n",
        "from keras.preprocessing.image import ImageDataGenerator\r\n",
        "\r\n",
        "train_datagen = ImageDataGenerator(rescale = 1./255,\r\n",
        "                                   shear_range = 0.2,\r\n",
        "                                   zoom_range = 0.2,\r\n",
        "                                   horizontal_flip = True)\r\n",
        "\r\n",
        "test_datagen = ImageDataGenerator(rescale = 1./255)\r\n",
        "\r\n",
        "training_set = train_datagen.flow_from_directory(train_path,\r\n",
        "                                                 target_size = (224, 224),\r\n",
        "                                                 batch_size = 32,\r\n",
        "                                                 class_mode = 'categorical')\r\n",
        "test_set = test_datagen.flow_from_directory(valid_path,\r\n",
        "                                            target_size = (224, 224),\r\n",
        "                                            batch_size = 32,\r\n",
        "                                            class_mode = 'categorical')\r\n",
        "\r\n",
        "#flow from directory is the method of imagedatagenerator\r\n",
        "# earlystopping_callback = EarlyStopping(monitor='val_accuracy',mode='auto')\r\n",
        "checkpoint_callback = ModelCheckpoint(filepath='CNN MobileNetV1.h5',monitor='val_accuracy', mode='max', save_best_only=True)\r\n",
        "history=model.fit_generator(training_set,validation_data=(test_set),epochs=10,steps_per_epoch=len(training_set)\r\n",
        ",validation_steps=len(test_set),callbacks=[checkpoint_callback])\r\n",
        "\r\n",
        "model = load_model('CNN MobileNetV1.h5')\r\n",
        "[train_loss, train_accuracy] = model.evaluate(training_set)\r\n",
        "print(\"Evaluation result on Train Data : Loss = {}, accuracy = {}\".format(train_loss, train_accuracy))\r\n",
        "[test_loss, test_acc] = model.evaluate(test_set)\r\n",
        "print(\"Evaluation result on Test Data : Loss = {}, accuracy = {}\".format(test_loss, test_acc))\r\n",
        "\r\n",
        "#Plot the loss curves\r\n",
        "plt.figure(figsize=[8,6])\r\n",
        "plt.plot(history.history['loss'],'r',linewidth=3.0)\r\n",
        "plt.plot(history.history['val_loss'],'b',linewidth=3.0)\r\n",
        "plt.legend(['Training loss', 'Validation Loss'],fontsize=18)\r\n",
        "plt.xlabel('Epochs ',fontsize=16)\r\n",
        "plt.ylabel('Loss',fontsize=16)\r\n",
        "plt.title('Loss Curves',fontsize=1)\r\n",
        "plt.show()\r\n",
        "\r\n",
        "#Plot the Accuracy Curves\r\n",
        "plt.figure(figsize=[8,6]) \r\n",
        "plt.plot(history.history['accuracy'],'r',linewidth=3.0) \r\n",
        "plt.plot(history.history['val_accuracy'],'b',linewidth=3.0)\r\n",
        "plt.legend(['Training Accuracy', 'Validation Accuracy'],fontsize=18)\r\n",
        "plt.xlabel('Epochs ',fontsize=16) \r\n",
        "plt.ylabel('Accuracy',fontsize=16)\r\n",
        "plt.title('Accuracy Curves',fontsize=16)\r\n",
        "plt.show()\r\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/mobilenet/mobilenet_1_0_224_tf_no_top.h5\n",
            "17227776/17225924 [==============================] - 0s 0us/step\n",
            "Total number of classes are 40\n",
            "Model: \"model\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "input_1 (InputLayer)         [(None, 224, 224, 3)]     0         \n",
            "_________________________________________________________________\n",
            "conv1 (Conv2D)               (None, 112, 112, 32)      864       \n",
            "_________________________________________________________________\n",
            "conv1_bn (BatchNormalization (None, 112, 112, 32)      128       \n",
            "_________________________________________________________________\n",
            "conv1_relu (ReLU)            (None, 112, 112, 32)      0         \n",
            "_________________________________________________________________\n",
            "conv_dw_1 (DepthwiseConv2D)  (None, 112, 112, 32)      288       \n",
            "_________________________________________________________________\n",
            "conv_dw_1_bn (BatchNormaliza (None, 112, 112, 32)      128       \n",
            "_________________________________________________________________\n",
            "conv_dw_1_relu (ReLU)        (None, 112, 112, 32)      0         \n",
            "_________________________________________________________________\n",
            "conv_pw_1 (Conv2D)           (None, 112, 112, 64)      2048      \n",
            "_________________________________________________________________\n",
            "conv_pw_1_bn (BatchNormaliza (None, 112, 112, 64)      256       \n",
            "_________________________________________________________________\n",
            "conv_pw_1_relu (ReLU)        (None, 112, 112, 64)      0         \n",
            "_________________________________________________________________\n",
            "conv_pad_2 (ZeroPadding2D)   (None, 113, 113, 64)      0         \n",
            "_________________________________________________________________\n",
            "conv_dw_2 (DepthwiseConv2D)  (None, 56, 56, 64)        576       \n",
            "_________________________________________________________________\n",
            "conv_dw_2_bn (BatchNormaliza (None, 56, 56, 64)        256       \n",
            "_________________________________________________________________\n",
            "conv_dw_2_relu (ReLU)        (None, 56, 56, 64)        0         \n",
            "_________________________________________________________________\n",
            "conv_pw_2 (Conv2D)           (None, 56, 56, 128)       8192      \n",
            "_________________________________________________________________\n",
            "conv_pw_2_bn (BatchNormaliza (None, 56, 56, 128)       512       \n",
            "_________________________________________________________________\n",
            "conv_pw_2_relu (ReLU)        (None, 56, 56, 128)       0         \n",
            "_________________________________________________________________\n",
            "conv_dw_3 (DepthwiseConv2D)  (None, 56, 56, 128)       1152      \n",
            "_________________________________________________________________\n",
            "conv_dw_3_bn (BatchNormaliza (None, 56, 56, 128)       512       \n",
            "_________________________________________________________________\n",
            "conv_dw_3_relu (ReLU)        (None, 56, 56, 128)       0         \n",
            "_________________________________________________________________\n",
            "conv_pw_3 (Conv2D)           (None, 56, 56, 128)       16384     \n",
            "_________________________________________________________________\n",
            "conv_pw_3_bn (BatchNormaliza (None, 56, 56, 128)       512       \n",
            "_________________________________________________________________\n",
            "conv_pw_3_relu (ReLU)        (None, 56, 56, 128)       0         \n",
            "_________________________________________________________________\n",
            "conv_pad_4 (ZeroPadding2D)   (None, 57, 57, 128)       0         \n",
            "_________________________________________________________________\n",
            "conv_dw_4 (DepthwiseConv2D)  (None, 28, 28, 128)       1152      \n",
            "_________________________________________________________________\n",
            "conv_dw_4_bn (BatchNormaliza (None, 28, 28, 128)       512       \n",
            "_________________________________________________________________\n",
            "conv_dw_4_relu (ReLU)        (None, 28, 28, 128)       0         \n",
            "_________________________________________________________________\n",
            "conv_pw_4 (Conv2D)           (None, 28, 28, 256)       32768     \n",
            "_________________________________________________________________\n",
            "conv_pw_4_bn (BatchNormaliza (None, 28, 28, 256)       1024      \n",
            "_________________________________________________________________\n",
            "conv_pw_4_relu (ReLU)        (None, 28, 28, 256)       0         \n",
            "_________________________________________________________________\n",
            "conv_dw_5 (DepthwiseConv2D)  (None, 28, 28, 256)       2304      \n",
            "_________________________________________________________________\n",
            "conv_dw_5_bn (BatchNormaliza (None, 28, 28, 256)       1024      \n",
            "_________________________________________________________________\n",
            "conv_dw_5_relu (ReLU)        (None, 28, 28, 256)       0         \n",
            "_________________________________________________________________\n",
            "conv_pw_5 (Conv2D)           (None, 28, 28, 256)       65536     \n",
            "_________________________________________________________________\n",
            "conv_pw_5_bn (BatchNormaliza (None, 28, 28, 256)       1024      \n",
            "_________________________________________________________________\n",
            "conv_pw_5_relu (ReLU)        (None, 28, 28, 256)       0         \n",
            "_________________________________________________________________\n",
            "conv_pad_6 (ZeroPadding2D)   (None, 29, 29, 256)       0         \n",
            "_________________________________________________________________\n",
            "conv_dw_6 (DepthwiseConv2D)  (None, 14, 14, 256)       2304      \n",
            "_________________________________________________________________\n",
            "conv_dw_6_bn (BatchNormaliza (None, 14, 14, 256)       1024      \n",
            "_________________________________________________________________\n",
            "conv_dw_6_relu (ReLU)        (None, 14, 14, 256)       0         \n",
            "_________________________________________________________________\n",
            "conv_pw_6 (Conv2D)           (None, 14, 14, 512)       131072    \n",
            "_________________________________________________________________\n",
            "conv_pw_6_bn (BatchNormaliza (None, 14, 14, 512)       2048      \n",
            "_________________________________________________________________\n",
            "conv_pw_6_relu (ReLU)        (None, 14, 14, 512)       0         \n",
            "_________________________________________________________________\n",
            "conv_dw_7 (DepthwiseConv2D)  (None, 14, 14, 512)       4608      \n",
            "_________________________________________________________________\n",
            "conv_dw_7_bn (BatchNormaliza (None, 14, 14, 512)       2048      \n",
            "_________________________________________________________________\n",
            "conv_dw_7_relu (ReLU)        (None, 14, 14, 512)       0         \n",
            "_________________________________________________________________\n",
            "conv_pw_7 (Conv2D)           (None, 14, 14, 512)       262144    \n",
            "_________________________________________________________________\n",
            "conv_pw_7_bn (BatchNormaliza (None, 14, 14, 512)       2048      \n",
            "_________________________________________________________________\n",
            "conv_pw_7_relu (ReLU)        (None, 14, 14, 512)       0         \n",
            "_________________________________________________________________\n",
            "conv_dw_8 (DepthwiseConv2D)  (None, 14, 14, 512)       4608      \n",
            "_________________________________________________________________\n",
            "conv_dw_8_bn (BatchNormaliza (None, 14, 14, 512)       2048      \n",
            "_________________________________________________________________\n",
            "conv_dw_8_relu (ReLU)        (None, 14, 14, 512)       0         \n",
            "_________________________________________________________________\n",
            "conv_pw_8 (Conv2D)           (None, 14, 14, 512)       262144    \n",
            "_________________________________________________________________\n",
            "conv_pw_8_bn (BatchNormaliza (None, 14, 14, 512)       2048      \n",
            "_________________________________________________________________\n",
            "conv_pw_8_relu (ReLU)        (None, 14, 14, 512)       0         \n",
            "_________________________________________________________________\n",
            "conv_dw_9 (DepthwiseConv2D)  (None, 14, 14, 512)       4608      \n",
            "_________________________________________________________________\n",
            "conv_dw_9_bn (BatchNormaliza (None, 14, 14, 512)       2048      \n",
            "_________________________________________________________________\n",
            "conv_dw_9_relu (ReLU)        (None, 14, 14, 512)       0         \n",
            "_________________________________________________________________\n",
            "conv_pw_9 (Conv2D)           (None, 14, 14, 512)       262144    \n",
            "_________________________________________________________________\n",
            "conv_pw_9_bn (BatchNormaliza (None, 14, 14, 512)       2048      \n",
            "_________________________________________________________________\n",
            "conv_pw_9_relu (ReLU)        (None, 14, 14, 512)       0         \n",
            "_________________________________________________________________\n",
            "conv_dw_10 (DepthwiseConv2D) (None, 14, 14, 512)       4608      \n",
            "_________________________________________________________________\n",
            "conv_dw_10_bn (BatchNormaliz (None, 14, 14, 512)       2048      \n",
            "_________________________________________________________________\n",
            "conv_dw_10_relu (ReLU)       (None, 14, 14, 512)       0         \n",
            "_________________________________________________________________\n",
            "conv_pw_10 (Conv2D)          (None, 14, 14, 512)       262144    \n",
            "_________________________________________________________________\n",
            "conv_pw_10_bn (BatchNormaliz (None, 14, 14, 512)       2048      \n",
            "_________________________________________________________________\n",
            "conv_pw_10_relu (ReLU)       (None, 14, 14, 512)       0         \n",
            "_________________________________________________________________\n",
            "conv_dw_11 (DepthwiseConv2D) (None, 14, 14, 512)       4608      \n",
            "_________________________________________________________________\n",
            "conv_dw_11_bn (BatchNormaliz (None, 14, 14, 512)       2048      \n",
            "_________________________________________________________________\n",
            "conv_dw_11_relu (ReLU)       (None, 14, 14, 512)       0         \n",
            "_________________________________________________________________\n",
            "conv_pw_11 (Conv2D)          (None, 14, 14, 512)       262144    \n",
            "_________________________________________________________________\n",
            "conv_pw_11_bn (BatchNormaliz (None, 14, 14, 512)       2048      \n",
            "_________________________________________________________________\n",
            "conv_pw_11_relu (ReLU)       (None, 14, 14, 512)       0         \n",
            "_________________________________________________________________\n",
            "conv_pad_12 (ZeroPadding2D)  (None, 15, 15, 512)       0         \n",
            "_________________________________________________________________\n",
            "conv_dw_12 (DepthwiseConv2D) (None, 7, 7, 512)         4608      \n",
            "_________________________________________________________________\n",
            "conv_dw_12_bn (BatchNormaliz (None, 7, 7, 512)         2048      \n",
            "_________________________________________________________________\n",
            "conv_dw_12_relu (ReLU)       (None, 7, 7, 512)         0         \n",
            "_________________________________________________________________\n",
            "conv_pw_12 (Conv2D)          (None, 7, 7, 1024)        524288    \n",
            "_________________________________________________________________\n",
            "conv_pw_12_bn (BatchNormaliz (None, 7, 7, 1024)        4096      \n",
            "_________________________________________________________________\n",
            "conv_pw_12_relu (ReLU)       (None, 7, 7, 1024)        0         \n",
            "_________________________________________________________________\n",
            "conv_dw_13 (DepthwiseConv2D) (None, 7, 7, 1024)        9216      \n",
            "_________________________________________________________________\n",
            "conv_dw_13_bn (BatchNormaliz (None, 7, 7, 1024)        4096      \n",
            "_________________________________________________________________\n",
            "conv_dw_13_relu (ReLU)       (None, 7, 7, 1024)        0         \n",
            "_________________________________________________________________\n",
            "conv_pw_13 (Conv2D)          (None, 7, 7, 1024)        1048576   \n",
            "_________________________________________________________________\n",
            "conv_pw_13_bn (BatchNormaliz (None, 7, 7, 1024)        4096      \n",
            "_________________________________________________________________\n",
            "conv_pw_13_relu (ReLU)       (None, 7, 7, 1024)        0         \n",
            "_________________________________________________________________\n",
            "flatten (Flatten)            (None, 50176)             0         \n",
            "_________________________________________________________________\n",
            "dense (Dense)                (None, 40)                2007080   \n",
            "=================================================================\n",
            "Total params: 5,235,944\n",
            "Trainable params: 2,007,080\n",
            "Non-trainable params: 3,228,864\n",
            "_________________________________________________________________\n",
            "Found 3675 images belonging to 40 classes.\n",
            "Found 408 images belonging to 40 classes.\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:1844: UserWarning: `Model.fit_generator` is deprecated and will be removed in a future version. Please use `Model.fit`, which supports generators.\n",
            "  warnings.warn('`Model.fit_generator` is deprecated and '\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/10\n",
            "115/115 [==============================] - 999s 9s/step - loss: 6.4913 - accuracy: 0.5977 - val_loss: 2.4033 - val_accuracy: 0.8309\n",
            "Epoch 2/10\n",
            "115/115 [==============================] - 196s 2s/step - loss: 0.7179 - accuracy: 0.9267 - val_loss: 2.1973 - val_accuracy: 0.8873\n",
            "Epoch 3/10\n",
            "115/115 [==============================] - 195s 2s/step - loss: 0.5304 - accuracy: 0.9457 - val_loss: 1.9458 - val_accuracy: 0.8897\n",
            "Epoch 4/10\n",
            "115/115 [==============================] - 196s 2s/step - loss: 0.3488 - accuracy: 0.9607 - val_loss: 2.2406 - val_accuracy: 0.8995\n",
            "Epoch 5/10\n",
            "115/115 [==============================] - 195s 2s/step - loss: 0.4258 - accuracy: 0.9652 - val_loss: 2.7449 - val_accuracy: 0.8848\n",
            "Epoch 6/10\n",
            "115/115 [==============================] - 195s 2s/step - loss: 0.2790 - accuracy: 0.9764 - val_loss: 2.1470 - val_accuracy: 0.9020\n",
            "Epoch 7/10\n",
            "115/115 [==============================] - 195s 2s/step - loss: 0.2598 - accuracy: 0.9751 - val_loss: 2.9270 - val_accuracy: 0.8995\n",
            "Epoch 8/10\n",
            "115/115 [==============================] - 194s 2s/step - loss: 0.1479 - accuracy: 0.9906 - val_loss: 2.5010 - val_accuracy: 0.9118\n",
            "Epoch 9/10\n",
            "115/115 [==============================] - 195s 2s/step - loss: 0.2292 - accuracy: 0.9816 - val_loss: 3.1065 - val_accuracy: 0.9093\n",
            "Epoch 10/10\n",
            "115/115 [==============================] - 196s 2s/step - loss: 0.2310 - accuracy: 0.9859 - val_loss: 2.7060 - val_accuracy: 0.9216\n",
            "115/115 [==============================] - 177s 2s/step - loss: 0.1088 - accuracy: 0.9913\n",
            "Evaluation result on Train Data : Loss = 0.10882510989904404, accuracy = 0.9912925362586975\n",
            "13/13 [==============================] - 16s 1s/step - loss: 2.7060 - accuracy: 0.9216\n",
            "Evaluation result on Test Data : Loss = 2.706021547317505, accuracy = 0.9215686321258545\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SuwIG_2yuq2R",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "47340cb8-1f65-4185-a799-d4649537ecb9"
      },
      "source": [
        "import cv2\r\n",
        "import tensorflow as tf\r\n",
        "\r\n",
        "\r\n",
        "def prepare(filepath):\r\n",
        "    img = image.load_img(filepath , target_size=(224, 224))\r\n",
        "    img_array = image.img_to_array(img)\r\n",
        "    img_array_expanded_dims = np.expand_dims(img_array, axis=0)\r\n",
        "    return tf.keras.applications.inception_resnet_v2.preprocess_input(img_array_expanded_dims)\r\n",
        "\r\n",
        "\r\n",
        "model = tf.keras.models.load_model(\"CNN MobileNetV1.h5\")\r\n",
        "\r\n",
        "prediction = model.predict([prepare('/content/drive/MyDrive/test imgs/pizzahut1.jpg')])\r\n",
        "print(prediction)  # will be a list in a list.\r\n",
        "highest=np.argmax(prediction)\r\n",
        "print(\"Highest position : \", highest)\r\n",
        "if(highest==0):\r\n",
        "    print(\"This class is Acer\")\r\n",
        "elif(highest==1):\r\n",
        "    print(\"This class is Albertsons\")\r\n",
        "elif(highest==2):\r\n",
        "    print(\"This class is Apple\")\r\n",
        "elif(highest==3):\r\n",
        "    print(\"This class is Asus\")\r\n",
        "elif(highest==4):\r\n",
        "    print(\"This class is Audi\")\r\n",
        "elif(highest==5):\r\n",
        "    print(\"This class is BMW\")\r\n",
        "elif(highest==6):\r\n",
        "    print(\"This class is Becel\")\r\n",
        "elif(highest==7):\r\n",
        "    print(\"This class is Burger King\")\r\n",
        "elif(highest==8):\r\n",
        "    print(\"This class is Cadbury\")\r\n",
        "elif(highest==9):\r\n",
        "    print(\"This class is Calpol\")\r\n",
        "elif(highest==10):\r\n",
        "    print(\"This class is Colgate\")\r\n",
        "elif(highest==11):\r\n",
        "    print(\"This class is Friskies\")\r\n",
        "elif(highest==12):\r\n",
        "    print(\"This class is Intel\")\r\n",
        "elif(highest==13):\r\n",
        "    print(\"This class is Jaguar\")\r\n",
        "elif(highest==14):\r\n",
        "    print(\"This class is KFC\")\r\n",
        "elif(highest==15):\r\n",
        "    print(\"This class is KitKat\")\r\n",
        "elif(highest==16):\r\n",
        "    print(\"This class is Knorr\")\r\n",
        "elif(highest==17):\r\n",
        "    print(\"This class is Lifebuoy\")\r\n",
        "elif(highest==18):\r\n",
        "    print(\"This class is Lipton\")\r\n",
        "elif(highest==19):\r\n",
        "    print(\"This class is Logitech\")\r\n",
        "elif(highest==20):\r\n",
        "    print(\"This class is Malteasers\")\r\n",
        "elif(highest==21):\r\n",
        "    print(\"This class is McDonald's\")\r\n",
        "elif(highest==22):\r\n",
        "    print(\"This class is Mercedes\")\r\n",
        "elif(highest==23):\r\n",
        "    print(\"This class is Oral-B\")\r\n",
        "elif(highest==24):\r\n",
        "    print(\"This class is Panadol\")\r\n",
        "elif(highest==25):\r\n",
        "    print(\"This class is Panago\")\r\n",
        "elif(highest==26):\r\n",
        "    print(\"This class is Perkin Elmer\")\r\n",
        "elif(highest==27):\r\n",
        "    print(\"This class is Pizza Hut\")\r\n",
        "elif(highest==28):\r\n",
        "    print(\"This class is Sensodyne\")\r\n",
        "elif(highest==29):\r\n",
        "    print(\"This class is TicTac\")\r\n",
        "elif(highest==30):\r\n",
        "    print(\"This class is Total\")\r\n",
        "elif(highest==31):\r\n",
        "    print(\"This class is Toyota\")\r\n",
        "elif(highest==32):\r\n",
        "    print(\"This class is Triaminic\")\r\n",
        "elif(highest==33):\r\n",
        "    print(\"This class is Wall's\")\r\n",
        "elif(highest==34):\r\n",
        "    print(\"This class is Coca Cola\")\r\n",
        "elif(highest==35):\r\n",
        "    print(\"This class is Hp\")\r\n",
        "elif(highest==36):\r\n",
        "    print(\"This class is Huawei\")\r\n",
        "elif(highest==37):\r\n",
        "    print(\"This class is Pepsi\")\r\n",
        "elif(highest==38):\r\n",
        "    print(\"This class is Samsung\")\r\n",
        "elif(highest==39):\r\n",
        "    print(\"This class is Sony\")\r\n",
        "else:\r\n",
        "    print(\"Invalid Class and Invalid Category\")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:11 out of the last 11 calls to <function Model.make_predict_function.<locals>.predict_function at 0x7f15471f2bf8> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has experimental_relax_shapes=True option that relaxes argument shapes that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n",
            "[[1.03106315e-30 2.27742862e-32 1.01413562e-26 3.45212592e-24\n",
            "  3.89598561e-24 1.88853192e-30 4.74150164e-32 8.56088340e-27\n",
            "  4.21717336e-27 1.76786462e-20 2.47634515e-01 7.35907679e-06\n",
            "  2.64558338e-33 4.25330108e-18 7.48397171e-01 3.84442834e-03\n",
            "  9.75457226e-09 4.63347850e-12 5.88562699e-10 7.70315873e-37\n",
            "  1.00341807e-04 1.91478692e-20 5.74501217e-25 0.00000000e+00\n",
            "  3.09697966e-25 3.97157418e-14 6.82877817e-32 6.10960445e-27\n",
            "  3.59071958e-27 2.18684815e-19 6.50836717e-16 0.00000000e+00\n",
            "  5.30245758e-09 3.78458607e-23 1.61692697e-05 5.08399406e-26\n",
            "  1.36489231e-11 3.03818689e-20 1.19075186e-20 8.02605645e-19]]\n",
            "Highest position :  14\n",
            "This class is KFC\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TCKGU-ppMXSh"
      },
      "source": [
        "from keras.applications.mobilenet import preprocess_input\r\n",
        "from tensorflow.keras.preprocessing import image\r\n",
        "from tensorflow.keras.applications import imagenet_utils\r\n",
        "\r\n",
        "def prepare_image(file):\r\n",
        "    img_path = '/content/drive/MyDrive/test imgs/'\r\n",
        "    img = image.load_img(img_path + file, target_size=(224, 224))\r\n",
        "    img_array = image.img_to_array(img)\r\n",
        "    img_array_expanded_dims = np.expand_dims(img_array, axis=0)\r\n",
        "    return keras.applications.mobilenet.preprocess_input(img_array_expanded_dims)\r\n",
        "    \r\n",
        "model = tf.keras.models.load_model(\"CNN MobileNetV1.h5\")\r\n",
        "preprocessed_image = prepare_image('mcd.jpg')\r\n",
        "predictions = model.predict(preprocessed_image)\r\n",
        "results = imagenet_utils.decode_predictions(predictions)"
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}