example_model/rotten_vs_fresh.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Salinan dari rotten-fresh.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "9Phj4dAK-4r9",
        "colab_type": "code",
        "outputId": "39610561-b6dc-4edf-aa94-261146d73237",
        "colab": {
          "resources": {
            "http://localhost:8080/nbextensions/google.colab/files.js": {
              "data": "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",
              "ok": true,
              "headers": [
                [
                  "content-type",
                  "application/javascript"
                ]
              ],
              "status": 200,
              "status_text": ""
            }
          },
          "base_uri": "https://localhost:8080/",
          "height": 146
        }
      },
      "source": [
        "# Run this cell and select the kaggle.json file downloaded\n",
        "# from the Kaggle account settings page.\n",
        "from google.colab import files\n",
        "files.upload()\n",
        "\n",
        "# Let's make sure the kaggle.json file is present.\n",
        "!ls -lha kaggle.json\n",
        "\n",
        "# Next, install the Kaggle API client.\n",
        "!pip install -q kaggle\n",
        "\n",
        "# The Kaggle API client expects this file to be in ~/.kaggle,\n",
        "# so move it there.\n",
        "!mkdir -p ~/.kaggle\n",
        "!cp kaggle.json ~/.kaggle/\n",
        "\n",
        "# This permissions change avoids a warning on Kaggle tool startup.\n",
        "!chmod 600 ~/.kaggle/kaggle.json\n",
        "\n",
        "! kaggle datasets download -d sriramr/fruits-fresh-and-rotten-for-classification"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "\n",
              "     <input type=\"file\" id=\"files-8960a579-8dc1-4181-8381-9bfdb0e26eb1\" name=\"files[]\" multiple disabled />\n",
              "     <output id=\"result-8960a579-8dc1-4181-8381-9bfdb0e26eb1\">\n",
              "      Upload widget is only available when the cell has been executed in the\n",
              "      current browser session. Please rerun this cell to enable.\n",
              "      </output>\n",
              "      <script src=\"/nbextensions/google.colab/files.js\"></script> "
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Saving kaggle.json to kaggle.json\n",
            "-rw-r--r-- 1 root root 69 Jun 11 13:20 kaggle.json\n",
            "Downloading fruits-fresh-and-rotten-for-classification.zip to /content\n",
            "100% 3.57G/3.58G [01:05<00:00, 47.7MB/s]\n",
            "100% 3.58G/3.58G [01:05<00:00, 58.6MB/s]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DjuaSZEkCPae",
        "colab_type": "code",
        "outputId": "f0778f52-bdb7-483b-d024-3bf1d6c4273b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "import zipfile\n",
        "import os\n",
        "\n",
        "curDir = os.getcwd()\n",
        "print(curDir)\n",
        "\n",
        "local_zip = '/content/fruits-fresh-and-rotten-for-classification.zip'\n",
        "zip_ref = zipfile.ZipFile(local_zip, 'r')\n",
        "zip_ref.extractall('/content')\n",
        "\n",
        "zip_ref.close()"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/content\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "e23Rdtr4p3fb",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 304
        },
        "outputId": "b68fff12-b6e1-44bf-e2e8-5f9dcb2e919d"
      },
      "source": [
        "import random\n",
        "from keras.preprocessing.image import load_img\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "filenames = os.listdir('/content/dataset/train/freshoranges')\n",
        "sample = random.choice(filenames)\n",
        "image = load_img('/content/dataset/train/freshoranges/'+sample)\n",
        "plt.imshow(image)\n",
        "print(image.size)"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Using TensorFlow backend.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "(356, 270)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6QrUqtGSSG5s",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import shutil\n",
        "\n",
        "shutil.rmtree('/content/dataset/dataset')\n",
        "os.remove('fruits-fresh-and-rotten-for-classification.zip')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "U9Vm18FjpbLp",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 52
        },
        "outputId": "1f0e3225-62c7-4fcf-a61a-3485b50d804d"
      },
      "source": [
        "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
        "\n",
        "train_datagen = ImageDataGenerator(rescale=1/255)\n",
        "validation_datagen = ImageDataGenerator(rescale=1/255)\n",
        "\n",
        "train_generator = train_datagen.flow_from_directory(\n",
        "        '/content/dataset/train', \n",
        "        target_size=(150, 150),\n",
        "        batch_size=15,\n",
        "\t      class_mode='categorical')\n",
        "\n",
        "validation_generator = validation_datagen.flow_from_directory(\n",
        "        '/content/dataset/test',\n",
        "        target_size=(150, 150),\n",
        "        batch_size=15,\n",
        "\t      class_mode='categorical')"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Found 10901 images belonging to 6 classes.\n",
            "Found 2698 images belonging to 6 classes.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cI6rDkXmNo5Z",
        "colab_type": "code",
        "outputId": "d4050e66-3099-4498-d183-8123af784a5f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 791
        }
      },
      "source": [
        "from keras.models import Sequential\n",
        "from keras.layers import Conv2D, MaxPooling2D\n",
        "from keras.layers import Activation, Dropout, Flatten, Dense, BatchNormalization, GlobalMaxPooling2D\n",
        "from keras import backend as K\n",
        "\n",
        "model = Sequential()\n",
        "model.add(Conv2D(64, (3, 3), input_shape=(150,150,3)))\n",
        "model.add(Activation('relu'))\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "model.add(Conv2D(64, (3, 3)))\n",
        "# model.add(BatchNormalization())\n",
        "model.add(Activation('relu'))\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "\n",
        "model.add(Conv2D(32, (3, 3)))\n",
        "model.add(Activation('relu'))\n",
        "model.add(Conv2D(32, (3, 3)))\n",
        "# model.add(BatchNormalization())\n",
        "model.add(Activation('relu'))\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "# model.add(GlobalMaxPooling2D())\n",
        "\n",
        "\n",
        "# model.add(Dropout(0.25))\n",
        "# model.add(Dense(128))\n",
        "# model.add(Activation('relu'))\n",
        "# model.add(Dropout(0.25))\n",
        "# model.add(Dense(128))\n",
        "# model.add(Activation('relu'))\n",
        "model.add(Dropout(0.25))\n",
        "model.add(Dense(256))\n",
        "model.add(Activation('relu'))\n",
        "model.add(Dropout(0.25))\n",
        "\n",
        "model.add(Flatten())\n",
        "model.add(Dense(6))\n",
        "model.add(Activation('softmax'))\n",
        "\n",
        "model.compile(loss='categorical_crossentropy', optimizer='adamax', metrics=['accuracy'])\n",
        "model.summary()"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_1\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d_1 (Conv2D)            (None, 148, 148, 64)      1792      \n",
            "_________________________________________________________________\n",
            "activation_1 (Activation)    (None, 148, 148, 64)      0         \n",
            "_________________________________________________________________\n",
            "max_pooling2d_1 (MaxPooling2 (None, 74, 74, 64)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_2 (Conv2D)            (None, 72, 72, 64)        36928     \n",
            "_________________________________________________________________\n",
            "activation_2 (Activation)    (None, 72, 72, 64)        0         \n",
            "_________________________________________________________________\n",
            "max_pooling2d_2 (MaxPooling2 (None, 36, 36, 64)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_3 (Conv2D)            (None, 34, 34, 32)        18464     \n",
            "_________________________________________________________________\n",
            "activation_3 (Activation)    (None, 34, 34, 32)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_4 (Conv2D)            (None, 32, 32, 32)        9248      \n",
            "_________________________________________________________________\n",
            "activation_4 (Activation)    (None, 32, 32, 32)        0         \n",
            "_________________________________________________________________\n",
            "max_pooling2d_3 (MaxPooling2 (None, 16, 16, 32)        0         \n",
            "_________________________________________________________________\n",
            "dropout_1 (Dropout)          (None, 16, 16, 32)        0         \n",
            "_________________________________________________________________\n",
            "dense_1 (Dense)              (None, 16, 16, 256)       8448      \n",
            "_________________________________________________________________\n",
            "activation_5 (Activation)    (None, 16, 16, 256)       0         \n",
            "_________________________________________________________________\n",
            "dropout_2 (Dropout)          (None, 16, 16, 256)       0         \n",
            "_________________________________________________________________\n",
            "flatten_1 (Flatten)          (None, 65536)             0         \n",
            "_________________________________________________________________\n",
            "dense_2 (Dense)              (None, 6)                 393222    \n",
            "_________________________________________________________________\n",
            "activation_6 (Activation)    (None, 6)                 0         \n",
            "=================================================================\n",
            "Total params: 468,102\n",
            "Trainable params: 468,102\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gVx6zlJ3WDJR",
        "colab_type": "code",
        "outputId": "a2aae0ed-de71-48e7-9462-45d507511af2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "history = model.fit_generator(\n",
        "              train_generator,    \n",
        "              epochs=30, \n",
        "              validation_data = validation_generator,\n",
        "              verbose=1\n",
        "              )"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/30\n",
            "727/727 [==============================] - 75s 103ms/step - loss: 0.5970 - accuracy: 0.7803 - val_loss: 0.1729 - val_accuracy: 0.8866\n",
            "Epoch 2/30\n",
            "727/727 [==============================] - 67s 92ms/step - loss: 0.2956 - accuracy: 0.8951 - val_loss: 0.1823 - val_accuracy: 0.9110\n",
            "Epoch 3/30\n",
            "727/727 [==============================] - 67s 92ms/step - loss: 0.2277 - accuracy: 0.9173 - val_loss: 0.2925 - val_accuracy: 0.9318\n",
            "Epoch 4/30\n",
            "727/727 [==============================] - 65s 89ms/step - loss: 0.1759 - accuracy: 0.9382 - val_loss: 0.1175 - val_accuracy: 0.9466\n",
            "Epoch 5/30\n",
            "727/727 [==============================] - 65s 89ms/step - loss: 0.1324 - accuracy: 0.9526 - val_loss: 0.0759 - val_accuracy: 0.9481\n",
            "Epoch 6/30\n",
            "727/727 [==============================] - 65s 89ms/step - loss: 0.1159 - accuracy: 0.9598 - val_loss: 0.2800 - val_accuracy: 0.9218\n",
            "Epoch 7/30\n",
            "727/727 [==============================] - 66s 90ms/step - loss: 0.0920 - accuracy: 0.9664 - val_loss: 0.0093 - val_accuracy: 0.9622\n",
            "Epoch 8/30\n",
            "727/727 [==============================] - 63s 87ms/step - loss: 0.0750 - accuracy: 0.9728 - val_loss: 0.1420 - val_accuracy: 0.9703\n",
            "Epoch 9/30\n",
            "727/727 [==============================] - 65s 89ms/step - loss: 0.0654 - accuracy: 0.9765 - val_loss: 7.4147e-04 - val_accuracy: 0.9703\n",
            "Epoch 10/30\n",
            "727/727 [==============================] - 64s 88ms/step - loss: 0.0557 - accuracy: 0.9795 - val_loss: 0.2779 - val_accuracy: 0.9544\n",
            "Epoch 11/30\n",
            "727/727 [==============================] - 64s 88ms/step - loss: 0.0464 - accuracy: 0.9849 - val_loss: 0.0124 - val_accuracy: 0.9715\n",
            "Epoch 12/30\n",
            "727/727 [==============================] - 63s 87ms/step - loss: 0.0434 - accuracy: 0.9841 - val_loss: 0.0627 - val_accuracy: 0.9696\n",
            "Epoch 13/30\n",
            "727/727 [==============================] - 63s 87ms/step - loss: 0.0382 - accuracy: 0.9867 - val_loss: 0.0028 - val_accuracy: 0.9807\n",
            "Epoch 14/30\n",
            "727/727 [==============================] - 63s 87ms/step - loss: 0.0332 - accuracy: 0.9883 - val_loss: 0.0654 - val_accuracy: 0.9789\n",
            "Epoch 15/30\n",
            "727/727 [==============================] - 64s 88ms/step - loss: 0.0256 - accuracy: 0.9916 - val_loss: 0.0021 - val_accuracy: 0.9748\n",
            "Epoch 16/30\n",
            "727/727 [==============================] - 64s 88ms/step - loss: 0.0270 - accuracy: 0.9903 - val_loss: 5.8414e-05 - val_accuracy: 0.9678\n",
            "Epoch 17/30\n",
            "727/727 [==============================] - 64s 88ms/step - loss: 0.0218 - accuracy: 0.9921 - val_loss: 0.1840 - val_accuracy: 0.9696\n",
            "Epoch 18/30\n",
            "727/727 [==============================] - 65s 90ms/step - loss: 0.0207 - accuracy: 0.9928 - val_loss: 0.0021 - val_accuracy: 0.9841\n",
            "Epoch 19/30\n",
            "727/727 [==============================] - 64s 88ms/step - loss: 0.0151 - accuracy: 0.9951 - val_loss: 0.0029 - val_accuracy: 0.9859\n",
            "Epoch 20/30\n",
            "727/727 [==============================] - 64s 88ms/step - loss: 0.0213 - accuracy: 0.9934 - val_loss: 1.7547e-04 - val_accuracy: 0.9874\n",
            "Epoch 21/30\n",
            "727/727 [==============================] - 66s 91ms/step - loss: 0.0189 - accuracy: 0.9937 - val_loss: 0.1833 - val_accuracy: 0.9800\n",
            "Epoch 22/30\n",
            "727/727 [==============================] - 66s 91ms/step - loss: 0.0122 - accuracy: 0.9958 - val_loss: 0.2228 - val_accuracy: 0.9874\n",
            "Epoch 23/30\n",
            "727/727 [==============================] - 67s 92ms/step - loss: 0.0107 - accuracy: 0.9965 - val_loss: 6.7006e-05 - val_accuracy: 0.9907\n",
            "Epoch 24/30\n",
            "727/727 [==============================] - 64s 89ms/step - loss: 0.0171 - accuracy: 0.9944 - val_loss: 0.0448 - val_accuracy: 0.9863\n",
            "Epoch 25/30\n",
            "727/727 [==============================] - 65s 89ms/step - loss: 0.0141 - accuracy: 0.9945 - val_loss: 0.1778 - val_accuracy: 0.9796\n",
            "Epoch 26/30\n",
            "727/727 [==============================] - 65s 90ms/step - loss: 0.0168 - accuracy: 0.9944 - val_loss: 0.0065 - val_accuracy: 0.9826\n",
            "Epoch 27/30\n",
            "727/727 [==============================] - 65s 89ms/step - loss: 0.0116 - accuracy: 0.9962 - val_loss: 0.0025 - val_accuracy: 0.9848\n",
            "Epoch 28/30\n",
            "727/727 [==============================] - 65s 90ms/step - loss: 0.0107 - accuracy: 0.9965 - val_loss: 9.5447e-04 - val_accuracy: 0.9833\n",
            "Epoch 29/30\n",
            "727/727 [==============================] - 63s 86ms/step - loss: 0.0155 - accuracy: 0.9952 - val_loss: 0.1329 - val_accuracy: 0.9889\n",
            "Epoch 30/30\n",
            "727/727 [==============================] - 64s 87ms/step - loss: 0.0066 - accuracy: 0.9983 - val_loss: 6.4307e-05 - val_accuracy: 0.9907\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eZR2de7ZWLd0",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 298
        },
        "outputId": "38222c8e-7ff6-442f-be3e-902fc2ba0827"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "acc = history.history['accuracy']\n",
        "val_acc = history.history['val_accuracy']\n",
        "loss = history.history['loss']\n",
        "val_loss = history.history['val_loss']\n",
        "\n",
        "epochs = range(len(acc))\n",
        "\n",
        "plt.plot(epochs, acc, 'r', label='Training accuracy')\n",
        "plt.plot(epochs, val_acc, 'b', label='Validation accuracy')\n",
        "plt.title('Training and validation accuracy')\n",
        "plt.legend(loc=0)\n",
        "plt.figure()\n",
        "\n",
        "\n",
        "plt.show()"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "i2BNYKAAYgt2",
        "colab_type": "code",
        "colab": {
          "resources": {
            "http://localhost:8080/nbextensions/google.colab/files.js": {
              "data": "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",
              "ok": true,
              "headers": [
                [
                  "content-type",
                  "application/javascript"
                ]
              ],
              "status": 200,
              "status_text": ""
            }
          },
          "base_uri": "https://localhost:8080/",
          "height": 93
        },
        "outputId": "b08e6cef-9cec-4600-bb3f-9d986cb07d4a"
      },
      "source": [
        "import numpy as np\n",
        "from google.colab import files\n",
        "from keras.preprocessing import image\n",
        "# https://storage.googleapis.com/laurencemoroney-blog.appspot.com/rps-validation.zip\n",
        "\n",
        "uploaded = files.upload()\n",
        "\n",
        "for fn in uploaded.keys():\n",
        " \n",
        "  # predicting images\n",
        "  path = fn\n",
        "  img = image.load_img(path, target_size=(150, 150,3))\n",
        "  x = image.img_to_array(img)\n",
        "  x = np.expand_dims(x, axis=0)\n",
        "\n",
        "  images = np.vstack([x])\n",
        "  classes = model.predict(images, batch_size=15)\n",
        "  # print(fn)\n",
        "  # print(classes)\n",
        "  class_list = ['fresh apple', 'fresh banana', 'fresh orange', 'rotten apple', 'rotten banana', 'rotten orange']\n",
        "  print('{} is a {}'.format(fn, class_list[classes.argmax(axis=-1)[0]]))\n"
      ],
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "\n",
              "     <input type=\"file\" id=\"files-de786d49-b05f-4ac7-a852-db960db30e0c\" name=\"files[]\" multiple disabled />\n",
              "     <output id=\"result-de786d49-b05f-4ac7-a852-db960db30e0c\">\n",
              "      Upload widget is only available when the cell has been executed in the\n",
              "      current browser session. Please rerun this cell to enable.\n",
              "      </output>\n",
              "      <script src=\"/nbextensions/google.colab/files.js\"></script> "
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Saving rotten-overripe-soft-banana-on-white-background-DBW8WH.jpg to rotten-overripe-soft-banana-on-white-background-DBW8WH (1).jpg\n",
            "rotten-overripe-soft-banana-on-white-background-DBW8WH.jpg is a rotten banana\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3phRVMUvB0NW",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model.save('rotten-fresh-model.h5') \n",
        "\n",
        "# Save the entire model as a SavedModel.\n",
        "# !mkdir -p saved_model\n",
        "# model.save('saved_model/my_model') "
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}