Submission_Image_Classification_Model_Deployment.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Submission Image Classification Model Deployment.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qfsy7GWvgz0z"
      },
      "source": [
        "## **Izdiharti Noni Pertiwi**\n",
        "dataset rockpaperscissors"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QHyTjIJIesaY",
        "outputId": "6ed148f1-7316-4919-f367-de6102de4ac2"
      },
      "source": [
        "!wget --no-check-certificate \\\n",
        "  https://dicodingacademy.blob.core.windows.net/picodiploma/ml_pemula_academy/rockpaperscissors.zip \\\n",
        "  -O /tmp/rockpaperscissors.zip"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--2021-07-16 13:59:33--  https://dicodingacademy.blob.core.windows.net/picodiploma/ml_pemula_academy/rockpaperscissors.zip\n",
            "Resolving dicodingacademy.blob.core.windows.net (dicodingacademy.blob.core.windows.net)... 52.239.197.36\n",
            "Connecting to dicodingacademy.blob.core.windows.net (dicodingacademy.blob.core.windows.net)|52.239.197.36|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 322873683 (308M) [application/zip]\n",
            "Saving to: ‘/tmp/rockpaperscissors.zip’\n",
            "\n",
            "/tmp/rockpapersciss 100%[===================>] 307.92M  3.02MB/s    in 45s     \n",
            "\n",
            "2021-07-16 14:00:19 (6.81 MB/s) - ‘/tmp/rockpaperscissors.zip’ saved [322873683/322873683]\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RWlxSCNZeuzb"
      },
      "source": [
        "# melakukan ekstraksi pada file zip\n",
        "import zipfile,os,shutil\n",
        "local_zip = '/tmp/rockpaperscissors.zip'\n",
        "zip_ref = zipfile.ZipFile(local_zip, 'r')\n",
        "zip_ref.extractall('/tmp')\n",
        "zip_ref.close()"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0XN1U6qke5ov"
      },
      "source": [
        "base_dir = '/tmp/rockpaperscissors/rps-cv-images'\n",
        "roc_dir = os.path.join(base_dir,'rock')\n",
        "pap_dir = os.path.join(base_dir, 'paper')\n",
        "sci_dir = os.path.join(base_dir, 'scissors')"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ylKvngTOdKVj",
        "outputId": "d36a8232-0b35-474c-9d57-1e52c77fcdec"
      },
      "source": [
        "total_rock = len(os.listdir(roc_dir))\n",
        "total_paper = len(os.listdir(pap_dir))\n",
        "total_scissors = len(os.listdir(sci_dir))\n",
        "\n",
        "# Mencetak jumlah data rock paper scissor\n",
        "print(\"Total Data Paper Image      : \",total_paper)\n",
        "print(\"Total Data Rock Image       : \",total_rock)\n",
        "print(\"Total Data Scissors Image  : \",total_scissors)"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Total Data Paper Image      :  712\n",
            "Total Data Rock Image       :  726\n",
            "Total Data Scissors Image  :  750\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tdfy96b1gVKO"
      },
      "source": [
        "import sklearn\n",
        "from sklearn import datasets\n",
        "from sklearn.model_selection import train_test_split"
      ],
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4iYINE0QdpF1"
      },
      "source": [
        "# Pembuatan Train dan Validation  data generator \n",
        "# Untuk augmentasi gambar\n",
        "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
        "#Ukuran validation 20% dari dataset\n",
        "val_size = 0.2\n",
        "\n",
        "Train_datagen = ImageDataGenerator(\n",
        "    rotation_range = 30,\n",
        "    brightness_range = [0.2,1.0],\n",
        "    shear_range = 0.2,\n",
        "    zoom_range = 0.2,\n",
        "    horizontal_flip = True,\n",
        "    fill_mode = \"nearest\",\n",
        "    rescale = 1./255,\n",
        "    validation_split = val_size\n",
        ")\n",
        "\n",
        "Validation_datagen = ImageDataGenerator(\n",
        "    rotation_range = 30,\n",
        "    brightness_range = [0.2,1.0],\n",
        "    shear_range = 0.2,\n",
        "    zoom_range = 0.2,\n",
        "    horizontal_flip = True,\n",
        "    fill_mode = \"nearest\",\n",
        "    rescale = 1./255,\n",
        "    validation_split = val_size\n",
        ")"
      ],
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ehyqMWP6dqIo",
        "outputId": "f690eca5-b64b-4524-b60a-b6dcb8197044"
      },
      "source": [
        "# Train dan Validation generator  dengan mode categorical\n",
        "Train_generator = Train_datagen.flow_from_directory(\n",
        "    base_dir,\n",
        "    target_size = (150,150),\n",
        "    color_mode = \"rgb\",\n",
        "    class_mode = \"categorical\",\n",
        "    batch_size = 16,\n",
        "    shuffle = True,\n",
        "    subset = \"training\"\n",
        ")\n",
        "\n",
        "Validation_generator = Validation_datagen.flow_from_directory(\n",
        "    base_dir,\n",
        "    target_size = (150,150),\n",
        "    color_mode = \"rgb\",\n",
        "    class_mode = \"categorical\",\n",
        "    batch_size = 16,\n",
        "    shuffle = False,\n",
        "    subset = \"validation\"\n",
        ")"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Found 1751 images belonging to 3 classes.\n",
            "Found 437 images belonging to 3 classes.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zIma0sebekX_"
      },
      "source": [
        "import tensorflow as tf\n",
        "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import Conv2D, MaxPooling2D, Activation, Dense, Flatten\n",
        "from tensorflow.keras.layers import Dropout\n",
        "from tensorflow.keras.optimizers import Adam"
      ],
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WCvOTRmMeKdj"
      },
      "source": [
        "# Pembuatan Model dengan tipe Sequential\n",
        "Model = Sequential(\n",
        "    [\n",
        "     # Activation menggunakan relu + softmax\n",
        "     Conv2D(32, (3,3), strides = (1,1), activation = 'relu' , input_shape = (150,150,3)),\n",
        "     MaxPooling2D(pool_size = (2,2), padding = 'valid'),\n",
        "     Conv2D(64, (3,3), strides = (1,1), activation = 'relu' ),\n",
        "     MaxPooling2D(pool_size = (2,2), padding = 'valid'),\n",
        "     Conv2D(128, (3,3), strides = (1,1), activation = 'relu' ),\n",
        "     MaxPooling2D(pool_size = (2,2), padding = 'valid'),\n",
        "     Flatten(),\n",
        "     # Droupout\n",
        "     Dropout(0.2),\n",
        "     Dense(128, activation = 'relu'),\n",
        "     Dense(3, activation='softmax')\n",
        "    ]\n",
        ")"
      ],
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "plZ4mp2jeOoo"
      },
      "source": [
        "# Penggunaan Optimizer 'Adam' dengan learning rate 0.00146\n",
        "Adam(learning_rate=0.00146, name='Adam')\n",
        "Model.compile(optimizer = 'Adam',loss = 'categorical_crossentropy',metrics = ['accuracy'])"
      ],
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_DJUzsDxe03M"
      },
      "source": [
        "ACCURACY_THRESHOLD = 0.92\n",
        "\n",
        "# Implement callback function to stop training\n",
        "# when accuracy reaches e.g. ACCURACY_THRESHOLD = 0.92\n",
        "class myCallback(tf.keras.callbacks.Callback):\n",
        "\tdef on_epoch_end(self, epoch, logs={}):\n",
        "\t\tif(logs.get('accuracy') > ACCURACY_THRESHOLD):\n",
        "\t\t\tprint(\"\\nReached %2.2f%% accuracy, so stopping training!!\" %(ACCURACY_THRESHOLD*100))\n",
        "\t\t\tself.model.stop_training = True\n",
        "\n",
        "# Instantiate a callback object     \n",
        "callbacks = myCallback()"
      ],
      "execution_count": 16,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "oVQm2pwqeR8X",
        "outputId": "77b92901-2435-443a-bd3e-0eb28e16b677"
      },
      "source": [
        "def scheduler(epoch, lr):\n",
        "  if epoch < 5:\n",
        "    return lr\n",
        "  else:\n",
        "    return lr * tf.math.exp(-0.1)\n",
        "\n",
        "lr_schedule = tf.keras.callbacks.LearningRateScheduler(scheduler, verbose=1)\n",
        "\n",
        "Model.summary()"
      ],
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d (Conv2D)              (None, 148, 148, 32)      896       \n",
            "_________________________________________________________________\n",
            "max_pooling2d (MaxPooling2D) (None, 74, 74, 32)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_1 (Conv2D)            (None, 72, 72, 64)        18496     \n",
            "_________________________________________________________________\n",
            "max_pooling2d_1 (MaxPooling2 (None, 36, 36, 64)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_2 (Conv2D)            (None, 34, 34, 128)       73856     \n",
            "_________________________________________________________________\n",
            "max_pooling2d_2 (MaxPooling2 (None, 17, 17, 128)       0         \n",
            "_________________________________________________________________\n",
            "flatten (Flatten)            (None, 36992)             0         \n",
            "_________________________________________________________________\n",
            "dropout (Dropout)            (None, 36992)             0         \n",
            "_________________________________________________________________\n",
            "dense (Dense)                (None, 128)               4735104   \n",
            "_________________________________________________________________\n",
            "dense_1 (Dense)              (None, 3)                 387       \n",
            "=================================================================\n",
            "Total params: 4,828,739\n",
            "Trainable params: 4,828,739\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Wl8-y8iMeUgs",
        "outputId": "efada0b1-bc12-48ae-f2c5-b898ec5ddbb7"
      },
      "source": [
        "# Proses Training \n",
        "batch_size = 16\n",
        "with tf.device(\"/device:GPU:0\"):\n",
        "  history = Model.fit(Train_generator, \n",
        "                    epochs =  40, \n",
        "                    steps_per_epoch = 1751//batch_size, \n",
        "                    validation_data = Validation_generator, \n",
        "                    verbose = 1, \n",
        "                    validation_steps = 437//batch_size,\n",
        "                    callbacks =[lr_schedule, callbacks])"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/40\n",
            "\n",
            "Epoch 00001: LearningRateScheduler reducing learning rate to 0.0010000000474974513.\n",
            "109/109 [==============================] - 82s 756ms/step - loss: 0.5476 - accuracy: 0.8012 - val_loss: 0.3862 - val_accuracy: 0.8542\n",
            "Epoch 2/40\n",
            "\n",
            "Epoch 00002: LearningRateScheduler reducing learning rate to 0.0010000000474974513.\n",
            "109/109 [==============================] - 82s 754ms/step - loss: 0.2993 - accuracy: 0.8882 - val_loss: 0.3677 - val_accuracy: 0.8449\n",
            "Epoch 3/40\n",
            "\n",
            "Epoch 00003: LearningRateScheduler reducing learning rate to 0.0010000000474974513.\n",
            "109/109 [==============================] - 82s 753ms/step - loss: 0.2238 - accuracy: 0.9233 - val_loss: 0.2087 - val_accuracy: 0.9282\n",
            "\n",
            "Reached 92.00% accuracy, so stopping training!!\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "resources": {
            "http://localhost:8080/nbextensions/google.colab/files.js": {
              "data": "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",
              "ok": true,
              "headers": [
                [
                  "content-type",
                  "application/javascript"
                ]
              ],
              "status": 200,
              "status_text": ""
            }
          },
          "base_uri": "https://localhost:8080/",
          "height": 415
        },
        "id": "XiBlBM2_gvTU",
        "outputId": "aad38c42-94e9-4d65-958c-418b1cdb7cb0"
      },
      "source": [
        "import numpy as np\n",
        "from google.colab import files\n",
        "from keras.preprocessing import image\n",
        "import matplotlib.pyplot as plt\n",
        "import matplotlib.image as mpimg\n",
        "import matplotlib.pyplot as plt\n",
        "import matplotlib.image as mpimg\n",
        "%matplotlib inline\n",
        "\n",
        "# Upload File ke google colabs\n",
        "uploaded = files.upload()\n",
        "\n",
        "# pengkondisian\n",
        "for file_upload in uploaded.keys():\n",
        "\n",
        "  path = file_upload\n",
        "  img = image.load_img(path, target_size=(150,150))\n",
        "  imgplot = plt.imshow(img)\n",
        "  x = image.img_to_array(img)\n",
        "  x = np.expand_dims(x, axis=0)\n",
        "\n",
        "  # Membuat numpy vstack array untuk hasil prediksi \n",
        "  images = np.vstack([x])\n",
        "  classes = Model.predict(images, batch_size=16)\n",
        "  \n",
        "  print(\"\\n\")\n",
        "  print('Hasil Prediksi : ',classes[0],'\\n')\n",
        "\n",
        "  if classes[0][0] == 1:\n",
        "    print('Kategori Gambar : Paper')\n",
        "  elif classes[0][1] == 1:\n",
        "    print('Kategori Gambar : Rock')\n",
        "  else:\n",
        "    print('Kategori Gambar : Scissor')"
      ],
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "\n",
              "     <input type=\"file\" id=\"files-4644a2d0-d0a9-463f-a61c-6dbb8fbc2c90\" name=\"files[]\" multiple disabled\n",
              "        style=\"border:none\" />\n",
              "     <output id=\"result-4644a2d0-d0a9-463f-a61c-6dbb8fbc2c90\">\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 WhatsApp Image 2021-07-01 at 20.32.11.jpeg to WhatsApp Image 2021-07-01 at 20.32.11 (1).jpeg\n",
            "\n",
            "\n",
            "Hasil Prediksi :  [0. 1. 0.] \n",
            "\n",
            "Kategori Gambar : Rock\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": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "id": "uvkuUuVQeZDj",
        "outputId": "c8710b3f-435f-442a-a847-d3fc4090ae89"
      },
      "source": [
        "# Mengambil Nilai Accuracy \n",
        "plt.plot(history.history['accuracy'])\n",
        "plt.plot(history.history['val_accuracy'])\n",
        "plt.title('Accuracy')\n",
        "plt.ylabel('accuracy')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(['train', 'val'], loc='upper left')\n",
        "plt.show()"
      ],
      "execution_count": 25,
      "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"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "id": "ITkLe_NPehr3",
        "outputId": "58982f9b-f5cf-49b4-c522-0107e2279840"
      },
      "source": [
        "plt.plot(history.history['loss'])\n",
        "plt.plot(history.history['val_loss'])\n",
        "plt.title('Loss')\n",
        "plt.ylabel('loss')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(['train', 'val'], loc='upper left')\n",
        "plt.show()"
      ],
      "execution_count": 26,
      "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"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "flVVnSkYemiG",
        "outputId": "3b3ef9dd-21f5-4d9f-bb5d-66bb44195cb8"
      },
      "source": [
        "import warnings\n",
        "# Menghilangkan Warning\n",
        "warnings.filterwarnings('ignore')\n",
        "# Convert Model.\n",
        "converter = tf.lite.TFLiteConverter.from_keras_model(Model)\n",
        "tflite_model = converter.convert()\n",
        "# Save the model.\n",
        "with tf.io.gfile.GFile('model.tflite', 'wb') as f:\n",
        "  f.write(tflite_model)"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "INFO:tensorflow:Assets written to: /tmp/tmpf8b6w6ss/assets\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "INFO:tensorflow:Assets written to: /tmp/tmpf8b6w6ss/assets\n"
          ],
          "name": "stderr"
        }
      ]
    }
  ]
}