RiceClassification.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0Et3PVTrHg1P"
      },
      "source": [
        "## Kaggle Load Dataset"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "0Bq49a8FFd4H"
      },
      "outputs": [],
      "source": [
        "from google.colab import files\n",
        "#Upload Kaggle API Key\n",
        "files.upload()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "id": "auDwPqjXFh4W"
      },
      "outputs": [],
      "source": [
        "!mkdir ~/.kaggle \n",
        "!cp kaggle.json ~/.kaggle/ \n",
        "!chmod 600 ~/.kaggle/kaggle.json\n",
        "# !kaggle datasets list"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Dos1hg6pFnfe",
        "outputId": "417ec7f6-844f-460c-98ab-bda28eefc3b6"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading rice-image-dataset.zip to /content\n",
            " 91% 200M/219M [00:02<00:00, 87.4MB/s]\n",
            "100% 219M/219M [00:02<00:00, 103MB/s] \n"
          ]
        }
      ],
      "source": [
        "!kaggle datasets download -d muratkokludataset/rice-image-dataset"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qCMJ9GOjHlPc"
      },
      "source": [
        "## Extract Dataset to new Directory"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "id": "F4ZuA1-WEgs2"
      },
      "outputs": [],
      "source": [
        "import os, zipfile"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "id": "C6-3IM0hFX6e"
      },
      "outputs": [],
      "source": [
        "zipPath = '../content/rice-image-dataset.zip'\n",
        "zipFile = zipfile.ZipFile(zipPath, 'r')\n",
        "zipFile.extractall('../content/RiceDataset/')\n",
        "zipFile.close()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "id": "Ydc0-cBHHQoF"
      },
      "outputs": [],
      "source": [
        "FILE_PATH = '../content/RiceDataset/Rice_Image_Dataset/'"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ah6YXXtAHqT8"
      },
      "source": [
        "## ImageDataGenerator"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "id": "tWKejjsQHcLc"
      },
      "outputs": [],
      "source": [
        "from tensorflow.keras.preprocessing.image import ImageDataGenerator"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 31,
      "metadata": {
        "id": "BPIa2OT6JMfI"
      },
      "outputs": [],
      "source": [
        "IMG_HEIGHT = 150\n",
        "IMG_WIDTH = 150\n",
        "BATCH_SIZE = 64"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 32,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6TXf3iaQHYb9",
        "outputId": "02cb6981-1ff1-4211-e08b-99e9906c76a3"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Found 60000 images belonging to 5 classes.\n",
            "Found 15000 images belonging to 5 classes.\n"
          ]
        }
      ],
      "source": [
        "train_datagen = ImageDataGenerator(\n",
        "    rescale=1.0/255,\n",
        "    rotation_range = 20,\n",
        "    horizontal_flip = True,\n",
        "    shear_range = 0.2,\n",
        "    fill_mode = 'nearest',\n",
        "    validation_split = 0.2\n",
        ")\n",
        "\n",
        "val_datagen = ImageDataGenerator(\n",
        "    rescale=1.0/255,\n",
        "    validation_split = 0.2\n",
        ")\n",
        "\n",
        "train_generator = train_datagen.flow_from_directory(\n",
        "    FILE_PATH,\n",
        "    target_size = (IMG_HEIGHT, IMG_WIDTH),\n",
        "    batch_size = BATCH_SIZE,\n",
        "    class_mode = 'categorical',\n",
        "    subset='training'\n",
        ")\n",
        "\n",
        "val_generator = val_datagen.flow_from_directory(\n",
        "    FILE_PATH,\n",
        "    target_size = (IMG_HEIGHT, IMG_WIDTH),\n",
        "    batch_size = BATCH_SIZE,\n",
        "    class_mode = 'categorical',\n",
        "    subset='validation'\n",
        ")\n",
        "\n",
        "target_names = ['Arborio', 'Basmati', 'Ipsala', 'Jasmine', 'Karacadag']"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kEHV_JhAQ1c8"
      },
      "source": [
        "## Model Architecture"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 23,
      "metadata": {
        "id": "jzY5mMy2MBZv"
      },
      "outputs": [],
      "source": [
        "# from tensorflow.keras.applications import ResNet50, ResNet152V2#MobileNetV2\n",
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import Flatten, Dense, Conv2D, MaxPooling2D, Dropout, BatchNormalization\n",
        "# from tensorflow.keras.layers.experimental.preprocessing import Rescaling\n",
        "# from tensorflow.keras.layers import Input"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 28,
      "metadata": {
        "id": "ZrnqBml4Qzla"
      },
      "outputs": [],
      "source": [
      
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model = Sequential([\n",
        "    Conv2D(filters = 16, kernel_size = (5, 5), padding = 'Same', activation = 'relu', input_shape = (IMG_HEIGHT, IMG_WIDTH, 3)),\n",
        "    MaxPooling2D(pool_size = (2,2)),\n",
        "    Dropout(0.2),\n",
        "    BatchNormalization(),\n",
        "\n",
        "    Conv2D(filters = 32, kernel_size = (3, 3), padding = 'Same', activation = 'relu'),\n",
        "    MaxPooling2D(pool_size = (2,2), strides = (2, 2)),\n",
        "    Dropout(0.2),\n",
        "\n",
        "    Conv2D(filters = 64, kernel_size = (3, 3), padding = 'Same', activation = 'relu'),\n",
        "    MaxPooling2D(pool_size = (2,2), strides = (2, 2)),\n",
        "    Dropout(0.2),\n",
        "\n",
        "    Flatten(),\n",
        "    Dense(64, activation='relu'),\n",
        "    Dropout(0.2),\n",
        "    Dense(5, activation='softmax')\n",
        "])"
      ],
      "metadata": {
        "id": "wpD9r7ACQVtF"
      },
      "execution_count": 39,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": 40,
      "metadata": {
        "id": "7OSjFmaSZsvs"
      },
      "outputs": [],
      "source": [
        "model.compile(\n",
        "    optimizer='adam',\n",
        "    loss='categorical_crossentropy',\n",
        "    metrics=['accuracy']\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 19,
      "metadata": {
        "id": "VTLVGdKeXOWk"
      },
      "outputs": [],
      "source": [
        "from tensorflow.keras.callbacks import Callback, EarlyStopping"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 21,
      "metadata": {
        "id": "Ma8ekiCiXMUG"
      },
      "outputs": [],
      "source": [
        "#Callback\n",
        "\n",
        "class TargetCallback(Callback):\n",
        "  def on_epoch_end(self, epoch, logs={}):\n",
        "    if(logs.get('accuracy') >= 0.95 and logs.get('val_accuracy') >= 0.95):\n",
        "      print(\"\\nAccuracy and Validation Accuracy has reached 95%!\\nStop Train!\")\n",
        "      self.model.stop_training = True\n",
        "Target = TargetCallback()\n",
        "\n",
        "EarlyStop = EarlyStopping(\n",
        "    monitor = 'val_loss',\n",
        "    min_delta = 0.0001,\n",
        "    patience = 5,\n",
        "    verbose = 1,\n",
        "    mode = 'auto'\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 41,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wpdreLarZJ3E",
        "outputId": "8ef75810-e4a7-44d5-cc5d-2ee48da284e7"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/100\n",
            "938/938 [==============================] - 1297s 1s/step - loss: 0.2447 - accuracy: 0.9134 - val_loss: 0.0988 - val_accuracy: 0.9626\n",
            "Epoch 2/100\n",
            "938/938 [==============================] - ETA: 0s - loss: 0.1264 - accuracy: 0.9588\n",
            "Accuracy and Validation Accuracy has reached 95%!\n",
            "Stop Train!\n",
            "938/938 [==============================] - 1299s 1s/step - loss: 0.1264 - accuracy: 0.9588 - val_loss: 0.0905 - val_accuracy: 0.9681\n"
          ]
        }
      ],
      "source": [
        "History = model.fit(\n",
        "    train_generator,\n",
        "    epochs =  100,\n",
        "    validation_data = val_generator,\n",
        "    callbacks = [Target, EarlyStop],\n",
        "    verbose = 1\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 42,
      "metadata": {
        "id": "ux2EImvjbv27"
      },
      "outputs": [],
      "source": [
        "import matplotlib.pyplot as plt"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 43,
      "metadata": {
        "id": "mfyDSTxqbQeE",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 608
        },
        "outputId": "a67e454e-8258-4e4d-984c-bb32484bb7b7"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Loss with Val_Loss Graph\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"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Acc with Val_Acc Graph\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEWCAYAAAB8LwAVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3dd3jUdbr38fedkBBKIEDooYQOiiIErCwoIqBrxUVwse0e8azrFtuqz9k9z5ZzLlCx97667lpW1/Ks0kQQGysBRVESAqEkoUhvIf1+/piJO8RABshkkszndV25+PW5vwF+93zL7/szd0dERKSquGgHICIi9ZMShIiIVEsJQkREqqUEISIi1VKCEBGRailBiIhItZQgJOaZWU8zczNrEsaxV5vZR3URl0i0KUFIg2Jm68ysxMxSq2z/PHiT7xmdyA6KpaWZ7TOzWdGOReRYKEFIQ7QWmFK5YmaDgebRC+d7JgLFwFgz61SXHxxOLUgkXEoQ0hD9BbgyZP0q4IXQA8ystZm9YGZbzWy9mf3WzOKC++LNbKaZbTOzXOC8as59xsw2mVmBmf2PmcUfQXxXAY8DXwJTq1z7DDP7xMx2mVmemV0d3N7MzO4JxrrbzD4KbhttZvlVrrHOzM4OLv/ezF4zsxfNbA9wtZmNMLNPg5+xycweNrPEkPOPM7N5ZrbDzLaY2f8xs05mVmhm7UKOGxr8/SUcQdmlEVGCkIZoMdDKzAYGb9yTgRerHPMQ0BroBYwikFCuCe67FvghcBKQAVxa5dw/A2VAn+Ax5wD/EU5gZtYDGA38NfhzZZV9s4KxtQeGAF8Ed88EhgGnAW2B3wAV4XwmcCHwGpAS/Mxy4EYgFTgVGANcH4whGXgPmA10CZZxvrtvBhYCk0KuewXwsruXhhmHNDJKENJQVdYixgIrgYLKHSFJ4w533+vu64B7CNzwIHATvN/d89x9BzA95NyOwLnAr919v7t/C9wXvF44rgC+dPdvgJeB48zspOC+y4H33P0ldy919+3u/kWwZvMT4FfuXuDu5e7+ibsXh/mZn7r7m+5e4e4H3H2puy9297Jg2Z8gkCQhkBg3u/s97l4U/P38K7jveYI1nuDvcAqB37PEKLVXSkP1F2ARkE6V5iUC35wTgPUh29YDXYPLXYC8Kvsq9Qieu8nMKrfFVTn+cK4EngJw9wIz+4BAk9PnQDdgTTXnpAJJh9gXjoNiM7N+wL0EakfNCfw/XxrcfagYAN4CHjezdKA/sNvdPzvKmKQRUA1CGiR3X0+gs/pc4B9Vdm8DSgnc7Ct159+1jE0EbpSh+yrlEehgTnX3lOBPK3c/rqaYzOw0oC9wh5ltNrPNwMnA5cHO4zygdzWnbgOKDrFvPyEd8MFv9u2rHFN1SubHgCygr7u3Av4PUJnt8gg0u32PuxcBrxKoRVyBag8xTwlCGrKfAme5+/7Qje5eTuBG979mlhxs+7+Jf/dTvAr80szSzKwNcHvIuZuAucA9ZtbKzOLMrLeZjaJmVwHzgEEE+heGAMcDzYAJBPoHzjazSWbWxMzamdkQd68AngXuNbMuwU70U82sKbAKSDKz84Kdxb8FmtYQRzKwB9hnZgOAn4Xs+yfQ2cx+bWZNg7+fk0P2vwBcDVyAEkTMU4KQBsvd17h75iF2/4LAt+9c4CPgbwRuwhBoApoDLAeW8f0ayJVAIvANsJNAB3Dnw8ViZkkE+jYecvfNIT9rCdxor3L3DQRqPDcDOwh0UJ8YvMQtwFfAkuC+O4E4d99NoIP5aQI1oP3AQaOaqnELgf6OvcGyvlK5w933Eui3OR/YDOQAZ4bs/5hA5/iyYC1NYpjphUEiEsrM3gf+5u5PRzsWiS4lCBH5jpkNJ9BM1i1Y25AYpiYmEQHAzJ4n8IzEr5UcBFSDEBGRQ1ANQkREqhXRB+XMbDzwABAPPO3uM6rs70FgZEl7AiM3prp7vpmdSeDp1UoDgMnu/uahPis1NdV79uxZyyUQEWncli5dus3dqz5bA0SwiSn4QM8qAkPq8gkM35sSnIKg8pi/A/909+fN7CzgGne/osp12gKrgTR3LzzU52VkZHhm5qFGPIqISHXMbKm7Z1S3L5JNTCOA1e6e6+4lBOalubDKMYOA94PLC6rZD4GJ1GYdLjmIiEjti2SC6MrBc8Tk8++5cCotBy4JLl8MJIdONxw0GXipug8ws2lmlmlmmVu3bq2FkEVEpFK0O6lvAUaZ2ecEZpssIDBVMQBm1hkYTOCp1+9x9yfdPcPdM9q3r7YJTUREjlIkO6kLOHhCtDRCpmQGcPeNBGsQZtYSmOjuu0IOmQS8ofnoRUTqXiRrEEuAvmaWHnyb1WTg7dADzCy18i1fwB38e66cSlM4RPOSiIhEVsQShLuXATcQaB5aCbzq7l+b2R/N7ILgYaOBbDNbBXQE/rfyfAu8fL4b8EGkYhQRkUNrNE9Sa5iriMiRO9wwV71RTkSkvisrgaJdULQbDgT/LNoFB3YGlpu3g4xrar7OEVKCEBGJNHco2R+8qYfe4EOWv7v5V7NcWsNjYGnDlSBERKKmorzKjf1QN/VD3Pwryg5//aatICkFmrUO/Nmud3A9JfBnUutqloPHJiRFpMhKECISO0oPHP5GfrgbfPGew187rsnBN+9mbaBNz0Pf1EOXk1pDXHyd/AqOhBKEiDQcFRVQsvfommkO7ILy4sNfP6HFwTfy1mnQ6fjqb+pVb/CJLcCsbn4PdUQJQkTqVnlplZv3ETTZFO8BrzjMxe3739RbdT74Rv7dN/oqzTdJraFJYp39GhoCJQgROTLugU7TI22mqVwu3X/468cnHnzzbtEeUvvW3A7fLAUSkyEu2jMINR5KECKx6LsO1zCbZqp+o6+oYfabxOSDb+Rte9XcDl+5nNCsbn4HUiMlCJGGqqw4jJt61W/xwaRQvPvw17b479+8U7rX3A7frE1gNE68bi2Ngf4WRaLFHYr3Hvlomsqbf1nR4a+f0PzgG3mrrtDhuEM001RZboQdrnLklCBEjkV5KRTtqdLRegRNNjV2uLY6+Oad2q+am3qbar7Rt4YmTevs1yCNkxKExDb34Nj4MNveq36jL9l3+OtXdrhW3tSbt4O2vWtuh09KCTTVqMNVokgJQhq+iorA8MejaaYp2g3lJYe/fmLLg2/ebXrW0A4f8u0+oZmaaqTBUoKQ+qFyMrLv3ch31vztvmgPcJhZiS2+mrHxXWtuh69cV4erxCj9y5fa4R5obgm3aabqctmBw1+/SbODb97JnaHDwPCGTia21Ld4kaOgBCH/Vjk2vnIK4SNppglrMrLWwYnIgjfv1D5Vbuoph26yUYerSJ1TgmhsSg8c2dw0ocslew9/7biEg2/ezdtC2/QammmCy01b1cvJyETk0JQg6puDJiM7gtE0lcs1TUaW2PLgb+eVDz/VNIVBUuvAuHo11YjEDCWISCgvrebmvavmZppwJiOzuO83v7Tqcph2+DYHj42PT6i734OINGhKENU5aDKyI2ymCWcysiZJB9/IW3aE1P7hDZ1MbKmx8SJSJ5QgCnfA6/9R5dt9GJORVX37U+VkZDUOnWwdsbc/iYjUJiWI+IRAckhKgZQeDfrtTyIitUkJomkyXPt+tKMQEal31JgtIiLVUoIQEZFqKUGIiEi1lCBERKRaEU0QZjbezLLNbLWZ3V7N/h5mNt/MvjSzhWaWFrKvu5nNNbOVZvaNmfWMZKwiInKwiCUIM4sHHgEmAIOAKWY2qMphM4EX3P0E4I/A9JB9LwB3u/tAYATwbaRiFRGR74tkDWIEsNrdc929BHgZuLDKMYOAyjGmCyr3BxNJE3efB+Du+9y9MIKxiohIFZFMEF2BvJD1/OC2UMuBS4LLFwPJZtYO6AfsMrN/mNnnZnZ3sEZyEDObZmaZZpa5devWCBRBRCR2RbuT+hZglJl9DowCCoByAg/wjQzuHw70Aq6uerK7P+nuGe6e0b59+zoLWkQkFkQyQRQA3ULW04LbvuPuG939Enc/Cfiv4LZdBGobXwSbp8qAN4GhEYxVRESqiGSCWAL0NbN0M0sEJgNvhx5gZqlmVhnDHcCzIeemmFllteAs4JsIxioi0iAVlZaTtyMyXbQRSxDBb/43AHOAlcCr7v61mf3RzC4IHjYayDazVUBH4H+D55YTaF6ab2ZfAQY8FalYRUQamvIK5++ZeZw1cyHX/3UZ7l7rn2GRuGg0ZGRkeGZmZrTDEBGJKHfn/axvuXN2Fqu27OPEtNbcNmEAp/VOParrmdlSd8+obp9mcxURaSCWbdjJjFlZfLZ2B+mpLXj0x0OZcHwnLEKvAlaCEBGp59Zs3cfds7OZ/fVmUls25X8uOp7LhncjIT6yA1GVIERE6qkte4q4/70cXs3MI6lJHDeN7cdPz0inRdO6uXUrQYiI1DN7ikp54oM1PPPRWsornCtO6cENZ/UhtWXTOo1DCUJEpJ4oLivnxcUbePj9HHYWlnLhkC7cPLY/3ds1j0o8ShAiIlFWUeG8tbyAmXNWUbDrACP7pnLb+AEc37V1VONSghARiRJ354NVW7lzdjYrN+3h+K6tmDFxMCP71o+pg5QgRESiYHneLmbMyuLT3O10b9ucB6ecxA8HdyYuLjJDVo+GEoSISB1au20/M+dm886Xm2jbIpHfnz+Iy0/uQWKTaM+d+n1KECIidWDr3mIenJ/DS59tILFJHL8c05drR6aTnJQQ7dAOSQlCRCSC9hWX8eSiXJ7+MJeSsgqmjOjOL8b0oUNyUrRDq5EShIhIBJSUVfDSZxt4cH4O2/eXcN7gztwyrj/pqS2iHVrYlCBERGpRRYXzz682MXNONht2FHJKr7Y8M2EgQ7qlRDu0I6YEISJSSz7K2caM2StZUbCHAZ2S+fM1wxnVr33EJtOLNCUIEZFjtKJgN3fOzuLDnG10TWnGvZNO5KIhXevVkNWjoQQhInKUNmwv5J552bz1xUZSmifw2/MGMvWUHiQlxEc7tFqhBCEicoS27yvmofdX89d/rSc+zvj5mb25blRvWtXjIatHQwlCRCRMhSVlPPPhWp5YlEthSRmXDe/Gr8/uR8dW9X/I6tFQghARqUFpeQWvLMnj/vdy2LavmHHHdeTWcQPo06FltEOLKCUIEZFDcHdmrdjM3XOyWbttP8N7tuGJK4YxrEebaIdWJ5QgRESq8ema7cyYncXyvF3069iSZ67K4KwBHRrskNWjoQQhIhJi5aY93Dk7i4XZW+ncOom7Lj2BiUPTiG/gQ1aPhhKEiAiQv7OQe+et4o3PC0hu2oQ7JgzgqtN6Npohq0dDCUJEYtrO/SU8smA1L3y6Hgym/aAX14/qQ+vmjWvI6tFQghCRmHSgpJxnP17L4wvXsL+kjEuHpfHrs/vRJaVZtEOrN5QgRCSmlJVX8NrSfO57bxVb9hRz9sAO3DpuAP07JUc7tHpHCUJEYoK7M/ebLdw1O4s1W/cztHsKD00Zyoj0ttEOrd6KaIIws/HAA0A88LS7z6iyvwfwLNAe2AFMdff84L5y4KvgoRvc/YJIxioijdeSdTuYMSuLpet30rt9C564YhjnDOoYU0NWj0bEEoSZxQOPAGOBfGCJmb3t7t+EHDYTeMHdnzezs4DpwBXBfQfcfUik4hORxi9ny17unJ3Neyu30CG5KdMvGcyPhqXRJL7+vf+5PopkDWIEsNrdcwHM7GXgQiA0QQwCbgouLwDejGA8IhIjNu0+wH3zVvHa0nxaJDbh1nH9+cnp6TRLjN0hq0cjkgmiK5AXsp4PnFzlmOXAJQSaoS4Gks2snbtvB5LMLBMoA2a4u5KHiBzW7sJSHv1gNX/+eB3u8JPT0/n5mX1o0yIx2qE1SNHupL4FeNjMrgYWAQVAeXBfD3cvMLNewPtm9pW7rwk92cymAdMAunfvXndRi0i9UlRazgufruORBWvYU1TKxUO6ctM5/Uhr0zzaoTVokUwQBUC3kPW04LbvuPtGAjUIzKwlMNHddwX3FQT/zDWzhcBJwJoq5z8JPAmQkZHhESmFiNRb5RXOP5blc9+8VWzcXcTo/u35zbgBDOrSKtqhNQqRTBBLgL5mlk4gMUwGLg89wMxSgR3uXgHcQWBEE2bWBih09+LgMacDd0UwVhFpQNyd97O+5c7ZWazaso8T01ozc9KJnNY7NdqhNSoRSxDuXmZmNwBzCAxzfdbdvzazPwKZ7v42MBqYbmZOoInp58HTBwJPmFkFEEegD+Kb732IiMScZRt2MmNWFp+t3UF6agse/fFQJhzfSUNWI8DcG0fLTEZGhmdmZkY7DBGJkDVb93H37Gxmf72Z1JZN+dXZfZk8vBsJGrJ6TMxsqbtnVLcv2p3UIiKHtWVPEfe/l8OrmXkkNYnjprH9+OkZ6bRoqttXpOk3LCL10p6iUp78IJenP8qlvMK54pQe3HBWH1JbNo12aDFDCUJE6pXisnJeXLyBh9/PYWdhKRec2IWbz+lHj3Ytoh1azFGCEJF6oaLCeWt5ATPnrKJg1wFG9k3ltvEDOL5r62iHFrOUIEQkqtydRTnbmDEri5Wb9nBcl1bMmDiYkX3bRzu0mKcEISJRszxvF3fOzuKTNdvp3rY5D045iR8O7kxcDL7/uT5SghCROrdu237unpvNO19uom2LRH5//iAuP7kHiU00ZLU+UYIQkTqzdW8xD87P4aXPNpDYJI5fjunLtSPTSU7S+5/roxoThJmdD7wTnA5DROSI7Ssu46lFuTz1YS4lZRVMGdGdX4zpQ4fkpGiHJocRTg3iMuB+M3udwHQZWRGOSUQaiZKyCl76bAMPzs9h+/4SzhvcmVvG9Sc9VUNWG4IaE4S7TzWzVsAU4M/BeZOeA15y972RDlBEGp6KCuefX21i5pxsNuwo5JRebXlmwkCGdEuJdmhyBMLqg3D3PWb2GtAM+DWBl/vcamYPuvtDkQxQRBqWj1cHhqx+VbCbAZ2S+fM1wxnVr70m02uAwumDuAC4BugDvACMcPdvzaw5gdeHKkGICCsKdnPn7Cw+zNlG15Rm3DvpRC4a0lVDVhuwcGoQE4H73H1R6EZ3LzSzn0YmLBFpKDZsL+Seedm89cVGUpon8NvzBjL1lB4kJej9zw1dOAni98CmyhUzawZ0dPd17j4/UoGJSP22fV8xD72/mr/+az3xccbPz+zNdaN600pDVhuNcBLE34HTQtbLg9uGRyQiEanXCkvKeObDtTyxKJfCkjIuG96NX5/dj46tNGS1sQknQTRx95LKFXcvMbPECMYkIvVQaXkFryzJ44H5OWzdW8y44zpy67gB9OnQMtqhSYSEkyC2mtkFwVeEYmYXAtsiG5aI1BfuzqwVm7l7TjZrt+1neM82PD51GMN6tIl2aBJh4SSI/wT+amYPAwbkAVdGNCoRqRcW525n+qwsluftol/HljxzVQZnDeigIasxIpwH5dYAp5hZy+D6vohHJSJRtXLTHu6ancWC7K10bp3EXZeewMShacRryGpMCetBOTM7DzgOSKr85uDuf4xgXCISBfk7C7l33ire+LyA5KZNuGPCAK46raeGrMaocB6UexxoDpwJPA1cCnwW4bhEpA7t3F/CIwtW88Kn68Fg2g96cf2oPrRuriGrsSycGsRp7n6CmX3p7n8ws3uAWZEOTEQi70BJOc99spbHFq5hf3EZE4emcePYfnRJaRbt0KQeCCdBFAX/LDSzLsB2oHPkQhKRSCsrr+C1pfnc994qtuwp5uyBHbh13AD6d0qOdmhSj4STIP6fmaUAdwPLAAeeimhUIhIR7s68b7Zw15xsVn+7j6HdU3hoylBGpLeNdmhSDx02QZhZHDDf3XcBr5vZP4Ekd99dJ9GJSK1Zsm4HM2ZlsXT9Tnq1b8HjU4cx7riOGrIqh3TYBOHuFWb2CHBScL0YKK6LwESkduRs2cuds7N5b+UWOiQ3Zfolg/nRsDSaxOv9z3J44TQxzTezicA/3N0jHZCI1I5Nuw9w37xVvLY0nxaJTbh1XH9+cno6zRI1ZFXCE06CuA64CSgzsyICT1O7u7eq6UQzGw88AMQDT7v7jCr7ewDPAu2BHcBUd88P2d+KwDsn3nT3G8Irkkhs211YymMfrOG5j9fiDtecns4NZ/ahTQtNoSZHJpwnqY9qWIOZxQOPAGOBfGCJmb3t7t+EHDYTeMHdnzezs4DpwBUh+/8EHPQeChGpXlFpOS98uo5HFqxhT1EpFw/pyo1j+9GtbfNohyYNVDgPyv2guu1VXyBUjRHAanfPDV7nZeBCAjWCSoMI1E4AFgBvhnzuMKAjMBvIqClOkVhVXuH8Y1k+981bxcbdRYzu357fjBvAoC41VvJFDiucJqZbQ5aTCNz4lwJn1XBeVwIT+1XKB06ucsxy4BICzVAXA8lm1g7YCdwDTAXOPtQHmNk0YBpA9+7dayqHSKPi7izI/pY7Z2WTvWUvJ6a1ZuakEzmtd2q0Q5NGIpwmpvND182sG3B/LX3+LcDDZnY1gaakAgIvJLoeeNfd8w83BM/dnwSeBMjIyFAHusSMZRt2MmNWFp+t3UHPds155PKhnDu4k4asSq0Ka7K+KvKBgWEcVwB0C1lPC277jrtvJFCDIDhb7ER332VmpwIjzex6oCWQaGb73P32o4hXpNFYs3Ufd8/OZvbXm0lt2ZQ/XXQ8k4d3I0FDViUCwumDeIjA09MAccAQAk9U12QJ0NfM0gkkhsnA5VWunQrscPcK4A4CI5pw9x+HHHM1kKHkILHs2z1F3D8/h1eW5JHUJI6bxvbjp2ek06Lp0XzHEwlPOP+6MkOWy4CX3P3jmk5y9zIzuwGYQ2CY67Pu/rWZ/RHIDL6hbjQw3cycQBPTz4+0ACKN2Z6iUp78IJenP8qlvMK54pQe3HBWH1JbNo12aBIDrKZn38ysBVDk7uXB9XigqbsX1kF8YcvIyPDMzMyaDxRpAIrLynlx8QYefj+HnYWlXHBiF24+px892rWIdmjSyJjZUnevdqRoWE9SExhJVPkmuWbAXOC02glPRCpVVDhvLS/gnrmryN95gDP6pHL7hAEc37V1tEOTGBROgkgKfc2ou+8zMz15I1KL3J1FOduYMSuLlZv2cFyXVky/ZDAj+7aPdmgSw8JJEPvNbKi7L4PvHmA7ENmwRGLHl/m7mDEri0/WbKdb22Y8MHkI55/QhTi9/1miLJwE8Wvg72a2kcA8TJ2AyyIalUgMWLdtP3fPzeadLzfRtkUivz9/EJef3IPEJhqyKvVDOA/KLTGzAUD/4KZsdy+NbFgijdfWvcU8OD+Hlz7bQGKTOH45pi/XjkwnOUnvf5b6JZznIH4O/NXdVwTX25jZFHd/NOLRiTQi+4rLeGpRLk99mEtxWQVTRnTjl2P60iE5KdqhiVQrnCama939kcoVd99pZtcCShAiYSgpq+Clzzbw4Pwctu8v4bzBnbn5nH70at8y2qGJHFY4CSLezKzyZUHB5yA0sbxIDSoqnH9+tYmZc7LZsKOQU3q15ZkJAxnSLSXaoYmEJZwEMRt4xcyeCK5fB8yKXEgiDd/HqwNDVr8q2M2ATsk8d81wRvdrr8n0pEEJJ0HcRmBK7f8Mrn9JYCSTiFSxomA3d87O4sOcbXRNaca9k07kwiFdideQVWmAwhnFVGFm/wJ6A5OAVOD1SAcm0pDk7Shk5txs3vpiIynNE/jteQOZekoPkhL0/mdpuA6ZIMysHzAl+LMNeAXA3c+sm9BE6r/t+4p5eMFqXly8nvg44/rRvfnP0b1ppSGr0ggcrgaRBXwI/NDdVwOY2Y11EpVIPVdYUsYzH67liUW5FJaUcdnwbvxqTD86tdaQVWk8DpcgLiHwDocFZjYbeJnAk9QiMau0vIJXluTxwPwctu4tZtxxHbl1XH/6dEiOdmgite6QCcLd3wTeDE73fSGBKTc6mNljwBvuPreOYhSJOndn9orN3D0nm9xt+xnesw2PTx3GsB5toh2aSMSE00m9H/gb8DczawP8iMDIJiUIiQmLc7czfVYWy/N20bdDS56+MoMxAztoyKo0ekf0vkJ33wk8GfwRadRWbtrDXbOzWJC9lc6tk7jr0hOYODRNQ1YlZuiFtiJVFOw6wL1zV/GPz/NJbtqEOyYM4KrTemrIqsQcJQiRoJ37S3h04Wqe/3Q9ANN+0IvrR/WhdXMNWZXYpAQhMe9ASTnPfbKWxxauYX9xGROHpnHj2H50SWkW7dBEokoJQmJWWXkFry/L5755OWzeU8TZAztw67gB9O+kIasioAQhMcjdmffNFu6ak83qb/cxtHsKD045iRHpbaMdmki9ogQhMSVz3Q6mz8pi6fqd9GrfgsenDmPccR01ZFWkGkoQEhNytuzlztnZvLdyCx2SmzL9ksH8aFgaTeL1/meRQ1GCkEZt0+4D3D8vh78vzaNFYhNuHdefn5yeTrNEDVkVqYkShDRKuwtLeeyDNTz38Vrc4ZrT07nhzD60aaGXIYqESwlCGpWi0nJe+HQdjyxYw56iUi4e0pUbx/ajW9vm0Q5NpMGJaIIws/HAA0A88LS7z6iyvwfwLNAe2AFMdff84PY3gDggAXjI3R+PZKzSsJVXOP9Yls9981axcXcRo/u35zfjBjCoS6tohybSYEUsQZhZPPAIMBbIB5aY2dvu/k3IYTOBF9z9eTM7C5gOXAFsAk5192IzawmsCJ67MVLxSsPk7izI/pY7Z2WTvWUvJ6a1ZuakEzmtd2q0QxNp8CJZgxgBrHb3XAAze5nAtOGhCWIQcFNweQHwJoC7l4Qc05RATULkIMs27GTGrCw+W7uDnu2a88jlQzl3cCcNWRWpJZFMEF2BvJD1fODkKscsJ/BiogeAi4FkM2vn7tvNrBvwDtAHuLW62oOZTQOmAXTv3r32SyD10pqt+7h7djazv95Masum/Omi45k8vBsJGrIqUqui3Ul9C/CwmV0NLAIKgHIAd88DTjCzLgReXPSau28JPdndv5t6PCMjw+sycKl73+4p4v75ObyyJI+kJnHcNLYfPz0jnRZNo/3PWKRxiuT/rAKgW8h6WnDbd4K1gksAgn0NE919V9VjzGwFMBJ4LYLxSj21p6iUJz/I5ZmP1lJWUcEVp+P6HegAABKXSURBVPTghrP6kNqyabRDE2nUIpkglgB9zSydQGKYDFweeoCZpQI73L0CuIPAiCbMLA3Y7u4Hgm+xOwO4L4KxSj1UXFbOi4s38PD7OewsLOWCE7tw8zn96NGuRbRDE4kJEUsQ7l5mZjcAcwgMc33W3b82sz8Cme7+NjAamG5mTqCJ6efB0wcC9wS3GzDT3b+KVKxSv1RUOG8v38jMudnk7zzAGX1SuX3CAI7v2jraoYnEFHNvHE33GRkZnpmZGe0w5Bi4O4tytjFjVhYrN+3huC6tuH3CAEb2bR/t0EQaLTNb6u4Z1e1T757UC1/m72LGrCw+WbOdbm2b8cDkIZx/Qhfi9P5nkahRgpCoWrdtP3fPzeadLzfRtkUivz9/EJef3IPEJhqyKhJtShASFVv3FvPQ+zn87V8bSIiP45dn9eHaH/QiOUnvfxapL5QgpE7tKy7jqUW5PPVhLsVlFUwZ0Y1fjulLh+SkaIcmIlUoQUidKCmr4KXPNvDg/By27y/hvMGdufmcfvRq3zLaoYnIIShBSERVVDjvfLWJmXOzWb+9kJPT2/LMuQMZ0i0l2qGJSA2UICRiPl4dGLL6VcFuBnRK5rlrhjO6X3tNpifSQChBSK1bUbCbO2dn8WHONrqmNOPeSSdy4ZCuxGvIqkiDogQhtSZvRyH3zM3mzS82ktI8gd+eN5Cpp/QgKUHvfxZpiJQg5Jht31fMwwtW8+Li9cTHGdeP7s11o3rTupmGrIo0ZEoQctQKS8p45sO1PLEol8KSMi4b3o1fjelHp9YasirSGChByBErLa/g1cw87n8vh617izlnUEd+M74/fTokRzs0EalFShASNndn9orN3D0nm9xt+xnesw2PTx3KsB5tox2aiESAEoSEZXHudqbPymJ53i76dmjJ01dmMGZgBw1ZFWnElCDksFZu2sNds7NYkL2Vzq2TuOvSE5g4NE1DVkVigBKEVKtg1wHunbuKf3yeT3LTJtw+YQBXn9ZTQ1ZFYogShBxk5/4SHl24muc/XQ/AtJG9+Nno3qQ0T4xyZCJS15QgBIADJeU898laHlu4hn3FZVw6NI0bx/ajS0qzaIcmIlGiBBHjysoreH1ZPvfNy2HzniLOHtiBW8cNoH8nDVkViXVKEDHK3Zn3zRbumpPN6m/3cVL3FB6YPISTe7WLdmgiUk8oQcSgzHU7mD4ri6Xrd9KrfQsenzqMccd11JBVETmIEkQMydmylztnZ/Peyi10SG7K9EsG86NhaTSJ1/ufReT7lCBiwKbdB7h/Xg5/X5pHi8Qm3DquPz85PZ1miRqyKiKHpgTRiO0uLOWxD9bw3MdrcYdrTk/n52f2oW0LDVkVkZopQTRCRaXlvPDpOh5ZsIY9RaVcPKQrN47tR7e2zaMdmog0IEoQjUh5hfPG5wXcOzebjbuLGNWvPbeNH8CgLq2iHZqINEBKEI2Au7Mg+1vunJVN9pa9nJDWmpk/OpHT+qRGOzSReq+0tJT8/HyKioqiHUpEJSUlkZaWRkJC+C/yimiCMLPxwANAPPC0u8+osr8H8CzQHtgBTHX3fDMbAjwGtALKgf9191ciGWtDtWzDTmbMyuKztTvo2a45j1w+lHMHd9KQVZEw5efnk5ycTM+ePRvt/xt3Z/v27eTn55Oenh72eRFLEGYWDzwCjAXygSVm9ra7fxNy2EzgBXd/3szOAqYDVwCFwJXunmNmXYClZjbH3XdFKt6GZs3Wfcyck82sFZtJbdmUP110PJOHdyNBQ1ZFjkhRUVGjTg4AZka7du3YunXrEZ0XyRrECGC1u+cCmNnLwIVAaIIYBNwUXF4AvAng7qsqD3D3jWb2LYFaRswniG/3FHH//BxeWZJHUpM4bjy7H/8xMp0WTdVaKHK0GnNyqHQ0ZYzkXaUrkBeyng+cXOWY5cAlBJqhLgaSzaydu2+vPMDMRgCJwJqqH2Bm04BpAN27d6/V4OubPUWlPPlBLs98tJayigquOKUHN5zVh9SWTaMdmog0UtFuj7gFGGVmnwOjgAICfQ4AmFln4C/ANe5eUfVkd3/S3TPcPaN9+/Z1FXOdKi4r59mP1jLqrgU8vGA1Zw/qyHs3jeL3Fxyn5CDSCOzatYtHH330iM8799xz2bUrso0qkaxBFADdQtbTgtu+4+4bCdQgMLOWwMTKfgYzawW8A/yXuy+OYJz1UkWF8/byjcycm03+zgOc0SeV28YPYHBa62iHJiK1qDJBXH/99QdtLysro0mTQ9+i33333UiHFtEEsQToa2bpBBLDZODy0APMLBXYEawd3EFgRBNmlgi8QaAD+7UIxljvuDuLcrYxY1YWKzft4bgurZh+yWBG9m2cNSSR+uQP/+9rvtm4p1avOahLK/7v+ccdcv/tt9/OmjVrGDJkCAkJCSQlJdGmTRuysrJYtWoVF110EXl5eRQVFfGrX/2KadOmAdCzZ08yMzPZt28fEyZM4IwzzuCTTz6ha9euvPXWWzRrduzvcolYgnD3MjO7AZhDYJjrs+7+tZn9Ech097eB0cB0M3NgEfDz4OmTgB8A7czs6uC2q939i0jFWx98mb+LGbOy+GTNdrq1bcYDk4dw/gldiNP7n0UarRkzZrBixQq++OILFi5cyHnnnceKFSu+G4767LPP0rZtWw4cOMDw4cOZOHEi7dodPC1/Tk4OL730Ek899RSTJk3i9ddfZ+rUqcccW0SHvrj7u8C7Vbb9d8jya8D3agju/iLwYiRjq0/WbdvPzLnZ/PPLTbRtkcjvzx/E5Sf3ILFJtLuIRGLL4b7p15URI0Yc9KzCgw8+yBtvvAFAXl4eOTk530sQ6enpDBkyBIBhw4axbt26WolFYyOjaOveYh56P4e//WsDCfFx/PKsPlz7g14kJ4X/pKOINC4tWrT4bnnhwoW89957fPrppzRv3pzRo0dX+8R306b/HrASHx/PgQMHaiUWJYgo2FdcxlOLcnnqw1yKyyqYMqIbvxzTlw7JSdEOTUTqWHJyMnv37q123+7du2nTpg3NmzcnKyuLxYvrdryOEkQdKimr4OUlG3hwfg7b9pVw3uDO3HxOP3q1bxnt0EQkStq1a8fpp5/O8ccfT7NmzejYseN3+8aPH8/jjz/OwIED6d+/P6ecckqdxmbuXqcfGCkZGRmemZkZ7TCqVVHhvPPVJmbOzWb99kJOTm/LHecOZEi3lGiHJhLzVq5cycCBA6MdRp2orqxmttTdM6o7XjWICPt4dWDI6lcFuxnQKZnnrhnO6H7tY+LRfhFp2JQgImRFwW7unJ3Fhznb6JrSjHsnnciFQ7oSryGrItJAKEHUsrwdhdwzN5s3v9hISvMEfnveQKae0oOkBL3/WUQaFiWIWrJ9XzEPL1jNi4vXEx9nXD+6N9eN6k3rZhqyKiINkxLEMSosKeOZD9fyxKJcCkvKuGx4N341ph+dWmvIqog0bEoQR6m0vIJXM/O4/70ctu4t5pxBHfnN+P706ZAc7dBERGqF5nI4Qu7OrK82Me6+RfzXGyvo0bY5r//sVJ68MkPJQUSO2NFO9w1w//33U1hYWMsR/ZsSxBFYnLudix79hJ/9dRnxccbTV2bw9/88lWE92kY7NBFpoOpzglATUxiyNu/hrtnZvJ/1LZ1bJ3HXpScwcWiahqyKNDazbofNX9XuNTsNhgkzDrk7dLrvsWPH0qFDB1599VWKi4u5+OKL+cMf/sD+/fuZNGkS+fn5lJeX87vf/Y4tW7awceNGzjzzTFJTU1mwYEHtxo0SxGEV7DrAvXNX8Y/P80lu2oTbJwzg6tN6asiqiNSa0Om+586dy2uvvcZnn32Gu3PBBRewaNEitm7dSpcuXXjnnXeAwBxNrVu35t5772XBggWkpqZGJDYliGrs3F/CowtX8/yn6wGYNrIXPxvdm5TmiVGOTEQi6jDf9OvC3LlzmTt3LieddBIA+/btIycnh5EjR3LzzTdz22238cMf/pCRI0fWSTxKECGKSst57uN1PLpwNfuKy7h0aBo3ju1Hl5RjfzOTiEhN3J077riD66677nv7li1bxrvvvstvf/tbxowZw3//939Xc4XapQQBlJVX8PqyfO6bl8PmPUWMGdCB34wfQP9OGpUkIpEVOt33uHHj+N3vfsePf/xjWrZsSUFBAQkJCZSVldG2bVumTp1KSkoKTz/99EHnqokpQvJ2FHLNn5ew+tt9nNQ9hQcmD+HkXu1qPlFEpBaETvc9YcIELr/8ck499VQAWrZsyYsvvsjq1au59dZbiYuLIyEhgcceewyAadOmMX78eLp06RKRTuqYn+67tLyC6/6ylEkZ3Rh3XEfNsioSYzTdt6b7PqSE+DievXp4tMMQEal39KCciIhUSwlCRGJeY2lqP5yjKaMShIjEtKSkJLZv396ok4S7s337dpKSjmyW6ZjvgxCR2JaWlkZ+fj5bt26NdigRlZSURFpa2hGdowQhIjEtISGB9PT0aIdRL6mJSUREqqUEISIi1VKCEBGRajWaJ6nNbCuw/hgukQpsq6VwGopYK3OslRdU5lhxLGXu4e7tq9vRaBLEsTKzzEM9bt5YxVqZY628oDLHikiVWU1MIiJSLSUIERGplhLEvz0Z7QCiINbKHGvlBZU5VkSkzOqDEBGRaqkGISIi1VKCEBGRasVUgjCz8WaWbWarzez2avY3NbNXgvv/ZWY96z7K2hVGmW8ys2/M7Eszm29mPaIRZ22qqcwhx000MzezBj8kMpwym9mk4N/112b2t7qOsbaF8W+7u5ktMLPPg/++z41GnLXFzJ41s2/NbMUh9puZPRj8fXxpZkOP+UPdPSZ+gHhgDdALSASWA4OqHHM98HhweTLwSrTjroMynwk0Dy7/LBbKHDwuGVgELAYyoh13Hfw99wU+B9oE1ztEO+46KPOTwM+Cy4OAddGO+xjL/ANgKLDiEPvPBWYBBpwC/OtYPzOWahAjgNXunuvuJcDLwIVVjrkQeD64/Bowxhr2S6prLLO7L3D3wuDqYuDI5gOuf8L5ewb4E3AnUFSXwUVIOGW+FnjE3XcCuPu3dRxjbQunzA60Ci63BjbWYXy1zt0XATsOc8iFwAsesBhIMbPOx/KZsZQgugJ5Iev5wW3VHuPuZcBuoF2dRBcZ4ZQ51E8JfANpyGosc7Dq3c3d36nLwCIonL/nfkA/M/vYzBab2fg6iy4ywinz74GpZpYPvAv8om5Ci5oj/f9eI70PQgAws6lABjAq2rFEkpnFAfcCV0c5lLrWhEAz02gCtcRFZjbY3XdFNarImgL82d3vMbNTgb+Y2fHuXhHtwBqKWKpBFADdQtbTgtuqPcbMmhColm6vk+giI5wyY2ZnA/8FXODuxXUUW6TUVOZk4HhgoZmtI9BW+3YD76gO5+85H3jb3UvdfS2wikDCaKjCKfNPgVcB3P1TIInApHaNVVj/349ELCWIJUBfM0s3s0QCndBvVznmbeCq4PKlwPse7P1poGoss5mdBDxBIDk09HZpqKHM7r7b3VPdvae79yTQ73KBu2dGJ9xaEc6/7TcJ1B4ws1QCTU65dRlkLQunzBuAMQBmNpBAgmjM7xV9G7gyOJrpFGC3u286lgvGTBOTu5eZ2Q3AHAIjIJ5196/N7I9Apru/DTxDoBq6mkBn0OToRXzswizz3UBL4O/B/vgN7n5B1II+RmGWuVEJs8xzgHPM7BugHLjV3Rts7TjMMt8MPGVmNxLosL66IX/hM7OXCCT51GC/yv8FEgDc/XEC/SznAquBQuCaY/7MBvz7EhGRCIqlJiYRETkCShAiIlItJQgREamWEoSIiFRLCUJERKqlBCFyBMys3My+CPk55GyxR3HtnoeaqVMkGmLmOQiRWnLA3YdEOwiRuqAahEgtMLN1ZnaXmX1lZp+ZWZ/g9p5m9n7I+za6B7d3NLM3zGx58Oe04KXizeyp4Dsb5ppZs6gVSmKeEoTIkWlWpYnpspB9u919MPAwcH9w20PA8+5+AvBX4MHg9geBD9z9RAJz/H8d3N6XwLTcxwG7gIkRLo/IIelJapEjYGb73L1lNdvXAWe5e66ZJQCb3b2dmW0DOrt7aXD7JndPNbOtQFro5IgWeIPhPHfvG1y/DUhw9/+JfMlEvk81CJHa44dYPhKhs+mWo35CiSIlCJHac1nIn58Glz/h35M+/hj4MLg8n8ArXjGzeDNrXVdBioRL305EjkwzM/siZH22u1cOdW1jZl8SqAVMCW77BfCcmd1KYKrpyhk2fwU8aWY/JVBT+BlwTFMzi9Q29UGI1IJgH0SGu2+LdiwitUVNTCIiUi3VIEREpFqqQYiISLWUIEREpFpKECIiUi0lCBERqZYShIiIVOv/A5G18RrIPzDBAAAAAElFTkSuQmCC\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "print(\"Loss with Val_Loss Graph\")\n",
        "plt.plot(History.history['loss'])\n",
        "plt.plot(History.history['val_loss'])\n",
        "plt.title('Model Loss')\n",
        "plt.ylabel('Loss')\n",
        "plt.xlabel('Epochs')\n",
        "plt.legend(['train', 'test'], loc = 'upper right')\n",
        "plt.show()\n",
        "\n",
        "print(\"Acc with Val_Acc Graph\")\n",
        "plt.plot(History.history['accuracy'])\n",
        "plt.plot(History.history['val_accuracy'])\n",
        "plt.title('Model Accuracy')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['train', 'test'], loc='lower right')\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 44,
      "metadata": {
        "id": "skGdAG_McVvt"
      },
      "outputs": [],
      "source": [
        "import tensorflow as tf\n",
        "import pathlib"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ISj3k22LsE5J"
      },
      "source": [
        "## Save Model and convert to TFLite\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 45,
      "metadata": {
        "id": "SwvkvIVNb0rD",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "afa3cd3f-ab4c-425c-8874-8c9adcc11c21"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "5411836"
            ]
          },
          "metadata": {},
          "execution_count": 45
        }
      ],
      "source": [
        "export_dir = 'saved_model/'\n",
        "tf.saved_model.save(model, export_dir)\n",
        "\n",
        "converter = tf.lite.TFLiteConverter.from_saved_model(export_dir)\n",
        "tflite_model = converter.convert()\n",
        "\n",
        "tflite_model_file = pathlib.Path('RiceModel.tflite')\n",
        "tflite_model_file.write_bytes(tflite_model)"
      ]
    }
  ],
  "metadata": {
    "colab": {
      "collapsed_sections": [
        "0Et3PVTrHg1P",
        "qCMJ9GOjHlPc"
      ],
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}