VGG_KTH_TIPS_2B.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "VGG_KTH-TIPS-2B.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_J-Du6kR1hUR",
        "outputId": "5fd33d4d-f81e-4f4d-fae4-dc67789e67a3"
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Mounted at /content/drive\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "l1jsgnWBCS76"
      },
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "import cv2 as cv\n",
        "import os\n",
        "import tensorflow as tf\n",
        "from tensorflow import keras\n",
        "\n",
        "%matplotlib inline"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pJFyi_kwCmwM"
      },
      "source": [
        "DataDir = '/content/drive/MyDrive/PS 1/Datasets/KTH-TIPS-2B/KTH-TIPS2-b'\n",
        "Categories = ['aluminium_foil', 'brown_bread', 'corduroy', 'cork', 'cotton', 'cracker', 'lettuce_leaf', 'linen', 'white_bread', 'wood','wool']\n",
        "Samples = ['sample_a']\n",
        "images = []\n",
        "y = np.array([])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "WG7l4RjZCsGg",
        "outputId": "de4f3108-00d8-407b-a008-64ffe34b1721"
      },
      "source": [
        "for category in Categories:\n",
        "  for Sample in Samples:\n",
        "    path = os.path.join(DataDir,category)\n",
        "    path = os.path.join(path,Sample)\n",
        "    class_num = Categories.index(category)\n",
        "    print(path)\n",
        "    for imgname in os.listdir(path):\n",
        "        img = cv.imread(os.path.join(path,imgname))\n",
        "        img = cv.resize(img, (300, 300))\n",
        "        images.append(img)\n",
        "        y=np.append(y,[class_num])\n",
        "      \n",
        "        \n",
        "  \n",
        "x = np.array(images)\n",
        "print(x.shape)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/content/drive/MyDrive/PS 1/Datasets/KTH-TIPS-2B/KTH-TIPS2-b/aluminium_foil/sample_a\n",
            "/content/drive/MyDrive/PS 1/Datasets/KTH-TIPS-2B/KTH-TIPS2-b/brown_bread/sample_a\n",
            "/content/drive/MyDrive/PS 1/Datasets/KTH-TIPS-2B/KTH-TIPS2-b/corduroy/sample_a\n",
            "/content/drive/MyDrive/PS 1/Datasets/KTH-TIPS-2B/KTH-TIPS2-b/cork/sample_a\n",
            "/content/drive/MyDrive/PS 1/Datasets/KTH-TIPS-2B/KTH-TIPS2-b/cotton/sample_a\n",
            "/content/drive/MyDrive/PS 1/Datasets/KTH-TIPS-2B/KTH-TIPS2-b/cracker/sample_a\n",
            "/content/drive/MyDrive/PS 1/Datasets/KTH-TIPS-2B/KTH-TIPS2-b/lettuce_leaf/sample_a\n",
            "/content/drive/MyDrive/PS 1/Datasets/KTH-TIPS-2B/KTH-TIPS2-b/linen/sample_a\n",
            "/content/drive/MyDrive/PS 1/Datasets/KTH-TIPS-2B/KTH-TIPS2-b/white_bread/sample_a\n",
            "/content/drive/MyDrive/PS 1/Datasets/KTH-TIPS-2B/KTH-TIPS2-b/wood/sample_a\n",
            "/content/drive/MyDrive/PS 1/Datasets/KTH-TIPS-2B/KTH-TIPS2-b/wool/sample_a\n",
            "(1188, 300, 300, 3)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6otTVgxpCvDC",
        "outputId": "cc7be358-7945-4c5a-9086-bb1680691b24"
      },
      "source": [
        "y = np.reshape(y,(-1,1))\n",
        "print(y.shape)\n",
        "print(y)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(1188, 1)\n",
            "[[ 0.]\n",
            " [ 0.]\n",
            " [ 0.]\n",
            " ...\n",
            " [10.]\n",
            " [10.]\n",
            " [10.]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "voKiahHhDAOA"
      },
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "X_train, X_test, y_train, y_test = train_test_split(x, y, test_size = 0.2)\n",
        "X_train, X_val, y_train, y_val = train_test_split(X_train, y_train, test_size = 0.25)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8FvFdqutDDRi",
        "outputId": "b201ba28-c8d2-4459-a477-adf8c42c3e07"
      },
      "source": [
        "print(X_train.shape)\n",
        "print(X_test.shape)\n",
        "print(X_val.shape)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(712, 300, 300, 3)\n",
            "(238, 300, 300, 3)\n",
            "(238, 300, 300, 3)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SPSDn4FxDECo",
        "outputId": "bfe91b26-c62b-4006-8efe-b283021b1322"
      },
      "source": [
        "print(y_train.shape)\n",
        "print(y_test.shape)\n",
        "print(y_val.shape)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(712, 1)\n",
            "(238, 1)\n",
            "(238, 1)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NKyZ9u0pDdDO"
      },
      "source": [
        "train_ds = tf.data.Dataset.from_tensor_slices((X_train, y_train))\n",
        "test_ds = tf.data.Dataset.from_tensor_slices((X_test, y_test))\n",
        "val_ds = tf.data.Dataset.from_tensor_slices((X_val, y_val))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JuIs5CjoDwVW",
        "outputId": "585ea5a5-e9f9-4546-d59a-d5f02af0a897"
      },
      "source": [
        "print(train_ds)\n",
        "print(train_ds)\n",
        "print(val_ds)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<TensorSliceDataset shapes: ((300, 300, 3), (1,)), types: (tf.uint8, tf.float64)>\n",
            "<TensorSliceDataset shapes: ((300, 300, 3), (1,)), types: (tf.uint8, tf.float64)>\n",
            "<TensorSliceDataset shapes: ((300, 300, 3), (1,)), types: (tf.uint8, tf.float64)>\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3Kn4rXswDwmW",
        "outputId": "e964db10-add1-4ef4-e099-036af03c18ee"
      },
      "source": [
        "train_ds_size = tf.data.experimental.cardinality(train_ds).numpy()\n",
        "test_ds_size = tf.data.experimental.cardinality(test_ds).numpy()\n",
        "val_ds_size = tf.data.experimental.cardinality(val_ds).numpy()\n",
        "\n",
        "print('Train size:', train_ds_size)\n",
        "print('Test size:', test_ds_size)\n",
        "print('Val size:', val_ds_size)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train size: 712\n",
            "Test size: 238\n",
            "Val size: 238\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aKXtkUOVDGN_"
      },
      "source": [
        "from tensorflow import keras\n",
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import Dense, Activation, Dropout, Flatten,\\\n",
        " Conv2D, MaxPooling2D,BatchNormalization\n",
        "import keras,os\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Dense, Conv2D, MaxPool2D , Flatten\n",
        "from keras.preprocessing.image import ImageDataGenerator\n",
        "import numpy as np\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YOnN9XHwDStq"
      },
      "source": [
        "model = Sequential()\n",
        "model.add(Conv2D(input_shape=(224,224,3),filters=64,kernel_size=(3,3),padding=\"same\", activation=\"relu\"))\n",
        "model.add(Conv2D(filters=64,kernel_size=(3,3),padding=\"same\", activation=\"relu\"))\n",
        "model.add(MaxPool2D(pool_size=(2,2),strides=(2,2)))\n",
        "model.add(Conv2D(filters=128, kernel_size=(3,3), padding=\"same\", activation=\"relu\"))\n",
        "model.add(Conv2D(filters=128, kernel_size=(3,3), padding=\"same\", activation=\"relu\"))\n",
        "model.add(MaxPool2D(pool_size=(2,2),strides=(2,2)))\n",
        "model.add(Conv2D(filters=256, kernel_size=(3,3), padding=\"same\", activation=\"relu\"))\n",
        "model.add(Conv2D(filters=256, kernel_size=(3,3), padding=\"same\", activation=\"relu\"))\n",
        "model.add(Conv2D(filters=256, kernel_size=(3,3), padding=\"same\", activation=\"relu\"))\n",
        "model.add(MaxPool2D(pool_size=(2,2),strides=(2,2)))\n",
        "model.add(Conv2D(filters=512, kernel_size=(3,3), padding=\"same\", activation=\"relu\"))\n",
        "model.add(Conv2D(filters=512, kernel_size=(3,3), padding=\"same\", activation=\"relu\"))\n",
        "model.add(Conv2D(filters=512, kernel_size=(3,3), padding=\"same\", activation=\"relu\"))\n",
        "model.add(MaxPool2D(pool_size=(2,2),strides=(2,2)))\n",
        "model.add(Conv2D(filters=512, kernel_size=(3,3), padding=\"same\", activation=\"relu\"))\n",
        "model.add(Conv2D(filters=512, kernel_size=(3,3), padding=\"same\", activation=\"relu\"))\n",
        "model.add(Conv2D(filters=512, kernel_size=(3,3), padding=\"same\", activation=\"relu\"))\n",
        "model.add(MaxPool2D(pool_size=(2,2),strides=(2,2))\n",
        "model.add(Flatten())\n",
        "model.add(Dense(units=4096,activation=\"relu\"))\n",
        "model.add(Dense(units=4096,activation=\"relu\"))\n",
        "model.add(Dense(units=11, activation=\"softmax\"))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ttsvNtPMJzYc"
      },
      "source": [
        "model = keras.models.Sequential([\n",
        "    keras.layers.Conv2D(input_shape=(224,224,3),filters=64,kernel_size=(3,3),padding=\"same\", activation=\"relu\"),\n",
        "    keras.layers.Conv2D(filters=64,kernel_size=(3,3),padding=\"same\", activation=\"relu\"),\n",
        "    keras.layers.MaxPool2D(pool_size=(2,2),strides=(2,2)),\n",
        "    keras.layers.Conv2D(filters=128, kernel_size=(3,3), padding=\"same\", activation=\"relu\"),\n",
        "    keras.layers.Conv2D(filters=128, kernel_size=(3,3), padding=\"same\", activation=\"relu\"),\n",
        "    keras.layers.MaxPool2D(pool_size=(2,2),strides=(2,2)),\n",
        "    keras.layers.Conv2D(filters=256, kernel_size=(3,3), padding=\"same\", activation=\"relu\"),\n",
        "    keras.layers.Conv2D(filters=256, kernel_size=(3,3), padding=\"same\", activation=\"relu\"),\n",
        "    keras.layers.Conv2D(filters=256, kernel_size=(3,3), padding=\"same\", activation=\"relu\"),\n",
        "    keras.layers.MaxPool2D(pool_size=(2,2),strides=(2,2)),\n",
        "    keras.layers.Conv2D(filters=512, kernel_size=(3,3), padding=\"same\", activation=\"relu\"),\n",
        "    keras.layers.Conv2D(filters=512, kernel_size=(3,3), padding=\"same\", activation=\"relu\"),\n",
        "    keras.layers.Conv2D(filters=512, kernel_size=(3,3), padding=\"same\", activation=\"relu\"),\n",
        "    keras.layers.MaxPool2D(pool_size=(2,2),strides=(2,2)),\n",
        "    keras.layers.Conv2D(filters=512, kernel_size=(3,3), padding=\"same\", activation=\"relu\"),\n",
        "    keras.layers.Conv2D(filters=512, kernel_size=(3,3), padding=\"same\", activation=\"relu\"),\n",
        "    keras.layers.Conv2D(filters=512, kernel_size=(3,3), padding=\"same\", activation=\"relu\"),\n",
        "    keras.layers.MaxPool2D(pool_size=(2,2),strides=(2,2)),\n",
        "    keras.layers.Flatten(),\n",
        "    keras.layers.Dense(4096, activation='relu'),\n",
        "    keras.layers.Dropout(0.5),\n",
        "    keras.layers.Dense(4096, activation='relu'),\n",
        "    keras.layers.Dropout(0.5),\n",
        "    keras.layers.Dense(11, activation='softmax')\n",
        "])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "h56oF-v6D4cG",
        "outputId": "1a21719d-c046-4a8f-d286-30e49db9cff6"
      },
      "source": [
        "model.compile(\n",
        "    loss='sparse_categorical_crossentropy',\n",
        "    optimizer=tf.optimizers.SGD(lr=0.003),\n",
        "    metrics=['accuracy']    \n",
        ")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/optimizer_v2/optimizer_v2.py:375: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead.\n",
            "  \"The `lr` argument is deprecated, use `learning_rate` instead.\")\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FUo0r-akD_HU",
        "outputId": "2f880f58-3b01-44ac-a802-7aa9240f6dbf"
      },
      "source": [
        "model.summary()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d (Conv2D)              (None, 224, 224, 64)      1792      \n",
            "_________________________________________________________________\n",
            "conv2d_1 (Conv2D)            (None, 224, 224, 64)      36928     \n",
            "_________________________________________________________________\n",
            "max_pooling2d (MaxPooling2D) (None, 112, 112, 64)      0         \n",
            "_________________________________________________________________\n",
            "conv2d_2 (Conv2D)            (None, 112, 112, 128)     73856     \n",
            "_________________________________________________________________\n",
            "conv2d_3 (Conv2D)            (None, 112, 112, 128)     147584    \n",
            "_________________________________________________________________\n",
            "max_pooling2d_1 (MaxPooling2 (None, 56, 56, 128)       0         \n",
            "_________________________________________________________________\n",
            "conv2d_4 (Conv2D)            (None, 56, 56, 256)       295168    \n",
            "_________________________________________________________________\n",
            "conv2d_5 (Conv2D)            (None, 56, 56, 256)       590080    \n",
            "_________________________________________________________________\n",
            "conv2d_6 (Conv2D)            (None, 56, 56, 256)       590080    \n",
            "_________________________________________________________________\n",
            "max_pooling2d_2 (MaxPooling2 (None, 28, 28, 256)       0         \n",
            "_________________________________________________________________\n",
            "conv2d_7 (Conv2D)            (None, 28, 28, 512)       1180160   \n",
            "_________________________________________________________________\n",
            "conv2d_8 (Conv2D)            (None, 28, 28, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "conv2d_9 (Conv2D)            (None, 28, 28, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "max_pooling2d_3 (MaxPooling2 (None, 14, 14, 512)       0         \n",
            "_________________________________________________________________\n",
            "conv2d_10 (Conv2D)           (None, 14, 14, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "conv2d_11 (Conv2D)           (None, 14, 14, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "conv2d_12 (Conv2D)           (None, 14, 14, 512)       2359808   \n",
            "_________________________________________________________________\n",
            "max_pooling2d_4 (MaxPooling2 (None, 7, 7, 512)         0         \n",
            "_________________________________________________________________\n",
            "flatten (Flatten)            (None, 25088)             0         \n",
            "_________________________________________________________________\n",
            "dense (Dense)                (None, 4096)              102764544 \n",
            "_________________________________________________________________\n",
            "dropout (Dropout)            (None, 4096)              0         \n",
            "_________________________________________________________________\n",
            "dense_1 (Dense)              (None, 4096)              16781312  \n",
            "_________________________________________________________________\n",
            "dropout_1 (Dropout)          (None, 4096)              0         \n",
            "_________________________________________________________________\n",
            "dense_2 (Dense)              (None, 11)                45067     \n",
            "=================================================================\n",
            "Total params: 134,305,611\n",
            "Trainable params: 134,305,611\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5cGDCk_DEB6o"
      },
      "source": [
        "def process_image(image,label):\n",
        "    image=tf.image.per_image_standardization(image)\n",
        "    image=tf.image.resize(image,(224,224))\n",
        "    \n",
        "    return image,label"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_R8s3nODEHq2"
      },
      "source": [
        "train_ds = (train_ds\n",
        "          .map(process_image)\n",
        "          .shuffle(buffer_size=train_ds_size)\n",
        "          .batch(batch_size=30,drop_remainder=True)\n",
        "         )\n",
        "\n",
        "test_ds = (test_ds\n",
        "           .map(process_image)\n",
        "          .shuffle(buffer_size=test_ds_size)\n",
        "          .batch(batch_size=30,drop_remainder=True)\n",
        "         )\n",
        "\n",
        "val_ds = (val_ds\n",
        "          .map(process_image)\n",
        "          .shuffle(buffer_size=val_ds_size)\n",
        "          .batch(batch_size=30,drop_remainder=True)\n",
        "         )"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "oqFoG7dUEQhc",
        "outputId": "ed6cb6e7-6fff-49e8-8587-7364967f7479"
      },
      "source": [
        "print(train_ds)\n",
        "print(test_ds)\n",
        "print(val_ds)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<BatchDataset shapes: ((30, 224, 224, 3), (30, 1)), types: (tf.float32, tf.float64)>\n",
            "<BatchDataset shapes: ((30, 224, 224, 3), (30, 1)), types: (tf.float32, tf.float64)>\n",
            "<BatchDataset shapes: ((30, 224, 224, 3), (30, 1)), types: (tf.float32, tf.float64)>\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PdGzxx4pESrL",
        "outputId": "c8ea0339-e884-42d2-91f1-8fab9f046248"
      },
      "source": [
        "history = model.fit(train_ds,\n",
        "          epochs=200,\n",
        "          validation_data=val_ds,\n",
        "          validation_freq=1)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/200\n",
            "23/23 [==============================] - 80s 916ms/step - loss: 2.3981 - accuracy: 0.0906 - val_loss: 2.3974 - val_accuracy: 0.1714\n",
            "Epoch 2/200\n",
            "23/23 [==============================] - 21s 878ms/step - loss: 2.3978 - accuracy: 0.1087 - val_loss: 2.3972 - val_accuracy: 0.2286\n",
            "Epoch 3/200\n",
            "23/23 [==============================] - 21s 880ms/step - loss: 2.3974 - accuracy: 0.1255 - val_loss: 2.3968 - val_accuracy: 0.2571\n",
            "Epoch 4/200\n",
            "23/23 [==============================] - 21s 885ms/step - loss: 2.3971 - accuracy: 0.1374 - val_loss: 2.3965 - val_accuracy: 0.2143\n",
            "Epoch 5/200\n",
            "23/23 [==============================] - 21s 875ms/step - loss: 2.3962 - accuracy: 0.1370 - val_loss: 2.3964 - val_accuracy: 0.1714\n",
            "Epoch 6/200\n",
            "23/23 [==============================] - 21s 880ms/step - loss: 2.3962 - accuracy: 0.1181 - val_loss: 2.3959 - val_accuracy: 0.1619\n",
            "Epoch 7/200\n",
            "23/23 [==============================] - 21s 883ms/step - loss: 2.3954 - accuracy: 0.1447 - val_loss: 2.3958 - val_accuracy: 0.1286\n",
            "Epoch 8/200\n",
            "23/23 [==============================] - 21s 884ms/step - loss: 2.3950 - accuracy: 0.1491 - val_loss: 2.3953 - val_accuracy: 0.1238\n",
            "Epoch 9/200\n",
            "23/23 [==============================] - 21s 877ms/step - loss: 2.3951 - accuracy: 0.1349 - val_loss: 2.3949 - val_accuracy: 0.1190\n",
            "Epoch 10/200\n",
            "23/23 [==============================] - 21s 877ms/step - loss: 2.3935 - accuracy: 0.2271 - val_loss: 2.3946 - val_accuracy: 0.1048\n",
            "Epoch 11/200\n",
            "23/23 [==============================] - 21s 880ms/step - loss: 2.3942 - accuracy: 0.1488 - val_loss: 2.3941 - val_accuracy: 0.1238\n",
            "Epoch 12/200\n",
            "23/23 [==============================] - 21s 877ms/step - loss: 2.3939 - accuracy: 0.1515 - val_loss: 2.3935 - val_accuracy: 0.1429\n",
            "Epoch 13/200\n",
            "23/23 [==============================] - 21s 876ms/step - loss: 2.3924 - accuracy: 0.1717 - val_loss: 2.3932 - val_accuracy: 0.1905\n",
            "Epoch 14/200\n",
            "23/23 [==============================] - 21s 881ms/step - loss: 2.3931 - accuracy: 0.1916 - val_loss: 2.3923 - val_accuracy: 0.2762\n",
            "Epoch 15/200\n",
            "23/23 [==============================] - 21s 883ms/step - loss: 2.3908 - accuracy: 0.2155 - val_loss: 2.3922 - val_accuracy: 0.2810\n",
            "Epoch 16/200\n",
            "23/23 [==============================] - 21s 885ms/step - loss: 2.3898 - accuracy: 0.2272 - val_loss: 2.3910 - val_accuracy: 0.2952\n",
            "Epoch 17/200\n",
            "23/23 [==============================] - 21s 879ms/step - loss: 2.3897 - accuracy: 0.2432 - val_loss: 2.3907 - val_accuracy: 0.2619\n",
            "Epoch 18/200\n",
            "23/23 [==============================] - 21s 876ms/step - loss: 2.3894 - accuracy: 0.2082 - val_loss: 2.3884 - val_accuracy: 0.3000\n",
            "Epoch 19/200\n",
            "23/23 [==============================] - 21s 879ms/step - loss: 2.3885 - accuracy: 0.2598 - val_loss: 2.3880 - val_accuracy: 0.3000\n",
            "Epoch 20/200\n",
            "23/23 [==============================] - 21s 884ms/step - loss: 2.3868 - accuracy: 0.2830 - val_loss: 2.3878 - val_accuracy: 0.3000\n",
            "Epoch 21/200\n",
            "23/23 [==============================] - 21s 878ms/step - loss: 2.3864 - accuracy: 0.2384 - val_loss: 2.3862 - val_accuracy: 0.3143\n",
            "Epoch 22/200\n",
            "23/23 [==============================] - 21s 876ms/step - loss: 2.3855 - accuracy: 0.2636 - val_loss: 2.3848 - val_accuracy: 0.3000\n",
            "Epoch 23/200\n",
            "23/23 [==============================] - 21s 876ms/step - loss: 2.3833 - accuracy: 0.2632 - val_loss: 2.3830 - val_accuracy: 0.3286\n",
            "Epoch 24/200\n",
            "23/23 [==============================] - 21s 875ms/step - loss: 2.3817 - accuracy: 0.2518 - val_loss: 2.3813 - val_accuracy: 0.3429\n",
            "Epoch 25/200\n",
            "23/23 [==============================] - 21s 879ms/step - loss: 2.3799 - accuracy: 0.2841 - val_loss: 2.3798 - val_accuracy: 0.3286\n",
            "Epoch 26/200\n",
            "23/23 [==============================] - 21s 877ms/step - loss: 2.3789 - accuracy: 0.2699 - val_loss: 2.3766 - val_accuracy: 0.3333\n",
            "Epoch 27/200\n",
            "23/23 [==============================] - 21s 881ms/step - loss: 2.3775 - accuracy: 0.2798 - val_loss: 2.3728 - val_accuracy: 0.3095\n",
            "Epoch 28/200\n",
            "23/23 [==============================] - 21s 877ms/step - loss: 2.3729 - accuracy: 0.2767 - val_loss: 2.3693 - val_accuracy: 0.3333\n",
            "Epoch 29/200\n",
            "23/23 [==============================] - 21s 881ms/step - loss: 2.3673 - accuracy: 0.3361 - val_loss: 2.3636 - val_accuracy: 0.3333\n",
            "Epoch 30/200\n",
            "23/23 [==============================] - 21s 883ms/step - loss: 2.3603 - accuracy: 0.2946 - val_loss: 2.3598 - val_accuracy: 0.2857\n",
            "Epoch 31/200\n",
            "23/23 [==============================] - 21s 885ms/step - loss: 2.3568 - accuracy: 0.3161 - val_loss: 2.3506 - val_accuracy: 0.2381\n",
            "Epoch 32/200\n",
            "23/23 [==============================] - 21s 886ms/step - loss: 2.3478 - accuracy: 0.2646 - val_loss: 2.3347 - val_accuracy: 0.2048\n",
            "Epoch 33/200\n",
            "23/23 [==============================] - 21s 877ms/step - loss: 2.3230 - accuracy: 0.3042 - val_loss: 2.3127 - val_accuracy: 0.1762\n",
            "Epoch 34/200\n",
            "23/23 [==============================] - 21s 881ms/step - loss: 2.2993 - accuracy: 0.2440 - val_loss: 2.2536 - val_accuracy: 0.1619\n",
            "Epoch 35/200\n",
            "23/23 [==============================] - 21s 881ms/step - loss: 2.2333 - accuracy: 0.2065 - val_loss: 2.1485 - val_accuracy: 0.1476\n",
            "Epoch 36/200\n",
            "23/23 [==============================] - 21s 878ms/step - loss: 2.0927 - accuracy: 0.2305 - val_loss: 2.0203 - val_accuracy: 0.2143\n",
            "Epoch 37/200\n",
            "23/23 [==============================] - 21s 879ms/step - loss: 1.9802 - accuracy: 0.2675 - val_loss: 2.0020 - val_accuracy: 0.1238\n",
            "Epoch 38/200\n",
            "23/23 [==============================] - 21s 882ms/step - loss: 1.9402 - accuracy: 0.2597 - val_loss: 1.9070 - val_accuracy: 0.1667\n",
            "Epoch 39/200\n",
            "23/23 [==============================] - 21s 884ms/step - loss: 1.9071 - accuracy: 0.2380 - val_loss: 1.9316 - val_accuracy: 0.2762\n",
            "Epoch 40/200\n",
            "23/23 [==============================] - 21s 886ms/step - loss: 1.8465 - accuracy: 0.2951 - val_loss: 1.8085 - val_accuracy: 0.3000\n",
            "Epoch 41/200\n",
            "23/23 [==============================] - 21s 876ms/step - loss: 1.7853 - accuracy: 0.3564 - val_loss: 1.7327 - val_accuracy: 0.4143\n",
            "Epoch 42/200\n",
            "23/23 [==============================] - 22s 909ms/step - loss: 1.7586 - accuracy: 0.2674 - val_loss: 1.8733 - val_accuracy: 0.2905\n",
            "Epoch 43/200\n",
            "23/23 [==============================] - 21s 882ms/step - loss: 1.8020 - accuracy: 0.3187 - val_loss: 1.7646 - val_accuracy: 0.4381\n",
            "Epoch 44/200\n",
            "23/23 [==============================] - 21s 877ms/step - loss: 1.6927 - accuracy: 0.3661 - val_loss: 1.5738 - val_accuracy: 0.3571\n",
            "Epoch 45/200\n",
            "23/23 [==============================] - 21s 874ms/step - loss: 1.7894 - accuracy: 0.3212 - val_loss: 2.4817 - val_accuracy: 0.1286\n",
            "Epoch 46/200\n",
            "23/23 [==============================] - 21s 879ms/step - loss: 1.8728 - accuracy: 0.3156 - val_loss: 1.5734 - val_accuracy: 0.3714\n",
            "Epoch 47/200\n",
            "23/23 [==============================] - 21s 882ms/step - loss: 1.6648 - accuracy: 0.3855 - val_loss: 1.4762 - val_accuracy: 0.4190\n",
            "Epoch 48/200\n",
            "23/23 [==============================] - 21s 887ms/step - loss: 1.7300 - accuracy: 0.3761 - val_loss: 1.4817 - val_accuracy: 0.4714\n",
            "Epoch 49/200\n",
            "23/23 [==============================] - 21s 877ms/step - loss: 1.6116 - accuracy: 0.3758 - val_loss: 1.3956 - val_accuracy: 0.5381\n",
            "Epoch 50/200\n",
            "23/23 [==============================] - 21s 879ms/step - loss: 1.4828 - accuracy: 0.4249 - val_loss: 1.5516 - val_accuracy: 0.4190\n",
            "Epoch 51/200\n",
            "23/23 [==============================] - 21s 876ms/step - loss: 1.5807 - accuracy: 0.3756 - val_loss: 1.4858 - val_accuracy: 0.4429\n",
            "Epoch 52/200\n",
            "23/23 [==============================] - 21s 880ms/step - loss: 1.4270 - accuracy: 0.4656 - val_loss: 1.2916 - val_accuracy: 0.5429\n",
            "Epoch 53/200\n",
            "23/23 [==============================] - 21s 877ms/step - loss: 1.3731 - accuracy: 0.5036 - val_loss: 1.1809 - val_accuracy: 0.5571\n",
            "Epoch 54/200\n",
            "23/23 [==============================] - 21s 875ms/step - loss: 1.4714 - accuracy: 0.4446 - val_loss: 1.6138 - val_accuracy: 0.3238\n",
            "Epoch 55/200\n",
            "23/23 [==============================] - 22s 907ms/step - loss: 1.3356 - accuracy: 0.5256 - val_loss: 1.0315 - val_accuracy: 0.6714\n",
            "Epoch 56/200\n",
            "23/23 [==============================] - 21s 881ms/step - loss: 1.1603 - accuracy: 0.5359 - val_loss: 0.9587 - val_accuracy: 0.6286\n",
            "Epoch 57/200\n",
            "23/23 [==============================] - 21s 884ms/step - loss: 1.0038 - accuracy: 0.6110 - val_loss: 0.9656 - val_accuracy: 0.6476\n",
            "Epoch 58/200\n",
            "23/23 [==============================] - 21s 879ms/step - loss: 1.1464 - accuracy: 0.5415 - val_loss: 0.8452 - val_accuracy: 0.7429\n",
            "Epoch 59/200\n",
            "23/23 [==============================] - 21s 876ms/step - loss: 0.9267 - accuracy: 0.6558 - val_loss: 0.9662 - val_accuracy: 0.6524\n",
            "Epoch 60/200\n",
            "23/23 [==============================] - 21s 881ms/step - loss: 1.0993 - accuracy: 0.5540 - val_loss: 0.9717 - val_accuracy: 0.6524\n",
            "Epoch 61/200\n",
            "23/23 [==============================] - 21s 879ms/step - loss: 0.9587 - accuracy: 0.6371 - val_loss: 0.7711 - val_accuracy: 0.7762\n",
            "Epoch 62/200\n",
            "23/23 [==============================] - 21s 876ms/step - loss: 0.8704 - accuracy: 0.6732 - val_loss: 0.7128 - val_accuracy: 0.7571\n",
            "Epoch 63/200\n",
            "23/23 [==============================] - 21s 880ms/step - loss: 0.8640 - accuracy: 0.6980 - val_loss: 1.7847 - val_accuracy: 0.3857\n",
            "Epoch 64/200\n",
            "23/23 [==============================] - 21s 883ms/step - loss: 1.0513 - accuracy: 0.6396 - val_loss: 0.9568 - val_accuracy: 0.6571\n",
            "Epoch 65/200\n",
            "23/23 [==============================] - 21s 886ms/step - loss: 0.8000 - accuracy: 0.7135 - val_loss: 1.0982 - val_accuracy: 0.5810\n",
            "Epoch 66/200\n",
            "23/23 [==============================] - 21s 886ms/step - loss: 0.7727 - accuracy: 0.7420 - val_loss: 0.7288 - val_accuracy: 0.7952\n",
            "Epoch 67/200\n",
            "23/23 [==============================] - 21s 885ms/step - loss: 0.9645 - accuracy: 0.6715 - val_loss: 0.5474 - val_accuracy: 0.8381\n",
            "Epoch 68/200\n",
            "23/23 [==============================] - 21s 879ms/step - loss: 0.6778 - accuracy: 0.7533 - val_loss: 0.6983 - val_accuracy: 0.8190\n",
            "Epoch 69/200\n",
            "23/23 [==============================] - 21s 881ms/step - loss: 0.6884 - accuracy: 0.7398 - val_loss: 0.7383 - val_accuracy: 0.7524\n",
            "Epoch 70/200\n",
            "23/23 [==============================] - 22s 908ms/step - loss: 0.6496 - accuracy: 0.7754 - val_loss: 0.6225 - val_accuracy: 0.7667\n",
            "Epoch 71/200\n",
            "23/23 [==============================] - 21s 878ms/step - loss: 0.6152 - accuracy: 0.7920 - val_loss: 0.4611 - val_accuracy: 0.8619\n",
            "Epoch 72/200\n",
            "23/23 [==============================] - 21s 881ms/step - loss: 0.6186 - accuracy: 0.7966 - val_loss: 0.4983 - val_accuracy: 0.8286\n",
            "Epoch 73/200\n",
            "23/23 [==============================] - 21s 879ms/step - loss: 0.4818 - accuracy: 0.8223 - val_loss: 0.6278 - val_accuracy: 0.7762\n",
            "Epoch 74/200\n",
            "23/23 [==============================] - 21s 882ms/step - loss: 0.6007 - accuracy: 0.7621 - val_loss: 0.5417 - val_accuracy: 0.8095\n",
            "Epoch 75/200\n",
            "23/23 [==============================] - 21s 877ms/step - loss: 0.4313 - accuracy: 0.8515 - val_loss: 0.6177 - val_accuracy: 0.7810\n",
            "Epoch 76/200\n",
            "23/23 [==============================] - 21s 880ms/step - loss: 0.7171 - accuracy: 0.7723 - val_loss: 0.5031 - val_accuracy: 0.8333\n",
            "Epoch 77/200\n",
            "23/23 [==============================] - 21s 877ms/step - loss: 0.6146 - accuracy: 0.7776 - val_loss: 0.6618 - val_accuracy: 0.7762\n",
            "Epoch 78/200\n",
            "23/23 [==============================] - 21s 878ms/step - loss: 0.6691 - accuracy: 0.7819 - val_loss: 0.4242 - val_accuracy: 0.8667\n",
            "Epoch 79/200\n",
            "23/23 [==============================] - 21s 874ms/step - loss: 0.7083 - accuracy: 0.7877 - val_loss: 0.4989 - val_accuracy: 0.8333\n",
            "Epoch 80/200\n",
            "23/23 [==============================] - 21s 880ms/step - loss: 0.4842 - accuracy: 0.8244 - val_loss: 0.3987 - val_accuracy: 0.8667\n",
            "Epoch 81/200\n",
            "23/23 [==============================] - 21s 882ms/step - loss: 0.4388 - accuracy: 0.8394 - val_loss: 0.4354 - val_accuracy: 0.8810\n",
            "Epoch 82/200\n",
            "23/23 [==============================] - 21s 880ms/step - loss: 0.4031 - accuracy: 0.8621 - val_loss: 0.4963 - val_accuracy: 0.8571\n",
            "Epoch 83/200\n",
            "23/23 [==============================] - 21s 877ms/step - loss: 0.4506 - accuracy: 0.8671 - val_loss: 0.4560 - val_accuracy: 0.8524\n",
            "Epoch 84/200\n",
            "23/23 [==============================] - 21s 881ms/step - loss: 0.4459 - accuracy: 0.8478 - val_loss: 0.3531 - val_accuracy: 0.9000\n",
            "Epoch 85/200\n",
            "23/23 [==============================] - 21s 876ms/step - loss: 0.9891 - accuracy: 0.6899 - val_loss: 0.4507 - val_accuracy: 0.8667\n",
            "Epoch 86/200\n",
            "23/23 [==============================] - 21s 880ms/step - loss: 0.3743 - accuracy: 0.8675 - val_loss: 0.3754 - val_accuracy: 0.8667\n",
            "Epoch 87/200\n",
            "23/23 [==============================] - 21s 883ms/step - loss: 0.3006 - accuracy: 0.8941 - val_loss: 0.3100 - val_accuracy: 0.9048\n",
            "Epoch 88/200\n",
            "23/23 [==============================] - 21s 885ms/step - loss: 0.4323 - accuracy: 0.8773 - val_loss: 0.3280 - val_accuracy: 0.8857\n",
            "Epoch 89/200\n",
            "23/23 [==============================] - 21s 878ms/step - loss: 0.3001 - accuracy: 0.8973 - val_loss: 0.3405 - val_accuracy: 0.9000\n",
            "Epoch 90/200\n",
            "23/23 [==============================] - 21s 876ms/step - loss: 0.2865 - accuracy: 0.8889 - val_loss: 3.1220 - val_accuracy: 0.4476\n",
            "Epoch 91/200\n",
            "23/23 [==============================] - 21s 881ms/step - loss: 1.9592 - accuracy: 0.5023 - val_loss: 0.6996 - val_accuracy: 0.7714\n",
            "Epoch 92/200\n",
            "23/23 [==============================] - 21s 884ms/step - loss: 0.5510 - accuracy: 0.8096 - val_loss: 0.6854 - val_accuracy: 0.7857\n",
            "Epoch 93/200\n",
            "23/23 [==============================] - 21s 885ms/step - loss: 0.4464 - accuracy: 0.8465 - val_loss: 0.4104 - val_accuracy: 0.8667\n",
            "Epoch 94/200\n",
            "23/23 [==============================] - 21s 885ms/step - loss: 0.3522 - accuracy: 0.8778 - val_loss: 0.3636 - val_accuracy: 0.9000\n",
            "Epoch 95/200\n",
            "23/23 [==============================] - 21s 880ms/step - loss: 0.2001 - accuracy: 0.9434 - val_loss: 0.3768 - val_accuracy: 0.8905\n",
            "Epoch 96/200\n",
            "23/23 [==============================] - 21s 879ms/step - loss: 0.3142 - accuracy: 0.8937 - val_loss: 0.3308 - val_accuracy: 0.8857\n",
            "Epoch 97/200\n",
            "23/23 [==============================] - 21s 877ms/step - loss: 0.2326 - accuracy: 0.9131 - val_loss: 0.3570 - val_accuracy: 0.8905\n",
            "Epoch 98/200\n",
            "23/23 [==============================] - 21s 879ms/step - loss: 0.1972 - accuracy: 0.9290 - val_loss: 0.2935 - val_accuracy: 0.9095\n",
            "Epoch 99/200\n",
            "23/23 [==============================] - 21s 882ms/step - loss: 0.1175 - accuracy: 0.9701 - val_loss: 1.5081 - val_accuracy: 0.5762\n",
            "Epoch 100/200\n",
            "23/23 [==============================] - 21s 881ms/step - loss: 0.5549 - accuracy: 0.8417 - val_loss: 0.3650 - val_accuracy: 0.9000\n",
            "Epoch 101/200\n",
            "23/23 [==============================] - 21s 881ms/step - loss: 0.1919 - accuracy: 0.9506 - val_loss: 0.2253 - val_accuracy: 0.9095\n",
            "Epoch 102/200\n",
            "23/23 [==============================] - 21s 883ms/step - loss: 0.1971 - accuracy: 0.9239 - val_loss: 0.2824 - val_accuracy: 0.9143\n",
            "Epoch 103/200\n",
            "23/23 [==============================] - 21s 882ms/step - loss: 0.1812 - accuracy: 0.9370 - val_loss: 0.3692 - val_accuracy: 0.9000\n",
            "Epoch 104/200\n",
            "23/23 [==============================] - 21s 886ms/step - loss: 0.1849 - accuracy: 0.9303 - val_loss: 0.2571 - val_accuracy: 0.9238\n",
            "Epoch 105/200\n",
            "23/23 [==============================] - 21s 887ms/step - loss: 0.1603 - accuracy: 0.9483 - val_loss: 0.3579 - val_accuracy: 0.9238\n",
            "Epoch 106/200\n",
            "23/23 [==============================] - 21s 886ms/step - loss: 0.1560 - accuracy: 0.9542 - val_loss: 0.3133 - val_accuracy: 0.9190\n",
            "Epoch 107/200\n",
            "23/23 [==============================] - 21s 885ms/step - loss: 0.2474 - accuracy: 0.9346 - val_loss: 1.1538 - val_accuracy: 0.6333\n",
            "Epoch 108/200\n",
            "23/23 [==============================] - 21s 885ms/step - loss: 0.8445 - accuracy: 0.7326 - val_loss: 0.5387 - val_accuracy: 0.8714\n",
            "Epoch 109/200\n",
            "23/23 [==============================] - 21s 886ms/step - loss: 0.2338 - accuracy: 0.9347 - val_loss: 0.4191 - val_accuracy: 0.8429\n",
            "Epoch 110/200\n",
            "23/23 [==============================] - 21s 880ms/step - loss: 0.1836 - accuracy: 0.9309 - val_loss: 0.5609 - val_accuracy: 0.8286\n",
            "Epoch 111/200\n",
            "23/23 [==============================] - 21s 877ms/step - loss: 0.2295 - accuracy: 0.9560 - val_loss: 0.2757 - val_accuracy: 0.9143\n",
            "Epoch 112/200\n",
            "23/23 [==============================] - 21s 880ms/step - loss: 0.1086 - accuracy: 0.9692 - val_loss: 0.3079 - val_accuracy: 0.9095\n",
            "Epoch 113/200\n",
            "23/23 [==============================] - 22s 906ms/step - loss: 0.1139 - accuracy: 0.9595 - val_loss: 0.2541 - val_accuracy: 0.9238\n",
            "Epoch 114/200\n",
            "23/23 [==============================] - 21s 883ms/step - loss: 0.0843 - accuracy: 0.9711 - val_loss: 0.5771 - val_accuracy: 0.8429\n",
            "Epoch 115/200\n",
            "23/23 [==============================] - 21s 878ms/step - loss: 0.2140 - accuracy: 0.9248 - val_loss: 0.3008 - val_accuracy: 0.9238\n",
            "Epoch 116/200\n",
            "23/23 [==============================] - 21s 877ms/step - loss: 0.1162 - accuracy: 0.9700 - val_loss: 0.2532 - val_accuracy: 0.9333\n",
            "Epoch 117/200\n",
            "23/23 [==============================] - 21s 881ms/step - loss: 0.1275 - accuracy: 0.9671 - val_loss: 0.3095 - val_accuracy: 0.9333\n",
            "Epoch 118/200\n",
            "23/23 [==============================] - 21s 885ms/step - loss: 0.0753 - accuracy: 0.9666 - val_loss: 0.2751 - val_accuracy: 0.9190\n",
            "Epoch 119/200\n",
            "23/23 [==============================] - 21s 885ms/step - loss: 0.0780 - accuracy: 0.9726 - val_loss: 3.9294 - val_accuracy: 0.5095\n",
            "Epoch 120/200\n",
            "23/23 [==============================] - 21s 885ms/step - loss: 1.7042 - accuracy: 0.5919 - val_loss: 0.2643 - val_accuracy: 0.9381\n",
            "Epoch 121/200\n",
            "23/23 [==============================] - 21s 878ms/step - loss: 0.1192 - accuracy: 0.9824 - val_loss: 0.3449 - val_accuracy: 0.9190\n",
            "Epoch 122/200\n",
            "23/23 [==============================] - 21s 880ms/step - loss: 0.0636 - accuracy: 0.9810 - val_loss: 0.4600 - val_accuracy: 0.9000\n",
            "Epoch 123/200\n",
            "23/23 [==============================] - 21s 875ms/step - loss: 0.1147 - accuracy: 0.9711 - val_loss: 0.2698 - val_accuracy: 0.9333\n",
            "Epoch 124/200\n",
            "23/23 [==============================] - 21s 877ms/step - loss: 0.1201 - accuracy: 0.9606 - val_loss: 0.2906 - val_accuracy: 0.9238\n",
            "Epoch 125/200\n",
            "23/23 [==============================] - 21s 875ms/step - loss: 0.0590 - accuracy: 0.9757 - val_loss: 0.3464 - val_accuracy: 0.9238\n",
            "Epoch 126/200\n",
            "23/23 [==============================] - 21s 874ms/step - loss: 0.0905 - accuracy: 0.9755 - val_loss: 0.3117 - val_accuracy: 0.9286\n",
            "Epoch 127/200\n",
            "23/23 [==============================] - 21s 881ms/step - loss: 0.0622 - accuracy: 0.9802 - val_loss: 0.3923 - val_accuracy: 0.9333\n",
            "Epoch 128/200\n",
            "23/23 [==============================] - 21s 883ms/step - loss: 0.0480 - accuracy: 0.9799 - val_loss: 0.3524 - val_accuracy: 0.9381\n",
            "Epoch 129/200\n",
            "23/23 [==============================] - 21s 879ms/step - loss: 0.0225 - accuracy: 0.9940 - val_loss: 0.3320 - val_accuracy: 0.9429\n",
            "Epoch 130/200\n",
            "23/23 [==============================] - 21s 881ms/step - loss: 0.0185 - accuracy: 0.9977 - val_loss: 0.3236 - val_accuracy: 0.9381\n",
            "Epoch 131/200\n",
            "23/23 [==============================] - 21s 886ms/step - loss: 0.0303 - accuracy: 0.9898 - val_loss: 0.3788 - val_accuracy: 0.9190\n",
            "Epoch 132/200\n",
            "23/23 [==============================] - 21s 886ms/step - loss: 0.0688 - accuracy: 0.9775 - val_loss: 7.8207 - val_accuracy: 0.2810\n",
            "Epoch 133/200\n",
            "23/23 [==============================] - 21s 885ms/step - loss: 2.6420 - accuracy: 0.4175 - val_loss: 0.6969 - val_accuracy: 0.7333\n",
            "Epoch 134/200\n",
            "23/23 [==============================] - 21s 880ms/step - loss: 0.5203 - accuracy: 0.7917 - val_loss: 0.5290 - val_accuracy: 0.8333\n",
            "Epoch 135/200\n",
            "23/23 [==============================] - 21s 881ms/step - loss: 0.2344 - accuracy: 0.9203 - val_loss: 0.4010 - val_accuracy: 0.9143\n",
            "Epoch 136/200\n",
            "23/23 [==============================] - 21s 885ms/step - loss: 0.0902 - accuracy: 0.9632 - val_loss: 0.2697 - val_accuracy: 0.9333\n",
            "Epoch 137/200\n",
            "23/23 [==============================] - 21s 880ms/step - loss: 0.0818 - accuracy: 0.9740 - val_loss: 0.4236 - val_accuracy: 0.9286\n",
            "Epoch 138/200\n",
            "23/23 [==============================] - 21s 881ms/step - loss: 0.4730 - accuracy: 0.9009 - val_loss: 0.2895 - val_accuracy: 0.9190\n",
            "Epoch 139/200\n",
            "23/23 [==============================] - 21s 878ms/step - loss: 0.0640 - accuracy: 0.9926 - val_loss: 0.1918 - val_accuracy: 0.9286\n",
            "Epoch 140/200\n",
            "23/23 [==============================] - 21s 875ms/step - loss: 0.1109 - accuracy: 0.9624 - val_loss: 0.2440 - val_accuracy: 0.9429\n",
            "Epoch 141/200\n",
            "23/23 [==============================] - 21s 881ms/step - loss: 0.0975 - accuracy: 0.9735 - val_loss: 0.3980 - val_accuracy: 0.9143\n",
            "Epoch 142/200\n",
            "23/23 [==============================] - 21s 877ms/step - loss: 0.0675 - accuracy: 0.9732 - val_loss: 0.3062 - val_accuracy: 0.9286\n",
            "Epoch 143/200\n",
            "23/23 [==============================] - 21s 881ms/step - loss: 0.0360 - accuracy: 0.9925 - val_loss: 0.3285 - val_accuracy: 0.9333\n",
            "Epoch 144/200\n",
            "23/23 [==============================] - 21s 881ms/step - loss: 0.0322 - accuracy: 0.9918 - val_loss: 0.3578 - val_accuracy: 0.9476\n",
            "Epoch 145/200\n",
            "23/23 [==============================] - 21s 878ms/step - loss: 0.0531 - accuracy: 0.9830 - val_loss: 0.3323 - val_accuracy: 0.9238\n",
            "Epoch 146/200\n",
            "23/23 [==============================] - 21s 881ms/step - loss: 0.0228 - accuracy: 0.9955 - val_loss: 0.2378 - val_accuracy: 0.9381\n",
            "Epoch 147/200\n",
            "23/23 [==============================] - 21s 880ms/step - loss: 0.0146 - accuracy: 0.9949 - val_loss: 0.2909 - val_accuracy: 0.9333\n",
            "Epoch 148/200\n",
            "23/23 [==============================] - 21s 884ms/step - loss: 0.0131 - accuracy: 0.9989 - val_loss: 0.3839 - val_accuracy: 0.9333\n",
            "Epoch 149/200\n",
            "23/23 [==============================] - 21s 884ms/step - loss: 0.0100 - accuracy: 0.9997 - val_loss: 0.3381 - val_accuracy: 0.9476\n",
            "Epoch 150/200\n",
            "23/23 [==============================] - 21s 887ms/step - loss: 0.0785 - accuracy: 0.9732 - val_loss: 0.2745 - val_accuracy: 0.9286\n",
            "Epoch 151/200\n",
            "23/23 [==============================] - 21s 879ms/step - loss: 0.0385 - accuracy: 0.9987 - val_loss: 0.2813 - val_accuracy: 0.9333\n",
            "Epoch 152/200\n",
            "23/23 [==============================] - 21s 878ms/step - loss: 0.0182 - accuracy: 0.9982 - val_loss: 0.3063 - val_accuracy: 0.9429\n",
            "Epoch 153/200\n",
            "23/23 [==============================] - 21s 879ms/step - loss: 0.0174 - accuracy: 0.9941 - val_loss: 0.2986 - val_accuracy: 0.9381\n",
            "Epoch 154/200\n",
            "23/23 [==============================] - 21s 881ms/step - loss: 0.0166 - accuracy: 0.9961 - val_loss: 0.2813 - val_accuracy: 0.9476\n",
            "Epoch 155/200\n",
            "23/23 [==============================] - 21s 879ms/step - loss: 0.0229 - accuracy: 0.9917 - val_loss: 0.2742 - val_accuracy: 0.9333\n",
            "Epoch 156/200\n",
            "23/23 [==============================] - 21s 883ms/step - loss: 0.0062 - accuracy: 1.0000 - val_loss: 0.3292 - val_accuracy: 0.9333\n",
            "Epoch 157/200\n",
            "23/23 [==============================] - 21s 886ms/step - loss: 0.0170 - accuracy: 0.9958 - val_loss: 0.3087 - val_accuracy: 0.9333\n",
            "Epoch 158/200\n",
            "23/23 [==============================] - 21s 887ms/step - loss: 0.0050 - accuracy: 1.0000 - val_loss: 0.2498 - val_accuracy: 0.9524\n",
            "Epoch 159/200\n",
            "23/23 [==============================] - 21s 880ms/step - loss: 0.7617 - accuracy: 0.8067 - val_loss: 0.9398 - val_accuracy: 0.7286\n",
            "Epoch 160/200\n",
            "23/23 [==============================] - 21s 878ms/step - loss: 0.5220 - accuracy: 0.8211 - val_loss: 0.2654 - val_accuracy: 0.9333\n",
            "Epoch 161/200\n",
            "23/23 [==============================] - 21s 882ms/step - loss: 0.0697 - accuracy: 0.9903 - val_loss: 0.2902 - val_accuracy: 0.9381\n",
            "Epoch 162/200\n",
            "23/23 [==============================] - 21s 885ms/step - loss: 0.0302 - accuracy: 0.9929 - val_loss: 0.2903 - val_accuracy: 0.9333\n",
            "Epoch 163/200\n",
            "23/23 [==============================] - 21s 888ms/step - loss: 0.0138 - accuracy: 0.9982 - val_loss: 0.3142 - val_accuracy: 0.9429\n",
            "Epoch 164/200\n",
            "23/23 [==============================] - 21s 888ms/step - loss: 0.0161 - accuracy: 0.9951 - val_loss: 0.2969 - val_accuracy: 0.9381\n",
            "Epoch 165/200\n",
            "23/23 [==============================] - 21s 880ms/step - loss: 0.0248 - accuracy: 0.9911 - val_loss: 0.3703 - val_accuracy: 0.9381\n",
            "Epoch 166/200\n",
            "23/23 [==============================] - 21s 884ms/step - loss: 0.0071 - accuracy: 0.9997 - val_loss: 0.3501 - val_accuracy: 0.9333\n",
            "Epoch 167/200\n",
            "23/23 [==============================] - 21s 880ms/step - loss: 0.0061 - accuracy: 0.9991 - val_loss: 0.3743 - val_accuracy: 0.9429\n",
            "Epoch 168/200\n",
            "23/23 [==============================] - 21s 879ms/step - loss: 0.0126 - accuracy: 0.9971 - val_loss: 0.2969 - val_accuracy: 0.9238\n",
            "Epoch 169/200\n",
            "23/23 [==============================] - 21s 877ms/step - loss: 0.2356 - accuracy: 0.9310 - val_loss: 0.2760 - val_accuracy: 0.9286\n",
            "Epoch 170/200\n",
            "23/23 [==============================] - 21s 880ms/step - loss: 0.0164 - accuracy: 0.9986 - val_loss: 0.2815 - val_accuracy: 0.9524\n",
            "Epoch 171/200\n",
            "23/23 [==============================] - 21s 882ms/step - loss: 0.0164 - accuracy: 0.9954 - val_loss: 0.3126 - val_accuracy: 0.9381\n",
            "Epoch 172/200\n",
            "23/23 [==============================] - 21s 882ms/step - loss: 0.0054 - accuracy: 0.9999 - val_loss: 0.3265 - val_accuracy: 0.9095\n",
            "Epoch 173/200\n",
            "23/23 [==============================] - 21s 884ms/step - loss: 0.0130 - accuracy: 0.9993 - val_loss: 0.2736 - val_accuracy: 0.9476\n",
            "Epoch 174/200\n",
            "23/23 [==============================] - 21s 887ms/step - loss: 0.0083 - accuracy: 0.9993 - val_loss: 0.3582 - val_accuracy: 0.9333\n",
            "Epoch 175/200\n",
            "23/23 [==============================] - 21s 886ms/step - loss: 0.0056 - accuracy: 1.0000 - val_loss: 0.3453 - val_accuracy: 0.9429\n",
            "Epoch 176/200\n",
            "23/23 [==============================] - 21s 889ms/step - loss: 0.0972 - accuracy: 0.9777 - val_loss: 0.2867 - val_accuracy: 0.9524\n",
            "Epoch 177/200\n",
            "23/23 [==============================] - 21s 887ms/step - loss: 0.0081 - accuracy: 0.9997 - val_loss: 0.4627 - val_accuracy: 0.9286\n",
            "Epoch 178/200\n",
            "23/23 [==============================] - 21s 887ms/step - loss: 0.0257 - accuracy: 0.9902 - val_loss: 0.3720 - val_accuracy: 0.9381\n",
            "Epoch 179/200\n",
            "23/23 [==============================] - 21s 888ms/step - loss: 0.0161 - accuracy: 0.9933 - val_loss: 0.2537 - val_accuracy: 0.9429\n",
            "Epoch 180/200\n",
            "23/23 [==============================] - 21s 881ms/step - loss: 0.0029 - accuracy: 0.9998 - val_loss: 0.3223 - val_accuracy: 0.9429\n",
            "Epoch 181/200\n",
            "23/23 [==============================] - 21s 878ms/step - loss: 0.0061 - accuracy: 1.0000 - val_loss: 0.3182 - val_accuracy: 0.9429\n",
            "Epoch 182/200\n",
            "23/23 [==============================] - 22s 906ms/step - loss: 0.0053 - accuracy: 1.0000 - val_loss: 0.3873 - val_accuracy: 0.9286\n",
            "Epoch 183/200\n",
            "23/23 [==============================] - 21s 878ms/step - loss: 0.0148 - accuracy: 0.9918 - val_loss: 0.3454 - val_accuracy: 0.9286\n",
            "Epoch 184/200\n",
            "23/23 [==============================] - 21s 879ms/step - loss: 0.0052 - accuracy: 0.9991 - val_loss: 0.2756 - val_accuracy: 0.9571\n",
            "Epoch 185/200\n",
            "23/23 [==============================] - 21s 882ms/step - loss: 0.0078 - accuracy: 0.9985 - val_loss: 0.3762 - val_accuracy: 0.9429\n",
            "Epoch 186/200\n",
            "23/23 [==============================] - 21s 881ms/step - loss: 0.0061 - accuracy: 0.9973 - val_loss: 0.4586 - val_accuracy: 0.9238\n",
            "Epoch 187/200\n",
            "23/23 [==============================] - 21s 883ms/step - loss: 0.0224 - accuracy: 0.9882 - val_loss: 0.3333 - val_accuracy: 0.9429\n",
            "Epoch 188/200\n",
            "23/23 [==============================] - 21s 880ms/step - loss: 0.0249 - accuracy: 0.9929 - val_loss: 0.3807 - val_accuracy: 0.9381\n",
            "Epoch 189/200\n",
            "23/23 [==============================] - 21s 881ms/step - loss: 0.0040 - accuracy: 1.0000 - val_loss: 0.3436 - val_accuracy: 0.9333\n",
            "Epoch 190/200\n",
            "23/23 [==============================] - 21s 881ms/step - loss: 0.0024 - accuracy: 1.0000 - val_loss: 0.3894 - val_accuracy: 0.9333\n",
            "Epoch 191/200\n",
            "23/23 [==============================] - 21s 878ms/step - loss: 0.0060 - accuracy: 0.9973 - val_loss: 0.3970 - val_accuracy: 0.9333\n",
            "Epoch 192/200\n",
            "23/23 [==============================] - 21s 883ms/step - loss: 0.0038 - accuracy: 1.0000 - val_loss: 0.3367 - val_accuracy: 0.9429\n",
            "Epoch 193/200\n",
            "23/23 [==============================] - 21s 880ms/step - loss: 0.0040 - accuracy: 0.9993 - val_loss: 0.3746 - val_accuracy: 0.9476\n",
            "Epoch 194/200\n",
            "23/23 [==============================] - 21s 884ms/step - loss: 0.0200 - accuracy: 0.9955 - val_loss: 0.2729 - val_accuracy: 0.9381\n",
            "Epoch 195/200\n",
            "23/23 [==============================] - 21s 885ms/step - loss: 0.0562 - accuracy: 0.9838 - val_loss: 0.2574 - val_accuracy: 0.9524\n",
            "Epoch 196/200\n",
            "23/23 [==============================] - 21s 887ms/step - loss: 0.0047 - accuracy: 0.9997 - val_loss: 0.4380 - val_accuracy: 0.9286\n",
            "Epoch 197/200\n",
            "23/23 [==============================] - 21s 880ms/step - loss: 0.0076 - accuracy: 0.9977 - val_loss: 0.3602 - val_accuracy: 0.9476\n",
            "Epoch 198/200\n",
            "23/23 [==============================] - 21s 883ms/step - loss: 0.0034 - accuracy: 1.0000 - val_loss: 0.3448 - val_accuracy: 0.9429\n",
            "Epoch 199/200\n",
            "23/23 [==============================] - 21s 885ms/step - loss: 0.0027 - accuracy: 0.9991 - val_loss: 0.3639 - val_accuracy: 0.9524\n",
            "Epoch 200/200\n",
            "23/23 [==============================] - 21s 888ms/step - loss: 0.0015 - accuracy: 1.0000 - val_loss: 0.3663 - val_accuracy: 0.9571\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "id": "T4ihguEOEXfu",
        "outputId": "bdd0ea6e-16b8-453c-b77f-0bd66f3437be"
      },
      "source": [
        "plt.plot(history.history['accuracy'], label='Training accuracy')\n",
        "plt.plot(history.history['val_accuracy'], label = 'Testing accuracy')\n",
        "plt.title('Model Accuracy')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.legend(loc='lower right')\n",
        "plt.show()\n",
        "%matplotlib inline"
      ],
      "execution_count": null,
      "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": {
        "id": "m7Rjkx7eEb_Z"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}