G3_Train_Unfreeze_500.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "machine_shape": "hm",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU",
    "gpuClass": "premium"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/Wanita-8943/Main_Project/blob/main/G3_Train_Unfreeze_500.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "KKSs7cyoPHcD"
      },
      "outputs": [],
      "source": [
        "import pandas as pd\n",
        "import shutil"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Klm1B91QlvKd",
        "outputId": "20ae35fb-59ad-46e0-fb23-55ae11f3df47"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow.keras import models\n",
        "from tensorflow.keras import layers\n",
        "from tensorflow.keras import optimizers\n",
        "import os\n",
        "import glob\n",
        "import shutil\n",
        "import sys\n",
        "import numpy as np\n",
        "from skimage.io import imread\n",
        "import matplotlib.pyplot as plt\n",
        "from IPython.display import Image\n",
        "%matplotlib inline"
      ],
      "metadata": {
        "id": "1pX9g1HxPM2f"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "batch_size = 16\n",
        "width = 150\n",
        "height = 150\n",
        "epochs = 500\n",
        "NUM_TRAIN = 2850\n",
        "NUM_TEST = 950\n",
        "dropout_rate = 0.2\n",
        "input_shape = (height, width, 3)"
      ],
      "metadata": {
        "id": "eSFtvGyvPM6O"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#ดึงข้อมูลใน Github มาใช้\n",
        "import os\n",
        "%cd /content\n",
        "if not os.path.isdir(\"efficientnet_keras_transfer_learning\"):\n",
        " !git clone https://github.com/Wanita-8943/efficientnet_keras_transfer_learning\n",
        "%cd efficientnet_keras_transfer_learning/"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lb4K4CsMPNAW",
        "outputId": "4daaf065-f8c7-4d6e-f5f3-71fb813f60fd"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "/content\n",
            "Cloning into 'efficientnet_keras_transfer_learning'...\n",
            "remote: Enumerating objects: 837, done.\u001b[K\n",
            "remote: Total 837 (delta 0), reused 0 (delta 0), pack-reused 837\u001b[K\n",
            "Receiving objects: 100% (837/837), 13.85 MiB | 19.43 MiB/s, done.\n",
            "Resolving deltas: 100% (497/497), done.\n",
            "/content/efficientnet_keras_transfer_learning\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Options: EfficientNetB0, EfficientNetB1, EfficientNetB2, EfficientNetB3\n",
        "# Higher the number, the more complex the model is.\n",
        "from efficientnet import EfficientNetB0 as Net\n",
        "from efficientnet import center_crop_and_resize, preprocess_input"
      ],
      "metadata": {
        "id": "eyBg0dLKPND3"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# loading pretrained conv base model\n",
        "# โหลดโมเดล มาโดยตัด output ของโมเดลออก เเต่ยังใช้ input อันเดิม\n",
        "# เเละโหลด weight ของโมเดล มาด้วยที่ชื่อว่า imagenet\n",
        "conv_base = Net(weights='imagenet', include_top=False, input_shape=input_shape)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9VBe9W9-H54h",
        "outputId": "0505accd-0e21-4de8-89ab-c0bc0028e204"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:tensorflow:From /usr/local/lib/python3.8/dist-packages/tensorflow/python/autograph/pyct/static_analysis/liveness.py:83: Analyzer.lamba_check (from tensorflow.python.autograph.pyct.static_analysis.liveness) is deprecated and will be removed after 2023-09-23.\n",
            "Instructions for updating:\n",
            "Lambda fuctions will be no more assumed to be used in the statement where they are used, or at least in the same block. https://github.com/tensorflow/tensorflow/issues/56089\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading data from https://github.com/qubvel/efficientnet/releases/download/v0.0.1/efficientnet-b0_imagenet_1000_notop.h5\n",
            "16717576/16717576 [==============================] - 1s 0us/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "base_dir = '/content/drive/MyDrive/TVT_Gender'\n",
        "os.makedirs(base_dir, exist_ok=True)\n",
        "\n",
        "# Directories for our training,\n",
        "# validation and test splits\n",
        "train_dir = os.path.join(base_dir, 'train')\n",
        "os.makedirs(train_dir, exist_ok=True)\n",
        "validation_dir = os.path.join(base_dir, 'validation')\n",
        "os.makedirs(validation_dir, exist_ok=True)\n",
        "test_dir = os.path.join(base_dir, 'test')\n",
        "os.makedirs(test_dir, exist_ok=True)"
      ],
      "metadata": {
        "id": "Jwpq_-KvPef8"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#load model"
      ],
      "metadata": {
        "id": "od-ZSNm5PoGy"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import sys\n",
        "sys.path.append('/content/drive/MyDrive/cut_panoramic/Model/Classification/Gender/2_รอบที่2_Gender_250.h5')\n",
        "\n",
        "from efficientnet.layers import Swish, DropConnect\n",
        "from efficientnet.model import ConvKernalInitializer\n",
        "from tensorflow.keras.utils import get_custom_objects\n",
        "\n",
        "get_custom_objects().update({\n",
        "    'ConvKernalInitializer': ConvKernalInitializer,\n",
        "    'Swish': Swish,\n",
        "    'DropConnect':DropConnect\n",
        "})"
      ],
      "metadata": {
        "id": "n5iPL5MNPkhE"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#load model \n",
        "from tensorflow.keras.models import load_model\n",
        "model = load_model('/content/drive/MyDrive/cut_panoramic/Model/Classification/Gender/2_รอบที่2_Gender_250.h5')\n",
        "height = width = model.input_shape[1]"
      ],
      "metadata": {
        "id": "plYz49xMPkly"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "z6IOPBflFbvc",
        "outputId": "42954c78-584e-499e-c82b-607183103f4e"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " efficientnet-b0 (Functional  (None, 5, 5, 1280)       4049564   \n",
            " )                                                               \n",
            "                                                                 \n",
            " gap (GlobalMaxPooling2D)    (None, 1280)              0         \n",
            "                                                                 \n",
            " dropout_out (Dropout)       (None, 1280)              0         \n",
            "                                                                 \n",
            " fc_out (Dense)              (None, 2)                 2562      \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 4,052,126\n",
            "Trainable params: 2,562\n",
            "Non-trainable params: 4,049,564\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Train ด้วย ImageDataGenerator ของ Keras ซึ่งจะเพิ่มข้อมูลเสริมระหว่างการฝึกเพื่อลดโอกาสเกิด overfitting\n",
        "#overfitting เกิดจากข้อมูลที่ซับซ้อนกันเกินไป\n",
        "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
        "\n",
        "train_datagen = ImageDataGenerator(\n",
        "      rescale=1./255, #โมเดลส่วนใหญ่ต้องใช้ RGB ในช่วง 0–1\n",
        "      rotation_range=40,\n",
        "      width_shift_range=0.2,\n",
        "      height_shift_range=0.2,\n",
        "      shear_range=0.2,\n",
        "      zoom_range=0.2,\n",
        "      horizontal_flip=True,\n",
        "      fill_mode='nearest')\n",
        "\n",
        "# Note that the validation data should not be augmented!\n",
        "test_datagen = ImageDataGenerator(rescale=1./255)\n",
        "\n",
        "train_generator = train_datagen.flow_from_directory(\n",
        "        # This is the target directory #ไดเรกเป้าหมาย\n",
        "        train_dir,\n",
        "        # รูปภาพทั้งหมดจะถูกปรับขนาดตามความสูงและความกว้างของเป้าหมาย\n",
        "        target_size=(height, width),\n",
        "        batch_size=batch_size,\n",
        "        # Since we use categorical_crossentropy loss, we need categorical labels\n",
        "        #เนื่องจากเราใช้ categorical_crossentropy loss เราจึงต้องมีป้ายกำกับตามหมวดหมู่\n",
        "        class_mode='categorical')\n",
        "\n",
        "validation_generator = test_datagen.flow_from_directory( #การดึงภาพจาก Directory มาเข้าโมเดล \n",
        "        validation_dir,\n",
        "        target_size=(height, width),\n",
        "        batch_size=batch_size,\n",
        "        class_mode='categorical')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KBMZbdr2Pgw4",
        "outputId": "31da46d9-5253-4ea0-f744-73d4aafaa199"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Found 2850 images belonging to 2 classes.\n",
            "Found 950 images belonging to 2 classes.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# multiply_16\n",
        "# set 'multiply_16' and following layers trainable\n",
        "model.trainable = True\n",
        "\n",
        "set_trainable = False\n",
        "for layer in conv_base.layers:\n",
        "    if layer.name == 'multiply_16':\n",
        "        set_trainable = True\n",
        "    if set_trainable:\n",
        "        layer.trainable = True\n",
        "    else:\n",
        "        layer.trainable = False  "
      ],
      "metadata": {
        "id": "A0yv7vmgGhSI"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lJReTlOCGkRE",
        "outputId": "8dff7075-ae5f-4c5f-ced6-c8088a0522c4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " efficientnet-b0 (Functional  (None, 5, 5, 1280)       4049564   \n",
            " )                                                               \n",
            "                                                                 \n",
            " gap (GlobalMaxPooling2D)    (None, 1280)              0         \n",
            "                                                                 \n",
            " dropout_out (Dropout)       (None, 1280)              0         \n",
            "                                                                 \n",
            " fc_out (Dense)              (None, 2)                 2562      \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 4,052,126\n",
            "Trainable params: 4,010,110\n",
            "Non-trainable params: 42,016\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "history = model.fit_generator(\n",
        "      train_generator,\n",
        "      steps_per_epoch= NUM_TRAIN //batch_size,\n",
        "      epochs=epochs,\n",
        "      validation_data=validation_generator,\n",
        "      validation_steps= NUM_TEST //batch_size,\n",
        "      verbose=1,\n",
        "      use_multiprocessing=True,\n",
        "      workers=4)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "C9Cf1dwyP1PD",
        "outputId": "a4d19ad7-e21c-4fec-a05f-c813fb0c4dbf"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "<ipython-input-15-caa7b37242a8>:1: UserWarning: `Model.fit_generator` is deprecated and will be removed in a future version. Please use `Model.fit`, which supports generators.\n",
            "  history = model.fit_generator(\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/500\n",
            "178/178 [==============================] - 235s 1s/step - loss: 0.5889 - acc: 0.6817 - val_loss: 0.5472 - val_acc: 0.7256\n",
            "Epoch 2/500\n",
            "178/178 [==============================] - 16s 89ms/step - loss: 0.5931 - acc: 0.6800 - val_loss: 0.5377 - val_acc: 0.7278\n",
            "Epoch 3/500\n",
            "178/178 [==============================] - 16s 89ms/step - loss: 0.5977 - acc: 0.6845 - val_loss: 0.5430 - val_acc: 0.7246\n",
            "Epoch 4/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.6065 - acc: 0.6715 - val_loss: 0.5455 - val_acc: 0.7214\n",
            "Epoch 5/500\n",
            "178/178 [==============================] - 16s 91ms/step - loss: 0.5983 - acc: 0.6673 - val_loss: 0.5598 - val_acc: 0.7150\n",
            "Epoch 6/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5858 - acc: 0.6930 - val_loss: 0.5681 - val_acc: 0.7055\n",
            "Epoch 7/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.6012 - acc: 0.6729 - val_loss: 0.5482 - val_acc: 0.7225\n",
            "Epoch 8/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5868 - acc: 0.6997 - val_loss: 0.5396 - val_acc: 0.7267\n",
            "Epoch 9/500\n",
            "178/178 [==============================] - 16s 89ms/step - loss: 0.5941 - acc: 0.6867 - val_loss: 0.5498 - val_acc: 0.7203\n",
            "Epoch 10/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5918 - acc: 0.6817 - val_loss: 0.5486 - val_acc: 0.7193\n",
            "Epoch 11/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5948 - acc: 0.6810 - val_loss: 0.5465 - val_acc: 0.7235\n",
            "Epoch 12/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.6071 - acc: 0.6699 - val_loss: 0.5378 - val_acc: 0.7246\n",
            "Epoch 13/500\n",
            "178/178 [==============================] - 16s 89ms/step - loss: 0.5938 - acc: 0.6937 - val_loss: 0.5444 - val_acc: 0.7246\n",
            "Epoch 14/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5925 - acc: 0.6757 - val_loss: 0.5493 - val_acc: 0.7214\n",
            "Epoch 15/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6019 - acc: 0.6708 - val_loss: 0.5476 - val_acc: 0.7256\n",
            "Epoch 16/500\n",
            "178/178 [==============================] - 16s 89ms/step - loss: 0.6008 - acc: 0.6821 - val_loss: 0.5558 - val_acc: 0.7172\n",
            "Epoch 17/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.6086 - acc: 0.6634 - val_loss: 0.5563 - val_acc: 0.7182\n",
            "Epoch 18/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.6002 - acc: 0.6796 - val_loss: 0.5463 - val_acc: 0.7214\n",
            "Epoch 19/500\n",
            "178/178 [==============================] - 16s 89ms/step - loss: 0.6032 - acc: 0.6814 - val_loss: 0.5561 - val_acc: 0.7161\n",
            "Epoch 20/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5923 - acc: 0.6934 - val_loss: 0.5417 - val_acc: 0.7256\n",
            "Epoch 21/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5812 - acc: 0.6934 - val_loss: 0.5376 - val_acc: 0.7331\n",
            "Epoch 22/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5996 - acc: 0.6782 - val_loss: 0.5428 - val_acc: 0.7256\n",
            "Epoch 23/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5851 - acc: 0.6831 - val_loss: 0.5425 - val_acc: 0.7267\n",
            "Epoch 24/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.6008 - acc: 0.6715 - val_loss: 0.5384 - val_acc: 0.7299\n",
            "Epoch 25/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5956 - acc: 0.6874 - val_loss: 0.5425 - val_acc: 0.7256\n",
            "Epoch 26/500\n",
            "178/178 [==============================] - 12s 63ms/step - loss: 0.5989 - acc: 0.6821 - val_loss: 0.5509 - val_acc: 0.7108\n",
            "Epoch 27/500\n",
            "178/178 [==============================] - 16s 89ms/step - loss: 0.6064 - acc: 0.6757 - val_loss: 0.5430 - val_acc: 0.7256\n",
            "Epoch 28/500\n",
            "178/178 [==============================] - 16s 89ms/step - loss: 0.6090 - acc: 0.6761 - val_loss: 0.5389 - val_acc: 0.7278\n",
            "Epoch 29/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5993 - acc: 0.6796 - val_loss: 0.5604 - val_acc: 0.7076\n",
            "Epoch 30/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5999 - acc: 0.6807 - val_loss: 0.5413 - val_acc: 0.7267\n",
            "Epoch 31/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5965 - acc: 0.6916 - val_loss: 0.5623 - val_acc: 0.7055\n",
            "Epoch 32/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5910 - acc: 0.6948 - val_loss: 0.5603 - val_acc: 0.7087\n",
            "Epoch 33/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5982 - acc: 0.6842 - val_loss: 0.5725 - val_acc: 0.7023\n",
            "Epoch 34/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5949 - acc: 0.6817 - val_loss: 0.5590 - val_acc: 0.7087\n",
            "Epoch 35/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5869 - acc: 0.6930 - val_loss: 0.5753 - val_acc: 0.7002\n",
            "Epoch 36/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.6047 - acc: 0.6800 - val_loss: 0.5559 - val_acc: 0.7097\n",
            "Epoch 37/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.6064 - acc: 0.6680 - val_loss: 0.5460 - val_acc: 0.7193\n",
            "Epoch 38/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5857 - acc: 0.6983 - val_loss: 0.5454 - val_acc: 0.7203\n",
            "Epoch 39/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5954 - acc: 0.6782 - val_loss: 0.5464 - val_acc: 0.7193\n",
            "Epoch 40/500\n",
            "178/178 [==============================] - 12s 64ms/step - loss: 0.5907 - acc: 0.6898 - val_loss: 0.5418 - val_acc: 0.7246\n",
            "Epoch 41/500\n",
            "178/178 [==============================] - 17s 90ms/step - loss: 0.5907 - acc: 0.6948 - val_loss: 0.5459 - val_acc: 0.7214\n",
            "Epoch 42/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5984 - acc: 0.6870 - val_loss: 0.5602 - val_acc: 0.7119\n",
            "Epoch 43/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.6033 - acc: 0.6810 - val_loss: 0.5537 - val_acc: 0.7119\n",
            "Epoch 44/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5887 - acc: 0.6817 - val_loss: 0.5463 - val_acc: 0.7214\n",
            "Epoch 45/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5880 - acc: 0.6909 - val_loss: 0.5478 - val_acc: 0.7150\n",
            "Epoch 46/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5921 - acc: 0.6972 - val_loss: 0.5485 - val_acc: 0.7140\n",
            "Epoch 47/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5948 - acc: 0.6948 - val_loss: 0.5503 - val_acc: 0.7129\n",
            "Epoch 48/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5995 - acc: 0.6853 - val_loss: 0.5488 - val_acc: 0.7140\n",
            "Epoch 49/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5865 - acc: 0.7011 - val_loss: 0.5336 - val_acc: 0.7256\n",
            "Epoch 50/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5953 - acc: 0.6856 - val_loss: 0.5368 - val_acc: 0.7278\n",
            "Epoch 51/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5910 - acc: 0.6867 - val_loss: 0.5534 - val_acc: 0.7140\n",
            "Epoch 52/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5958 - acc: 0.6909 - val_loss: 0.5375 - val_acc: 0.7246\n",
            "Epoch 53/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5824 - acc: 0.6912 - val_loss: 0.5475 - val_acc: 0.7182\n",
            "Epoch 54/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.6029 - acc: 0.6800 - val_loss: 0.5433 - val_acc: 0.7214\n",
            "Epoch 55/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5922 - acc: 0.6838 - val_loss: 0.5451 - val_acc: 0.7203\n",
            "Epoch 56/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5979 - acc: 0.6761 - val_loss: 0.5415 - val_acc: 0.7246\n",
            "Epoch 57/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.6041 - acc: 0.6870 - val_loss: 0.5409 - val_acc: 0.7203\n",
            "Epoch 58/500\n",
            "178/178 [==============================] - 16s 89ms/step - loss: 0.6014 - acc: 0.6817 - val_loss: 0.5361 - val_acc: 0.7256\n",
            "Epoch 59/500\n",
            "178/178 [==============================] - 12s 66ms/step - loss: 0.5896 - acc: 0.6923 - val_loss: 0.5507 - val_acc: 0.7140\n",
            "Epoch 60/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5947 - acc: 0.6793 - val_loss: 0.5408 - val_acc: 0.7214\n",
            "Epoch 61/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5966 - acc: 0.6835 - val_loss: 0.5453 - val_acc: 0.7246\n",
            "Epoch 62/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6050 - acc: 0.6694 - val_loss: 0.5398 - val_acc: 0.7267\n",
            "Epoch 63/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5828 - acc: 0.6941 - val_loss: 0.5411 - val_acc: 0.7278\n",
            "Epoch 64/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5971 - acc: 0.6860 - val_loss: 0.5510 - val_acc: 0.7140\n",
            "Epoch 65/500\n",
            "178/178 [==============================] - 16s 89ms/step - loss: 0.5852 - acc: 0.6976 - val_loss: 0.5441 - val_acc: 0.7225\n",
            "Epoch 66/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5928 - acc: 0.6937 - val_loss: 0.5605 - val_acc: 0.7087\n",
            "Epoch 67/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5981 - acc: 0.6905 - val_loss: 0.5338 - val_acc: 0.7278\n",
            "Epoch 68/500\n",
            "178/178 [==============================] - 16s 89ms/step - loss: 0.5976 - acc: 0.6838 - val_loss: 0.5466 - val_acc: 0.7172\n",
            "Epoch 69/500\n",
            "178/178 [==============================] - 16s 89ms/step - loss: 0.6003 - acc: 0.6725 - val_loss: 0.5469 - val_acc: 0.7203\n",
            "Epoch 70/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5763 - acc: 0.6934 - val_loss: 0.5415 - val_acc: 0.7256\n",
            "Epoch 71/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5903 - acc: 0.6948 - val_loss: 0.5391 - val_acc: 0.7299\n",
            "Epoch 72/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5920 - acc: 0.6796 - val_loss: 0.5333 - val_acc: 0.7267\n",
            "Epoch 73/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5970 - acc: 0.6800 - val_loss: 0.5557 - val_acc: 0.7150\n",
            "Epoch 74/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6064 - acc: 0.6725 - val_loss: 0.5402 - val_acc: 0.7256\n",
            "Epoch 75/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5915 - acc: 0.6902 - val_loss: 0.5462 - val_acc: 0.7193\n",
            "Epoch 76/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5996 - acc: 0.6793 - val_loss: 0.5633 - val_acc: 0.7034\n",
            "Epoch 77/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.6019 - acc: 0.6750 - val_loss: 0.5589 - val_acc: 0.7076\n",
            "Epoch 78/500\n",
            "178/178 [==============================] - 16s 89ms/step - loss: 0.6135 - acc: 0.6697 - val_loss: 0.5528 - val_acc: 0.7140\n",
            "Epoch 79/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6056 - acc: 0.6761 - val_loss: 0.5538 - val_acc: 0.7150\n",
            "Epoch 80/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5944 - acc: 0.6835 - val_loss: 0.5595 - val_acc: 0.7076\n",
            "Epoch 81/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5968 - acc: 0.6927 - val_loss: 0.5456 - val_acc: 0.7140\n",
            "Epoch 82/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5963 - acc: 0.6831 - val_loss: 0.5433 - val_acc: 0.7256\n",
            "Epoch 83/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5812 - acc: 0.6958 - val_loss: 0.5352 - val_acc: 0.7288\n",
            "Epoch 84/500\n",
            "178/178 [==============================] - 17s 90ms/step - loss: 0.5974 - acc: 0.6771 - val_loss: 0.5392 - val_acc: 0.7299\n",
            "Epoch 85/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6098 - acc: 0.6704 - val_loss: 0.5446 - val_acc: 0.7225\n",
            "Epoch 86/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.6058 - acc: 0.6658 - val_loss: 0.5324 - val_acc: 0.7278\n",
            "Epoch 87/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5995 - acc: 0.6856 - val_loss: 0.5452 - val_acc: 0.7225\n",
            "Epoch 88/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5995 - acc: 0.6800 - val_loss: 0.5327 - val_acc: 0.7309\n",
            "Epoch 89/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5952 - acc: 0.6838 - val_loss: 0.5476 - val_acc: 0.7172\n",
            "Epoch 90/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5938 - acc: 0.6725 - val_loss: 0.5592 - val_acc: 0.7129\n",
            "Epoch 91/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5934 - acc: 0.6920 - val_loss: 0.5424 - val_acc: 0.7182\n",
            "Epoch 92/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6000 - acc: 0.6771 - val_loss: 0.5455 - val_acc: 0.7182\n",
            "Epoch 93/500\n",
            "178/178 [==============================] - 17s 90ms/step - loss: 0.6027 - acc: 0.6881 - val_loss: 0.5324 - val_acc: 0.7278\n",
            "Epoch 94/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5904 - acc: 0.6860 - val_loss: 0.5338 - val_acc: 0.7320\n",
            "Epoch 95/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6030 - acc: 0.6807 - val_loss: 0.5259 - val_acc: 0.7320\n",
            "Epoch 96/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6037 - acc: 0.6740 - val_loss: 0.5481 - val_acc: 0.7193\n",
            "Epoch 97/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5995 - acc: 0.6763 - val_loss: 0.5382 - val_acc: 0.7214\n",
            "Epoch 98/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5966 - acc: 0.6761 - val_loss: 0.5370 - val_acc: 0.7299\n",
            "Epoch 99/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5998 - acc: 0.6803 - val_loss: 0.5453 - val_acc: 0.7256\n",
            "Epoch 100/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5827 - acc: 0.6937 - val_loss: 0.5376 - val_acc: 0.7256\n",
            "Epoch 101/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5951 - acc: 0.6814 - val_loss: 0.5442 - val_acc: 0.7193\n",
            "Epoch 102/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6013 - acc: 0.6800 - val_loss: 0.5555 - val_acc: 0.7108\n",
            "Epoch 103/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5955 - acc: 0.6909 - val_loss: 0.5455 - val_acc: 0.7214\n",
            "Epoch 104/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6057 - acc: 0.6821 - val_loss: 0.5364 - val_acc: 0.7225\n",
            "Epoch 105/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5923 - acc: 0.6930 - val_loss: 0.5431 - val_acc: 0.7182\n",
            "Epoch 106/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5787 - acc: 0.7025 - val_loss: 0.5388 - val_acc: 0.7214\n",
            "Epoch 107/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5924 - acc: 0.6828 - val_loss: 0.5471 - val_acc: 0.7193\n",
            "Epoch 108/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5796 - acc: 0.6937 - val_loss: 0.5398 - val_acc: 0.7182\n",
            "Epoch 109/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5844 - acc: 0.6789 - val_loss: 0.5361 - val_acc: 0.7299\n",
            "Epoch 110/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5961 - acc: 0.6838 - val_loss: 0.5353 - val_acc: 0.7278\n",
            "Epoch 111/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5958 - acc: 0.6835 - val_loss: 0.5396 - val_acc: 0.7246\n",
            "Epoch 112/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6035 - acc: 0.6743 - val_loss: 0.5352 - val_acc: 0.7246\n",
            "Epoch 113/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.6011 - acc: 0.6736 - val_loss: 0.5487 - val_acc: 0.7161\n",
            "Epoch 114/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5888 - acc: 0.6884 - val_loss: 0.5464 - val_acc: 0.7214\n",
            "Epoch 115/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6031 - acc: 0.6747 - val_loss: 0.5525 - val_acc: 0.7140\n",
            "Epoch 116/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5982 - acc: 0.6828 - val_loss: 0.5317 - val_acc: 0.7288\n",
            "Epoch 117/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5924 - acc: 0.6912 - val_loss: 0.5576 - val_acc: 0.7097\n",
            "Epoch 118/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5888 - acc: 0.6916 - val_loss: 0.5421 - val_acc: 0.7182\n",
            "Epoch 119/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5951 - acc: 0.6842 - val_loss: 0.5473 - val_acc: 0.7193\n",
            "Epoch 120/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6030 - acc: 0.6729 - val_loss: 0.5485 - val_acc: 0.7150\n",
            "Epoch 121/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5881 - acc: 0.6920 - val_loss: 0.5432 - val_acc: 0.7182\n",
            "Epoch 122/500\n",
            "178/178 [==============================] - 12s 65ms/step - loss: 0.5959 - acc: 0.6902 - val_loss: 0.5527 - val_acc: 0.7150\n",
            "Epoch 123/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5964 - acc: 0.6793 - val_loss: 0.5483 - val_acc: 0.7140\n",
            "Epoch 124/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5902 - acc: 0.7047 - val_loss: 0.5633 - val_acc: 0.7044\n",
            "Epoch 125/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5843 - acc: 0.6983 - val_loss: 0.5344 - val_acc: 0.7246\n",
            "Epoch 126/500\n",
            "178/178 [==============================] - 17s 90ms/step - loss: 0.5922 - acc: 0.6842 - val_loss: 0.5335 - val_acc: 0.7299\n",
            "Epoch 127/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5955 - acc: 0.6768 - val_loss: 0.5318 - val_acc: 0.7288\n",
            "Epoch 128/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5821 - acc: 0.6937 - val_loss: 0.5340 - val_acc: 0.7331\n",
            "Epoch 129/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5912 - acc: 0.6810 - val_loss: 0.5260 - val_acc: 0.7362\n",
            "Epoch 130/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6033 - acc: 0.6814 - val_loss: 0.5299 - val_acc: 0.7288\n",
            "Epoch 131/500\n",
            "178/178 [==============================] - 18s 96ms/step - loss: 0.5851 - acc: 0.6898 - val_loss: 0.5379 - val_acc: 0.7214\n",
            "Epoch 132/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5977 - acc: 0.6800 - val_loss: 0.5330 - val_acc: 0.7256\n",
            "Epoch 133/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5892 - acc: 0.6800 - val_loss: 0.5538 - val_acc: 0.7140\n",
            "Epoch 134/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5951 - acc: 0.6814 - val_loss: 0.5727 - val_acc: 0.7023\n",
            "Epoch 135/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5853 - acc: 0.6867 - val_loss: 0.5481 - val_acc: 0.7225\n",
            "Epoch 136/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6020 - acc: 0.6729 - val_loss: 0.5546 - val_acc: 0.7119\n",
            "Epoch 137/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5923 - acc: 0.6930 - val_loss: 0.5743 - val_acc: 0.6970\n",
            "Epoch 138/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6038 - acc: 0.6694 - val_loss: 0.5425 - val_acc: 0.7150\n",
            "Epoch 139/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5976 - acc: 0.6814 - val_loss: 0.5459 - val_acc: 0.7119\n",
            "Epoch 140/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5864 - acc: 0.6920 - val_loss: 0.5479 - val_acc: 0.7119\n",
            "Epoch 141/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5975 - acc: 0.6810 - val_loss: 0.5416 - val_acc: 0.7172\n",
            "Epoch 142/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5978 - acc: 0.6817 - val_loss: 0.5469 - val_acc: 0.7161\n",
            "Epoch 143/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6150 - acc: 0.6722 - val_loss: 0.5545 - val_acc: 0.7150\n",
            "Epoch 144/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6133 - acc: 0.6694 - val_loss: 0.5362 - val_acc: 0.7256\n",
            "Epoch 145/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5940 - acc: 0.6821 - val_loss: 0.5481 - val_acc: 0.7161\n",
            "Epoch 146/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5919 - acc: 0.6838 - val_loss: 0.5421 - val_acc: 0.7140\n",
            "Epoch 147/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6033 - acc: 0.6743 - val_loss: 0.5468 - val_acc: 0.7150\n",
            "Epoch 148/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5953 - acc: 0.6835 - val_loss: 0.5420 - val_acc: 0.7129\n",
            "Epoch 149/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5827 - acc: 0.6934 - val_loss: 0.5533 - val_acc: 0.7129\n",
            "Epoch 150/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.5957 - acc: 0.6905 - val_loss: 0.5690 - val_acc: 0.6981\n",
            "Epoch 151/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6007 - acc: 0.6853 - val_loss: 0.5490 - val_acc: 0.7182\n",
            "Epoch 152/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6090 - acc: 0.6704 - val_loss: 0.5525 - val_acc: 0.7214\n",
            "Epoch 153/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6040 - acc: 0.6835 - val_loss: 0.5311 - val_acc: 0.7267\n",
            "Epoch 154/500\n",
            "178/178 [==============================] - 16s 90ms/step - loss: 0.6138 - acc: 0.6761 - val_loss: 0.5513 - val_acc: 0.7172\n",
            "Epoch 155/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6010 - acc: 0.6750 - val_loss: 0.5476 - val_acc: 0.7150\n",
            "Epoch 156/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6074 - acc: 0.6701 - val_loss: 0.5490 - val_acc: 0.7119\n",
            "Epoch 157/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6010 - acc: 0.6729 - val_loss: 0.5663 - val_acc: 0.7002\n",
            "Epoch 158/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5950 - acc: 0.6824 - val_loss: 0.5458 - val_acc: 0.7140\n",
            "Epoch 159/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5907 - acc: 0.6881 - val_loss: 0.5597 - val_acc: 0.7023\n",
            "Epoch 160/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5957 - acc: 0.6701 - val_loss: 0.5436 - val_acc: 0.7246\n",
            "Epoch 161/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5982 - acc: 0.6778 - val_loss: 0.5487 - val_acc: 0.7140\n",
            "Epoch 162/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6105 - acc: 0.6757 - val_loss: 0.5386 - val_acc: 0.7203\n",
            "Epoch 163/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5902 - acc: 0.6927 - val_loss: 0.5509 - val_acc: 0.7140\n",
            "Epoch 164/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6059 - acc: 0.6722 - val_loss: 0.5617 - val_acc: 0.7034\n",
            "Epoch 165/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5980 - acc: 0.6803 - val_loss: 0.5545 - val_acc: 0.7097\n",
            "Epoch 166/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5862 - acc: 0.6842 - val_loss: 0.5472 - val_acc: 0.7140\n",
            "Epoch 167/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5927 - acc: 0.6916 - val_loss: 0.5580 - val_acc: 0.7108\n",
            "Epoch 168/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6130 - acc: 0.6616 - val_loss: 0.5492 - val_acc: 0.7140\n",
            "Epoch 169/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5839 - acc: 0.6951 - val_loss: 0.5569 - val_acc: 0.7108\n",
            "Epoch 170/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5953 - acc: 0.6849 - val_loss: 0.5492 - val_acc: 0.7182\n",
            "Epoch 171/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5970 - acc: 0.6912 - val_loss: 0.5526 - val_acc: 0.7108\n",
            "Epoch 172/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5879 - acc: 0.6909 - val_loss: 0.5384 - val_acc: 0.7182\n",
            "Epoch 173/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6003 - acc: 0.6729 - val_loss: 0.5540 - val_acc: 0.7140\n",
            "Epoch 174/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6037 - acc: 0.6722 - val_loss: 0.5417 - val_acc: 0.7172\n",
            "Epoch 175/500\n",
            "178/178 [==============================] - 17s 90ms/step - loss: 0.5894 - acc: 0.6898 - val_loss: 0.5391 - val_acc: 0.7182\n",
            "Epoch 176/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6069 - acc: 0.6768 - val_loss: 0.5401 - val_acc: 0.7172\n",
            "Epoch 177/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5925 - acc: 0.6870 - val_loss: 0.5383 - val_acc: 0.7193\n",
            "Epoch 178/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5856 - acc: 0.6898 - val_loss: 0.5399 - val_acc: 0.7214\n",
            "Epoch 179/500\n",
            "178/178 [==============================] - 17s 94ms/step - loss: 0.6015 - acc: 0.6778 - val_loss: 0.5383 - val_acc: 0.7203\n",
            "Epoch 180/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6088 - acc: 0.6715 - val_loss: 0.5533 - val_acc: 0.7119\n",
            "Epoch 181/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6009 - acc: 0.6821 - val_loss: 0.5390 - val_acc: 0.7214\n",
            "Epoch 182/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5956 - acc: 0.6863 - val_loss: 0.5312 - val_acc: 0.7225\n",
            "Epoch 183/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5970 - acc: 0.6856 - val_loss: 0.5415 - val_acc: 0.7193\n",
            "Epoch 184/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6047 - acc: 0.6736 - val_loss: 0.5423 - val_acc: 0.7203\n",
            "Epoch 185/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5956 - acc: 0.6810 - val_loss: 0.5725 - val_acc: 0.7013\n",
            "Epoch 186/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5971 - acc: 0.6722 - val_loss: 0.5408 - val_acc: 0.7193\n",
            "Epoch 187/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5893 - acc: 0.6870 - val_loss: 0.5394 - val_acc: 0.7235\n",
            "Epoch 188/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5877 - acc: 0.6983 - val_loss: 0.5404 - val_acc: 0.7225\n",
            "Epoch 189/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5942 - acc: 0.6842 - val_loss: 0.5415 - val_acc: 0.7182\n",
            "Epoch 190/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5836 - acc: 0.6962 - val_loss: 0.5438 - val_acc: 0.7193\n",
            "Epoch 191/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5868 - acc: 0.6898 - val_loss: 0.5406 - val_acc: 0.7214\n",
            "Epoch 192/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5879 - acc: 0.6853 - val_loss: 0.5534 - val_acc: 0.7182\n",
            "Epoch 193/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5999 - acc: 0.6757 - val_loss: 0.5429 - val_acc: 0.7193\n",
            "Epoch 194/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5939 - acc: 0.6891 - val_loss: 0.5331 - val_acc: 0.7225\n",
            "Epoch 195/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5928 - acc: 0.6835 - val_loss: 0.5393 - val_acc: 0.7203\n",
            "Epoch 196/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5921 - acc: 0.6810 - val_loss: 0.5506 - val_acc: 0.7150\n",
            "Epoch 197/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5927 - acc: 0.6867 - val_loss: 0.5382 - val_acc: 0.7225\n",
            "Epoch 198/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5901 - acc: 0.6944 - val_loss: 0.5348 - val_acc: 0.7309\n",
            "Epoch 199/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5897 - acc: 0.6895 - val_loss: 0.5341 - val_acc: 0.7320\n",
            "Epoch 200/500\n",
            "178/178 [==============================] - 12s 67ms/step - loss: 0.5928 - acc: 0.6853 - val_loss: 0.5458 - val_acc: 0.7172\n",
            "Epoch 201/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5911 - acc: 0.6962 - val_loss: 0.5425 - val_acc: 0.7150\n",
            "Epoch 202/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5862 - acc: 0.6881 - val_loss: 0.5443 - val_acc: 0.7119\n",
            "Epoch 203/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5893 - acc: 0.6874 - val_loss: 0.5411 - val_acc: 0.7150\n",
            "Epoch 204/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5981 - acc: 0.6853 - val_loss: 0.5333 - val_acc: 0.7267\n",
            "Epoch 205/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5993 - acc: 0.6831 - val_loss: 0.5316 - val_acc: 0.7278\n",
            "Epoch 206/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5886 - acc: 0.6909 - val_loss: 0.5341 - val_acc: 0.7225\n",
            "Epoch 207/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5974 - acc: 0.6867 - val_loss: 0.5350 - val_acc: 0.7225\n",
            "Epoch 208/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5994 - acc: 0.6782 - val_loss: 0.5346 - val_acc: 0.7246\n",
            "Epoch 209/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5845 - acc: 0.7004 - val_loss: 0.5369 - val_acc: 0.7246\n",
            "Epoch 210/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5976 - acc: 0.6803 - val_loss: 0.5403 - val_acc: 0.7214\n",
            "Epoch 211/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5924 - acc: 0.6867 - val_loss: 0.5350 - val_acc: 0.7214\n",
            "Epoch 212/500\n",
            "178/178 [==============================] - 17s 94ms/step - loss: 0.6013 - acc: 0.6800 - val_loss: 0.5304 - val_acc: 0.7331\n",
            "Epoch 213/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5964 - acc: 0.6835 - val_loss: 0.5350 - val_acc: 0.7267\n",
            "Epoch 214/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5825 - acc: 0.6870 - val_loss: 0.5335 - val_acc: 0.7256\n",
            "Epoch 215/500\n",
            "178/178 [==============================] - 18s 96ms/step - loss: 0.6052 - acc: 0.6747 - val_loss: 0.5377 - val_acc: 0.7235\n",
            "Epoch 216/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5892 - acc: 0.6937 - val_loss: 0.5370 - val_acc: 0.7225\n",
            "Epoch 217/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5958 - acc: 0.6835 - val_loss: 0.5415 - val_acc: 0.7182\n",
            "Epoch 218/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5930 - acc: 0.6835 - val_loss: 0.5495 - val_acc: 0.7140\n",
            "Epoch 219/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5969 - acc: 0.6817 - val_loss: 0.5363 - val_acc: 0.7182\n",
            "Epoch 220/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5903 - acc: 0.6958 - val_loss: 0.5556 - val_acc: 0.7129\n",
            "Epoch 221/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5943 - acc: 0.6927 - val_loss: 0.5421 - val_acc: 0.7150\n",
            "Epoch 222/500\n",
            "178/178 [==============================] - 17s 90ms/step - loss: 0.5881 - acc: 0.6923 - val_loss: 0.5444 - val_acc: 0.7150\n",
            "Epoch 223/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5920 - acc: 0.6778 - val_loss: 0.5411 - val_acc: 0.7161\n",
            "Epoch 224/500\n",
            "178/178 [==============================] - 17s 90ms/step - loss: 0.5893 - acc: 0.6877 - val_loss: 0.5430 - val_acc: 0.7150\n",
            "Epoch 225/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5994 - acc: 0.6754 - val_loss: 0.5333 - val_acc: 0.7267\n",
            "Epoch 226/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6066 - acc: 0.6757 - val_loss: 0.5581 - val_acc: 0.7066\n",
            "Epoch 227/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5818 - acc: 0.6891 - val_loss: 0.5362 - val_acc: 0.7256\n",
            "Epoch 228/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5966 - acc: 0.6807 - val_loss: 0.5377 - val_acc: 0.7193\n",
            "Epoch 229/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5928 - acc: 0.6912 - val_loss: 0.5323 - val_acc: 0.7246\n",
            "Epoch 230/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5864 - acc: 0.6905 - val_loss: 0.5402 - val_acc: 0.7129\n",
            "Epoch 231/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5941 - acc: 0.6874 - val_loss: 0.5448 - val_acc: 0.7108\n",
            "Epoch 232/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5852 - acc: 0.6920 - val_loss: 0.5468 - val_acc: 0.7172\n",
            "Epoch 233/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5968 - acc: 0.6842 - val_loss: 0.5501 - val_acc: 0.7150\n",
            "Epoch 234/500\n",
            "178/178 [==============================] - 12s 66ms/step - loss: 0.5971 - acc: 0.6898 - val_loss: 0.5419 - val_acc: 0.7182\n",
            "Epoch 235/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5927 - acc: 0.6888 - val_loss: 0.5424 - val_acc: 0.7108\n",
            "Epoch 236/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5924 - acc: 0.6828 - val_loss: 0.5358 - val_acc: 0.7246\n",
            "Epoch 237/500\n",
            "178/178 [==============================] - 12s 65ms/step - loss: 0.5832 - acc: 0.6856 - val_loss: 0.5488 - val_acc: 0.7182\n",
            "Epoch 238/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5959 - acc: 0.6842 - val_loss: 0.5310 - val_acc: 0.7256\n",
            "Epoch 239/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5951 - acc: 0.6863 - val_loss: 0.5309 - val_acc: 0.7278\n",
            "Epoch 240/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5845 - acc: 0.6951 - val_loss: 0.5293 - val_acc: 0.7309\n",
            "Epoch 241/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5905 - acc: 0.6912 - val_loss: 0.5417 - val_acc: 0.7172\n",
            "Epoch 242/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5992 - acc: 0.6884 - val_loss: 0.5504 - val_acc: 0.7108\n",
            "Epoch 243/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5874 - acc: 0.6884 - val_loss: 0.5309 - val_acc: 0.7278\n",
            "Epoch 244/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5999 - acc: 0.6771 - val_loss: 0.5327 - val_acc: 0.7299\n",
            "Epoch 245/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5984 - acc: 0.6831 - val_loss: 0.5449 - val_acc: 0.7172\n",
            "Epoch 246/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5944 - acc: 0.6800 - val_loss: 0.5420 - val_acc: 0.7161\n",
            "Epoch 247/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6032 - acc: 0.6729 - val_loss: 0.5460 - val_acc: 0.7140\n",
            "Epoch 248/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5909 - acc: 0.6849 - val_loss: 0.5381 - val_acc: 0.7182\n",
            "Epoch 249/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6032 - acc: 0.6870 - val_loss: 0.5334 - val_acc: 0.7225\n",
            "Epoch 250/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5892 - acc: 0.6962 - val_loss: 0.5412 - val_acc: 0.7193\n",
            "Epoch 251/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5934 - acc: 0.6895 - val_loss: 0.5386 - val_acc: 0.7172\n",
            "Epoch 252/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5934 - acc: 0.6842 - val_loss: 0.5510 - val_acc: 0.7129\n",
            "Epoch 253/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5876 - acc: 0.6824 - val_loss: 0.5502 - val_acc: 0.7140\n",
            "Epoch 254/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5948 - acc: 0.6817 - val_loss: 0.5310 - val_acc: 0.7193\n",
            "Epoch 255/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5875 - acc: 0.6891 - val_loss: 0.5382 - val_acc: 0.7150\n",
            "Epoch 256/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.6004 - acc: 0.6824 - val_loss: 0.5406 - val_acc: 0.7161\n",
            "Epoch 257/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6192 - acc: 0.6634 - val_loss: 0.5329 - val_acc: 0.7214\n",
            "Epoch 258/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5858 - acc: 0.6965 - val_loss: 0.5502 - val_acc: 0.7150\n",
            "Epoch 259/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6008 - acc: 0.6838 - val_loss: 0.5462 - val_acc: 0.7172\n",
            "Epoch 260/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5858 - acc: 0.6941 - val_loss: 0.5384 - val_acc: 0.7172\n",
            "Epoch 261/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5871 - acc: 0.6951 - val_loss: 0.5370 - val_acc: 0.7140\n",
            "Epoch 262/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5928 - acc: 0.6909 - val_loss: 0.5385 - val_acc: 0.7150\n",
            "Epoch 263/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6039 - acc: 0.6761 - val_loss: 0.5421 - val_acc: 0.7161\n",
            "Epoch 264/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6046 - acc: 0.6680 - val_loss: 0.5397 - val_acc: 0.7150\n",
            "Epoch 265/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6050 - acc: 0.6796 - val_loss: 0.5419 - val_acc: 0.7097\n",
            "Epoch 266/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5851 - acc: 0.6902 - val_loss: 0.5422 - val_acc: 0.7108\n",
            "Epoch 267/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6046 - acc: 0.6715 - val_loss: 0.5355 - val_acc: 0.7172\n",
            "Epoch 268/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5916 - acc: 0.6948 - val_loss: 0.5407 - val_acc: 0.7161\n",
            "Epoch 269/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5986 - acc: 0.6793 - val_loss: 0.5395 - val_acc: 0.7140\n",
            "Epoch 270/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5963 - acc: 0.6824 - val_loss: 0.5396 - val_acc: 0.7161\n",
            "Epoch 271/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6056 - acc: 0.6793 - val_loss: 0.5407 - val_acc: 0.7129\n",
            "Epoch 272/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5879 - acc: 0.6941 - val_loss: 0.5453 - val_acc: 0.7182\n",
            "Epoch 273/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5888 - acc: 0.6891 - val_loss: 0.5424 - val_acc: 0.7161\n",
            "Epoch 274/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5877 - acc: 0.6881 - val_loss: 0.5596 - val_acc: 0.7066\n",
            "Epoch 275/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.6013 - acc: 0.6916 - val_loss: 0.5378 - val_acc: 0.7150\n",
            "Epoch 276/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5893 - acc: 0.6870 - val_loss: 0.5340 - val_acc: 0.7182\n",
            "Epoch 277/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6035 - acc: 0.6768 - val_loss: 0.5386 - val_acc: 0.7225\n",
            "Epoch 278/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5919 - acc: 0.6888 - val_loss: 0.5477 - val_acc: 0.7140\n",
            "Epoch 279/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5991 - acc: 0.6803 - val_loss: 0.5463 - val_acc: 0.7182\n",
            "Epoch 280/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5996 - acc: 0.6740 - val_loss: 0.5338 - val_acc: 0.7150\n",
            "Epoch 281/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5973 - acc: 0.6754 - val_loss: 0.5363 - val_acc: 0.7214\n",
            "Epoch 282/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5984 - acc: 0.6835 - val_loss: 0.5430 - val_acc: 0.7172\n",
            "Epoch 283/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5931 - acc: 0.6902 - val_loss: 0.5384 - val_acc: 0.7214\n",
            "Epoch 284/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.6006 - acc: 0.6831 - val_loss: 0.5324 - val_acc: 0.7235\n",
            "Epoch 285/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6024 - acc: 0.6775 - val_loss: 0.5453 - val_acc: 0.7150\n",
            "Epoch 286/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5891 - acc: 0.6870 - val_loss: 0.5496 - val_acc: 0.7108\n",
            "Epoch 287/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5986 - acc: 0.6838 - val_loss: 0.5334 - val_acc: 0.7172\n",
            "Epoch 288/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6009 - acc: 0.6754 - val_loss: 0.5441 - val_acc: 0.7161\n",
            "Epoch 289/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5993 - acc: 0.6796 - val_loss: 0.5366 - val_acc: 0.7246\n",
            "Epoch 290/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6041 - acc: 0.6923 - val_loss: 0.5391 - val_acc: 0.7225\n",
            "Epoch 291/500\n",
            "178/178 [==============================] - 12s 64ms/step - loss: 0.5998 - acc: 0.6796 - val_loss: 0.5454 - val_acc: 0.7225\n",
            "Epoch 292/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6004 - acc: 0.6778 - val_loss: 0.5481 - val_acc: 0.7150\n",
            "Epoch 293/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5873 - acc: 0.6874 - val_loss: 0.5574 - val_acc: 0.7055\n",
            "Epoch 294/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6040 - acc: 0.6824 - val_loss: 0.5458 - val_acc: 0.7150\n",
            "Epoch 295/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5833 - acc: 0.6969 - val_loss: 0.5377 - val_acc: 0.7203\n",
            "Epoch 296/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5986 - acc: 0.6803 - val_loss: 0.5486 - val_acc: 0.7140\n",
            "Epoch 297/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5999 - acc: 0.6789 - val_loss: 0.5358 - val_acc: 0.7203\n",
            "Epoch 298/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5909 - acc: 0.6840 - val_loss: 0.5436 - val_acc: 0.7193\n",
            "Epoch 299/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5872 - acc: 0.6888 - val_loss: 0.5441 - val_acc: 0.7214\n",
            "Epoch 300/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5919 - acc: 0.6884 - val_loss: 0.5338 - val_acc: 0.7214\n",
            "Epoch 301/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5937 - acc: 0.6814 - val_loss: 0.5372 - val_acc: 0.7203\n",
            "Epoch 302/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6009 - acc: 0.6711 - val_loss: 0.5525 - val_acc: 0.7097\n",
            "Epoch 303/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6025 - acc: 0.6658 - val_loss: 0.5477 - val_acc: 0.7140\n",
            "Epoch 304/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5875 - acc: 0.6909 - val_loss: 0.5479 - val_acc: 0.7203\n",
            "Epoch 305/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5994 - acc: 0.6754 - val_loss: 0.5498 - val_acc: 0.7150\n",
            "Epoch 306/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5947 - acc: 0.6803 - val_loss: 0.5395 - val_acc: 0.7235\n",
            "Epoch 307/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6026 - acc: 0.6676 - val_loss: 0.5533 - val_acc: 0.7108\n",
            "Epoch 308/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5993 - acc: 0.6817 - val_loss: 0.5407 - val_acc: 0.7256\n",
            "Epoch 309/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5891 - acc: 0.6803 - val_loss: 0.5280 - val_acc: 0.7341\n",
            "Epoch 310/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5966 - acc: 0.6793 - val_loss: 0.5318 - val_acc: 0.7225\n",
            "Epoch 311/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5890 - acc: 0.6920 - val_loss: 0.5438 - val_acc: 0.7203\n",
            "Epoch 312/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5943 - acc: 0.6853 - val_loss: 0.5396 - val_acc: 0.7193\n",
            "Epoch 313/500\n",
            "178/178 [==============================] - 17s 94ms/step - loss: 0.5908 - acc: 0.6870 - val_loss: 0.5339 - val_acc: 0.7129\n",
            "Epoch 314/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6001 - acc: 0.6835 - val_loss: 0.5383 - val_acc: 0.7225\n",
            "Epoch 315/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5846 - acc: 0.6990 - val_loss: 0.5311 - val_acc: 0.7267\n",
            "Epoch 316/500\n",
            "178/178 [==============================] - 17s 96ms/step - loss: 0.5962 - acc: 0.6849 - val_loss: 0.5629 - val_acc: 0.7119\n",
            "Epoch 317/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6046 - acc: 0.6757 - val_loss: 0.5368 - val_acc: 0.7235\n",
            "Epoch 318/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5951 - acc: 0.6856 - val_loss: 0.5417 - val_acc: 0.7267\n",
            "Epoch 319/500\n",
            "178/178 [==============================] - 17s 94ms/step - loss: 0.6011 - acc: 0.6793 - val_loss: 0.5367 - val_acc: 0.7140\n",
            "Epoch 320/500\n",
            "178/178 [==============================] - 18s 97ms/step - loss: 0.5981 - acc: 0.6814 - val_loss: 0.5418 - val_acc: 0.7235\n",
            "Epoch 321/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.6025 - acc: 0.6771 - val_loss: 0.5411 - val_acc: 0.7267\n",
            "Epoch 322/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5974 - acc: 0.6831 - val_loss: 0.5350 - val_acc: 0.7140\n",
            "Epoch 323/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5962 - acc: 0.6803 - val_loss: 0.5565 - val_acc: 0.7119\n",
            "Epoch 324/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5960 - acc: 0.6821 - val_loss: 0.5396 - val_acc: 0.7161\n",
            "Epoch 325/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.6039 - acc: 0.6775 - val_loss: 0.5467 - val_acc: 0.7182\n",
            "Epoch 326/500\n",
            "178/178 [==============================] - 17s 94ms/step - loss: 0.6011 - acc: 0.6775 - val_loss: 0.5564 - val_acc: 0.7097\n",
            "Epoch 327/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5788 - acc: 0.6994 - val_loss: 0.5647 - val_acc: 0.7055\n",
            "Epoch 328/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5819 - acc: 0.7001 - val_loss: 0.5429 - val_acc: 0.7172\n",
            "Epoch 329/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6119 - acc: 0.6665 - val_loss: 0.5308 - val_acc: 0.7235\n",
            "Epoch 330/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5901 - acc: 0.6891 - val_loss: 0.5396 - val_acc: 0.7182\n",
            "Epoch 331/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5863 - acc: 0.6831 - val_loss: 0.5368 - val_acc: 0.7182\n",
            "Epoch 332/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5861 - acc: 0.6835 - val_loss: 0.5306 - val_acc: 0.7267\n",
            "Epoch 333/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5962 - acc: 0.6930 - val_loss: 0.5356 - val_acc: 0.7203\n",
            "Epoch 334/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5894 - acc: 0.6905 - val_loss: 0.5464 - val_acc: 0.7172\n",
            "Epoch 335/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5905 - acc: 0.6902 - val_loss: 0.5514 - val_acc: 0.7140\n",
            "Epoch 336/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5924 - acc: 0.6870 - val_loss: 0.5474 - val_acc: 0.7129\n",
            "Epoch 337/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5836 - acc: 0.6976 - val_loss: 0.5375 - val_acc: 0.7129\n",
            "Epoch 338/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6017 - acc: 0.6842 - val_loss: 0.5555 - val_acc: 0.7097\n",
            "Epoch 339/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5891 - acc: 0.6884 - val_loss: 0.5412 - val_acc: 0.7161\n",
            "Epoch 340/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5996 - acc: 0.6842 - val_loss: 0.5387 - val_acc: 0.7140\n",
            "Epoch 341/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6063 - acc: 0.6711 - val_loss: 0.5430 - val_acc: 0.7235\n",
            "Epoch 342/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5923 - acc: 0.6888 - val_loss: 0.5490 - val_acc: 0.7182\n",
            "Epoch 343/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5921 - acc: 0.6853 - val_loss: 0.5349 - val_acc: 0.7150\n",
            "Epoch 344/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5933 - acc: 0.6853 - val_loss: 0.5371 - val_acc: 0.7150\n",
            "Epoch 345/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6025 - acc: 0.6785 - val_loss: 0.5391 - val_acc: 0.7172\n",
            "Epoch 346/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5854 - acc: 0.6916 - val_loss: 0.5284 - val_acc: 0.7309\n",
            "Epoch 347/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5912 - acc: 0.6874 - val_loss: 0.5354 - val_acc: 0.7172\n",
            "Epoch 348/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5887 - acc: 0.6951 - val_loss: 0.5320 - val_acc: 0.7214\n",
            "Epoch 349/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5948 - acc: 0.6877 - val_loss: 0.5430 - val_acc: 0.7150\n",
            "Epoch 350/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.6030 - acc: 0.6817 - val_loss: 0.5549 - val_acc: 0.7097\n",
            "Epoch 351/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5929 - acc: 0.6835 - val_loss: 0.5342 - val_acc: 0.7182\n",
            "Epoch 352/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6077 - acc: 0.6750 - val_loss: 0.5428 - val_acc: 0.7119\n",
            "Epoch 353/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.6047 - acc: 0.6736 - val_loss: 0.5362 - val_acc: 0.7182\n",
            "Epoch 354/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5980 - acc: 0.6927 - val_loss: 0.5308 - val_acc: 0.7256\n",
            "Epoch 355/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5997 - acc: 0.6778 - val_loss: 0.5389 - val_acc: 0.7182\n",
            "Epoch 356/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5920 - acc: 0.6796 - val_loss: 0.5369 - val_acc: 0.7150\n",
            "Epoch 357/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5927 - acc: 0.6969 - val_loss: 0.5423 - val_acc: 0.7140\n",
            "Epoch 358/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5950 - acc: 0.6842 - val_loss: 0.5488 - val_acc: 0.7172\n",
            "Epoch 359/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5957 - acc: 0.6842 - val_loss: 0.5518 - val_acc: 0.7097\n",
            "Epoch 360/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6017 - acc: 0.6747 - val_loss: 0.5396 - val_acc: 0.7140\n",
            "Epoch 361/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5999 - acc: 0.6785 - val_loss: 0.5648 - val_acc: 0.7066\n",
            "Epoch 362/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5930 - acc: 0.6853 - val_loss: 0.5509 - val_acc: 0.7172\n",
            "Epoch 363/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5938 - acc: 0.6793 - val_loss: 0.5397 - val_acc: 0.7193\n",
            "Epoch 364/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5905 - acc: 0.6860 - val_loss: 0.5430 - val_acc: 0.7214\n",
            "Epoch 365/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5974 - acc: 0.6941 - val_loss: 0.5572 - val_acc: 0.7087\n",
            "Epoch 366/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5777 - acc: 0.6912 - val_loss: 0.5580 - val_acc: 0.7108\n",
            "Epoch 367/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5948 - acc: 0.6909 - val_loss: 0.5418 - val_acc: 0.7214\n",
            "Epoch 368/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.6056 - acc: 0.6708 - val_loss: 0.5467 - val_acc: 0.7203\n",
            "Epoch 369/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6089 - acc: 0.6655 - val_loss: 0.5457 - val_acc: 0.7193\n",
            "Epoch 370/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6030 - acc: 0.6867 - val_loss: 0.5430 - val_acc: 0.7150\n",
            "Epoch 371/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5874 - acc: 0.6881 - val_loss: 0.5414 - val_acc: 0.7076\n",
            "Epoch 372/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5977 - acc: 0.6888 - val_loss: 0.5447 - val_acc: 0.7172\n",
            "Epoch 373/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6075 - acc: 0.6715 - val_loss: 0.5354 - val_acc: 0.7182\n",
            "Epoch 374/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5787 - acc: 0.7032 - val_loss: 0.5355 - val_acc: 0.7161\n",
            "Epoch 375/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5940 - acc: 0.6817 - val_loss: 0.5524 - val_acc: 0.7140\n",
            "Epoch 376/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5999 - acc: 0.6718 - val_loss: 0.5545 - val_acc: 0.7161\n",
            "Epoch 377/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6042 - acc: 0.6810 - val_loss: 0.5515 - val_acc: 0.7182\n",
            "Epoch 378/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5871 - acc: 0.6856 - val_loss: 0.5516 - val_acc: 0.7172\n",
            "Epoch 379/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5945 - acc: 0.6856 - val_loss: 0.5318 - val_acc: 0.7225\n",
            "Epoch 380/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5971 - acc: 0.6817 - val_loss: 0.5403 - val_acc: 0.7108\n",
            "Epoch 381/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5932 - acc: 0.6803 - val_loss: 0.5378 - val_acc: 0.7161\n",
            "Epoch 382/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5958 - acc: 0.6838 - val_loss: 0.5420 - val_acc: 0.7108\n",
            "Epoch 383/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5911 - acc: 0.6881 - val_loss: 0.5436 - val_acc: 0.7150\n",
            "Epoch 384/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5861 - acc: 0.6842 - val_loss: 0.5393 - val_acc: 0.7140\n",
            "Epoch 385/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6007 - acc: 0.6747 - val_loss: 0.5458 - val_acc: 0.7119\n",
            "Epoch 386/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5907 - acc: 0.6803 - val_loss: 0.5398 - val_acc: 0.7129\n",
            "Epoch 387/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.6068 - acc: 0.6768 - val_loss: 0.5400 - val_acc: 0.7108\n",
            "Epoch 388/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6010 - acc: 0.6888 - val_loss: 0.5424 - val_acc: 0.7097\n",
            "Epoch 389/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5855 - acc: 0.6972 - val_loss: 0.5366 - val_acc: 0.7182\n",
            "Epoch 390/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6017 - acc: 0.6807 - val_loss: 0.5409 - val_acc: 0.7097\n",
            "Epoch 391/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.6029 - acc: 0.6729 - val_loss: 0.5331 - val_acc: 0.7235\n",
            "Epoch 392/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5972 - acc: 0.6775 - val_loss: 0.5381 - val_acc: 0.7182\n",
            "Epoch 393/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.6035 - acc: 0.6757 - val_loss: 0.5508 - val_acc: 0.7182\n",
            "Epoch 394/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5979 - acc: 0.6810 - val_loss: 0.5367 - val_acc: 0.7203\n",
            "Epoch 395/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5991 - acc: 0.6810 - val_loss: 0.5466 - val_acc: 0.7108\n",
            "Epoch 396/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5920 - acc: 0.6845 - val_loss: 0.5424 - val_acc: 0.7087\n",
            "Epoch 397/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5973 - acc: 0.6835 - val_loss: 0.5315 - val_acc: 0.7299\n",
            "Epoch 398/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5872 - acc: 0.6853 - val_loss: 0.5357 - val_acc: 0.7172\n",
            "Epoch 399/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.6008 - acc: 0.6764 - val_loss: 0.5346 - val_acc: 0.7161\n",
            "Epoch 400/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5937 - acc: 0.6842 - val_loss: 0.5303 - val_acc: 0.7246\n",
            "Epoch 401/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.6020 - acc: 0.6810 - val_loss: 0.5345 - val_acc: 0.7203\n",
            "Epoch 402/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5937 - acc: 0.6785 - val_loss: 0.5380 - val_acc: 0.7129\n",
            "Epoch 403/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5935 - acc: 0.6849 - val_loss: 0.5334 - val_acc: 0.7182\n",
            "Epoch 404/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5868 - acc: 0.6923 - val_loss: 0.5448 - val_acc: 0.7225\n",
            "Epoch 405/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5983 - acc: 0.6867 - val_loss: 0.5339 - val_acc: 0.7203\n",
            "Epoch 406/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5923 - acc: 0.6814 - val_loss: 0.5417 - val_acc: 0.7076\n",
            "Epoch 407/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5971 - acc: 0.6775 - val_loss: 0.5445 - val_acc: 0.7246\n",
            "Epoch 408/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5948 - acc: 0.6870 - val_loss: 0.5435 - val_acc: 0.7193\n",
            "Epoch 409/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5982 - acc: 0.6803 - val_loss: 0.5334 - val_acc: 0.7150\n",
            "Epoch 410/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5882 - acc: 0.6838 - val_loss: 0.5283 - val_acc: 0.7246\n",
            "Epoch 411/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6028 - acc: 0.6831 - val_loss: 0.5764 - val_acc: 0.6992\n",
            "Epoch 412/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5985 - acc: 0.6785 - val_loss: 0.5329 - val_acc: 0.7193\n",
            "Epoch 413/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5920 - acc: 0.6860 - val_loss: 0.5469 - val_acc: 0.7203\n",
            "Epoch 414/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5941 - acc: 0.6807 - val_loss: 0.5357 - val_acc: 0.7161\n",
            "Epoch 415/500\n",
            "178/178 [==============================] - 17s 91ms/step - loss: 0.5959 - acc: 0.6761 - val_loss: 0.5302 - val_acc: 0.7225\n",
            "Epoch 416/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5961 - acc: 0.6761 - val_loss: 0.5453 - val_acc: 0.7172\n",
            "Epoch 417/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5958 - acc: 0.6821 - val_loss: 0.5467 - val_acc: 0.7182\n",
            "Epoch 418/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5821 - acc: 0.6958 - val_loss: 0.5372 - val_acc: 0.7150\n",
            "Epoch 419/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5902 - acc: 0.6905 - val_loss: 0.5278 - val_acc: 0.7256\n",
            "Epoch 420/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5910 - acc: 0.6895 - val_loss: 0.5472 - val_acc: 0.7182\n",
            "Epoch 421/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5868 - acc: 0.6867 - val_loss: 0.5503 - val_acc: 0.7193\n",
            "Epoch 422/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5996 - acc: 0.6881 - val_loss: 0.5358 - val_acc: 0.7203\n",
            "Epoch 423/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.6003 - acc: 0.6768 - val_loss: 0.5254 - val_acc: 0.7341\n",
            "Epoch 424/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5938 - acc: 0.6863 - val_loss: 0.5268 - val_acc: 0.7288\n",
            "Epoch 425/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5885 - acc: 0.6845 - val_loss: 0.5348 - val_acc: 0.7193\n",
            "Epoch 426/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.6009 - acc: 0.6747 - val_loss: 0.5278 - val_acc: 0.7278\n",
            "Epoch 427/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.6011 - acc: 0.6711 - val_loss: 0.5380 - val_acc: 0.7225\n",
            "Epoch 428/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6085 - acc: 0.6782 - val_loss: 0.5275 - val_acc: 0.7235\n",
            "Epoch 429/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5894 - acc: 0.6881 - val_loss: 0.5280 - val_acc: 0.7267\n",
            "Epoch 430/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5879 - acc: 0.6838 - val_loss: 0.5299 - val_acc: 0.7235\n",
            "Epoch 431/500\n",
            "178/178 [==============================] - 17s 95ms/step - loss: 0.6067 - acc: 0.6694 - val_loss: 0.5250 - val_acc: 0.7331\n",
            "Epoch 432/500\n",
            "178/178 [==============================] - 17s 94ms/step - loss: 0.6046 - acc: 0.6740 - val_loss: 0.5366 - val_acc: 0.7150\n",
            "Epoch 433/500\n",
            "178/178 [==============================] - 17s 94ms/step - loss: 0.5943 - acc: 0.6793 - val_loss: 0.5282 - val_acc: 0.7256\n",
            "Epoch 434/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5947 - acc: 0.6997 - val_loss: 0.5303 - val_acc: 0.7235\n",
            "Epoch 435/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.6008 - acc: 0.6810 - val_loss: 0.5302 - val_acc: 0.7203\n",
            "Epoch 436/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.6040 - acc: 0.6775 - val_loss: 0.5315 - val_acc: 0.7214\n",
            "Epoch 437/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5897 - acc: 0.6912 - val_loss: 0.5294 - val_acc: 0.7246\n",
            "Epoch 438/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5975 - acc: 0.6789 - val_loss: 0.5301 - val_acc: 0.7256\n",
            "Epoch 439/500\n",
            "178/178 [==============================] - 18s 99ms/step - loss: 0.5835 - acc: 0.6916 - val_loss: 0.5421 - val_acc: 0.7066\n",
            "Epoch 440/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5929 - acc: 0.6849 - val_loss: 0.5315 - val_acc: 0.7235\n",
            "Epoch 441/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5978 - acc: 0.6807 - val_loss: 0.5306 - val_acc: 0.7203\n",
            "Epoch 442/500\n",
            "178/178 [==============================] - 17s 96ms/step - loss: 0.5910 - acc: 0.6884 - val_loss: 0.5320 - val_acc: 0.7225\n",
            "Epoch 443/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.6071 - acc: 0.6637 - val_loss: 0.5467 - val_acc: 0.7161\n",
            "Epoch 444/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5952 - acc: 0.6895 - val_loss: 0.5373 - val_acc: 0.7119\n",
            "Epoch 445/500\n",
            "178/178 [==============================] - 18s 96ms/step - loss: 0.5798 - acc: 0.6920 - val_loss: 0.5367 - val_acc: 0.7140\n",
            "Epoch 446/500\n",
            "178/178 [==============================] - 18s 98ms/step - loss: 0.5884 - acc: 0.6856 - val_loss: 0.5557 - val_acc: 0.7140\n",
            "Epoch 447/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5975 - acc: 0.6951 - val_loss: 0.5309 - val_acc: 0.7299\n",
            "Epoch 448/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5980 - acc: 0.6785 - val_loss: 0.5400 - val_acc: 0.7129\n",
            "Epoch 449/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5956 - acc: 0.6824 - val_loss: 0.5487 - val_acc: 0.7203\n",
            "Epoch 450/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5865 - acc: 0.6838 - val_loss: 0.5423 - val_acc: 0.7182\n",
            "Epoch 451/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5982 - acc: 0.6747 - val_loss: 0.5384 - val_acc: 0.7182\n",
            "Epoch 452/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.6021 - acc: 0.6782 - val_loss: 0.5346 - val_acc: 0.7193\n",
            "Epoch 453/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.6040 - acc: 0.6722 - val_loss: 0.5390 - val_acc: 0.7214\n",
            "Epoch 454/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5998 - acc: 0.6800 - val_loss: 0.5364 - val_acc: 0.7203\n",
            "Epoch 455/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5935 - acc: 0.6856 - val_loss: 0.5461 - val_acc: 0.7172\n",
            "Epoch 456/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5955 - acc: 0.6944 - val_loss: 0.5305 - val_acc: 0.7214\n",
            "Epoch 457/500\n",
            "178/178 [==============================] - 17s 94ms/step - loss: 0.5982 - acc: 0.6838 - val_loss: 0.5746 - val_acc: 0.7034\n",
            "Epoch 458/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5873 - acc: 0.6842 - val_loss: 0.5330 - val_acc: 0.7172\n",
            "Epoch 459/500\n",
            "178/178 [==============================] - 17s 94ms/step - loss: 0.5827 - acc: 0.6948 - val_loss: 0.5419 - val_acc: 0.7203\n",
            "Epoch 460/500\n",
            "178/178 [==============================] - 17s 95ms/step - loss: 0.5900 - acc: 0.6905 - val_loss: 0.5391 - val_acc: 0.7203\n",
            "Epoch 461/500\n",
            "178/178 [==============================] - 17s 94ms/step - loss: 0.5945 - acc: 0.6768 - val_loss: 0.5385 - val_acc: 0.7193\n",
            "Epoch 462/500\n",
            "178/178 [==============================] - 17s 94ms/step - loss: 0.6026 - acc: 0.6743 - val_loss: 0.5402 - val_acc: 0.7193\n",
            "Epoch 463/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5978 - acc: 0.6877 - val_loss: 0.5353 - val_acc: 0.7203\n",
            "Epoch 464/500\n",
            "178/178 [==============================] - 17s 94ms/step - loss: 0.6095 - acc: 0.6613 - val_loss: 0.5349 - val_acc: 0.7203\n",
            "Epoch 465/500\n",
            "178/178 [==============================] - 17s 94ms/step - loss: 0.5935 - acc: 0.6863 - val_loss: 0.5401 - val_acc: 0.7182\n",
            "Epoch 466/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5959 - acc: 0.6821 - val_loss: 0.5564 - val_acc: 0.7161\n",
            "Epoch 467/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.6026 - acc: 0.6771 - val_loss: 0.5434 - val_acc: 0.7214\n",
            "Epoch 468/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5888 - acc: 0.6867 - val_loss: 0.5403 - val_acc: 0.7214\n",
            "Epoch 469/500\n",
            "178/178 [==============================] - 17s 94ms/step - loss: 0.5865 - acc: 0.6923 - val_loss: 0.5303 - val_acc: 0.7193\n",
            "Epoch 470/500\n",
            "178/178 [==============================] - 17s 94ms/step - loss: 0.5860 - acc: 0.6955 - val_loss: 0.5280 - val_acc: 0.7246\n",
            "Epoch 471/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5982 - acc: 0.6814 - val_loss: 0.5559 - val_acc: 0.7140\n",
            "Epoch 472/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5885 - acc: 0.6845 - val_loss: 0.5499 - val_acc: 0.7203\n",
            "Epoch 473/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5939 - acc: 0.6849 - val_loss: 0.5459 - val_acc: 0.7203\n",
            "Epoch 474/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5952 - acc: 0.6853 - val_loss: 0.5446 - val_acc: 0.7182\n",
            "Epoch 475/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5924 - acc: 0.6718 - val_loss: 0.5475 - val_acc: 0.7161\n",
            "Epoch 476/500\n",
            "178/178 [==============================] - 17s 94ms/step - loss: 0.5932 - acc: 0.6785 - val_loss: 0.5519 - val_acc: 0.7193\n",
            "Epoch 477/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5936 - acc: 0.6835 - val_loss: 0.5500 - val_acc: 0.7214\n",
            "Epoch 478/500\n",
            "178/178 [==============================] - 17s 95ms/step - loss: 0.5811 - acc: 0.6965 - val_loss: 0.5488 - val_acc: 0.7182\n",
            "Epoch 479/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5955 - acc: 0.6845 - val_loss: 0.5491 - val_acc: 0.7161\n",
            "Epoch 480/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5881 - acc: 0.6912 - val_loss: 0.5426 - val_acc: 0.7140\n",
            "Epoch 481/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5796 - acc: 0.6920 - val_loss: 0.5420 - val_acc: 0.7161\n",
            "Epoch 482/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5913 - acc: 0.6856 - val_loss: 0.5304 - val_acc: 0.7256\n",
            "Epoch 483/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5929 - acc: 0.6845 - val_loss: 0.5339 - val_acc: 0.7172\n",
            "Epoch 484/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5879 - acc: 0.6800 - val_loss: 0.5303 - val_acc: 0.7267\n",
            "Epoch 485/500\n",
            "178/178 [==============================] - 17s 94ms/step - loss: 0.5970 - acc: 0.6817 - val_loss: 0.5318 - val_acc: 0.7225\n",
            "Epoch 486/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6020 - acc: 0.6814 - val_loss: 0.5520 - val_acc: 0.7225\n",
            "Epoch 487/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.6000 - acc: 0.6824 - val_loss: 0.5394 - val_acc: 0.7161\n",
            "Epoch 488/500\n",
            "178/178 [==============================] - 17s 95ms/step - loss: 0.5901 - acc: 0.6857 - val_loss: 0.5345 - val_acc: 0.7182\n",
            "Epoch 489/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6066 - acc: 0.6722 - val_loss: 0.5366 - val_acc: 0.7172\n",
            "Epoch 490/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.6014 - acc: 0.6764 - val_loss: 0.5385 - val_acc: 0.7119\n",
            "Epoch 491/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5974 - acc: 0.6817 - val_loss: 0.5348 - val_acc: 0.7119\n",
            "Epoch 492/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.6020 - acc: 0.6807 - val_loss: 0.5726 - val_acc: 0.6970\n",
            "Epoch 493/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5956 - acc: 0.6800 - val_loss: 0.5360 - val_acc: 0.7150\n",
            "Epoch 494/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5952 - acc: 0.6824 - val_loss: 0.5464 - val_acc: 0.7193\n",
            "Epoch 495/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5948 - acc: 0.6761 - val_loss: 0.5433 - val_acc: 0.7161\n",
            "Epoch 496/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5863 - acc: 0.6891 - val_loss: 0.5430 - val_acc: 0.7150\n",
            "Epoch 497/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5908 - acc: 0.6884 - val_loss: 0.5312 - val_acc: 0.7246\n",
            "Epoch 498/500\n",
            "178/178 [==============================] - 17s 92ms/step - loss: 0.5993 - acc: 0.6828 - val_loss: 0.5363 - val_acc: 0.7193\n",
            "Epoch 499/500\n",
            "178/178 [==============================] - 17s 93ms/step - loss: 0.5829 - acc: 0.6983 - val_loss: 0.5432 - val_acc: 0.7182\n",
            "Epoch 500/500\n",
            "178/178 [==============================] - 17s 95ms/step - loss: 0.6038 - acc: 0.6817 - val_loss: 0.5427 - val_acc: 0.7172\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "acc = history.history['acc']\n",
        "val_acc = history.history['val_acc']\n",
        "loss = history.history['loss']\n",
        "val_loss = history.history['val_loss']\n",
        "\n",
        "epochs_x = range(len(acc))\n",
        "\n",
        "plt.plot(epochs_x, acc, 'go', label='Training acc')\n",
        "plt.plot(epochs_x, val_acc, 'k', label='Validation acc')\n",
        "plt.title('Training and validation accuracy')\n",
        "plt.legend()\n",
        "\n",
        "plt.figure()\n",
        "\n",
        "plt.plot(epochs_x, loss, 'go', label='Training loss')\n",
        "plt.plot(epochs_x, val_loss, 'k', label='Validation loss')\n",
        "plt.title('Training and validation loss')\n",
        "plt.legend()\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "kwylTJpTP5XI",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 545
        },
        "outputId": "961d5046-a7d7-4678-9a62-e33c5b50da56"
      },
      "execution_count": null,
      "outputs": [
        {
          "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": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#Download the model"
      ],
      "metadata": {
        "id": "lD-vKaoHQAFd"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "os.makedirs('/content/drive/My Drive/cut_panoramic/Model', exist_ok=True)\n",
        "model.save('/content/drive/My Drive/cut_panoramic/Model/Classification/Gender/G3_New_Train_500_Unfreeze.h5')"
      ],
      "metadata": {
        "id": "74dL7-HLP_Sh"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# from google.colab import files\n",
        "# files.download('/content/drive/My Drive/cut_panoramic/Model/2_รอบที่2_Gender_250.h5')"
      ],
      "metadata": {
        "id": "qcPW-brHQDpc"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "r5WoXVo8mVpF"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}