facedetection_with_autoencoder.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4",
      "mount_file_id": "1-nZP2xSd4w9tBPldVXXpsDICuHGxZtZG",
      "authorship_tag": "ABX9TyPv8VABYLlHiEYrVx8x7sly",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/masoudshahrian/Deep-learning-code/blob/main/facedetection_with_autoencoder.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import tensorflow as tf\n",
        "from tensorflow.keras import layers, models\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import cv2\n",
        "import glob"
      ],
      "metadata": {
        "id": "9499vY3iFlRC"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Nvcy7jbBFhxX"
      },
      "outputs": [],
      "source": [
        "# Set your image dataset\n",
        "path=\"/content/drive/MyDrive/Imagetest\"\n",
        "\n",
        "# Load and preprocess the dataset\n",
        "def load_images(path, img_size=(64, 64)):  #Reduce image size\n",
        "    images = []\n",
        "    for file in glob.glob(path + \"/*.jpg\"):\n",
        "        img = cv2.imread(file)\n",
        "        img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)\n",
        "        img = cv2.resize(img, img_size)\n",
        "        images.append(img)\n",
        "    return np.array(images)\n",
        "\n",
        "\n",
        "# Load and preprocess the new image\n",
        "def load_and_preprocess_image(image_path, img_size=(64, 64)):\n",
        "    img = cv2.imread(image_path)\n",
        "    img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)\n",
        "    img = cv2.resize(img, img_size)\n",
        "    return img\n",
        "\n",
        "\n",
        "# تقسیم تصاویر به دو نیمه\n",
        "def split_images(images):\n",
        "    upper_half = []\n",
        "    lower_half = []\n",
        "    for img in images:\n",
        "        h, w, _ = img.shape\n",
        "        upper_half.append(img[:h//2, :])\n",
        "        lower_half.append(img[h//2:, :])\n",
        "    return np.array(upper_half), np.array(lower_half)\n",
        "\n",
        "\n",
        "# Split the image into upper and lower halves\n",
        "def split_image(img):\n",
        "    h, w, _ = img.shape\n",
        "    upper_half = img[:h//2, :]\n",
        "    return upper_half\n",
        "\n",
        "\n",
        "# Combine 2 halfs of image and create complete image\n",
        "def combine_and_display(upper_half, lower_half_pred):\n",
        "    full_face = np.vstack((upper_half, lower_half_pred))\n",
        "    plt.imshow(full_face)\n",
        "    plt.axis('off')\n",
        "    plt.show()\n",
        "\n",
        "\n",
        "# Load and preprocessing\n",
        "path_to_images =path  # \"/path/to/your/dataset\"  # Your imege dataset path\n",
        "images = load_images(path_to_images)\n",
        "upper_half, lower_half = split_images(images)\n",
        "\n",
        "# Normalizing Data\n",
        "upper_half = upper_half / 255.0\n",
        "lower_half = lower_half / 255.0\n",
        "\n",
        "# Initiate autoencoder\n",
        "input_shape = upper_half.shape[1:]\n",
        "\n",
        "encoder_input = layers.Input(shape=input_shape)\n",
        "x = layers.Conv2D(32, (3, 3), activation='relu', padding='same')(encoder_input)\n",
        "x = layers.MaxPooling2D((2, 2), padding='same')(x)\n",
        "x = layers.Conv2D(64, (3, 3), activation='relu', padding='same')(x)\n",
        "x = layers.MaxPooling2D((2, 2), padding='same')(x)\n",
        "x = layers.Flatten()(x)\n",
        "\n",
        "encoder_output = layers.Dense(256, activation='relu')(x)  # Number of nodes\n",
        "\n",
        "encoder = models.Model(encoder_input, encoder_output)\n"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# encoder_inp////////ut\n",
        "encoder_output"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "O0-6gDO_J0Ng",
        "outputId": "c9ee5d18-9b70-4601-8612-1dd10ca1fd89"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<KerasTensor: shape=(None, 256) dtype=float32 (created by layer 'dense')>"
            ]
          },
          "metadata": {},
          "execution_count": 3
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "decoder_input = layers.Input(shape=(256,))  # Decoder number\n",
        "x = layers.Dense(encoder_output.shape[1], activation='relu')(decoder_input)\n",
        "x = layers.Dense((input_shape[0]//4) * (input_shape[1]//4) * 64, activation='relu')(x)\n",
        "x = layers.Reshape((input_shape[0]//4, input_shape[1]//4, 64))(x)\n",
        "x = layers.Conv2DTranspose(64, (3, 3), activation='relu', padding='same')(x)\n",
        "x = layers.UpSampling2D((2, 2))(x)\n",
        "x = layers.Conv2DTranspose(32, (3, 3), activation='relu', padding='same')(x)\n",
        "x = layers.UpSampling2D((2, 2))(x)\n",
        "decoder_output = layers.Conv2DTranspose(3, (3, 3), activation='sigmoid', padding='same')(x)\n",
        "\n",
        "decoder = models.Model(decoder_input, decoder_output)\n",
        "\n",
        "autoencoder_input = layers.Input(shape=input_shape)\n",
        "encoded_img = encoder(autoencoder_input)\n",
        "decoded_img = decoder(encoded_img)\n",
        "\n",
        "autoencoder = models.Model(autoencoder_input, decoded_img)\n",
        "autoencoder.compile(optimizer='adam', loss='mse')\n",
        "\n",
        "# Training model\n",
        "autoencoder.fit(upper_half, lower_half, epochs=500, batch_size=32, validation_split=0.1)  # Epoch & batch size\n",
        "\n",
        "# ذخیره مدل\n",
        "autoencoder.save('autoencoder_model.h5')\n",
        "\n",
        "# Showing results\n",
        "def display_results(model, upper_half, lower_half):\n",
        "    predictions = model.predict(upper_half)\n",
        "    fig, axes = plt.subplots(3, 10, figsize=(15, 5))\n",
        "    for i in range(10):\n",
        "        axes[0, i].imshow(upper_half[i])\n",
        "        axes[0, i].axis('off')\n",
        "        axes[1, i].imshow(lower_half[i])\n",
        "        axes[1, i].axis('off')\n",
        "        axes[2, i].imshow(predictions[i])\n",
        "        axes[2, i].axis('off')\n",
        "    plt.show()\n",
        "\n",
        "display_results(autoencoder, upper_half, lower_half)\n",
        "\n",
        "\n",
        "\n",
        "# # ذخیره مدل\n",
        "autoencoder.save('autoencoder_model.h5')\n",
        "\n",
        "# بارگذاری مدل و انجام پیش‌بینی\n",
        "loaded_model = tf.keras.models.load_model('autoencoder_model.h5')\n",
        "\n",
        "# Path to the new image\n",
        "image_path =\"/content/test1.png\" #\"/content/resized_image.jpg\"# \"/path/to/new/image.jpg\"  # مسیر تصویر جدید\n",
        "\n",
        "# Load and preprocess the new image\n",
        "new_image = load_and_preprocess_image(image_path)\n",
        "upper_half = split_image(new_image)\n",
        "\n",
        "# Normalize the upper half\n",
        "upper_half_normalized = upper_half / 255.0\n",
        "\n",
        "# Predict the lower half using the model\n",
        "upper_half_normalized = np.expand_dims(upper_half_normalized, axis=0)  # Add batch dimension\n",
        "lower_half_pred = loaded_model.predict(upper_half_normalized)[0]\n",
        "\n",
        "# Combine the upper and predicted lower halves and display the full face\n",
        "combine_and_display(upper_half, lower_half_pred)\n",
        "\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "4jYyQFnqF9kW",
        "outputId": "e5d28b82-7cc7-485a-e23c-69f1b367b491"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/500\n",
            "148/148 [==============================] - 9s 12ms/step - loss: 0.0434 - val_loss: 0.0323\n",
            "Epoch 2/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0294 - val_loss: 0.0260\n",
            "Epoch 3/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0249 - val_loss: 0.0255\n",
            "Epoch 4/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0220 - val_loss: 0.0227\n",
            "Epoch 5/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0191 - val_loss: 0.0203\n",
            "Epoch 6/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0171 - val_loss: 0.0186\n",
            "Epoch 7/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0152 - val_loss: 0.0178\n",
            "Epoch 8/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0138 - val_loss: 0.0166\n",
            "Epoch 9/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0123 - val_loss: 0.0159\n",
            "Epoch 10/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0114 - val_loss: 0.0155\n",
            "Epoch 11/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0103 - val_loss: 0.0146\n",
            "Epoch 12/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0095 - val_loss: 0.0140\n",
            "Epoch 13/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0087 - val_loss: 0.0135\n",
            "Epoch 14/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0082 - val_loss: 0.0133\n",
            "Epoch 15/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0077 - val_loss: 0.0131\n",
            "Epoch 16/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0071 - val_loss: 0.0128\n",
            "Epoch 17/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0069 - val_loss: 0.0124\n",
            "Epoch 18/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0065 - val_loss: 0.0147\n",
            "Epoch 19/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0066 - val_loss: 0.0121\n",
            "Epoch 20/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0060 - val_loss: 0.0119\n",
            "Epoch 21/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0056 - val_loss: 0.0118\n",
            "Epoch 22/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0054 - val_loss: 0.0119\n",
            "Epoch 23/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0052 - val_loss: 0.0116\n",
            "Epoch 24/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0050 - val_loss: 0.0118\n",
            "Epoch 25/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0049 - val_loss: 0.0120\n",
            "Epoch 26/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0049 - val_loss: 0.0117\n",
            "Epoch 27/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0047 - val_loss: 0.0112\n",
            "Epoch 28/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0043 - val_loss: 0.0111\n",
            "Epoch 29/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0043 - val_loss: 0.0109\n",
            "Epoch 30/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0041 - val_loss: 0.0107\n",
            "Epoch 31/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0040 - val_loss: 0.0108\n",
            "Epoch 32/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0040 - val_loss: 0.0110\n",
            "Epoch 33/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0040 - val_loss: 0.0109\n",
            "Epoch 34/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0039 - val_loss: 0.0107\n",
            "Epoch 35/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0039 - val_loss: 0.0108\n",
            "Epoch 36/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0038 - val_loss: 0.0109\n",
            "Epoch 37/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0037 - val_loss: 0.0107\n",
            "Epoch 38/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0036 - val_loss: 0.0106\n",
            "Epoch 39/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0036 - val_loss: 0.0108\n",
            "Epoch 40/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0034 - val_loss: 0.0115\n",
            "Epoch 41/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0036 - val_loss: 0.0111\n",
            "Epoch 42/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0035 - val_loss: 0.0106\n",
            "Epoch 43/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0034 - val_loss: 0.0106\n",
            "Epoch 44/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0034 - val_loss: 0.0106\n",
            "Epoch 45/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0031 - val_loss: 0.0103\n",
            "Epoch 46/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0030 - val_loss: 0.0102\n",
            "Epoch 47/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0030 - val_loss: 0.0106\n",
            "Epoch 48/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0031 - val_loss: 0.0103\n",
            "Epoch 49/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0031 - val_loss: 0.0107\n",
            "Epoch 50/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0031 - val_loss: 0.0107\n",
            "Epoch 51/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0032 - val_loss: 0.0107\n",
            "Epoch 52/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0032 - val_loss: 0.0103\n",
            "Epoch 53/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0031 - val_loss: 0.0105\n",
            "Epoch 54/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0029 - val_loss: 0.0101\n",
            "Epoch 55/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0027 - val_loss: 0.0101\n",
            "Epoch 56/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0026 - val_loss: 0.0101\n",
            "Epoch 57/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0027 - val_loss: 0.0101\n",
            "Epoch 58/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0027 - val_loss: 0.0101\n",
            "Epoch 59/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0027 - val_loss: 0.0101\n",
            "Epoch 60/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0027 - val_loss: 0.0102\n",
            "Epoch 61/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0027 - val_loss: 0.0102\n",
            "Epoch 62/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0027 - val_loss: 0.0115\n",
            "Epoch 63/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0034 - val_loss: 0.0103\n",
            "Epoch 64/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0029 - val_loss: 0.0105\n",
            "Epoch 65/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0027 - val_loss: 0.0103\n",
            "Epoch 66/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0025 - val_loss: 0.0100\n",
            "Epoch 67/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0024 - val_loss: 0.0099\n",
            "Epoch 68/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0023 - val_loss: 0.0100\n",
            "Epoch 69/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0023 - val_loss: 0.0098\n",
            "Epoch 70/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0023 - val_loss: 0.0099\n",
            "Epoch 71/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0024 - val_loss: 0.0102\n",
            "Epoch 72/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0025 - val_loss: 0.0101\n",
            "Epoch 73/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0027 - val_loss: 0.0103\n",
            "Epoch 74/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0029 - val_loss: 0.0102\n",
            "Epoch 75/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0026 - val_loss: 0.0099\n",
            "Epoch 76/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0024 - val_loss: 0.0101\n",
            "Epoch 77/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0023 - val_loss: 0.0098\n",
            "Epoch 78/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0022 - val_loss: 0.0098\n",
            "Epoch 79/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0021 - val_loss: 0.0099\n",
            "Epoch 80/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0022 - val_loss: 0.0100\n",
            "Epoch 81/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0022 - val_loss: 0.0100\n",
            "Epoch 82/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0023 - val_loss: 0.0099\n",
            "Epoch 83/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0025 - val_loss: 0.0100\n",
            "Epoch 84/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0024 - val_loss: 0.0107\n",
            "Epoch 85/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0024 - val_loss: 0.0102\n",
            "Epoch 86/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0023 - val_loss: 0.0100\n",
            "Epoch 87/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0022 - val_loss: 0.0100\n",
            "Epoch 88/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0021 - val_loss: 0.0099\n",
            "Epoch 89/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0021 - val_loss: 0.0100\n",
            "Epoch 90/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0021 - val_loss: 0.0098\n",
            "Epoch 91/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0021 - val_loss: 0.0099\n",
            "Epoch 92/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0021 - val_loss: 0.0099\n",
            "Epoch 93/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0021 - val_loss: 0.0100\n",
            "Epoch 94/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0022 - val_loss: 0.0100\n",
            "Epoch 95/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0022 - val_loss: 0.0099\n",
            "Epoch 96/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0022 - val_loss: 0.0098\n",
            "Epoch 97/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0021 - val_loss: 0.0101\n",
            "Epoch 98/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0021 - val_loss: 0.0098\n",
            "Epoch 99/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0022 - val_loss: 0.0099\n",
            "Epoch 100/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0021 - val_loss: 0.0098\n",
            "Epoch 101/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0020 - val_loss: 0.0098\n",
            "Epoch 102/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0020 - val_loss: 0.0098\n",
            "Epoch 103/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0021 - val_loss: 0.0099\n",
            "Epoch 104/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0021 - val_loss: 0.0099\n",
            "Epoch 105/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0021 - val_loss: 0.0098\n",
            "Epoch 106/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0020 - val_loss: 0.0097\n",
            "Epoch 107/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0019 - val_loss: 0.0097\n",
            "Epoch 108/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0019 - val_loss: 0.0097\n",
            "Epoch 109/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0020 - val_loss: 0.0098\n",
            "Epoch 110/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0021 - val_loss: 0.0099\n",
            "Epoch 111/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0020 - val_loss: 0.0098\n",
            "Epoch 112/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0019 - val_loss: 0.0099\n",
            "Epoch 113/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0019 - val_loss: 0.0098\n",
            "Epoch 114/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0019 - val_loss: 0.0098\n",
            "Epoch 115/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0019 - val_loss: 0.0098\n",
            "Epoch 116/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0019 - val_loss: 0.0099\n",
            "Epoch 117/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0020 - val_loss: 0.0097\n",
            "Epoch 118/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0021 - val_loss: 0.0101\n",
            "Epoch 119/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0020 - val_loss: 0.0100\n",
            "Epoch 120/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0020 - val_loss: 0.0101\n",
            "Epoch 121/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0019 - val_loss: 0.0098\n",
            "Epoch 122/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0019 - val_loss: 0.0097\n",
            "Epoch 123/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0018 - val_loss: 0.0097\n",
            "Epoch 124/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0018 - val_loss: 0.0096\n",
            "Epoch 125/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0018 - val_loss: 0.0096\n",
            "Epoch 126/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0018 - val_loss: 0.0097\n",
            "Epoch 127/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0019 - val_loss: 0.0100\n",
            "Epoch 128/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0019 - val_loss: 0.0097\n",
            "Epoch 129/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0018 - val_loss: 0.0096\n",
            "Epoch 130/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0018 - val_loss: 0.0098\n",
            "Epoch 131/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0018 - val_loss: 0.0096\n",
            "Epoch 132/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0018 - val_loss: 0.0097\n",
            "Epoch 133/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0018 - val_loss: 0.0096\n",
            "Epoch 134/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0018 - val_loss: 0.0099\n",
            "Epoch 135/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0018 - val_loss: 0.0098\n",
            "Epoch 136/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0018 - val_loss: 0.0097\n",
            "Epoch 137/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0018 - val_loss: 0.0098\n",
            "Epoch 138/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0018 - val_loss: 0.0098\n",
            "Epoch 139/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0018 - val_loss: 0.0096\n",
            "Epoch 140/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0018 - val_loss: 0.0097\n",
            "Epoch 141/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0018 - val_loss: 0.0097\n",
            "Epoch 142/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0017 - val_loss: 0.0096\n",
            "Epoch 143/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0017 - val_loss: 0.0096\n",
            "Epoch 144/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0017 - val_loss: 0.0096\n",
            "Epoch 145/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0017 - val_loss: 0.0096\n",
            "Epoch 146/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0017 - val_loss: 0.0096\n",
            "Epoch 147/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0018 - val_loss: 0.0098\n",
            "Epoch 148/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0018 - val_loss: 0.0097\n",
            "Epoch 149/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0017 - val_loss: 0.0096\n",
            "Epoch 150/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0018 - val_loss: 0.0097\n",
            "Epoch 151/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0017 - val_loss: 0.0097\n",
            "Epoch 152/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0017 - val_loss: 0.0096\n",
            "Epoch 153/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0017 - val_loss: 0.0097\n",
            "Epoch 154/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0017 - val_loss: 0.0096\n",
            "Epoch 155/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0016 - val_loss: 0.0095\n",
            "Epoch 156/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0016 - val_loss: 0.0096\n",
            "Epoch 157/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0016 - val_loss: 0.0096\n",
            "Epoch 158/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0017 - val_loss: 0.0097\n",
            "Epoch 159/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0017 - val_loss: 0.0096\n",
            "Epoch 160/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0017 - val_loss: 0.0097\n",
            "Epoch 161/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0017 - val_loss: 0.0096\n",
            "Epoch 162/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0017 - val_loss: 0.0097\n",
            "Epoch 163/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0016 - val_loss: 0.0099\n",
            "Epoch 164/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0018 - val_loss: 0.0097\n",
            "Epoch 165/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0017 - val_loss: 0.0096\n",
            "Epoch 166/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0016 - val_loss: 0.0096\n",
            "Epoch 167/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0016 - val_loss: 0.0096\n",
            "Epoch 168/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0015 - val_loss: 0.0095\n",
            "Epoch 169/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0015 - val_loss: 0.0096\n",
            "Epoch 170/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0016 - val_loss: 0.0096\n",
            "Epoch 171/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0016 - val_loss: 0.0096\n",
            "Epoch 172/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0017 - val_loss: 0.0097\n",
            "Epoch 173/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0017 - val_loss: 0.0097\n",
            "Epoch 174/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0016 - val_loss: 0.0097\n",
            "Epoch 175/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0016 - val_loss: 0.0095\n",
            "Epoch 176/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0016 - val_loss: 0.0096\n",
            "Epoch 177/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0015 - val_loss: 0.0096\n",
            "Epoch 178/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0015 - val_loss: 0.0096\n",
            "Epoch 179/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0016 - val_loss: 0.0096\n",
            "Epoch 180/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0017 - val_loss: 0.0097\n",
            "Epoch 181/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0017 - val_loss: 0.0097\n",
            "Epoch 182/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0017 - val_loss: 0.0095\n",
            "Epoch 183/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0016 - val_loss: 0.0095\n",
            "Epoch 184/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0015 - val_loss: 0.0095\n",
            "Epoch 185/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0015 - val_loss: 0.0095\n",
            "Epoch 186/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0015 - val_loss: 0.0095\n",
            "Epoch 187/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0015 - val_loss: 0.0097\n",
            "Epoch 188/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0016 - val_loss: 0.0096\n",
            "Epoch 189/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0016 - val_loss: 0.0096\n",
            "Epoch 190/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0016 - val_loss: 0.0096\n",
            "Epoch 191/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0015 - val_loss: 0.0096\n",
            "Epoch 192/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0015 - val_loss: 0.0095\n",
            "Epoch 193/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0015 - val_loss: 0.0095\n",
            "Epoch 194/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0015 - val_loss: 0.0095\n",
            "Epoch 195/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0015 - val_loss: 0.0096\n",
            "Epoch 196/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0016 - val_loss: 0.0096\n",
            "Epoch 197/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0016 - val_loss: 0.0096\n",
            "Epoch 198/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0016 - val_loss: 0.0095\n",
            "Epoch 199/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0016 - val_loss: 0.0097\n",
            "Epoch 200/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0016 - val_loss: 0.0095\n",
            "Epoch 201/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0015 - val_loss: 0.0094\n",
            "Epoch 202/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 203/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0094\n",
            "Epoch 204/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0015 - val_loss: 0.0095\n",
            "Epoch 205/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0015 - val_loss: 0.0096\n",
            "Epoch 206/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0015 - val_loss: 0.0096\n",
            "Epoch 207/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0015 - val_loss: 0.0094\n",
            "Epoch 208/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0016 - val_loss: 0.0094\n",
            "Epoch 209/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0015 - val_loss: 0.0095\n",
            "Epoch 210/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0015 - val_loss: 0.0095\n",
            "Epoch 211/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 212/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 213/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 214/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0015 - val_loss: 0.0096\n",
            "Epoch 215/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0015 - val_loss: 0.0095\n",
            "Epoch 216/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0015 - val_loss: 0.0095\n",
            "Epoch 217/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0015 - val_loss: 0.0095\n",
            "Epoch 218/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0016 - val_loss: 0.0095\n",
            "Epoch 219/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0015 - val_loss: 0.0095\n",
            "Epoch 220/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0014 - val_loss: 0.0094\n",
            "Epoch 221/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0094\n",
            "Epoch 222/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 223/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 224/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0015 - val_loss: 0.0095\n",
            "Epoch 225/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0015 - val_loss: 0.0095\n",
            "Epoch 226/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0015 - val_loss: 0.0097\n",
            "Epoch 227/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0015 - val_loss: 0.0096\n",
            "Epoch 228/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0015 - val_loss: 0.0095\n",
            "Epoch 229/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0015 - val_loss: 0.0095\n",
            "Epoch 230/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0014 - val_loss: 0.0094\n",
            "Epoch 231/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 232/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 233/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0096\n",
            "Epoch 234/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 235/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0096\n",
            "Epoch 236/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 237/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0015 - val_loss: 0.0095\n",
            "Epoch 238/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0096\n",
            "Epoch 239/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0094\n",
            "Epoch 240/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 241/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0014 - val_loss: 0.0094\n",
            "Epoch 242/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 243/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0014 - val_loss: 0.0094\n",
            "Epoch 244/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 245/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0014 - val_loss: 0.0094\n",
            "Epoch 246/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 247/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 248/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 249/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 250/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0094\n",
            "Epoch 251/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 252/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0014 - val_loss: 0.0094\n",
            "Epoch 253/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 254/500\n",
            "148/148 [==============================] - 2s 11ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 255/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 256/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 257/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0015 - val_loss: 0.0095\n",
            "Epoch 258/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0015 - val_loss: 0.0094\n",
            "Epoch 259/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 260/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 261/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 262/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 263/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 264/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 265/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 266/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 267/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 268/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 269/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0014 - val_loss: 0.0097\n",
            "Epoch 270/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0014 - val_loss: 0.0096\n",
            "Epoch 271/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0014 - val_loss: 0.0094\n",
            "Epoch 272/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 273/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 274/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 275/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 276/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0096\n",
            "Epoch 277/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0014 - val_loss: 0.0094\n",
            "Epoch 278/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 279/500\n",
            "148/148 [==============================] - 2s 11ms/step - loss: 0.0014 - val_loss: 0.0098\n",
            "Epoch 280/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0094\n",
            "Epoch 281/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 282/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 283/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0096\n",
            "Epoch 284/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 285/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0094\n",
            "Epoch 286/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 287/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 288/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0013 - val_loss: 0.0096\n",
            "Epoch 289/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 290/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 291/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 292/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 293/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 294/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 295/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 296/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 297/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 298/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 299/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 300/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 301/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 302/500\n",
            "148/148 [==============================] - 2s 11ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 303/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0096\n",
            "Epoch 304/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 305/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 306/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 307/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 308/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 309/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 310/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 311/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 312/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0014 - val_loss: 0.0094\n",
            "Epoch 313/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 314/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 315/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 316/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 317/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 318/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 319/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 320/500\n",
            "148/148 [==============================] - 1s 7ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 321/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0102\n",
            "Epoch 322/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 323/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 324/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 325/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 326/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 327/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 328/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 329/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0098\n",
            "Epoch 330/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0014 - val_loss: 0.0095\n",
            "Epoch 331/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 332/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 333/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 334/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 335/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 336/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0012 - val_loss: 0.0098\n",
            "Epoch 337/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 338/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 339/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 340/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 341/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0096\n",
            "Epoch 342/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 343/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 344/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 345/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 346/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 347/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 348/500\n",
            "148/148 [==============================] - 2s 11ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 349/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 350/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 351/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0096\n",
            "Epoch 352/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 353/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 354/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 355/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 356/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 357/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 358/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 359/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 360/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 361/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 362/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 363/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 364/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0096\n",
            "Epoch 365/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 366/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 367/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 368/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 369/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 370/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 371/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 372/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 373/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 374/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 375/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 376/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0097\n",
            "Epoch 377/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 378/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 379/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 380/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 381/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 382/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 383/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 384/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 385/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 386/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 387/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 388/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 389/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 390/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 391/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 392/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 393/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 394/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 395/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 396/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0093\n",
            "Epoch 397/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 398/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 399/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 400/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 401/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 402/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 403/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 404/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 405/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 406/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 407/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 408/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 409/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 410/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 411/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 412/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0094\n",
            "Epoch 413/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0096\n",
            "Epoch 414/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 415/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 416/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 417/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0011 - val_loss: 0.0095\n",
            "Epoch 418/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 419/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 420/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0096\n",
            "Epoch 421/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 422/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 423/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 424/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0095\n",
            "Epoch 425/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 426/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0095\n",
            "Epoch 427/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 428/500\n",
            "148/148 [==============================] - 2s 11ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 429/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 430/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 431/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 432/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 433/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 434/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0095\n",
            "Epoch 435/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 436/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 437/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 438/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 439/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 440/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 441/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 442/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0011 - val_loss: 0.0095\n",
            "Epoch 443/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 444/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 445/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0096\n",
            "Epoch 446/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 447/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 448/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 449/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 450/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0096\n",
            "Epoch 451/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 452/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 453/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 454/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 455/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0093\n",
            "Epoch 456/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0095\n",
            "Epoch 457/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 458/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0097\n",
            "Epoch 459/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0013 - val_loss: 0.0095\n",
            "Epoch 460/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 461/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 462/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0011 - val_loss: 0.0093\n",
            "Epoch 463/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0011 - val_loss: 0.0093\n",
            "Epoch 464/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0010 - val_loss: 0.0093\n",
            "Epoch 465/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0010 - val_loss: 0.0094\n",
            "Epoch 466/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 467/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0095\n",
            "Epoch 468/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0096\n",
            "Epoch 469/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0096\n",
            "Epoch 470/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 471/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 472/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 473/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 474/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0011 - val_loss: 0.0093\n",
            "Epoch 475/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 476/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0011 - val_loss: 0.0095\n",
            "Epoch 477/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 478/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 479/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 480/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 481/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 482/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 483/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 484/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0095\n",
            "Epoch 485/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 486/500\n",
            "148/148 [==============================] - 2s 10ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 487/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 488/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 489/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 490/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 491/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 492/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 493/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 494/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0096\n",
            "Epoch 495/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 496/500\n",
            "148/148 [==============================] - 1s 9ms/step - loss: 0.0012 - val_loss: 0.0094\n",
            "Epoch 497/500\n",
            "148/148 [==============================] - 2s 11ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 498/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0012 - val_loss: 0.0095\n",
            "Epoch 499/500\n",
            "148/148 [==============================] - 1s 10ms/step - loss: 0.0011 - val_loss: 0.0094\n",
            "Epoch 500/500\n",
            "148/148 [==============================] - 1s 8ms/step - loss: 0.0011 - val_loss: 0.0095\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/keras/src/engine/training.py:3103: UserWarning: You are saving your model as an HDF5 file via `model.save()`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')`.\n",
            "  saving_api.save_model(\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "164/164 [==============================] - 1s 3ms/step\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1500x500 with 30 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1/1 [==============================] - 0s 380ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# autoencoder.save('autoencoder_model.h5')\n"
      ],
      "metadata": {
        "id": "NgSMvm9nXliC"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Load and preprocess the new image\n",
        "def load_and_preprocess_image(image_path, img_size=(64, 64)):\n",
        "    img = cv2.imread(image_path)\n",
        "    img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)\n",
        "    img = cv2.resize(img, img_size)\n",
        "    return img\n",
        "\n",
        "# Split the image into upper and lower halves\n",
        "def split_image(img):\n",
        "    h, w, _ = img.shape\n",
        "    upper_half = img[:h//2, :]\n",
        "    return upper_half\n",
        "\n",
        "# ترکیب نیمه بالا و پایین و نمایش تصویر کامل\n",
        "def combine_and_display(upper_half, lower_half_pred):\n",
        "    full_face = np.vstack((upper_half, lower_half_pred))\n",
        "    plt.imshow(full_face)\n",
        "    plt.axis('off')\n",
        "    plt.show()\n",
        "\n",
        "# Load the trained model\n",
        "loaded_model = tf.keras.models.load_model('autoencoder_model.h5')\n",
        "\n",
        "# Path to the new image\n",
        "image_path = \"/content/test1.png\"  # مسیر تصویر جدید\n",
        "\n",
        "# Load and preprocess the new image\n",
        "new_image = load_and_preprocess_image(image_path)\n",
        "upper_half = split_image(new_image)\n",
        "\n",
        "# Normalize the upper half\n",
        "upper_half_normalized = upper_half / 255.0\n",
        "\n",
        "# Predict the lower half using the model\n",
        "upper_half_normalized = np.expand_dims(upper_half_normalized, axis=0)  # Add batch dimension\n",
        "lower_half_pred = loaded_model.predict(upper_half_normalized)[0]\n",
        "\n",
        "# Combine the upper and predicted lower halves and display the full face\n",
        "combine_and_display(upper_half, lower_half_pred)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 478
        },
        "id": "wd__jgd0EJ1i",
        "outputId": "0a671dc4-d612-48d8-d1d5-777f48cf4be2"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:tensorflow:6 out of the last 171 calls to <function Model.make_predict_function.<locals>.predict_function at 0x7b3600d06a70> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has reduce_retracing=True option that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1/1 [==============================] - 0s 245ms/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:matplotlib.image:Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    }
  ]
}