chapter_3_regression.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "chapter_3_regression.ipynb",
      "provenance": [],
      "authorship_tag": "ABX9TyNDVV1zDiHSR74rHLWXL/9x",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "TPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/ademaldemir/deeplearning/blob/master/chapter_3_regression.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Chbl0aH8GV1D",
        "colab_type": "text"
      },
      "source": [
        "# 3.6. Predicting house prices: a regression example"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nG2pytTTGSvG",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from keras.datasets import boston_housing\n",
        "import numpy as np\n"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YhF8Ao0VHVr2",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "d43ee92c-491f-4690-9c24-6490c34ec1a1"
      },
      "source": [
        "(train_data, train_targets), (test_data, test_targets) = boston_housing.load_data()"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/boston_housing.npz\n",
            "57344/57026 [==============================] - 0s 0us/step\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IMA4cdgOHftJ",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "5b96e725-7160-4334-932e-6870f6840455"
      },
      "source": [
        "train_data.shape"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(404, 13)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "i8MtomDnHi7J",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "93c6dca6-2684-44de-bfa9-17e7b3892361"
      },
      "source": [
        "test_data.shape"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(102, 13)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pao5l460HkJp",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        },
        "outputId": "03fd637d-4bc5-42d1-a38e-23e1d50c9383"
      },
      "source": [
        "train_data[0]"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([  1.23247,   0.     ,   8.14   ,   0.     ,   0.538  ,   6.142  ,\n",
              "        91.7    ,   3.9769 ,   4.     , 307.     ,  21.     , 396.9    ,\n",
              "        18.72   ])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aMm25c66HpvX",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "e2716449-7d72-4de6-f25f-1b3eb0a663a0"
      },
      "source": [
        "train_targets[0]"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "15.2"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fFNpm2yFH3aE",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "abf47222-3bba-4e18-d1c9-84dd921fedb4"
      },
      "source": [
        "np.max(train_data)"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "711.0"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4hn1IadvIBQM",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "617b068d-1b23-46d6-844f-205b4b3db347"
      },
      "source": [
        "np.max(test_data)"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "711.0"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lR-vz3eGIeUQ",
        "colab_type": "text"
      },
      "source": [
        "## Normalizing the data\n",
        "\n",
        "It would be problematic to feed into a neural network values that all take wildly different ranges. The network might be able to automatically adapt to such heterogeneous data, but it would definitely make learning more difficult. A widespread best practice to deal with such data is to do feature-wise normalization: for each feature in the input data (a column in the input data matrix), you subtract the mean of the feature and divide by the standard deviation, so that the feature is centered around 0 and has a unit standard deviation. This is easily done in Numpy.\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mte8RBJXIIrm",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "e1e18b1f-8091-48d2-b1fa-ebe52744706e"
      },
      "source": [
        "mean = train_data.mean(axis=0)\n",
        "mean[0]"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "3.7451105693069318"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mu23cgOvIvoM",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "train_data -= mean"
      ],
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hLngG8zbIzGV",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "std = train_data.std(axis=0)\n",
        "train_data /= std"
      ],
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sRalCHbZI9Cm",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "test_data -= mean\n",
        "test_data /= std"
      ],
      "execution_count": 13,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TFB4TDOOJXS5",
        "colab_type": "text"
      },
      "source": [
        "## BUILDING YOUR NETWORK"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "euOA24NyKY5d",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "748b424b-8bbe-447a-ee8d-576a37516f64"
      },
      "source": [
        "train_data.shape[1]"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "13"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "R_TLY5hWJKlB",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from keras.models import Sequential\n",
        "from keras.layers import Dense\n",
        "\n",
        "def build_model():\n",
        "    model = Sequential()\n",
        "    model.add(Dense(64, activation='relu', input_shape=(train_data.shape[1],)))\n",
        "    model.add(Dense(64, activation='relu'))\n",
        "    model.add(Dense(1))\n",
        "    model.compile(optimizer='rmsprop', loss='mse', metrics=['mae'])\n",
        "    return model"
      ],
      "execution_count": 15,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5j457n7QMnlT",
        "colab_type": "text"
      },
      "source": [
        "## VALIDATING YOUR APPROACH USING K-FOLD VALIDATION"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IaVRgOlKU0au",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 85
        },
        "outputId": "692b1a3f-be2b-4e77-eb9a-eb3854c9cbf8"
      },
      "source": [
        "import numpy as np\n",
        "\n",
        "k = 4\n",
        "num_val_samples = len(train_data) // k\n",
        "num_epochs = 100\n",
        "all_scores = []\n",
        "for i in range(k):\n",
        "    print('processing fold #', i)\n",
        "    val_data = train_data[i * num_val_samples: (i + 1) * num_val_samples]    \n",
        "    val_targets = train_targets[i * num_val_samples: (i + 1) * num_val_samples]\n",
        "\n",
        "    partial_train_data = np.concatenate(                                     \n",
        "        [train_data[:i * num_val_samples],\n",
        "         train_data[(i + 1) * num_val_samples:]],\n",
        "        axis=0)\n",
        "    partial_train_targets = np.concatenate(\n",
        "        [train_targets[:i * num_val_samples],\n",
        "         train_targets[(i + 1) * num_val_samples:]],\n",
        "        axis=0)\n",
        "\n",
        "    model = build_model()                                                    \n",
        "    model.fit(partial_train_data, partial_train_targets,                     \n",
        "              epochs=num_epochs, batch_size=1, verbose=0)\n",
        "    val_mse, val_mae = model.evaluate(val_data, val_targets, verbose=0)      \n",
        "    all_scores.append(val_mae)"
      ],
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "processing fold # 0\n",
            "processing fold # 1\n",
            "processing fold # 2\n",
            "processing fold # 3\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LBGNXB1XOdV_",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "825b1d59-45fb-4b49-d84d-2c7f89ba6b12"
      },
      "source": [
        "all_scores"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[1.940312147140503, 3.009253978729248, 2.555952310562134, 2.339728593826294]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 27
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "i3NG4QjIO7io",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "e2a2f834-c6c5-44a2-8446-d13256af45b9"
      },
      "source": [
        "np.mean(all_scores)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "2.4613117575645447"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 28
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "R2M7TmeMLYN4",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "24f15289-b0e5-45d2-ab68-645fd57d40c9"
      },
      "source": [
        "num_epochs = 500\n",
        "all_mae_histories = []\n",
        "for i in range(k):\n",
        "    print('processing fold #', i)\n",
        "    val_data = train_data[i * num_val_samples: (i + 1) * num_val_samples]    \n",
        "    val_targets = train_targets[i * num_val_samples: (i + 1) * num_val_samples]\n",
        "    partial_train_data = np.concatenate(                                     \n",
        "        [train_data[:i * num_val_samples],\n",
        "         train_data[(i + 1) * num_val_samples:]],\n",
        "        axis=0)\n",
        "    partial_train_targets = np.concatenate(\n",
        "        [train_targets[:i * num_val_samples],\n",
        "         train_targets[(i + 1) * num_val_samples:]],\n",
        "        axis=0)\n",
        "\n",
        "    model = build_model()                                                    \n",
        "    history = model.fit(partial_train_data, partial_train_targets,           \n",
        "                        validation_data=(val_data, val_targets),\n",
        "                        epochs=num_epochs, batch_size=1, verbose=1)\n",
        "    mae_history = history.history['val_mae']\n",
        "    all_mae_histories.append(mae_history)"
      ],
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "processing fold # 0\n",
            "Epoch 1/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 211.9861 - mae: 11.0441 - val_loss: 41.2114 - val_mae: 3.9471\n",
            "Epoch 2/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 33.2180 - mae: 3.7827 - val_loss: 29.2155 - val_mae: 3.3515\n",
            "Epoch 3/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 23.3078 - mae: 3.1555 - val_loss: 20.7992 - val_mae: 2.8922\n",
            "Epoch 4/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 18.6543 - mae: 2.8209 - val_loss: 18.2394 - val_mae: 2.4567\n",
            "Epoch 5/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 17.6592 - mae: 2.6880 - val_loss: 16.3207 - val_mae: 2.4264\n",
            "Epoch 6/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 15.8381 - mae: 2.5922 - val_loss: 14.5610 - val_mae: 2.3227\n",
            "Epoch 7/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 14.7358 - mae: 2.4696 - val_loss: 15.1806 - val_mae: 2.3988\n",
            "Epoch 8/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 14.0793 - mae: 2.4121 - val_loss: 16.8021 - val_mae: 2.6550\n",
            "Epoch 9/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 13.9661 - mae: 2.4224 - val_loss: 13.0983 - val_mae: 2.2806\n",
            "Epoch 10/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 13.0036 - mae: 2.3676 - val_loss: 11.1862 - val_mae: 2.2805\n",
            "Epoch 11/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 12.3057 - mae: 2.3374 - val_loss: 12.0803 - val_mae: 2.1544\n",
            "Epoch 12/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 11.6899 - mae: 2.2490 - val_loss: 10.7572 - val_mae: 2.0939\n",
            "Epoch 13/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 11.7109 - mae: 2.2158 - val_loss: 11.2415 - val_mae: 2.1053\n",
            "Epoch 14/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 11.1905 - mae: 2.2692 - val_loss: 10.6190 - val_mae: 2.1290\n",
            "Epoch 15/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 10.9671 - mae: 2.1441 - val_loss: 8.9260 - val_mae: 2.1289\n",
            "Epoch 16/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 10.6057 - mae: 2.2286 - val_loss: 12.1439 - val_mae: 2.2456\n",
            "Epoch 17/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 10.7945 - mae: 2.1503 - val_loss: 10.5586 - val_mae: 2.2983\n",
            "Epoch 18/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 10.3495 - mae: 2.1684 - val_loss: 9.9353 - val_mae: 2.1639\n",
            "Epoch 19/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 9.7924 - mae: 2.0227 - val_loss: 9.0994 - val_mae: 2.1772\n",
            "Epoch 20/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 9.8998 - mae: 2.1042 - val_loss: 9.2880 - val_mae: 2.2340\n",
            "Epoch 21/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 9.7407 - mae: 2.0782 - val_loss: 8.1701 - val_mae: 2.0494\n",
            "Epoch 22/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 9.5671 - mae: 2.0430 - val_loss: 9.1517 - val_mae: 2.1150\n",
            "Epoch 23/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 9.6790 - mae: 2.0435 - val_loss: 9.4036 - val_mae: 2.0729\n",
            "Epoch 24/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 9.0397 - mae: 2.0180 - val_loss: 8.2979 - val_mae: 1.9910\n",
            "Epoch 25/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 9.1266 - mae: 2.0043 - val_loss: 8.3235 - val_mae: 1.9313\n",
            "Epoch 26/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 8.6198 - mae: 2.0108 - val_loss: 10.7574 - val_mae: 2.2247\n",
            "Epoch 27/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 9.0239 - mae: 1.9658 - val_loss: 8.1763 - val_mae: 2.0316\n",
            "Epoch 28/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 8.1666 - mae: 1.9288 - val_loss: 9.6021 - val_mae: 2.2082\n",
            "Epoch 29/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 8.0419 - mae: 1.9838 - val_loss: 10.1492 - val_mae: 2.1639\n",
            "Epoch 30/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.9082 - mae: 1.8561 - val_loss: 8.7128 - val_mae: 2.0254\n",
            "Epoch 31/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.6667 - mae: 1.9277 - val_loss: 8.5890 - val_mae: 2.1995\n",
            "Epoch 32/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.2691 - mae: 1.8170 - val_loss: 14.9029 - val_mae: 3.0284\n",
            "Epoch 33/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.5265 - mae: 1.8502 - val_loss: 8.4514 - val_mae: 2.0568\n",
            "Epoch 34/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.4197 - mae: 1.8141 - val_loss: 8.5214 - val_mae: 2.0806\n",
            "Epoch 35/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.3352 - mae: 1.8416 - val_loss: 8.6821 - val_mae: 2.0549\n",
            "Epoch 36/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.3485 - mae: 1.8235 - val_loss: 9.1746 - val_mae: 2.3254\n",
            "Epoch 37/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.4852 - mae: 1.7916 - val_loss: 10.4581 - val_mae: 2.3177\n",
            "Epoch 38/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.3206 - mae: 1.8066 - val_loss: 9.0890 - val_mae: 2.2018\n",
            "Epoch 39/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.8619 - mae: 1.7521 - val_loss: 9.1822 - val_mae: 2.1408\n",
            "Epoch 40/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.8170 - mae: 1.7185 - val_loss: 9.7656 - val_mae: 2.3308\n",
            "Epoch 41/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.4372 - mae: 1.7150 - val_loss: 10.8049 - val_mae: 2.2406\n",
            "Epoch 42/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.0735 - mae: 1.7678 - val_loss: 8.3813 - val_mae: 2.1363\n",
            "Epoch 43/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.4441 - mae: 1.6992 - val_loss: 8.9906 - val_mae: 2.2687\n",
            "Epoch 44/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.6602 - mae: 1.7072 - val_loss: 10.7081 - val_mae: 2.5307\n",
            "Epoch 45/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.6242 - mae: 1.6670 - val_loss: 9.9197 - val_mae: 2.4270\n",
            "Epoch 46/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.2121 - mae: 1.6370 - val_loss: 9.9412 - val_mae: 2.4999\n",
            "Epoch 47/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.4828 - mae: 1.7427 - val_loss: 9.6716 - val_mae: 2.2514\n",
            "Epoch 48/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.5394 - mae: 1.6879 - val_loss: 9.7133 - val_mae: 2.3821\n",
            "Epoch 49/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.4702 - mae: 1.6533 - val_loss: 10.1172 - val_mae: 2.2682\n",
            "Epoch 50/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.4131 - mae: 1.6455 - val_loss: 8.8615 - val_mae: 2.1390\n",
            "Epoch 51/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.1018 - mae: 1.6421 - val_loss: 8.9932 - val_mae: 2.1783\n",
            "Epoch 52/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.2033 - mae: 1.6578 - val_loss: 9.1447 - val_mae: 2.0928\n",
            "Epoch 53/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.3766 - mae: 1.7096 - val_loss: 8.6061 - val_mae: 2.1219\n",
            "Epoch 54/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.8894 - mae: 1.6043 - val_loss: 11.8450 - val_mae: 2.6244\n",
            "Epoch 55/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.9292 - mae: 1.6547 - val_loss: 8.5687 - val_mae: 2.0187\n",
            "Epoch 56/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.9937 - mae: 1.6230 - val_loss: 8.7199 - val_mae: 2.1383\n",
            "Epoch 57/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.9943 - mae: 1.6301 - val_loss: 8.9104 - val_mae: 2.2766\n",
            "Epoch 58/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.6867 - mae: 1.5866 - val_loss: 8.0213 - val_mae: 2.1104\n",
            "Epoch 59/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.8108 - mae: 1.6118 - val_loss: 9.1143 - val_mae: 2.2710\n",
            "Epoch 60/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.8334 - mae: 1.6021 - val_loss: 8.6717 - val_mae: 2.3118\n",
            "Epoch 61/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.7972 - mae: 1.6293 - val_loss: 11.1192 - val_mae: 2.4287\n",
            "Epoch 62/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.7343 - mae: 1.5917 - val_loss: 8.8458 - val_mae: 2.0575\n",
            "Epoch 63/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.5043 - mae: 1.5098 - val_loss: 8.6633 - val_mae: 2.0826\n",
            "Epoch 64/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.1200 - mae: 1.5094 - val_loss: 10.3186 - val_mae: 2.3848\n",
            "Epoch 65/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.7928 - mae: 1.5690 - val_loss: 10.8564 - val_mae: 2.3363\n",
            "Epoch 66/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.2817 - mae: 1.5366 - val_loss: 10.4916 - val_mae: 2.4404\n",
            "Epoch 67/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.3547 - mae: 1.5724 - val_loss: 9.6467 - val_mae: 2.1693\n",
            "Epoch 68/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.1639 - mae: 1.6074 - val_loss: 8.7801 - val_mae: 2.2338\n",
            "Epoch 69/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.0858 - mae: 1.5268 - val_loss: 9.6719 - val_mae: 2.4572\n",
            "Epoch 70/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.1626 - mae: 1.5619 - val_loss: 9.6054 - val_mae: 2.2761\n",
            "Epoch 71/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.1683 - mae: 1.5165 - val_loss: 8.4192 - val_mae: 2.1017\n",
            "Epoch 72/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.2221 - mae: 1.5556 - val_loss: 11.3593 - val_mae: 2.7773\n",
            "Epoch 73/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.3566 - mae: 1.5521 - val_loss: 8.5823 - val_mae: 2.1535\n",
            "Epoch 74/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.8391 - mae: 1.4738 - val_loss: 8.4002 - val_mae: 2.0959\n",
            "Epoch 75/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.1427 - mae: 1.5250 - val_loss: 7.7155 - val_mae: 2.1278\n",
            "Epoch 76/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.8515 - mae: 1.4792 - val_loss: 8.9448 - val_mae: 2.1603\n",
            "Epoch 77/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.0360 - mae: 1.4848 - val_loss: 8.8825 - val_mae: 2.1646\n",
            "Epoch 78/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.7874 - mae: 1.4667 - val_loss: 7.8181 - val_mae: 2.1462\n",
            "Epoch 79/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.8390 - mae: 1.5231 - val_loss: 8.4180 - val_mae: 2.0960\n",
            "Epoch 80/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.1222 - mae: 1.4757 - val_loss: 7.8342 - val_mae: 1.9749\n",
            "Epoch 81/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.6152 - mae: 1.4354 - val_loss: 8.6625 - val_mae: 2.2547\n",
            "Epoch 82/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.7041 - mae: 1.5287 - val_loss: 7.9441 - val_mae: 2.0130\n",
            "Epoch 83/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.9412 - mae: 1.4197 - val_loss: 8.3652 - val_mae: 2.2026\n",
            "Epoch 84/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.6101 - mae: 1.4390 - val_loss: 10.0704 - val_mae: 2.4498\n",
            "Epoch 85/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.5858 - mae: 1.4504 - val_loss: 10.8824 - val_mae: 2.4425\n",
            "Epoch 86/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.1851 - mae: 1.4242 - val_loss: 11.1940 - val_mae: 2.4748\n",
            "Epoch 87/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.3211 - mae: 1.4758 - val_loss: 8.2504 - val_mae: 2.1840\n",
            "Epoch 88/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.2123 - mae: 1.3843 - val_loss: 9.2405 - val_mae: 2.1849\n",
            "Epoch 89/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.5873 - mae: 1.4165 - val_loss: 9.7718 - val_mae: 2.4093\n",
            "Epoch 90/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.3189 - mae: 1.4798 - val_loss: 8.3076 - val_mae: 1.9987\n",
            "Epoch 91/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.3792 - mae: 1.3703 - val_loss: 9.4267 - val_mae: 2.2678\n",
            "Epoch 92/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.3851 - mae: 1.4384 - val_loss: 10.6285 - val_mae: 2.4612\n",
            "Epoch 93/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.9842 - mae: 1.3623 - val_loss: 9.7985 - val_mae: 2.2788\n",
            "Epoch 94/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.0748 - mae: 1.3636 - val_loss: 9.2321 - val_mae: 2.1601\n",
            "Epoch 95/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.2132 - mae: 1.4020 - val_loss: 8.5272 - val_mae: 2.1543\n",
            "Epoch 96/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.1861 - mae: 1.3919 - val_loss: 10.3693 - val_mae: 2.3065\n",
            "Epoch 97/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.1310 - mae: 1.3913 - val_loss: 8.3666 - val_mae: 2.0966\n",
            "Epoch 98/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.2699 - mae: 1.4419 - val_loss: 8.4657 - val_mae: 2.1763\n",
            "Epoch 99/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.8678 - mae: 1.3904 - val_loss: 8.2513 - val_mae: 2.1669\n",
            "Epoch 100/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.3090 - mae: 1.3573 - val_loss: 10.2993 - val_mae: 2.4828\n",
            "Epoch 101/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.0218 - mae: 1.3475 - val_loss: 8.6904 - val_mae: 2.2149\n",
            "Epoch 102/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.8522 - mae: 1.3173 - val_loss: 8.7151 - val_mae: 2.1381\n",
            "Epoch 103/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.9185 - mae: 1.3605 - val_loss: 10.8274 - val_mae: 2.3695\n",
            "Epoch 104/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.8397 - mae: 1.3885 - val_loss: 8.3453 - val_mae: 2.1475\n",
            "Epoch 105/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.9595 - mae: 1.3465 - val_loss: 10.8249 - val_mae: 2.5925\n",
            "Epoch 106/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.9221 - mae: 1.3170 - val_loss: 9.2281 - val_mae: 2.1819\n",
            "Epoch 107/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.0327 - mae: 1.3709 - val_loss: 8.9805 - val_mae: 2.2634\n",
            "Epoch 108/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.7398 - mae: 1.3585 - val_loss: 9.3477 - val_mae: 2.2164\n",
            "Epoch 109/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.9034 - mae: 1.3207 - val_loss: 8.1817 - val_mae: 2.1450\n",
            "Epoch 110/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.9708 - mae: 1.3713 - val_loss: 9.1416 - val_mae: 2.2477\n",
            "Epoch 111/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.8257 - mae: 1.3151 - val_loss: 11.1164 - val_mae: 2.4871\n",
            "Epoch 112/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.2609 - mae: 1.2554 - val_loss: 8.5513 - val_mae: 2.0847\n",
            "Epoch 113/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.6877 - mae: 1.3183 - val_loss: 10.4592 - val_mae: 2.3726\n",
            "Epoch 114/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.9017 - mae: 1.3582 - val_loss: 9.3018 - val_mae: 2.0953\n",
            "Epoch 115/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.3493 - mae: 1.2543 - val_loss: 9.1218 - val_mae: 2.2727\n",
            "Epoch 116/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.7569 - mae: 1.2947 - val_loss: 13.1061 - val_mae: 2.8315\n",
            "Epoch 117/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.5513 - mae: 1.2874 - val_loss: 8.7143 - val_mae: 2.2825\n",
            "Epoch 118/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.4600 - mae: 1.2871 - val_loss: 9.8406 - val_mae: 2.3814\n",
            "Epoch 119/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.5136 - mae: 1.2815 - val_loss: 9.4807 - val_mae: 2.1041\n",
            "Epoch 120/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.6418 - mae: 1.3737 - val_loss: 10.2666 - val_mae: 2.4525\n",
            "Epoch 121/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.5363 - mae: 1.3121 - val_loss: 8.6427 - val_mae: 2.0230\n",
            "Epoch 122/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.5715 - mae: 1.3201 - val_loss: 9.9832 - val_mae: 2.3039\n",
            "Epoch 123/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.3248 - mae: 1.2934 - val_loss: 14.3295 - val_mae: 2.8439\n",
            "Epoch 124/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.3341 - mae: 1.2746 - val_loss: 9.2875 - val_mae: 2.2080\n",
            "Epoch 125/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.4283 - mae: 1.3058 - val_loss: 8.6679 - val_mae: 2.1133\n",
            "Epoch 126/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.3680 - mae: 1.2885 - val_loss: 10.3994 - val_mae: 2.4591\n",
            "Epoch 127/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.3135 - mae: 1.2770 - val_loss: 11.0896 - val_mae: 2.5092\n",
            "Epoch 128/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.3307 - mae: 1.2526 - val_loss: 9.1199 - val_mae: 2.1312\n",
            "Epoch 129/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.2745 - mae: 1.2602 - val_loss: 8.5225 - val_mae: 2.1153\n",
            "Epoch 130/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.1776 - mae: 1.2792 - val_loss: 10.6043 - val_mae: 2.4524\n",
            "Epoch 131/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.2032 - mae: 1.2717 - val_loss: 9.8180 - val_mae: 2.3138\n",
            "Epoch 132/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.2820 - mae: 1.2795 - val_loss: 10.9561 - val_mae: 2.4013\n",
            "Epoch 133/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.2026 - mae: 1.2868 - val_loss: 11.3092 - val_mae: 2.4284\n",
            "Epoch 134/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.2436 - mae: 1.3167 - val_loss: 9.1118 - val_mae: 2.0767\n",
            "Epoch 135/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.8682 - mae: 1.2015 - val_loss: 12.3965 - val_mae: 2.6775\n",
            "Epoch 136/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.8945 - mae: 1.2767 - val_loss: 9.1182 - val_mae: 2.1906\n",
            "Epoch 137/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.2398 - mae: 1.2625 - val_loss: 11.7782 - val_mae: 2.5801\n",
            "Epoch 138/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.9852 - mae: 1.2226 - val_loss: 10.3488 - val_mae: 2.3718\n",
            "Epoch 139/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.0673 - mae: 1.2284 - val_loss: 10.7806 - val_mae: 2.3596\n",
            "Epoch 140/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.9565 - mae: 1.1900 - val_loss: 9.9935 - val_mae: 2.1705\n",
            "Epoch 141/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.0083 - mae: 1.2540 - val_loss: 9.4643 - val_mae: 2.1688\n",
            "Epoch 142/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.8607 - mae: 1.1987 - val_loss: 9.6588 - val_mae: 2.1786\n",
            "Epoch 143/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.7932 - mae: 1.2385 - val_loss: 11.6173 - val_mae: 2.4002\n",
            "Epoch 144/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.9426 - mae: 1.2228 - val_loss: 9.5834 - val_mae: 2.1460\n",
            "Epoch 145/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.9724 - mae: 1.2564 - val_loss: 11.0322 - val_mae: 2.3631\n",
            "Epoch 146/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.0411 - mae: 1.2291 - val_loss: 11.1473 - val_mae: 2.3469\n",
            "Epoch 147/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.7102 - mae: 1.1669 - val_loss: 9.8968 - val_mae: 2.2546\n",
            "Epoch 148/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.1589 - mae: 1.2112 - val_loss: 10.3345 - val_mae: 2.3558\n",
            "Epoch 149/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.5493 - mae: 1.1292 - val_loss: 10.9499 - val_mae: 2.4154\n",
            "Epoch 150/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.0229 - mae: 1.2174 - val_loss: 12.6350 - val_mae: 2.5447\n",
            "Epoch 151/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.8481 - mae: 1.1996 - val_loss: 10.6928 - val_mae: 2.2165\n",
            "Epoch 152/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.6725 - mae: 1.2111 - val_loss: 9.8841 - val_mae: 2.1287\n",
            "Epoch 153/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.9147 - mae: 1.1894 - val_loss: 10.5478 - val_mae: 2.3543\n",
            "Epoch 154/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.8119 - mae: 1.1474 - val_loss: 11.4996 - val_mae: 2.4245\n",
            "Epoch 155/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2445 - mae: 1.1005 - val_loss: 11.3648 - val_mae: 2.3828\n",
            "Epoch 156/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.7809 - mae: 1.1889 - val_loss: 10.3748 - val_mae: 2.2212\n",
            "Epoch 157/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.7843 - mae: 1.1594 - val_loss: 10.9798 - val_mae: 2.2895\n",
            "Epoch 158/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.5516 - mae: 1.1511 - val_loss: 10.1832 - val_mae: 2.2252\n",
            "Epoch 159/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.4146 - mae: 1.1415 - val_loss: 11.4309 - val_mae: 2.3529\n",
            "Epoch 160/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.7649 - mae: 1.1872 - val_loss: 10.6509 - val_mae: 2.2879\n",
            "Epoch 161/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.7309 - mae: 1.1407 - val_loss: 12.0082 - val_mae: 2.4274\n",
            "Epoch 162/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.6069 - mae: 1.1534 - val_loss: 10.8406 - val_mae: 2.2356\n",
            "Epoch 163/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.6415 - mae: 1.1460 - val_loss: 10.9365 - val_mae: 2.4544\n",
            "Epoch 164/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.5721 - mae: 1.1464 - val_loss: 10.1004 - val_mae: 2.2472\n",
            "Epoch 165/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.5759 - mae: 1.1520 - val_loss: 11.2913 - val_mae: 2.4251\n",
            "Epoch 166/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.5064 - mae: 1.1136 - val_loss: 10.6820 - val_mae: 2.2410\n",
            "Epoch 167/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.5426 - mae: 1.1074 - val_loss: 11.1877 - val_mae: 2.3808\n",
            "Epoch 168/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.5090 - mae: 1.1295 - val_loss: 11.0196 - val_mae: 2.2951\n",
            "Epoch 169/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2915 - mae: 1.0345 - val_loss: 11.4066 - val_mae: 2.4693\n",
            "Epoch 170/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3516 - mae: 1.1148 - val_loss: 10.6976 - val_mae: 2.3612\n",
            "Epoch 171/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.4977 - mae: 1.1235 - val_loss: 10.2052 - val_mae: 2.2499\n",
            "Epoch 172/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3437 - mae: 1.1085 - val_loss: 12.2072 - val_mae: 2.4366\n",
            "Epoch 173/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1896 - mae: 1.0690 - val_loss: 9.9949 - val_mae: 2.2060\n",
            "Epoch 174/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.4928 - mae: 1.1558 - val_loss: 10.1552 - val_mae: 2.1401\n",
            "Epoch 175/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1869 - mae: 1.0559 - val_loss: 10.4468 - val_mae: 2.1924\n",
            "Epoch 176/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2389 - mae: 1.0541 - val_loss: 11.4671 - val_mae: 2.3403\n",
            "Epoch 177/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2352 - mae: 1.0450 - val_loss: 9.2717 - val_mae: 2.1809\n",
            "Epoch 178/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3513 - mae: 1.0705 - val_loss: 9.1490 - val_mae: 2.2242\n",
            "Epoch 179/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3427 - mae: 1.0733 - val_loss: 10.3625 - val_mae: 2.2987\n",
            "Epoch 180/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1979 - mae: 1.0862 - val_loss: 9.6332 - val_mae: 2.1843\n",
            "Epoch 181/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.4109 - mae: 1.0926 - val_loss: 10.4622 - val_mae: 2.2361\n",
            "Epoch 182/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2861 - mae: 1.0618 - val_loss: 10.6813 - val_mae: 2.2169\n",
            "Epoch 183/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1476 - mae: 1.0378 - val_loss: 10.2746 - val_mae: 2.1975\n",
            "Epoch 184/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2231 - mae: 1.0669 - val_loss: 9.9475 - val_mae: 2.1718\n",
            "Epoch 185/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3046 - mae: 1.0656 - val_loss: 10.7699 - val_mae: 2.3230\n",
            "Epoch 186/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3362 - mae: 1.0593 - val_loss: 11.6738 - val_mae: 2.3605\n",
            "Epoch 187/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0146 - mae: 1.0165 - val_loss: 10.4465 - val_mae: 2.1874\n",
            "Epoch 188/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0396 - mae: 1.0348 - val_loss: 10.9360 - val_mae: 2.2691\n",
            "Epoch 189/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2243 - mae: 1.0423 - val_loss: 11.0368 - val_mae: 2.3496\n",
            "Epoch 190/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8889 - mae: 1.0078 - val_loss: 10.7228 - val_mae: 2.3165\n",
            "Epoch 191/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1640 - mae: 1.0570 - val_loss: 10.5544 - val_mae: 2.3384\n",
            "Epoch 192/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9805 - mae: 1.0261 - val_loss: 11.5257 - val_mae: 2.3289\n",
            "Epoch 193/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1030 - mae: 1.0695 - val_loss: 10.8367 - val_mae: 2.2575\n",
            "Epoch 194/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0435 - mae: 1.0855 - val_loss: 11.5357 - val_mae: 2.3489\n",
            "Epoch 195/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2294 - mae: 1.0409 - val_loss: 14.7910 - val_mae: 2.7347\n",
            "Epoch 196/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8579 - mae: 1.0033 - val_loss: 10.8086 - val_mae: 2.2983\n",
            "Epoch 197/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9504 - mae: 1.0018 - val_loss: 10.4955 - val_mae: 2.1813\n",
            "Epoch 198/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8828 - mae: 0.9996 - val_loss: 14.4755 - val_mae: 2.6733\n",
            "Epoch 199/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0824 - mae: 1.0854 - val_loss: 12.9525 - val_mae: 2.6362\n",
            "Epoch 200/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0069 - mae: 1.0300 - val_loss: 11.4186 - val_mae: 2.3000\n",
            "Epoch 201/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9794 - mae: 1.0196 - val_loss: 11.4277 - val_mae: 2.3765\n",
            "Epoch 202/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0219 - mae: 1.0518 - val_loss: 11.6762 - val_mae: 2.3926\n",
            "Epoch 203/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7532 - mae: 0.9825 - val_loss: 10.5784 - val_mae: 2.3740\n",
            "Epoch 204/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8393 - mae: 1.0014 - val_loss: 11.0093 - val_mae: 2.2356\n",
            "Epoch 205/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8729 - mae: 1.0044 - val_loss: 10.3098 - val_mae: 2.2399\n",
            "Epoch 206/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9480 - mae: 0.9785 - val_loss: 11.7226 - val_mae: 2.2832\n",
            "Epoch 207/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7989 - mae: 0.9993 - val_loss: 11.2447 - val_mae: 2.2448\n",
            "Epoch 208/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8958 - mae: 0.9962 - val_loss: 11.7375 - val_mae: 2.3721\n",
            "Epoch 209/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7855 - mae: 0.9617 - val_loss: 10.7209 - val_mae: 2.3380\n",
            "Epoch 210/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7804 - mae: 0.9873 - val_loss: 11.7657 - val_mae: 2.3743\n",
            "Epoch 211/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7755 - mae: 0.9667 - val_loss: 11.6531 - val_mae: 2.4243\n",
            "Epoch 212/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7332 - mae: 1.0028 - val_loss: 11.7127 - val_mae: 2.3509\n",
            "Epoch 213/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8413 - mae: 0.9773 - val_loss: 10.5508 - val_mae: 2.3034\n",
            "Epoch 214/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8688 - mae: 0.9814 - val_loss: 12.1792 - val_mae: 2.4577\n",
            "Epoch 215/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7652 - mae: 0.9945 - val_loss: 12.0634 - val_mae: 2.3475\n",
            "Epoch 216/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6778 - mae: 0.9528 - val_loss: 12.6256 - val_mae: 2.5366\n",
            "Epoch 217/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0630 - mae: 0.9955 - val_loss: 11.1947 - val_mae: 2.3012\n",
            "Epoch 218/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9198 - mae: 1.0322 - val_loss: 12.2324 - val_mae: 2.4235\n",
            "Epoch 219/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7675 - mae: 0.9826 - val_loss: 11.8656 - val_mae: 2.3761\n",
            "Epoch 220/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9232 - mae: 0.9915 - val_loss: 11.0952 - val_mae: 2.2888\n",
            "Epoch 221/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7522 - mae: 0.9687 - val_loss: 11.7943 - val_mae: 2.3169\n",
            "Epoch 222/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7648 - mae: 0.9835 - val_loss: 11.7554 - val_mae: 2.4490\n",
            "Epoch 223/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7916 - mae: 0.9488 - val_loss: 11.7877 - val_mae: 2.4430\n",
            "Epoch 224/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6424 - mae: 0.9372 - val_loss: 12.0674 - val_mae: 2.3292\n",
            "Epoch 225/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5985 - mae: 0.9196 - val_loss: 11.9887 - val_mae: 2.4159\n",
            "Epoch 226/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7158 - mae: 0.9260 - val_loss: 12.6190 - val_mae: 2.3855\n",
            "Epoch 227/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7418 - mae: 0.9444 - val_loss: 12.3616 - val_mae: 2.4281\n",
            "Epoch 228/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5566 - mae: 0.9281 - val_loss: 12.5882 - val_mae: 2.5032\n",
            "Epoch 229/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7446 - mae: 0.9291 - val_loss: 11.6343 - val_mae: 2.4412\n",
            "Epoch 230/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7381 - mae: 0.9645 - val_loss: 12.2883 - val_mae: 2.4475\n",
            "Epoch 231/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6043 - mae: 0.9107 - val_loss: 11.8212 - val_mae: 2.4299\n",
            "Epoch 232/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6135 - mae: 0.9108 - val_loss: 11.9986 - val_mae: 2.4732\n",
            "Epoch 233/500\n",
            "303/303 [==============================] - 0s 2ms/step - loss: 1.6978 - mae: 0.9336 - val_loss: 11.6984 - val_mae: 2.3226\n",
            "Epoch 234/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5866 - mae: 0.9745 - val_loss: 12.1086 - val_mae: 2.3789\n",
            "Epoch 235/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5901 - mae: 0.9110 - val_loss: 11.2158 - val_mae: 2.4074\n",
            "Epoch 236/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5527 - mae: 0.9178 - val_loss: 11.6811 - val_mae: 2.3457\n",
            "Epoch 237/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5517 - mae: 0.8959 - val_loss: 11.6019 - val_mae: 2.3553\n",
            "Epoch 238/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6470 - mae: 0.9640 - val_loss: 10.4769 - val_mae: 2.2643\n",
            "Epoch 239/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4443 - mae: 0.8880 - val_loss: 11.7410 - val_mae: 2.3368\n",
            "Epoch 240/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4495 - mae: 0.8792 - val_loss: 13.3359 - val_mae: 2.4280\n",
            "Epoch 241/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6340 - mae: 0.9480 - val_loss: 13.4943 - val_mae: 2.6790\n",
            "Epoch 242/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4912 - mae: 0.8974 - val_loss: 11.7111 - val_mae: 2.2813\n",
            "Epoch 243/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6857 - mae: 0.8881 - val_loss: 12.6851 - val_mae: 2.4328\n",
            "Epoch 244/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4267 - mae: 0.8643 - val_loss: 11.9130 - val_mae: 2.4139\n",
            "Epoch 245/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5150 - mae: 0.8964 - val_loss: 12.5667 - val_mae: 2.4084\n",
            "Epoch 246/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4602 - mae: 0.8895 - val_loss: 12.5671 - val_mae: 2.4874\n",
            "Epoch 247/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5545 - mae: 0.8989 - val_loss: 14.4585 - val_mae: 2.4739\n",
            "Epoch 248/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4034 - mae: 0.8773 - val_loss: 12.7948 - val_mae: 2.5290\n",
            "Epoch 249/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4467 - mae: 0.8693 - val_loss: 11.7416 - val_mae: 2.3664\n",
            "Epoch 250/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3536 - mae: 0.8605 - val_loss: 12.8405 - val_mae: 2.4850\n",
            "Epoch 251/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5270 - mae: 0.9097 - val_loss: 13.3903 - val_mae: 2.5089\n",
            "Epoch 252/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6822 - mae: 0.8974 - val_loss: 12.7922 - val_mae: 2.4760\n",
            "Epoch 253/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4816 - mae: 0.9148 - val_loss: 13.0343 - val_mae: 2.5569\n",
            "Epoch 254/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4399 - mae: 0.8783 - val_loss: 12.6739 - val_mae: 2.3483\n",
            "Epoch 255/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4474 - mae: 0.8885 - val_loss: 12.6078 - val_mae: 2.4486\n",
            "Epoch 256/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3554 - mae: 0.8541 - val_loss: 14.0501 - val_mae: 2.6679\n",
            "Epoch 257/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6350 - mae: 0.9206 - val_loss: 12.2919 - val_mae: 2.4399\n",
            "Epoch 258/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2693 - mae: 0.8612 - val_loss: 12.9697 - val_mae: 2.3318\n",
            "Epoch 259/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4543 - mae: 0.8801 - val_loss: 12.5055 - val_mae: 2.3832\n",
            "Epoch 260/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4381 - mae: 0.8694 - val_loss: 12.9087 - val_mae: 2.5020\n",
            "Epoch 261/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3766 - mae: 0.8554 - val_loss: 12.3858 - val_mae: 2.4007\n",
            "Epoch 262/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3646 - mae: 0.8646 - val_loss: 12.8076 - val_mae: 2.4123\n",
            "Epoch 263/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4474 - mae: 0.8597 - val_loss: 13.8481 - val_mae: 2.5504\n",
            "Epoch 264/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2967 - mae: 0.8479 - val_loss: 12.5309 - val_mae: 2.3389\n",
            "Epoch 265/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4392 - mae: 0.8838 - val_loss: 15.1838 - val_mae: 2.6197\n",
            "Epoch 266/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2856 - mae: 0.8470 - val_loss: 14.5142 - val_mae: 2.7000\n",
            "Epoch 267/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4048 - mae: 0.8615 - val_loss: 12.5309 - val_mae: 2.3866\n",
            "Epoch 268/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4106 - mae: 0.8550 - val_loss: 12.6493 - val_mae: 2.3843\n",
            "Epoch 269/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2377 - mae: 0.8204 - val_loss: 12.0853 - val_mae: 2.3684\n",
            "Epoch 270/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3701 - mae: 0.8523 - val_loss: 12.6461 - val_mae: 2.5198\n",
            "Epoch 271/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5164 - mae: 0.9014 - val_loss: 13.2532 - val_mae: 2.6322\n",
            "Epoch 272/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3008 - mae: 0.8580 - val_loss: 12.9058 - val_mae: 2.3875\n",
            "Epoch 273/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3216 - mae: 0.8112 - val_loss: 13.0737 - val_mae: 2.5152\n",
            "Epoch 274/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4232 - mae: 0.8819 - val_loss: 12.3441 - val_mae: 2.3937\n",
            "Epoch 275/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2921 - mae: 0.8021 - val_loss: 12.1540 - val_mae: 2.3470\n",
            "Epoch 276/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3750 - mae: 0.8542 - val_loss: 11.3182 - val_mae: 2.2506\n",
            "Epoch 277/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2616 - mae: 0.8019 - val_loss: 12.4964 - val_mae: 2.4407\n",
            "Epoch 278/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3483 - mae: 0.8764 - val_loss: 12.3184 - val_mae: 2.4111\n",
            "Epoch 279/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3912 - mae: 0.8580 - val_loss: 11.9540 - val_mae: 2.3427\n",
            "Epoch 280/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2701 - mae: 0.8164 - val_loss: 14.2092 - val_mae: 2.4909\n",
            "Epoch 281/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2696 - mae: 0.8090 - val_loss: 12.7743 - val_mae: 2.5023\n",
            "Epoch 282/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3502 - mae: 0.8306 - val_loss: 13.0428 - val_mae: 2.4488\n",
            "Epoch 283/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4588 - mae: 0.8805 - val_loss: 14.1038 - val_mae: 2.6033\n",
            "Epoch 284/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2606 - mae: 0.7995 - val_loss: 13.6894 - val_mae: 2.5586\n",
            "Epoch 285/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3269 - mae: 0.8358 - val_loss: 13.5319 - val_mae: 2.5423\n",
            "Epoch 286/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2078 - mae: 0.8155 - val_loss: 15.4691 - val_mae: 2.6199\n",
            "Epoch 287/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2546 - mae: 0.8162 - val_loss: 14.2204 - val_mae: 2.5559\n",
            "Epoch 288/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2268 - mae: 0.7842 - val_loss: 11.9582 - val_mae: 2.3776\n",
            "Epoch 289/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2511 - mae: 0.8296 - val_loss: 12.0409 - val_mae: 2.2998\n",
            "Epoch 290/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1444 - mae: 0.7930 - val_loss: 15.7219 - val_mae: 2.8807\n",
            "Epoch 291/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3749 - mae: 0.8514 - val_loss: 12.8118 - val_mae: 2.4467\n",
            "Epoch 292/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3730 - mae: 0.8723 - val_loss: 12.6124 - val_mae: 2.4539\n",
            "Epoch 293/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2047 - mae: 0.8082 - val_loss: 12.7879 - val_mae: 2.4830\n",
            "Epoch 294/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3363 - mae: 0.8454 - val_loss: 13.5917 - val_mae: 2.6929\n",
            "Epoch 295/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4226 - mae: 0.8518 - val_loss: 12.6858 - val_mae: 2.4555\n",
            "Epoch 296/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3015 - mae: 0.8626 - val_loss: 12.6945 - val_mae: 2.4287\n",
            "Epoch 297/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1126 - mae: 0.7775 - val_loss: 12.5049 - val_mae: 2.4889\n",
            "Epoch 298/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4020 - mae: 0.8404 - val_loss: 13.4456 - val_mae: 2.5505\n",
            "Epoch 299/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2586 - mae: 0.8107 - val_loss: 13.7988 - val_mae: 2.4136\n",
            "Epoch 300/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1674 - mae: 0.8049 - val_loss: 13.6833 - val_mae: 2.6180\n",
            "Epoch 301/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3672 - mae: 0.8594 - val_loss: 12.9748 - val_mae: 2.4849\n",
            "Epoch 302/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1364 - mae: 0.7451 - val_loss: 14.6058 - val_mae: 2.6769\n",
            "Epoch 303/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2365 - mae: 0.8007 - val_loss: 13.4047 - val_mae: 2.4528\n",
            "Epoch 304/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2575 - mae: 0.8024 - val_loss: 13.3146 - val_mae: 2.5069\n",
            "Epoch 305/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0981 - mae: 0.7918 - val_loss: 14.6072 - val_mae: 2.7814\n",
            "Epoch 306/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1511 - mae: 0.7951 - val_loss: 11.8595 - val_mae: 2.3525\n",
            "Epoch 307/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2314 - mae: 0.8084 - val_loss: 13.1448 - val_mae: 2.4888\n",
            "Epoch 308/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1151 - mae: 0.7712 - val_loss: 12.7996 - val_mae: 2.4289\n",
            "Epoch 309/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2020 - mae: 0.8084 - val_loss: 13.6502 - val_mae: 2.5146\n",
            "Epoch 310/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1092 - mae: 0.7835 - val_loss: 13.4438 - val_mae: 2.5031\n",
            "Epoch 311/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1161 - mae: 0.7580 - val_loss: 15.2105 - val_mae: 2.6589\n",
            "Epoch 312/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2797 - mae: 0.8353 - val_loss: 13.4237 - val_mae: 2.4845\n",
            "Epoch 313/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1271 - mae: 0.7803 - val_loss: 14.1811 - val_mae: 2.6129\n",
            "Epoch 314/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2195 - mae: 0.7921 - val_loss: 13.8995 - val_mae: 2.4887\n",
            "Epoch 315/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1457 - mae: 0.7938 - val_loss: 13.3636 - val_mae: 2.5175\n",
            "Epoch 316/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2591 - mae: 0.8123 - val_loss: 13.6552 - val_mae: 2.4352\n",
            "Epoch 317/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2619 - mae: 0.8232 - val_loss: 12.5144 - val_mae: 2.3308\n",
            "Epoch 318/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1015 - mae: 0.7285 - val_loss: 13.1948 - val_mae: 2.4220\n",
            "Epoch 319/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2553 - mae: 0.7891 - val_loss: 12.7240 - val_mae: 2.4925\n",
            "Epoch 320/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0589 - mae: 0.7546 - val_loss: 12.8962 - val_mae: 2.5139\n",
            "Epoch 321/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1743 - mae: 0.7891 - val_loss: 13.6147 - val_mae: 2.4156\n",
            "Epoch 322/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0088 - mae: 0.7249 - val_loss: 12.5054 - val_mae: 2.3810\n",
            "Epoch 323/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2219 - mae: 0.8273 - val_loss: 13.1170 - val_mae: 2.4616\n",
            "Epoch 324/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3971 - mae: 0.8052 - val_loss: 13.3338 - val_mae: 2.5019\n",
            "Epoch 325/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0729 - mae: 0.7716 - val_loss: 13.6109 - val_mae: 2.4918\n",
            "Epoch 326/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1714 - mae: 0.7863 - val_loss: 13.4435 - val_mae: 2.5034\n",
            "Epoch 327/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2957 - mae: 0.8246 - val_loss: 13.4203 - val_mae: 2.5485\n",
            "Epoch 328/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9733 - mae: 0.7101 - val_loss: 13.5361 - val_mae: 2.4131\n",
            "Epoch 329/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1543 - mae: 0.7846 - val_loss: 14.2476 - val_mae: 2.5919\n",
            "Epoch 330/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1777 - mae: 0.7497 - val_loss: 14.0319 - val_mae: 2.6231\n",
            "Epoch 331/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2072 - mae: 0.7883 - val_loss: 15.4286 - val_mae: 2.7734\n",
            "Epoch 332/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0827 - mae: 0.7605 - val_loss: 14.3178 - val_mae: 2.5031\n",
            "Epoch 333/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1913 - mae: 0.7994 - val_loss: 13.3018 - val_mae: 2.3908\n",
            "Epoch 334/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9307 - mae: 0.7254 - val_loss: 14.2272 - val_mae: 2.5136\n",
            "Epoch 335/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2135 - mae: 0.8038 - val_loss: 13.2113 - val_mae: 2.5619\n",
            "Epoch 336/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0228 - mae: 0.7378 - val_loss: 14.9227 - val_mae: 2.5917\n",
            "Epoch 337/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0300 - mae: 0.7486 - val_loss: 15.1922 - val_mae: 2.7431\n",
            "Epoch 338/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1197 - mae: 0.7701 - val_loss: 13.5942 - val_mae: 2.4487\n",
            "Epoch 339/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1276 - mae: 0.8083 - val_loss: 14.6865 - val_mae: 2.5099\n",
            "Epoch 340/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1182 - mae: 0.7519 - val_loss: 15.1576 - val_mae: 2.6920\n",
            "Epoch 341/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1580 - mae: 0.7735 - val_loss: 15.5537 - val_mae: 2.6541\n",
            "Epoch 342/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0520 - mae: 0.7512 - val_loss: 13.6967 - val_mae: 2.5514\n",
            "Epoch 343/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0568 - mae: 0.7596 - val_loss: 13.6241 - val_mae: 2.4663\n",
            "Epoch 344/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0903 - mae: 0.7439 - val_loss: 13.8101 - val_mae: 2.5634\n",
            "Epoch 345/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0093 - mae: 0.7394 - val_loss: 14.0766 - val_mae: 2.4622\n",
            "Epoch 346/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0815 - mae: 0.7454 - val_loss: 14.2653 - val_mae: 2.5612\n",
            "Epoch 347/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0900 - mae: 0.7526 - val_loss: 13.5664 - val_mae: 2.4418\n",
            "Epoch 348/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9753 - mae: 0.7283 - val_loss: 14.5789 - val_mae: 2.5920\n",
            "Epoch 349/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0041 - mae: 0.7171 - val_loss: 13.7643 - val_mae: 2.6435\n",
            "Epoch 350/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0644 - mae: 0.7701 - val_loss: 13.8643 - val_mae: 2.6378\n",
            "Epoch 351/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9852 - mae: 0.7158 - val_loss: 13.5366 - val_mae: 2.5576\n",
            "Epoch 352/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0027 - mae: 0.7555 - val_loss: 13.3954 - val_mae: 2.4518\n",
            "Epoch 353/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0660 - mae: 0.7461 - val_loss: 14.3624 - val_mae: 2.5562\n",
            "Epoch 354/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0586 - mae: 0.7717 - val_loss: 14.1065 - val_mae: 2.6163\n",
            "Epoch 355/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0439 - mae: 0.7382 - val_loss: 13.2442 - val_mae: 2.4727\n",
            "Epoch 356/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0488 - mae: 0.7616 - val_loss: 15.7434 - val_mae: 2.7459\n",
            "Epoch 357/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9980 - mae: 0.7542 - val_loss: 13.0362 - val_mae: 2.4762\n",
            "Epoch 358/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9434 - mae: 0.7136 - val_loss: 14.1181 - val_mae: 2.5351\n",
            "Epoch 359/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9572 - mae: 0.7258 - val_loss: 14.7594 - val_mae: 2.7077\n",
            "Epoch 360/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0936 - mae: 0.7304 - val_loss: 14.8162 - val_mae: 2.7049\n",
            "Epoch 361/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0269 - mae: 0.7444 - val_loss: 14.7119 - val_mae: 2.5613\n",
            "Epoch 362/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0757 - mae: 0.7402 - val_loss: 14.3991 - val_mae: 2.4838\n",
            "Epoch 363/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0618 - mae: 0.7575 - val_loss: 14.6796 - val_mae: 2.4881\n",
            "Epoch 364/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1003 - mae: 0.7519 - val_loss: 13.6583 - val_mae: 2.4192\n",
            "Epoch 365/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9378 - mae: 0.7154 - val_loss: 15.0792 - val_mae: 2.5776\n",
            "Epoch 366/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8745 - mae: 0.6986 - val_loss: 15.7203 - val_mae: 2.8314\n",
            "Epoch 367/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0374 - mae: 0.7489 - val_loss: 13.8551 - val_mae: 2.5392\n",
            "Epoch 368/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9951 - mae: 0.7265 - val_loss: 14.2017 - val_mae: 2.5958\n",
            "Epoch 369/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0621 - mae: 0.7549 - val_loss: 14.2402 - val_mae: 2.5453\n",
            "Epoch 370/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0765 - mae: 0.7239 - val_loss: 14.6641 - val_mae: 2.6637\n",
            "Epoch 371/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8923 - mae: 0.7097 - val_loss: 13.6742 - val_mae: 2.5175\n",
            "Epoch 372/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0514 - mae: 0.7346 - val_loss: 13.9718 - val_mae: 2.5259\n",
            "Epoch 373/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9738 - mae: 0.7231 - val_loss: 15.6837 - val_mae: 2.6958\n",
            "Epoch 374/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9516 - mae: 0.7267 - val_loss: 13.4299 - val_mae: 2.5934\n",
            "Epoch 375/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8862 - mae: 0.7024 - val_loss: 15.2069 - val_mae: 2.6631\n",
            "Epoch 376/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0096 - mae: 0.7407 - val_loss: 13.9537 - val_mae: 2.6051\n",
            "Epoch 377/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9234 - mae: 0.6940 - val_loss: 14.0757 - val_mae: 2.5363\n",
            "Epoch 378/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0449 - mae: 0.7368 - val_loss: 15.1129 - val_mae: 2.7257\n",
            "Epoch 379/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0239 - mae: 0.7167 - val_loss: 14.2985 - val_mae: 2.5497\n",
            "Epoch 380/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9711 - mae: 0.6811 - val_loss: 14.4801 - val_mae: 2.6423\n",
            "Epoch 381/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9254 - mae: 0.7087 - val_loss: 14.0196 - val_mae: 2.5666\n",
            "Epoch 382/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9680 - mae: 0.7354 - val_loss: 13.4086 - val_mae: 2.5220\n",
            "Epoch 383/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0311 - mae: 0.7354 - val_loss: 15.2966 - val_mae: 2.5907\n",
            "Epoch 384/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9021 - mae: 0.7056 - val_loss: 13.8567 - val_mae: 2.4753\n",
            "Epoch 385/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8748 - mae: 0.6837 - val_loss: 13.8130 - val_mae: 2.6628\n",
            "Epoch 386/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8993 - mae: 0.7006 - val_loss: 14.7415 - val_mae: 2.5907\n",
            "Epoch 387/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0565 - mae: 0.7421 - val_loss: 13.9140 - val_mae: 2.5427\n",
            "Epoch 388/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9426 - mae: 0.7092 - val_loss: 13.4602 - val_mae: 2.5380\n",
            "Epoch 389/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9595 - mae: 0.7133 - val_loss: 14.0853 - val_mae: 2.5680\n",
            "Epoch 390/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0123 - mae: 0.7212 - val_loss: 14.1748 - val_mae: 2.6852\n",
            "Epoch 391/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9153 - mae: 0.7223 - val_loss: 13.0485 - val_mae: 2.5383\n",
            "Epoch 392/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0095 - mae: 0.7021 - val_loss: 14.7885 - val_mae: 2.6860\n",
            "Epoch 393/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0394 - mae: 0.7096 - val_loss: 13.7640 - val_mae: 2.5021\n",
            "Epoch 394/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9590 - mae: 0.7108 - val_loss: 13.8423 - val_mae: 2.6815\n",
            "Epoch 395/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9720 - mae: 0.7231 - val_loss: 14.4359 - val_mae: 2.6336\n",
            "Epoch 396/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9473 - mae: 0.6994 - val_loss: 13.5617 - val_mae: 2.4628\n",
            "Epoch 397/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8489 - mae: 0.6899 - val_loss: 14.5976 - val_mae: 2.7312\n",
            "Epoch 398/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9921 - mae: 0.7273 - val_loss: 14.5377 - val_mae: 2.6120\n",
            "Epoch 399/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9461 - mae: 0.7203 - val_loss: 14.8906 - val_mae: 2.6999\n",
            "Epoch 400/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0926 - mae: 0.7419 - val_loss: 14.5176 - val_mae: 2.5128\n",
            "Epoch 401/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9988 - mae: 0.7417 - val_loss: 13.5360 - val_mae: 2.5044\n",
            "Epoch 402/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9806 - mae: 0.7215 - val_loss: 14.9920 - val_mae: 2.6659\n",
            "Epoch 403/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8717 - mae: 0.7096 - val_loss: 14.3010 - val_mae: 2.5562\n",
            "Epoch 404/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0721 - mae: 0.7173 - val_loss: 14.5116 - val_mae: 2.5225\n",
            "Epoch 405/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0020 - mae: 0.6870 - val_loss: 14.8001 - val_mae: 2.6762\n",
            "Epoch 406/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9529 - mae: 0.6887 - val_loss: 15.2807 - val_mae: 2.6796\n",
            "Epoch 407/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9580 - mae: 0.7334 - val_loss: 14.7213 - val_mae: 2.5702\n",
            "Epoch 408/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9062 - mae: 0.6964 - val_loss: 13.7674 - val_mae: 2.4888\n",
            "Epoch 409/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9205 - mae: 0.7050 - val_loss: 14.1935 - val_mae: 2.5799\n",
            "Epoch 410/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9307 - mae: 0.7033 - val_loss: 14.4046 - val_mae: 2.5405\n",
            "Epoch 411/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8518 - mae: 0.6910 - val_loss: 13.8393 - val_mae: 2.6588\n",
            "Epoch 412/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8814 - mae: 0.7182 - val_loss: 14.3061 - val_mae: 2.5732\n",
            "Epoch 413/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9631 - mae: 0.7305 - val_loss: 13.8181 - val_mae: 2.6807\n",
            "Epoch 414/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9790 - mae: 0.7128 - val_loss: 14.0413 - val_mae: 2.6087\n",
            "Epoch 415/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9156 - mae: 0.6893 - val_loss: 12.6425 - val_mae: 2.4625\n",
            "Epoch 416/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9974 - mae: 0.7164 - val_loss: 14.6791 - val_mae: 2.6284\n",
            "Epoch 417/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8621 - mae: 0.6791 - val_loss: 13.6465 - val_mae: 2.4587\n",
            "Epoch 418/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8119 - mae: 0.6770 - val_loss: 14.3426 - val_mae: 2.7748\n",
            "Epoch 419/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8514 - mae: 0.6805 - val_loss: 13.9137 - val_mae: 2.6099\n",
            "Epoch 420/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9885 - mae: 0.6900 - val_loss: 14.0227 - val_mae: 2.6272\n",
            "Epoch 421/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8845 - mae: 0.7084 - val_loss: 13.8099 - val_mae: 2.4855\n",
            "Epoch 422/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8271 - mae: 0.6606 - val_loss: 15.0161 - val_mae: 2.6812\n",
            "Epoch 423/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8514 - mae: 0.6775 - val_loss: 15.0998 - val_mae: 2.5696\n",
            "Epoch 424/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8222 - mae: 0.6665 - val_loss: 14.4607 - val_mae: 2.5486\n",
            "Epoch 425/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8891 - mae: 0.7103 - val_loss: 14.9561 - val_mae: 2.7609\n",
            "Epoch 426/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9314 - mae: 0.6708 - val_loss: 14.8182 - val_mae: 2.6205\n",
            "Epoch 427/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7585 - mae: 0.6541 - val_loss: 14.1110 - val_mae: 2.6604\n",
            "Epoch 428/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9526 - mae: 0.6995 - val_loss: 14.2566 - val_mae: 2.5880\n",
            "Epoch 429/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9490 - mae: 0.7056 - val_loss: 14.1670 - val_mae: 2.5950\n",
            "Epoch 430/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8936 - mae: 0.6934 - val_loss: 14.6074 - val_mae: 2.5625\n",
            "Epoch 431/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8512 - mae: 0.6971 - val_loss: 15.2620 - val_mae: 2.7460\n",
            "Epoch 432/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8756 - mae: 0.6965 - val_loss: 13.5262 - val_mae: 2.5582\n",
            "Epoch 433/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8246 - mae: 0.6765 - val_loss: 14.0393 - val_mae: 2.6389\n",
            "Epoch 434/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8890 - mae: 0.7005 - val_loss: 13.4213 - val_mae: 2.6091\n",
            "Epoch 435/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8759 - mae: 0.6880 - val_loss: 15.3864 - val_mae: 2.7689\n",
            "Epoch 436/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8440 - mae: 0.6446 - val_loss: 14.5556 - val_mae: 2.6462\n",
            "Epoch 437/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7877 - mae: 0.6361 - val_loss: 14.4999 - val_mae: 2.5678\n",
            "Epoch 438/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7866 - mae: 0.6602 - val_loss: 14.2952 - val_mae: 2.6225\n",
            "Epoch 439/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9706 - mae: 0.7071 - val_loss: 13.6261 - val_mae: 2.5808\n",
            "Epoch 440/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8517 - mae: 0.6834 - val_loss: 16.2054 - val_mae: 2.8398\n",
            "Epoch 441/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8235 - mae: 0.6461 - val_loss: 14.7756 - val_mae: 2.6246\n",
            "Epoch 442/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8874 - mae: 0.6784 - val_loss: 14.2939 - val_mae: 2.7538\n",
            "Epoch 443/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8201 - mae: 0.6672 - val_loss: 15.4245 - val_mae: 2.7079\n",
            "Epoch 444/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8412 - mae: 0.6644 - val_loss: 13.9838 - val_mae: 2.6162\n",
            "Epoch 445/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8602 - mae: 0.6704 - val_loss: 14.7845 - val_mae: 2.7271\n",
            "Epoch 446/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0134 - mae: 0.6755 - val_loss: 14.0566 - val_mae: 2.6678\n",
            "Epoch 447/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9230 - mae: 0.6836 - val_loss: 14.5606 - val_mae: 2.7801\n",
            "Epoch 448/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8040 - mae: 0.6536 - val_loss: 12.8201 - val_mae: 2.5632\n",
            "Epoch 449/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7841 - mae: 0.6545 - val_loss: 16.2472 - val_mae: 2.8234\n",
            "Epoch 450/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8245 - mae: 0.6729 - val_loss: 14.6277 - val_mae: 2.7298\n",
            "Epoch 451/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9333 - mae: 0.7151 - val_loss: 13.5383 - val_mae: 2.5625\n",
            "Epoch 452/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7881 - mae: 0.6469 - val_loss: 16.7237 - val_mae: 2.8964\n",
            "Epoch 453/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7798 - mae: 0.6708 - val_loss: 13.8328 - val_mae: 2.7453\n",
            "Epoch 454/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8972 - mae: 0.6862 - val_loss: 14.3767 - val_mae: 2.6834\n",
            "Epoch 455/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8748 - mae: 0.6678 - val_loss: 15.3393 - val_mae: 2.7481\n",
            "Epoch 456/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8079 - mae: 0.6820 - val_loss: 14.4961 - val_mae: 2.6576\n",
            "Epoch 457/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8079 - mae: 0.6475 - val_loss: 15.1931 - val_mae: 2.7392\n",
            "Epoch 458/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8161 - mae: 0.6588 - val_loss: 14.5283 - val_mae: 2.7664\n",
            "Epoch 459/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0306 - mae: 0.6911 - val_loss: 14.3844 - val_mae: 2.7200\n",
            "Epoch 460/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8793 - mae: 0.6493 - val_loss: 14.4880 - val_mae: 2.6409\n",
            "Epoch 461/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8686 - mae: 0.6636 - val_loss: 14.2547 - val_mae: 2.6903\n",
            "Epoch 462/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8259 - mae: 0.6563 - val_loss: 14.2177 - val_mae: 2.5988\n",
            "Epoch 463/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7815 - mae: 0.6328 - val_loss: 16.3095 - val_mae: 2.6891\n",
            "Epoch 464/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9846 - mae: 0.6867 - val_loss: 14.2660 - val_mae: 2.5893\n",
            "Epoch 465/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7819 - mae: 0.6145 - val_loss: 14.3624 - val_mae: 2.6475\n",
            "Epoch 466/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8774 - mae: 0.6840 - val_loss: 13.8541 - val_mae: 2.6181\n",
            "Epoch 467/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8206 - mae: 0.6622 - val_loss: 14.8865 - val_mae: 2.7806\n",
            "Epoch 468/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9225 - mae: 0.6617 - val_loss: 14.8258 - val_mae: 2.6389\n",
            "Epoch 469/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7923 - mae: 0.6338 - val_loss: 15.1528 - val_mae: 2.8416\n",
            "Epoch 470/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7946 - mae: 0.6361 - val_loss: 15.1742 - val_mae: 2.7224\n",
            "Epoch 471/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7837 - mae: 0.6491 - val_loss: 14.0906 - val_mae: 2.6325\n",
            "Epoch 472/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7420 - mae: 0.6532 - val_loss: 15.5462 - val_mae: 2.9447\n",
            "Epoch 473/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8710 - mae: 0.6705 - val_loss: 13.9511 - val_mae: 2.6374\n",
            "Epoch 474/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9869 - mae: 0.6791 - val_loss: 13.9053 - val_mae: 2.6007\n",
            "Epoch 475/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7014 - mae: 0.6221 - val_loss: 14.0911 - val_mae: 2.5974\n",
            "Epoch 476/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7411 - mae: 0.6331 - val_loss: 16.1560 - val_mae: 2.8232\n",
            "Epoch 477/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8188 - mae: 0.6750 - val_loss: 14.1603 - val_mae: 2.7906\n",
            "Epoch 478/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7728 - mae: 0.6411 - val_loss: 14.1175 - val_mae: 2.6521\n",
            "Epoch 479/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8509 - mae: 0.6648 - val_loss: 14.2448 - val_mae: 2.5424\n",
            "Epoch 480/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7484 - mae: 0.6316 - val_loss: 14.3068 - val_mae: 2.7523\n",
            "Epoch 481/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7271 - mae: 0.6212 - val_loss: 13.9934 - val_mae: 2.5651\n",
            "Epoch 482/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7908 - mae: 0.6517 - val_loss: 13.8467 - val_mae: 2.6489\n",
            "Epoch 483/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8280 - mae: 0.6602 - val_loss: 13.6185 - val_mae: 2.6270\n",
            "Epoch 484/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7522 - mae: 0.6352 - val_loss: 13.1858 - val_mae: 2.6350\n",
            "Epoch 485/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9295 - mae: 0.6807 - val_loss: 13.2227 - val_mae: 2.6354\n",
            "Epoch 486/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8185 - mae: 0.6517 - val_loss: 14.1520 - val_mae: 2.6380\n",
            "Epoch 487/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8116 - mae: 0.6628 - val_loss: 13.8160 - val_mae: 2.6004\n",
            "Epoch 488/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7519 - mae: 0.6322 - val_loss: 13.9834 - val_mae: 2.6416\n",
            "Epoch 489/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8887 - mae: 0.6731 - val_loss: 14.1938 - val_mae: 2.6719\n",
            "Epoch 490/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7351 - mae: 0.6164 - val_loss: 14.9539 - val_mae: 2.6588\n",
            "Epoch 491/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7533 - mae: 0.6537 - val_loss: 14.5651 - val_mae: 2.6639\n",
            "Epoch 492/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8479 - mae: 0.6684 - val_loss: 15.7259 - val_mae: 2.7621\n",
            "Epoch 493/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7684 - mae: 0.6086 - val_loss: 14.2383 - val_mae: 2.7284\n",
            "Epoch 494/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8691 - mae: 0.6806 - val_loss: 13.7644 - val_mae: 2.6548\n",
            "Epoch 495/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8494 - mae: 0.6380 - val_loss: 15.0103 - val_mae: 2.7160\n",
            "Epoch 496/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7517 - mae: 0.6262 - val_loss: 13.8485 - val_mae: 2.6816\n",
            "Epoch 497/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7960 - mae: 0.6504 - val_loss: 15.7052 - val_mae: 2.7323\n",
            "Epoch 498/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8052 - mae: 0.6527 - val_loss: 14.5833 - val_mae: 2.6951\n",
            "Epoch 499/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7917 - mae: 0.6430 - val_loss: 14.0980 - val_mae: 2.6540\n",
            "Epoch 500/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8459 - mae: 0.6605 - val_loss: 14.0586 - val_mae: 2.6338\n",
            "processing fold # 1\n",
            "Epoch 1/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 220.7412 - mae: 11.1328 - val_loss: 32.0854 - val_mae: 4.3085\n",
            "Epoch 2/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 32.9278 - mae: 3.7948 - val_loss: 21.0124 - val_mae: 3.5553\n",
            "Epoch 3/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 23.1247 - mae: 3.0540 - val_loss: 15.3123 - val_mae: 2.8747\n",
            "Epoch 4/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 17.7679 - mae: 2.6716 - val_loss: 12.8700 - val_mae: 2.6739\n",
            "Epoch 5/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 15.6908 - mae: 2.4994 - val_loss: 12.1587 - val_mae: 2.6042\n",
            "Epoch 6/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 14.0212 - mae: 2.3481 - val_loss: 11.4282 - val_mae: 2.5758\n",
            "Epoch 7/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 13.3129 - mae: 2.2775 - val_loss: 11.9376 - val_mae: 2.6273\n",
            "Epoch 8/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 12.3567 - mae: 2.1965 - val_loss: 12.6859 - val_mae: 2.7281\n",
            "Epoch 9/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 11.5994 - mae: 2.2030 - val_loss: 10.8519 - val_mae: 2.5081\n",
            "Epoch 10/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 11.1520 - mae: 2.1669 - val_loss: 11.4899 - val_mae: 2.5877\n",
            "Epoch 11/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 10.9368 - mae: 2.1307 - val_loss: 11.7985 - val_mae: 2.6038\n",
            "Epoch 12/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 10.6377 - mae: 2.1317 - val_loss: 10.5126 - val_mae: 2.4784\n",
            "Epoch 13/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 9.9722 - mae: 1.9893 - val_loss: 13.3480 - val_mae: 2.8359\n",
            "Epoch 14/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 9.9344 - mae: 2.0289 - val_loss: 10.5283 - val_mae: 2.4967\n",
            "Epoch 15/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 9.7281 - mae: 2.0229 - val_loss: 11.4833 - val_mae: 2.6040\n",
            "Epoch 16/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 9.5985 - mae: 2.0452 - val_loss: 10.0522 - val_mae: 2.4504\n",
            "Epoch 17/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 9.5367 - mae: 1.9533 - val_loss: 10.7878 - val_mae: 2.5308\n",
            "Epoch 18/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 9.6479 - mae: 1.9709 - val_loss: 9.4042 - val_mae: 2.3104\n",
            "Epoch 19/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 9.2810 - mae: 2.0122 - val_loss: 9.7056 - val_mae: 2.3843\n",
            "Epoch 20/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 9.1293 - mae: 1.9848 - val_loss: 9.9875 - val_mae: 2.4547\n",
            "Epoch 21/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 8.9566 - mae: 1.9757 - val_loss: 9.3291 - val_mae: 2.3377\n",
            "Epoch 22/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 8.8464 - mae: 1.8505 - val_loss: 9.2010 - val_mae: 2.3025\n",
            "Epoch 23/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 8.4098 - mae: 1.8911 - val_loss: 10.0589 - val_mae: 2.4687\n",
            "Epoch 24/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 8.9475 - mae: 1.8565 - val_loss: 10.9423 - val_mae: 2.5807\n",
            "Epoch 25/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 8.6110 - mae: 1.8767 - val_loss: 12.0930 - val_mae: 2.7425\n",
            "Epoch 26/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 8.3818 - mae: 1.8742 - val_loss: 9.9718 - val_mae: 2.4503\n",
            "Epoch 27/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 8.2958 - mae: 1.8990 - val_loss: 9.5768 - val_mae: 2.4182\n",
            "Epoch 28/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 8.3613 - mae: 1.8959 - val_loss: 9.7692 - val_mae: 2.4506\n",
            "Epoch 29/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.8954 - mae: 1.8570 - val_loss: 8.4357 - val_mae: 2.2563\n",
            "Epoch 30/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.8492 - mae: 1.7939 - val_loss: 8.9858 - val_mae: 2.3183\n",
            "Epoch 31/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 8.1462 - mae: 1.8069 - val_loss: 9.3546 - val_mae: 2.4173\n",
            "Epoch 32/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.8404 - mae: 1.7933 - val_loss: 10.4133 - val_mae: 2.5332\n",
            "Epoch 33/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.5467 - mae: 1.7971 - val_loss: 9.0063 - val_mae: 2.3551\n",
            "Epoch 34/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.5622 - mae: 1.8139 - val_loss: 8.7790 - val_mae: 2.2971\n",
            "Epoch 35/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.6340 - mae: 1.7873 - val_loss: 9.3573 - val_mae: 2.3703\n",
            "Epoch 36/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.6182 - mae: 1.7569 - val_loss: 9.9919 - val_mae: 2.4363\n",
            "Epoch 37/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.4753 - mae: 1.7417 - val_loss: 9.7654 - val_mae: 2.4539\n",
            "Epoch 38/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.3067 - mae: 1.7894 - val_loss: 10.0693 - val_mae: 2.5012\n",
            "Epoch 39/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.9279 - mae: 1.6916 - val_loss: 10.9206 - val_mae: 2.5831\n",
            "Epoch 40/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.0255 - mae: 1.7676 - val_loss: 8.8423 - val_mae: 2.3044\n",
            "Epoch 41/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.5503 - mae: 1.7232 - val_loss: 9.9824 - val_mae: 2.4835\n",
            "Epoch 42/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.8204 - mae: 1.7195 - val_loss: 9.8091 - val_mae: 2.4949\n",
            "Epoch 43/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.8079 - mae: 1.6719 - val_loss: 9.4669 - val_mae: 2.3456\n",
            "Epoch 44/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.8382 - mae: 1.6884 - val_loss: 9.2994 - val_mae: 2.3912\n",
            "Epoch 45/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.3580 - mae: 1.6229 - val_loss: 9.3518 - val_mae: 2.4280\n",
            "Epoch 46/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.8260 - mae: 1.6883 - val_loss: 9.9313 - val_mae: 2.4860\n",
            "Epoch 47/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.6052 - mae: 1.6183 - val_loss: 8.8415 - val_mae: 2.3297\n",
            "Epoch 48/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.8270 - mae: 1.6100 - val_loss: 11.9924 - val_mae: 2.7065\n",
            "Epoch 49/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.4713 - mae: 1.6776 - val_loss: 8.7011 - val_mae: 2.3368\n",
            "Epoch 50/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.4820 - mae: 1.6416 - val_loss: 9.3433 - val_mae: 2.4199\n",
            "Epoch 51/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.5352 - mae: 1.6177 - val_loss: 9.2986 - val_mae: 2.3350\n",
            "Epoch 52/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.3945 - mae: 1.6683 - val_loss: 9.7211 - val_mae: 2.4882\n",
            "Epoch 53/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.8119 - mae: 1.6527 - val_loss: 9.1148 - val_mae: 2.4227\n",
            "Epoch 54/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.8928 - mae: 1.5762 - val_loss: 7.8986 - val_mae: 2.2205\n",
            "Epoch 55/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.1244 - mae: 1.6197 - val_loss: 8.4496 - val_mae: 2.2845\n",
            "Epoch 56/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.5517 - mae: 1.5265 - val_loss: 9.3528 - val_mae: 2.4341\n",
            "Epoch 57/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.8401 - mae: 1.5817 - val_loss: 9.5544 - val_mae: 2.4539\n",
            "Epoch 58/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.7646 - mae: 1.5899 - val_loss: 11.9645 - val_mae: 2.8180\n",
            "Epoch 59/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.7496 - mae: 1.5721 - val_loss: 8.7599 - val_mae: 2.3251\n",
            "Epoch 60/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.4027 - mae: 1.4600 - val_loss: 12.7806 - val_mae: 2.8055\n",
            "Epoch 61/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.3316 - mae: 1.5120 - val_loss: 10.1255 - val_mae: 2.5406\n",
            "Epoch 62/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.7595 - mae: 1.5562 - val_loss: 13.1021 - val_mae: 2.8778\n",
            "Epoch 63/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.0638 - mae: 1.4934 - val_loss: 10.5602 - val_mae: 2.4848\n",
            "Epoch 64/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.1879 - mae: 1.5157 - val_loss: 10.6700 - val_mae: 2.4658\n",
            "Epoch 65/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.1797 - mae: 1.5006 - val_loss: 9.1175 - val_mae: 2.3194\n",
            "Epoch 66/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.9263 - mae: 1.4454 - val_loss: 10.3368 - val_mae: 2.4791\n",
            "Epoch 67/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.1851 - mae: 1.4707 - val_loss: 8.7203 - val_mae: 2.2624\n",
            "Epoch 68/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.8461 - mae: 1.4619 - val_loss: 11.2867 - val_mae: 2.6448\n",
            "Epoch 69/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.7084 - mae: 1.4722 - val_loss: 10.6940 - val_mae: 2.5792\n",
            "Epoch 70/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.5255 - mae: 1.4817 - val_loss: 12.9552 - val_mae: 2.9006\n",
            "Epoch 71/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.7772 - mae: 1.4564 - val_loss: 8.5914 - val_mae: 2.2124\n",
            "Epoch 72/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.3958 - mae: 1.4482 - val_loss: 8.9903 - val_mae: 2.2742\n",
            "Epoch 73/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.4470 - mae: 1.4399 - val_loss: 8.9331 - val_mae: 2.3315\n",
            "Epoch 74/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.8494 - mae: 1.4404 - val_loss: 10.2167 - val_mae: 2.4163\n",
            "Epoch 75/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.4840 - mae: 1.3206 - val_loss: 10.0234 - val_mae: 2.3941\n",
            "Epoch 76/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.2359 - mae: 1.4045 - val_loss: 10.1703 - val_mae: 2.4295\n",
            "Epoch 77/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.4906 - mae: 1.3488 - val_loss: 9.8640 - val_mae: 2.4820\n",
            "Epoch 78/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.3648 - mae: 1.4071 - val_loss: 10.1485 - val_mae: 2.4395\n",
            "Epoch 79/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.1876 - mae: 1.3658 - val_loss: 10.0647 - val_mae: 2.4454\n",
            "Epoch 80/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.5860 - mae: 1.3721 - val_loss: 10.8031 - val_mae: 2.5046\n",
            "Epoch 81/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.4101 - mae: 1.3915 - val_loss: 12.2220 - val_mae: 2.7081\n",
            "Epoch 82/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.4782 - mae: 1.3431 - val_loss: 9.3565 - val_mae: 2.2357\n",
            "Epoch 83/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.3946 - mae: 1.3553 - val_loss: 9.0315 - val_mae: 2.2137\n",
            "Epoch 84/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.9438 - mae: 1.3509 - val_loss: 9.5507 - val_mae: 2.2844\n",
            "Epoch 85/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.8513 - mae: 1.3443 - val_loss: 13.4089 - val_mae: 2.6578\n",
            "Epoch 86/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.9011 - mae: 1.3071 - val_loss: 10.0779 - val_mae: 2.3822\n",
            "Epoch 87/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.9311 - mae: 1.3619 - val_loss: 11.8488 - val_mae: 2.5319\n",
            "Epoch 88/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.1196 - mae: 1.3459 - val_loss: 10.1241 - val_mae: 2.3554\n",
            "Epoch 89/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.2497 - mae: 1.3056 - val_loss: 13.2142 - val_mae: 2.6285\n",
            "Epoch 90/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.1000 - mae: 1.3295 - val_loss: 11.6307 - val_mae: 2.5130\n",
            "Epoch 91/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.8462 - mae: 1.3062 - val_loss: 10.7356 - val_mae: 2.4403\n",
            "Epoch 92/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.8216 - mae: 1.2738 - val_loss: 10.1541 - val_mae: 2.3633\n",
            "Epoch 93/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.8343 - mae: 1.2806 - val_loss: 12.6704 - val_mae: 2.6454\n",
            "Epoch 94/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.6971 - mae: 1.2776 - val_loss: 9.6331 - val_mae: 2.2928\n",
            "Epoch 95/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.6872 - mae: 1.2660 - val_loss: 10.9100 - val_mae: 2.4192\n",
            "Epoch 96/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.5552 - mae: 1.2168 - val_loss: 11.3283 - val_mae: 2.5316\n",
            "Epoch 97/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.7332 - mae: 1.3231 - val_loss: 11.2400 - val_mae: 2.4434\n",
            "Epoch 98/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.6369 - mae: 1.2818 - val_loss: 12.3118 - val_mae: 2.5369\n",
            "Epoch 99/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.4309 - mae: 1.2542 - val_loss: 11.2394 - val_mae: 2.4510\n",
            "Epoch 100/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.5020 - mae: 1.2374 - val_loss: 11.3336 - val_mae: 2.3673\n",
            "Epoch 101/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.3717 - mae: 1.2549 - val_loss: 9.3581 - val_mae: 2.3203\n",
            "Epoch 102/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.7421 - mae: 1.2638 - val_loss: 9.6679 - val_mae: 2.3466\n",
            "Epoch 103/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.3822 - mae: 1.2486 - val_loss: 10.3806 - val_mae: 2.3664\n",
            "Epoch 104/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.3100 - mae: 1.2170 - val_loss: 9.5138 - val_mae: 2.2925\n",
            "Epoch 105/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.4891 - mae: 1.2200 - val_loss: 9.6580 - val_mae: 2.3117\n",
            "Epoch 106/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.5329 - mae: 1.2036 - val_loss: 12.3936 - val_mae: 2.6726\n",
            "Epoch 107/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.3566 - mae: 1.2760 - val_loss: 11.2496 - val_mae: 2.4079\n",
            "Epoch 108/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.3400 - mae: 1.2464 - val_loss: 12.1769 - val_mae: 2.5514\n",
            "Epoch 109/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.2374 - mae: 1.2038 - val_loss: 12.3898 - val_mae: 2.5438\n",
            "Epoch 110/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.3260 - mae: 1.2334 - val_loss: 10.6083 - val_mae: 2.4815\n",
            "Epoch 111/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.2061 - mae: 1.2108 - val_loss: 11.1610 - val_mae: 2.4521\n",
            "Epoch 112/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.2890 - mae: 1.2328 - val_loss: 11.1639 - val_mae: 2.4276\n",
            "Epoch 113/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.0013 - mae: 1.2213 - val_loss: 12.0009 - val_mae: 2.6295\n",
            "Epoch 114/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.2257 - mae: 1.2113 - val_loss: 8.9297 - val_mae: 2.2908\n",
            "Epoch 115/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.2486 - mae: 1.2562 - val_loss: 11.3151 - val_mae: 2.3938\n",
            "Epoch 116/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.0626 - mae: 1.1843 - val_loss: 12.1141 - val_mae: 2.4186\n",
            "Epoch 117/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.8260 - mae: 1.1253 - val_loss: 10.7727 - val_mae: 2.3154\n",
            "Epoch 118/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.9862 - mae: 1.1910 - val_loss: 10.8300 - val_mae: 2.3346\n",
            "Epoch 119/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.0040 - mae: 1.1629 - val_loss: 11.4419 - val_mae: 2.4528\n",
            "Epoch 120/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.9147 - mae: 1.1595 - val_loss: 11.4312 - val_mae: 2.3913\n",
            "Epoch 121/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.8879 - mae: 1.1706 - val_loss: 18.0538 - val_mae: 2.9869\n",
            "Epoch 122/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.7824 - mae: 1.1876 - val_loss: 12.6156 - val_mae: 2.4346\n",
            "Epoch 123/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.8153 - mae: 1.1569 - val_loss: 12.8019 - val_mae: 2.4748\n",
            "Epoch 124/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.8720 - mae: 1.1458 - val_loss: 10.8249 - val_mae: 2.3586\n",
            "Epoch 125/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.6624 - mae: 1.1692 - val_loss: 14.2722 - val_mae: 2.6609\n",
            "Epoch 126/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.8722 - mae: 1.1498 - val_loss: 12.9451 - val_mae: 2.5494\n",
            "Epoch 127/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.5519 - mae: 1.1127 - val_loss: 15.7678 - val_mae: 2.7440\n",
            "Epoch 128/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.6575 - mae: 1.1380 - val_loss: 14.7753 - val_mae: 2.5672\n",
            "Epoch 129/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.7754 - mae: 1.1765 - val_loss: 13.2057 - val_mae: 2.5966\n",
            "Epoch 130/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.9928 - mae: 1.1476 - val_loss: 12.4397 - val_mae: 2.3352\n",
            "Epoch 131/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.8437 - mae: 1.1463 - val_loss: 14.1500 - val_mae: 2.5627\n",
            "Epoch 132/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.6148 - mae: 1.1099 - val_loss: 10.9070 - val_mae: 2.3441\n",
            "Epoch 133/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.4927 - mae: 1.0831 - val_loss: 13.1754 - val_mae: 2.6729\n",
            "Epoch 134/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.6129 - mae: 1.1075 - val_loss: 13.5577 - val_mae: 2.3845\n",
            "Epoch 135/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2393 - mae: 1.0107 - val_loss: 14.5087 - val_mae: 2.6494\n",
            "Epoch 136/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.5755 - mae: 1.1562 - val_loss: 10.2272 - val_mae: 2.2627\n",
            "Epoch 137/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.5874 - mae: 1.1188 - val_loss: 17.5552 - val_mae: 2.8282\n",
            "Epoch 138/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2967 - mae: 1.0920 - val_loss: 16.5384 - val_mae: 2.8625\n",
            "Epoch 139/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.4165 - mae: 1.1473 - val_loss: 20.4417 - val_mae: 3.1897\n",
            "Epoch 140/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2658 - mae: 1.0811 - val_loss: 12.9690 - val_mae: 2.4702\n",
            "Epoch 141/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2747 - mae: 1.0873 - val_loss: 13.5156 - val_mae: 2.7754\n",
            "Epoch 142/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3914 - mae: 1.0902 - val_loss: 13.3872 - val_mae: 2.5417\n",
            "Epoch 143/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2412 - mae: 1.0602 - val_loss: 11.4897 - val_mae: 2.4186\n",
            "Epoch 144/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3456 - mae: 1.1023 - val_loss: 13.3403 - val_mae: 2.4669\n",
            "Epoch 145/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.5207 - mae: 1.1051 - val_loss: 12.3745 - val_mae: 2.6201\n",
            "Epoch 146/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3106 - mae: 1.0372 - val_loss: 13.1415 - val_mae: 2.5287\n",
            "Epoch 147/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2370 - mae: 1.0540 - val_loss: 14.7149 - val_mae: 2.6222\n",
            "Epoch 148/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3693 - mae: 1.0917 - val_loss: 13.5176 - val_mae: 2.5120\n",
            "Epoch 149/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1543 - mae: 1.0874 - val_loss: 13.8847 - val_mae: 2.5317\n",
            "Epoch 150/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9790 - mae: 1.0043 - val_loss: 17.9717 - val_mae: 2.7273\n",
            "Epoch 151/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2084 - mae: 1.0541 - val_loss: 12.8748 - val_mae: 2.4386\n",
            "Epoch 152/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2426 - mae: 1.0265 - val_loss: 18.7781 - val_mae: 2.9278\n",
            "Epoch 153/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1950 - mae: 1.0639 - val_loss: 14.9325 - val_mae: 2.7237\n",
            "Epoch 154/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1759 - mae: 1.0446 - val_loss: 16.4917 - val_mae: 2.6632\n",
            "Epoch 155/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1770 - mae: 1.0477 - val_loss: 13.4722 - val_mae: 2.5959\n",
            "Epoch 156/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9764 - mae: 1.0167 - val_loss: 15.6557 - val_mae: 2.5987\n",
            "Epoch 157/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2082 - mae: 1.0344 - val_loss: 10.1367 - val_mae: 2.2778\n",
            "Epoch 158/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9948 - mae: 0.9740 - val_loss: 16.6981 - val_mae: 2.6235\n",
            "Epoch 159/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8157 - mae: 1.0028 - val_loss: 14.1109 - val_mae: 2.6159\n",
            "Epoch 160/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9158 - mae: 1.0599 - val_loss: 12.6227 - val_mae: 2.3833\n",
            "Epoch 161/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9709 - mae: 1.0304 - val_loss: 14.9204 - val_mae: 2.5884\n",
            "Epoch 162/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8663 - mae: 1.0041 - val_loss: 12.4824 - val_mae: 2.3706\n",
            "Epoch 163/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9548 - mae: 0.9777 - val_loss: 18.6412 - val_mae: 2.9614\n",
            "Epoch 164/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1265 - mae: 1.0541 - val_loss: 14.9600 - val_mae: 2.5185\n",
            "Epoch 165/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9638 - mae: 0.9650 - val_loss: 14.5545 - val_mae: 2.6112\n",
            "Epoch 166/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7956 - mae: 0.9575 - val_loss: 14.3577 - val_mae: 2.6202\n",
            "Epoch 167/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0340 - mae: 1.0008 - val_loss: 16.0244 - val_mae: 2.7189\n",
            "Epoch 168/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9149 - mae: 0.9615 - val_loss: 14.2477 - val_mae: 2.5242\n",
            "Epoch 169/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9355 - mae: 1.0135 - val_loss: 19.0734 - val_mae: 3.0186\n",
            "Epoch 170/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8467 - mae: 0.9825 - val_loss: 12.1622 - val_mae: 2.3229\n",
            "Epoch 171/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9397 - mae: 1.0073 - val_loss: 17.5473 - val_mae: 3.0699\n",
            "Epoch 172/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9274 - mae: 0.9543 - val_loss: 16.3505 - val_mae: 2.6872\n",
            "Epoch 173/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7673 - mae: 0.9924 - val_loss: 15.5466 - val_mae: 2.5710\n",
            "Epoch 174/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6663 - mae: 0.9580 - val_loss: 13.9090 - val_mae: 2.5976\n",
            "Epoch 175/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8637 - mae: 0.9868 - val_loss: 14.9376 - val_mae: 2.6334\n",
            "Epoch 176/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8650 - mae: 0.9920 - val_loss: 14.3895 - val_mae: 2.5323\n",
            "Epoch 177/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8070 - mae: 0.9904 - val_loss: 16.6160 - val_mae: 2.6210\n",
            "Epoch 178/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6738 - mae: 0.9084 - val_loss: 15.1438 - val_mae: 2.8096\n",
            "Epoch 179/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8331 - mae: 0.9512 - val_loss: 12.5440 - val_mae: 2.4667\n",
            "Epoch 180/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8720 - mae: 0.9878 - val_loss: 15.2114 - val_mae: 2.6341\n",
            "Epoch 181/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8858 - mae: 1.0096 - val_loss: 17.1232 - val_mae: 2.8385\n",
            "Epoch 182/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6126 - mae: 0.9636 - val_loss: 13.6855 - val_mae: 2.5319\n",
            "Epoch 183/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7386 - mae: 0.9318 - val_loss: 16.0300 - val_mae: 2.6909\n",
            "Epoch 184/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7932 - mae: 0.9750 - val_loss: 17.0747 - val_mae: 2.8429\n",
            "Epoch 185/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8841 - mae: 0.9764 - val_loss: 12.4706 - val_mae: 2.5748\n",
            "Epoch 186/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7889 - mae: 0.9780 - val_loss: 18.1992 - val_mae: 2.8963\n",
            "Epoch 187/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6756 - mae: 0.9357 - val_loss: 16.0904 - val_mae: 2.7483\n",
            "Epoch 188/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5622 - mae: 0.9097 - val_loss: 16.2962 - val_mae: 2.8001\n",
            "Epoch 189/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5300 - mae: 0.9257 - val_loss: 14.1199 - val_mae: 2.6290\n",
            "Epoch 190/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6805 - mae: 0.9610 - val_loss: 16.0681 - val_mae: 2.6582\n",
            "Epoch 191/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6828 - mae: 0.9167 - val_loss: 17.7586 - val_mae: 2.6697\n",
            "Epoch 192/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6435 - mae: 0.9266 - val_loss: 16.9420 - val_mae: 2.5849\n",
            "Epoch 193/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5807 - mae: 0.9241 - val_loss: 18.0994 - val_mae: 2.8327\n",
            "Epoch 194/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6917 - mae: 0.9619 - val_loss: 16.8923 - val_mae: 2.6833\n",
            "Epoch 195/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5927 - mae: 0.9251 - val_loss: 12.9565 - val_mae: 2.5274\n",
            "Epoch 196/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7687 - mae: 0.9189 - val_loss: 15.6813 - val_mae: 2.5721\n",
            "Epoch 197/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5334 - mae: 0.9199 - val_loss: 15.7370 - val_mae: 2.6691\n",
            "Epoch 198/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5148 - mae: 0.9043 - val_loss: 15.2395 - val_mae: 2.6100\n",
            "Epoch 199/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5721 - mae: 0.9174 - val_loss: 19.6660 - val_mae: 3.1512\n",
            "Epoch 200/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5964 - mae: 0.9384 - val_loss: 16.5319 - val_mae: 2.5750\n",
            "Epoch 201/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5841 - mae: 0.9216 - val_loss: 16.4688 - val_mae: 2.7929\n",
            "Epoch 202/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4178 - mae: 0.9162 - val_loss: 20.9978 - val_mae: 2.8838\n",
            "Epoch 203/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5500 - mae: 0.9300 - val_loss: 10.3702 - val_mae: 2.3562\n",
            "Epoch 204/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6521 - mae: 0.9473 - val_loss: 12.5600 - val_mae: 2.4288\n",
            "Epoch 205/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3817 - mae: 0.8780 - val_loss: 12.8646 - val_mae: 2.5859\n",
            "Epoch 206/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5054 - mae: 0.8802 - val_loss: 12.3812 - val_mae: 2.4359\n",
            "Epoch 207/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5379 - mae: 0.9460 - val_loss: 13.5394 - val_mae: 2.5822\n",
            "Epoch 208/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5808 - mae: 0.8785 - val_loss: 20.0555 - val_mae: 3.0231\n",
            "Epoch 209/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6050 - mae: 0.9125 - val_loss: 15.8531 - val_mae: 2.7523\n",
            "Epoch 210/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6976 - mae: 0.9434 - val_loss: 15.2344 - val_mae: 2.5524\n",
            "Epoch 211/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3448 - mae: 0.8360 - val_loss: 19.1251 - val_mae: 2.9555\n",
            "Epoch 212/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5800 - mae: 0.9380 - val_loss: 13.8958 - val_mae: 2.6388\n",
            "Epoch 213/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5515 - mae: 0.9005 - val_loss: 14.7828 - val_mae: 2.5513\n",
            "Epoch 214/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4832 - mae: 0.9024 - val_loss: 12.6545 - val_mae: 2.4664\n",
            "Epoch 215/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4307 - mae: 0.9053 - val_loss: 17.8669 - val_mae: 2.7253\n",
            "Epoch 216/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8032 - mae: 0.9226 - val_loss: 15.3756 - val_mae: 2.6337\n",
            "Epoch 217/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2910 - mae: 0.8604 - val_loss: 21.2301 - val_mae: 3.0299\n",
            "Epoch 218/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4999 - mae: 0.8555 - val_loss: 15.6027 - val_mae: 2.6521\n",
            "Epoch 219/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4471 - mae: 0.8791 - val_loss: 13.6377 - val_mae: 2.5833\n",
            "Epoch 220/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5492 - mae: 0.8836 - val_loss: 14.2070 - val_mae: 2.5476\n",
            "Epoch 221/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4305 - mae: 0.8926 - val_loss: 16.2209 - val_mae: 2.8049\n",
            "Epoch 222/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4783 - mae: 0.8974 - val_loss: 15.2125 - val_mae: 2.6261\n",
            "Epoch 223/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3605 - mae: 0.8559 - val_loss: 17.5103 - val_mae: 2.7230\n",
            "Epoch 224/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4073 - mae: 0.8259 - val_loss: 15.8014 - val_mae: 2.6936\n",
            "Epoch 225/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3666 - mae: 0.8766 - val_loss: 16.4859 - val_mae: 2.7714\n",
            "Epoch 226/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2883 - mae: 0.8550 - val_loss: 19.9489 - val_mae: 2.7695\n",
            "Epoch 227/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3326 - mae: 0.8539 - val_loss: 15.3884 - val_mae: 2.6827\n",
            "Epoch 228/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6007 - mae: 0.8874 - val_loss: 16.3927 - val_mae: 2.4587\n",
            "Epoch 229/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4869 - mae: 0.8631 - val_loss: 22.6763 - val_mae: 3.1147\n",
            "Epoch 230/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4491 - mae: 0.8861 - val_loss: 21.8758 - val_mae: 2.8357\n",
            "Epoch 231/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4956 - mae: 0.9039 - val_loss: 13.2730 - val_mae: 2.4983\n",
            "Epoch 232/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2598 - mae: 0.7998 - val_loss: 23.2503 - val_mae: 3.0906\n",
            "Epoch 233/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2033 - mae: 0.8068 - val_loss: 21.7068 - val_mae: 2.7481\n",
            "Epoch 234/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2263 - mae: 0.8297 - val_loss: 19.1769 - val_mae: 2.8290\n",
            "Epoch 235/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4773 - mae: 0.8572 - val_loss: 19.1838 - val_mae: 2.7641\n",
            "Epoch 236/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4005 - mae: 0.8568 - val_loss: 14.8817 - val_mae: 2.5898\n",
            "Epoch 237/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4682 - mae: 0.8764 - val_loss: 19.7025 - val_mae: 2.8598\n",
            "Epoch 238/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2728 - mae: 0.8433 - val_loss: 17.2885 - val_mae: 2.6496\n",
            "Epoch 239/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2998 - mae: 0.8283 - val_loss: 16.6436 - val_mae: 2.6550\n",
            "Epoch 240/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1868 - mae: 0.8014 - val_loss: 17.4444 - val_mae: 2.6211\n",
            "Epoch 241/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4446 - mae: 0.8953 - val_loss: 14.8515 - val_mae: 2.6123\n",
            "Epoch 242/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2629 - mae: 0.8501 - val_loss: 20.8005 - val_mae: 2.9526\n",
            "Epoch 243/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3325 - mae: 0.8729 - val_loss: 14.8304 - val_mae: 2.6463\n",
            "Epoch 244/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3805 - mae: 0.8573 - val_loss: 15.5496 - val_mae: 2.5217\n",
            "Epoch 245/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4167 - mae: 0.8605 - val_loss: 16.8373 - val_mae: 2.6443\n",
            "Epoch 246/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2177 - mae: 0.8230 - val_loss: 15.8825 - val_mae: 2.5599\n",
            "Epoch 247/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4109 - mae: 0.8576 - val_loss: 14.2996 - val_mae: 2.5736\n",
            "Epoch 248/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3942 - mae: 0.8695 - val_loss: 16.2063 - val_mae: 2.6560\n",
            "Epoch 249/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4392 - mae: 0.8984 - val_loss: 16.3668 - val_mae: 2.7076\n",
            "Epoch 250/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3433 - mae: 0.8431 - val_loss: 14.5463 - val_mae: 2.5221\n",
            "Epoch 251/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2759 - mae: 0.8346 - val_loss: 16.1959 - val_mae: 2.8330\n",
            "Epoch 252/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2270 - mae: 0.8421 - val_loss: 15.4273 - val_mae: 2.6474\n",
            "Epoch 253/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1807 - mae: 0.8225 - val_loss: 14.9818 - val_mae: 2.5768\n",
            "Epoch 254/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2944 - mae: 0.8424 - val_loss: 16.2117 - val_mae: 2.7469\n",
            "Epoch 255/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3272 - mae: 0.8463 - val_loss: 15.9050 - val_mae: 2.5810\n",
            "Epoch 256/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1590 - mae: 0.8003 - val_loss: 16.7397 - val_mae: 2.5018\n",
            "Epoch 257/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1969 - mae: 0.7927 - val_loss: 23.5423 - val_mae: 2.9834\n",
            "Epoch 258/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3936 - mae: 0.8626 - val_loss: 22.8845 - val_mae: 3.1685\n",
            "Epoch 259/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2384 - mae: 0.8151 - val_loss: 16.6080 - val_mae: 2.6105\n",
            "Epoch 260/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2932 - mae: 0.8163 - val_loss: 19.8928 - val_mae: 2.8771\n",
            "Epoch 261/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3260 - mae: 0.8060 - val_loss: 16.2949 - val_mae: 2.5857\n",
            "Epoch 262/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2072 - mae: 0.8141 - val_loss: 16.8613 - val_mae: 2.6387\n",
            "Epoch 263/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2305 - mae: 0.8089 - val_loss: 21.3914 - val_mae: 2.8315\n",
            "Epoch 264/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2325 - mae: 0.8039 - val_loss: 22.1299 - val_mae: 2.8988\n",
            "Epoch 265/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4809 - mae: 0.8408 - val_loss: 14.9781 - val_mae: 2.4957\n",
            "Epoch 266/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2513 - mae: 0.8042 - val_loss: 17.2241 - val_mae: 2.7193\n",
            "Epoch 267/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1225 - mae: 0.7749 - val_loss: 20.2423 - val_mae: 2.8849\n",
            "Epoch 268/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4607 - mae: 0.8791 - val_loss: 14.4720 - val_mae: 2.5261\n",
            "Epoch 269/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2431 - mae: 0.8136 - val_loss: 15.8644 - val_mae: 2.7457\n",
            "Epoch 270/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3044 - mae: 0.8345 - val_loss: 16.7423 - val_mae: 2.7645\n",
            "Epoch 271/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2491 - mae: 0.8062 - val_loss: 14.4805 - val_mae: 2.5322\n",
            "Epoch 272/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2118 - mae: 0.8010 - val_loss: 17.5720 - val_mae: 3.1332\n",
            "Epoch 273/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4586 - mae: 0.8478 - val_loss: 17.4021 - val_mae: 2.8201\n",
            "Epoch 274/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1515 - mae: 0.7946 - val_loss: 14.4837 - val_mae: 2.5512\n",
            "Epoch 275/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3483 - mae: 0.8166 - val_loss: 20.1066 - val_mae: 3.2371\n",
            "Epoch 276/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2402 - mae: 0.8025 - val_loss: 13.7299 - val_mae: 2.6322\n",
            "Epoch 277/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1244 - mae: 0.8009 - val_loss: 16.4763 - val_mae: 2.7366\n",
            "Epoch 278/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1812 - mae: 0.7948 - val_loss: 14.4978 - val_mae: 2.6751\n",
            "Epoch 279/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1796 - mae: 0.7872 - val_loss: 14.6504 - val_mae: 2.5757\n",
            "Epoch 280/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0942 - mae: 0.7626 - val_loss: 15.8769 - val_mae: 2.7373\n",
            "Epoch 281/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3508 - mae: 0.8203 - val_loss: 21.3964 - val_mae: 2.9247\n",
            "Epoch 282/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1515 - mae: 0.8028 - val_loss: 15.8731 - val_mae: 2.7062\n",
            "Epoch 283/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2224 - mae: 0.8011 - val_loss: 15.0779 - val_mae: 2.6635\n",
            "Epoch 284/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0314 - mae: 0.7702 - val_loss: 18.3984 - val_mae: 2.7609\n",
            "Epoch 285/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1651 - mae: 0.8037 - val_loss: 16.0302 - val_mae: 2.7075\n",
            "Epoch 286/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0587 - mae: 0.7666 - val_loss: 15.6713 - val_mae: 2.5913\n",
            "Epoch 287/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3245 - mae: 0.8513 - val_loss: 15.9493 - val_mae: 2.5589\n",
            "Epoch 288/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1253 - mae: 0.7563 - val_loss: 17.7877 - val_mae: 2.6553\n",
            "Epoch 289/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2232 - mae: 0.7953 - val_loss: 16.6392 - val_mae: 2.5509\n",
            "Epoch 290/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1601 - mae: 0.7813 - val_loss: 15.5808 - val_mae: 2.5296\n",
            "Epoch 291/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0475 - mae: 0.7353 - val_loss: 16.9427 - val_mae: 2.6249\n",
            "Epoch 292/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0698 - mae: 0.7944 - val_loss: 17.9367 - val_mae: 2.7034\n",
            "Epoch 293/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1347 - mae: 0.7743 - val_loss: 18.4902 - val_mae: 2.7303\n",
            "Epoch 294/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1028 - mae: 0.7781 - val_loss: 16.2843 - val_mae: 2.5626\n",
            "Epoch 295/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0185 - mae: 0.7497 - val_loss: 17.6356 - val_mae: 2.6786\n",
            "Epoch 296/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1269 - mae: 0.7743 - val_loss: 19.8353 - val_mae: 2.8435\n",
            "Epoch 297/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1430 - mae: 0.7878 - val_loss: 16.6480 - val_mae: 2.6454\n",
            "Epoch 298/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1531 - mae: 0.7661 - val_loss: 20.1119 - val_mae: 2.7541\n",
            "Epoch 299/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0840 - mae: 0.7830 - val_loss: 18.2762 - val_mae: 2.8155\n",
            "Epoch 300/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1275 - mae: 0.7902 - val_loss: 12.7999 - val_mae: 2.4094\n",
            "Epoch 301/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0529 - mae: 0.7443 - val_loss: 18.7070 - val_mae: 3.0149\n",
            "Epoch 302/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1552 - mae: 0.7981 - val_loss: 18.4157 - val_mae: 2.9691\n",
            "Epoch 303/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0723 - mae: 0.7411 - val_loss: 14.3577 - val_mae: 2.4583\n",
            "Epoch 304/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0811 - mae: 0.7495 - val_loss: 14.5070 - val_mae: 2.4837\n",
            "Epoch 305/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1249 - mae: 0.7757 - val_loss: 14.4364 - val_mae: 2.5219\n",
            "Epoch 306/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9890 - mae: 0.7191 - val_loss: 19.8773 - val_mae: 2.7934\n",
            "Epoch 307/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0868 - mae: 0.7530 - val_loss: 16.6374 - val_mae: 2.6572\n",
            "Epoch 308/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0236 - mae: 0.7446 - val_loss: 21.6954 - val_mae: 2.9777\n",
            "Epoch 309/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0507 - mae: 0.7514 - val_loss: 18.6370 - val_mae: 2.6483\n",
            "Epoch 310/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0884 - mae: 0.7814 - val_loss: 20.7368 - val_mae: 2.7911\n",
            "Epoch 311/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2033 - mae: 0.7753 - val_loss: 17.8962 - val_mae: 2.7882\n",
            "Epoch 312/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9325 - mae: 0.7348 - val_loss: 22.9432 - val_mae: 3.2203\n",
            "Epoch 313/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9857 - mae: 0.7350 - val_loss: 12.1330 - val_mae: 2.3661\n",
            "Epoch 314/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1782 - mae: 0.7655 - val_loss: 17.0227 - val_mae: 2.7695\n",
            "Epoch 315/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1646 - mae: 0.7637 - val_loss: 17.9498 - val_mae: 2.7629\n",
            "Epoch 316/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1017 - mae: 0.7732 - val_loss: 18.2133 - val_mae: 2.7532\n",
            "Epoch 317/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0710 - mae: 0.7556 - val_loss: 14.4538 - val_mae: 2.4873\n",
            "Epoch 318/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1090 - mae: 0.7630 - val_loss: 15.7056 - val_mae: 2.5991\n",
            "Epoch 319/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9724 - mae: 0.7398 - val_loss: 14.8246 - val_mae: 2.5430\n",
            "Epoch 320/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9450 - mae: 0.7180 - val_loss: 19.5295 - val_mae: 2.8738\n",
            "Epoch 321/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0132 - mae: 0.7389 - val_loss: 15.8434 - val_mae: 2.6085\n",
            "Epoch 322/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1412 - mae: 0.7389 - val_loss: 18.2191 - val_mae: 2.7529\n",
            "Epoch 323/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0694 - mae: 0.8011 - val_loss: 20.8276 - val_mae: 2.9173\n",
            "Epoch 324/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1057 - mae: 0.7456 - val_loss: 20.3520 - val_mae: 2.9330\n",
            "Epoch 325/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0947 - mae: 0.7547 - val_loss: 23.8770 - val_mae: 2.8747\n",
            "Epoch 326/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9402 - mae: 0.7048 - val_loss: 25.0586 - val_mae: 2.9956\n",
            "Epoch 327/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9558 - mae: 0.7293 - val_loss: 22.9200 - val_mae: 3.0045\n",
            "Epoch 328/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9756 - mae: 0.7078 - val_loss: 21.1662 - val_mae: 2.8888\n",
            "Epoch 329/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1559 - mae: 0.7706 - val_loss: 18.0980 - val_mae: 2.8378\n",
            "Epoch 330/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9921 - mae: 0.7278 - val_loss: 20.5922 - val_mae: 2.8409\n",
            "Epoch 331/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9915 - mae: 0.7323 - val_loss: 16.1056 - val_mae: 2.5718\n",
            "Epoch 332/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0277 - mae: 0.7627 - val_loss: 23.5589 - val_mae: 3.0361\n",
            "Epoch 333/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0223 - mae: 0.7469 - val_loss: 16.4025 - val_mae: 2.5632\n",
            "Epoch 334/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1181 - mae: 0.7457 - val_loss: 18.3822 - val_mae: 2.7290\n",
            "Epoch 335/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0435 - mae: 0.7505 - val_loss: 19.1455 - val_mae: 2.7853\n",
            "Epoch 336/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0659 - mae: 0.7672 - val_loss: 24.4323 - val_mae: 3.2845\n",
            "Epoch 337/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9856 - mae: 0.7349 - val_loss: 18.4912 - val_mae: 2.7817\n",
            "Epoch 338/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9324 - mae: 0.7021 - val_loss: 21.4211 - val_mae: 2.8556\n",
            "Epoch 339/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0247 - mae: 0.7595 - val_loss: 19.6560 - val_mae: 2.8954\n",
            "Epoch 340/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1570 - mae: 0.7430 - val_loss: 16.5357 - val_mae: 2.6130\n",
            "Epoch 341/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0391 - mae: 0.7540 - val_loss: 16.1939 - val_mae: 2.6173\n",
            "Epoch 342/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9716 - mae: 0.7364 - val_loss: 23.7090 - val_mae: 3.0338\n",
            "Epoch 343/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9897 - mae: 0.7137 - val_loss: 23.9073 - val_mae: 2.9742\n",
            "Epoch 344/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9978 - mae: 0.7251 - val_loss: 15.9171 - val_mae: 2.6357\n",
            "Epoch 345/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0348 - mae: 0.7257 - val_loss: 16.4900 - val_mae: 2.6204\n",
            "Epoch 346/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0215 - mae: 0.7254 - val_loss: 19.4884 - val_mae: 2.8102\n",
            "Epoch 347/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9176 - mae: 0.6905 - val_loss: 18.7266 - val_mae: 2.7654\n",
            "Epoch 348/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9526 - mae: 0.7136 - val_loss: 20.7168 - val_mae: 2.8832\n",
            "Epoch 349/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9666 - mae: 0.7341 - val_loss: 19.3249 - val_mae: 2.8306\n",
            "Epoch 350/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2056 - mae: 0.7456 - val_loss: 15.3528 - val_mae: 2.6083\n",
            "Epoch 351/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0183 - mae: 0.7455 - val_loss: 14.3695 - val_mae: 2.6109\n",
            "Epoch 352/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8487 - mae: 0.6980 - val_loss: 15.3708 - val_mae: 2.5538\n",
            "Epoch 353/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0187 - mae: 0.7347 - val_loss: 25.4250 - val_mae: 3.0496\n",
            "Epoch 354/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2127 - mae: 0.7644 - val_loss: 17.8815 - val_mae: 2.7054\n",
            "Epoch 355/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9267 - mae: 0.7256 - val_loss: 15.3676 - val_mae: 2.6694\n",
            "Epoch 356/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9141 - mae: 0.7252 - val_loss: 20.2340 - val_mae: 2.8737\n",
            "Epoch 357/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0090 - mae: 0.7160 - val_loss: 17.3926 - val_mae: 2.6337\n",
            "Epoch 358/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0003 - mae: 0.7355 - val_loss: 18.5354 - val_mae: 2.7785\n",
            "Epoch 359/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9403 - mae: 0.7136 - val_loss: 16.1141 - val_mae: 2.6728\n",
            "Epoch 360/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9975 - mae: 0.7108 - val_loss: 16.4147 - val_mae: 2.6178\n",
            "Epoch 361/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9623 - mae: 0.7017 - val_loss: 22.1052 - val_mae: 2.9244\n",
            "Epoch 362/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9291 - mae: 0.6828 - val_loss: 20.7353 - val_mae: 2.7791\n",
            "Epoch 363/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9339 - mae: 0.7126 - val_loss: 23.6614 - val_mae: 2.9283\n",
            "Epoch 364/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0050 - mae: 0.7028 - val_loss: 21.0133 - val_mae: 2.8676\n",
            "Epoch 365/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9645 - mae: 0.6850 - val_loss: 14.7490 - val_mae: 2.5598\n",
            "Epoch 366/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0182 - mae: 0.7445 - val_loss: 24.4320 - val_mae: 3.0661\n",
            "Epoch 367/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9239 - mae: 0.7000 - val_loss: 18.7735 - val_mae: 2.8265\n",
            "Epoch 368/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9645 - mae: 0.6948 - val_loss: 18.7526 - val_mae: 2.8801\n",
            "Epoch 369/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0002 - mae: 0.7526 - val_loss: 19.6096 - val_mae: 2.8905\n",
            "Epoch 370/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0898 - mae: 0.6723 - val_loss: 17.2283 - val_mae: 2.7197\n",
            "Epoch 371/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9761 - mae: 0.7217 - val_loss: 23.5690 - val_mae: 3.1400\n",
            "Epoch 372/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8189 - mae: 0.6802 - val_loss: 18.8285 - val_mae: 2.6912\n",
            "Epoch 373/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0105 - mae: 0.7204 - val_loss: 18.2124 - val_mae: 2.6848\n",
            "Epoch 374/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8443 - mae: 0.6934 - val_loss: 24.0399 - val_mae: 3.0932\n",
            "Epoch 375/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9652 - mae: 0.7012 - val_loss: 17.7090 - val_mae: 2.6715\n",
            "Epoch 376/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8838 - mae: 0.7229 - val_loss: 21.9398 - val_mae: 2.9919\n",
            "Epoch 377/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8574 - mae: 0.6959 - val_loss: 15.4300 - val_mae: 2.6487\n",
            "Epoch 378/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1602 - mae: 0.7306 - val_loss: 20.5269 - val_mae: 2.9815\n",
            "Epoch 379/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9578 - mae: 0.7074 - val_loss: 21.5906 - val_mae: 2.8725\n",
            "Epoch 380/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9977 - mae: 0.6762 - val_loss: 18.3597 - val_mae: 2.7469\n",
            "Epoch 381/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0417 - mae: 0.7469 - val_loss: 19.9312 - val_mae: 2.8226\n",
            "Epoch 382/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8296 - mae: 0.6684 - val_loss: 22.3755 - val_mae: 2.8988\n",
            "Epoch 383/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8737 - mae: 0.6806 - val_loss: 15.8629 - val_mae: 2.7269\n",
            "Epoch 384/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0325 - mae: 0.7273 - val_loss: 18.6598 - val_mae: 2.7732\n",
            "Epoch 385/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9000 - mae: 0.6875 - val_loss: 20.0806 - val_mae: 2.9336\n",
            "Epoch 386/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9566 - mae: 0.7137 - val_loss: 17.3441 - val_mae: 2.6475\n",
            "Epoch 387/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8196 - mae: 0.6841 - val_loss: 16.9239 - val_mae: 2.5882\n",
            "Epoch 388/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8526 - mae: 0.7090 - val_loss: 15.8908 - val_mae: 2.6562\n",
            "Epoch 389/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0696 - mae: 0.7502 - val_loss: 18.0893 - val_mae: 2.7558\n",
            "Epoch 390/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9410 - mae: 0.7213 - val_loss: 18.9840 - val_mae: 2.9871\n",
            "Epoch 391/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9305 - mae: 0.7063 - val_loss: 17.9339 - val_mae: 2.7129\n",
            "Epoch 392/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8311 - mae: 0.6789 - val_loss: 17.4684 - val_mae: 2.8348\n",
            "Epoch 393/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8848 - mae: 0.6668 - val_loss: 24.9506 - val_mae: 3.0377\n",
            "Epoch 394/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9334 - mae: 0.6821 - val_loss: 20.3242 - val_mae: 2.9234\n",
            "Epoch 395/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8260 - mae: 0.6804 - val_loss: 23.0731 - val_mae: 3.0655\n",
            "Epoch 396/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8666 - mae: 0.6747 - val_loss: 18.9056 - val_mae: 2.7187\n",
            "Epoch 397/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0022 - mae: 0.7173 - val_loss: 18.6697 - val_mae: 2.8685\n",
            "Epoch 398/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8525 - mae: 0.6750 - val_loss: 26.9563 - val_mae: 3.2186\n",
            "Epoch 399/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9503 - mae: 0.7251 - val_loss: 20.0162 - val_mae: 2.7474\n",
            "Epoch 400/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8918 - mae: 0.6663 - val_loss: 20.0780 - val_mae: 2.9345\n",
            "Epoch 401/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9676 - mae: 0.6926 - val_loss: 19.2870 - val_mae: 2.7689\n",
            "Epoch 402/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7473 - mae: 0.6256 - val_loss: 19.9652 - val_mae: 2.8442\n",
            "Epoch 403/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9190 - mae: 0.6944 - val_loss: 18.1085 - val_mae: 2.7692\n",
            "Epoch 404/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7644 - mae: 0.6282 - val_loss: 18.4440 - val_mae: 2.7098\n",
            "Epoch 405/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8752 - mae: 0.6956 - val_loss: 21.2936 - val_mae: 2.9190\n",
            "Epoch 406/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.6770 - mae: 0.6224 - val_loss: 23.8565 - val_mae: 3.0878\n",
            "Epoch 407/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0486 - mae: 0.7207 - val_loss: 21.9129 - val_mae: 2.9478\n",
            "Epoch 408/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8977 - mae: 0.6715 - val_loss: 17.1948 - val_mae: 2.6157\n",
            "Epoch 409/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8361 - mae: 0.6734 - val_loss: 20.5269 - val_mae: 2.9963\n",
            "Epoch 410/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8018 - mae: 0.6808 - val_loss: 17.3475 - val_mae: 2.6984\n",
            "Epoch 411/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7838 - mae: 0.6595 - val_loss: 17.5740 - val_mae: 2.7475\n",
            "Epoch 412/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9609 - mae: 0.6753 - val_loss: 18.3064 - val_mae: 2.7343\n",
            "Epoch 413/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9298 - mae: 0.6722 - val_loss: 20.2420 - val_mae: 2.8094\n",
            "Epoch 414/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8077 - mae: 0.6656 - val_loss: 20.0130 - val_mae: 2.7848\n",
            "Epoch 415/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8979 - mae: 0.6924 - val_loss: 19.2944 - val_mae: 2.7326\n",
            "Epoch 416/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9111 - mae: 0.6892 - val_loss: 25.0992 - val_mae: 3.1074\n",
            "Epoch 417/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9745 - mae: 0.7185 - val_loss: 17.8171 - val_mae: 2.8208\n",
            "Epoch 418/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8402 - mae: 0.6806 - val_loss: 24.0377 - val_mae: 2.9722\n",
            "Epoch 419/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8021 - mae: 0.6590 - val_loss: 26.4659 - val_mae: 3.1402\n",
            "Epoch 420/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8235 - mae: 0.6576 - val_loss: 22.4320 - val_mae: 2.8305\n",
            "Epoch 421/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9175 - mae: 0.7184 - val_loss: 20.7984 - val_mae: 3.0340\n",
            "Epoch 422/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7707 - mae: 0.6239 - val_loss: 18.8606 - val_mae: 2.6888\n",
            "Epoch 423/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9860 - mae: 0.7001 - val_loss: 24.6563 - val_mae: 3.0931\n",
            "Epoch 424/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7658 - mae: 0.6708 - val_loss: 22.0193 - val_mae: 2.8436\n",
            "Epoch 425/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8355 - mae: 0.6512 - val_loss: 22.5144 - val_mae: 2.9418\n",
            "Epoch 426/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8294 - mae: 0.6591 - val_loss: 25.9154 - val_mae: 3.0321\n",
            "Epoch 427/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8288 - mae: 0.6558 - val_loss: 22.6895 - val_mae: 2.8714\n",
            "Epoch 428/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7528 - mae: 0.6306 - val_loss: 22.6242 - val_mae: 2.9211\n",
            "Epoch 429/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9108 - mae: 0.7091 - val_loss: 18.2800 - val_mae: 2.7628\n",
            "Epoch 430/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8421 - mae: 0.6727 - val_loss: 20.3119 - val_mae: 2.8406\n",
            "Epoch 431/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8962 - mae: 0.7025 - val_loss: 24.7328 - val_mae: 3.1254\n",
            "Epoch 432/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8125 - mae: 0.6456 - val_loss: 19.6052 - val_mae: 2.7521\n",
            "Epoch 433/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9551 - mae: 0.7045 - val_loss: 18.1729 - val_mae: 2.6075\n",
            "Epoch 434/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7689 - mae: 0.6701 - val_loss: 20.9827 - val_mae: 2.7970\n",
            "Epoch 435/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7390 - mae: 0.6257 - val_loss: 27.4957 - val_mae: 3.1485\n",
            "Epoch 436/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9372 - mae: 0.7086 - val_loss: 25.4240 - val_mae: 3.0070\n",
            "Epoch 437/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8701 - mae: 0.6622 - val_loss: 21.9813 - val_mae: 2.8999\n",
            "Epoch 438/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7694 - mae: 0.6696 - val_loss: 21.2706 - val_mae: 2.6867\n",
            "Epoch 439/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7572 - mae: 0.6474 - val_loss: 21.6626 - val_mae: 2.9316\n",
            "Epoch 440/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8949 - mae: 0.6740 - val_loss: 21.0578 - val_mae: 2.8245\n",
            "Epoch 441/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7427 - mae: 0.6549 - val_loss: 27.3686 - val_mae: 3.0034\n",
            "Epoch 442/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7299 - mae: 0.6438 - val_loss: 27.6093 - val_mae: 3.0875\n",
            "Epoch 443/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9026 - mae: 0.6870 - val_loss: 26.4794 - val_mae: 3.0894\n",
            "Epoch 444/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8988 - mae: 0.6878 - val_loss: 22.1026 - val_mae: 2.8613\n",
            "Epoch 445/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8066 - mae: 0.6507 - val_loss: 21.8738 - val_mae: 2.7759\n",
            "Epoch 446/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8947 - mae: 0.6893 - val_loss: 22.3114 - val_mae: 2.8787\n",
            "Epoch 447/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7614 - mae: 0.6230 - val_loss: 23.0611 - val_mae: 2.9656\n",
            "Epoch 448/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8580 - mae: 0.7005 - val_loss: 22.7847 - val_mae: 2.8570\n",
            "Epoch 449/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8927 - mae: 0.6741 - val_loss: 23.8174 - val_mae: 2.9204\n",
            "Epoch 450/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7665 - mae: 0.6451 - val_loss: 23.0553 - val_mae: 2.8439\n",
            "Epoch 451/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8324 - mae: 0.6596 - val_loss: 18.7872 - val_mae: 2.6742\n",
            "Epoch 452/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7656 - mae: 0.6224 - val_loss: 24.4580 - val_mae: 3.1022\n",
            "Epoch 453/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8184 - mae: 0.6635 - val_loss: 20.3448 - val_mae: 2.7671\n",
            "Epoch 454/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8905 - mae: 0.6705 - val_loss: 17.8982 - val_mae: 2.7298\n",
            "Epoch 455/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8998 - mae: 0.6741 - val_loss: 25.0998 - val_mae: 3.0842\n",
            "Epoch 456/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9667 - mae: 0.6827 - val_loss: 19.2835 - val_mae: 2.5883\n",
            "Epoch 457/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7916 - mae: 0.6494 - val_loss: 21.7808 - val_mae: 2.8971\n",
            "Epoch 458/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8091 - mae: 0.6530 - val_loss: 20.7532 - val_mae: 2.9088\n",
            "Epoch 459/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8337 - mae: 0.6684 - val_loss: 21.2505 - val_mae: 2.8679\n",
            "Epoch 460/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8442 - mae: 0.6593 - val_loss: 25.6585 - val_mae: 3.3389\n",
            "Epoch 461/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8656 - mae: 0.6525 - val_loss: 20.2744 - val_mae: 2.7297\n",
            "Epoch 462/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7968 - mae: 0.6195 - val_loss: 21.1119 - val_mae: 2.7602\n",
            "Epoch 463/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7980 - mae: 0.6577 - val_loss: 21.6942 - val_mae: 2.8705\n",
            "Epoch 464/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8509 - mae: 0.6667 - val_loss: 19.9045 - val_mae: 2.7466\n",
            "Epoch 465/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7767 - mae: 0.6324 - val_loss: 23.7488 - val_mae: 2.9148\n",
            "Epoch 466/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8000 - mae: 0.6543 - val_loss: 26.9734 - val_mae: 3.0652\n",
            "Epoch 467/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8077 - mae: 0.6250 - val_loss: 23.9069 - val_mae: 2.9023\n",
            "Epoch 468/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7801 - mae: 0.6479 - val_loss: 22.5873 - val_mae: 2.7585\n",
            "Epoch 469/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8488 - mae: 0.6613 - val_loss: 20.4635 - val_mae: 2.6684\n",
            "Epoch 470/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7613 - mae: 0.6368 - val_loss: 25.4648 - val_mae: 3.0735\n",
            "Epoch 471/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7738 - mae: 0.6496 - val_loss: 19.6727 - val_mae: 2.6822\n",
            "Epoch 472/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8183 - mae: 0.6640 - val_loss: 25.0914 - val_mae: 3.0266\n",
            "Epoch 473/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8240 - mae: 0.6388 - val_loss: 19.6395 - val_mae: 2.7550\n",
            "Epoch 474/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7533 - mae: 0.5995 - val_loss: 20.4314 - val_mae: 2.7617\n",
            "Epoch 475/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.6783 - mae: 0.6029 - val_loss: 26.3662 - val_mae: 2.9621\n",
            "Epoch 476/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8671 - mae: 0.6648 - val_loss: 22.7580 - val_mae: 2.9262\n",
            "Epoch 477/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8669 - mae: 0.6782 - val_loss: 25.3970 - val_mae: 3.0152\n",
            "Epoch 478/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8838 - mae: 0.6677 - val_loss: 26.7642 - val_mae: 3.0820\n",
            "Epoch 479/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.6668 - mae: 0.5987 - val_loss: 27.3409 - val_mae: 3.1238\n",
            "Epoch 480/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7266 - mae: 0.6370 - val_loss: 22.6953 - val_mae: 2.9655\n",
            "Epoch 481/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8399 - mae: 0.6616 - val_loss: 20.8087 - val_mae: 2.7953\n",
            "Epoch 482/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8564 - mae: 0.6545 - val_loss: 22.4709 - val_mae: 2.8961\n",
            "Epoch 483/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8054 - mae: 0.6472 - val_loss: 22.7428 - val_mae: 2.8463\n",
            "Epoch 484/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7944 - mae: 0.6627 - val_loss: 20.5005 - val_mae: 2.7654\n",
            "Epoch 485/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8312 - mae: 0.6557 - val_loss: 20.6794 - val_mae: 2.7916\n",
            "Epoch 486/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.6558 - mae: 0.6172 - val_loss: 20.7685 - val_mae: 2.8161\n",
            "Epoch 487/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7566 - mae: 0.6614 - val_loss: 25.7567 - val_mae: 3.1826\n",
            "Epoch 488/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8060 - mae: 0.6309 - val_loss: 30.5012 - val_mae: 3.3761\n",
            "Epoch 489/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7353 - mae: 0.6426 - val_loss: 23.2553 - val_mae: 2.6793\n",
            "Epoch 490/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8625 - mae: 0.6730 - val_loss: 24.3425 - val_mae: 3.0342\n",
            "Epoch 491/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7352 - mae: 0.6446 - val_loss: 25.0700 - val_mae: 2.9603\n",
            "Epoch 492/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7856 - mae: 0.6546 - val_loss: 26.3366 - val_mae: 2.9513\n",
            "Epoch 493/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7799 - mae: 0.6498 - val_loss: 25.2561 - val_mae: 2.9153\n",
            "Epoch 494/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7700 - mae: 0.6611 - val_loss: 18.9207 - val_mae: 2.6193\n",
            "Epoch 495/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.6903 - mae: 0.6104 - val_loss: 21.4975 - val_mae: 2.8465\n",
            "Epoch 496/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7520 - mae: 0.6398 - val_loss: 30.0455 - val_mae: 3.2220\n",
            "Epoch 497/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.6884 - mae: 0.6379 - val_loss: 25.8201 - val_mae: 2.9730\n",
            "Epoch 498/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7331 - mae: 0.6301 - val_loss: 24.9513 - val_mae: 3.0397\n",
            "Epoch 499/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7631 - mae: 0.6533 - val_loss: 24.1770 - val_mae: 2.9193\n",
            "Epoch 500/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7427 - mae: 0.6259 - val_loss: 24.5187 - val_mae: 2.8742\n",
            "processing fold # 2\n",
            "Epoch 1/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 218.1121 - mae: 11.0961 - val_loss: 32.2149 - val_mae: 3.9717\n",
            "Epoch 2/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 32.0702 - mae: 3.7068 - val_loss: 20.4275 - val_mae: 2.8837\n",
            "Epoch 3/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 22.5084 - mae: 3.1329 - val_loss: 17.6368 - val_mae: 2.6565\n",
            "Epoch 4/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 18.8444 - mae: 2.8728 - val_loss: 18.1407 - val_mae: 2.7580\n",
            "Epoch 5/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 15.7999 - mae: 2.6496 - val_loss: 16.3358 - val_mae: 2.6410\n",
            "Epoch 6/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 14.1635 - mae: 2.5491 - val_loss: 16.4763 - val_mae: 2.6291\n",
            "Epoch 7/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 12.4710 - mae: 2.3639 - val_loss: 16.5633 - val_mae: 2.7966\n",
            "Epoch 8/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 12.2642 - mae: 2.3933 - val_loss: 16.4135 - val_mae: 2.6955\n",
            "Epoch 9/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 11.1014 - mae: 2.2582 - val_loss: 16.7620 - val_mae: 2.6964\n",
            "Epoch 10/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 9.9839 - mae: 2.1814 - val_loss: 18.2483 - val_mae: 2.7500\n",
            "Epoch 11/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 10.1585 - mae: 2.2528 - val_loss: 15.8300 - val_mae: 2.6413\n",
            "Epoch 12/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 10.1970 - mae: 2.1408 - val_loss: 15.2454 - val_mae: 2.5733\n",
            "Epoch 13/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 9.3363 - mae: 2.1356 - val_loss: 16.7303 - val_mae: 2.8512\n",
            "Epoch 14/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 9.3192 - mae: 2.0682 - val_loss: 15.8364 - val_mae: 2.6408\n",
            "Epoch 15/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 8.7339 - mae: 2.1128 - val_loss: 16.7573 - val_mae: 2.7744\n",
            "Epoch 16/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 9.1724 - mae: 2.0622 - val_loss: 15.1232 - val_mae: 2.6247\n",
            "Epoch 17/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 8.5463 - mae: 2.0498 - val_loss: 16.2281 - val_mae: 2.7911\n",
            "Epoch 18/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 8.2579 - mae: 2.0146 - val_loss: 16.6557 - val_mae: 2.7420\n",
            "Epoch 19/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 8.2921 - mae: 1.9448 - val_loss: 17.2967 - val_mae: 2.7379\n",
            "Epoch 20/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 8.0075 - mae: 1.9512 - val_loss: 16.5480 - val_mae: 2.6416\n",
            "Epoch 21/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.8385 - mae: 1.9275 - val_loss: 16.5230 - val_mae: 2.7718\n",
            "Epoch 22/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.6694 - mae: 1.9208 - val_loss: 17.2599 - val_mae: 3.0029\n",
            "Epoch 23/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.8562 - mae: 1.8966 - val_loss: 14.2198 - val_mae: 2.5192\n",
            "Epoch 24/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.5637 - mae: 1.8553 - val_loss: 16.3101 - val_mae: 2.7872\n",
            "Epoch 25/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.1952 - mae: 1.8245 - val_loss: 17.5053 - val_mae: 2.6790\n",
            "Epoch 26/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.5040 - mae: 1.8397 - val_loss: 16.1122 - val_mae: 2.8336\n",
            "Epoch 27/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.3917 - mae: 1.8318 - val_loss: 15.4653 - val_mae: 2.7210\n",
            "Epoch 28/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.2972 - mae: 1.7838 - val_loss: 15.8600 - val_mae: 2.6845\n",
            "Epoch 29/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.5695 - mae: 1.7863 - val_loss: 15.0137 - val_mae: 2.6518\n",
            "Epoch 30/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.8750 - mae: 1.7267 - val_loss: 14.8927 - val_mae: 2.4888\n",
            "Epoch 31/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.6511 - mae: 1.7578 - val_loss: 17.6986 - val_mae: 2.7137\n",
            "Epoch 32/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.0488 - mae: 1.7406 - val_loss: 15.4307 - val_mae: 2.6801\n",
            "Epoch 33/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.6964 - mae: 1.7378 - val_loss: 14.3301 - val_mae: 2.5171\n",
            "Epoch 34/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.3523 - mae: 1.7157 - val_loss: 15.3739 - val_mae: 2.5447\n",
            "Epoch 35/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.6183 - mae: 1.7179 - val_loss: 14.9243 - val_mae: 2.5575\n",
            "Epoch 36/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.0757 - mae: 1.7051 - val_loss: 15.0791 - val_mae: 2.5810\n",
            "Epoch 37/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.7832 - mae: 1.6449 - val_loss: 13.4790 - val_mae: 2.3972\n",
            "Epoch 38/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.1493 - mae: 1.6777 - val_loss: 14.4385 - val_mae: 2.5032\n",
            "Epoch 39/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.6823 - mae: 1.6467 - val_loss: 14.6975 - val_mae: 2.5126\n",
            "Epoch 40/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.6595 - mae: 1.6182 - val_loss: 16.1361 - val_mae: 2.6713\n",
            "Epoch 41/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.1198 - mae: 1.6706 - val_loss: 15.3567 - val_mae: 2.5057\n",
            "Epoch 42/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.8323 - mae: 1.6570 - val_loss: 14.8322 - val_mae: 2.6510\n",
            "Epoch 43/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.5498 - mae: 1.6604 - val_loss: 15.0179 - val_mae: 2.5507\n",
            "Epoch 44/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.5021 - mae: 1.6173 - val_loss: 15.0228 - val_mae: 2.5903\n",
            "Epoch 45/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.4038 - mae: 1.5551 - val_loss: 16.2778 - val_mae: 2.7448\n",
            "Epoch 46/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.2800 - mae: 1.6078 - val_loss: 15.7833 - val_mae: 2.7086\n",
            "Epoch 47/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.0178 - mae: 1.5489 - val_loss: 14.2725 - val_mae: 2.5243\n",
            "Epoch 48/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.2484 - mae: 1.5374 - val_loss: 14.6149 - val_mae: 2.4713\n",
            "Epoch 49/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.5048 - mae: 1.5969 - val_loss: 16.0309 - val_mae: 2.7884\n",
            "Epoch 50/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.3084 - mae: 1.5317 - val_loss: 14.7442 - val_mae: 2.6560\n",
            "Epoch 51/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.9545 - mae: 1.5091 - val_loss: 17.5245 - val_mae: 2.7258\n",
            "Epoch 52/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.1646 - mae: 1.5075 - val_loss: 14.7508 - val_mae: 2.5740\n",
            "Epoch 53/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.9781 - mae: 1.5174 - val_loss: 14.9764 - val_mae: 2.6331\n",
            "Epoch 54/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.0113 - mae: 1.5737 - val_loss: 15.6703 - val_mae: 2.6699\n",
            "Epoch 55/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.8522 - mae: 1.5216 - val_loss: 14.8209 - val_mae: 2.5789\n",
            "Epoch 56/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.8297 - mae: 1.5347 - val_loss: 15.4328 - val_mae: 2.5533\n",
            "Epoch 57/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.6903 - mae: 1.5022 - val_loss: 14.7501 - val_mae: 2.5294\n",
            "Epoch 58/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.5880 - mae: 1.4943 - val_loss: 14.9767 - val_mae: 2.6880\n",
            "Epoch 59/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.7543 - mae: 1.4615 - val_loss: 14.0263 - val_mae: 2.5304\n",
            "Epoch 60/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.3324 - mae: 1.4319 - val_loss: 16.1282 - val_mae: 2.9551\n",
            "Epoch 61/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.5448 - mae: 1.5261 - val_loss: 13.8946 - val_mae: 2.5213\n",
            "Epoch 62/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.5053 - mae: 1.5158 - val_loss: 14.3725 - val_mae: 2.5055\n",
            "Epoch 63/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.7166 - mae: 1.4568 - val_loss: 13.4547 - val_mae: 2.4644\n",
            "Epoch 64/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.4166 - mae: 1.3874 - val_loss: 15.5510 - val_mae: 2.6767\n",
            "Epoch 65/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.2070 - mae: 1.4446 - val_loss: 15.9416 - val_mae: 2.6933\n",
            "Epoch 66/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.4905 - mae: 1.4516 - val_loss: 15.0810 - val_mae: 2.5962\n",
            "Epoch 67/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.1522 - mae: 1.4089 - val_loss: 17.6634 - val_mae: 2.7388\n",
            "Epoch 68/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.4693 - mae: 1.4565 - val_loss: 15.0883 - val_mae: 2.6039\n",
            "Epoch 69/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.6133 - mae: 1.4687 - val_loss: 14.6287 - val_mae: 2.5833\n",
            "Epoch 70/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.9230 - mae: 1.3768 - val_loss: 16.2932 - val_mae: 2.9032\n",
            "Epoch 71/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.1805 - mae: 1.4199 - val_loss: 14.7467 - val_mae: 2.5498\n",
            "Epoch 72/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.2266 - mae: 1.4218 - val_loss: 14.9628 - val_mae: 2.6530\n",
            "Epoch 73/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.3999 - mae: 1.4177 - val_loss: 14.7539 - val_mae: 2.6238\n",
            "Epoch 74/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.7948 - mae: 1.3054 - val_loss: 14.0911 - val_mae: 2.6402\n",
            "Epoch 75/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.6997 - mae: 1.3652 - val_loss: 14.1401 - val_mae: 2.6019\n",
            "Epoch 76/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.9312 - mae: 1.3268 - val_loss: 12.9055 - val_mae: 2.4620\n",
            "Epoch 77/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.9582 - mae: 1.3470 - val_loss: 15.2236 - val_mae: 2.5573\n",
            "Epoch 78/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.8928 - mae: 1.3525 - val_loss: 14.1520 - val_mae: 2.5983\n",
            "Epoch 79/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.0881 - mae: 1.3977 - val_loss: 14.8545 - val_mae: 2.5994\n",
            "Epoch 80/500\n",
            "303/303 [==============================] - 0s 2ms/step - loss: 3.5969 - mae: 1.3146 - val_loss: 14.6655 - val_mae: 2.5979\n",
            "Epoch 81/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.8228 - mae: 1.3729 - val_loss: 16.2335 - val_mae: 2.6951\n",
            "Epoch 82/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.7010 - mae: 1.3443 - val_loss: 15.7952 - val_mae: 2.7242\n",
            "Epoch 83/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.5808 - mae: 1.3369 - val_loss: 16.0859 - val_mae: 2.8367\n",
            "Epoch 84/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.6126 - mae: 1.3226 - val_loss: 16.0386 - val_mae: 2.8279\n",
            "Epoch 85/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.6338 - mae: 1.2932 - val_loss: 14.0147 - val_mae: 2.6496\n",
            "Epoch 86/500\n",
            "303/303 [==============================] - 0s 2ms/step - loss: 3.4202 - mae: 1.3060 - val_loss: 15.9670 - val_mae: 2.6605\n",
            "Epoch 87/500\n",
            "303/303 [==============================] - 1s 2ms/step - loss: 3.7246 - mae: 1.2752 - val_loss: 13.9978 - val_mae: 2.6007\n",
            "Epoch 88/500\n",
            "303/303 [==============================] - 1s 2ms/step - loss: 3.5319 - mae: 1.3096 - val_loss: 14.9299 - val_mae: 2.6452\n",
            "Epoch 89/500\n",
            "303/303 [==============================] - 1s 2ms/step - loss: 3.5547 - mae: 1.3112 - val_loss: 15.9970 - val_mae: 2.9042\n",
            "Epoch 90/500\n",
            "303/303 [==============================] - 0s 2ms/step - loss: 3.4441 - mae: 1.2873 - val_loss: 15.0138 - val_mae: 2.5285\n",
            "Epoch 91/500\n",
            "303/303 [==============================] - 1s 2ms/step - loss: 3.2347 - mae: 1.2466 - val_loss: 14.4065 - val_mae: 2.4762\n",
            "Epoch 92/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.1198 - mae: 1.2784 - val_loss: 13.1751 - val_mae: 2.4914\n",
            "Epoch 93/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.3706 - mae: 1.3094 - val_loss: 14.7833 - val_mae: 2.4591\n",
            "Epoch 94/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.3253 - mae: 1.2661 - val_loss: 14.7804 - val_mae: 2.5989\n",
            "Epoch 95/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.4156 - mae: 1.2870 - val_loss: 13.3985 - val_mae: 2.5685\n",
            "Epoch 96/500\n",
            "303/303 [==============================] - 0s 2ms/step - loss: 3.1334 - mae: 1.2342 - val_loss: 15.3069 - val_mae: 2.6726\n",
            "Epoch 97/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.3769 - mae: 1.2544 - val_loss: 14.2667 - val_mae: 2.5362\n",
            "Epoch 98/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.0512 - mae: 1.2435 - val_loss: 14.3864 - val_mae: 2.5397\n",
            "Epoch 99/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.3222 - mae: 1.2837 - val_loss: 15.0534 - val_mae: 2.6883\n",
            "Epoch 100/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.2406 - mae: 1.2389 - val_loss: 14.1209 - val_mae: 2.5712\n",
            "Epoch 101/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.1572 - mae: 1.2316 - val_loss: 13.5735 - val_mae: 2.5503\n",
            "Epoch 102/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.1249 - mae: 1.2510 - val_loss: 14.3468 - val_mae: 2.6347\n",
            "Epoch 103/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.9887 - mae: 1.1557 - val_loss: 15.0658 - val_mae: 2.5525\n",
            "Epoch 104/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.1035 - mae: 1.2554 - val_loss: 14.8594 - val_mae: 2.6867\n",
            "Epoch 105/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.7664 - mae: 1.1808 - val_loss: 13.8864 - val_mae: 2.5732\n",
            "Epoch 106/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.2141 - mae: 1.2867 - val_loss: 14.9598 - val_mae: 2.5242\n",
            "Epoch 107/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.0159 - mae: 1.1890 - val_loss: 17.4737 - val_mae: 2.7868\n",
            "Epoch 108/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.1943 - mae: 1.2445 - val_loss: 14.2228 - val_mae: 2.5890\n",
            "Epoch 109/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.1016 - mae: 1.2126 - val_loss: 14.6419 - val_mae: 2.6816\n",
            "Epoch 110/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.0368 - mae: 1.2240 - val_loss: 13.9065 - val_mae: 2.5794\n",
            "Epoch 111/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.8966 - mae: 1.2279 - val_loss: 14.6990 - val_mae: 2.6896\n",
            "Epoch 112/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.9933 - mae: 1.1904 - val_loss: 15.2150 - val_mae: 2.6901\n",
            "Epoch 113/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.6976 - mae: 1.1556 - val_loss: 16.1742 - val_mae: 2.7893\n",
            "Epoch 114/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.9629 - mae: 1.2001 - val_loss: 14.8671 - val_mae: 2.5927\n",
            "Epoch 115/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.6093 - mae: 1.1468 - val_loss: 15.6264 - val_mae: 2.7404\n",
            "Epoch 116/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.7056 - mae: 1.1623 - val_loss: 14.2821 - val_mae: 2.4742\n",
            "Epoch 117/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.7008 - mae: 1.2050 - val_loss: 15.0003 - val_mae: 2.5807\n",
            "Epoch 118/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.0737 - mae: 1.1262 - val_loss: 17.0388 - val_mae: 2.9886\n",
            "Epoch 119/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.4616 - mae: 1.1249 - val_loss: 14.5870 - val_mae: 2.6225\n",
            "Epoch 120/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.6363 - mae: 1.1804 - val_loss: 16.1027 - val_mae: 2.8141\n",
            "Epoch 121/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.7290 - mae: 1.1232 - val_loss: 17.0351 - val_mae: 2.9971\n",
            "Epoch 122/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.7617 - mae: 1.1482 - val_loss: 15.4886 - val_mae: 2.6769\n",
            "Epoch 123/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.9257 - mae: 1.2026 - val_loss: 15.4888 - val_mae: 2.7822\n",
            "Epoch 124/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.5522 - mae: 1.1537 - val_loss: 15.1833 - val_mae: 2.7023\n",
            "Epoch 125/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3915 - mae: 1.0772 - val_loss: 18.9441 - val_mae: 2.9585\n",
            "Epoch 126/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.8237 - mae: 1.1803 - val_loss: 14.0609 - val_mae: 2.6229\n",
            "Epoch 127/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.4596 - mae: 1.1575 - val_loss: 16.3891 - val_mae: 2.7158\n",
            "Epoch 128/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.6437 - mae: 1.1425 - val_loss: 14.3646 - val_mae: 2.6311\n",
            "Epoch 129/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.5647 - mae: 1.1508 - val_loss: 15.1252 - val_mae: 2.7469\n",
            "Epoch 130/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3351 - mae: 1.1016 - val_loss: 15.7889 - val_mae: 2.7133\n",
            "Epoch 131/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.5966 - mae: 1.1245 - val_loss: 14.7832 - val_mae: 2.7138\n",
            "Epoch 132/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.6290 - mae: 1.1518 - val_loss: 14.5908 - val_mae: 2.5447\n",
            "Epoch 133/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3304 - mae: 1.0797 - val_loss: 14.1091 - val_mae: 2.6617\n",
            "Epoch 134/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2480 - mae: 1.1015 - val_loss: 15.4172 - val_mae: 2.6512\n",
            "Epoch 135/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.4198 - mae: 1.0405 - val_loss: 15.7653 - val_mae: 2.7068\n",
            "Epoch 136/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2447 - mae: 1.0464 - val_loss: 15.8371 - val_mae: 2.7583\n",
            "Epoch 137/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.4776 - mae: 1.1432 - val_loss: 14.3279 - val_mae: 2.6212\n",
            "Epoch 138/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1822 - mae: 1.1123 - val_loss: 14.8168 - val_mae: 2.6652\n",
            "Epoch 139/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.4418 - mae: 1.0976 - val_loss: 14.4604 - val_mae: 2.5882\n",
            "Epoch 140/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2469 - mae: 1.0167 - val_loss: 16.1407 - val_mae: 2.9523\n",
            "Epoch 141/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3376 - mae: 1.1247 - val_loss: 14.8662 - val_mae: 2.6039\n",
            "Epoch 142/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2490 - mae: 1.0558 - val_loss: 17.1458 - val_mae: 2.9458\n",
            "Epoch 143/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1691 - mae: 1.0478 - val_loss: 16.5100 - val_mae: 2.7000\n",
            "Epoch 144/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.4356 - mae: 1.1087 - val_loss: 15.8256 - val_mae: 2.8114\n",
            "Epoch 145/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.4952 - mae: 1.0817 - val_loss: 15.3707 - val_mae: 2.6530\n",
            "Epoch 146/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2027 - mae: 1.0745 - val_loss: 14.3411 - val_mae: 2.6110\n",
            "Epoch 147/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0243 - mae: 1.0394 - val_loss: 14.5695 - val_mae: 2.6017\n",
            "Epoch 148/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.4674 - mae: 1.0893 - val_loss: 17.0004 - val_mae: 3.0023\n",
            "Epoch 149/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3022 - mae: 1.0994 - val_loss: 14.5835 - val_mae: 2.5863\n",
            "Epoch 150/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2438 - mae: 1.0348 - val_loss: 15.4206 - val_mae: 2.6614\n",
            "Epoch 151/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2143 - mae: 1.0677 - val_loss: 14.5071 - val_mae: 2.6525\n",
            "Epoch 152/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3113 - mae: 1.0190 - val_loss: 13.9768 - val_mae: 2.6014\n",
            "Epoch 153/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1567 - mae: 1.0201 - val_loss: 15.0127 - val_mae: 2.6232\n",
            "Epoch 154/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1701 - mae: 1.0154 - val_loss: 14.5428 - val_mae: 2.6221\n",
            "Epoch 155/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0395 - mae: 1.0371 - val_loss: 15.8076 - val_mae: 2.6884\n",
            "Epoch 156/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1935 - mae: 1.0837 - val_loss: 15.0742 - val_mae: 2.6623\n",
            "Epoch 157/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1557 - mae: 1.0035 - val_loss: 14.8885 - val_mae: 2.6220\n",
            "Epoch 158/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0114 - mae: 0.9888 - val_loss: 13.6670 - val_mae: 2.5565\n",
            "Epoch 159/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0908 - mae: 1.0119 - val_loss: 14.0034 - val_mae: 2.5336\n",
            "Epoch 160/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9915 - mae: 0.9861 - val_loss: 14.3463 - val_mae: 2.6106\n",
            "Epoch 161/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9582 - mae: 1.0323 - val_loss: 15.2112 - val_mae: 2.7445\n",
            "Epoch 162/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1427 - mae: 1.0322 - val_loss: 14.8949 - val_mae: 2.6470\n",
            "Epoch 163/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9547 - mae: 1.0024 - val_loss: 13.8646 - val_mae: 2.6748\n",
            "Epoch 164/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1817 - mae: 1.0390 - val_loss: 14.4359 - val_mae: 2.6423\n",
            "Epoch 165/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8874 - mae: 0.9682 - val_loss: 17.1702 - val_mae: 3.1642\n",
            "Epoch 166/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9492 - mae: 0.9733 - val_loss: 14.9769 - val_mae: 2.7733\n",
            "Epoch 167/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0197 - mae: 0.9899 - val_loss: 15.7310 - val_mae: 2.7443\n",
            "Epoch 168/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8755 - mae: 0.9856 - val_loss: 15.7208 - val_mae: 2.8173\n",
            "Epoch 169/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8818 - mae: 1.0011 - val_loss: 17.8297 - val_mae: 2.9532\n",
            "Epoch 170/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9030 - mae: 0.9799 - val_loss: 14.5132 - val_mae: 2.6636\n",
            "Epoch 171/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0120 - mae: 0.9882 - val_loss: 15.4256 - val_mae: 2.7017\n",
            "Epoch 172/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7195 - mae: 0.9480 - val_loss: 16.2250 - val_mae: 2.7404\n",
            "Epoch 173/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7010 - mae: 0.9382 - val_loss: 15.4895 - val_mae: 2.8651\n",
            "Epoch 174/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9302 - mae: 0.9428 - val_loss: 15.0501 - val_mae: 2.7071\n",
            "Epoch 175/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9002 - mae: 0.9528 - val_loss: 14.6706 - val_mae: 2.7038\n",
            "Epoch 176/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7162 - mae: 0.9544 - val_loss: 14.9335 - val_mae: 2.6610\n",
            "Epoch 177/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9780 - mae: 0.9616 - val_loss: 14.7256 - val_mae: 2.6122\n",
            "Epoch 178/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7722 - mae: 0.9718 - val_loss: 14.8454 - val_mae: 2.6218\n",
            "Epoch 179/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7355 - mae: 0.9229 - val_loss: 14.4283 - val_mae: 2.6229\n",
            "Epoch 180/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9083 - mae: 0.9768 - val_loss: 15.0399 - val_mae: 2.6163\n",
            "Epoch 181/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5216 - mae: 0.9000 - val_loss: 15.7848 - val_mae: 2.8622\n",
            "Epoch 182/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9227 - mae: 0.9484 - val_loss: 14.4533 - val_mae: 2.6859\n",
            "Epoch 183/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7196 - mae: 0.9577 - val_loss: 15.8984 - val_mae: 2.7662\n",
            "Epoch 184/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0434 - mae: 0.9578 - val_loss: 14.0629 - val_mae: 2.6449\n",
            "Epoch 185/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5056 - mae: 0.8585 - val_loss: 14.7431 - val_mae: 2.6997\n",
            "Epoch 186/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8348 - mae: 0.9439 - val_loss: 14.4964 - val_mae: 2.6551\n",
            "Epoch 187/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9011 - mae: 0.9436 - val_loss: 14.9197 - val_mae: 2.7651\n",
            "Epoch 188/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5763 - mae: 0.9044 - val_loss: 16.3155 - val_mae: 2.8192\n",
            "Epoch 189/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7990 - mae: 0.9348 - val_loss: 15.0551 - val_mae: 2.7549\n",
            "Epoch 190/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6414 - mae: 0.9012 - val_loss: 15.0193 - val_mae: 2.6635\n",
            "Epoch 191/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7702 - mae: 0.9547 - val_loss: 15.0645 - val_mae: 2.8028\n",
            "Epoch 192/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6086 - mae: 0.8961 - val_loss: 14.3546 - val_mae: 2.7357\n",
            "Epoch 193/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6290 - mae: 0.9453 - val_loss: 15.6862 - val_mae: 2.7487\n",
            "Epoch 194/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7317 - mae: 0.9651 - val_loss: 15.0689 - val_mae: 2.7273\n",
            "Epoch 195/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5424 - mae: 0.8933 - val_loss: 14.4523 - val_mae: 2.7487\n",
            "Epoch 196/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5089 - mae: 0.8631 - val_loss: 16.3420 - val_mae: 2.8101\n",
            "Epoch 197/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6431 - mae: 0.8855 - val_loss: 14.8529 - val_mae: 2.6910\n",
            "Epoch 198/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7306 - mae: 0.9405 - val_loss: 15.7376 - val_mae: 2.7558\n",
            "Epoch 199/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6906 - mae: 0.9546 - val_loss: 15.0767 - val_mae: 2.7377\n",
            "Epoch 200/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6119 - mae: 0.9113 - val_loss: 14.1606 - val_mae: 2.6774\n",
            "Epoch 201/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4005 - mae: 0.8571 - val_loss: 14.2968 - val_mae: 2.6505\n",
            "Epoch 202/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7181 - mae: 0.9209 - val_loss: 14.3845 - val_mae: 2.6693\n",
            "Epoch 203/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5718 - mae: 0.8856 - val_loss: 16.7986 - val_mae: 2.9165\n",
            "Epoch 204/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4531 - mae: 0.8554 - val_loss: 13.7828 - val_mae: 2.6254\n",
            "Epoch 205/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6471 - mae: 0.8932 - val_loss: 14.9530 - val_mae: 2.6355\n",
            "Epoch 206/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7203 - mae: 0.9322 - val_loss: 16.6613 - val_mae: 2.8448\n",
            "Epoch 207/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6872 - mae: 0.9255 - val_loss: 14.5517 - val_mae: 2.6902\n",
            "Epoch 208/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7205 - mae: 0.9143 - val_loss: 14.7246 - val_mae: 2.6556\n",
            "Epoch 209/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5393 - mae: 0.8684 - val_loss: 14.2125 - val_mae: 2.6252\n",
            "Epoch 210/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5684 - mae: 0.8687 - val_loss: 14.7695 - val_mae: 2.6876\n",
            "Epoch 211/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4807 - mae: 0.8811 - val_loss: 16.0263 - val_mae: 2.7899\n",
            "Epoch 212/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6628 - mae: 0.8661 - val_loss: 15.0888 - val_mae: 2.8065\n",
            "Epoch 213/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6448 - mae: 0.9288 - val_loss: 16.3139 - val_mae: 2.8288\n",
            "Epoch 214/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3885 - mae: 0.8700 - val_loss: 15.2406 - val_mae: 2.7729\n",
            "Epoch 215/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5417 - mae: 0.9107 - val_loss: 14.6035 - val_mae: 2.7243\n",
            "Epoch 216/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5737 - mae: 0.8852 - val_loss: 17.7400 - val_mae: 2.8686\n",
            "Epoch 217/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3855 - mae: 0.8755 - val_loss: 15.1220 - val_mae: 2.7281\n",
            "Epoch 218/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6317 - mae: 0.8698 - val_loss: 15.4598 - val_mae: 2.7369\n",
            "Epoch 219/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6510 - mae: 0.8891 - val_loss: 16.2181 - val_mae: 2.7875\n",
            "Epoch 220/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4486 - mae: 0.8295 - val_loss: 14.4825 - val_mae: 2.6716\n",
            "Epoch 221/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5387 - mae: 0.8732 - val_loss: 14.8391 - val_mae: 2.6530\n",
            "Epoch 222/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5980 - mae: 0.8689 - val_loss: 14.7384 - val_mae: 2.6622\n",
            "Epoch 223/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4645 - mae: 0.8776 - val_loss: 14.9807 - val_mae: 2.6282\n",
            "Epoch 224/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4568 - mae: 0.8596 - val_loss: 15.8180 - val_mae: 2.7130\n",
            "Epoch 225/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4115 - mae: 0.7814 - val_loss: 15.3930 - val_mae: 2.7644\n",
            "Epoch 226/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4979 - mae: 0.8482 - val_loss: 15.4121 - val_mae: 2.6858\n",
            "Epoch 227/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5304 - mae: 0.8582 - val_loss: 14.3722 - val_mae: 2.6791\n",
            "Epoch 228/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5000 - mae: 0.8466 - val_loss: 14.4331 - val_mae: 2.6288\n",
            "Epoch 229/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4587 - mae: 0.8429 - val_loss: 15.8094 - val_mae: 2.7381\n",
            "Epoch 230/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4264 - mae: 0.8575 - val_loss: 16.3068 - val_mae: 2.7666\n",
            "Epoch 231/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3012 - mae: 0.8220 - val_loss: 16.6105 - val_mae: 2.7531\n",
            "Epoch 232/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3317 - mae: 0.8103 - val_loss: 15.0288 - val_mae: 2.7348\n",
            "Epoch 233/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4243 - mae: 0.8590 - val_loss: 14.5096 - val_mae: 2.6981\n",
            "Epoch 234/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4044 - mae: 0.8473 - val_loss: 14.8449 - val_mae: 2.6802\n",
            "Epoch 235/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3490 - mae: 0.8287 - val_loss: 14.2794 - val_mae: 2.7265\n",
            "Epoch 236/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5251 - mae: 0.8475 - val_loss: 14.2751 - val_mae: 2.6517\n",
            "Epoch 237/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5021 - mae: 0.8261 - val_loss: 15.4690 - val_mae: 2.7689\n",
            "Epoch 238/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3288 - mae: 0.8102 - val_loss: 15.1057 - val_mae: 2.7442\n",
            "Epoch 239/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6594 - mae: 0.8409 - val_loss: 15.8426 - val_mae: 2.7267\n",
            "Epoch 240/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3433 - mae: 0.7987 - val_loss: 16.1036 - val_mae: 2.7506\n",
            "Epoch 241/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3357 - mae: 0.8201 - val_loss: 13.8963 - val_mae: 2.6012\n",
            "Epoch 242/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2696 - mae: 0.8171 - val_loss: 16.0778 - val_mae: 2.7906\n",
            "Epoch 243/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5061 - mae: 0.8382 - val_loss: 14.9443 - val_mae: 2.6770\n",
            "Epoch 244/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4924 - mae: 0.8214 - val_loss: 14.8423 - val_mae: 2.7456\n",
            "Epoch 245/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3556 - mae: 0.7801 - val_loss: 14.6139 - val_mae: 2.7576\n",
            "Epoch 246/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2068 - mae: 0.8052 - val_loss: 16.4963 - val_mae: 2.9186\n",
            "Epoch 247/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3877 - mae: 0.8141 - val_loss: 15.0389 - val_mae: 2.7482\n",
            "Epoch 248/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3949 - mae: 0.8331 - val_loss: 13.6846 - val_mae: 2.5982\n",
            "Epoch 249/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5154 - mae: 0.8582 - val_loss: 16.0889 - val_mae: 2.9238\n",
            "Epoch 250/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3832 - mae: 0.8483 - val_loss: 14.8270 - val_mae: 2.6228\n",
            "Epoch 251/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3712 - mae: 0.8270 - val_loss: 15.0752 - val_mae: 2.7728\n",
            "Epoch 252/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2941 - mae: 0.7918 - val_loss: 14.6709 - val_mae: 2.7102\n",
            "Epoch 253/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4086 - mae: 0.8663 - val_loss: 14.0654 - val_mae: 2.6247\n",
            "Epoch 254/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1887 - mae: 0.7642 - val_loss: 14.6743 - val_mae: 2.7069\n",
            "Epoch 255/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2393 - mae: 0.8010 - val_loss: 13.5730 - val_mae: 2.6348\n",
            "Epoch 256/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2100 - mae: 0.7811 - val_loss: 14.2589 - val_mae: 2.6879\n",
            "Epoch 257/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2656 - mae: 0.8054 - val_loss: 15.9716 - val_mae: 2.7659\n",
            "Epoch 258/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2396 - mae: 0.8325 - val_loss: 14.7960 - val_mae: 2.6653\n",
            "Epoch 259/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2912 - mae: 0.7843 - val_loss: 14.5346 - val_mae: 2.6701\n",
            "Epoch 260/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2815 - mae: 0.8194 - val_loss: 15.9383 - val_mae: 2.8237\n",
            "Epoch 261/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2597 - mae: 0.8083 - val_loss: 15.6344 - val_mae: 2.8232\n",
            "Epoch 262/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4135 - mae: 0.8409 - val_loss: 14.2110 - val_mae: 2.6430\n",
            "Epoch 263/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1092 - mae: 0.7833 - val_loss: 14.4535 - val_mae: 2.6633\n",
            "Epoch 264/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2879 - mae: 0.8140 - val_loss: 14.4680 - val_mae: 2.6821\n",
            "Epoch 265/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1701 - mae: 0.7504 - val_loss: 13.8973 - val_mae: 2.5651\n",
            "Epoch 266/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3578 - mae: 0.8286 - val_loss: 16.1257 - val_mae: 2.8651\n",
            "Epoch 267/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3343 - mae: 0.8050 - val_loss: 13.7187 - val_mae: 2.6286\n",
            "Epoch 268/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2274 - mae: 0.7847 - val_loss: 15.6904 - val_mae: 2.8237\n",
            "Epoch 269/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1069 - mae: 0.7577 - val_loss: 14.9463 - val_mae: 2.6252\n",
            "Epoch 270/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2673 - mae: 0.7828 - val_loss: 14.8706 - val_mae: 2.6546\n",
            "Epoch 271/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2788 - mae: 0.7729 - val_loss: 14.7421 - val_mae: 2.6940\n",
            "Epoch 272/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2400 - mae: 0.7373 - val_loss: 14.3037 - val_mae: 2.5730\n",
            "Epoch 273/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2782 - mae: 0.7951 - val_loss: 14.5524 - val_mae: 2.6446\n",
            "Epoch 274/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2201 - mae: 0.7684 - val_loss: 14.0805 - val_mae: 2.6626\n",
            "Epoch 275/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1952 - mae: 0.8207 - val_loss: 14.5453 - val_mae: 2.6947\n",
            "Epoch 276/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1098 - mae: 0.7564 - val_loss: 13.9002 - val_mae: 2.6097\n",
            "Epoch 277/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2532 - mae: 0.7799 - val_loss: 14.1201 - val_mae: 2.7060\n",
            "Epoch 278/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2110 - mae: 0.7602 - val_loss: 14.9659 - val_mae: 2.7319\n",
            "Epoch 279/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1855 - mae: 0.7669 - val_loss: 15.6116 - val_mae: 2.8193\n",
            "Epoch 280/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4315 - mae: 0.7586 - val_loss: 13.7355 - val_mae: 2.6298\n",
            "Epoch 281/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1068 - mae: 0.7073 - val_loss: 14.3145 - val_mae: 2.6426\n",
            "Epoch 282/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1467 - mae: 0.8020 - val_loss: 14.6690 - val_mae: 2.7871\n",
            "Epoch 283/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2031 - mae: 0.7461 - val_loss: 15.2655 - val_mae: 2.7619\n",
            "Epoch 284/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1820 - mae: 0.7567 - val_loss: 14.9825 - val_mae: 2.7919\n",
            "Epoch 285/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2432 - mae: 0.7610 - val_loss: 14.1376 - val_mae: 2.6432\n",
            "Epoch 286/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1668 - mae: 0.7817 - val_loss: 14.7414 - val_mae: 2.8014\n",
            "Epoch 287/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1103 - mae: 0.7480 - val_loss: 14.6081 - val_mae: 2.6981\n",
            "Epoch 288/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3171 - mae: 0.8129 - val_loss: 15.4373 - val_mae: 2.7646\n",
            "Epoch 289/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1544 - mae: 0.7551 - val_loss: 15.2878 - val_mae: 2.6923\n",
            "Epoch 290/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0784 - mae: 0.7386 - val_loss: 16.2336 - val_mae: 2.7882\n",
            "Epoch 291/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2210 - mae: 0.7469 - val_loss: 15.2720 - val_mae: 2.6985\n",
            "Epoch 292/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2967 - mae: 0.7891 - val_loss: 15.4653 - val_mae: 2.7272\n",
            "Epoch 293/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1290 - mae: 0.7454 - val_loss: 13.8622 - val_mae: 2.6879\n",
            "Epoch 294/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4443 - mae: 0.7885 - val_loss: 14.8423 - val_mae: 2.7134\n",
            "Epoch 295/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2799 - mae: 0.7673 - val_loss: 14.4250 - val_mae: 2.6541\n",
            "Epoch 296/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1346 - mae: 0.7580 - val_loss: 14.2654 - val_mae: 2.6687\n",
            "Epoch 297/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0663 - mae: 0.7255 - val_loss: 14.7622 - val_mae: 2.7950\n",
            "Epoch 298/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1711 - mae: 0.7676 - val_loss: 13.7113 - val_mae: 2.6118\n",
            "Epoch 299/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2126 - mae: 0.7507 - val_loss: 14.5922 - val_mae: 2.7102\n",
            "Epoch 300/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1679 - mae: 0.7838 - val_loss: 17.5389 - val_mae: 3.0192\n",
            "Epoch 301/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2334 - mae: 0.7753 - val_loss: 14.0608 - val_mae: 2.6930\n",
            "Epoch 302/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2949 - mae: 0.8057 - val_loss: 14.8966 - val_mae: 2.7471\n",
            "Epoch 303/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0984 - mae: 0.7526 - val_loss: 15.6706 - val_mae: 2.6765\n",
            "Epoch 304/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3059 - mae: 0.7764 - val_loss: 14.2743 - val_mae: 2.6551\n",
            "Epoch 305/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2258 - mae: 0.7712 - val_loss: 15.0302 - val_mae: 2.7171\n",
            "Epoch 306/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2773 - mae: 0.7609 - val_loss: 15.8566 - val_mae: 2.7700\n",
            "Epoch 307/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1347 - mae: 0.7450 - val_loss: 14.7499 - val_mae: 2.7084\n",
            "Epoch 308/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0231 - mae: 0.7181 - val_loss: 14.6255 - val_mae: 2.7258\n",
            "Epoch 309/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2228 - mae: 0.7471 - val_loss: 14.8568 - val_mae: 2.7182\n",
            "Epoch 310/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1307 - mae: 0.7584 - val_loss: 14.3465 - val_mae: 2.7031\n",
            "Epoch 311/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0968 - mae: 0.7251 - val_loss: 14.9905 - val_mae: 2.7431\n",
            "Epoch 312/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1315 - mae: 0.7429 - val_loss: 14.8438 - val_mae: 2.7351\n",
            "Epoch 313/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1778 - mae: 0.7498 - val_loss: 14.6906 - val_mae: 2.7030\n",
            "Epoch 314/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0700 - mae: 0.7351 - val_loss: 14.1403 - val_mae: 2.6475\n",
            "Epoch 315/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2047 - mae: 0.7464 - val_loss: 13.6925 - val_mae: 2.6106\n",
            "Epoch 316/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1436 - mae: 0.7579 - val_loss: 13.5555 - val_mae: 2.5860\n",
            "Epoch 317/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1395 - mae: 0.7385 - val_loss: 13.8342 - val_mae: 2.5579\n",
            "Epoch 318/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0534 - mae: 0.7283 - val_loss: 15.7179 - val_mae: 2.8622\n",
            "Epoch 319/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1584 - mae: 0.7679 - val_loss: 14.9515 - val_mae: 2.6839\n",
            "Epoch 320/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0667 - mae: 0.7474 - val_loss: 14.6613 - val_mae: 2.7367\n",
            "Epoch 321/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1379 - mae: 0.7263 - val_loss: 15.4355 - val_mae: 2.7716\n",
            "Epoch 322/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0672 - mae: 0.7265 - val_loss: 14.0374 - val_mae: 2.6884\n",
            "Epoch 323/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0302 - mae: 0.7304 - val_loss: 14.9850 - val_mae: 2.7629\n",
            "Epoch 324/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0857 - mae: 0.7661 - val_loss: 13.5984 - val_mae: 2.6066\n",
            "Epoch 325/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9722 - mae: 0.6994 - val_loss: 14.5344 - val_mae: 2.7039\n",
            "Epoch 326/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0577 - mae: 0.7559 - val_loss: 15.2922 - val_mae: 2.7630\n",
            "Epoch 327/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2167 - mae: 0.7600 - val_loss: 14.4059 - val_mae: 2.6350\n",
            "Epoch 328/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1270 - mae: 0.7528 - val_loss: 14.7749 - val_mae: 2.6974\n",
            "Epoch 329/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9913 - mae: 0.7172 - val_loss: 14.7678 - val_mae: 2.7289\n",
            "Epoch 330/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0656 - mae: 0.7445 - val_loss: 14.5460 - val_mae: 2.6514\n",
            "Epoch 331/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9861 - mae: 0.7279 - val_loss: 14.7043 - val_mae: 2.7999\n",
            "Epoch 332/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1205 - mae: 0.7376 - val_loss: 13.5024 - val_mae: 2.6325\n",
            "Epoch 333/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1951 - mae: 0.7496 - val_loss: 15.0399 - val_mae: 2.6967\n",
            "Epoch 334/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1631 - mae: 0.7662 - val_loss: 14.2895 - val_mae: 2.6429\n",
            "Epoch 335/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9606 - mae: 0.7338 - val_loss: 14.1725 - val_mae: 2.6547\n",
            "Epoch 336/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0426 - mae: 0.7265 - val_loss: 14.3196 - val_mae: 2.7433\n",
            "Epoch 337/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0784 - mae: 0.7482 - val_loss: 15.4389 - val_mae: 2.7668\n",
            "Epoch 338/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9110 - mae: 0.6571 - val_loss: 15.7952 - val_mae: 2.9558\n",
            "Epoch 339/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1178 - mae: 0.7178 - val_loss: 16.3560 - val_mae: 2.9710\n",
            "Epoch 340/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1540 - mae: 0.7519 - val_loss: 15.3701 - val_mae: 2.7954\n",
            "Epoch 341/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9411 - mae: 0.6809 - val_loss: 14.3653 - val_mae: 2.7851\n",
            "Epoch 342/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2858 - mae: 0.7516 - val_loss: 14.0545 - val_mae: 2.6823\n",
            "Epoch 343/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9287 - mae: 0.6829 - val_loss: 14.3259 - val_mae: 2.7315\n",
            "Epoch 344/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0440 - mae: 0.6988 - val_loss: 14.7211 - val_mae: 2.7564\n",
            "Epoch 345/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0098 - mae: 0.7029 - val_loss: 14.3821 - val_mae: 2.7166\n",
            "Epoch 346/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1386 - mae: 0.7571 - val_loss: 13.8897 - val_mae: 2.6999\n",
            "Epoch 347/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9972 - mae: 0.7261 - val_loss: 14.7688 - val_mae: 2.7123\n",
            "Epoch 348/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0312 - mae: 0.7409 - val_loss: 14.1895 - val_mae: 2.6923\n",
            "Epoch 349/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0997 - mae: 0.7288 - val_loss: 14.7945 - val_mae: 2.7247\n",
            "Epoch 350/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0579 - mae: 0.7448 - val_loss: 14.3217 - val_mae: 2.6895\n",
            "Epoch 351/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1241 - mae: 0.7567 - val_loss: 14.6290 - val_mae: 2.8223\n",
            "Epoch 352/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1879 - mae: 0.7845 - val_loss: 14.9563 - val_mae: 2.7764\n",
            "Epoch 353/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0118 - mae: 0.7238 - val_loss: 14.1480 - val_mae: 2.6549\n",
            "Epoch 354/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0145 - mae: 0.7070 - val_loss: 15.2806 - val_mae: 2.7576\n",
            "Epoch 355/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0442 - mae: 0.7252 - val_loss: 13.9919 - val_mae: 2.6808\n",
            "Epoch 356/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9740 - mae: 0.6897 - val_loss: 15.2128 - val_mae: 2.7058\n",
            "Epoch 357/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9205 - mae: 0.6805 - val_loss: 14.9448 - val_mae: 2.7729\n",
            "Epoch 358/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0863 - mae: 0.7057 - val_loss: 14.8627 - val_mae: 2.7747\n",
            "Epoch 359/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9352 - mae: 0.6903 - val_loss: 15.0182 - val_mae: 2.8135\n",
            "Epoch 360/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0061 - mae: 0.7020 - val_loss: 14.2454 - val_mae: 2.7122\n",
            "Epoch 361/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9948 - mae: 0.7128 - val_loss: 14.3932 - val_mae: 2.6812\n",
            "Epoch 362/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9354 - mae: 0.7173 - val_loss: 16.2713 - val_mae: 2.8041\n",
            "Epoch 363/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9262 - mae: 0.7000 - val_loss: 17.3863 - val_mae: 2.9997\n",
            "Epoch 364/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0262 - mae: 0.7444 - val_loss: 14.1846 - val_mae: 2.6306\n",
            "Epoch 365/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9100 - mae: 0.6736 - val_loss: 14.1421 - val_mae: 2.6898\n",
            "Epoch 366/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2005 - mae: 0.7680 - val_loss: 14.5912 - val_mae: 2.6850\n",
            "Epoch 367/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9134 - mae: 0.6988 - val_loss: 15.8347 - val_mae: 2.8249\n",
            "Epoch 368/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0362 - mae: 0.6995 - val_loss: 15.0646 - val_mae: 2.7221\n",
            "Epoch 369/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8422 - mae: 0.6841 - val_loss: 14.1552 - val_mae: 2.6457\n",
            "Epoch 370/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0177 - mae: 0.7421 - val_loss: 14.4683 - val_mae: 2.6902\n",
            "Epoch 371/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8256 - mae: 0.6460 - val_loss: 13.4765 - val_mae: 2.5909\n",
            "Epoch 372/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0429 - mae: 0.7220 - val_loss: 15.8975 - val_mae: 2.8476\n",
            "Epoch 373/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8670 - mae: 0.6805 - val_loss: 14.2956 - val_mae: 2.7090\n",
            "Epoch 374/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1769 - mae: 0.7393 - val_loss: 14.8436 - val_mae: 2.7316\n",
            "Epoch 375/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8573 - mae: 0.6735 - val_loss: 17.1892 - val_mae: 2.9194\n",
            "Epoch 376/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9858 - mae: 0.7069 - val_loss: 14.2966 - val_mae: 2.7778\n",
            "Epoch 377/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8916 - mae: 0.6761 - val_loss: 13.8017 - val_mae: 2.6017\n",
            "Epoch 378/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0479 - mae: 0.6987 - val_loss: 13.8731 - val_mae: 2.6869\n",
            "Epoch 379/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8159 - mae: 0.6784 - val_loss: 15.0571 - val_mae: 2.7658\n",
            "Epoch 380/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8740 - mae: 0.6689 - val_loss: 15.1708 - val_mae: 2.7498\n",
            "Epoch 381/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0865 - mae: 0.7386 - val_loss: 16.5947 - val_mae: 2.8293\n",
            "Epoch 382/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8978 - mae: 0.6933 - val_loss: 14.9965 - val_mae: 2.7045\n",
            "Epoch 383/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9269 - mae: 0.7019 - val_loss: 14.5333 - val_mae: 2.7825\n",
            "Epoch 384/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9215 - mae: 0.6548 - val_loss: 14.4854 - val_mae: 2.6903\n",
            "Epoch 385/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0133 - mae: 0.6865 - val_loss: 14.6701 - val_mae: 2.7108\n",
            "Epoch 386/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9211 - mae: 0.6870 - val_loss: 14.9900 - val_mae: 2.8109\n",
            "Epoch 387/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8987 - mae: 0.6812 - val_loss: 14.3145 - val_mae: 2.7550\n",
            "Epoch 388/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0169 - mae: 0.7036 - val_loss: 15.4798 - val_mae: 2.8171\n",
            "Epoch 389/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9776 - mae: 0.6850 - val_loss: 15.7702 - val_mae: 2.8511\n",
            "Epoch 390/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9165 - mae: 0.6854 - val_loss: 14.8882 - val_mae: 2.7867\n",
            "Epoch 391/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8396 - mae: 0.6474 - val_loss: 14.7203 - val_mae: 2.8121\n",
            "Epoch 392/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1320 - mae: 0.6870 - val_loss: 15.1642 - val_mae: 2.7732\n",
            "Epoch 393/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8584 - mae: 0.6569 - val_loss: 14.3332 - val_mae: 2.7248\n",
            "Epoch 394/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9662 - mae: 0.7113 - val_loss: 14.1996 - val_mae: 2.7192\n",
            "Epoch 395/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9372 - mae: 0.6949 - val_loss: 14.7877 - val_mae: 2.7752\n",
            "Epoch 396/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0489 - mae: 0.6867 - val_loss: 14.1673 - val_mae: 2.6944\n",
            "Epoch 397/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8636 - mae: 0.6636 - val_loss: 14.5331 - val_mae: 2.6958\n",
            "Epoch 398/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9895 - mae: 0.6972 - val_loss: 14.3123 - val_mae: 2.7230\n",
            "Epoch 399/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9558 - mae: 0.6794 - val_loss: 16.1366 - val_mae: 2.8918\n",
            "Epoch 400/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9151 - mae: 0.6765 - val_loss: 14.7639 - val_mae: 2.7151\n",
            "Epoch 401/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9906 - mae: 0.6797 - val_loss: 15.4475 - val_mae: 2.7868\n",
            "Epoch 402/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8223 - mae: 0.6440 - val_loss: 15.2233 - val_mae: 2.8346\n",
            "Epoch 403/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8864 - mae: 0.6580 - val_loss: 14.5519 - val_mae: 2.7824\n",
            "Epoch 404/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8876 - mae: 0.6600 - val_loss: 16.5410 - val_mae: 2.8992\n",
            "Epoch 405/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0103 - mae: 0.6731 - val_loss: 16.3120 - val_mae: 2.8145\n",
            "Epoch 406/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9394 - mae: 0.7126 - val_loss: 16.7846 - val_mae: 2.9573\n",
            "Epoch 407/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8181 - mae: 0.6469 - val_loss: 16.3007 - val_mae: 2.9090\n",
            "Epoch 408/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8891 - mae: 0.6670 - val_loss: 15.6824 - val_mae: 2.8101\n",
            "Epoch 409/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8900 - mae: 0.7147 - val_loss: 16.3813 - val_mae: 2.9191\n",
            "Epoch 410/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9977 - mae: 0.6717 - val_loss: 15.5450 - val_mae: 2.7957\n",
            "Epoch 411/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8406 - mae: 0.6515 - val_loss: 16.4923 - val_mae: 2.9094\n",
            "Epoch 412/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9149 - mae: 0.6650 - val_loss: 16.9225 - val_mae: 2.9573\n",
            "Epoch 413/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9544 - mae: 0.6618 - val_loss: 15.4455 - val_mae: 2.8079\n",
            "Epoch 414/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9535 - mae: 0.6940 - val_loss: 15.9596 - val_mae: 2.7928\n",
            "Epoch 415/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8821 - mae: 0.6760 - val_loss: 16.4393 - val_mae: 2.8295\n",
            "Epoch 416/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8430 - mae: 0.6783 - val_loss: 17.0164 - val_mae: 2.9361\n",
            "Epoch 417/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7984 - mae: 0.6692 - val_loss: 16.3020 - val_mae: 2.8816\n",
            "Epoch 418/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9037 - mae: 0.6676 - val_loss: 16.0158 - val_mae: 2.7895\n",
            "Epoch 419/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8670 - mae: 0.6645 - val_loss: 15.7719 - val_mae: 2.7826\n",
            "Epoch 420/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7885 - mae: 0.6478 - val_loss: 15.5387 - val_mae: 2.7714\n",
            "Epoch 421/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8438 - mae: 0.6471 - val_loss: 15.8918 - val_mae: 2.8535\n",
            "Epoch 422/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8221 - mae: 0.6393 - val_loss: 16.0788 - val_mae: 2.8504\n",
            "Epoch 423/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9580 - mae: 0.6621 - val_loss: 16.9354 - val_mae: 2.8783\n",
            "Epoch 424/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9006 - mae: 0.6752 - val_loss: 15.4303 - val_mae: 2.8265\n",
            "Epoch 425/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0214 - mae: 0.6620 - val_loss: 14.9187 - val_mae: 2.7974\n",
            "Epoch 426/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7854 - mae: 0.6367 - val_loss: 15.9113 - val_mae: 2.8500\n",
            "Epoch 427/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9482 - mae: 0.6600 - val_loss: 15.3092 - val_mae: 2.7982\n",
            "Epoch 428/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9298 - mae: 0.6843 - val_loss: 15.5777 - val_mae: 2.7629\n",
            "Epoch 429/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7560 - mae: 0.6100 - val_loss: 17.1715 - val_mae: 2.9309\n",
            "Epoch 430/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9530 - mae: 0.6807 - val_loss: 15.2779 - val_mae: 2.8003\n",
            "Epoch 431/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0406 - mae: 0.6911 - val_loss: 15.6720 - val_mae: 2.8570\n",
            "Epoch 432/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8660 - mae: 0.6791 - val_loss: 16.1977 - val_mae: 2.9196\n",
            "Epoch 433/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8379 - mae: 0.6388 - val_loss: 18.7174 - val_mae: 3.0733\n",
            "Epoch 434/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9406 - mae: 0.6520 - val_loss: 16.7994 - val_mae: 2.9110\n",
            "Epoch 435/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7763 - mae: 0.6373 - val_loss: 15.5417 - val_mae: 2.8057\n",
            "Epoch 436/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8962 - mae: 0.6532 - val_loss: 16.0977 - val_mae: 2.8349\n",
            "Epoch 437/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8726 - mae: 0.6763 - val_loss: 14.9895 - val_mae: 2.7652\n",
            "Epoch 438/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7633 - mae: 0.6246 - val_loss: 15.2413 - val_mae: 2.8093\n",
            "Epoch 439/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8773 - mae: 0.6649 - val_loss: 16.1435 - val_mae: 2.8372\n",
            "Epoch 440/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8383 - mae: 0.6498 - val_loss: 17.1270 - val_mae: 2.9068\n",
            "Epoch 441/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9209 - mae: 0.6551 - val_loss: 16.7574 - val_mae: 2.9070\n",
            "Epoch 442/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8193 - mae: 0.6648 - val_loss: 15.6145 - val_mae: 2.7498\n",
            "Epoch 443/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9217 - mae: 0.6544 - val_loss: 16.1350 - val_mae: 2.8723\n",
            "Epoch 444/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8071 - mae: 0.6503 - val_loss: 16.3514 - val_mae: 2.8302\n",
            "Epoch 445/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8170 - mae: 0.6355 - val_loss: 15.5243 - val_mae: 2.7409\n",
            "Epoch 446/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7990 - mae: 0.6431 - val_loss: 14.7739 - val_mae: 2.7944\n",
            "Epoch 447/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8749 - mae: 0.6237 - val_loss: 16.8420 - val_mae: 2.9107\n",
            "Epoch 448/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7832 - mae: 0.6345 - val_loss: 15.4264 - val_mae: 2.8709\n",
            "Epoch 449/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7907 - mae: 0.6219 - val_loss: 15.6864 - val_mae: 2.8111\n",
            "Epoch 450/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7619 - mae: 0.6179 - val_loss: 15.9753 - val_mae: 2.8116\n",
            "Epoch 451/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7595 - mae: 0.6288 - val_loss: 16.1682 - val_mae: 2.8541\n",
            "Epoch 452/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8148 - mae: 0.6080 - val_loss: 15.9318 - val_mae: 2.8292\n",
            "Epoch 453/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7971 - mae: 0.6287 - val_loss: 16.1295 - val_mae: 2.8765\n",
            "Epoch 454/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7917 - mae: 0.6583 - val_loss: 15.3577 - val_mae: 2.8192\n",
            "Epoch 455/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8152 - mae: 0.6385 - val_loss: 16.0945 - val_mae: 2.8127\n",
            "Epoch 456/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9223 - mae: 0.6412 - val_loss: 15.4749 - val_mae: 2.8144\n",
            "Epoch 457/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8692 - mae: 0.6449 - val_loss: 17.0402 - val_mae: 2.9665\n",
            "Epoch 458/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8921 - mae: 0.6779 - val_loss: 16.0271 - val_mae: 2.8629\n",
            "Epoch 459/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8793 - mae: 0.6447 - val_loss: 16.4694 - val_mae: 2.9358\n",
            "Epoch 460/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7569 - mae: 0.6154 - val_loss: 15.7404 - val_mae: 2.8263\n",
            "Epoch 461/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8156 - mae: 0.6363 - val_loss: 15.3867 - val_mae: 2.8632\n",
            "Epoch 462/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8892 - mae: 0.6591 - val_loss: 15.1177 - val_mae: 2.7593\n",
            "Epoch 463/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.6895 - mae: 0.6084 - val_loss: 15.3751 - val_mae: 2.8082\n",
            "Epoch 464/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8150 - mae: 0.6737 - val_loss: 15.7352 - val_mae: 2.8205\n",
            "Epoch 465/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8203 - mae: 0.6280 - val_loss: 16.7764 - val_mae: 2.9368\n",
            "Epoch 466/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8040 - mae: 0.6208 - val_loss: 16.1238 - val_mae: 2.8322\n",
            "Epoch 467/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7905 - mae: 0.6439 - val_loss: 16.0846 - val_mae: 2.8865\n",
            "Epoch 468/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7900 - mae: 0.6652 - val_loss: 15.6720 - val_mae: 2.7629\n",
            "Epoch 469/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7835 - mae: 0.6586 - val_loss: 15.9622 - val_mae: 2.7827\n",
            "Epoch 470/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7673 - mae: 0.6329 - val_loss: 16.1111 - val_mae: 2.8547\n",
            "Epoch 471/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7525 - mae: 0.6319 - val_loss: 15.5566 - val_mae: 2.7830\n",
            "Epoch 472/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.6899 - mae: 0.6150 - val_loss: 15.1843 - val_mae: 2.7830\n",
            "Epoch 473/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8081 - mae: 0.6281 - val_loss: 16.2653 - val_mae: 2.8776\n",
            "Epoch 474/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8428 - mae: 0.6453 - val_loss: 15.9746 - val_mae: 2.8893\n",
            "Epoch 475/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8448 - mae: 0.6338 - val_loss: 16.6508 - val_mae: 2.8838\n",
            "Epoch 476/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7510 - mae: 0.6322 - val_loss: 15.6922 - val_mae: 2.8209\n",
            "Epoch 477/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.6712 - mae: 0.5934 - val_loss: 15.4988 - val_mae: 2.8409\n",
            "Epoch 478/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8814 - mae: 0.6456 - val_loss: 16.5284 - val_mae: 2.8702\n",
            "Epoch 479/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7804 - mae: 0.6258 - val_loss: 15.4113 - val_mae: 2.7908\n",
            "Epoch 480/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7625 - mae: 0.6020 - val_loss: 15.9415 - val_mae: 2.8181\n",
            "Epoch 481/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7963 - mae: 0.6237 - val_loss: 17.2021 - val_mae: 3.0088\n",
            "Epoch 482/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7600 - mae: 0.6273 - val_loss: 17.2534 - val_mae: 2.9113\n",
            "Epoch 483/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7649 - mae: 0.6414 - val_loss: 16.2658 - val_mae: 2.8323\n",
            "Epoch 484/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8049 - mae: 0.6211 - val_loss: 16.7678 - val_mae: 2.9288\n",
            "Epoch 485/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.6958 - mae: 0.6327 - val_loss: 16.6872 - val_mae: 2.9081\n",
            "Epoch 486/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7767 - mae: 0.6003 - val_loss: 16.0834 - val_mae: 2.8615\n",
            "Epoch 487/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8013 - mae: 0.6207 - val_loss: 16.0884 - val_mae: 2.7988\n",
            "Epoch 488/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7937 - mae: 0.6418 - val_loss: 15.7254 - val_mae: 2.8208\n",
            "Epoch 489/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8042 - mae: 0.6306 - val_loss: 15.7354 - val_mae: 2.7711\n",
            "Epoch 490/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8391 - mae: 0.6365 - val_loss: 16.1082 - val_mae: 2.8001\n",
            "Epoch 491/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.6590 - mae: 0.5775 - val_loss: 17.3523 - val_mae: 2.9408\n",
            "Epoch 492/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7592 - mae: 0.6253 - val_loss: 16.1107 - val_mae: 2.8043\n",
            "Epoch 493/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7239 - mae: 0.6298 - val_loss: 18.6254 - val_mae: 3.1258\n",
            "Epoch 494/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8839 - mae: 0.6308 - val_loss: 16.4261 - val_mae: 2.7688\n",
            "Epoch 495/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8243 - mae: 0.6260 - val_loss: 15.8991 - val_mae: 2.8400\n",
            "Epoch 496/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7371 - mae: 0.6282 - val_loss: 16.2636 - val_mae: 2.7966\n",
            "Epoch 497/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.7213 - mae: 0.6383 - val_loss: 16.5134 - val_mae: 2.8427\n",
            "Epoch 498/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.6654 - mae: 0.6066 - val_loss: 16.9568 - val_mae: 2.8886\n",
            "Epoch 499/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9490 - mae: 0.6202 - val_loss: 16.3258 - val_mae: 2.8871\n",
            "Epoch 500/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.6640 - mae: 0.5772 - val_loss: 15.7578 - val_mae: 2.8422\n",
            "processing fold # 3\n",
            "Epoch 1/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 172.9893 - mae: 9.7962 - val_loss: 51.6674 - val_mae: 5.1868\n",
            "Epoch 2/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 26.4707 - mae: 3.5193 - val_loss: 31.2858 - val_mae: 3.7193\n",
            "Epoch 3/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 19.6291 - mae: 2.9513 - val_loss: 25.9591 - val_mae: 3.5313\n",
            "Epoch 4/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 16.0696 - mae: 2.6905 - val_loss: 23.5496 - val_mae: 3.2721\n",
            "Epoch 5/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 14.5219 - mae: 2.5096 - val_loss: 22.3513 - val_mae: 2.9697\n",
            "Epoch 6/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 13.5536 - mae: 2.4488 - val_loss: 19.3622 - val_mae: 2.8665\n",
            "Epoch 7/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 12.3238 - mae: 2.3778 - val_loss: 19.8772 - val_mae: 2.8674\n",
            "Epoch 8/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 11.7188 - mae: 2.3436 - val_loss: 21.5338 - val_mae: 3.0490\n",
            "Epoch 9/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 11.9176 - mae: 2.2110 - val_loss: 20.4430 - val_mae: 2.9037\n",
            "Epoch 10/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 10.6521 - mae: 2.1600 - val_loss: 19.3123 - val_mae: 2.8213\n",
            "Epoch 11/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 10.9656 - mae: 2.1142 - val_loss: 16.1213 - val_mae: 2.7788\n",
            "Epoch 12/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 10.0047 - mae: 2.1277 - val_loss: 16.0300 - val_mae: 2.6641\n",
            "Epoch 13/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 10.0080 - mae: 2.0853 - val_loss: 18.1863 - val_mae: 2.7743\n",
            "Epoch 14/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 10.0769 - mae: 2.0683 - val_loss: 17.6303 - val_mae: 2.7376\n",
            "Epoch 15/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 10.2027 - mae: 2.0065 - val_loss: 17.0784 - val_mae: 2.7633\n",
            "Epoch 16/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 9.2252 - mae: 1.9691 - val_loss: 19.0710 - val_mae: 2.9487\n",
            "Epoch 17/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 9.2957 - mae: 2.0063 - val_loss: 15.1260 - val_mae: 2.6547\n",
            "Epoch 18/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 9.5702 - mae: 2.0244 - val_loss: 14.6865 - val_mae: 2.5035\n",
            "Epoch 19/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 9.4524 - mae: 2.0195 - val_loss: 13.5533 - val_mae: 2.4424\n",
            "Epoch 20/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 8.9447 - mae: 1.9923 - val_loss: 13.3895 - val_mae: 2.5173\n",
            "Epoch 21/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 8.8186 - mae: 1.9395 - val_loss: 15.9061 - val_mae: 2.7684\n",
            "Epoch 22/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 8.7115 - mae: 1.9125 - val_loss: 13.2966 - val_mae: 2.5275\n",
            "Epoch 23/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 8.4687 - mae: 1.8979 - val_loss: 13.1846 - val_mae: 2.4129\n",
            "Epoch 24/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 8.6480 - mae: 1.8756 - val_loss: 12.8791 - val_mae: 2.3882\n",
            "Epoch 25/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 8.2139 - mae: 1.8834 - val_loss: 13.3287 - val_mae: 2.5247\n",
            "Epoch 26/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 8.1560 - mae: 1.8930 - val_loss: 12.0924 - val_mae: 2.3480\n",
            "Epoch 27/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 8.6441 - mae: 1.8489 - val_loss: 12.9299 - val_mae: 2.4253\n",
            "Epoch 28/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.7879 - mae: 1.8175 - val_loss: 13.2196 - val_mae: 2.4793\n",
            "Epoch 29/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.8707 - mae: 1.8821 - val_loss: 15.2145 - val_mae: 2.7349\n",
            "Epoch 30/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.2825 - mae: 1.7379 - val_loss: 16.9337 - val_mae: 2.8411\n",
            "Epoch 31/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.6266 - mae: 1.8754 - val_loss: 13.2497 - val_mae: 2.5053\n",
            "Epoch 32/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.5342 - mae: 1.8016 - val_loss: 12.5563 - val_mae: 2.4662\n",
            "Epoch 33/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.5089 - mae: 1.7831 - val_loss: 12.2430 - val_mae: 2.3201\n",
            "Epoch 34/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.5877 - mae: 1.8278 - val_loss: 12.8638 - val_mae: 2.4254\n",
            "Epoch 35/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.4912 - mae: 1.7859 - val_loss: 11.9705 - val_mae: 2.3209\n",
            "Epoch 36/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.7633 - mae: 1.7079 - val_loss: 11.3582 - val_mae: 2.3521\n",
            "Epoch 37/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.4796 - mae: 1.7668 - val_loss: 11.3175 - val_mae: 2.2821\n",
            "Epoch 38/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.7222 - mae: 1.6784 - val_loss: 12.4702 - val_mae: 2.4441\n",
            "Epoch 39/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.0629 - mae: 1.7086 - val_loss: 11.6940 - val_mae: 2.3245\n",
            "Epoch 40/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.0160 - mae: 1.6797 - val_loss: 12.0028 - val_mae: 2.3659\n",
            "Epoch 41/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.5573 - mae: 1.6472 - val_loss: 14.4134 - val_mae: 2.6712\n",
            "Epoch 42/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 7.1098 - mae: 1.6906 - val_loss: 11.5720 - val_mae: 2.2924\n",
            "Epoch 43/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.4756 - mae: 1.6205 - val_loss: 11.9676 - val_mae: 2.2918\n",
            "Epoch 44/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.6466 - mae: 1.6934 - val_loss: 14.0640 - val_mae: 2.5201\n",
            "Epoch 45/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.6847 - mae: 1.6469 - val_loss: 12.1149 - val_mae: 2.3802\n",
            "Epoch 46/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.4373 - mae: 1.6719 - val_loss: 11.8772 - val_mae: 2.3630\n",
            "Epoch 47/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.9790 - mae: 1.6009 - val_loss: 11.4271 - val_mae: 2.3477\n",
            "Epoch 48/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.0889 - mae: 1.5839 - val_loss: 13.6361 - val_mae: 2.6841\n",
            "Epoch 49/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.4759 - mae: 1.6245 - val_loss: 12.2857 - val_mae: 2.4103\n",
            "Epoch 50/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.1250 - mae: 1.6075 - val_loss: 12.0367 - val_mae: 2.3940\n",
            "Epoch 51/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.0903 - mae: 1.6015 - val_loss: 13.7057 - val_mae: 2.5927\n",
            "Epoch 52/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.8319 - mae: 1.5407 - val_loss: 12.1180 - val_mae: 2.4732\n",
            "Epoch 53/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.1753 - mae: 1.6251 - val_loss: 11.2907 - val_mae: 2.2725\n",
            "Epoch 54/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.0009 - mae: 1.5267 - val_loss: 11.6576 - val_mae: 2.3150\n",
            "Epoch 55/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.0439 - mae: 1.5566 - val_loss: 11.4203 - val_mae: 2.2842\n",
            "Epoch 56/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.9220 - mae: 1.5547 - val_loss: 10.9118 - val_mae: 2.2774\n",
            "Epoch 57/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 6.0186 - mae: 1.6039 - val_loss: 10.7419 - val_mae: 2.2274\n",
            "Epoch 58/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.7030 - mae: 1.5339 - val_loss: 12.8347 - val_mae: 2.4393\n",
            "Epoch 59/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.8065 - mae: 1.5481 - val_loss: 10.3091 - val_mae: 2.1905\n",
            "Epoch 60/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.9138 - mae: 1.5411 - val_loss: 11.6786 - val_mae: 2.3275\n",
            "Epoch 61/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.3553 - mae: 1.5042 - val_loss: 11.1065 - val_mae: 2.3550\n",
            "Epoch 62/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.5444 - mae: 1.5053 - val_loss: 11.4513 - val_mae: 2.3554\n",
            "Epoch 63/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.4481 - mae: 1.4877 - val_loss: 11.5225 - val_mae: 2.4225\n",
            "Epoch 64/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.6976 - mae: 1.5088 - val_loss: 10.8351 - val_mae: 2.3134\n",
            "Epoch 65/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.3742 - mae: 1.4950 - val_loss: 11.4292 - val_mae: 2.4286\n",
            "Epoch 66/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.6055 - mae: 1.4881 - val_loss: 14.0678 - val_mae: 2.7412\n",
            "Epoch 67/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.7874 - mae: 1.4876 - val_loss: 12.4806 - val_mae: 2.6076\n",
            "Epoch 68/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.4105 - mae: 1.4780 - val_loss: 10.8957 - val_mae: 2.2648\n",
            "Epoch 69/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.6816 - mae: 1.4750 - val_loss: 11.0858 - val_mae: 2.2945\n",
            "Epoch 70/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.4581 - mae: 1.4830 - val_loss: 11.0222 - val_mae: 2.3406\n",
            "Epoch 71/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.2174 - mae: 1.4573 - val_loss: 11.9224 - val_mae: 2.3934\n",
            "Epoch 72/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.4819 - mae: 1.4919 - val_loss: 14.3695 - val_mae: 2.8233\n",
            "Epoch 73/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.3024 - mae: 1.4576 - val_loss: 12.5750 - val_mae: 2.5526\n",
            "Epoch 74/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.1478 - mae: 1.4805 - val_loss: 13.0399 - val_mae: 2.6469\n",
            "Epoch 75/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.1210 - mae: 1.4727 - val_loss: 12.2796 - val_mae: 2.5367\n",
            "Epoch 76/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.8114 - mae: 1.4237 - val_loss: 15.4981 - val_mae: 2.9250\n",
            "Epoch 77/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.1055 - mae: 1.4500 - val_loss: 10.9967 - val_mae: 2.3027\n",
            "Epoch 78/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.2605 - mae: 1.4237 - val_loss: 11.2242 - val_mae: 2.3818\n",
            "Epoch 79/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.7785 - mae: 1.3539 - val_loss: 14.9191 - val_mae: 2.8057\n",
            "Epoch 80/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.0996 - mae: 1.4442 - val_loss: 12.7038 - val_mae: 2.5937\n",
            "Epoch 81/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.0379 - mae: 1.4586 - val_loss: 10.9814 - val_mae: 2.2657\n",
            "Epoch 82/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 5.0347 - mae: 1.4263 - val_loss: 11.0271 - val_mae: 2.3046\n",
            "Epoch 83/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.5736 - mae: 1.4051 - val_loss: 12.7637 - val_mae: 2.6187\n",
            "Epoch 84/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.6978 - mae: 1.4434 - val_loss: 10.9742 - val_mae: 2.3082\n",
            "Epoch 85/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.7614 - mae: 1.3796 - val_loss: 10.7056 - val_mae: 2.2490\n",
            "Epoch 86/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.3896 - mae: 1.3800 - val_loss: 10.8947 - val_mae: 2.2978\n",
            "Epoch 87/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.6401 - mae: 1.3663 - val_loss: 11.4215 - val_mae: 2.4675\n",
            "Epoch 88/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.5114 - mae: 1.3647 - val_loss: 12.9335 - val_mae: 2.5102\n",
            "Epoch 89/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.9022 - mae: 1.3876 - val_loss: 12.4805 - val_mae: 2.4895\n",
            "Epoch 90/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.9666 - mae: 1.4021 - val_loss: 11.9994 - val_mae: 2.4501\n",
            "Epoch 91/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.5983 - mae: 1.3807 - val_loss: 11.3964 - val_mae: 2.3125\n",
            "Epoch 92/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.7175 - mae: 1.3153 - val_loss: 11.6565 - val_mae: 2.4155\n",
            "Epoch 93/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.6068 - mae: 1.4222 - val_loss: 11.3757 - val_mae: 2.3358\n",
            "Epoch 94/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.8338 - mae: 1.4260 - val_loss: 11.6647 - val_mae: 2.3885\n",
            "Epoch 95/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.6494 - mae: 1.3151 - val_loss: 13.4849 - val_mae: 2.6767\n",
            "Epoch 96/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.7427 - mae: 1.3346 - val_loss: 11.4822 - val_mae: 2.3921\n",
            "Epoch 97/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.3370 - mae: 1.3173 - val_loss: 11.0776 - val_mae: 2.3608\n",
            "Epoch 98/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.7638 - mae: 1.3029 - val_loss: 14.3285 - val_mae: 2.7169\n",
            "Epoch 99/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.6408 - mae: 1.3318 - val_loss: 11.8230 - val_mae: 2.4492\n",
            "Epoch 100/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.7782 - mae: 1.3314 - val_loss: 10.9787 - val_mae: 2.3086\n",
            "Epoch 101/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.5310 - mae: 1.2924 - val_loss: 12.9707 - val_mae: 2.6393\n",
            "Epoch 102/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.0558 - mae: 1.2960 - val_loss: 11.9854 - val_mae: 2.4335\n",
            "Epoch 103/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.4001 - mae: 1.2585 - val_loss: 11.0793 - val_mae: 2.2972\n",
            "Epoch 104/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.2708 - mae: 1.3138 - val_loss: 11.2952 - val_mae: 2.4319\n",
            "Epoch 105/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.3235 - mae: 1.2776 - val_loss: 11.6106 - val_mae: 2.4738\n",
            "Epoch 106/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.4212 - mae: 1.3206 - val_loss: 11.0696 - val_mae: 2.3040\n",
            "Epoch 107/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.2831 - mae: 1.2797 - val_loss: 11.0301 - val_mae: 2.2793\n",
            "Epoch 108/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.2636 - mae: 1.3459 - val_loss: 11.8957 - val_mae: 2.4703\n",
            "Epoch 109/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.2567 - mae: 1.3130 - val_loss: 11.5046 - val_mae: 2.3114\n",
            "Epoch 110/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.1994 - mae: 1.2323 - val_loss: 12.5893 - val_mae: 2.4941\n",
            "Epoch 111/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.1706 - mae: 1.3028 - val_loss: 11.6933 - val_mae: 2.4245\n",
            "Epoch 112/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.1696 - mae: 1.2776 - val_loss: 12.3823 - val_mae: 2.4349\n",
            "Epoch 113/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.3429 - mae: 1.2737 - val_loss: 11.0078 - val_mae: 2.3659\n",
            "Epoch 114/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.0803 - mae: 1.2621 - val_loss: 11.7085 - val_mae: 2.4432\n",
            "Epoch 115/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.1084 - mae: 1.2695 - val_loss: 12.1171 - val_mae: 2.4727\n",
            "Epoch 116/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.0708 - mae: 1.2002 - val_loss: 11.5085 - val_mae: 2.3893\n",
            "Epoch 117/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.8708 - mae: 1.2770 - val_loss: 11.8541 - val_mae: 2.4145\n",
            "Epoch 118/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.9086 - mae: 1.2500 - val_loss: 11.2890 - val_mae: 2.4130\n",
            "Epoch 119/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.8854 - mae: 1.2202 - val_loss: 11.2292 - val_mae: 2.3729\n",
            "Epoch 120/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.0413 - mae: 1.2558 - val_loss: 11.0546 - val_mae: 2.3798\n",
            "Epoch 121/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.9786 - mae: 1.2466 - val_loss: 11.1962 - val_mae: 2.3422\n",
            "Epoch 122/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.0289 - mae: 1.2066 - val_loss: 10.7017 - val_mae: 2.3381\n",
            "Epoch 123/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.6612 - mae: 1.2152 - val_loss: 11.4009 - val_mae: 2.3687\n",
            "Epoch 124/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.9993 - mae: 1.2760 - val_loss: 11.3300 - val_mae: 2.3625\n",
            "Epoch 125/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.7247 - mae: 1.2245 - val_loss: 11.2114 - val_mae: 2.3815\n",
            "Epoch 126/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.9693 - mae: 1.2396 - val_loss: 11.7519 - val_mae: 2.3995\n",
            "Epoch 127/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 4.2675 - mae: 1.2141 - val_loss: 12.7470 - val_mae: 2.5791\n",
            "Epoch 128/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.9891 - mae: 1.2262 - val_loss: 12.1691 - val_mae: 2.4357\n",
            "Epoch 129/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.7467 - mae: 1.2023 - val_loss: 13.7195 - val_mae: 2.6908\n",
            "Epoch 130/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.7452 - mae: 1.2140 - val_loss: 13.3054 - val_mae: 2.5611\n",
            "Epoch 131/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.7044 - mae: 1.2142 - val_loss: 12.2771 - val_mae: 2.4679\n",
            "Epoch 132/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.5546 - mae: 1.1848 - val_loss: 12.6439 - val_mae: 2.5443\n",
            "Epoch 133/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.5661 - mae: 1.1590 - val_loss: 11.5583 - val_mae: 2.3319\n",
            "Epoch 134/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.7986 - mae: 1.2105 - val_loss: 11.9578 - val_mae: 2.4359\n",
            "Epoch 135/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.8062 - mae: 1.2165 - val_loss: 12.0383 - val_mae: 2.4786\n",
            "Epoch 136/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.8181 - mae: 1.2472 - val_loss: 11.6566 - val_mae: 2.4140\n",
            "Epoch 137/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.5419 - mae: 1.1819 - val_loss: 11.6470 - val_mae: 2.4080\n",
            "Epoch 138/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.3539 - mae: 1.2151 - val_loss: 18.9185 - val_mae: 3.1229\n",
            "Epoch 139/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.9539 - mae: 1.2244 - val_loss: 11.7911 - val_mae: 2.3904\n",
            "Epoch 140/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.6317 - mae: 1.1714 - val_loss: 11.5325 - val_mae: 2.4237\n",
            "Epoch 141/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.7261 - mae: 1.2136 - val_loss: 11.0456 - val_mae: 2.3130\n",
            "Epoch 142/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.8855 - mae: 1.1728 - val_loss: 12.2926 - val_mae: 2.4773\n",
            "Epoch 143/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.5560 - mae: 1.1465 - val_loss: 12.2526 - val_mae: 2.4784\n",
            "Epoch 144/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.5792 - mae: 1.1924 - val_loss: 13.7416 - val_mae: 2.6415\n",
            "Epoch 145/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.8333 - mae: 1.2049 - val_loss: 15.0262 - val_mae: 2.9005\n",
            "Epoch 146/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.2077 - mae: 1.1437 - val_loss: 12.6549 - val_mae: 2.5921\n",
            "Epoch 147/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.7080 - mae: 1.1650 - val_loss: 12.5544 - val_mae: 2.5715\n",
            "Epoch 148/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.6655 - mae: 1.1680 - val_loss: 12.7028 - val_mae: 2.5255\n",
            "Epoch 149/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.5586 - mae: 1.1483 - val_loss: 12.0773 - val_mae: 2.4423\n",
            "Epoch 150/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.2441 - mae: 1.1085 - val_loss: 14.5971 - val_mae: 2.8535\n",
            "Epoch 151/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.4378 - mae: 1.1633 - val_loss: 11.5439 - val_mae: 2.4249\n",
            "Epoch 152/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.2359 - mae: 1.1821 - val_loss: 11.4086 - val_mae: 2.3982\n",
            "Epoch 153/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.3387 - mae: 1.1577 - val_loss: 12.5820 - val_mae: 2.5243\n",
            "Epoch 154/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.2396 - mae: 1.1428 - val_loss: 11.2517 - val_mae: 2.3374\n",
            "Epoch 155/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.1341 - mae: 1.1239 - val_loss: 11.3193 - val_mae: 2.3291\n",
            "Epoch 156/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.3609 - mae: 1.1638 - val_loss: 14.5273 - val_mae: 2.7716\n",
            "Epoch 157/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.1673 - mae: 1.1764 - val_loss: 12.4145 - val_mae: 2.4746\n",
            "Epoch 158/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.3533 - mae: 1.1316 - val_loss: 11.9559 - val_mae: 2.4267\n",
            "Epoch 159/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.1829 - mae: 1.0961 - val_loss: 15.7031 - val_mae: 2.9084\n",
            "Epoch 160/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.3156 - mae: 1.1720 - val_loss: 12.9302 - val_mae: 2.5938\n",
            "Epoch 161/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.2906 - mae: 1.1153 - val_loss: 12.6381 - val_mae: 2.5040\n",
            "Epoch 162/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.3359 - mae: 1.1681 - val_loss: 16.1393 - val_mae: 2.9846\n",
            "Epoch 163/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.2079 - mae: 1.1347 - val_loss: 12.2252 - val_mae: 2.5318\n",
            "Epoch 164/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.3462 - mae: 1.1667 - val_loss: 12.4457 - val_mae: 2.4835\n",
            "Epoch 165/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.0778 - mae: 1.1040 - val_loss: 12.0994 - val_mae: 2.5119\n",
            "Epoch 166/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.3380 - mae: 1.1134 - val_loss: 11.6847 - val_mae: 2.3900\n",
            "Epoch 167/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.1417 - mae: 1.1514 - val_loss: 14.4719 - val_mae: 2.7740\n",
            "Epoch 168/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.2323 - mae: 1.1371 - val_loss: 11.7705 - val_mae: 2.4226\n",
            "Epoch 169/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.0132 - mae: 1.0893 - val_loss: 11.9152 - val_mae: 2.4128\n",
            "Epoch 170/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.1037 - mae: 1.1152 - val_loss: 13.6913 - val_mae: 2.6559\n",
            "Epoch 171/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.1621 - mae: 1.1189 - val_loss: 12.9843 - val_mae: 2.4800\n",
            "Epoch 172/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.9477 - mae: 1.0907 - val_loss: 11.8244 - val_mae: 2.4827\n",
            "Epoch 173/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.1806 - mae: 1.1626 - val_loss: 12.6547 - val_mae: 2.5237\n",
            "Epoch 174/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.2640 - mae: 1.1030 - val_loss: 17.5246 - val_mae: 3.1201\n",
            "Epoch 175/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.2880 - mae: 1.1104 - val_loss: 12.1309 - val_mae: 2.4683\n",
            "Epoch 176/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.0726 - mae: 1.1131 - val_loss: 12.9465 - val_mae: 2.6428\n",
            "Epoch 177/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.2020 - mae: 1.1145 - val_loss: 11.6501 - val_mae: 2.4141\n",
            "Epoch 178/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.1177 - mae: 1.1201 - val_loss: 12.9885 - val_mae: 2.5873\n",
            "Epoch 179/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.1501 - mae: 1.0958 - val_loss: 11.9780 - val_mae: 2.3840\n",
            "Epoch 180/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.9039 - mae: 1.0927 - val_loss: 12.8842 - val_mae: 2.6262\n",
            "Epoch 181/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.0479 - mae: 1.0401 - val_loss: 12.6849 - val_mae: 2.5440\n",
            "Epoch 182/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.0017 - mae: 1.1172 - val_loss: 12.5861 - val_mae: 2.5506\n",
            "Epoch 183/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.0932 - mae: 1.0831 - val_loss: 11.5183 - val_mae: 2.3945\n",
            "Epoch 184/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.1224 - mae: 1.0887 - val_loss: 12.4497 - val_mae: 2.4963\n",
            "Epoch 185/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.9395 - mae: 1.0770 - val_loss: 11.6904 - val_mae: 2.3780\n",
            "Epoch 186/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 3.0202 - mae: 1.1315 - val_loss: 13.3748 - val_mae: 2.6502\n",
            "Epoch 187/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.8426 - mae: 1.0633 - val_loss: 12.0607 - val_mae: 2.4674\n",
            "Epoch 188/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.7625 - mae: 1.0626 - val_loss: 11.7441 - val_mae: 2.4176\n",
            "Epoch 189/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.9283 - mae: 1.0977 - val_loss: 12.1147 - val_mae: 2.5047\n",
            "Epoch 190/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.8744 - mae: 1.1082 - val_loss: 11.5798 - val_mae: 2.3606\n",
            "Epoch 191/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.7645 - mae: 1.0355 - val_loss: 14.2451 - val_mae: 2.6796\n",
            "Epoch 192/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.8110 - mae: 1.1048 - val_loss: 12.3584 - val_mae: 2.5130\n",
            "Epoch 193/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.8277 - mae: 1.0244 - val_loss: 12.5095 - val_mae: 2.5359\n",
            "Epoch 194/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.7687 - mae: 1.0271 - val_loss: 12.6155 - val_mae: 2.5522\n",
            "Epoch 195/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.7904 - mae: 1.1481 - val_loss: 12.6441 - val_mae: 2.5224\n",
            "Epoch 196/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.7953 - mae: 0.9916 - val_loss: 14.7657 - val_mae: 2.7858\n",
            "Epoch 197/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.7555 - mae: 1.0885 - val_loss: 12.0091 - val_mae: 2.4219\n",
            "Epoch 198/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.7045 - mae: 1.0285 - val_loss: 12.6440 - val_mae: 2.5413\n",
            "Epoch 199/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.6385 - mae: 1.0034 - val_loss: 13.4240 - val_mae: 2.6997\n",
            "Epoch 200/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.6802 - mae: 1.0832 - val_loss: 13.3190 - val_mae: 2.6295\n",
            "Epoch 201/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.6727 - mae: 1.0366 - val_loss: 12.7243 - val_mae: 2.6041\n",
            "Epoch 202/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.6814 - mae: 1.0540 - val_loss: 12.5561 - val_mae: 2.5277\n",
            "Epoch 203/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.7182 - mae: 1.0461 - val_loss: 13.3152 - val_mae: 2.5832\n",
            "Epoch 204/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.6878 - mae: 1.0866 - val_loss: 13.4034 - val_mae: 2.7101\n",
            "Epoch 205/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.6223 - mae: 1.0336 - val_loss: 12.9137 - val_mae: 2.5482\n",
            "Epoch 206/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.7371 - mae: 1.1128 - val_loss: 14.4567 - val_mae: 2.7341\n",
            "Epoch 207/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.6294 - mae: 1.0331 - val_loss: 12.7295 - val_mae: 2.5730\n",
            "Epoch 208/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.4270 - mae: 1.0350 - val_loss: 12.7697 - val_mae: 2.5578\n",
            "Epoch 209/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.6656 - mae: 1.0061 - val_loss: 14.5000 - val_mae: 2.7359\n",
            "Epoch 210/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.6279 - mae: 1.0325 - val_loss: 12.1770 - val_mae: 2.4417\n",
            "Epoch 211/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3377 - mae: 0.9690 - val_loss: 15.0500 - val_mae: 2.8242\n",
            "Epoch 212/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3458 - mae: 1.0259 - val_loss: 12.8232 - val_mae: 2.5499\n",
            "Epoch 213/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.5899 - mae: 0.9869 - val_loss: 13.5587 - val_mae: 2.6639\n",
            "Epoch 214/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.4962 - mae: 1.0316 - val_loss: 13.2606 - val_mae: 2.6975\n",
            "Epoch 215/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.6870 - mae: 1.0425 - val_loss: 12.4197 - val_mae: 2.4964\n",
            "Epoch 216/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2942 - mae: 1.0058 - val_loss: 11.9666 - val_mae: 2.4588\n",
            "Epoch 217/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.4410 - mae: 0.9835 - val_loss: 13.4082 - val_mae: 2.6306\n",
            "Epoch 218/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3743 - mae: 1.0095 - val_loss: 12.8328 - val_mae: 2.5995\n",
            "Epoch 219/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3151 - mae: 0.9722 - val_loss: 13.1849 - val_mae: 2.6216\n",
            "Epoch 220/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3321 - mae: 1.0251 - val_loss: 14.9771 - val_mae: 2.8279\n",
            "Epoch 221/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3843 - mae: 1.0045 - val_loss: 12.9433 - val_mae: 2.5753\n",
            "Epoch 222/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3337 - mae: 0.9890 - val_loss: 13.7155 - val_mae: 2.6655\n",
            "Epoch 223/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2889 - mae: 1.0142 - val_loss: 12.9829 - val_mae: 2.5455\n",
            "Epoch 224/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.6426 - mae: 1.0476 - val_loss: 12.1638 - val_mae: 2.4788\n",
            "Epoch 225/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.4263 - mae: 1.0144 - val_loss: 12.4291 - val_mae: 2.4871\n",
            "Epoch 226/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2482 - mae: 0.9893 - val_loss: 12.9053 - val_mae: 2.5415\n",
            "Epoch 227/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3678 - mae: 0.9780 - val_loss: 13.8216 - val_mae: 2.6588\n",
            "Epoch 228/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2515 - mae: 1.0012 - val_loss: 12.3254 - val_mae: 2.4938\n",
            "Epoch 229/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1479 - mae: 0.9497 - val_loss: 13.8694 - val_mae: 2.6166\n",
            "Epoch 230/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2647 - mae: 1.0131 - val_loss: 12.5516 - val_mae: 2.5031\n",
            "Epoch 231/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3112 - mae: 1.0513 - val_loss: 14.0929 - val_mae: 2.7324\n",
            "Epoch 232/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2663 - mae: 0.9788 - val_loss: 12.5262 - val_mae: 2.5520\n",
            "Epoch 233/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2167 - mae: 0.9737 - val_loss: 13.0774 - val_mae: 2.6027\n",
            "Epoch 234/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3766 - mae: 1.0006 - val_loss: 15.1669 - val_mae: 2.8883\n",
            "Epoch 235/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1991 - mae: 1.0020 - val_loss: 13.9174 - val_mae: 2.6803\n",
            "Epoch 236/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3246 - mae: 0.9867 - val_loss: 14.0675 - val_mae: 2.7576\n",
            "Epoch 237/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0503 - mae: 0.9059 - val_loss: 14.9981 - val_mae: 2.8249\n",
            "Epoch 238/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3077 - mae: 0.9858 - val_loss: 13.5968 - val_mae: 2.6890\n",
            "Epoch 239/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3107 - mae: 1.0025 - val_loss: 12.8055 - val_mae: 2.5553\n",
            "Epoch 240/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.4269 - mae: 0.9814 - val_loss: 13.6432 - val_mae: 2.6370\n",
            "Epoch 241/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.3703 - mae: 0.9537 - val_loss: 14.1504 - val_mae: 2.7346\n",
            "Epoch 242/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1989 - mae: 0.9789 - val_loss: 13.7451 - val_mae: 2.6453\n",
            "Epoch 243/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2188 - mae: 0.9384 - val_loss: 14.3388 - val_mae: 2.8172\n",
            "Epoch 244/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0985 - mae: 0.9863 - val_loss: 15.3284 - val_mae: 2.9669\n",
            "Epoch 245/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0690 - mae: 0.9440 - val_loss: 11.9690 - val_mae: 2.4628\n",
            "Epoch 246/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2367 - mae: 0.9917 - val_loss: 14.6493 - val_mae: 2.8008\n",
            "Epoch 247/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0672 - mae: 0.9856 - val_loss: 13.5260 - val_mae: 2.6328\n",
            "Epoch 248/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2754 - mae: 0.9727 - val_loss: 13.1369 - val_mae: 2.5701\n",
            "Epoch 249/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1011 - mae: 0.9485 - val_loss: 13.1665 - val_mae: 2.5619\n",
            "Epoch 250/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9290 - mae: 0.9579 - val_loss: 16.2748 - val_mae: 2.9890\n",
            "Epoch 251/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9691 - mae: 0.9095 - val_loss: 13.8556 - val_mae: 2.6754\n",
            "Epoch 252/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0826 - mae: 0.9641 - val_loss: 13.1151 - val_mae: 2.6131\n",
            "Epoch 253/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0332 - mae: 0.9716 - val_loss: 12.6294 - val_mae: 2.5396\n",
            "Epoch 254/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9356 - mae: 0.9108 - val_loss: 13.8629 - val_mae: 2.6716\n",
            "Epoch 255/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9978 - mae: 0.9238 - val_loss: 13.1600 - val_mae: 2.6389\n",
            "Epoch 256/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9601 - mae: 0.9127 - val_loss: 12.6198 - val_mae: 2.5665\n",
            "Epoch 257/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.2759 - mae: 0.9492 - val_loss: 13.1269 - val_mae: 2.5691\n",
            "Epoch 258/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1162 - mae: 0.9554 - val_loss: 12.8392 - val_mae: 2.5443\n",
            "Epoch 259/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9335 - mae: 0.8829 - val_loss: 13.2509 - val_mae: 2.6022\n",
            "Epoch 260/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1057 - mae: 0.9428 - val_loss: 14.5734 - val_mae: 2.8162\n",
            "Epoch 261/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0097 - mae: 0.9474 - val_loss: 14.1635 - val_mae: 2.7233\n",
            "Epoch 262/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9434 - mae: 0.9321 - val_loss: 14.4928 - val_mae: 2.7912\n",
            "Epoch 263/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0621 - mae: 0.9564 - val_loss: 13.3233 - val_mae: 2.5753\n",
            "Epoch 264/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0324 - mae: 0.9577 - val_loss: 14.1720 - val_mae: 2.7590\n",
            "Epoch 265/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0447 - mae: 0.9582 - val_loss: 13.5470 - val_mae: 2.6992\n",
            "Epoch 266/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9614 - mae: 0.9349 - val_loss: 13.8857 - val_mae: 2.7385\n",
            "Epoch 267/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8321 - mae: 0.9154 - val_loss: 13.8005 - val_mae: 2.6758\n",
            "Epoch 268/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9706 - mae: 0.9060 - val_loss: 13.7319 - val_mae: 2.6653\n",
            "Epoch 269/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1734 - mae: 0.9424 - val_loss: 13.1039 - val_mae: 2.6268\n",
            "Epoch 270/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0871 - mae: 0.9411 - val_loss: 14.1830 - val_mae: 2.7369\n",
            "Epoch 271/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0818 - mae: 0.9463 - val_loss: 13.5591 - val_mae: 2.6893\n",
            "Epoch 272/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9410 - mae: 0.9142 - val_loss: 14.6752 - val_mae: 2.8118\n",
            "Epoch 273/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8035 - mae: 0.8753 - val_loss: 14.9625 - val_mae: 2.8512\n",
            "Epoch 274/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9642 - mae: 0.9692 - val_loss: 13.4928 - val_mae: 2.6478\n",
            "Epoch 275/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0213 - mae: 0.8985 - val_loss: 15.4081 - val_mae: 2.8852\n",
            "Epoch 276/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1587 - mae: 0.9410 - val_loss: 12.8470 - val_mae: 2.5998\n",
            "Epoch 277/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0968 - mae: 0.9574 - val_loss: 14.9047 - val_mae: 2.8195\n",
            "Epoch 278/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0428 - mae: 0.9241 - val_loss: 13.4368 - val_mae: 2.7029\n",
            "Epoch 279/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7071 - mae: 0.8904 - val_loss: 15.8825 - val_mae: 2.9424\n",
            "Epoch 280/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9339 - mae: 0.9540 - val_loss: 14.9688 - val_mae: 2.7611\n",
            "Epoch 281/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6611 - mae: 0.8951 - val_loss: 13.3986 - val_mae: 2.6623\n",
            "Epoch 282/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0482 - mae: 0.9145 - val_loss: 12.7310 - val_mae: 2.6041\n",
            "Epoch 283/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8980 - mae: 0.9406 - val_loss: 15.1356 - val_mae: 2.8705\n",
            "Epoch 284/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.0165 - mae: 0.9466 - val_loss: 12.9483 - val_mae: 2.6052\n",
            "Epoch 285/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9886 - mae: 0.9645 - val_loss: 13.2500 - val_mae: 2.6747\n",
            "Epoch 286/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6912 - mae: 0.9071 - val_loss: 12.6859 - val_mae: 2.5855\n",
            "Epoch 287/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8272 - mae: 0.9381 - val_loss: 15.4178 - val_mae: 2.9558\n",
            "Epoch 288/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.9121 - mae: 0.9141 - val_loss: 13.3256 - val_mae: 2.7332\n",
            "Epoch 289/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8183 - mae: 0.9391 - val_loss: 15.3998 - val_mae: 2.9580\n",
            "Epoch 290/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 2.1012 - mae: 0.9312 - val_loss: 13.5818 - val_mae: 2.7189\n",
            "Epoch 291/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6468 - mae: 0.9136 - val_loss: 16.1800 - val_mae: 3.1384\n",
            "Epoch 292/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7617 - mae: 0.9092 - val_loss: 17.1029 - val_mae: 3.1538\n",
            "Epoch 293/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6103 - mae: 0.8823 - val_loss: 13.7527 - val_mae: 2.7160\n",
            "Epoch 294/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5833 - mae: 0.8892 - val_loss: 12.2726 - val_mae: 2.5608\n",
            "Epoch 295/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6863 - mae: 0.8902 - val_loss: 11.9390 - val_mae: 2.6004\n",
            "Epoch 296/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8104 - mae: 0.9140 - val_loss: 12.6217 - val_mae: 2.6396\n",
            "Epoch 297/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6382 - mae: 0.8777 - val_loss: 12.4246 - val_mae: 2.6121\n",
            "Epoch 298/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8617 - mae: 0.9408 - val_loss: 13.0402 - val_mae: 2.7309\n",
            "Epoch 299/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6419 - mae: 0.8665 - val_loss: 18.5424 - val_mae: 3.4505\n",
            "Epoch 300/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8159 - mae: 0.9282 - val_loss: 13.9408 - val_mae: 2.8742\n",
            "Epoch 301/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7215 - mae: 0.8954 - val_loss: 15.3463 - val_mae: 2.9452\n",
            "Epoch 302/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6643 - mae: 0.9016 - val_loss: 12.5480 - val_mae: 2.6529\n",
            "Epoch 303/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8760 - mae: 0.8964 - val_loss: 13.3477 - val_mae: 2.6544\n",
            "Epoch 304/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7689 - mae: 0.9161 - val_loss: 13.2076 - val_mae: 2.6813\n",
            "Epoch 305/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8678 - mae: 0.9032 - val_loss: 14.4107 - val_mae: 2.8200\n",
            "Epoch 306/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6720 - mae: 0.8919 - val_loss: 12.7940 - val_mae: 2.6708\n",
            "Epoch 307/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6650 - mae: 0.8926 - val_loss: 12.2546 - val_mae: 2.6119\n",
            "Epoch 308/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7718 - mae: 0.8722 - val_loss: 13.3163 - val_mae: 2.6941\n",
            "Epoch 309/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6909 - mae: 0.8876 - val_loss: 16.9246 - val_mae: 3.1518\n",
            "Epoch 310/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7462 - mae: 0.9162 - val_loss: 15.0920 - val_mae: 2.9080\n",
            "Epoch 311/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.8103 - mae: 0.8903 - val_loss: 13.1500 - val_mae: 2.6946\n",
            "Epoch 312/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6102 - mae: 0.8725 - val_loss: 13.3662 - val_mae: 2.7586\n",
            "Epoch 313/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5172 - mae: 0.8593 - val_loss: 12.7734 - val_mae: 2.6488\n",
            "Epoch 314/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6727 - mae: 0.8728 - val_loss: 13.3548 - val_mae: 2.7565\n",
            "Epoch 315/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6501 - mae: 0.8595 - val_loss: 14.7330 - val_mae: 2.8485\n",
            "Epoch 316/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5569 - mae: 0.9050 - val_loss: 13.7363 - val_mae: 2.7748\n",
            "Epoch 317/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.7005 - mae: 0.8646 - val_loss: 13.0711 - val_mae: 2.7586\n",
            "Epoch 318/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6015 - mae: 0.8771 - val_loss: 12.7004 - val_mae: 2.6565\n",
            "Epoch 319/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5474 - mae: 0.8878 - val_loss: 12.2492 - val_mae: 2.6116\n",
            "Epoch 320/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6787 - mae: 0.9141 - val_loss: 13.1580 - val_mae: 2.7188\n",
            "Epoch 321/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5843 - mae: 0.8657 - val_loss: 13.3806 - val_mae: 2.7845\n",
            "Epoch 322/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5191 - mae: 0.8763 - val_loss: 14.9908 - val_mae: 2.9215\n",
            "Epoch 323/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6134 - mae: 0.8908 - val_loss: 12.3690 - val_mae: 2.5986\n",
            "Epoch 324/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5513 - mae: 0.8633 - val_loss: 14.4252 - val_mae: 2.9016\n",
            "Epoch 325/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6012 - mae: 0.8639 - val_loss: 12.4591 - val_mae: 2.5979\n",
            "Epoch 326/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6408 - mae: 0.8630 - val_loss: 14.6548 - val_mae: 2.7694\n",
            "Epoch 327/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4153 - mae: 0.8262 - val_loss: 13.1526 - val_mae: 2.7108\n",
            "Epoch 328/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5166 - mae: 0.8641 - val_loss: 12.4956 - val_mae: 2.6533\n",
            "Epoch 329/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5555 - mae: 0.8663 - val_loss: 13.5689 - val_mae: 2.7672\n",
            "Epoch 330/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4931 - mae: 0.8550 - val_loss: 13.6713 - val_mae: 2.7664\n",
            "Epoch 331/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4295 - mae: 0.8674 - val_loss: 13.8469 - val_mae: 2.7390\n",
            "Epoch 332/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6344 - mae: 0.8712 - val_loss: 13.1415 - val_mae: 2.7435\n",
            "Epoch 333/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5534 - mae: 0.8512 - val_loss: 13.1785 - val_mae: 2.6658\n",
            "Epoch 334/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5156 - mae: 0.8461 - val_loss: 13.3436 - val_mae: 2.7615\n",
            "Epoch 335/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4647 - mae: 0.8348 - val_loss: 13.3429 - val_mae: 2.7323\n",
            "Epoch 336/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5647 - mae: 0.8490 - val_loss: 12.2687 - val_mae: 2.6192\n",
            "Epoch 337/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2948 - mae: 0.8192 - val_loss: 14.1144 - val_mae: 2.8228\n",
            "Epoch 338/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4916 - mae: 0.8587 - val_loss: 14.1044 - val_mae: 2.8434\n",
            "Epoch 339/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.6449 - mae: 0.8795 - val_loss: 12.9007 - val_mae: 2.6905\n",
            "Epoch 340/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4891 - mae: 0.8640 - val_loss: 12.9498 - val_mae: 2.7132\n",
            "Epoch 341/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3809 - mae: 0.8232 - val_loss: 15.0062 - val_mae: 2.9776\n",
            "Epoch 342/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4882 - mae: 0.8351 - val_loss: 12.4396 - val_mae: 2.6916\n",
            "Epoch 343/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2857 - mae: 0.8270 - val_loss: 12.8572 - val_mae: 2.7403\n",
            "Epoch 344/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5306 - mae: 0.8686 - val_loss: 14.2153 - val_mae: 2.8281\n",
            "Epoch 345/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4970 - mae: 0.8535 - val_loss: 13.5679 - val_mae: 2.7676\n",
            "Epoch 346/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4687 - mae: 0.8433 - val_loss: 14.9427 - val_mae: 2.9322\n",
            "Epoch 347/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3570 - mae: 0.8624 - val_loss: 14.3791 - val_mae: 2.8889\n",
            "Epoch 348/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5310 - mae: 0.8613 - val_loss: 13.0340 - val_mae: 2.7339\n",
            "Epoch 349/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3427 - mae: 0.8182 - val_loss: 15.6235 - val_mae: 3.0899\n",
            "Epoch 350/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5719 - mae: 0.8805 - val_loss: 13.6856 - val_mae: 2.8302\n",
            "Epoch 351/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2112 - mae: 0.7635 - val_loss: 13.9272 - val_mae: 2.8272\n",
            "Epoch 352/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3991 - mae: 0.8275 - val_loss: 13.6517 - val_mae: 2.7520\n",
            "Epoch 353/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4780 - mae: 0.8392 - val_loss: 12.7023 - val_mae: 2.6767\n",
            "Epoch 354/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3757 - mae: 0.8491 - val_loss: 16.8631 - val_mae: 3.2357\n",
            "Epoch 355/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3225 - mae: 0.8389 - val_loss: 13.0739 - val_mae: 2.6623\n",
            "Epoch 356/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3303 - mae: 0.8172 - val_loss: 12.3078 - val_mae: 2.6157\n",
            "Epoch 357/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3371 - mae: 0.8032 - val_loss: 13.1036 - val_mae: 2.7165\n",
            "Epoch 358/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.4244 - mae: 0.8377 - val_loss: 12.0007 - val_mae: 2.6528\n",
            "Epoch 359/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1909 - mae: 0.7998 - val_loss: 14.8520 - val_mae: 2.8459\n",
            "Epoch 360/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3388 - mae: 0.8547 - val_loss: 13.5145 - val_mae: 2.7874\n",
            "Epoch 361/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.5375 - mae: 0.8350 - val_loss: 12.4655 - val_mae: 2.6712\n",
            "Epoch 362/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2796 - mae: 0.8133 - val_loss: 12.3755 - val_mae: 2.7322\n",
            "Epoch 363/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3909 - mae: 0.8245 - val_loss: 13.9929 - val_mae: 2.8582\n",
            "Epoch 364/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2851 - mae: 0.8227 - val_loss: 11.9994 - val_mae: 2.5698\n",
            "Epoch 365/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3626 - mae: 0.8493 - val_loss: 12.5330 - val_mae: 2.6640\n",
            "Epoch 366/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3360 - mae: 0.8421 - val_loss: 12.0371 - val_mae: 2.7150\n",
            "Epoch 367/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3398 - mae: 0.8180 - val_loss: 12.5231 - val_mae: 2.6846\n",
            "Epoch 368/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1833 - mae: 0.8130 - val_loss: 13.5104 - val_mae: 2.7602\n",
            "Epoch 369/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2455 - mae: 0.8073 - val_loss: 12.8340 - val_mae: 2.7485\n",
            "Epoch 370/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2792 - mae: 0.8237 - val_loss: 13.6172 - val_mae: 2.7826\n",
            "Epoch 371/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1540 - mae: 0.7706 - val_loss: 12.3162 - val_mae: 2.6617\n",
            "Epoch 372/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2662 - mae: 0.8135 - val_loss: 13.3440 - val_mae: 2.8088\n",
            "Epoch 373/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3448 - mae: 0.8053 - val_loss: 12.1073 - val_mae: 2.6281\n",
            "Epoch 374/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2214 - mae: 0.8366 - val_loss: 13.3415 - val_mae: 2.8252\n",
            "Epoch 375/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3537 - mae: 0.8180 - val_loss: 12.4874 - val_mae: 2.7380\n",
            "Epoch 376/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2251 - mae: 0.7875 - val_loss: 12.8490 - val_mae: 2.7886\n",
            "Epoch 377/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2887 - mae: 0.8226 - val_loss: 12.4312 - val_mae: 2.6716\n",
            "Epoch 378/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2012 - mae: 0.8231 - val_loss: 13.2274 - val_mae: 2.7689\n",
            "Epoch 379/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1591 - mae: 0.7495 - val_loss: 14.4131 - val_mae: 2.9400\n",
            "Epoch 380/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2780 - mae: 0.8032 - val_loss: 15.0599 - val_mae: 2.9746\n",
            "Epoch 381/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1106 - mae: 0.7796 - val_loss: 13.0287 - val_mae: 2.7262\n",
            "Epoch 382/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2924 - mae: 0.8024 - val_loss: 15.6067 - val_mae: 3.1028\n",
            "Epoch 383/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3145 - mae: 0.7965 - val_loss: 12.9893 - val_mae: 2.7040\n",
            "Epoch 384/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2223 - mae: 0.7728 - val_loss: 12.0276 - val_mae: 2.6526\n",
            "Epoch 385/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2220 - mae: 0.7642 - val_loss: 12.9719 - val_mae: 2.7458\n",
            "Epoch 386/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2965 - mae: 0.8218 - val_loss: 13.5552 - val_mae: 2.7169\n",
            "Epoch 387/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1627 - mae: 0.7791 - val_loss: 12.2278 - val_mae: 2.6619\n",
            "Epoch 388/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2570 - mae: 0.7679 - val_loss: 14.7419 - val_mae: 2.9123\n",
            "Epoch 389/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2155 - mae: 0.7884 - val_loss: 12.9017 - val_mae: 2.7334\n",
            "Epoch 390/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2131 - mae: 0.7555 - val_loss: 12.3259 - val_mae: 2.6458\n",
            "Epoch 391/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2890 - mae: 0.8396 - val_loss: 13.4892 - val_mae: 2.8026\n",
            "Epoch 392/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2430 - mae: 0.7851 - val_loss: 14.6692 - val_mae: 2.9708\n",
            "Epoch 393/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1107 - mae: 0.7715 - val_loss: 12.7253 - val_mae: 2.7994\n",
            "Epoch 394/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2446 - mae: 0.7851 - val_loss: 12.9987 - val_mae: 2.7475\n",
            "Epoch 395/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1816 - mae: 0.7611 - val_loss: 13.3376 - val_mae: 2.7883\n",
            "Epoch 396/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.3150 - mae: 0.8080 - val_loss: 13.0827 - val_mae: 2.7964\n",
            "Epoch 397/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2953 - mae: 0.8179 - val_loss: 13.5526 - val_mae: 2.8654\n",
            "Epoch 398/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2968 - mae: 0.8089 - val_loss: 12.4845 - val_mae: 2.7271\n",
            "Epoch 399/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1288 - mae: 0.7692 - val_loss: 14.1234 - val_mae: 2.9901\n",
            "Epoch 400/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1932 - mae: 0.8280 - val_loss: 13.2071 - val_mae: 2.7920\n",
            "Epoch 401/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1354 - mae: 0.7775 - val_loss: 12.2272 - val_mae: 2.6609\n",
            "Epoch 402/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2042 - mae: 0.7740 - val_loss: 13.6984 - val_mae: 2.8639\n",
            "Epoch 403/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1007 - mae: 0.7690 - val_loss: 14.2784 - val_mae: 2.8431\n",
            "Epoch 404/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0722 - mae: 0.7793 - val_loss: 14.7525 - val_mae: 2.8592\n",
            "Epoch 405/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0282 - mae: 0.7566 - val_loss: 13.0069 - val_mae: 2.7840\n",
            "Epoch 406/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0718 - mae: 0.7450 - val_loss: 13.2354 - val_mae: 2.7788\n",
            "Epoch 407/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1469 - mae: 0.7841 - val_loss: 13.5630 - val_mae: 2.7584\n",
            "Epoch 408/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2482 - mae: 0.8277 - val_loss: 12.0713 - val_mae: 2.6404\n",
            "Epoch 409/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1252 - mae: 0.7681 - val_loss: 14.0017 - val_mae: 2.8854\n",
            "Epoch 410/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2281 - mae: 0.7845 - val_loss: 12.7126 - val_mae: 2.7200\n",
            "Epoch 411/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1254 - mae: 0.7838 - val_loss: 12.9496 - val_mae: 2.7438\n",
            "Epoch 412/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0622 - mae: 0.7478 - val_loss: 15.1753 - val_mae: 3.0072\n",
            "Epoch 413/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2217 - mae: 0.8098 - val_loss: 12.8550 - val_mae: 2.6866\n",
            "Epoch 414/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1310 - mae: 0.7794 - val_loss: 14.0118 - val_mae: 2.8807\n",
            "Epoch 415/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0993 - mae: 0.7375 - val_loss: 12.7081 - val_mae: 2.7224\n",
            "Epoch 416/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0346 - mae: 0.7533 - val_loss: 13.3188 - val_mae: 2.8552\n",
            "Epoch 417/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1277 - mae: 0.7605 - val_loss: 13.0256 - val_mae: 2.7877\n",
            "Epoch 418/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1636 - mae: 0.7913 - val_loss: 13.3464 - val_mae: 2.7811\n",
            "Epoch 419/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2067 - mae: 0.8062 - val_loss: 12.0395 - val_mae: 2.6206\n",
            "Epoch 420/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1584 - mae: 0.7740 - val_loss: 12.8553 - val_mae: 2.7379\n",
            "Epoch 421/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1617 - mae: 0.7465 - val_loss: 14.4924 - val_mae: 2.8906\n",
            "Epoch 422/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0719 - mae: 0.7681 - val_loss: 11.9252 - val_mae: 2.5903\n",
            "Epoch 423/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1803 - mae: 0.8001 - val_loss: 12.2690 - val_mae: 2.7311\n",
            "Epoch 424/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1801 - mae: 0.7719 - val_loss: 12.8031 - val_mae: 2.7109\n",
            "Epoch 425/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9845 - mae: 0.7537 - val_loss: 13.2319 - val_mae: 2.6725\n",
            "Epoch 426/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.2018 - mae: 0.7751 - val_loss: 13.8861 - val_mae: 2.7587\n",
            "Epoch 427/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0499 - mae: 0.7665 - val_loss: 15.8177 - val_mae: 3.0513\n",
            "Epoch 428/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1833 - mae: 0.7710 - val_loss: 15.0781 - val_mae: 2.9923\n",
            "Epoch 429/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0535 - mae: 0.7595 - val_loss: 15.1091 - val_mae: 3.0211\n",
            "Epoch 430/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0993 - mae: 0.7653 - val_loss: 12.1581 - val_mae: 2.6708\n",
            "Epoch 431/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0825 - mae: 0.7577 - val_loss: 13.4106 - val_mae: 2.8170\n",
            "Epoch 432/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9817 - mae: 0.7116 - val_loss: 12.5370 - val_mae: 2.6899\n",
            "Epoch 433/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1239 - mae: 0.7719 - val_loss: 11.8504 - val_mae: 2.6522\n",
            "Epoch 434/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1374 - mae: 0.7669 - val_loss: 12.2331 - val_mae: 2.6774\n",
            "Epoch 435/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1246 - mae: 0.7854 - val_loss: 15.7860 - val_mae: 3.1061\n",
            "Epoch 436/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1133 - mae: 0.7555 - val_loss: 13.1028 - val_mae: 2.7873\n",
            "Epoch 437/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0669 - mae: 0.7419 - val_loss: 12.8841 - val_mae: 2.6946\n",
            "Epoch 438/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0881 - mae: 0.7882 - val_loss: 12.9262 - val_mae: 2.6714\n",
            "Epoch 439/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9869 - mae: 0.7054 - val_loss: 11.4354 - val_mae: 2.5241\n",
            "Epoch 440/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0624 - mae: 0.7542 - val_loss: 14.0718 - val_mae: 2.7786\n",
            "Epoch 441/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0788 - mae: 0.7463 - val_loss: 13.1745 - val_mae: 2.7573\n",
            "Epoch 442/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1309 - mae: 0.7677 - val_loss: 13.6395 - val_mae: 2.8051\n",
            "Epoch 443/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0378 - mae: 0.7423 - val_loss: 12.7974 - val_mae: 2.6661\n",
            "Epoch 444/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0470 - mae: 0.7341 - val_loss: 13.9105 - val_mae: 2.8668\n",
            "Epoch 445/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0487 - mae: 0.7451 - val_loss: 14.1744 - val_mae: 2.8385\n",
            "Epoch 446/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1381 - mae: 0.7392 - val_loss: 12.6406 - val_mae: 2.6834\n",
            "Epoch 447/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9956 - mae: 0.7236 - val_loss: 15.9711 - val_mae: 3.0167\n",
            "Epoch 448/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0250 - mae: 0.7474 - val_loss: 13.0052 - val_mae: 2.6413\n",
            "Epoch 449/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9502 - mae: 0.7147 - val_loss: 16.5659 - val_mae: 3.1287\n",
            "Epoch 450/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1037 - mae: 0.7729 - val_loss: 12.9188 - val_mae: 2.7477\n",
            "Epoch 451/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0339 - mae: 0.7542 - val_loss: 13.8105 - val_mae: 2.7902\n",
            "Epoch 452/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9892 - mae: 0.7344 - val_loss: 13.6921 - val_mae: 2.8270\n",
            "Epoch 453/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8889 - mae: 0.7055 - val_loss: 13.5364 - val_mae: 2.7804\n",
            "Epoch 454/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9494 - mae: 0.6880 - val_loss: 12.8253 - val_mae: 2.6824\n",
            "Epoch 455/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0759 - mae: 0.7536 - val_loss: 13.8282 - val_mae: 2.8585\n",
            "Epoch 456/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1496 - mae: 0.7215 - val_loss: 12.9194 - val_mae: 2.6297\n",
            "Epoch 457/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9245 - mae: 0.7102 - val_loss: 14.2563 - val_mae: 2.8766\n",
            "Epoch 458/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8890 - mae: 0.7081 - val_loss: 13.2724 - val_mae: 2.7737\n",
            "Epoch 459/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0081 - mae: 0.7390 - val_loss: 14.6852 - val_mae: 2.8960\n",
            "Epoch 460/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8874 - mae: 0.7049 - val_loss: 14.5066 - val_mae: 2.8306\n",
            "Epoch 461/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0384 - mae: 0.7378 - val_loss: 15.3508 - val_mae: 3.0372\n",
            "Epoch 462/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9584 - mae: 0.7370 - val_loss: 13.7261 - val_mae: 2.7924\n",
            "Epoch 463/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8490 - mae: 0.7016 - val_loss: 15.1651 - val_mae: 2.9605\n",
            "Epoch 464/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9583 - mae: 0.7359 - val_loss: 14.6084 - val_mae: 2.8977\n",
            "Epoch 465/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0107 - mae: 0.7249 - val_loss: 13.1230 - val_mae: 2.7616\n",
            "Epoch 466/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9373 - mae: 0.6881 - val_loss: 15.1171 - val_mae: 2.9959\n",
            "Epoch 467/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0564 - mae: 0.7540 - val_loss: 13.5953 - val_mae: 2.8089\n",
            "Epoch 468/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9543 - mae: 0.7177 - val_loss: 14.1589 - val_mae: 2.8546\n",
            "Epoch 469/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9031 - mae: 0.6727 - val_loss: 14.1229 - val_mae: 2.8672\n",
            "Epoch 470/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.1356 - mae: 0.7464 - val_loss: 13.4020 - val_mae: 2.7562\n",
            "Epoch 471/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0230 - mae: 0.7351 - val_loss: 13.4358 - val_mae: 2.7944\n",
            "Epoch 472/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9902 - mae: 0.7454 - val_loss: 13.9321 - val_mae: 2.8799\n",
            "Epoch 473/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0085 - mae: 0.7036 - val_loss: 13.4253 - val_mae: 2.8021\n",
            "Epoch 474/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0672 - mae: 0.7370 - val_loss: 13.6612 - val_mae: 2.7583\n",
            "Epoch 475/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8915 - mae: 0.7064 - val_loss: 13.8905 - val_mae: 2.8522\n",
            "Epoch 476/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8925 - mae: 0.7033 - val_loss: 12.7721 - val_mae: 2.6928\n",
            "Epoch 477/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9763 - mae: 0.7116 - val_loss: 14.4507 - val_mae: 2.8870\n",
            "Epoch 478/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8928 - mae: 0.6753 - val_loss: 13.2297 - val_mae: 2.7940\n",
            "Epoch 479/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9286 - mae: 0.7148 - val_loss: 13.4336 - val_mae: 2.7748\n",
            "Epoch 480/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9891 - mae: 0.7005 - val_loss: 12.9757 - val_mae: 2.7400\n",
            "Epoch 481/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9294 - mae: 0.7127 - val_loss: 13.8381 - val_mae: 2.8411\n",
            "Epoch 482/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8704 - mae: 0.6713 - val_loss: 14.1790 - val_mae: 2.9077\n",
            "Epoch 483/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9954 - mae: 0.7022 - val_loss: 12.9664 - val_mae: 2.7280\n",
            "Epoch 484/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9715 - mae: 0.7102 - val_loss: 14.9213 - val_mae: 2.8748\n",
            "Epoch 485/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8419 - mae: 0.6772 - val_loss: 13.0467 - val_mae: 2.6979\n",
            "Epoch 486/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9774 - mae: 0.7383 - val_loss: 13.7089 - val_mae: 2.7246\n",
            "Epoch 487/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9370 - mae: 0.6933 - val_loss: 14.0046 - val_mae: 2.7750\n",
            "Epoch 488/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9954 - mae: 0.7548 - val_loss: 13.2314 - val_mae: 2.8024\n",
            "Epoch 489/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8948 - mae: 0.6722 - val_loss: 13.3487 - val_mae: 2.7929\n",
            "Epoch 490/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9634 - mae: 0.7344 - val_loss: 13.7396 - val_mae: 2.7685\n",
            "Epoch 491/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0138 - mae: 0.7595 - val_loss: 16.5455 - val_mae: 2.9969\n",
            "Epoch 492/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8671 - mae: 0.6933 - val_loss: 13.1976 - val_mae: 2.7545\n",
            "Epoch 493/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0055 - mae: 0.7040 - val_loss: 13.7994 - val_mae: 2.7303\n",
            "Epoch 494/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9358 - mae: 0.7293 - val_loss: 15.9269 - val_mae: 2.9590\n",
            "Epoch 495/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9488 - mae: 0.7038 - val_loss: 12.8237 - val_mae: 2.7357\n",
            "Epoch 496/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9539 - mae: 0.7237 - val_loss: 14.5908 - val_mae: 2.9197\n",
            "Epoch 497/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9232 - mae: 0.7157 - val_loss: 15.1167 - val_mae: 2.9581\n",
            "Epoch 498/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.8133 - mae: 0.6793 - val_loss: 13.6156 - val_mae: 2.7718\n",
            "Epoch 499/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 0.9643 - mae: 0.7439 - val_loss: 13.2504 - val_mae: 2.6886\n",
            "Epoch 500/500\n",
            "303/303 [==============================] - 0s 1ms/step - loss: 1.0290 - mae: 0.7350 - val_loss: 15.2055 - val_mae: 2.9644\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "grWQIfOoO88q",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "average_mae_history = [np.mean([x[i] for x in all_mae_histories]) for i in range(num_epochs)]"
      ],
      "execution_count": 22,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UAxl9UaISQZ9",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        },
        "outputId": "38cea1ba-b402-474e-fb27-f7da936462a7"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "plt.plot(range(1, len(average_mae_history) + 1), average_mae_history)\n",
        "plt.xlabel('Epochs')\n",
        "plt.ylabel('Validation MAE')\n",
        "plt.show()"
      ],
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4x_xdYp0SQ7Z",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        },
        "outputId": "e224cede-4bdf-4675-b1c4-7e16127d9090"
      },
      "source": [
        "def smooth_curve(points, factor=0.9):\n",
        "  smoothed_points = []\n",
        "  for point in points:\n",
        "    if smoothed_points:\n",
        "      previous = smoothed_points[-1]\n",
        "      smoothed_points.append(previous * factor + point * (1 - factor))\n",
        "    else:\n",
        "      smoothed_points.append(point)\n",
        "  return smoothed_points\n",
        "\n",
        "smooth_mae_history = smooth_curve(average_mae_history[10:])\n",
        "\n",
        "plt.plot(range(1, len(smooth_mae_history) + 1), smooth_mae_history)\n",
        "plt.xlabel('Epochs')\n",
        "plt.ylabel('Validation MAE')\n",
        "plt.show()"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yWiuUAFHVd8g",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "153629d9-6540-4902-d7f0-8435b82d001d"
      },
      "source": [
        "model = build_model()                                                     \n",
        "model.fit(train_data, train_targets,                                      \n",
        "          epochs=80, batch_size=16, verbose=0)\n",
        "test_mse_score, test_mae_score = model.evaluate(test_data, test_targets)"
      ],
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "4/4 [==============================] - 0s 2ms/step - loss: 18.4007 - mae: 2.8988\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9YUWmTEZVkkJ",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "30c5da4c-2120-4131-d788-eb8cba91e80e"
      },
      "source": [
        "test_mae_score"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "2.898832321166992"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "536HfWu5V4mo",
        "colab_type": "text"
      },
      "source": [
        "# WRAPPING UP\n",
        "\n",
        "Here’s what you should take away from this example:\n",
        "\n",
        "* Regression is done using different loss functions than what we used for classification. Mean squared error (MSE) is a loss function commonly used for regression.\n",
        "* Similarly, evaluation metrics to be used for regression differ from those used for classification; naturally, the concept of accuracy doesn’t apply for regression. A common regression metric is mean absolute error (MAE).\n",
        "* When features in the input data have values in different ranges, each feature should be scaled independently as a preprocessing step.\n",
        "* When there is little data available, using K-fold validation is a great way to reliably evaluate a model.\n",
        "* When little training data is available, it’s preferable to use a small network with few hidden layers (typically only one or two), in order to avoid severe overfitting."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "M89E_AREV_yo",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}