{ "nbformat": 4, "nbformat_minor": 0, "metadata": { "colab": { "name": "Final Neural Network prediction.ipynb", "provenance": [] }, "kernelspec": { "name": "python3", "display_name": "Python 3" } }, "cells": [ { "cell_type": "code", "metadata": { "id": "Wa2VS7ydJoFL" }, "source": [ "import pandas_datareader as web\n", "import seaborn as sns\n", "import datetime\n", "import pandas as pd\n", "from sklearn import preprocessing\n", "from sklearn.model_selection import train_test_split\n", "import matplotlib.pyplot as plt\n", "\n", "global X_test " ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "QfW7zjVPJ1z8" }, "source": [ "#Reading the data into Pandas\n", "df = pd.read_csv('/content/karjat.csv')" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 314 }, "id": "pXs2DGk6J4kD", "outputId": "bfa84e58-f0d8-4c08-87d9-e8c7980b02f2" }, "source": [ "#Pre-Processing and Understanding the Data\n", "#Getting to know mean, std daviation, Interquartile ranges of data\n", "df.describe()\n" ], "execution_count": null, "outputs": [ { "output_type": "execute_result", "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Maximum TemperatureMinimum TemperatureTemperatureHeat IndexPrecipitationWind SpeedCloud CoverRelative Humidity
count1642.0000001642.0000001642.0000001642.0000001642.0000001642.0000001642.0000001642.000000
mean33.43976919.42198527.00249736.81979354.44903822.67259440.60115766.713965
std2.7141934.5241972.2767644.025920119.38922611.90800630.57516612.470943
min24.0000000.90000018.60000027.1000000.0000003.6000000.00000031.750000
25%31.10000015.90000025.80000034.1000000.00000018.40000013.10000057.052500
50%33.10000021.10000027.20000036.9000000.10000022.30000032.30000065.265000
75%35.10000022.70000028.60000039.30000045.83000025.90000069.50000078.007500
max42.20000028.30000032.20000053.800000959.240000179.60000097.20000093.260000
\n", "
" ], "text/plain": [ " Maximum Temperature Minimum Temperature ... Cloud Cover Relative Humidity\n", "count 1642.000000 1642.000000 ... 1642.000000 1642.000000\n", "mean 33.439769 19.421985 ... 40.601157 66.713965\n", "std 2.714193 4.524197 ... 30.575166 12.470943\n", "min 24.000000 0.900000 ... 0.000000 31.750000\n", "25% 31.100000 15.900000 ... 13.100000 57.052500\n", "50% 33.100000 21.100000 ... 32.300000 65.265000\n", "75% 35.100000 22.700000 ... 69.500000 78.007500\n", "max 42.200000 28.300000 ... 97.200000 93.260000\n", "\n", "[8 rows x 8 columns]" ] }, "metadata": { "tags": [] }, "execution_count": 29 } ] }, { "cell_type": "code", "metadata": { "id": "0o8bStvcJ5_E" }, "source": [ "##Checking if data has any missing values \n", "##print (df.isna().sum())\n" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "xqBBrABcJ7Ba" }, "source": [ "#We have 'Period' column which we want to convert to single numeric value before processing\n", "le = preprocessing.LabelEncoder()\n", "date=le.fit_transform(df['Period'])\n", "date=pd.DataFrame(date)\n" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "id": "OL-zHdnxJ8UM" }, "source": [ "\n", "#Droping the original date column from dataframe\n", "df=df.drop(['Location'],axis=1)\n", "df=df.drop('Address',axis=1)\n", "df=df.drop('Maximum Temperature',axis=1)\n", "df=df.drop('Minimum Temperature',axis=1)\n", "df=df.drop('Heat Index',axis=1)\n", "df=df.drop('Precipitation',axis=1)\n", "df=df.drop('Wind Speed',axis=1)\n", "df=df.drop('Cloud Cover',axis=1)\n", "df=df.drop('Relative Humidity',axis=1)" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 359 }, "id": "jtUHgp_eJ90r", "outputId": "f6dc8284-73b2-469b-ee35-1a200219f1d2" }, "source": [ "#Adding new encoded date column to our dataframe\n", "df['Period']=date\n", "#Once again checking the dataframe\n", "df.head(10)\n" ], "execution_count": null, "outputs": [ { "output_type": "execute_result", "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
PeriodTemperature
03725.4
137624.8
271724.3
3105826.3
4139925.9
5152824.0
6155923.0
7159025.0
8162125.1
96826.7
\n", "
" ], "text/plain": [ " Period Temperature\n", "0 37 25.4\n", "1 376 24.8\n", "2 717 24.3\n", "3 1058 26.3\n", "4 1399 25.9\n", "5 1528 24.0\n", "6 1559 23.0\n", "7 1590 25.0\n", "8 1621 25.1\n", "9 68 26.7" ] }, "metadata": { "tags": [] }, "execution_count": 33 } ] }, { "cell_type": "code", "metadata": { "id": "nlnvcchkJ_UG" }, "source": [ "#Actual processing starts here\n", "#Seperating dependent and independent columns \n", "y_actual=df.pop('Temperature')\n", "\n", "X=df" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "_9323Ze_KBvS", "outputId": "b93be44f-923e-4607-cd98-534ffd3a4b18" }, "source": [ "#Now Spliting the data into train test \n", "print ('Splitting the data into train test')\n", "X_train, X_test, y_train, y_test = train_test_split(X, y_actual, test_size=0.30, random_state=0)\n", "\n", "print ('Spliting Done.')\n", "print ('---------------------------------------------------------------------')\n", "print('X_train shape ',X_train.shape)\n", "print('X_test.shape ',X_test.shape)\n", "print(' y_train.shape ',y_train.shape)\n", "print('y_test.shape ', y_test.shape)\n", "\n", "print ('\\n\\n\\n')\n", "print ('--------------------------------------------------------------------------------------------')\n", "print ('Neural Network .. ')" ], "execution_count": null, "outputs": [ { "output_type": "stream", "text": [ "Splitting the data into train test\n", "Spliting Done.\n", "---------------------------------------------------------------------\n", "X_train shape (1149, 1)\n", "X_test.shape (493, 1)\n", " y_train.shape (1149,)\n", "y_test.shape (493,)\n", "\n", "\n", "\n", "\n", "--------------------------------------------------------------------------------------------\n", "Neural Network .. \n" ], "name": "stdout" } ] }, { "cell_type": "code", "metadata": { "id": "fvWlyTjEKEEc" }, "source": [ "import keras\n", "from keras.models import Sequential\n", "from keras.layers import Dense, Dropout\n", "from keras.optimizers import Adam\n", "from sklearn.preprocessing import StandardScaler\n", "from keras.wrappers.scikit_learn import KerasRegressor" ], "execution_count": null, "outputs": [] }, { "cell_type": "code", "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "elLLhXmhKFOT", "outputId": "e10432cd-8a66-475e-c833-9c04496dd29e" }, "source": [ "print (' ')\n", "def build_regressor():\n", " regressor = Sequential()\n", " regressor.add(Dense(units=5, activation='relu', input_dim=1))\n", " regressor.add(Dense(units=15, activation='relu'))\n", " regressor.add(Dense(units=1))\n", " regressor.compile(optimizer='adam', loss='mean_squared_error', metrics=['mae'])\n", " return regressor" ], "execution_count": null, "outputs": [ { "output_type": "stream", "text": [ " \n" ], "name": "stdout" } ] }, { "cell_type": "code", "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 1000 }, "id": "Gg9ZGdsFKHaq", "outputId": "6a959719-3219-4b99-82d7-2ba540c67399" }, "source": [ "def cnn():\n", " global X_test\n", " regressor = KerasRegressor(build_fn=build_regressor, batch_size=32, epochs=500)\n", " print ('Training the CNN Model...')\n", " history = regressor.fit(X_train, y_train)\n", " print ('Training Done')\n", " y_pred = regressor.predict(X_test)\n", " \n", " \n", " #Saving the result in excel\n", " result2 = pd.DataFrame()\n", " result2['Actual_values'] = y_test\n", " result2['Predict_values'] = y_pred\n", " result2['Period'] = le.inverse_transform(X_test['Period'])\n", " result2.head(10)\n", " result2.to_excel('Results_CNN.xlsx', index=False)\n", " \n", " plt.figure(figsize=(16,8))\n", " plt.title('Predicted Temperature Values using NN.')\n", " plt.plot(y_pred)\n", " plt.show()\n", "\n", "cnn()" ], "execution_count": null, "outputs": [ { "output_type": "stream", "text": [ "Training the CNN Model...\n", "Epoch 1/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 279.0621 - mae: 14.7699\n", "Epoch 2/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 211.9123 - mae: 12.1667\n", "Epoch 3/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 218.0576 - mae: 12.4903\n", "Epoch 4/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 194.2386 - mae: 11.6236\n", "Epoch 5/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 208.1153 - mae: 12.1709\n", "Epoch 6/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 205.8811 - mae: 12.0214\n", "Epoch 7/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 202.9065 - mae: 11.9390\n", "Epoch 8/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 200.1639 - mae: 11.8437\n", "Epoch 9/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 204.0377 - mae: 12.0880\n", "Epoch 10/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 194.2572 - mae: 11.6219\n", "Epoch 11/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 193.3072 - mae: 11.7112\n", "Epoch 12/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 180.4089 - mae: 11.2391\n", "Epoch 13/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 183.5659 - mae: 11.3790\n", "Epoch 14/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 200.1705 - mae: 11.9349\n", "Epoch 15/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 189.3466 - mae: 11.5642\n", "Epoch 16/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 186.4536 - mae: 11.5121\n", "Epoch 17/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 182.6394 - mae: 11.3618\n", "Epoch 18/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 183.0726 - mae: 11.3738\n", "Epoch 19/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 190.6540 - mae: 11.7405\n", "Epoch 20/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 170.5429 - mae: 10.9358\n", "Epoch 21/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 178.6833 - mae: 11.2016\n", "Epoch 22/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 178.1088 - mae: 11.1247\n", "Epoch 23/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 169.0413 - mae: 10.8795\n", "Epoch 24/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 170.3852 - mae: 10.8992\n", "Epoch 25/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 170.3944 - mae: 10.9377\n", "Epoch 26/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 166.8693 - mae: 10.7725\n", "Epoch 27/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 168.8513 - mae: 11.0017\n", "Epoch 28/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 170.0227 - mae: 10.9859\n", "Epoch 29/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 166.5854 - mae: 10.8612\n", "Epoch 30/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 169.6390 - mae: 10.9325\n", "Epoch 31/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 166.0459 - mae: 10.9219\n", "Epoch 32/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 155.6238 - mae: 10.3555\n", "Epoch 33/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 155.9009 - mae: 10.5042\n", "Epoch 34/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 157.8084 - mae: 10.6066\n", "Epoch 35/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 160.4327 - mae: 10.6518\n", "Epoch 36/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 150.8383 - mae: 10.3457\n", "Epoch 37/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 154.3426 - mae: 10.4430\n", "Epoch 38/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 146.3718 - mae: 10.1717\n", "Epoch 39/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 146.5977 - mae: 10.0473\n", "Epoch 40/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 148.1808 - mae: 10.3142\n", "Epoch 41/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 149.0166 - mae: 10.2803\n", "Epoch 42/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 140.3740 - mae: 9.9120\n", "Epoch 43/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 138.9175 - mae: 9.9573\n", "Epoch 44/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 135.0847 - mae: 9.7895\n", "Epoch 45/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 135.4719 - mae: 9.8082\n", "Epoch 46/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 133.9024 - mae: 9.6543\n", "Epoch 47/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 126.6546 - mae: 9.4716\n", "Epoch 48/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 126.8151 - mae: 9.4319\n", "Epoch 49/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 123.8165 - mae: 9.3865\n", "Epoch 50/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 122.9658 - mae: 9.2922\n", "Epoch 51/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 120.4032 - mae: 9.2159\n", "Epoch 52/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 119.4950 - mae: 9.2241\n", "Epoch 53/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 121.2059 - mae: 9.2647\n", "Epoch 54/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 118.7676 - mae: 9.1943\n", "Epoch 55/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 113.9606 - mae: 9.0571\n", "Epoch 56/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 111.6136 - mae: 8.8441\n", "Epoch 57/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 108.0592 - mae: 8.6614\n", "Epoch 58/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 104.8177 - mae: 8.5840\n", "Epoch 59/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 109.7415 - mae: 8.9108\n", "Epoch 60/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 104.7365 - mae: 8.6821\n", "Epoch 61/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 102.2955 - mae: 8.5328\n", "Epoch 62/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 96.9496 - mae: 8.2725\n", "Epoch 63/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 95.2658 - mae: 8.1849\n", "Epoch 64/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 94.4696 - mae: 8.1289\n", "Epoch 65/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 88.7216 - mae: 7.8865\n", "Epoch 66/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 88.2820 - mae: 7.8840\n", "Epoch 67/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 87.5142 - mae: 7.8452\n", "Epoch 68/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 84.3065 - mae: 7.6849\n", "Epoch 69/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 80.8092 - mae: 7.5148\n", "Epoch 70/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 83.7547 - mae: 7.6691\n", "Epoch 71/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 74.5664 - mae: 7.2070\n", "Epoch 72/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 77.3155 - mae: 7.3562\n", "Epoch 73/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 74.6474 - mae: 7.2569\n", "Epoch 74/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 71.7180 - mae: 7.1387\n", "Epoch 75/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 68.0042 - mae: 6.9376\n", "Epoch 76/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 69.1923 - mae: 6.9824\n", "Epoch 77/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 62.9197 - mae: 6.6071\n", "Epoch 78/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 61.6972 - mae: 6.5782\n", "Epoch 79/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 62.5539 - mae: 6.6484\n", "Epoch 80/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 58.4784 - mae: 6.4152\n", "Epoch 81/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 55.0328 - mae: 6.1396\n", "Epoch 82/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 56.1807 - mae: 6.2426\n", "Epoch 83/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 53.1191 - mae: 6.0858\n", "Epoch 84/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 51.1812 - mae: 6.0132\n", "Epoch 85/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 47.3761 - mae: 5.7633\n", "Epoch 86/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 46.1643 - mae: 5.6836\n", "Epoch 87/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 41.4803 - mae: 5.3428\n", "Epoch 88/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 42.9237 - mae: 5.5300\n", "Epoch 89/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 42.0332 - mae: 5.4568\n", "Epoch 90/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 38.4672 - mae: 5.2137\n", "Epoch 91/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 39.5544 - mae: 5.2831\n", "Epoch 92/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 35.4791 - mae: 4.9826\n", "Epoch 93/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 33.3324 - mae: 4.7970\n", "Epoch 94/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 33.6206 - mae: 4.9215\n", "Epoch 95/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 32.8745 - mae: 4.8137\n", "Epoch 96/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 31.3830 - mae: 4.7038\n", "Epoch 97/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 29.9062 - mae: 4.5760\n", "Epoch 98/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 27.8692 - mae: 4.3784\n", "Epoch 99/500\n", "36/36 [==============================] - 0s 3ms/step - loss: 25.7679 - mae: 4.2315\n", "Epoch 100/500\n", "36/36 [==============================] - 0s 3ms/step - loss: 24.8135 - mae: 4.1266\n", "Epoch 101/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 23.3186 - mae: 4.0433\n", "Epoch 102/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 23.5291 - mae: 4.0854\n", "Epoch 103/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 21.0501 - mae: 3.8249\n", "Epoch 104/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 20.2240 - mae: 3.7721\n", "Epoch 105/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 18.3640 - mae: 3.5519\n", "Epoch 106/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 17.7413 - mae: 3.4853\n", "Epoch 107/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 16.9648 - mae: 3.4411\n", "Epoch 108/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 16.5582 - mae: 3.3890\n", "Epoch 109/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 14.8946 - mae: 3.2183\n", "Epoch 110/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 13.8921 - mae: 3.1032\n", "Epoch 111/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 13.7679 - mae: 3.0833\n", "Epoch 112/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 12.7269 - mae: 2.9689\n", "Epoch 113/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 12.5800 - mae: 2.9404\n", "Epoch 114/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 12.1117 - mae: 2.9048\n", "Epoch 115/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 11.1527 - mae: 2.7503\n", "Epoch 116/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 10.2530 - mae: 2.5783\n", "Epoch 117/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 9.5033 - mae: 2.5409\n", "Epoch 118/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 9.1536 - mae: 2.5164\n", "Epoch 119/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 9.3318 - mae: 2.5197\n", "Epoch 120/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 8.3280 - mae: 2.3855\n", "Epoch 121/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 7.6669 - mae: 2.2458\n", "Epoch 122/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 7.5082 - mae: 2.2521\n", "Epoch 123/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 6.9904 - mae: 2.1564\n", "Epoch 124/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 6.5189 - mae: 2.0894\n", "Epoch 125/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 6.6024 - mae: 2.1000\n", "Epoch 126/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 7.0815 - mae: 2.1527\n", "Epoch 127/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 5.6175 - mae: 1.8845\n", "Epoch 128/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 5.6053 - mae: 1.9184\n", "Epoch 129/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 5.7198 - mae: 1.9330\n", "Epoch 130/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 5.1665 - mae: 1.8334\n", "Epoch 131/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 5.2837 - mae: 1.8235\n", "Epoch 132/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 4.8646 - mae: 1.7515\n", "Epoch 133/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 5.1366 - mae: 1.8146\n", "Epoch 134/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 4.7993 - mae: 1.7777\n", "Epoch 135/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 4.5782 - mae: 1.6847\n", "Epoch 136/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 4.5885 - mae: 1.6921\n", "Epoch 137/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 4.6645 - mae: 1.7089\n", "Epoch 138/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 4.3893 - mae: 1.6721\n", "Epoch 139/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 4.5310 - mae: 1.6847\n", "Epoch 140/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 4.3318 - mae: 1.6580\n", "Epoch 141/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 4.1245 - mae: 1.6226\n", "Epoch 142/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 4.2141 - mae: 1.6232\n", "Epoch 143/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 4.3534 - mae: 1.6542\n", "Epoch 144/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.9410 - mae: 1.5714\n", "Epoch 145/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 4.0269 - mae: 1.5900\n", "Epoch 146/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.8861 - mae: 1.5252\n", "Epoch 147/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 4.1662 - mae: 1.6092\n", "Epoch 148/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 4.2271 - mae: 1.5985\n", "Epoch 149/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7872 - mae: 1.5140\n", "Epoch 150/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7548 - mae: 1.4838\n", "Epoch 151/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7050 - mae: 1.5029\n", "Epoch 152/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.8002 - mae: 1.5323\n", "Epoch 153/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.9388 - mae: 1.5246\n", "Epoch 154/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 4.0161 - mae: 1.5398\n", "Epoch 155/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.9633 - mae: 1.5377\n", "Epoch 156/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.9122 - mae: 1.5290\n", "Epoch 157/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.7774 - mae: 1.5279\n", "Epoch 158/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.8671 - mae: 1.5172\n", "Epoch 159/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.8193 - mae: 1.5320\n", "Epoch 160/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6971 - mae: 1.4857\n", "Epoch 161/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.8077 - mae: 1.4956\n", "Epoch 162/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.5709 - mae: 1.4459\n", "Epoch 163/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.7687 - mae: 1.5170\n", "Epoch 164/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.1963 - mae: 1.3978\n", "Epoch 165/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5545 - mae: 1.4586\n", "Epoch 166/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6968 - mae: 1.4993\n", "Epoch 167/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.7519 - mae: 1.4905\n", "Epoch 168/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.4780 - mae: 1.4382\n", "Epoch 169/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.5226 - mae: 1.4353\n", "Epoch 170/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5693 - mae: 1.4541\n", "Epoch 171/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6563 - mae: 1.4577\n", "Epoch 172/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 4.1397 - mae: 1.5694\n", "Epoch 173/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5390 - mae: 1.4630\n", "Epoch 174/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7185 - mae: 1.4393\n", "Epoch 175/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.3569 - mae: 1.4133\n", "Epoch 176/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.2024 - mae: 1.3743\n", "Epoch 177/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.1662 - mae: 1.3622\n", "Epoch 178/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.9383 - mae: 1.5187\n", "Epoch 179/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.7844 - mae: 1.4592\n", "Epoch 180/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.7099 - mae: 1.4518\n", "Epoch 181/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.5002 - mae: 1.4187\n", "Epoch 182/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4666 - mae: 1.3925\n", "Epoch 183/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.9016 - mae: 1.5027\n", "Epoch 184/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.5504 - mae: 1.4555\n", "Epoch 185/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.7675 - mae: 1.4573\n", "Epoch 186/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.6190 - mae: 1.4362\n", "Epoch 187/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4681 - mae: 1.4002\n", "Epoch 188/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7177 - mae: 1.4355\n", "Epoch 189/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.7416 - mae: 1.4539\n", "Epoch 190/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.5239 - mae: 1.4098\n", "Epoch 191/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5527 - mae: 1.4455\n", "Epoch 192/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.3555 - mae: 1.4126\n", "Epoch 193/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.5754 - mae: 1.4459\n", "Epoch 194/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6068 - mae: 1.4397\n", "Epoch 195/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.5230 - mae: 1.4019\n", "Epoch 196/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5066 - mae: 1.4091\n", "Epoch 197/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4848 - mae: 1.4320\n", "Epoch 198/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.5449 - mae: 1.4294\n", "Epoch 199/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.9047 - mae: 1.4507\n", "Epoch 200/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.3455 - mae: 1.3818\n", "Epoch 201/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5188 - mae: 1.4151\n", "Epoch 202/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.5899 - mae: 1.4200\n", "Epoch 203/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.3916 - mae: 1.3955\n", "Epoch 204/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6553 - mae: 1.4193\n", "Epoch 205/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4723 - mae: 1.4118\n", "Epoch 206/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5355 - mae: 1.4098\n", "Epoch 207/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.8898 - mae: 1.4861\n", "Epoch 208/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6842 - mae: 1.4665\n", "Epoch 209/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.6706 - mae: 1.4485\n", "Epoch 210/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.3794 - mae: 1.3951\n", "Epoch 211/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6012 - mae: 1.4267\n", "Epoch 212/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.5387 - mae: 1.4160\n", "Epoch 213/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.2343 - mae: 1.3398\n", "Epoch 214/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4378 - mae: 1.4265\n", "Epoch 215/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7764 - mae: 1.4496\n", "Epoch 216/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.7615 - mae: 1.4381\n", "Epoch 217/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6369 - mae: 1.4026\n", "Epoch 218/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.1416 - mae: 1.3324\n", "Epoch 219/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.3017 - mae: 1.3853\n", "Epoch 220/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5158 - mae: 1.4277\n", "Epoch 221/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4010 - mae: 1.3995\n", "Epoch 222/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.6178 - mae: 1.4268\n", "Epoch 223/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.9025 - mae: 1.4625\n", "Epoch 224/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.4381 - mae: 1.4081\n", "Epoch 225/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.6625 - mae: 1.4477\n", "Epoch 226/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.8596 - mae: 1.4919\n", "Epoch 227/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.8018 - mae: 1.4716\n", "Epoch 228/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7889 - mae: 1.4427\n", "Epoch 229/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.6909 - mae: 1.4547\n", "Epoch 230/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.7217 - mae: 1.4269\n", "Epoch 231/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5433 - mae: 1.4117\n", "Epoch 232/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7283 - mae: 1.4443\n", "Epoch 233/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6020 - mae: 1.4364\n", "Epoch 234/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6196 - mae: 1.4266\n", "Epoch 235/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.8303 - mae: 1.4620\n", "Epoch 236/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4757 - mae: 1.4143\n", "Epoch 237/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.6349 - mae: 1.4348\n", "Epoch 238/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.4497 - mae: 1.3937\n", "Epoch 239/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5575 - mae: 1.4159\n", "Epoch 240/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6106 - mae: 1.4312\n", "Epoch 241/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.4777 - mae: 1.4274\n", "Epoch 242/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5741 - mae: 1.4152\n", "Epoch 243/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7051 - mae: 1.4295\n", "Epoch 244/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.7211 - mae: 1.4609\n", "Epoch 245/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5361 - mae: 1.4335\n", "Epoch 246/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6790 - mae: 1.4346\n", "Epoch 247/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.3833 - mae: 1.3859\n", "Epoch 248/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.5287 - mae: 1.4137\n", "Epoch 249/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.6381 - mae: 1.4040\n", "Epoch 250/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5443 - mae: 1.4334\n", "Epoch 251/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6561 - mae: 1.4360\n", "Epoch 252/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.9627 - mae: 1.4823\n", "Epoch 253/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6198 - mae: 1.4237\n", "Epoch 254/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.2362 - mae: 1.3392\n", "Epoch 255/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.4394 - mae: 1.3943\n", "Epoch 256/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4533 - mae: 1.3845\n", "Epoch 257/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6452 - mae: 1.4044\n", "Epoch 258/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.8027 - mae: 1.4521\n", "Epoch 259/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.9811 - mae: 1.4949\n", "Epoch 260/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5408 - mae: 1.4057\n", "Epoch 261/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.4094 - mae: 1.4121\n", "Epoch 262/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.4768 - mae: 1.3936\n", "Epoch 263/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.8673 - mae: 1.4994\n", "Epoch 264/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.4126 - mae: 1.3863\n", "Epoch 265/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.2887 - mae: 1.3820\n", "Epoch 266/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6788 - mae: 1.4264\n", "Epoch 267/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4893 - mae: 1.4006\n", "Epoch 268/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4289 - mae: 1.3760\n", "Epoch 269/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.6211 - mae: 1.4492\n", "Epoch 270/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.6488 - mae: 1.4583\n", "Epoch 271/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.2856 - mae: 1.3643\n", "Epoch 272/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.3826 - mae: 1.3774\n", "Epoch 273/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.3454 - mae: 1.3836\n", "Epoch 274/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.5098 - mae: 1.3906\n", "Epoch 275/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4052 - mae: 1.3848\n", "Epoch 276/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4037 - mae: 1.3872\n", "Epoch 277/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4087 - mae: 1.4052\n", "Epoch 278/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6599 - mae: 1.4463\n", "Epoch 279/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.5617 - mae: 1.4059\n", "Epoch 280/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5298 - mae: 1.4037\n", "Epoch 281/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.8215 - mae: 1.4478\n", "Epoch 282/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6421 - mae: 1.4192\n", "Epoch 283/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.3050 - mae: 1.3603\n", "Epoch 284/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.3113 - mae: 1.3863\n", "Epoch 285/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6208 - mae: 1.4388\n", "Epoch 286/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.5480 - mae: 1.4114\n", "Epoch 287/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.6136 - mae: 1.3900\n", "Epoch 288/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.3968 - mae: 1.4016\n", "Epoch 289/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.3819 - mae: 1.3754\n", "Epoch 290/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7737 - mae: 1.4906\n", "Epoch 291/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6412 - mae: 1.4341\n", "Epoch 292/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5826 - mae: 1.3986\n", "Epoch 293/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.6425 - mae: 1.4209\n", "Epoch 294/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6537 - mae: 1.4424\n", "Epoch 295/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6325 - mae: 1.4388\n", "Epoch 296/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7367 - mae: 1.4451\n", "Epoch 297/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.3241 - mae: 1.3825\n", "Epoch 298/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4420 - mae: 1.3932\n", "Epoch 299/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.5182 - mae: 1.4110\n", "Epoch 300/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.5779 - mae: 1.3950\n", "Epoch 301/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7790 - mae: 1.4862\n", "Epoch 302/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7829 - mae: 1.4591\n", "Epoch 303/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7746 - mae: 1.4482\n", "Epoch 304/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.5904 - mae: 1.4311\n", "Epoch 305/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.3911 - mae: 1.3941\n", "Epoch 306/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.4050 - mae: 1.3531\n", "Epoch 307/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.8462 - mae: 1.4792\n", "Epoch 308/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4320 - mae: 1.3813\n", "Epoch 309/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7648 - mae: 1.4414\n", "Epoch 310/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.8352 - mae: 1.4722\n", "Epoch 311/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5196 - mae: 1.4158\n", "Epoch 312/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.5463 - mae: 1.4100\n", "Epoch 313/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4538 - mae: 1.4181\n", "Epoch 314/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.1341 - mae: 1.3231\n", "Epoch 315/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4226 - mae: 1.3934\n", "Epoch 316/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5578 - mae: 1.3885\n", "Epoch 317/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.4651 - mae: 1.3783\n", "Epoch 318/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5444 - mae: 1.3969\n", "Epoch 319/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.7921 - mae: 1.4471\n", "Epoch 320/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4965 - mae: 1.4350\n", "Epoch 321/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5221 - mae: 1.4279\n", "Epoch 322/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7200 - mae: 1.4570\n", "Epoch 323/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4077 - mae: 1.3851\n", "Epoch 324/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.9371 - mae: 1.4967\n", "Epoch 325/500\n", "36/36 [==============================] - 0s 9ms/step - loss: 3.5588 - mae: 1.4299\n", "Epoch 326/500\n", "36/36 [==============================] - 0s 11ms/step - loss: 3.6518 - mae: 1.4172\n", "Epoch 327/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.8062 - mae: 1.4375\n", "Epoch 328/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4781 - mae: 1.4003\n", "Epoch 329/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.2518 - mae: 1.3565\n", "Epoch 330/500\n", "36/36 [==============================] - 0s 6ms/step - loss: 3.5089 - mae: 1.4080\n", "Epoch 331/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.4939 - mae: 1.4215\n", "Epoch 332/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4224 - mae: 1.3878\n", "Epoch 333/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.7799 - mae: 1.4516\n", "Epoch 334/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.2162 - mae: 1.3472\n", "Epoch 335/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4004 - mae: 1.3891\n", "Epoch 336/500\n", "36/36 [==============================] - 0s 3ms/step - loss: 3.5844 - mae: 1.4309\n", "Epoch 337/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4287 - mae: 1.4102\n", "Epoch 338/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6819 - mae: 1.4540\n", "Epoch 339/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.4987 - mae: 1.4206\n", "Epoch 340/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4819 - mae: 1.4067\n", "Epoch 341/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5398 - mae: 1.4363\n", "Epoch 342/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6026 - mae: 1.4230\n", "Epoch 343/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7363 - mae: 1.4351\n", "Epoch 344/500\n", "36/36 [==============================] - 0s 5ms/step - loss: 3.7515 - mae: 1.4663\n", "Epoch 345/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.3528 - mae: 1.3850\n", "Epoch 346/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7189 - mae: 1.4410\n", "Epoch 347/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.6506 - mae: 1.4457\n", "Epoch 348/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.3062 - mae: 1.3482\n", "Epoch 349/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.8847 - mae: 1.4673\n", "Epoch 350/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7583 - mae: 1.4488\n", "Epoch 351/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.5707 - mae: 1.4243\n", "Epoch 352/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5299 - mae: 1.4121\n", "Epoch 353/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6041 - mae: 1.4105\n", "Epoch 354/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.8048 - mae: 1.4557\n", "Epoch 355/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.7047 - mae: 1.4153\n", "Epoch 356/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.2750 - mae: 1.3761\n", "Epoch 357/500\n", "36/36 [==============================] - 0s 4ms/step - loss: 3.5078 - mae: 1.3904\n", "Epoch 358/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5551 - mae: 1.4355\n", "Epoch 359/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.6110 - mae: 1.4249\n", "Epoch 360/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6454 - mae: 1.4311\n", "Epoch 361/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6162 - mae: 1.4541\n", "Epoch 362/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.5998 - mae: 1.4331\n", "Epoch 363/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.1045 - mae: 1.3340\n", "Epoch 364/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4750 - mae: 1.4119\n", "Epoch 365/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.2192 - mae: 1.3469\n", "Epoch 366/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6063 - mae: 1.4386\n", "Epoch 367/500\n", "36/36 [==============================] - 0s 3ms/step - loss: 3.6762 - mae: 1.4376\n", "Epoch 368/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.8595 - mae: 1.4814\n", "Epoch 369/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6991 - mae: 1.4489\n", "Epoch 370/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4685 - mae: 1.3871\n", "Epoch 371/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.3583 - mae: 1.3991\n", "Epoch 372/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7415 - mae: 1.4291\n", "Epoch 373/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5929 - mae: 1.4234\n", "Epoch 374/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.9234 - mae: 1.4729\n", "Epoch 375/500\n", "36/36 [==============================] - 0s 3ms/step - loss: 3.6273 - mae: 1.4402\n", "Epoch 376/500\n", "36/36 [==============================] - 0s 5ms/step - loss: 3.6178 - mae: 1.4456\n", "Epoch 377/500\n", "36/36 [==============================] - 0s 3ms/step - loss: 3.7578 - mae: 1.4838\n", "Epoch 378/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7482 - mae: 1.4392\n", "Epoch 379/500\n", "36/36 [==============================] - 0s 3ms/step - loss: 3.6029 - mae: 1.4200\n", "Epoch 380/500\n", "36/36 [==============================] - 0s 3ms/step - loss: 3.5536 - mae: 1.4279\n", "Epoch 381/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.2112 - mae: 1.3253\n", "Epoch 382/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7413 - mae: 1.4666\n", "Epoch 383/500\n", "36/36 [==============================] - 0s 3ms/step - loss: 3.6776 - mae: 1.4437\n", "Epoch 384/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6060 - mae: 1.3963\n", "Epoch 385/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5433 - mae: 1.4255\n", "Epoch 386/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6623 - mae: 1.4329\n", "Epoch 387/500\n", "36/36 [==============================] - 0s 3ms/step - loss: 3.5392 - mae: 1.4181\n", "Epoch 388/500\n", "36/36 [==============================] - 0s 3ms/step - loss: 3.8578 - mae: 1.4676\n", "Epoch 389/500\n", "36/36 [==============================] - 0s 3ms/step - loss: 3.9003 - mae: 1.4429\n", "Epoch 390/500\n", "36/36 [==============================] - 0s 3ms/step - loss: 3.5714 - mae: 1.4257\n", "Epoch 391/500\n", "36/36 [==============================] - 0s 3ms/step - loss: 3.5077 - mae: 1.4076\n", "Epoch 392/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5514 - mae: 1.4082\n", "Epoch 393/500\n", "36/36 [==============================] - 0s 3ms/step - loss: 3.4374 - mae: 1.3990\n", "Epoch 394/500\n", "36/36 [==============================] - 0s 3ms/step - loss: 3.7092 - mae: 1.4377\n", "Epoch 395/500\n", "36/36 [==============================] - 0s 3ms/step - loss: 3.5141 - mae: 1.4239\n", "Epoch 396/500\n", "36/36 [==============================] - 0s 3ms/step - loss: 3.4980 - mae: 1.3940\n", "Epoch 397/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.3912 - mae: 1.4024\n", "Epoch 398/500\n", "36/36 [==============================] - 0s 6ms/step - loss: 3.4483 - mae: 1.4017\n", "Epoch 399/500\n", "36/36 [==============================] - 0s 3ms/step - loss: 3.7251 - mae: 1.4482\n", "Epoch 400/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6032 - mae: 1.4098\n", "Epoch 401/500\n", "36/36 [==============================] - 0s 3ms/step - loss: 3.4085 - mae: 1.3562\n", "Epoch 402/500\n", "36/36 [==============================] - 0s 3ms/step - loss: 3.8176 - mae: 1.4719\n", "Epoch 403/500\n", "36/36 [==============================] - 0s 3ms/step - loss: 3.3764 - mae: 1.3805\n", "Epoch 404/500\n", "36/36 [==============================] - 0s 3ms/step - loss: 3.6528 - mae: 1.4567\n", "Epoch 405/500\n", "36/36 [==============================] - 0s 4ms/step - loss: 3.7080 - mae: 1.4490\n", "Epoch 406/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.3395 - mae: 1.3751\n", "Epoch 407/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5167 - mae: 1.4199\n", "Epoch 408/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5829 - mae: 1.4240\n", "Epoch 409/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7214 - mae: 1.4492\n", "Epoch 410/500\n", "36/36 [==============================] - 0s 3ms/step - loss: 3.5668 - mae: 1.4215\n", "Epoch 411/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.8139 - mae: 1.4414\n", "Epoch 412/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 4.0559 - mae: 1.4948\n", "Epoch 413/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6380 - mae: 1.4035\n", "Epoch 414/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5839 - mae: 1.4214\n", "Epoch 415/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5993 - mae: 1.4317\n", "Epoch 416/500\n", "36/36 [==============================] - 0s 3ms/step - loss: 3.4911 - mae: 1.3933\n", "Epoch 417/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.3748 - mae: 1.3742\n", "Epoch 418/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5217 - mae: 1.4175\n", "Epoch 419/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6469 - mae: 1.4249\n", "Epoch 420/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6943 - mae: 1.4348\n", "Epoch 421/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7600 - mae: 1.4369\n", "Epoch 422/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6959 - mae: 1.4532\n", "Epoch 423/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4773 - mae: 1.4091\n", "Epoch 424/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6859 - mae: 1.4393\n", "Epoch 425/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.3885 - mae: 1.3572\n", "Epoch 426/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.8545 - mae: 1.4599\n", "Epoch 427/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7112 - mae: 1.4456\n", "Epoch 428/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5166 - mae: 1.4222\n", "Epoch 429/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5126 - mae: 1.4114\n", "Epoch 430/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.3532 - mae: 1.3619\n", "Epoch 431/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6093 - mae: 1.4144\n", "Epoch 432/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.5987 - mae: 1.4307\n", "Epoch 433/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4686 - mae: 1.3978\n", "Epoch 434/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.1994 - mae: 1.3417\n", "Epoch 435/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.2727 - mae: 1.3670\n", "Epoch 436/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5122 - mae: 1.4432\n", "Epoch 437/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5827 - mae: 1.4288\n", "Epoch 438/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7049 - mae: 1.4381\n", "Epoch 439/500\n", "36/36 [==============================] - 0s 3ms/step - loss: 3.3737 - mae: 1.3786\n", "Epoch 440/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5140 - mae: 1.3919\n", "Epoch 441/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4147 - mae: 1.3862\n", "Epoch 442/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.3586 - mae: 1.3804\n", "Epoch 443/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.9619 - mae: 1.4858\n", "Epoch 444/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4523 - mae: 1.4020\n", "Epoch 445/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6105 - mae: 1.4196\n", "Epoch 446/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5114 - mae: 1.4054\n", "Epoch 447/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6318 - mae: 1.4181\n", "Epoch 448/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.5704 - mae: 1.4174\n", "Epoch 449/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4497 - mae: 1.4007\n", "Epoch 450/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6459 - mae: 1.4318\n", "Epoch 451/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7384 - mae: 1.4733\n", "Epoch 452/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5836 - mae: 1.4622\n", "Epoch 453/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.5289 - mae: 1.4246\n", "Epoch 454/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.3960 - mae: 1.4100\n", "Epoch 455/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.3303 - mae: 1.3755\n", "Epoch 456/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5566 - mae: 1.3948\n", "Epoch 457/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7051 - mae: 1.4401\n", "Epoch 458/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.3271 - mae: 1.3796\n", "Epoch 459/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6922 - mae: 1.4280\n", "Epoch 460/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7817 - mae: 1.4581\n", "Epoch 461/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7344 - mae: 1.4633\n", "Epoch 462/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.4826 - mae: 1.4061\n", "Epoch 463/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6336 - mae: 1.4193\n", "Epoch 464/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.3514 - mae: 1.3929\n", "Epoch 465/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.7179 - mae: 1.4460\n", "Epoch 466/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5846 - mae: 1.4244\n", "Epoch 467/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6907 - mae: 1.4306\n", "Epoch 468/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4206 - mae: 1.3835\n", "Epoch 469/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6383 - mae: 1.4285\n", "Epoch 470/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4773 - mae: 1.3963\n", "Epoch 471/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5952 - mae: 1.4196\n", "Epoch 472/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6941 - mae: 1.4412\n", "Epoch 473/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7083 - mae: 1.4697\n", "Epoch 474/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5177 - mae: 1.4141\n", "Epoch 475/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.3560 - mae: 1.3723\n", "Epoch 476/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5397 - mae: 1.4045\n", "Epoch 477/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.2715 - mae: 1.3827\n", "Epoch 478/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7242 - mae: 1.4754\n", "Epoch 479/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.3702 - mae: 1.3913\n", "Epoch 480/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4928 - mae: 1.4074\n", "Epoch 481/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5246 - mae: 1.4115\n", "Epoch 482/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.8048 - mae: 1.4472\n", "Epoch 483/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.1067 - mae: 1.3239\n", "Epoch 484/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4090 - mae: 1.3904\n", "Epoch 485/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4326 - mae: 1.3893\n", "Epoch 486/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.2704 - mae: 1.3728\n", "Epoch 487/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.5872 - mae: 1.4319\n", "Epoch 488/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7672 - mae: 1.4496\n", "Epoch 489/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.3966 - mae: 1.3896\n", "Epoch 490/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4789 - mae: 1.4178\n", "Epoch 491/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.2898 - mae: 1.3659\n", "Epoch 492/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5226 - mae: 1.4026\n", "Epoch 493/500\n", "36/36 [==============================] - 0s 1ms/step - loss: 3.6141 - mae: 1.4379\n", "Epoch 494/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.4354 - mae: 1.4044\n", "Epoch 495/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7728 - mae: 1.4539\n", "Epoch 496/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.5828 - mae: 1.4168\n", "Epoch 497/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.1168 - mae: 1.3414\n", "Epoch 498/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.6027 - mae: 1.4135\n", "Epoch 499/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.7028 - mae: 1.4425\n", "Epoch 500/500\n", "36/36 [==============================] - 0s 2ms/step - loss: 3.3174 - mae: 1.3821\n", "Training Done\n" ], "name": "stdout" }, { "output_type": "display_data", "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "tags": [], "needs_background": "light" } } ] } ] }