1  {
   2    "nbformat": 4,
   3    "nbformat_minor": 0,
   4    "metadata": {
   5      "accelerator": "GPU",
   6      "colab": {
   7        "name": "S+P Week 4 Lesson 5.ipynb",
   8        "provenance": [],
   9        "collapsed_sections": [],
  10        "toc_visible": true
  11      },
  12      "kernelspec": {
  13        "display_name": "Python 3",
  14        "name": "python3"
  15      }
  16    },
  17    "cells": [
  18      {
  19        "cell_type": "code",
  20        "metadata": {
  21          "id": "zX4Kg8DUTKWO"
  22        },
  23        "source": [
  24          "#@title Licensed under the Apache License, Version 2.0 (the \"License\");\n",
  25          "# you may not use this file except in compliance with the License.\n",
  26          "# You may obtain a copy of the License at\n",
  27          "#\n",
  28          "# https://www.apache.org/licenses/LICENSE-2.0\n",
  29          "#\n",
  30          "# Unless required by applicable law or agreed to in writing, software\n",
  31          "# distributed under the License is distributed on an \"AS IS\" BASIS,\n",
  32          "# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
  33          "# See the License for the specific language governing permissions and\n",
  34          "# limitations under the License."
  35        ],
  36        "execution_count": null,
  37        "outputs": []
  38      },
  39      {
  40        "cell_type": "code",
  41        "metadata": {
  42          "colab": {
  43            "base_uri": "https://localhost:8080/",
  44            "height": 34
  45          },
  46          "id": "56XEQOGknrAk",
  47          "outputId": "d52053ae-8ab5-4863-97c9-50073f912f46"
  48        },
  49        "source": [
  50          "import tensorflow as tf\n",
  51          "print(tf.__version__)"
  52        ],
  53        "execution_count": null,
  54        "outputs": [
  55          {
  56            "output_type": "stream",
  57            "text": [
  58              "2.0.0-beta1\n"
  59            ],
  60            "name": "stdout"
  61          }
  62        ]
  63      },
  64      {
  65        "cell_type": "code",
  66        "metadata": {
  67          "id": "sLl52leVp5wU"
  68        },
  69        "source": [
  70          "import numpy as np\n",
  71          "import matplotlib.pyplot as plt\n",
  72          "def plot_series(time, series, format=\"-\", start=0, end=None):\n",
  73          "    plt.plot(time[start:end], series[start:end], format)\n",
  74          "    plt.xlabel(\"Time\")\n",
  75          "    plt.ylabel(\"Value\")\n",
  76          "    plt.grid(True)"
  77        ],
  78        "execution_count": null,
  79        "outputs": []
  80      },
  81      {
  82        "cell_type": "code",
  83        "metadata": {
  84          "colab": {
  85            "base_uri": "https://localhost:8080/",
  86            "height": 208
  87          },
  88          "id": "tP7oqUdkk0gY",
  89          "outputId": "549237ab-672c-49d5-b36e-bd196eae772f"
  90        },
  91        "source": [
  92          "!wget --no-check-certificate \\\n",
  93          "    https://storage.googleapis.com/laurencemoroney-blog.appspot.com/Sunspots.csv \\\n",
  94          "    -O /tmp/sunspots.csv"
  95        ],
  96        "execution_count": null,
  97        "outputs": [
  98          {
  99            "output_type": "stream",
 100            "text": [
 101              "--2019-07-02 23:12:41--  https://storage.googleapis.com/laurencemoroney-blog.appspot.com/Sunspots.csv\n",
 102              "Resolving storage.googleapis.com (storage.googleapis.com)... 172.217.194.128, 2404:6800:4003:c04::80\n",
 103              "Connecting to storage.googleapis.com (storage.googleapis.com)|172.217.194.128|:443... connected.\n",
 104              "HTTP request sent, awaiting response... 200 OK\n",
 105              "Length: 70827 (69K) [application/octet-stream]\n",
 106              "Saving to: ‘/tmp/sunspots.csv’\n",
 107              "\n",
 108              "/tmp/sunspots.csv   100%[===================>]  69.17K  --.-KB/s    in 0.001s  \n",
 109              "\n",
 110              "2019-07-02 23:12:42 (116 MB/s) - ‘/tmp/sunspots.csv’ saved [70827/70827]\n",
 111              "\n"
 112            ],
 113            "name": "stdout"
 114          }
 115        ]
 116      },
 117      {
 118        "cell_type": "code",
 119        "metadata": {
 120          "colab": {
 121            "base_uri": "https://localhost:8080/",
 122            "height": 392
 123          },
 124          "id": "NcG9r1eClbTh",
 125          "outputId": "f3545940-e807-4109-f8be-a95a3a16d4cf"
 126        },
 127        "source": [
 128          "import csv\n",
 129          "time_step = []\n",
 130          "sunspots = []\n",
 131          "\n",
 132          "with open('/tmp/sunspots.csv') as csvfile:\n",
 133          "  reader = csv.reader(csvfile, delimiter=',')\n",
 134          "  next(reader)\n",
 135          "  for row in reader:\n",
 136          "    sunspots.append(float(row[2]))\n",
 137          "    time_step.append(int(row[0]))\n",
 138          "\n",
 139          "series = np.array(sunspots)\n",
 140          "time = np.array(time_step)\n",
 141          "plt.figure(figsize=(10, 6))\n",
 142          "plot_series(time, series)"
 143        ],
 144        "execution_count": null,
 145        "outputs": [
 146          {
 147            "output_type": "display_data",
 148            "data": {
 149              "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmcAAAF3CAYAAADgjOwXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\niGOOfQAAAABJRU5ErkJggg==\n",
 150              "text/plain": [
 151                "<Figure size 720x432 with 1 Axes>"
 152              ]
 153            },
 154            "metadata": {
 155              "tags": []
 156            }
 157          }
 158        ]
 159      },
 160      {
 161        "cell_type": "code",
 162        "metadata": {
 163          "colab": {
 164            "base_uri": "https://localhost:8080/",
 165            "height": 392
 166          },
 167          "id": "VinZVwUa8WO7",
 168          "outputId": "e95494c5-13c5-4b60-85f0-f9021720cff3"
 169        },
 170        "source": [
 171          "series = np.array(sunspots)\n",
 172          "time = np.array(time_step)\n",
 173          "plt.figure(figsize=(10, 6))\n",
 174          "plot_series(time, series)"
 175        ],
 176        "execution_count": null,
 177        "outputs": [
 178          {
 179            "output_type": "display_data",
 180            "data": {
 181              "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmcAAAF3CAYAAADgjOwXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\niGOOfQAAAABJRU5ErkJggg==\n",
 182              "text/plain": [
 183                "<Figure size 720x432 with 1 Axes>"
 184              ]
 185            },
 186            "metadata": {
 187              "tags": []
 188            }
 189          }
 190        ]
 191      },
 192      {
 193        "cell_type": "code",
 194        "metadata": {
 195          "id": "L92YRw_IpCFG"
 196        },
 197        "source": [
 198          "split_time = 3000\n",
 199          "time_train = time[:split_time]\n",
 200          "x_train = series[:split_time]\n",
 201          "time_valid = time[split_time:]\n",
 202          "x_valid = series[split_time:]\n",
 203          "\n",
 204          "window_size = 30\n",
 205          "batch_size = 32\n",
 206          "shuffle_buffer_size = 1000\n"
 207        ],
 208        "execution_count": null,
 209        "outputs": []
 210      },
 211      {
 212        "cell_type": "code",
 213        "metadata": {
 214          "id": "lJwUUZscnG38"
 215        },
 216        "source": [
 217          "def windowed_dataset(series, window_size, batch_size, shuffle_buffer):\n",
 218          "    series = tf.expand_dims(series, axis=-1)\n",
 219          "    ds = tf.data.Dataset.from_tensor_slices(series)\n",
 220          "    ds = ds.window(window_size + 1, shift=1, drop_remainder=True)\n",
 221          "    ds = ds.flat_map(lambda w: w.batch(window_size + 1))\n",
 222          "    ds = ds.shuffle(shuffle_buffer)\n",
 223          "    ds = ds.map(lambda w: (w[:-1], w[1:]))\n",
 224          "    return ds.batch(batch_size).prefetch(1)"
 225        ],
 226        "execution_count": null,
 227        "outputs": []
 228      },
 229      {
 230        "cell_type": "code",
 231        "metadata": {
 232          "id": "4XwGrf-A_wF0"
 233        },
 234        "source": [
 235          "def model_forecast(model, series, window_size):\n",
 236          "    ds = tf.data.Dataset.from_tensor_slices(series)\n",
 237          "    ds = ds.window(window_size, shift=1, drop_remainder=True)\n",
 238          "    ds = ds.flat_map(lambda w: w.batch(window_size))\n",
 239          "    ds = ds.batch(32).prefetch(1)\n",
 240          "    forecast = model.predict(ds)\n",
 241          "    return forecast"
 242        ],
 243        "execution_count": null,
 244        "outputs": []
 245      },
 246      {
 247        "cell_type": "code",
 248        "metadata": {
 249          "colab": {
 250            "base_uri": "https://localhost:8080/",
 251            "height": 1000
 252          },
 253          "id": "AclfYY3Mn6Ph",
 254          "outputId": "02a3801f-9033-459f-c1c6-f930e65fc01f"
 255        },
 256        "source": [
 257          "tf.keras.backend.clear_session()\n",
 258          "tf.random.set_seed(51)\n",
 259          "np.random.seed(51)\n",
 260          "window_size = 64\n",
 261          "batch_size = 256\n",
 262          "train_set = windowed_dataset(x_train, window_size, batch_size, shuffle_buffer_size)\n",
 263          "print(train_set)\n",
 264          "print(x_train.shape)\n",
 265          "\n",
 266          "model = tf.keras.models.Sequential([\n",
 267          "  tf.keras.layers.Conv1D(filters=32, kernel_size=5,\n",
 268          "                      strides=1, padding=\"causal\",\n",
 269          "                      activation=\"relu\",\n",
 270          "                      input_shape=[None, 1]),\n",
 271          "  tf.keras.layers.LSTM(64, return_sequences=True),\n",
 272          "  tf.keras.layers.LSTM(64, return_sequences=True),\n",
 273          "  tf.keras.layers.Dense(30, activation=\"relu\"),\n",
 274          "  tf.keras.layers.Dense(10, activation=\"relu\"),\n",
 275          "  tf.keras.layers.Dense(1),\n",
 276          "  tf.keras.layers.Lambda(lambda x: x * 400)\n",
 277          "])\n",
 278          "\n",
 279          "lr_schedule = tf.keras.callbacks.LearningRateScheduler(\n",
 280          "    lambda epoch: 1e-8 * 10**(epoch / 20))\n",
 281          "optimizer = tf.keras.optimizers.SGD(lr=1e-8, momentum=0.9)\n",
 282          "model.compile(loss=tf.keras.losses.Huber(),\n",
 283          "              optimizer=optimizer,\n",
 284          "              metrics=[\"mae\"])\n",
 285          "history = model.fit(train_set, epochs=100, callbacks=[lr_schedule])\n"
 286        ],
 287        "execution_count": null,
 288        "outputs": [
 289          {
 290            "output_type": "stream",
 291            "text": [
 292              "<PrefetchDataset shapes: ((None, None, 1), (None, None, 1)), types: (tf.float64, tf.float64)>\n",
 293              "(3000,)\n",
 294              "Epoch 1/100\n",
 295              "12/12 [==============================] - 7s 604ms/step - loss: 79.3289 - mae: 79.6432\n",
 296              "Epoch 2/100\n",
 297              "12/12 [==============================] - 4s 356ms/step - loss: 77.6025 - mae: 77.9456\n",
 298              "Epoch 3/100\n",
 299              "12/12 [==============================] - 4s 345ms/step - loss: 75.0294 - mae: 75.3777\n",
 300              "Epoch 4/100\n",
 301              "12/12 [==============================] - 4s 342ms/step - loss: 71.9486 - mae: 72.2988\n",
 302              "Epoch 5/100\n",
 303              "12/12 [==============================] - 4s 336ms/step - loss: 68.5473 - mae: 68.8979\n",
 304              "Epoch 6/100\n",
 305              "12/12 [==============================] - 4s 341ms/step - loss: 64.9847 - mae: 65.3328\n",
 306              "Epoch 7/100\n",
 307              "12/12 [==============================] - 4s 346ms/step - loss: 61.4695 - mae: 61.8105\n",
 308              "Epoch 8/100\n",
 309              "12/12 [==============================] - 4s 339ms/step - loss: 58.1538 - mae: 58.4827\n",
 310              "Epoch 9/100\n",
 311              "12/12 [==============================] - 4s 354ms/step - loss: 55.1324 - mae: 55.4492\n",
 312              "Epoch 10/100\n",
 313              "12/12 [==============================] - 4s 340ms/step - loss: 52.4293 - mae: 52.7354\n",
 314              "Epoch 11/100\n",
 315              "12/12 [==============================] - 4s 343ms/step - loss: 50.0465 - mae: 50.3450\n",
 316              "Epoch 12/100\n",
 317              "12/12 [==============================] - 4s 348ms/step - loss: 47.9855 - mae: 48.2800\n",
 318              "Epoch 13/100\n",
 319              "12/12 [==============================] - 4s 350ms/step - loss: 46.2026 - mae: 46.4946\n",
 320              "Epoch 14/100\n",
 321              "12/12 [==============================] - 4s 352ms/step - loss: 44.7008 - mae: 44.9915\n",
 322              "Epoch 15/100\n",
 323              "12/12 [==============================] - 4s 340ms/step - loss: 43.4629 - mae: 43.7563\n",
 324              "Epoch 16/100\n",
 325              "12/12 [==============================] - 4s 342ms/step - loss: 42.4297 - mae: 42.7283\n",
 326              "Epoch 17/100\n",
 327              "12/12 [==============================] - 4s 348ms/step - loss: 41.5160 - mae: 41.8219\n",
 328              "Epoch 18/100\n",
 329              "12/12 [==============================] - 4s 346ms/step - loss: 40.6779 - mae: 40.9926\n",
 330              "Epoch 19/100\n",
 331              "12/12 [==============================] - 4s 350ms/step - loss: 39.8902 - mae: 40.2143\n",
 332              "Epoch 20/100\n",
 333              "12/12 [==============================] - 4s 344ms/step - loss: 39.1337 - mae: 39.4679\n",
 334              "Epoch 21/100\n",
 335              "12/12 [==============================] - 4s 344ms/step - loss: 38.3574 - mae: 38.7057\n",
 336              "Epoch 22/100\n",
 337              "12/12 [==============================] - 4s 340ms/step - loss: 37.4999 - mae: 37.8590\n",
 338              "Epoch 23/100\n",
 339              "12/12 [==============================] - 4s 343ms/step - loss: 36.4614 - mae: 36.8280\n",
 340              "Epoch 24/100\n",
 341              "12/12 [==============================] - 4s 342ms/step - loss: 35.3582 - mae: 35.7350\n",
 342              "Epoch 25/100\n",
 343              "12/12 [==============================] - 4s 360ms/step - loss: 34.2471 - mae: 34.6287\n",
 344              "Epoch 26/100\n",
 345              "12/12 [==============================] - 4s 348ms/step - loss: 33.2302 - mae: 33.6199\n",
 346              "Epoch 27/100\n",
 347              "12/12 [==============================] - 4s 354ms/step - loss: 32.2114 - mae: 32.6066\n",
 348              "Epoch 28/100\n",
 349              "12/12 [==============================] - 4s 347ms/step - loss: 31.3240 - mae: 31.7227\n",
 350              "Epoch 29/100\n",
 351              "12/12 [==============================] - 4s 356ms/step - loss: 30.5670 - mae: 30.9680\n",
 352              "Epoch 30/100\n",
 353              "12/12 [==============================] - 4s 346ms/step - loss: 30.0500 - mae: 30.4513\n",
 354              "Epoch 31/100\n",
 355              "12/12 [==============================] - 4s 342ms/step - loss: 29.4792 - mae: 29.8943\n",
 356              "Epoch 32/100\n",
 357              "12/12 [==============================] - 4s 341ms/step - loss: 29.2023 - mae: 29.6147\n",
 358              "Epoch 33/100\n",
 359              "12/12 [==============================] - 4s 343ms/step - loss: 28.6435 - mae: 29.0801\n",
 360              "Epoch 34/100\n",
 361              "12/12 [==============================] - 4s 342ms/step - loss: 28.5520 - mae: 28.9969\n",
 362              "Epoch 35/100\n",
 363              "12/12 [==============================] - 4s 336ms/step - loss: 28.2922 - mae: 28.7440\n",
 364              "Epoch 36/100\n",
 365              "12/12 [==============================] - 4s 341ms/step - loss: 27.9763 - mae: 28.4263\n",
 366              "Epoch 37/100\n",
 367              "12/12 [==============================] - 4s 342ms/step - loss: 27.5422 - mae: 28.0023\n",
 368              "Epoch 38/100\n",
 369              "12/12 [==============================] - 4s 340ms/step - loss: 27.1134 - mae: 27.5632\n",
 370              "Epoch 39/100\n",
 371              "12/12 [==============================] - 4s 343ms/step - loss: 26.9710 - mae: 27.4343\n",
 372              "Epoch 40/100\n",
 373              "12/12 [==============================] - 4s 336ms/step - loss: 26.9833 - mae: 27.4235\n",
 374              "Epoch 41/100\n",
 375              "12/12 [==============================] - 4s 336ms/step - loss: 26.1409 - mae: 26.5900\n",
 376              "Epoch 42/100\n",
 377              "12/12 [==============================] - 4s 338ms/step - loss: 25.6200 - mae: 26.0847\n",
 378              "Epoch 43/100\n",
 379              "12/12 [==============================] - 4s 334ms/step - loss: 25.5947 - mae: 26.0638\n",
 380              "Epoch 44/100\n",
 381              "12/12 [==============================] - 4s 338ms/step - loss: 25.4066 - mae: 25.8794\n",
 382              "Epoch 45/100\n",
 383              "12/12 [==============================] - 4s 336ms/step - loss: 24.9945 - mae: 25.4612\n",
 384              "Epoch 46/100\n",
 385              "12/12 [==============================] - 4s 332ms/step - loss: 24.4322 - mae: 24.9142\n",
 386              "Epoch 47/100\n",
 387              "12/12 [==============================] - 4s 333ms/step - loss: 24.0959 - mae: 24.5685\n",
 388              "Epoch 48/100\n",
 389              "12/12 [==============================] - 4s 338ms/step - loss: 23.8872 - mae: 24.3501\n",
 390              "Epoch 49/100\n",
 391              "12/12 [==============================] - 4s 336ms/step - loss: 23.4193 - mae: 23.9016\n",
 392              "Epoch 50/100\n",
 393              "12/12 [==============================] - 4s 333ms/step - loss: 23.0463 - mae: 23.5124\n",
 394              "Epoch 51/100\n",
 395              "12/12 [==============================] - 4s 333ms/step - loss: 22.6978 - mae: 23.1679\n",
 396              "Epoch 52/100\n",
 397              "12/12 [==============================] - 4s 334ms/step - loss: 22.3717 - mae: 22.8404\n",
 398              "Epoch 53/100\n",
 399              "12/12 [==============================] - 4s 336ms/step - loss: 21.9212 - mae: 22.4039\n",
 400              "Epoch 54/100\n",
 401              "12/12 [==============================] - 4s 336ms/step - loss: 21.4567 - mae: 21.9503\n",
 402              "Epoch 55/100\n",
 403              "12/12 [==============================] - 4s 338ms/step - loss: 21.2488 - mae: 21.7416\n",
 404              "Epoch 56/100\n",
 405              "12/12 [==============================] - 4s 340ms/step - loss: 20.8880 - mae: 21.3779\n",
 406              "Epoch 57/100\n",
 407              "12/12 [==============================] - 4s 337ms/step - loss: 20.9349 - mae: 21.4147\n",
 408              "Epoch 58/100\n",
 409              "12/12 [==============================] - 4s 331ms/step - loss: 21.1731 - mae: 21.6536\n",
 410              "Epoch 59/100\n",
 411              "12/12 [==============================] - 4s 337ms/step - loss: 20.5169 - mae: 21.0077\n",
 412              "Epoch 60/100\n",
 413              "12/12 [==============================] - 4s 338ms/step - loss: 20.2004 - mae: 20.6985\n",
 414              "Epoch 61/100\n",
 415              "12/12 [==============================] - 4s 342ms/step - loss: 19.7558 - mae: 20.2489\n",
 416              "Epoch 62/100\n",
 417              "12/12 [==============================] - 4s 343ms/step - loss: 19.6051 - mae: 20.1048\n",
 418              "Epoch 63/100\n",
 419              "12/12 [==============================] - 4s 341ms/step - loss: 20.2526 - mae: 20.6180\n",
 420              "Epoch 64/100\n",
 421              "12/12 [==============================] - 4s 332ms/step - loss: 21.8298 - mae: 22.2950\n",
 422              "Epoch 65/100\n",
 423              "12/12 [==============================] - 4s 334ms/step - loss: 20.0344 - mae: 20.5010\n",
 424              "Epoch 66/100\n",
 425              "12/12 [==============================] - 4s 338ms/step - loss: 19.8413 - mae: 20.3550\n",
 426              "Epoch 67/100\n",
 427              "12/12 [==============================] - 4s 335ms/step - loss: 19.2233 - mae: 19.6374\n",
 428              "Epoch 68/100\n",
 429              "12/12 [==============================] - 4s 335ms/step - loss: 19.8518 - mae: 20.3470\n",
 430              "Epoch 69/100\n",
 431              "12/12 [==============================] - 4s 334ms/step - loss: 20.8911 - mae: 21.3336\n",
 432              "Epoch 70/100\n",
 433              "12/12 [==============================] - 4s 332ms/step - loss: 19.7789 - mae: 20.3114\n",
 434              "Epoch 71/100\n",
 435              "12/12 [==============================] - 4s 336ms/step - loss: 20.0997 - mae: 20.6617\n",
 436              "Epoch 72/100\n",
 437              "12/12 [==============================] - 4s 336ms/step - loss: 22.0256 - mae: 22.5009\n",
 438              "Epoch 73/100\n",
 439              "12/12 [==============================] - 4s 343ms/step - loss: 20.1406 - mae: 20.3911\n",
 440              "Epoch 74/100\n",
 441              "12/12 [==============================] - 4s 340ms/step - loss: 19.8856 - mae: 20.4056\n",
 442              "Epoch 75/100\n",
 443              "12/12 [==============================] - 4s 326ms/step - loss: 19.3671 - mae: 19.9150\n",
 444              "Epoch 76/100\n",
 445              "12/12 [==============================] - 4s 335ms/step - loss: 24.5668 - mae: 25.1798\n",
 446              "Epoch 77/100\n",
 447              "12/12 [==============================] - 4s 334ms/step - loss: 20.3807 - mae: 20.9050\n",
 448              "Epoch 78/100\n",
 449              "12/12 [==============================] - 4s 336ms/step - loss: 21.3236 - mae: 21.8123\n",
 450              "Epoch 79/100\n",
 451              "12/12 [==============================] - 4s 324ms/step - loss: 23.4973 - mae: 24.1804\n",
 452              "Epoch 80/100\n",
 453              "12/12 [==============================] - 4s 330ms/step - loss: 21.4505 - mae: 22.0003\n",
 454              "Epoch 81/100\n",
 455              "12/12 [==============================] - 4s 329ms/step - loss: 22.8129 - mae: 23.1917\n",
 456              "Epoch 82/100\n",
 457              "12/12 [==============================] - 4s 328ms/step - loss: 26.1930 - mae: 26.7929\n",
 458              "Epoch 83/100\n",
 459              "12/12 [==============================] - 4s 333ms/step - loss: 27.5889 - mae: 27.6906\n",
 460              "Epoch 84/100\n",
 461              "12/12 [==============================] - 4s 329ms/step - loss: 57.0122 - mae: 56.4861\n",
 462              "Epoch 85/100\n",
 463              "12/12 [==============================] - 4s 323ms/step - loss: 57.7061 - mae: 58.1530\n",
 464              "Epoch 86/100\n",
 465              "12/12 [==============================] - 4s 325ms/step - loss: 51.5218 - mae: 51.7742\n",
 466              "Epoch 87/100\n",
 467              "12/12 [==============================] - 4s 326ms/step - loss: 49.4597 - mae: 49.1312\n",
 468              "Epoch 88/100\n",
 469              "12/12 [==============================] - 4s 326ms/step - loss: 63.3244 - mae: 60.3740\n",
 470              "Epoch 89/100\n",
 471              "12/12 [==============================] - 4s 330ms/step - loss: 59.7555 - mae: 60.1709\n",
 472              "Epoch 90/100\n",
 473              "12/12 [==============================] - 4s 328ms/step - loss: 55.3084 - mae: 55.5208\n",
 474              "Epoch 91/100\n",
 475              "12/12 [==============================] - 4s 322ms/step - loss: 54.5066 - mae: 54.7503\n",
 476              "Epoch 92/100\n",
 477              "12/12 [==============================] - 4s 323ms/step - loss: 53.7672 - mae: 54.0108\n",
 478              "Epoch 93/100\n",
 479              "12/12 [==============================] - 4s 330ms/step - loss: 52.6478 - mae: 52.9221\n",
 480              "Epoch 94/100\n",
 481              "12/12 [==============================] - 4s 321ms/step - loss: 57.2889 - mae: 56.9231\n",
 482              "Epoch 95/100\n",
 483              "12/12 [==============================] - 4s 335ms/step - loss: 59.4641 - mae: 59.7467\n",
 484              "Epoch 96/100\n",
 485              "12/12 [==============================] - 4s 333ms/step - loss: 56.2535 - mae: 56.5364\n",
 486              "Epoch 97/100\n",
 487              "12/12 [==============================] - 4s 334ms/step - loss: 57.8982 - mae: 58.1364\n",
 488              "Epoch 98/100\n",
 489              "12/12 [==============================] - 4s 331ms/step - loss: 58.4549 - mae: 58.6909\n",
 490              "Epoch 99/100\n",
 491              "12/12 [==============================] - 4s 332ms/step - loss: 54.5750 - mae: 54.8093\n",
 492              "Epoch 100/100\n",
 493              "12/12 [==============================] - 4s 322ms/step - loss: 54.9520 - mae: 55.1922\n"
 494            ],
 495            "name": "stdout"
 496          }
 497        ]
 498      },
 499      {
 500        "cell_type": "code",
 501        "metadata": {
 502          "colab": {
 503            "base_uri": "https://localhost:8080/",
 504            "height": 290
 505          },
 506          "id": "vVcKmg7Q_7rD",
 507          "outputId": "14773dd5-ab46-469a-dc02-96f4b7607b07"
 508        },
 509        "source": [
 510          "plt.semilogx(history.history[\"lr\"], history.history[\"loss\"])\n",
 511          "plt.axis([1e-8, 1e-4, 0, 60])"
 512        ],
 513        "execution_count": null,
 514        "outputs": [
 515          {
 516            "output_type": "execute_result",
 517            "data": {
 518              "text/plain": [
 519                "[1e-08, 0.0001, 0, 60]"
 520              ]
 521            },
 522            "metadata": {
 523              "tags": []
 524            },
 525            "execution_count": 32
 526          },
 527          {
 528            "output_type": "display_data",
 529            "data": {
 530              "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAEACAYAAABfxaZOAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASUVORK5CYII=\n",
 531              "text/plain": [
 532                "<Figure size 432x288 with 1 Axes>"
 533              ]
 534            },
 535            "metadata": {
 536              "tags": []
 537            }
 538          }
 539        ]
 540      },
 541      {
 542        "cell_type": "code",
 543        "metadata": {
 544          "colab": {
 545            "base_uri": "https://localhost:8080/",
 546            "height": 1000
 547          },
 548          "id": "QsksvkcXAAgq",
 549          "outputId": "be930f4c-30b1-4e16-92c0-4778fbca4c27"
 550        },
 551        "source": [
 552          "tf.keras.backend.clear_session()\n",
 553          "tf.random.set_seed(51)\n",
 554          "np.random.seed(51)\n",
 555          "train_set = windowed_dataset(x_train, window_size=60, batch_size=100, shuffle_buffer=shuffle_buffer_size)\n",
 556          "model = tf.keras.models.Sequential([\n",
 557          "  tf.keras.layers.Conv1D(filters=60, kernel_size=5,\n",
 558          "                      strides=1, padding=\"causal\",\n",
 559          "                      activation=\"relu\",\n",
 560          "                      input_shape=[None, 1]),\n",
 561          "  tf.keras.layers.LSTM(60, return_sequences=True),\n",
 562          "  tf.keras.layers.LSTM(60, return_sequences=True),\n",
 563          "  tf.keras.layers.Dense(30, activation=\"relu\"),\n",
 564          "  tf.keras.layers.Dense(10, activation=\"relu\"),\n",
 565          "  tf.keras.layers.Dense(1),\n",
 566          "  tf.keras.layers.Lambda(lambda x: x * 400)\n",
 567          "])\n",
 568          "\n",
 569          "\n",
 570          "optimizer = tf.keras.optimizers.SGD(lr=1e-5, momentum=0.9)\n",
 571          "model.compile(loss=tf.keras.losses.Huber(),\n",
 572          "              optimizer=optimizer,\n",
 573          "              metrics=[\"mae\"])\n",
 574          "history = model.fit(train_set,epochs=500)"
 575        ],
 576        "execution_count": null,
 577        "outputs": [
 578          {
 579            "output_type": "stream",
 580            "text": [
 581              "Epoch 1/500\n",
 582              "30/30 [==============================] - 7s 227ms/step - loss: 37.8896 - mae: 38.6059\n",
 583              "Epoch 2/500\n",
 584              "30/30 [==============================] - 4s 128ms/step - loss: 24.3137 - mae: 24.8470\n",
 585              "Epoch 3/500\n",
 586              "30/30 [==============================] - 4s 124ms/step - loss: 21.0376 - mae: 21.5412\n",
 587              "Epoch 4/500\n",
 588              "30/30 [==============================] - 4s 124ms/step - loss: 20.2604 - mae: 20.7556\n",
 589              "Epoch 5/500\n",
 590              "30/30 [==============================] - 4s 124ms/step - loss: 19.6226 - mae: 20.1157\n",
 591              "Epoch 6/500\n",
 592              "30/30 [==============================] - 4s 125ms/step - loss: 19.2406 - mae: 19.7160\n",
 593              "Epoch 7/500\n",
 594              "30/30 [==============================] - 4s 126ms/step - loss: 18.8684 - mae: 19.3572\n",
 595              "Epoch 8/500\n",
 596              "30/30 [==============================] - 4s 123ms/step - loss: 18.6414 - mae: 19.1303\n",
 597              "Epoch 9/500\n",
 598              "30/30 [==============================] - 4s 124ms/step - loss: 18.3319 - mae: 18.8184\n",
 599              "Epoch 10/500\n",
 600              "30/30 [==============================] - 4s 127ms/step - loss: 18.0562 - mae: 18.5398\n",
 601              "Epoch 11/500\n",
 602              "30/30 [==============================] - 4s 124ms/step - loss: 17.9810 - mae: 18.4535\n",
 603              "Epoch 12/500\n",
 604              "30/30 [==============================] - 4s 124ms/step - loss: 18.1221 - mae: 18.5999\n",
 605              "Epoch 13/500\n",
 606              "30/30 [==============================] - 4s 123ms/step - loss: 18.0651 - mae: 18.5252\n",
 607              "Epoch 14/500\n",
 608              "30/30 [==============================] - 4s 124ms/step - loss: 17.7861 - mae: 18.2709\n",
 609              "Epoch 15/500\n",
 610              "30/30 [==============================] - 4s 125ms/step - loss: 17.8317 - mae: 18.2976\n",
 611              "Epoch 16/500\n",
 612              "30/30 [==============================] - 4s 122ms/step - loss: 17.8733 - mae: 18.3504\n",
 613              "Epoch 17/500\n",
 614              "30/30 [==============================] - 4s 123ms/step - loss: 17.8476 - mae: 18.3366\n",
 615              "Epoch 18/500\n",
 616              "30/30 [==============================] - 4s 124ms/step - loss: 17.4424 - mae: 17.9248\n",
 617              "Epoch 19/500\n",
 618              "30/30 [==============================] - 4s 127ms/step - loss: 17.3558 - mae: 17.8342\n",
 619              "Epoch 20/500\n",
 620              "30/30 [==============================] - 4s 124ms/step - loss: 17.6471 - mae: 18.1257\n",
 621              "Epoch 21/500\n",
 622              "30/30 [==============================] - 4s 124ms/step - loss: 17.5979 - mae: 18.0850\n",
 623              "Epoch 22/500\n",
 624              "30/30 [==============================] - 4s 127ms/step - loss: 17.2557 - mae: 17.7383\n",
 625              "Epoch 23/500\n",
 626              "30/30 [==============================] - 4s 127ms/step - loss: 17.4762 - mae: 17.9535\n",
 627              "Epoch 24/500\n",
 628              "30/30 [==============================] - 4s 126ms/step - loss: 17.4770 - mae: 17.9421\n",
 629              "Epoch 25/500\n",
 630              "30/30 [==============================] - 4s 123ms/step - loss: 17.5380 - mae: 18.0132\n",
 631              "Epoch 26/500\n",
 632              "30/30 [==============================] - 4s 125ms/step - loss: 17.2549 - mae: 17.7340\n",
 633              "Epoch 27/500\n",
 634              "30/30 [==============================] - 4s 127ms/step - loss: 17.1579 - mae: 17.6354\n",
 635              "Epoch 28/500\n",
 636              "30/30 [==============================] - 4s 125ms/step - loss: 17.2286 - mae: 17.7137\n",
 637              "Epoch 29/500\n",
 638              "30/30 [==============================] - 4s 125ms/step - loss: 17.0821 - mae: 17.5609\n",
 639              "Epoch 30/500\n",
 640              "30/30 [==============================] - 4s 124ms/step - loss: 17.1587 - mae: 17.6405\n",
 641              "Epoch 31/500\n",
 642              "30/30 [==============================] - 4s 123ms/step - loss: 17.1101 - mae: 17.5918\n",
 643              "Epoch 32/500\n",
 644              "30/30 [==============================] - 4s 125ms/step - loss: 17.0714 - mae: 17.5552\n",
 645              "Epoch 33/500\n",
 646              "30/30 [==============================] - 4s 124ms/step - loss: 17.0534 - mae: 17.5376\n",
 647              "Epoch 34/500\n",
 648              "30/30 [==============================] - 4s 125ms/step - loss: 17.0213 - mae: 17.5062\n",
 649              "Epoch 35/500\n",
 650              "30/30 [==============================] - 4s 125ms/step - loss: 16.9862 - mae: 17.4670\n",
 651              "Epoch 36/500\n",
 652              "30/30 [==============================] - 4s 123ms/step - loss: 16.9095 - mae: 17.3828\n",
 653              "Epoch 37/500\n",
 654              "30/30 [==============================] - 4s 124ms/step - loss: 17.1090 - mae: 17.5953\n",
 655              "Epoch 38/500\n",
 656              "30/30 [==============================] - 4s 125ms/step - loss: 16.9883 - mae: 17.4645\n",
 657              "Epoch 39/500\n",
 658              "30/30 [==============================] - 4s 122ms/step - loss: 17.0849 - mae: 17.5704\n",
 659              "Epoch 40/500\n",
 660              "30/30 [==============================] - 4s 123ms/step - loss: 16.8545 - mae: 17.3374\n",
 661              "Epoch 41/500\n",
 662              "30/30 [==============================] - 4s 124ms/step - loss: 16.9221 - mae: 17.4053\n",
 663              "Epoch 42/500\n",
 664              "30/30 [==============================] - 4s 124ms/step - loss: 16.9210 - mae: 17.3926\n",
 665              "Epoch 43/500\n",
 666              "30/30 [==============================] - 4s 125ms/step - loss: 17.1371 - mae: 17.6142\n",
 667              "Epoch 44/500\n",
 668              "30/30 [==============================] - 4s 126ms/step - loss: 16.8160 - mae: 17.2977\n",
 669              "Epoch 45/500\n",
 670              "30/30 [==============================] - 4s 124ms/step - loss: 16.8466 - mae: 17.3287\n",
 671              "Epoch 46/500\n",
 672              "30/30 [==============================] - 4s 128ms/step - loss: 16.8060 - mae: 17.2885\n",
 673              "Epoch 47/500\n",
 674              "30/30 [==============================] - 4s 126ms/step - loss: 16.8747 - mae: 17.3591\n",
 675              "Epoch 48/500\n",
 676              "30/30 [==============================] - 4s 125ms/step - loss: 16.7896 - mae: 17.2722\n",
 677              "Epoch 49/500\n",
 678              "30/30 [==============================] - 4s 124ms/step - loss: 16.7441 - mae: 17.2259\n",
 679              "Epoch 50/500\n",
 680              "30/30 [==============================] - 4s 123ms/step - loss: 16.7234 - mae: 17.2055\n",
 681              "Epoch 51/500\n",
 682              "30/30 [==============================] - 4s 124ms/step - loss: 16.7007 - mae: 17.1832\n",
 683              "Epoch 52/500\n",
 684              "30/30 [==============================] - 4s 121ms/step - loss: 16.7465 - mae: 17.2262\n",
 685              "Epoch 53/500\n",
 686              "30/30 [==============================] - 4s 122ms/step - loss: 16.7266 - mae: 17.2071\n",
 687              "Epoch 54/500\n",
 688              "30/30 [==============================] - 4s 122ms/step - loss: 16.6735 - mae: 17.1553\n",
 689              "Epoch 55/500\n",
 690              "30/30 [==============================] - 4s 123ms/step - loss: 16.6545 - mae: 17.1368\n",
 691              "Epoch 56/500\n",
 692              "30/30 [==============================] - 4s 121ms/step - loss: 16.7774 - mae: 17.2513\n",
 693              "Epoch 57/500\n",
 694              "30/30 [==============================] - 4s 122ms/step - loss: 16.9464 - mae: 17.4319\n",
 695              "Epoch 58/500\n",
 696              "30/30 [==============================] - 4s 121ms/step - loss: 16.6419 - mae: 17.1241\n",
 697              "Epoch 59/500\n",
 698              "30/30 [==============================] - 4s 121ms/step - loss: 16.7839 - mae: 17.2674\n",
 699              "Epoch 60/500\n",
 700              "30/30 [==============================] - 4s 123ms/step - loss: 16.6947 - mae: 17.1792\n",
 701              "Epoch 61/500\n",
 702              "30/30 [==============================] - 4s 121ms/step - loss: 16.5983 - mae: 17.0809\n",
 703              "Epoch 62/500\n",
 704              "30/30 [==============================] - 4s 122ms/step - loss: 16.6056 - mae: 17.0871\n",
 705              "Epoch 63/500\n",
 706              "30/30 [==============================] - 4s 125ms/step - loss: 16.6297 - mae: 17.1106\n",
 707              "Epoch 64/500\n",
 708              "30/30 [==============================] - 4s 122ms/step - loss: 16.6054 - mae: 17.0876\n",
 709              "Epoch 65/500\n",
 710              "30/30 [==============================] - 4s 124ms/step - loss: 16.7204 - mae: 17.1796\n",
 711              "Epoch 66/500\n",
 712              "30/30 [==============================] - 4s 125ms/step - loss: 17.1396 - mae: 17.6145\n",
 713              "Epoch 67/500\n",
 714              "30/30 [==============================] - 4s 124ms/step - loss: 16.8063 - mae: 17.2874\n",
 715              "Epoch 68/500\n",
 716              "30/30 [==============================] - 4s 124ms/step - loss: 16.7011 - mae: 17.1843\n",
 717              "Epoch 69/500\n",
 718              "30/30 [==============================] - 4s 124ms/step - loss: 16.6115 - mae: 17.0915\n",
 719              "Epoch 70/500\n",
 720              "30/30 [==============================] - 4s 122ms/step - loss: 16.7030 - mae: 17.1865\n",
 721              "Epoch 71/500\n",
 722              "30/30 [==============================] - 4s 124ms/step - loss: 16.5724 - mae: 17.0473\n",
 723              "Epoch 72/500\n",
 724              "30/30 [==============================] - 4s 123ms/step - loss: 16.7748 - mae: 17.2468\n",
 725              "Epoch 73/500\n",
 726              "30/30 [==============================] - 4s 122ms/step - loss: 16.8454 - mae: 17.3317\n",
 727              "Epoch 74/500\n",
 728              "30/30 [==============================] - 4s 127ms/step - loss: 16.7583 - mae: 17.2402\n",
 729              "Epoch 75/500\n",
 730              "30/30 [==============================] - 4s 122ms/step - loss: 16.7685 - mae: 17.2508\n",
 731              "Epoch 76/500\n",
 732              "30/30 [==============================] - 4s 124ms/step - loss: 16.7273 - mae: 17.2112\n",
 733              "Epoch 77/500\n",
 734              "30/30 [==============================] - 4s 123ms/step - loss: 16.6941 - mae: 17.1731\n",
 735              "Epoch 78/500\n",
 736              "30/30 [==============================] - 4s 122ms/step - loss: 16.6840 - mae: 17.1684\n",
 737              "Epoch 79/500\n",
 738              "30/30 [==============================] - 4s 121ms/step - loss: 16.5582 - mae: 17.0339\n",
 739              "Epoch 80/500\n",
 740              "30/30 [==============================] - 4s 123ms/step - loss: 16.7058 - mae: 17.1874\n",
 741              "Epoch 81/500\n",
 742              "30/30 [==============================] - 4s 123ms/step - loss: 16.6919 - mae: 17.1723\n",
 743              "Epoch 82/500\n",
 744              "30/30 [==============================] - 4s 122ms/step - loss: 16.5479 - mae: 17.0205\n",
 745              "Epoch 83/500\n",
 746              "30/30 [==============================] - 4s 120ms/step - loss: 16.7521 - mae: 17.2320\n",
 747              "Epoch 84/500\n",
 748              "30/30 [==============================] - 4s 123ms/step - loss: 16.4951 - mae: 16.9707\n",
 749              "Epoch 85/500\n",
 750              "30/30 [==============================] - 4s 122ms/step - loss: 16.6451 - mae: 17.1292\n",
 751              "Epoch 86/500\n",
 752              "30/30 [==============================] - 4s 122ms/step - loss: 16.5167 - mae: 16.9910\n",
 753              "Epoch 87/500\n",
 754              "30/30 [==============================] - 4s 124ms/step - loss: 16.6550 - mae: 17.1361\n",
 755              "Epoch 88/500\n",
 756              "30/30 [==============================] - 4s 125ms/step - loss: 16.4751 - mae: 16.9509\n",
 757              "Epoch 89/500\n",
 758              "30/30 [==============================] - 4s 122ms/step - loss: 16.6485 - mae: 17.1282\n",
 759              "Epoch 90/500\n",
 760              "30/30 [==============================] - 4s 122ms/step - loss: 16.4391 - mae: 16.9191\n",
 761              "Epoch 91/500\n",
 762              "30/30 [==============================] - 4s 121ms/step - loss: 16.5336 - mae: 17.0095\n",
 763              "Epoch 92/500\n",
 764              "30/30 [==============================] - 4s 122ms/step - loss: 16.6515 - mae: 17.1336\n",
 765              "Epoch 93/500\n",
 766              "30/30 [==============================] - 4s 123ms/step - loss: 16.3804 - mae: 16.8587\n",
 767              "Epoch 94/500\n",
 768              "30/30 [==============================] - 4s 122ms/step - loss: 16.5093 - mae: 16.9866\n",
 769              "Epoch 95/500\n",
 770              "30/30 [==============================] - 4s 121ms/step - loss: 16.6147 - mae: 17.0979\n",
 771              "Epoch 96/500\n",
 772              "30/30 [==============================] - 4s 121ms/step - loss: 16.3839 - mae: 16.8618\n",
 773              "Epoch 97/500\n",
 774              "30/30 [==============================] - 4s 122ms/step - loss: 16.4718 - mae: 16.9538\n",
 775              "Epoch 98/500\n",
 776              "30/30 [==============================] - 4s 121ms/step - loss: 16.4700 - mae: 16.9527\n",
 777              "Epoch 99/500\n",
 778              "30/30 [==============================] - 4s 122ms/step - loss: 16.4232 - mae: 16.8979\n",
 779              "Epoch 100/500\n",
 780              "30/30 [==============================] - 4s 121ms/step - loss: 16.5540 - mae: 17.0356\n",
 781              "Epoch 101/500\n",
 782              "30/30 [==============================] - 4s 120ms/step - loss: 16.3595 - mae: 16.8408\n",
 783              "Epoch 102/500\n",
 784              "30/30 [==============================] - 4s 121ms/step - loss: 16.3689 - mae: 16.8438\n",
 785              "Epoch 103/500\n",
 786              "30/30 [==============================] - 4s 121ms/step - loss: 16.5204 - mae: 17.0023\n",
 787              "Epoch 104/500\n",
 788              "30/30 [==============================] - 4s 121ms/step - loss: 16.3450 - mae: 16.8264\n",
 789              "Epoch 105/500\n",
 790              "30/30 [==============================] - 4s 119ms/step - loss: 16.3489 - mae: 16.8237\n",
 791              "Epoch 106/500\n",
 792              "30/30 [==============================] - 4s 122ms/step - loss: 16.4868 - mae: 16.9686\n",
 793              "Epoch 107/500\n",
 794              "30/30 [==============================] - 4s 123ms/step - loss: 16.3381 - mae: 16.8181\n",
 795              "Epoch 108/500\n",
 796              "30/30 [==============================] - 4s 121ms/step - loss: 16.3243 - mae: 16.8056\n",
 797              "Epoch 109/500\n",
 798              "30/30 [==============================] - 4s 124ms/step - loss: 16.3170 - mae: 16.7912\n",
 799              "Epoch 110/500\n",
 800              "30/30 [==============================] - 4s 123ms/step - loss: 16.4376 - mae: 16.9183\n",
 801              "Epoch 111/500\n",
 802              "30/30 [==============================] - 4s 124ms/step - loss: 16.3139 - mae: 16.7933\n",
 803              "Epoch 112/500\n",
 804              "30/30 [==============================] - 4s 121ms/step - loss: 16.2933 - mae: 16.7746\n",
 805              "Epoch 113/500\n",
 806              "30/30 [==============================] - 4s 126ms/step - loss: 16.2759 - mae: 16.7499\n",
 807              "Epoch 114/500\n",
 808              "30/30 [==============================] - 4s 123ms/step - loss: 16.3587 - mae: 16.8399\n",
 809              "Epoch 115/500\n",
 810              "30/30 [==============================] - 4s 122ms/step - loss: 16.3335 - mae: 16.8130\n",
 811              "Epoch 116/500\n",
 812              "30/30 [==============================] - 4s 123ms/step - loss: 16.3356 - mae: 16.8166\n",
 813              "Epoch 117/500\n",
 814              "30/30 [==============================] - 4s 121ms/step - loss: 16.3131 - mae: 16.7935\n",
 815              "Epoch 118/500\n",
 816              "30/30 [==============================] - 4s 120ms/step - loss: 16.3025 - mae: 16.7807\n",
 817              "Epoch 119/500\n",
 818              "30/30 [==============================] - 4s 121ms/step - loss: 16.3293 - mae: 16.8110\n",
 819              "Epoch 120/500\n",
 820              "30/30 [==============================] - 4s 130ms/step - loss: 16.2662 - mae: 16.7443\n",
 821              "Epoch 121/500\n",
 822              "30/30 [==============================] - 4s 122ms/step - loss: 16.2946 - mae: 16.7746\n",
 823              "Epoch 122/500\n",
 824              "30/30 [==============================] - 4s 120ms/step - loss: 16.2834 - mae: 16.7636\n",
 825              "Epoch 123/500\n",
 826              "30/30 [==============================] - 4s 122ms/step - loss: 16.2643 - mae: 16.7419\n",
 827              "Epoch 124/500\n",
 828              "30/30 [==============================] - 4s 122ms/step - loss: 16.2874 - mae: 16.7686\n",
 829              "Epoch 125/500\n",
 830              "30/30 [==============================] - 4s 122ms/step - loss: 16.2382 - mae: 16.7163\n",
 831              "Epoch 126/500\n",
 832              "30/30 [==============================] - 4s 121ms/step - loss: 16.2553 - mae: 16.7338\n",
 833              "Epoch 127/500\n",
 834              "30/30 [==============================] - 4s 121ms/step - loss: 16.2661 - mae: 16.7471\n",
 835              "Epoch 128/500\n",
 836              "30/30 [==============================] - 4s 120ms/step - loss: 16.2142 - mae: 16.6920\n",
 837              "Epoch 129/500\n",
 838              "30/30 [==============================] - 4s 122ms/step - loss: 16.2324 - mae: 16.7110\n",
 839              "Epoch 130/500\n",
 840              "30/30 [==============================] - 4s 123ms/step - loss: 16.2428 - mae: 16.7236\n",
 841              "Epoch 131/500\n",
 842              "30/30 [==============================] - 4s 125ms/step - loss: 16.1994 - mae: 16.6772\n",
 843              "Epoch 132/500\n",
 844              "30/30 [==============================] - 4s 125ms/step - loss: 16.2097 - mae: 16.6881\n",
 845              "Epoch 133/500\n",
 846              "30/30 [==============================] - 4s 123ms/step - loss: 16.2114 - mae: 16.6909\n",
 847              "Epoch 134/500\n",
 848              "30/30 [==============================] - 4s 121ms/step - loss: 16.2081 - mae: 16.6888\n",
 849              "Epoch 135/500\n",
 850              "30/30 [==============================] - 4s 120ms/step - loss: 16.1895 - mae: 16.6692\n",
 851              "Epoch 136/500\n",
 852              "30/30 [==============================] - 4s 121ms/step - loss: 16.1818 - mae: 16.6611\n",
 853              "Epoch 137/500\n",
 854              "30/30 [==============================] - 4s 121ms/step - loss: 16.1790 - mae: 16.6592\n",
 855              "Epoch 138/500\n",
 856              "30/30 [==============================] - 4s 122ms/step - loss: 16.1787 - mae: 16.6596\n",
 857              "Epoch 139/500\n",
 858              "30/30 [==============================] - 4s 122ms/step - loss: 16.1677 - mae: 16.6483\n",
 859              "Epoch 140/500\n",
 860              "30/30 [==============================] - 4s 122ms/step - loss: 16.1585 - mae: 16.6390\n",
 861              "Epoch 141/500\n",
 862              "30/30 [==============================] - 4s 121ms/step - loss: 16.1530 - mae: 16.6336\n",
 863              "Epoch 142/500\n",
 864              "30/30 [==============================] - 4s 121ms/step - loss: 16.1488 - mae: 16.6297\n",
 865              "Epoch 143/500\n",
 866              "30/30 [==============================] - 4s 120ms/step - loss: 16.1394 - mae: 16.6198\n",
 867              "Epoch 144/500\n",
 868              "30/30 [==============================] - 4s 119ms/step - loss: 16.1282 - mae: 16.6080\n",
 869              "Epoch 145/500\n",
 870              "30/30 [==============================] - 4s 121ms/step - loss: 16.1201 - mae: 16.5993\n",
 871              "Epoch 146/500\n",
 872              "30/30 [==============================] - 4s 119ms/step - loss: 16.1094 - mae: 16.5885\n",
 873              "Epoch 147/500\n",
 874              "30/30 [==============================] - 4s 122ms/step - loss: 16.1041 - mae: 16.5837\n",
 875              "Epoch 148/500\n",
 876              "30/30 [==============================] - 4s 125ms/step - loss: 16.1018 - mae: 16.5819\n",
 877              "Epoch 149/500\n",
 878              "30/30 [==============================] - 4s 125ms/step - loss: 16.1007 - mae: 16.5819\n",
 879              "Epoch 150/500\n",
 880              "30/30 [==============================] - 4s 123ms/step - loss: 16.0912 - mae: 16.5717\n",
 881              "Epoch 151/500\n",
 882              "30/30 [==============================] - 4s 124ms/step - loss: 16.0811 - mae: 16.5595\n",
 883              "Epoch 152/500\n",
 884              "30/30 [==============================] - 4s 123ms/step - loss: 16.0813 - mae: 16.5618\n",
 885              "Epoch 153/500\n",
 886              "30/30 [==============================] - 4s 124ms/step - loss: 16.0740 - mae: 16.5521\n",
 887              "Epoch 154/500\n",
 888              "30/30 [==============================] - 4s 124ms/step - loss: 16.0738 - mae: 16.5542\n",
 889              "Epoch 155/500\n",
 890              "30/30 [==============================] - 4s 122ms/step - loss: 16.0673 - mae: 16.5470\n",
 891              "Epoch 156/500\n",
 892              "30/30 [==============================] - 4s 121ms/step - loss: 16.0547 - mae: 16.5342\n",
 893              "Epoch 157/500\n",
 894              "30/30 [==============================] - 4s 120ms/step - loss: 16.0631 - mae: 16.5443\n",
 895              "Epoch 158/500\n",
 896              "30/30 [==============================] - 4s 122ms/step - loss: 16.0476 - mae: 16.5277\n",
 897              "Epoch 159/500\n",
 898              "30/30 [==============================] - 4s 120ms/step - loss: 16.0457 - mae: 16.5262\n",
 899              "Epoch 160/500\n",
 900              "30/30 [==============================] - 4s 122ms/step - loss: 16.0407 - mae: 16.5212\n",
 901              "Epoch 161/500\n",
 902              "30/30 [==============================] - 4s 121ms/step - loss: 16.0406 - mae: 16.5212\n",
 903              "Epoch 162/500\n",
 904              "30/30 [==============================] - 4s 120ms/step - loss: 16.0308 - mae: 16.5112\n",
 905              "Epoch 163/500\n",
 906              "30/30 [==============================] - 4s 122ms/step - loss: 16.0211 - mae: 16.5010\n",
 907              "Epoch 164/500\n",
 908              "30/30 [==============================] - 4s 121ms/step - loss: 16.0258 - mae: 16.5066\n",
 909              "Epoch 165/500\n",
 910              "30/30 [==============================] - 4s 121ms/step - loss: 16.0089 - mae: 16.4893\n",
 911              "Epoch 166/500\n",
 912              "30/30 [==============================] - 4s 122ms/step - loss: 16.0012 - mae: 16.4811\n",
 913              "Epoch 167/500\n",
 914              "30/30 [==============================] - 4s 122ms/step - loss: 16.0038 - mae: 16.4842\n",
 915              "Epoch 168/500\n",
 916              "30/30 [==============================] - 4s 120ms/step - loss: 16.0009 - mae: 16.4819\n",
 917              "Epoch 169/500\n",
 918              "30/30 [==============================] - 4s 120ms/step - loss: 15.9829 - mae: 16.4635\n",
 919              "Epoch 170/500\n",
 920              "30/30 [==============================] - 4s 121ms/step - loss: 15.9702 - mae: 16.4492\n",
 921              "Epoch 171/500\n",
 922              "30/30 [==============================] - 4s 121ms/step - loss: 15.9883 - mae: 16.4692\n",
 923              "Epoch 172/500\n",
 924              "30/30 [==============================] - 4s 121ms/step - loss: 15.9848 - mae: 16.4660\n",
 925              "Epoch 173/500\n",
 926              "30/30 [==============================] - 4s 121ms/step - loss: 15.9826 - mae: 16.4636\n",
 927              "Epoch 174/500\n",
 928              "30/30 [==============================] - 4s 122ms/step - loss: 15.9399 - mae: 16.4206\n",
 929              "Epoch 175/500\n",
 930              "30/30 [==============================] - 4s 122ms/step - loss: 15.9276 - mae: 16.4063\n",
 931              "Epoch 176/500\n",
 932              "30/30 [==============================] - 4s 124ms/step - loss: 15.9437 - mae: 16.4219\n",
 933              "Epoch 177/500\n",
 934              "30/30 [==============================] - 4s 122ms/step - loss: 15.9963 - mae: 16.4761\n",
 935              "Epoch 178/500\n",
 936              "30/30 [==============================] - 4s 120ms/step - loss: 15.8955 - mae: 16.3753\n",
 937              "Epoch 179/500\n",
 938              "30/30 [==============================] - 4s 121ms/step - loss: 15.8952 - mae: 16.3746\n",
 939              "Epoch 180/500\n",
 940              "30/30 [==============================] - 4s 121ms/step - loss: 15.9256 - mae: 16.4038\n",
 941              "Epoch 181/500\n",
 942              "30/30 [==============================] - 4s 121ms/step - loss: 15.9787 - mae: 16.4589\n",
 943              "Epoch 182/500\n",
 944              "30/30 [==============================] - 4s 122ms/step - loss: 15.8715 - mae: 16.3513\n",
 945              "Epoch 183/500\n",
 946              "30/30 [==============================] - 4s 120ms/step - loss: 15.8658 - mae: 16.3448\n",
 947              "Epoch 184/500\n",
 948              "30/30 [==============================] - 4s 121ms/step - loss: 15.9083 - mae: 16.3859\n",
 949              "Epoch 185/500\n",
 950              "30/30 [==============================] - 4s 122ms/step - loss: 15.9975 - mae: 16.4783\n",
 951              "Epoch 186/500\n",
 952              "30/30 [==============================] - 4s 122ms/step - loss: 15.8581 - mae: 16.3380\n",
 953              "Epoch 187/500\n",
 954              "30/30 [==============================] - 4s 121ms/step - loss: 15.8307 - mae: 16.3093\n",
 955              "Epoch 188/500\n",
 956              "30/30 [==============================] - 4s 122ms/step - loss: 15.8913 - mae: 16.3703\n",
 957              "Epoch 189/500\n",
 958              "30/30 [==============================] - 4s 121ms/step - loss: 15.9673 - mae: 16.4482\n",
 959              "Epoch 190/500\n",
 960              "30/30 [==============================] - 4s 121ms/step - loss: 15.8334 - mae: 16.3133\n",
 961              "Epoch 191/500\n",
 962              "30/30 [==============================] - 4s 122ms/step - loss: 15.8173 - mae: 16.2958\n",
 963              "Epoch 192/500\n",
 964              "30/30 [==============================] - 4s 122ms/step - loss: 15.8946 - mae: 16.3758\n",
 965              "Epoch 193/500\n",
 966              "30/30 [==============================] - 4s 124ms/step - loss: 15.9014 - mae: 16.3825\n",
 967              "Epoch 194/500\n",
 968              "30/30 [==============================] - 4s 120ms/step - loss: 15.8233 - mae: 16.3032\n",
 969              "Epoch 195/500\n",
 970              "30/30 [==============================] - 4s 123ms/step - loss: 15.8660 - mae: 16.3468\n",
 971              "Epoch 196/500\n",
 972              "30/30 [==============================] - 4s 123ms/step - loss: 15.8840 - mae: 16.3654\n",
 973              "Epoch 197/500\n",
 974              "30/30 [==============================] - 4s 122ms/step - loss: 15.8200 - mae: 16.3001\n",
 975              "Epoch 198/500\n",
 976              "30/30 [==============================] - 4s 123ms/step - loss: 15.8679 - mae: 16.3492\n",
 977              "Epoch 199/500\n",
 978              "30/30 [==============================] - 4s 120ms/step - loss: 15.8248 - mae: 16.3053\n",
 979              "Epoch 200/500\n",
 980              "30/30 [==============================] - 4s 122ms/step - loss: 15.8398 - mae: 16.3209\n",
 981              "Epoch 201/500\n",
 982              "30/30 [==============================] - 4s 121ms/step - loss: 15.8340 - mae: 16.3149\n",
 983              "Epoch 202/500\n",
 984              "30/30 [==============================] - 4s 121ms/step - loss: 15.8175 - mae: 16.2981\n",
 985              "Epoch 203/500\n",
 986              "30/30 [==============================] - 4s 121ms/step - loss: 15.8267 - mae: 16.3077\n",
 987              "Epoch 204/500\n",
 988              "30/30 [==============================] - 4s 120ms/step - loss: 15.8203 - mae: 16.3011\n",
 989              "Epoch 205/500\n",
 990              "30/30 [==============================] - 4s 121ms/step - loss: 15.8102 - mae: 16.2910\n",
 991              "Epoch 206/500\n",
 992              "30/30 [==============================] - 4s 121ms/step - loss: 15.8049 - mae: 16.2856\n",
 993              "Epoch 207/500\n",
 994              "30/30 [==============================] - 4s 123ms/step - loss: 15.8057 - mae: 16.2866\n",
 995              "Epoch 208/500\n",
 996              "30/30 [==============================] - 4s 122ms/step - loss: 15.7899 - mae: 16.2703\n",
 997              "Epoch 209/500\n",
 998              "30/30 [==============================] - 4s 123ms/step - loss: 15.7938 - mae: 16.2747\n",
 999              "Epoch 210/500\n",
1000              "30/30 [==============================] - 4s 125ms/step - loss: 15.7787 - mae: 16.2590\n",
1001              "Epoch 211/500\n",
1002              "30/30 [==============================] - 4s 123ms/step - loss: 15.7846 - mae: 16.2656\n",
1003              "Epoch 212/500\n",
1004              "30/30 [==============================] - 4s 125ms/step - loss: 15.7629 - mae: 16.2430\n",
1005              "Epoch 213/500\n",
1006              "30/30 [==============================] - 4s 123ms/step - loss: 15.7739 - mae: 16.2549\n",
1007              "Epoch 214/500\n",
1008              "30/30 [==============================] - 4s 123ms/step - loss: 15.7475 - mae: 16.2275\n",
1009              "Epoch 215/500\n",
1010              "30/30 [==============================] - 4s 127ms/step - loss: 15.7614 - mae: 16.2423\n",
1011              "Epoch 216/500\n",
1012              "30/30 [==============================] - 4s 125ms/step - loss: 15.7434 - mae: 16.2237\n",
1013              "Epoch 217/500\n",
1014              "30/30 [==============================] - 4s 127ms/step - loss: 15.7475 - mae: 16.2282\n",
1015              "Epoch 218/500\n",
1016              "30/30 [==============================] - 4s 124ms/step - loss: 15.7366 - mae: 16.2171\n",
1017              "Epoch 219/500\n",
1018              "30/30 [==============================] - 4s 124ms/step - loss: 15.7324 - mae: 16.2129\n",
1019              "Epoch 220/500\n",
1020              "30/30 [==============================] - 4s 122ms/step - loss: 15.7298 - mae: 16.2104\n",
1021              "Epoch 221/500\n",
1022              "30/30 [==============================] - 4s 122ms/step - loss: 15.7192 - mae: 16.1995\n",
1023              "Epoch 222/500\n",
1024              "30/30 [==============================] - 4s 123ms/step - loss: 15.7181 - mae: 16.1986\n",
1025              "Epoch 223/500\n",
1026              "30/30 [==============================] - 4s 120ms/step - loss: 15.7064 - mae: 16.1867\n",
1027              "Epoch 224/500\n",
1028              "30/30 [==============================] - 4s 122ms/step - loss: 15.7106 - mae: 16.1913\n",
1029              "Epoch 225/500\n",
1030              "30/30 [==============================] - 4s 119ms/step - loss: 15.6878 - mae: 16.1673\n",
1031              "Epoch 226/500\n",
1032              "30/30 [==============================] - 4s 122ms/step - loss: 15.7170 - mae: 16.1980\n",
1033              "Epoch 227/500\n",
1034              "30/30 [==============================] - 4s 121ms/step - loss: 15.6728 - mae: 16.1526\n",
1035              "Epoch 228/500\n",
1036              "30/30 [==============================] - 4s 121ms/step - loss: 15.6800 - mae: 16.1591\n",
1037              "Epoch 229/500\n",
1038              "30/30 [==============================] - 4s 123ms/step - loss: 15.7136 - mae: 16.1947\n",
1039              "Epoch 230/500\n",
1040              "30/30 [==============================] - 4s 123ms/step - loss: 15.6479 - mae: 16.1268\n",
1041              "Epoch 231/500\n",
1042              "30/30 [==============================] - 4s 123ms/step - loss: 15.6909 - mae: 16.1716\n",
1043              "Epoch 232/500\n",
1044              "30/30 [==============================] - 4s 122ms/step - loss: 15.6519 - mae: 16.1307\n",
1045              "Epoch 233/500\n",
1046              "30/30 [==============================] - 4s 122ms/step - loss: 15.6942 - mae: 16.1751\n",
1047              "Epoch 234/500\n",
1048              "30/30 [==============================] - 4s 124ms/step - loss: 15.6283 - mae: 16.1075\n",
1049              "Epoch 235/500\n",
1050              "30/30 [==============================] - 4s 123ms/step - loss: 15.6675 - mae: 16.1479\n",
1051              "Epoch 236/500\n",
1052              "30/30 [==============================] - 4s 121ms/step - loss: 15.6505 - mae: 16.1308\n",
1053              "Epoch 237/500\n",
1054              "30/30 [==============================] - 4s 123ms/step - loss: 15.6232 - mae: 16.1016\n",
1055              "Epoch 238/500\n",
1056              "30/30 [==============================] - 4s 123ms/step - loss: 15.6880 - mae: 16.1693\n",
1057              "Epoch 239/500\n",
1058              "30/30 [==============================] - 4s 122ms/step - loss: 15.6040 - mae: 16.0836\n",
1059              "Epoch 240/500\n",
1060              "30/30 [==============================] - 4s 123ms/step - loss: 15.6020 - mae: 16.0791\n",
1061              "Epoch 241/500\n",
1062              "30/30 [==============================] - 4s 123ms/step - loss: 15.7205 - mae: 16.2026\n",
1063              "Epoch 242/500\n",
1064              "30/30 [==============================] - 4s 123ms/step - loss: 15.5684 - mae: 16.0477\n",
1065              "Epoch 243/500\n",
1066              "30/30 [==============================] - 4s 120ms/step - loss: 15.6027 - mae: 16.0807\n",
1067              "Epoch 244/500\n",
1068              "30/30 [==============================] - 4s 118ms/step - loss: 15.6791 - mae: 16.1607\n",
1069              "Epoch 245/500\n",
1070              "30/30 [==============================] - 4s 119ms/step - loss: 15.5565 - mae: 16.0351\n",
1071              "Epoch 246/500\n",
1072              "30/30 [==============================] - 4s 120ms/step - loss: 15.6079 - mae: 16.0873\n",
1073              "Epoch 247/500\n",
1074              "30/30 [==============================] - 4s 119ms/step - loss: 15.6296 - mae: 16.1104\n",
1075              "Epoch 248/500\n",
1076              "30/30 [==============================] - 4s 119ms/step - loss: 15.5492 - mae: 16.0255\n",
1077              "Epoch 249/500\n",
1078              "30/30 [==============================] - 4s 119ms/step - loss: 15.7041 - mae: 16.1854\n",
1079              "Epoch 250/500\n",
1080              "30/30 [==============================] - 4s 120ms/step - loss: 15.5773 - mae: 16.0574\n",
1081              "Epoch 251/500\n",
1082              "30/30 [==============================] - 4s 119ms/step - loss: 15.5485 - mae: 16.0264\n",
1083              "Epoch 252/500\n",
1084              "30/30 [==============================] - 4s 119ms/step - loss: 15.6576 - mae: 16.1391\n",
1085              "Epoch 253/500\n",
1086              "30/30 [==============================] - 4s 119ms/step - loss: 15.5203 - mae: 15.9956\n",
1087              "Epoch 254/500\n",
1088              "30/30 [==============================] - 4s 120ms/step - loss: 15.7034 - mae: 16.1817\n",
1089              "Epoch 255/500\n",
1090              "30/30 [==============================] - 4s 119ms/step - loss: 15.7020 - mae: 16.1762\n",
1091              "Epoch 256/500\n",
1092              "30/30 [==============================] - 4s 119ms/step - loss: 15.8131 - mae: 16.2952\n",
1093              "Epoch 257/500\n",
1094              "30/30 [==============================] - 4s 119ms/step - loss: 15.6627 - mae: 16.1399\n",
1095              "Epoch 258/500\n",
1096              "30/30 [==============================] - 4s 119ms/step - loss: 15.6867 - mae: 16.1643\n",
1097              "Epoch 259/500\n",
1098              "30/30 [==============================] - 4s 123ms/step - loss: 15.6816 - mae: 16.1597\n",
1099              "Epoch 260/500\n",
1100              "30/30 [==============================] - 4s 120ms/step - loss: 15.6614 - mae: 16.1395\n",
1101              "Epoch 261/500\n",
1102              "30/30 [==============================] - 4s 118ms/step - loss: 15.6439 - mae: 16.1222\n",
1103              "Epoch 262/500\n",
1104              "30/30 [==============================] - 4s 121ms/step - loss: 15.6316 - mae: 16.1103\n",
1105              "Epoch 263/500\n",
1106              "30/30 [==============================] - 4s 121ms/step - loss: 15.6152 - mae: 16.0937\n",
1107              "Epoch 264/500\n",
1108              "30/30 [==============================] - 4s 120ms/step - loss: 15.6076 - mae: 16.0862\n",
1109              "Epoch 265/500\n",
1110              "30/30 [==============================] - 4s 119ms/step - loss: 15.5959 - mae: 16.0745\n",
1111              "Epoch 266/500\n",
1112              "30/30 [==============================] - 4s 119ms/step - loss: 15.5820 - mae: 16.0606\n",
1113              "Epoch 267/500\n",
1114              "30/30 [==============================] - 4s 120ms/step - loss: 15.5753 - mae: 16.0537\n",
1115              "Epoch 268/500\n",
1116              "30/30 [==============================] - 4s 119ms/step - loss: 15.5692 - mae: 16.0477\n",
1117              "Epoch 269/500\n",
1118              "30/30 [==============================] - 4s 120ms/step - loss: 15.5557 - mae: 16.0341\n",
1119              "Epoch 270/500\n",
1120              "30/30 [==============================] - 4s 120ms/step - loss: 15.5537 - mae: 16.0321\n",
1121              "Epoch 271/500\n",
1122              "30/30 [==============================] - 4s 118ms/step - loss: 15.5481 - mae: 16.0263\n",
1123              "Epoch 272/500\n",
1124              "30/30 [==============================] - 4s 119ms/step - loss: 15.5427 - mae: 16.0210\n",
1125              "Epoch 273/500\n",
1126              "30/30 [==============================] - 4s 119ms/step - loss: 15.5361 - mae: 16.0143\n",
1127              "Epoch 274/500\n",
1128              "30/30 [==============================] - 4s 118ms/step - loss: 15.5344 - mae: 16.0125\n",
1129              "Epoch 275/500\n",
1130              "30/30 [==============================] - 4s 120ms/step - loss: 15.5312 - mae: 16.0094\n",
1131              "Epoch 276/500\n",
1132              "30/30 [==============================] - 4s 121ms/step - loss: 15.5252 - mae: 16.0034\n",
1133              "Epoch 277/500\n",
1134              "30/30 [==============================] - 4s 122ms/step - loss: 15.5214 - mae: 15.9995\n",
1135              "Epoch 278/500\n",
1136              "30/30 [==============================] - 4s 121ms/step - loss: 15.5235 - mae: 16.0013\n",
1137              "Epoch 279/500\n",
1138              "30/30 [==============================] - 4s 124ms/step - loss: 15.5371 - mae: 16.0152\n",
1139              "Epoch 280/500\n",
1140              "30/30 [==============================] - 4s 122ms/step - loss: 15.5311 - mae: 16.0098\n",
1141              "Epoch 281/500\n",
1142              "30/30 [==============================] - 4s 122ms/step - loss: 15.4956 - mae: 15.9742\n",
1143              "Epoch 282/500\n",
1144              "30/30 [==============================] - 4s 121ms/step - loss: 15.4718 - mae: 15.9500\n",
1145              "Epoch 283/500\n",
1146              "30/30 [==============================] - 4s 120ms/step - loss: 15.4657 - mae: 15.9436\n",
1147              "Epoch 284/500\n",
1148              "30/30 [==============================] - 4s 123ms/step - loss: 15.4793 - mae: 15.9568\n",
1149              "Epoch 285/500\n",
1150              "30/30 [==============================] - 4s 122ms/step - loss: 15.4928 - mae: 15.9707\n",
1151              "Epoch 286/500\n",
1152              "30/30 [==============================] - 4s 123ms/step - loss: 15.4799 - mae: 15.9578\n",
1153              "Epoch 287/500\n",
1154              "30/30 [==============================] - 4s 120ms/step - loss: 15.4595 - mae: 15.9372\n",
1155              "Epoch 288/500\n",
1156              "30/30 [==============================] - 4s 119ms/step - loss: 15.4465 - mae: 15.9239\n",
1157              "Epoch 289/500\n",
1158              "30/30 [==============================] - 4s 119ms/step - loss: 15.4430 - mae: 15.9203\n",
1159              "Epoch 290/500\n",
1160              "30/30 [==============================] - 4s 119ms/step - loss: 15.4625 - mae: 15.9400\n",
1161              "Epoch 291/500\n",
1162              "30/30 [==============================] - 4s 120ms/step - loss: 15.4482 - mae: 15.9257\n",
1163              "Epoch 292/500\n",
1164              "30/30 [==============================] - 4s 120ms/step - loss: 15.4369 - mae: 15.9142\n",
1165              "Epoch 293/500\n",
1166              "30/30 [==============================] - 4s 120ms/step - loss: 15.4310 - mae: 15.9082\n",
1167              "Epoch 294/500\n",
1168              "30/30 [==============================] - 4s 119ms/step - loss: 15.4284 - mae: 15.9056\n",
1169              "Epoch 295/500\n",
1170              "30/30 [==============================] - 4s 120ms/step - loss: 15.4357 - mae: 15.9132\n",
1171              "Epoch 296/500\n",
1172              "30/30 [==============================] - 4s 119ms/step - loss: 15.4250 - mae: 15.9024\n",
1173              "Epoch 297/500\n",
1174              "30/30 [==============================] - 4s 119ms/step - loss: 15.4190 - mae: 15.8964\n",
1175              "Epoch 298/500\n",
1176              "30/30 [==============================] - 4s 119ms/step - loss: 15.4139 - mae: 15.8913\n",
1177              "Epoch 299/500\n",
1178              "30/30 [==============================] - 4s 118ms/step - loss: 15.4082 - mae: 15.8856\n",
1179              "Epoch 300/500\n",
1180              "30/30 [==============================] - 4s 122ms/step - loss: 15.4030 - mae: 15.8803\n",
1181              "Epoch 301/500\n",
1182              "30/30 [==============================] - 4s 122ms/step - loss: 15.3958 - mae: 15.8730\n",
1183              "Epoch 302/500\n",
1184              "30/30 [==============================] - 4s 123ms/step - loss: 15.3982 - mae: 15.8755\n",
1185              "Epoch 303/500\n",
1186              "30/30 [==============================] - 4s 123ms/step - loss: 15.3922 - mae: 15.8696\n",
1187              "Epoch 304/500\n",
1188              "30/30 [==============================] - 4s 121ms/step - loss: 15.3838 - mae: 15.8611\n",
1189              "Epoch 305/500\n",
1190              "30/30 [==============================] - 4s 120ms/step - loss: 15.3839 - mae: 15.8611\n",
1191              "Epoch 306/500\n",
1192              "30/30 [==============================] - 4s 120ms/step - loss: 15.3831 - mae: 15.8605\n",
1193              "Epoch 307/500\n",
1194              "30/30 [==============================] - 4s 120ms/step - loss: 15.3762 - mae: 15.8535\n",
1195              "Epoch 308/500\n",
1196              "30/30 [==============================] - 4s 122ms/step - loss: 15.3701 - mae: 15.8472\n",
1197              "Epoch 309/500\n",
1198              "30/30 [==============================] - 4s 120ms/step - loss: 15.3772 - mae: 15.8545\n",
1199              "Epoch 310/500\n",
1200              "30/30 [==============================] - 4s 121ms/step - loss: 15.3637 - mae: 15.8411\n",
1201              "Epoch 311/500\n",
1202              "30/30 [==============================] - 4s 119ms/step - loss: 15.3589 - mae: 15.8360\n",
1203              "Epoch 312/500\n",
1204              "30/30 [==============================] - 4s 121ms/step - loss: 15.3624 - mae: 15.8399\n",
1205              "Epoch 313/500\n",
1206              "30/30 [==============================] - 4s 121ms/step - loss: 15.3499 - mae: 15.8272\n",
1207              "Epoch 314/500\n",
1208              "30/30 [==============================] - 4s 121ms/step - loss: 15.3468 - mae: 15.8242\n",
1209              "Epoch 315/500\n",
1210              "30/30 [==============================] - 4s 118ms/step - loss: 15.3412 - mae: 15.8184\n",
1211              "Epoch 316/500\n",
1212              "30/30 [==============================] - 4s 119ms/step - loss: 15.3393 - mae: 15.8166\n",
1213              "Epoch 317/500\n",
1214              "30/30 [==============================] - 4s 119ms/step - loss: 15.3366 - mae: 15.8141\n",
1215              "Epoch 318/500\n",
1216              "30/30 [==============================] - 4s 119ms/step - loss: 15.3301 - mae: 15.8077\n",
1217              "Epoch 319/500\n",
1218              "30/30 [==============================] - 4s 120ms/step - loss: 15.3201 - mae: 15.7975\n",
1219              "Epoch 320/500\n",
1220              "30/30 [==============================] - 4s 120ms/step - loss: 15.3077 - mae: 15.7847\n",
1221              "Epoch 321/500\n",
1222              "30/30 [==============================] - 4s 118ms/step - loss: 15.2985 - mae: 15.7755\n",
1223              "Epoch 322/500\n",
1224              "30/30 [==============================] - 4s 119ms/step - loss: 15.2941 - mae: 15.7707\n",
1225              "Epoch 323/500\n",
1226              "30/30 [==============================] - 4s 119ms/step - loss: 15.3009 - mae: 15.7774\n",
1227              "Epoch 324/500\n",
1228              "30/30 [==============================] - 4s 119ms/step - loss: 15.3103 - mae: 15.7870\n",
1229              "Epoch 325/500\n",
1230              "30/30 [==============================] - 4s 121ms/step - loss: 15.3121 - mae: 15.7896\n",
1231              "Epoch 326/500\n",
1232              "30/30 [==============================] - 4s 121ms/step - loss: 15.3036 - mae: 15.7814\n",
1233              "Epoch 327/500\n",
1234              "30/30 [==============================] - 4s 121ms/step - loss: 15.2835 - mae: 15.7608\n",
1235              "Epoch 328/500\n",
1236              "30/30 [==============================] - 4s 121ms/step - loss: 15.2697 - mae: 15.7463\n",
1237              "Epoch 329/500\n",
1238              "30/30 [==============================] - 4s 123ms/step - loss: 15.2716 - mae: 15.7484\n",
1239              "Epoch 330/500\n",
1240              "30/30 [==============================] - 4s 124ms/step - loss: 15.2779 - mae: 15.7546\n",
1241              "Epoch 331/500\n",
1242              "30/30 [==============================] - 4s 122ms/step - loss: 15.2865 - mae: 15.7638\n",
1243              "Epoch 332/500\n",
1244              "30/30 [==============================] - 4s 120ms/step - loss: 15.2755 - mae: 15.7530\n",
1245              "Epoch 333/500\n",
1246              "30/30 [==============================] - 4s 121ms/step - loss: 15.2614 - mae: 15.7385\n",
1247              "Epoch 334/500\n",
1248              "30/30 [==============================] - 4s 119ms/step - loss: 15.2483 - mae: 15.7254\n",
1249              "Epoch 335/500\n",
1250              "30/30 [==============================] - 4s 121ms/step - loss: 15.2461 - mae: 15.7229\n",
1251              "Epoch 336/500\n",
1252              "30/30 [==============================] - 4s 121ms/step - loss: 15.2464 - mae: 15.7233\n",
1253              "Epoch 337/500\n",
1254              "30/30 [==============================] - 4s 121ms/step - loss: 15.2452 - mae: 15.7221\n",
1255              "Epoch 338/500\n",
1256              "30/30 [==============================] - 4s 120ms/step - loss: 15.2482 - mae: 15.7248\n",
1257              "Epoch 339/500\n",
1258              "30/30 [==============================] - 4s 121ms/step - loss: 15.2445 - mae: 15.7213\n",
1259              "Epoch 340/500\n",
1260              "30/30 [==============================] - 4s 121ms/step - loss: 15.2363 - mae: 15.7131\n",
1261              "Epoch 341/500\n",
1262              "30/30 [==============================] - 4s 121ms/step - loss: 15.2319 - mae: 15.7086\n",
1263              "Epoch 342/500\n",
1264              "30/30 [==============================] - 4s 123ms/step - loss: 15.2252 - mae: 15.7017\n",
1265              "Epoch 343/500\n",
1266              "30/30 [==============================] - 4s 122ms/step - loss: 15.2247 - mae: 15.7012\n",
1267              "Epoch 344/500\n",
1268              "30/30 [==============================] - 4s 125ms/step - loss: 15.2252 - mae: 15.7017\n",
1269              "Epoch 345/500\n",
1270              "30/30 [==============================] - 4s 122ms/step - loss: 15.2234 - mae: 15.7001\n",
1271              "Epoch 346/500\n",
1272              "30/30 [==============================] - 4s 124ms/step - loss: 15.2143 - mae: 15.6911\n",
1273              "Epoch 347/500\n",
1274              "30/30 [==============================] - 4s 122ms/step - loss: 15.2030 - mae: 15.6795\n",
1275              "Epoch 348/500\n",
1276              "30/30 [==============================] - 4s 121ms/step - loss: 15.1982 - mae: 15.6748\n",
1277              "Epoch 349/500\n",
1278              "30/30 [==============================] - 4s 120ms/step - loss: 15.2025 - mae: 15.6789\n",
1279              "Epoch 350/500\n",
1280              "30/30 [==============================] - 4s 121ms/step - loss: 15.1975 - mae: 15.6740\n",
1281              "Epoch 351/500\n",
1282              "30/30 [==============================] - 4s 122ms/step - loss: 15.1974 - mae: 15.6740\n",
1283              "Epoch 352/500\n",
1284              "30/30 [==============================] - 4s 124ms/step - loss: 15.1966 - mae: 15.6735\n",
1285              "Epoch 353/500\n",
1286              "30/30 [==============================] - 4s 124ms/step - loss: 15.1878 - mae: 15.6647\n",
1287              "Epoch 354/500\n",
1288              "30/30 [==============================] - 4s 122ms/step - loss: 15.1817 - mae: 15.6585\n",
1289              "Epoch 355/500\n",
1290              "30/30 [==============================] - 4s 122ms/step - loss: 15.1703 - mae: 15.6469\n",
1291              "Epoch 356/500\n",
1292              "30/30 [==============================] - 4s 122ms/step - loss: 15.1626 - mae: 15.6392\n",
1293              "Epoch 357/500\n",
1294              "30/30 [==============================] - 4s 122ms/step - loss: 15.1578 - mae: 15.6341\n",
1295              "Epoch 358/500\n",
1296              "30/30 [==============================] - 4s 122ms/step - loss: 15.1627 - mae: 15.6390\n",
1297              "Epoch 359/500\n",
1298              "30/30 [==============================] - 4s 120ms/step - loss: 15.1668 - mae: 15.6433\n",
1299              "Epoch 360/500\n",
1300              "30/30 [==============================] - 4s 121ms/step - loss: 15.1653 - mae: 15.6420\n",
1301              "Epoch 361/500\n",
1302              "30/30 [==============================] - 4s 123ms/step - loss: 15.1613 - mae: 15.6382\n",
1303              "Epoch 362/500\n",
1304              "30/30 [==============================] - 4s 123ms/step - loss: 15.1494 - mae: 15.6261\n",
1305              "Epoch 363/500\n",
1306              "30/30 [==============================] - 4s 121ms/step - loss: 15.1396 - mae: 15.6160\n",
1307              "Epoch 364/500\n",
1308              "30/30 [==============================] - 4s 121ms/step - loss: 15.1343 - mae: 15.6107\n",
1309              "Epoch 365/500\n",
1310              "30/30 [==============================] - 4s 121ms/step - loss: 15.1362 - mae: 15.6126\n",
1311              "Epoch 366/500\n",
1312              "30/30 [==============================] - 4s 122ms/step - loss: 15.1383 - mae: 15.6148\n",
1313              "Epoch 367/500\n",
1314              "30/30 [==============================] - 4s 124ms/step - loss: 15.1396 - mae: 15.6163\n",
1315              "Epoch 368/500\n",
1316              "30/30 [==============================] - 4s 127ms/step - loss: 15.1368 - mae: 15.6137\n",
1317              "Epoch 369/500\n",
1318              "30/30 [==============================] - 4s 123ms/step - loss: 15.1276 - mae: 15.6046\n",
1319              "Epoch 370/500\n",
1320              "30/30 [==============================] - 4s 121ms/step - loss: 15.1161 - mae: 15.5929\n",
1321              "Epoch 371/500\n",
1322              "30/30 [==============================] - 4s 121ms/step - loss: 15.1092 - mae: 15.5859\n",
1323              "Epoch 372/500\n",
1324              "30/30 [==============================] - 4s 120ms/step - loss: 15.1090 - mae: 15.5855\n",
1325              "Epoch 373/500\n",
1326              "30/30 [==============================] - 4s 122ms/step - loss: 15.1085 - mae: 15.5850\n",
1327              "Epoch 374/500\n",
1328              "30/30 [==============================] - 4s 120ms/step - loss: 15.1045 - mae: 15.5811\n",
1329              "Epoch 375/500\n",
1330              "30/30 [==============================] - 4s 121ms/step - loss: 15.0973 - mae: 15.5737\n",
1331              "Epoch 376/500\n",
1332              "30/30 [==============================] - 4s 119ms/step - loss: 15.1001 - mae: 15.5765\n",
1333              "Epoch 377/500\n",
1334              "30/30 [==============================] - 4s 122ms/step - loss: 15.0998 - mae: 15.5763\n",
1335              "Epoch 378/500\n",
1336              "30/30 [==============================] - 4s 120ms/step - loss: 15.0941 - mae: 15.5706\n",
1337              "Epoch 379/500\n",
1338              "30/30 [==============================] - 4s 118ms/step - loss: 15.0959 - mae: 15.5724\n",
1339              "Epoch 380/500\n",
1340              "30/30 [==============================] - 4s 118ms/step - loss: 15.0936 - mae: 15.5702\n",
1341              "Epoch 381/500\n",
1342              "30/30 [==============================] - 4s 118ms/step - loss: 15.0892 - mae: 15.5656\n",
1343              "Epoch 382/500\n",
1344              "30/30 [==============================] - 4s 117ms/step - loss: 15.0926 - mae: 15.5693\n",
1345              "Epoch 383/500\n",
1346              "30/30 [==============================] - 4s 119ms/step - loss: 15.0868 - mae: 15.5638\n",
1347              "Epoch 384/500\n",
1348              "30/30 [==============================] - 4s 117ms/step - loss: 15.0852 - mae: 15.5625\n",
1349              "Epoch 385/500\n",
1350              "30/30 [==============================] - 4s 120ms/step - loss: 15.0768 - mae: 15.5543\n",
1351              "Epoch 386/500\n",
1352              "30/30 [==============================] - 4s 125ms/step - loss: 15.0711 - mae: 15.5486\n",
1353              "Epoch 387/500\n",
1354              "30/30 [==============================] - 4s 122ms/step - loss: 15.0649 - mae: 15.5423\n",
1355              "Epoch 388/500\n",
1356              "30/30 [==============================] - 4s 121ms/step - loss: 15.0624 - mae: 15.5400\n",
1357              "Epoch 389/500\n",
1358              "30/30 [==============================] - 4s 119ms/step - loss: 15.0582 - mae: 15.5361\n",
1359              "Epoch 390/500\n",
1360              "30/30 [==============================] - 4s 120ms/step - loss: 15.0452 - mae: 15.5228\n",
1361              "Epoch 391/500\n",
1362              "30/30 [==============================] - 4s 119ms/step - loss: 15.0437 - mae: 15.5213\n",
1363              "Epoch 392/500\n",
1364              "30/30 [==============================] - 4s 122ms/step - loss: 15.0370 - mae: 15.5142\n",
1365              "Epoch 393/500\n",
1366              "30/30 [==============================] - 4s 119ms/step - loss: 15.0412 - mae: 15.5186\n",
1367              "Epoch 394/500\n",
1368              "30/30 [==============================] - 4s 121ms/step - loss: 15.0371 - mae: 15.5145\n",
1369              "Epoch 395/500\n",
1370              "30/30 [==============================] - 4s 121ms/step - loss: 15.0364 - mae: 15.5140\n",
1371              "Epoch 396/500\n",
1372              "30/30 [==============================] - 4s 121ms/step - loss: 15.0276 - mae: 15.5054\n",
1373              "Epoch 397/500\n",
1374              "30/30 [==============================] - 4s 121ms/step - loss: 15.0160 - mae: 15.4934\n",
1375              "Epoch 398/500\n",
1376              "30/30 [==============================] - 4s 119ms/step - loss: 15.0171 - mae: 15.4944\n",
1377              "Epoch 399/500\n",
1378              "30/30 [==============================] - 4s 122ms/step - loss: 15.0168 - mae: 15.4942\n",
1379              "Epoch 400/500\n",
1380              "30/30 [==============================] - 4s 120ms/step - loss: 15.0158 - mae: 15.4936\n",
1381              "Epoch 401/500\n",
1382              "30/30 [==============================] - 4s 119ms/step - loss: 15.0116 - mae: 15.4896\n",
1383              "Epoch 402/500\n",
1384              "30/30 [==============================] - 4s 121ms/step - loss: 15.0030 - mae: 15.4810\n",
1385              "Epoch 403/500\n",
1386              "30/30 [==============================] - 4s 123ms/step - loss: 14.9951 - mae: 15.4730\n",
1387              "Epoch 404/500\n",
1388              "30/30 [==============================] - 4s 121ms/step - loss: 14.9884 - mae: 15.4657\n",
1389              "Epoch 405/500\n",
1390              "30/30 [==============================] - 4s 122ms/step - loss: 14.9853 - mae: 15.4627\n",
1391              "Epoch 406/500\n",
1392              "30/30 [==============================] - 4s 123ms/step - loss: 14.9830 - mae: 15.4602\n",
1393              "Epoch 407/500\n",
1394              "30/30 [==============================] - 4s 121ms/step - loss: 14.9803 - mae: 15.4575\n",
1395              "Epoch 408/500\n",
1396              "30/30 [==============================] - 4s 122ms/step - loss: 14.9747 - mae: 15.4518\n",
1397              "Epoch 409/500\n",
1398              "30/30 [==============================] - 4s 122ms/step - loss: 14.9684 - mae: 15.4452\n",
1399              "Epoch 410/500\n",
1400              "30/30 [==============================] - 4s 122ms/step - loss: 14.9668 - mae: 15.4437\n",
1401              "Epoch 411/500\n",
1402              "30/30 [==============================] - 4s 122ms/step - loss: 14.9628 - mae: 15.4395\n",
1403              "Epoch 412/500\n",
1404              "30/30 [==============================] - 4s 120ms/step - loss: 14.9621 - mae: 15.4389\n",
1405              "Epoch 413/500\n",
1406              "30/30 [==============================] - 4s 121ms/step - loss: 14.9575 - mae: 15.4342\n",
1407              "Epoch 414/500\n",
1408              "30/30 [==============================] - 4s 121ms/step - loss: 14.9574 - mae: 15.4342\n",
1409              "Epoch 415/500\n",
1410              "30/30 [==============================] - 4s 122ms/step - loss: 14.9597 - mae: 15.4367\n",
1411              "Epoch 416/500\n",
1412              "30/30 [==============================] - 4s 121ms/step - loss: 14.9609 - mae: 15.4381\n",
1413              "Epoch 417/500\n",
1414              "30/30 [==============================] - 4s 123ms/step - loss: 14.9636 - mae: 15.4413\n",
1415              "Epoch 418/500\n",
1416              "30/30 [==============================] - 4s 122ms/step - loss: 14.9590 - mae: 15.4370\n",
1417              "Epoch 419/500\n",
1418              "30/30 [==============================] - 4s 124ms/step - loss: 14.9551 - mae: 15.4334\n",
1419              "Epoch 420/500\n",
1420              "30/30 [==============================] - 4s 121ms/step - loss: 14.9523 - mae: 15.4312\n",
1421              "Epoch 421/500\n",
1422              "30/30 [==============================] - 4s 121ms/step - loss: 14.9413 - mae: 15.4199\n",
1423              "Epoch 422/500\n",
1424              "30/30 [==============================] - 4s 121ms/step - loss: 14.9393 - mae: 15.4179\n",
1425              "Epoch 423/500\n",
1426              "30/30 [==============================] - 4s 121ms/step - loss: 14.9337 - mae: 15.4123\n",
1427              "Epoch 424/500\n",
1428              "30/30 [==============================] - 4s 120ms/step - loss: 14.9291 - mae: 15.4077\n",
1429              "Epoch 425/500\n",
1430              "30/30 [==============================] - 4s 120ms/step - loss: 14.9229 - mae: 15.4014\n",
1431              "Epoch 426/500\n",
1432              "30/30 [==============================] - 4s 122ms/step - loss: 14.9172 - mae: 15.3953\n",
1433              "Epoch 427/500\n",
1434              "30/30 [==============================] - 4s 125ms/step - loss: 14.9090 - mae: 15.3869\n",
1435              "Epoch 428/500\n",
1436              "30/30 [==============================] - 4s 124ms/step - loss: 14.9024 - mae: 15.3799\n",
1437              "Epoch 429/500\n",
1438              "30/30 [==============================] - 4s 120ms/step - loss: 14.9000 - mae: 15.3775\n",
1439              "Epoch 430/500\n",
1440              "30/30 [==============================] - 4s 119ms/step - loss: 14.8900 - mae: 15.3673\n",
1441              "Epoch 431/500\n",
1442              "30/30 [==============================] - 4s 120ms/step - loss: 14.8677 - mae: 15.3451\n",
1443              "Epoch 432/500\n",
1444              "30/30 [==============================] - 4s 124ms/step - loss: 14.8472 - mae: 15.3248\n",
1445              "Epoch 433/500\n",
1446              "30/30 [==============================] - 4s 121ms/step - loss: 14.8261 - mae: 15.3042\n",
1447              "Epoch 434/500\n",
1448              "30/30 [==============================] - 4s 121ms/step - loss: 14.8062 - mae: 15.2840\n",
1449              "Epoch 435/500\n",
1450              "30/30 [==============================] - 4s 123ms/step - loss: 14.7952 - mae: 15.2738\n",
1451              "Epoch 436/500\n",
1452              "30/30 [==============================] - 4s 124ms/step - loss: 14.7654 - mae: 15.2433\n",
1453              "Epoch 437/500\n",
1454              "30/30 [==============================] - 4s 123ms/step - loss: 14.7590 - mae: 15.2371\n",
1455              "Epoch 438/500\n",
1456              "30/30 [==============================] - 4s 122ms/step - loss: 14.7409 - mae: 15.2188\n",
1457              "Epoch 439/500\n",
1458              "30/30 [==============================] - 4s 122ms/step - loss: 14.7369 - mae: 15.2145\n",
1459              "Epoch 440/500\n",
1460              "30/30 [==============================] - 4s 123ms/step - loss: 14.7359 - mae: 15.2128\n",
1461              "Epoch 441/500\n",
1462              "30/30 [==============================] - 4s 122ms/step - loss: 14.7529 - mae: 15.2294\n",
1463              "Epoch 442/500\n",
1464              "30/30 [==============================] - 4s 122ms/step - loss: 14.7586 - mae: 15.2364\n",
1465              "Epoch 443/500\n",
1466              "30/30 [==============================] - 4s 122ms/step - loss: 14.7411 - mae: 15.2179\n",
1467              "Epoch 444/500\n",
1468              "30/30 [==============================] - 4s 125ms/step - loss: 14.7496 - mae: 15.2256\n",
1469              "Epoch 445/500\n",
1470              "30/30 [==============================] - 4s 121ms/step - loss: 14.7623 - mae: 15.2386\n",
1471              "Epoch 446/500\n",
1472              "30/30 [==============================] - 4s 120ms/step - loss: 14.7594 - mae: 15.2359\n",
1473              "Epoch 447/500\n",
1474              "30/30 [==============================] - 4s 120ms/step - loss: 14.7482 - mae: 15.2246\n",
1475              "Epoch 448/500\n",
1476              "30/30 [==============================] - 4s 121ms/step - loss: 14.7512 - mae: 15.2256\n",
1477              "Epoch 449/500\n",
1478              "30/30 [==============================] - 4s 122ms/step - loss: 14.7993 - mae: 15.2779\n",
1479              "Epoch 450/500\n",
1480              "30/30 [==============================] - 4s 120ms/step - loss: 14.7277 - mae: 15.2037\n",
1481              "Epoch 451/500\n",
1482              "30/30 [==============================] - 4s 119ms/step - loss: 14.7533 - mae: 15.2273\n",
1483              "Epoch 452/500\n",
1484              "30/30 [==============================] - 4s 120ms/step - loss: 14.7952 - mae: 15.2748\n",
1485              "Epoch 453/500\n",
1486              "30/30 [==============================] - 4s 120ms/step - loss: 14.7063 - mae: 15.1814\n",
1487              "Epoch 454/500\n",
1488              "30/30 [==============================] - 4s 120ms/step - loss: 14.7631 - mae: 15.2386\n",
1489              "Epoch 455/500\n",
1490              "30/30 [==============================] - 4s 120ms/step - loss: 14.7732 - mae: 15.2499\n",
1491              "Epoch 456/500\n",
1492              "30/30 [==============================] - 4s 121ms/step - loss: 14.7496 - mae: 15.2261\n",
1493              "Epoch 457/500\n",
1494              "30/30 [==============================] - 4s 120ms/step - loss: 14.7387 - mae: 15.2135\n",
1495              "Epoch 458/500\n",
1496              "30/30 [==============================] - 4s 118ms/step - loss: 14.7660 - mae: 15.2421\n",
1497              "Epoch 459/500\n",
1498              "30/30 [==============================] - 4s 123ms/step - loss: 14.7536 - mae: 15.2308\n",
1499              "Epoch 460/500\n",
1500              "30/30 [==============================] - 4s 125ms/step - loss: 14.7186 - mae: 15.1929\n",
1501              "Epoch 461/500\n",
1502              "30/30 [==============================] - 4s 123ms/step - loss: 14.7576 - mae: 15.2340\n",
1503              "Epoch 462/500\n",
1504              "30/30 [==============================] - 4s 121ms/step - loss: 14.7405 - mae: 15.2161\n",
1505              "Epoch 463/500\n",
1506              "30/30 [==============================] - 4s 122ms/step - loss: 14.7459 - mae: 15.2233\n",
1507              "Epoch 464/500\n",
1508              "30/30 [==============================] - 4s 119ms/step - loss: 14.7082 - mae: 15.1823\n",
1509              "Epoch 465/500\n",
1510              "30/30 [==============================] - 4s 122ms/step - loss: 14.7462 - mae: 15.2231\n",
1511              "Epoch 466/500\n",
1512              "30/30 [==============================] - 4s 118ms/step - loss: 14.7128 - mae: 15.1876\n",
1513              "Epoch 467/500\n",
1514              "30/30 [==============================] - 4s 118ms/step - loss: 14.7396 - mae: 15.2177\n",
1515              "Epoch 468/500\n",
1516              "30/30 [==============================] - 4s 118ms/step - loss: 14.6750 - mae: 15.1490\n",
1517              "Epoch 469/500\n",
1518              "30/30 [==============================] - 4s 120ms/step - loss: 14.7325 - mae: 15.2082\n",
1519              "Epoch 470/500\n",
1520              "30/30 [==============================] - 4s 121ms/step - loss: 14.7191 - mae: 15.1959\n",
1521              "Epoch 471/500\n",
1522              "30/30 [==============================] - 4s 122ms/step - loss: 14.6834 - mae: 15.1574\n",
1523              "Epoch 472/500\n",
1524              "30/30 [==============================] - 4s 122ms/step - loss: 14.7225 - mae: 15.2007\n",
1525              "Epoch 473/500\n",
1526              "30/30 [==============================] - 4s 121ms/step - loss: 14.6300 - mae: 15.1033\n",
1527              "Epoch 474/500\n",
1528              "30/30 [==============================] - 4s 119ms/step - loss: 14.7171 - mae: 15.1931\n",
1529              "Epoch 475/500\n",
1530              "30/30 [==============================] - 4s 118ms/step - loss: 14.6865 - mae: 15.1614\n",
1531              "Epoch 476/500\n",
1532              "30/30 [==============================] - 4s 121ms/step - loss: 14.6942 - mae: 15.1701\n",
1533              "Epoch 477/500\n",
1534              "30/30 [==============================] - 4s 121ms/step - loss: 14.6815 - mae: 15.1562\n",
1535              "Epoch 478/500\n",
1536              "30/30 [==============================] - 4s 120ms/step - loss: 14.6963 - mae: 15.1730\n",
1537              "Epoch 479/500\n",
1538              "30/30 [==============================] - 4s 120ms/step - loss: 14.6516 - mae: 15.1249\n",
1539              "Epoch 480/500\n",
1540              "30/30 [==============================] - 4s 121ms/step - loss: 14.7075 - mae: 15.1856\n",
1541              "Epoch 481/500\n",
1542              "30/30 [==============================] - 4s 121ms/step - loss: 14.6256 - mae: 15.0971\n",
1543              "Epoch 482/500\n",
1544              "30/30 [==============================] - 4s 121ms/step - loss: 14.7322 - mae: 15.2145\n",
1545              "Epoch 483/500\n",
1546              "30/30 [==============================] - 4s 120ms/step - loss: 14.5774 - mae: 15.0521\n",
1547              "Epoch 484/500\n",
1548              "30/30 [==============================] - 4s 123ms/step - loss: 14.6449 - mae: 15.1149\n",
1549              "Epoch 485/500\n",
1550              "30/30 [==============================] - 4s 123ms/step - loss: 14.8674 - mae: 15.3445\n",
1551              "Epoch 486/500\n",
1552              "30/30 [==============================] - 4s 120ms/step - loss: 14.6951 - mae: 15.1767\n",
1553              "Epoch 487/500\n",
1554              "30/30 [==============================] - 4s 120ms/step - loss: 14.5554 - mae: 15.0313\n",
1555              "Epoch 488/500\n",
1556              "30/30 [==============================] - 4s 119ms/step - loss: 14.5959 - mae: 15.0650\n",
1557              "Epoch 489/500\n",
1558              "30/30 [==============================] - 4s 119ms/step - loss: 14.8240 - mae: 15.3052\n",
1559              "Epoch 490/500\n",
1560              "30/30 [==============================] - 4s 120ms/step - loss: 14.5971 - mae: 15.0731\n",
1561              "Epoch 491/500\n",
1562              "30/30 [==============================] - 4s 120ms/step - loss: 14.6331 - mae: 15.1062\n",
1563              "Epoch 492/500\n",
1564              "30/30 [==============================] - 4s 120ms/step - loss: 14.6785 - mae: 15.1581\n",
1565              "Epoch 493/500\n",
1566              "30/30 [==============================] - 4s 121ms/step - loss: 14.5661 - mae: 15.0376\n",
1567              "Epoch 494/500\n",
1568              "30/30 [==============================] - 4s 121ms/step - loss: 14.6962 - mae: 15.1786\n",
1569              "Epoch 495/500\n",
1570              "30/30 [==============================] - 4s 118ms/step - loss: 14.5328 - mae: 15.0051\n",
1571              "Epoch 496/500\n",
1572              "30/30 [==============================] - 4s 119ms/step - loss: 14.6723 - mae: 15.1502\n",
1573              "Epoch 497/500\n",
1574              "30/30 [==============================] - 4s 119ms/step - loss: 14.5884 - mae: 15.0601\n",
1575              "Epoch 498/500\n",
1576              "30/30 [==============================] - 4s 120ms/step - loss: 14.6835 - mae: 15.1656\n",
1577              "Epoch 499/500\n",
1578              "30/30 [==============================] - 4s 119ms/step - loss: 14.5250 - mae: 14.9968\n",
1579              "Epoch 500/500\n",
1580              "30/30 [==============================] - 4s 120ms/step - loss: 14.6739 - mae: 15.1532\n"
1581            ],
1582            "name": "stdout"
1583          }
1584        ]
1585      },
1586      {
1587        "cell_type": "code",
1588        "metadata": {
1589          "id": "GaC6NNMRp0lb"
1590        },
1591        "source": [
1592          "rnn_forecast = model_forecast(model, series[..., np.newaxis], window_size)\n",
1593          "rnn_forecast = rnn_forecast[split_time - window_size:-1, -1, 0]"
1594        ],
1595        "execution_count": null,
1596        "outputs": []
1597      },
1598      {
1599        "cell_type": "code",
1600        "metadata": {
1601          "colab": {
1602            "base_uri": "https://localhost:8080/",
1603            "height": 392
1604          },
1605          "id": "PrktQX3hKYex",
1606          "outputId": "3126e6a2-6f0a-4501-a5ea-360aab623062"
1607        },
1608        "source": [
1609          "plt.figure(figsize=(10, 6))\n",
1610          "plot_series(time_valid, x_valid)\n",
1611          "plot_series(time_valid, rnn_forecast)"
1612        ],
1613        "execution_count": null,
1614        "outputs": [
1615          {
1616            "output_type": "display_data",
1617            "data": {
1618              "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmcAAAF3CAYAAADgjOwXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nfkSCMyGEEEIIPyLBmRBCCCGEH/n/pRwwxK0NdyYAAAAASUVORK5CYII=\n",
1619              "text/plain": [
1620                "<Figure size 720x432 with 1 Axes>"
1621              ]
1622            },
1623            "metadata": {
1624              "tags": []
1625            }
1626          }
1627        ]
1628      },
1629      {
1630        "cell_type": "code",
1631        "metadata": {
1632          "colab": {
1633            "base_uri": "https://localhost:8080/",
1634            "height": 34
1635          },
1636          "id": "13XrorC5wQoE",
1637          "outputId": "392b1fbb-cc0c-4c3f-b246-73de03c381dc"
1638        },
1639        "source": [
1640          "tf.keras.metrics.mean_absolute_error(x_valid, rnn_forecast).numpy()"
1641        ],
1642        "execution_count": null,
1643        "outputs": [
1644          {
1645            "output_type": "execute_result",
1646            "data": {
1647              "text/plain": [
1648                "15.40074"
1649              ]
1650            },
1651            "metadata": {
1652              "tags": []
1653            },
1654            "execution_count": 41
1655          }
1656        ]
1657      },
1658      {
1659        "cell_type": "code",
1660        "metadata": {
1661          "colab": {
1662            "base_uri": "https://localhost:8080/",
1663            "height": 609
1664          },
1665          "id": "MD2kyYUVt3O0",
1666          "outputId": "4d4ce37c-ebd7-43cc-d0f3-761c37774bda"
1667        },
1668        "source": [
1669          "import matplotlib.image  as mpimg\n",
1670          "import matplotlib.pyplot as plt\n",
1671          "\n",
1672          "#-----------------------------------------------------------\n",
1673          "# Retrieve a list of list results on training and test data\n",
1674          "# sets for each training epoch\n",
1675          "#-----------------------------------------------------------\n",
1676          "loss=history.history['loss']\n",
1677          "\n",
1678          "epochs=range(len(loss)) # Get number of epochs\n",
1679          "\n",
1680          "\n",
1681          "#------------------------------------------------\n",
1682          "# Plot training and validation loss per epoch\n",
1683          "#------------------------------------------------\n",
1684          "plt.plot(epochs, loss, 'r')\n",
1685          "plt.title('Training loss')\n",
1686          "plt.xlabel(\"Epochs\")\n",
1687          "plt.ylabel(\"Loss\")\n",
1688          "plt.legend([\"Loss\"])\n",
1689          "\n",
1690          "plt.figure()\n",
1691          "\n",
1692          "\n",
1693          "\n",
1694          "zoomed_loss = loss[200:]\n",
1695          "zoomed_epochs = range(200,500)\n",
1696          "\n",
1697          "\n",
1698          "#------------------------------------------------\n",
1699          "# Plot training and validation loss per epoch\n",
1700          "#------------------------------------------------\n",
1701          "plt.plot(zoomed_epochs, zoomed_loss, 'r')\n",
1702          "plt.title('Training loss')\n",
1703          "plt.xlabel(\"Epochs\")\n",
1704          "plt.ylabel(\"Loss\")\n",
1705          "plt.legend([\"Loss\"])\n",
1706          "\n",
1707          "plt.figure()"
1708        ],
1709        "execution_count": null,
1710        "outputs": [
1711          {
1712            "output_type": "execute_result",
1713            "data": {
1714              "text/plain": [
1715                "<Figure size 432x288 with 0 Axes>"
1716              ]
1717            },
1718            "metadata": {
1719              "tags": []
1720            },
1721            "execution_count": 42
1722          },
1723          {
1724            "output_type": "display_data",
1725            "data": {
1726              "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEWCAYAAABrDZDcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nZlZwDgIzs4JzEJiZFZyDwMys4P4/yTzCS0vyvfoAAAAASUVORK5CYII=\n",
1727              "text/plain": [
1728                "<Figure size 432x288 with 1 Axes>"
1729              ]
1730            },
1731            "metadata": {
1732              "tags": []
1733            }
1734          },
1735          {
1736            "output_type": "display_data",
1737            "data": {
1738              "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYwAAAEWCAYAAAB1xKBvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n6cCUtEQmmXf88fDLL6FX1bvvhoTSrx+