Прогноз_продаж_Этиацетат.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Прогноз продаж Этиацетат.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "machine_shape": "hm",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "TPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/StetskoSergey/pida5_stetsko/blob/master/%D0%9F%D1%80%D0%BE%D0%B3%D0%BD%D0%BE%D0%B7_%D0%BF%D1%80%D0%BE%D0%B4%D0%B0%D0%B6_%D0%AD%D1%82%D0%B8%D0%B0%D1%86%D0%B5%D1%82%D0%B0%D1%82.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "e-z6CrL2kzQw"
      },
      "source": [
        "Добрый день.\r\n",
        "Задача проверить какую точность может дать прогноз продаж товара, на пелче поставки 3 месяца. Есть данные по продажам продукта в кг за прошедшие 3 года, по неделям. \r\n",
        "Задача сделать прогнозную модель на 3 месяца вперед."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GNkzTFfynsmV",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 50
        },
        "outputId": "778f8f47-d565-414b-cd2b-3d6d9644bb01"
      },
      "source": [
        "%tensorflow_version 2.x\n",
        "import tensorflow as tf\n",
        "print(\"Tensorflow version \" + tf.__version__)\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "TensorFlow 2.x selected.\n",
            "Tensorflow version 2.1.0\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sLl52leVp5wU"
      },
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "\n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "def plot_series(time, series, format=\"-\", start=0, end=None):\n",
        "    plt.plot(time[start:end], series[start:end], format)\n",
        "    plt.xlabel(\"Time\")\n",
        "    plt.ylabel(\"Value\")\n",
        "    plt.grid(True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tP7oqUdkk0gY",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 388
        },
        "outputId": "05582f22-dde3-46c7-a077-22ebfbda8388"
      },
      "source": [
        "sales = pd.read_csv('sales.csv', names = ['time', 'sale'])\n",
        "\n",
        "series = np.array(sales['sale'])\n",
        "time = np.array(sales['time'])\n",
        "plt.figure(figsize=(10, 6))\n",
        "plot_series(time, series)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAncAAAFzCAYAAABcsy/pAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nUnkAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 720x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9SyXAHxNko_G"
      },
      "source": [
        ""
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lJwUUZscnG38"
      },
      "source": [
        "def univariate_data(dataset, start_index, end_index, history_size, target_size):\n",
        "  data = []\n",
        "  labels = []\n",
        "\n",
        "  start_index = start_index + history_size\n",
        "  if end_index is None:\n",
        "    end_index = len(dataset) - target_size\n",
        "\n",
        "  for i in range(start_index, end_index):\n",
        "    indices = range(i-history_size, i)\n",
        "    # Reshape data from (history_size,) to (history_size, 1)\n",
        "    data.append(np.reshape(dataset[indices], (history_size, 1)))\n",
        "    labels.append(dataset[i+target_size])\n",
        "  return np.array(data), np.array(labels)\n",
        "\n",
        "def multivariate_data(dataset, target, start_index, end_index, history_size,\n",
        "                      target_size, step, single_step=False):\n",
        "  data = []\n",
        "  labels = []\n",
        "\n",
        "  start_index = start_index + history_size\n",
        "  if end_index is None:\n",
        "    end_index = len(dataset) - target_size\n",
        "\n",
        "  for i in range(start_index, end_index):\n",
        "    indices = range(i-history_size, i, step)\n",
        "    data.append(dataset[indices])\n",
        "\n",
        "    if single_step:\n",
        "      labels.append(target[i+target_size])\n",
        "    else:\n",
        "      labels.append(target[i:i+target_size])\n",
        "\n",
        "  return np.array(data), np.array(labels)\n",
        "\n",
        "def make_window_dataset(ds, window_size=5, shift=1, stride=1):\n",
        "  windows = ds.window(window_size+1, shift=shift, stride=stride)\n",
        "\n",
        "  def sub_to_batch(sub):\n",
        "    return sub.batch(window_size+1, drop_remainder=True)\n",
        "\n",
        "  windows = windows.flat_map(sub_to_batch)\n",
        "  windows = windows.map(lambda w: (w[:-1], w[-1:]))\n",
        "  return windows\n",
        "\n",
        "def dense_10_step(batch):\n",
        "  # Shift features and labels one step relative to each other.\n",
        "  return batch[:-10], batch[10:]\n",
        "\n",
        "def windowed_dataset(series, window_size, batch_size, shuffle_buffer):\n",
        "    series = tf.expand_dims(series, axis=-1)\n",
        "    ds = tf.data.Dataset.from_tensor_slices(series)\n",
        "    ds = ds.window(window_size + 10, shift=1, drop_remainder=True)\n",
        "    ds = ds.flat_map(lambda w: w.batch(window_size + 10))\n",
        "    #ds = ds.shuffle(shuffle_buffer)\n",
        "    ds = ds.map(dense_10_step)\n",
        "    return ds.batch(batch_size).prefetch(1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "L92YRw_IpCFG"
      },
      "source": [
        "split_time = 104\n",
        "series = np.array(sales['sale'])\n",
        "\n",
        "train_mean = series[:split_time].mean()\n",
        "train_std = series[:split_time].std()\n",
        "uni_data = (series-train_mean)/train_std\n",
        "\n",
        "window_size = 52\n",
        "batch_size = 52\n",
        "shuffle_buffer_size = 1000\n",
        "\n",
        "train_set = windowed_dataset(uni_data[:split_time], window_size, batch_size, shuffle_buffer_size)\n",
        "valid_set = windowed_dataset(uni_data[split_time:], window_size, batch_size, shuffle_buffer_size)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "A5Ex2G8fQiLI"
      },
      "source": [
        "for x, y in train_set.take(2):\n",
        "  print(y)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zYIFQ7wRRsmB"
      },
      "source": [
        "model = tf.keras.models.Sequential([\n",
        "  tf.keras.layers.Conv1D(filters=5, kernel_size=5,\n",
        "                      strides=1, padding=\"causal\",\n",
        "                      activation=\"relu\",\n",
        "                      input_shape=[None, 1]),\n",
        "  tf.keras.layers.LSTM(64, return_sequences=True),\n",
        "  tf.keras.layers.LSTM(64, return_sequences=True),\n",
        "  tf.keras.layers.Dense(30, activation=\"relu\"),\n",
        "  tf.keras.layers.Dense(10, activation=\"relu\"),\n",
        "  tf.keras.layers.Dense(1),\n",
        "  #tf.keras.layers.Lambda(lambda x: x * 400)\n",
        "])\n",
        "\n",
        "lr_schedule = tf.keras.callbacks.LearningRateScheduler(\n",
        "    lambda epoch: 1e-6 * 10**(epoch / 20))\n",
        "optimizer = tf.keras.optimizers.SGD(lr=1e-6, momentum=0.9)\n",
        "model.compile(loss=tf.keras.losses.Huber(),\n",
        "              optimizer=optimizer,\n",
        "              metrics=[\"mae\"])\n",
        "history = model.fit(train_set, epochs=100, callbacks=[lr_schedule])\n",
        "\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-ufSGdsZSQAd"
      },
      "source": [
        "plt.semilogx(history.history[\"lr\"], history.history[\"loss\"])\n",
        "plt.axis([1e-8, 1e-1, 0, 6])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "luXQPXBCg6BF"
      },
      "source": [
        "def plot_train_history(history, title):\n",
        "  loss = history.history['loss']\n",
        "  val_loss = history.history['val_loss']\n",
        "\n",
        "  epochs = range(len(loss))\n",
        "\n",
        "  plt.figure()\n",
        "\n",
        "  plt.plot(epochs, loss, 'b', label='Training loss')\n",
        "  plt.plot(epochs, val_loss, 'r', label='Validation loss')\n",
        "  plt.title(title)\n",
        "  plt.legend()\n",
        "\n",
        "  plt.show()\n",
        "\n",
        "def multi_step_plot(history, true_future, prediction):\n",
        "  plt.figure(figsize=(12, 6))\n",
        "  num_in = create_time_steps(len(history))\n",
        "  num_out = len(true_future)\n",
        "\n",
        "  plt.plot(num_in, np.array(history[:, 1]), label='History')\n",
        "  plt.plot(np.arange(num_out)/STEP, np.array(true_future), 'bo',\n",
        "           label='True Future')\n",
        "  if prediction.any():\n",
        "    plt.plot(np.arange(num_out)/STEP, np.array(prediction), 'ro',\n",
        "             label='Predicted Future')\n",
        "  plt.legend(loc='upper left')\n",
        "  plt.show()  "
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4ryr6VrvT0LG"
      },
      "source": [
        "model = tf.keras.models.Sequential([\n",
        "  tf.keras.layers.Conv1D(filters=5, kernel_size=5,\n",
        "                      strides=1, padding=\"causal\",\n",
        "                      activation=\"relu\",\n",
        "                      input_shape=[None, 1]),\n",
        "  tf.keras.layers.LSTM(52 ,return_sequences=True),\n",
        "  tf.keras.layers.LSTM(52, return_sequences=True),\n",
        "  tf.keras.layers.Dense(52, activation=\"relu\"),\n",
        "  tf.keras.layers.Dense(30, activation=\"relu\"),\n",
        "  tf.keras.layers.Dense(10),\n",
        "  #tf.keras.layers.Lambda(lambda x: x * 400)\n",
        "])\n",
        "\n",
        "\n",
        "#optimizer=tf.keras.optimizers.RMSprop()\n",
        "model.compile(optimizer='adam', loss='mae')\n",
        "history = model.fit(train_set,epochs=1000)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "x_K3SeFxhM2w"
      },
      "source": [
        "plot_train_history(history, 'Multi-Step Training and validation loss')\n",
        "\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yVdjS15whgph"
      },
      "source": [
        "for x, y in valid_set.take(3):\n",
        "  print(y)\n",
        "  #multi_step_plot(x[0], y[0], model.predict(x)[0])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Hz9v7pddiMA1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "0f6a124d-7ce1-495b-9f9d-ab88302d68b6"
      },
      "source": [
        "valid_set.take(0)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<TakeDataset shapes: ((None, None, 1), (None, None, 1)), types: (tf.float64, tf.float64)>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 243
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4XwGrf-A_wF0"
      },
      "source": [
        "def model_forecast(model, series, window_size):\n",
        "    ds = tf.data.Dataset.from_tensor_slices(series)\n",
        "    ds = ds.window(window_size, shift=1, drop_remainder=True)\n",
        "    ds = ds.flat_map(lambda w: w.batch(window_size))\n",
        "    ds = ds.batch(1).prefetch(1)\n",
        "    forecast = model.predict(ds)\n",
        "    return forecast\n",
        "\n",
        "def create_time_steps(length):\n",
        "  return list(range(-length, 0))\n",
        "\n",
        "def show_plot(plot_data, delta, title):\n",
        "  labels = ['History', 'True Future', 'Model Prediction']\n",
        "  marker = ['.-', 'rx', 'go']\n",
        "  time_steps = create_time_steps(plot_data[0].shape[0])\n",
        "  if delta:\n",
        "    future = delta\n",
        "  else:\n",
        "    future = 0\n",
        "\n",
        "  plt.title(title)\n",
        "  for i, x in enumerate(plot_data):\n",
        "    if i:\n",
        "      plt.plot(future, plot_data[i], marker[i], markersize=10,\n",
        "               label=labels[i])\n",
        "    else:\n",
        "      plt.plot(time_steps, plot_data[i].flatten(), marker[i], label=labels[i])\n",
        "  plt.legend()\n",
        "  plt.xlim([time_steps[0], (future+5)*2])\n",
        "  plt.xlabel('Time-Step')\n",
        "  return plt      "
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FAKYO08wVmfW"
      },
      "source": [
        "rnn_forecast = model_forecast(model, uni_data[-window_size:, np.newaxis], window_size)\n",
        "#rnn_forecast = rnn_forecast[split_time - window_size:-1, -1, 0]\n",
        "rnn_forecast = rnn_forecast * train_std + train_mean"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EbG9CyXSUkei",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "6f9ab032-4d55-4d9d-ac49-62b93b5451c7"
      },
      "source": [
        "ds = tf.data.Dataset.from_tensor_slices(uni_data[-window_size:, np.newaxis])\n",
        "ds = ds.window(window_size, shift=1, drop_remainder=True)\n",
        "ds = ds.flat_map(lambda w: w.batch(window_size))\n",
        "for x in ds.take():\n",
        "  print(x.shape)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(52, 1)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9F8QJdNJV2jx",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 388
        },
        "outputId": "3278787a-a764-42b6-dd6e-95af80f15d3d"
      },
      "source": [
        "x_valid = series[split_time:]\n",
        "plt.figure(figsize=(10, 6))\n",
        "plot_series(time_valid, x_valid)\n",
        "plot_series(time_valid, rnn_forecast)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAncAAAFzCAYAAABcsy/pAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\npXrQbrYAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 720x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4r39rkS6XfKP",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "265b60e0-73a7-4f72-8b53-755aee6cf491"
      },
      "source": [
        "tf.keras.metrics.mean_absolute_error(x_valid, rnn_forecast).numpy()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "28265.066"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 252
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7qLbylu4gGnl"
      },
      "source": [
        "TRAIN_SPLIT = 104\n",
        "uni_data = series\n",
        "\n",
        "uni_train_mean = uni_data[:TRAIN_SPLIT].mean()\n",
        "uni_train_std = uni_data[:TRAIN_SPLIT].std()\n",
        "uni_data = (uni_data-uni_train_mean)/uni_train_std\n",
        "\n",
        "univariate_past_history = 52\n",
        "univariate_future_target = 0\n",
        "\n",
        "x_train_uni, y_train_uni = univariate_data(uni_data, 0, TRAIN_SPLIT,\n",
        "                                           univariate_past_history,\n",
        "                                           univariate_future_target)\n",
        "x_val_uni, y_val_uni = univariate_data(uni_data, TRAIN_SPLIT, None,\n",
        "                                       univariate_past_history,\n",
        "                                       univariate_future_target)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IiJUa8GIhfj5"
      },
      "source": [
        "BATCH_SIZE = 10\n",
        "BUFFER_SIZE = 10000\n",
        "\n",
        "train_univariate = tf.data.Dataset.from_tensor_slices((x_train_uni, y_train_uni))\n",
        "train_univariate = train_univariate.cache().shuffle(BUFFER_SIZE).batch(BATCH_SIZE).repeat()\n",
        "\n",
        "val_univariate = tf.data.Dataset.from_tensor_slices((x_val_uni, y_val_uni))\n",
        "val_univariate = val_univariate.batch(BATCH_SIZE).repeat()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "h-uB2B0Xhkhk"
      },
      "source": [
        "single_step_model = tf.keras.models.Sequential()\n",
        "single_step_model.add(tf.keras.layers.LSTM(32,\n",
        "                                           input_shape=(1,52), return_sequences = True))\n",
        "single_step_model.add(tf.keras.layers.Dense(1))\n",
        "\n",
        "single_step_model.compile(optimizer=tf.keras.optimizers.RMSprop(), loss='mae')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZsBod7r8lksh"
      },
      "source": [
        "for x, y in val_univariate.take(1):\n",
        "  print(single_step_model.predict(x).shape)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4WN9Da2Bhsn4",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 488
        },
        "outputId": "6e42743a-35de-4f24-ac08-896bbcba2d68"
      },
      "source": [
        "EVALUATION_INTERVAL = 200\n",
        "EPOCHS = 10\n",
        "\n",
        "single_step_history = single_step_model.fit(train_univariate, epochs=EPOCHS,\n",
        "                                            steps_per_epoch=EVALUATION_INTERVAL,\n",
        "                                            validation_data=val_univariate,\n",
        "                                            validation_steps=50)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "ValueError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-87-304fa2da7a3c>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      5\u001b[0m                                             \u001b[0msteps_per_epoch\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mEVALUATION_INTERVAL\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      6\u001b[0m                                             \u001b[0mvalidation_data\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mval_univariate\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 7\u001b[0;31m                                             validation_steps=50)\n\u001b[0m",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/keras/engine/training.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_freq, max_queue_size, workers, use_multiprocessing, **kwargs)\u001b[0m\n\u001b[1;32m    817\u001b[0m         \u001b[0mmax_queue_size\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mmax_queue_size\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    818\u001b[0m         \u001b[0mworkers\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mworkers\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 819\u001b[0;31m         use_multiprocessing=use_multiprocessing)\n\u001b[0m\u001b[1;32m    820\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    821\u001b[0m   def evaluate(self,\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/keras/engine/training_v2.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, model, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_freq, max_queue_size, workers, use_multiprocessing, **kwargs)\u001b[0m\n\u001b[1;32m    233\u001b[0m           \u001b[0mmax_queue_size\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mmax_queue_size\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    234\u001b[0m           \u001b[0mworkers\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mworkers\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 235\u001b[0;31m           use_multiprocessing=use_multiprocessing)\n\u001b[0m\u001b[1;32m    236\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    237\u001b[0m       \u001b[0mtotal_samples\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_get_total_number_of_samples\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtraining_data_adapter\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/keras/engine/training_v2.py\u001b[0m in \u001b[0;36m_process_training_inputs\u001b[0;34m(model, x, y, batch_size, epochs, sample_weights, class_weights, steps_per_epoch, validation_split, validation_data, validation_steps, shuffle, distribution_strategy, max_queue_size, workers, use_multiprocessing)\u001b[0m\n\u001b[1;32m    612\u001b[0m           \u001b[0mclass_weights\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mclass_weights\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    613\u001b[0m           \u001b[0msteps\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mvalidation_steps\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 614\u001b[0;31m           distribution_strategy=distribution_strategy)\n\u001b[0m\u001b[1;32m    615\u001b[0m     \u001b[0;32melif\u001b[0m \u001b[0mvalidation_steps\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    616\u001b[0m       raise ValueError('`validation_steps` should not be specified if '\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/keras/engine/training_v2.py\u001b[0m in \u001b[0;36m_process_inputs\u001b[0;34m(model, mode, x, y, batch_size, epochs, sample_weights, class_weights, shuffle, steps, distribution_strategy, max_queue_size, workers, use_multiprocessing)\u001b[0m\n\u001b[1;32m    704\u001b[0m       \u001b[0mmax_queue_size\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mmax_queue_size\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    705\u001b[0m       \u001b[0mworkers\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mworkers\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 706\u001b[0;31m       use_multiprocessing=use_multiprocessing)\n\u001b[0m\u001b[1;32m    707\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    708\u001b[0m   \u001b[0;32mreturn\u001b[0m \u001b[0madapter\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/keras/engine/data_adapter.py\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, x, y, sample_weights, standardize_function, **kwargs)\u001b[0m\n\u001b[1;32m    700\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    701\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mstandardize_function\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 702\u001b[0;31m       \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mstandardize_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    703\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    704\u001b[0m     \u001b[0;31m# Note that the dataset instance is immutable, its fine to reusing the user\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/keras/engine/training_v2.py\u001b[0m in \u001b[0;36mstandardize_function\u001b[0;34m(dataset)\u001b[0m\n\u001b[1;32m    682\u001b[0m           \u001b[0;32mreturn\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    683\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msample_weights\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 684\u001b[0;31m       \u001b[0;32mreturn\u001b[0m \u001b[0mdataset\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmap\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmap_fn\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnum_parallel_calls\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdataset_ops\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mAUTOTUNE\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    685\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    686\u001b[0m   \u001b[0;32mif\u001b[0m \u001b[0mmode\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0mModeKeys\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mPREDICT\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/data/ops/dataset_ops.py\u001b[0m in \u001b[0;36mmap\u001b[0;34m(self, map_func, num_parallel_calls)\u001b[0m\n\u001b[1;32m   1589\u001b[0m     \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1590\u001b[0m       return ParallelMapDataset(\n\u001b[0;32m-> 1591\u001b[0;31m           self, map_func, num_parallel_calls, preserve_cardinality=True)\n\u001b[0m\u001b[1;32m   1592\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1593\u001b[0m   \u001b[0;32mdef\u001b[0m \u001b[0mflat_map\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmap_func\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/data/ops/dataset_ops.py\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, input_dataset, map_func, num_parallel_calls, use_inter_op_parallelism, preserve_cardinality, use_legacy_function)\u001b[0m\n\u001b[1;32m   3924\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_transformation_name\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3925\u001b[0m         \u001b[0mdataset\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0minput_dataset\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 3926\u001b[0;31m         use_legacy_function=use_legacy_function)\n\u001b[0m\u001b[1;32m   3927\u001b[0m     self._num_parallel_calls = ops.convert_to_tensor(\n\u001b[1;32m   3928\u001b[0m         num_parallel_calls, dtype=dtypes.int32, name=\"num_parallel_calls\")\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/data/ops/dataset_ops.py\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, func, transformation_name, dataset, input_classes, input_shapes, input_types, input_structure, add_to_graph, use_legacy_function, defun_kwargs)\u001b[0m\n\u001b[1;32m   3145\u001b[0m       \u001b[0;32mwith\u001b[0m \u001b[0mtracking\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mresource_tracker_scope\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mresource_tracker\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3146\u001b[0m         \u001b[0;31m# TODO(b/141462134): Switch to using garbage collection.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 3147\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_function\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mwrapper_fn\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_get_concrete_function_internal\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   3148\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3149\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0madd_to_graph\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/eager/function.py\u001b[0m in \u001b[0;36m_get_concrete_function_internal\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   2393\u001b[0m     \u001b[0;34m\"\"\"Bypasses error checking when getting a graph function.\"\"\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2394\u001b[0m     graph_function = self._get_concrete_function_internal_garbage_collected(\n\u001b[0;32m-> 2395\u001b[0;31m         *args, **kwargs)\n\u001b[0m\u001b[1;32m   2396\u001b[0m     \u001b[0;31m# We're returning this concrete function to someone, and they may keep a\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2397\u001b[0m     \u001b[0;31m# reference to the FuncGraph without keeping a reference to the\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/eager/function.py\u001b[0m in \u001b[0;36m_get_concrete_function_internal_garbage_collected\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   2387\u001b[0m       \u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2388\u001b[0m     \u001b[0;32mwith\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_lock\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2389\u001b[0;31m       \u001b[0mgraph_function\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0m_\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0m_\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_maybe_define_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   2390\u001b[0m     \u001b[0;32mreturn\u001b[0m \u001b[0mgraph_function\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2391\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/eager/function.py\u001b[0m in \u001b[0;36m_maybe_define_function\u001b[0;34m(self, args, kwargs)\u001b[0m\n\u001b[1;32m   2701\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2702\u001b[0m       \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_function_cache\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmissed\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0madd\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcall_context_key\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2703\u001b[0;31m       \u001b[0mgraph_function\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_create_graph_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   2704\u001b[0m       \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_function_cache\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprimary\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mcache_key\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mgraph_function\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2705\u001b[0m       \u001b[0;32mreturn\u001b[0m \u001b[0mgraph_function\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/eager/function.py\u001b[0m in \u001b[0;36m_create_graph_function\u001b[0;34m(self, args, kwargs, override_flat_arg_shapes)\u001b[0m\n\u001b[1;32m   2591\u001b[0m             \u001b[0marg_names\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0marg_names\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2592\u001b[0m             \u001b[0moverride_flat_arg_shapes\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0moverride_flat_arg_shapes\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2593\u001b[0;31m             capture_by_value=self._capture_by_value),\n\u001b[0m\u001b[1;32m   2594\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_function_attributes\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2595\u001b[0m         \u001b[0;31m# Tell the ConcreteFunction to clean up its graph once it goes out of\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/framework/func_graph.py\u001b[0m in \u001b[0;36mfunc_graph_from_py_func\u001b[0;34m(name, python_func, args, kwargs, signature, func_graph, autograph, autograph_options, add_control_dependencies, arg_names, op_return_value, collections, capture_by_value, override_flat_arg_shapes)\u001b[0m\n\u001b[1;32m    976\u001b[0m                                           converted_func)\n\u001b[1;32m    977\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 978\u001b[0;31m       \u001b[0mfunc_outputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpython_func\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mfunc_args\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mfunc_kwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    979\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    980\u001b[0m       \u001b[0;31m# invariant: `func_outputs` contains only Tensors, CompositeTensors,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/data/ops/dataset_ops.py\u001b[0m in \u001b[0;36mwrapper_fn\u001b[0;34m(*args)\u001b[0m\n\u001b[1;32m   3138\u001b[0m           attributes=defun_kwargs)\n\u001b[1;32m   3139\u001b[0m       \u001b[0;32mdef\u001b[0m \u001b[0mwrapper_fn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m  \u001b[0;31m# pylint: disable=missing-docstring\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 3140\u001b[0;31m         \u001b[0mret\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_wrapper_helper\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   3141\u001b[0m         \u001b[0mret\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mstructure\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto_tensor_list\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_output_structure\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mret\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3142\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mops\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconvert_to_tensor\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mt\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mt\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mret\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/data/ops/dataset_ops.py\u001b[0m in \u001b[0;36m_wrapper_helper\u001b[0;34m(*args)\u001b[0m\n\u001b[1;32m   3080\u001b[0m         \u001b[0mnested_args\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mnested_args\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3081\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 3082\u001b[0;31m       \u001b[0mret\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mautograph\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtf_convert\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfunc\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mag_ctx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mnested_args\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   3083\u001b[0m       \u001b[0;31m# If `func` returns a list of tensors, `nest.flatten()` and\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3084\u001b[0m       \u001b[0;31m# `ops.convert_to_tensor()` would conspire to attempt to stack\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/autograph/impl/api.py\u001b[0m in \u001b[0;36mwrapper\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m    235\u001b[0m       \u001b[0;32mexcept\u001b[0m \u001b[0mException\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m  \u001b[0;31m# pylint:disable=broad-except\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    236\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mhasattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0me\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'ag_error_metadata'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 237\u001b[0;31m           \u001b[0;32mraise\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mag_error_metadata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto_exception\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0me\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    238\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    239\u001b[0m           \u001b[0;32mraise\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mValueError\u001b[0m: in converted code:\n\n    /tensorflow-2.1.0/python3.6/tensorflow_core/python/keras/engine/training_v2.py:677 map_fn\n        batch_size=None)\n    /tensorflow-2.1.0/python3.6/tensorflow_core/python/keras/engine/training.py:2410 _standardize_tensors\n        exception_prefix='input')\n    /tensorflow-2.1.0/python3.6/tensorflow_core/python/keras/engine/training_utils.py:573 standardize_input_data\n        'with shape ' + str(data_shape))\n\n    ValueError: Error when checking input: expected lstm_14_input to have 3 dimensions, but got array with shape (None, 1)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rPOIOGZPw2CQ",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 185
        },
        "outputId": "95ffd3e1-55ea-40a9-b8f5-bcc4874be81a"
      },
      "source": [
        "TRAIN_SPLIT = 104\n",
        "uni_data = series\n",
        "\n",
        "uni_train_mean = uni_data[:TRAIN_SPLIT].mean()\n",
        "uni_train_std = uni_data[:TRAIN_SPLIT].std()\n",
        "uni_data = (uni_data-uni_train_mean)/uni_train_std\n",
        "\n",
        "x_train = tf.data.Dataset.from_tensor_slices(uni_data)\n",
        "\n",
        "ds = make_window_dataset(x_train, window_size=10, shift = 1, stride=1)\n",
        "\n",
        "for example, i in ds.take(10):\n",
        "  print(i)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "tf.Tensor([0.35972211], shape=(1,), dtype=float64)\n",
            "tf.Tensor([-0.71460405], shape=(1,), dtype=float64)\n",
            "tf.Tensor([0.54521862], shape=(1,), dtype=float64)\n",
            "tf.Tensor([-0.09423895], shape=(1,), dtype=float64)\n",
            "tf.Tensor([0.71419639], shape=(1,), dtype=float64)\n",
            "tf.Tensor([0.63384889], shape=(1,), dtype=float64)\n",
            "tf.Tensor([0.31985252], shape=(1,), dtype=float64)\n",
            "tf.Tensor([0.08821586], shape=(1,), dtype=float64)\n",
            "tf.Tensor([-0.47051984], shape=(1,), dtype=float64)\n",
            "tf.Tensor([0.257568], shape=(1,), dtype=float64)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pHoZvDo57Nfn"
      },
      "source": [
        "TRAIN_SPLIT = 104\n",
        "uni_data = series\n",
        "\n",
        "uni_train_mean = uni_data[:TRAIN_SPLIT].mean()\n",
        "uni_train_std = uni_data[:TRAIN_SPLIT].std()\n",
        "x_train = (uni_data-uni_train_mean)/uni_train_std"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AclfYY3Mn6Ph"
      },
      "source": [
        "tf.keras.backend.clear_session()\n",
        "tf.random.set_seed(51)\n",
        "np.random.seed(51)\n",
        "window_size = 10\n",
        "batch_size = 256\n",
        "train_set = windowed_dataset(x_train, window_size, batch_size, shuffle_buffer_size)\n",
        "print(train_set)\n",
        "print(x_train.shape)\n",
        "\n",
        "model = tf.keras.models.Sequential([\n",
        "  tf.keras.layers.Conv1D(filters=52, kernel_size=5,\n",
        "                      strides=1, padding=\"causal\",\n",
        "                      activation=\"relu\",\n",
        "                      input_shape=[None, 1]),\n",
        "  tf.keras.layers.LSTM(64, return_sequences=True),\n",
        "  tf.keras.layers.LSTM(64, return_sequences=True),\n",
        "  tf.keras.layers.Dense(30, activation=\"relu\"),\n",
        "  tf.keras.layers.Dense(10, activation=\"relu\"),\n",
        "  tf.keras.layers.Dense(1),\n",
        "  tf.keras.layers.Lambda(lambda x: x * 400)\n",
        "])\n",
        "\n",
        "lr_schedule = tf.keras.callbacks.LearningRateScheduler(\n",
        "    lambda epoch: 1e-8 * 10**(epoch / 20))\n",
        "optimizer = tf.keras.optimizers.SGD(lr=1e-8, momentum=0.9)\n",
        "model.compile(loss=tf.keras.losses.Huber(),\n",
        "              optimizer=optimizer,\n",
        "              metrics=[\"mae\"])\n",
        "history = model.fit(train_set, epochs=100, callbacks=[lr_schedule])\n",
        "\n",
        "\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vVcKmg7Q_7rD",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 290
        },
        "outputId": "daa6a862-8af1-4464-c48d-faa2f37c3914"
      },
      "source": [
        "plt.semilogx(history.history[\"lr\"], history.history[\"loss\"])\n",
        "plt.axis([1e-8, 1e-3, 0, 6])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[1e-08, 0.001, 0, 6]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 134
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXYAAAEACAYAAACnJV25AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nACiGsANAMYQdAIoh7ABQzP8B30Y+mqqg2BQAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QsksvkcXAAgq",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 488
        },
        "outputId": "eeb02fb6-2184-44e0-dda2-36bbdd1a0fa1"
      },
      "source": [
        "tf.keras.backend.clear_session()\n",
        "tf.random.set_seed(51)\n",
        "np.random.seed(51)\n",
        "train_set = windowed_dataset(x_train, window_size=10, batch_size=100, shuffle_buffer=shuffle_buffer_size)\n",
        "model = tf.keras.models.Sequential([\n",
        "  #tf.keras.layers.Conv1D(filters=2, kernel_size=5,\n",
        "  #                    strides=1, padding=\"causal\",\n",
        "  #                    activation=\"relu\",\n",
        "  #                    input_shape=[None, 1]),\n",
        "  tf.keras.layers.LSTM(60, input_shape = (1,10) ,return_sequences=True),\n",
        "  tf.keras.layers.LSTM(60, return_sequences=True),\n",
        "  tf.keras.layers.Dense(30, activation=\"relu\"),\n",
        "  tf.keras.layers.Dense(10, activation=\"relu\"),\n",
        "  tf.keras.layers.Dense(1),\n",
        "  #tf.keras.layers.Lambda(lambda x: x * 400)\n",
        "])\n",
        "\n",
        "\n",
        "optimizer = tf.keras.optimizers.SGD(lr=1e-4, momentum=0.9)\n",
        "model.compile(loss=tf.keras.losses.Huber(),\n",
        "              optimizer=optimizer,\n",
        "              metrics=[\"mae\"])\n",
        "history = model.fit(train_set,epochs=500)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "ValueError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-116-57ede12ece7f>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     21\u001b[0m               \u001b[0moptimizer\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0moptimizer\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     22\u001b[0m               metrics=[\"mae\"])\n\u001b[0;32m---> 23\u001b[0;31m \u001b[0mhistory\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrain_set\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mepochs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m500\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/keras/engine/training.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_freq, max_queue_size, workers, use_multiprocessing, **kwargs)\u001b[0m\n\u001b[1;32m    817\u001b[0m         \u001b[0mmax_queue_size\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mmax_queue_size\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    818\u001b[0m         \u001b[0mworkers\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mworkers\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 819\u001b[0;31m         use_multiprocessing=use_multiprocessing)\n\u001b[0m\u001b[1;32m    820\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    821\u001b[0m   def evaluate(self,\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/keras/engine/training_v2.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, model, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_freq, max_queue_size, workers, use_multiprocessing, **kwargs)\u001b[0m\n\u001b[1;32m    233\u001b[0m           \u001b[0mmax_queue_size\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mmax_queue_size\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    234\u001b[0m           \u001b[0mworkers\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mworkers\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 235\u001b[0;31m           use_multiprocessing=use_multiprocessing)\n\u001b[0m\u001b[1;32m    236\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    237\u001b[0m       \u001b[0mtotal_samples\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_get_total_number_of_samples\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtraining_data_adapter\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/keras/engine/training_v2.py\u001b[0m in \u001b[0;36m_process_training_inputs\u001b[0;34m(model, x, y, batch_size, epochs, sample_weights, class_weights, steps_per_epoch, validation_split, validation_data, validation_steps, shuffle, distribution_strategy, max_queue_size, workers, use_multiprocessing)\u001b[0m\n\u001b[1;32m    591\u001b[0m         \u001b[0mmax_queue_size\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mmax_queue_size\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    592\u001b[0m         \u001b[0mworkers\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mworkers\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 593\u001b[0;31m         use_multiprocessing=use_multiprocessing)\n\u001b[0m\u001b[1;32m    594\u001b[0m     \u001b[0mval_adapter\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    595\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mvalidation_data\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/keras/engine/training_v2.py\u001b[0m in \u001b[0;36m_process_inputs\u001b[0;34m(model, mode, x, y, batch_size, epochs, sample_weights, class_weights, shuffle, steps, distribution_strategy, max_queue_size, workers, use_multiprocessing)\u001b[0m\n\u001b[1;32m    704\u001b[0m       \u001b[0mmax_queue_size\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mmax_queue_size\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    705\u001b[0m       \u001b[0mworkers\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mworkers\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 706\u001b[0;31m       use_multiprocessing=use_multiprocessing)\n\u001b[0m\u001b[1;32m    707\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    708\u001b[0m   \u001b[0;32mreturn\u001b[0m \u001b[0madapter\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/keras/engine/data_adapter.py\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, x, y, sample_weights, standardize_function, **kwargs)\u001b[0m\n\u001b[1;32m    700\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    701\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mstandardize_function\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 702\u001b[0;31m       \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mstandardize_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    703\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    704\u001b[0m     \u001b[0;31m# Note that the dataset instance is immutable, its fine to reusing the user\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/keras/engine/training_v2.py\u001b[0m in \u001b[0;36mstandardize_function\u001b[0;34m(dataset)\u001b[0m\n\u001b[1;32m    682\u001b[0m           \u001b[0;32mreturn\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    683\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msample_weights\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 684\u001b[0;31m       \u001b[0;32mreturn\u001b[0m \u001b[0mdataset\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmap\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmap_fn\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnum_parallel_calls\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdataset_ops\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mAUTOTUNE\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    685\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    686\u001b[0m   \u001b[0;32mif\u001b[0m \u001b[0mmode\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0mModeKeys\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mPREDICT\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/data/ops/dataset_ops.py\u001b[0m in \u001b[0;36mmap\u001b[0;34m(self, map_func, num_parallel_calls)\u001b[0m\n\u001b[1;32m   1589\u001b[0m     \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1590\u001b[0m       return ParallelMapDataset(\n\u001b[0;32m-> 1591\u001b[0;31m           self, map_func, num_parallel_calls, preserve_cardinality=True)\n\u001b[0m\u001b[1;32m   1592\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1593\u001b[0m   \u001b[0;32mdef\u001b[0m \u001b[0mflat_map\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmap_func\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/data/ops/dataset_ops.py\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, input_dataset, map_func, num_parallel_calls, use_inter_op_parallelism, preserve_cardinality, use_legacy_function)\u001b[0m\n\u001b[1;32m   3924\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_transformation_name\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3925\u001b[0m         \u001b[0mdataset\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0minput_dataset\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 3926\u001b[0;31m         use_legacy_function=use_legacy_function)\n\u001b[0m\u001b[1;32m   3927\u001b[0m     self._num_parallel_calls = ops.convert_to_tensor(\n\u001b[1;32m   3928\u001b[0m         num_parallel_calls, dtype=dtypes.int32, name=\"num_parallel_calls\")\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/data/ops/dataset_ops.py\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, func, transformation_name, dataset, input_classes, input_shapes, input_types, input_structure, add_to_graph, use_legacy_function, defun_kwargs)\u001b[0m\n\u001b[1;32m   3145\u001b[0m       \u001b[0;32mwith\u001b[0m \u001b[0mtracking\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mresource_tracker_scope\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mresource_tracker\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3146\u001b[0m         \u001b[0;31m# TODO(b/141462134): Switch to using garbage collection.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 3147\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_function\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mwrapper_fn\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_get_concrete_function_internal\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   3148\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3149\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0madd_to_graph\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/eager/function.py\u001b[0m in \u001b[0;36m_get_concrete_function_internal\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   2393\u001b[0m     \u001b[0;34m\"\"\"Bypasses error checking when getting a graph function.\"\"\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2394\u001b[0m     graph_function = self._get_concrete_function_internal_garbage_collected(\n\u001b[0;32m-> 2395\u001b[0;31m         *args, **kwargs)\n\u001b[0m\u001b[1;32m   2396\u001b[0m     \u001b[0;31m# We're returning this concrete function to someone, and they may keep a\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2397\u001b[0m     \u001b[0;31m# reference to the FuncGraph without keeping a reference to the\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/eager/function.py\u001b[0m in \u001b[0;36m_get_concrete_function_internal_garbage_collected\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   2387\u001b[0m       \u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2388\u001b[0m     \u001b[0;32mwith\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_lock\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2389\u001b[0;31m       \u001b[0mgraph_function\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0m_\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0m_\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_maybe_define_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   2390\u001b[0m     \u001b[0;32mreturn\u001b[0m \u001b[0mgraph_function\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2391\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/eager/function.py\u001b[0m in \u001b[0;36m_maybe_define_function\u001b[0;34m(self, args, kwargs)\u001b[0m\n\u001b[1;32m   2701\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2702\u001b[0m       \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_function_cache\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmissed\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0madd\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcall_context_key\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2703\u001b[0;31m       \u001b[0mgraph_function\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_create_graph_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   2704\u001b[0m       \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_function_cache\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprimary\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mcache_key\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mgraph_function\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2705\u001b[0m       \u001b[0;32mreturn\u001b[0m \u001b[0mgraph_function\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/eager/function.py\u001b[0m in \u001b[0;36m_create_graph_function\u001b[0;34m(self, args, kwargs, override_flat_arg_shapes)\u001b[0m\n\u001b[1;32m   2591\u001b[0m             \u001b[0marg_names\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0marg_names\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2592\u001b[0m             \u001b[0moverride_flat_arg_shapes\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0moverride_flat_arg_shapes\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2593\u001b[0;31m             capture_by_value=self._capture_by_value),\n\u001b[0m\u001b[1;32m   2594\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_function_attributes\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2595\u001b[0m         \u001b[0;31m# Tell the ConcreteFunction to clean up its graph once it goes out of\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/framework/func_graph.py\u001b[0m in \u001b[0;36mfunc_graph_from_py_func\u001b[0;34m(name, python_func, args, kwargs, signature, func_graph, autograph, autograph_options, add_control_dependencies, arg_names, op_return_value, collections, capture_by_value, override_flat_arg_shapes)\u001b[0m\n\u001b[1;32m    976\u001b[0m                                           converted_func)\n\u001b[1;32m    977\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 978\u001b[0;31m       \u001b[0mfunc_outputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpython_func\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mfunc_args\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mfunc_kwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    979\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    980\u001b[0m       \u001b[0;31m# invariant: `func_outputs` contains only Tensors, CompositeTensors,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/data/ops/dataset_ops.py\u001b[0m in \u001b[0;36mwrapper_fn\u001b[0;34m(*args)\u001b[0m\n\u001b[1;32m   3138\u001b[0m           attributes=defun_kwargs)\n\u001b[1;32m   3139\u001b[0m       \u001b[0;32mdef\u001b[0m \u001b[0mwrapper_fn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m  \u001b[0;31m# pylint: disable=missing-docstring\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 3140\u001b[0;31m         \u001b[0mret\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_wrapper_helper\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   3141\u001b[0m         \u001b[0mret\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mstructure\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto_tensor_list\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_output_structure\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mret\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3142\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mops\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconvert_to_tensor\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mt\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mt\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mret\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/data/ops/dataset_ops.py\u001b[0m in \u001b[0;36m_wrapper_helper\u001b[0;34m(*args)\u001b[0m\n\u001b[1;32m   3080\u001b[0m         \u001b[0mnested_args\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mnested_args\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3081\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 3082\u001b[0;31m       \u001b[0mret\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mautograph\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtf_convert\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfunc\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mag_ctx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mnested_args\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   3083\u001b[0m       \u001b[0;31m# If `func` returns a list of tensors, `nest.flatten()` and\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3084\u001b[0m       \u001b[0;31m# `ops.convert_to_tensor()` would conspire to attempt to stack\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/tensorflow-2.1.0/python3.6/tensorflow_core/python/autograph/impl/api.py\u001b[0m in \u001b[0;36mwrapper\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m    235\u001b[0m       \u001b[0;32mexcept\u001b[0m \u001b[0mException\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m  \u001b[0;31m# pylint:disable=broad-except\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    236\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mhasattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0me\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'ag_error_metadata'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 237\u001b[0;31m           \u001b[0;32mraise\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mag_error_metadata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto_exception\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0me\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    238\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    239\u001b[0m           \u001b[0;32mraise\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mValueError\u001b[0m: in converted code:\n\n    /tensorflow-2.1.0/python3.6/tensorflow_core/python/keras/engine/training_v2.py:677 map_fn\n        batch_size=None)\n    /tensorflow-2.1.0/python3.6/tensorflow_core/python/keras/engine/training.py:2410 _standardize_tensors\n        exception_prefix='input')\n    /tensorflow-2.1.0/python3.6/tensorflow_core/python/keras/engine/training_utils.py:582 standardize_input_data\n        str(data_shape))\n\n    ValueError: Error when checking input: expected lstm_input to have shape (1, 10) but got array with shape (None, 1)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GaC6NNMRp0lb"
      },
      "source": [
        "rnn_forecast = model_forecast(model, series[..., np.newaxis], window_size)\n",
        "rnn_forecast = rnn_forecast[split_time - window_size:-1, -1, 0]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PrktQX3hKYex",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 388
        },
        "outputId": "8f8a14a8-1725-4ebf-a53b-4cb0a65a4199"
      },
      "source": [
        "plt.figure(figsize=(10, 6))\n",
        "plot_series(time_valid, x_valid)\n",
        "plot_series(time_valid, rnn_forecast)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAncAAAFzCAYAAABcsy/pAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQgghhIgjEtwJIYQQQsSR/w9P5uMFtXP1xQAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 720x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "13XrorC5wQoE",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "392b1fbb-cc0c-4c3f-b246-73de03c381dc"
      },
      "source": [
        "tf.keras.metrics.mean_absolute_error(x_valid, rnn_forecast).numpy()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "15.40074"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 41
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MD2kyYUVt3O0",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 609
        },
        "outputId": "4d4ce37c-ebd7-43cc-d0f3-761c37774bda"
      },
      "source": [
        "import matplotlib.image  as mpimg\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "#-----------------------------------------------------------\n",
        "# Retrieve a list of list results on training and test data\n",
        "# sets for each training epoch\n",
        "#-----------------------------------------------------------\n",
        "loss=history.history['loss']\n",
        "\n",
        "epochs=range(len(loss)) # Get number of epochs\n",
        "\n",
        "\n",
        "#------------------------------------------------\n",
        "# Plot training and validation loss per epoch\n",
        "#------------------------------------------------\n",
        "plt.plot(epochs, loss, 'r')\n",
        "plt.title('Training loss')\n",
        "plt.xlabel(\"Epochs\")\n",
        "plt.ylabel(\"Loss\")\n",
        "plt.legend([\"Loss\"])\n",
        "\n",
        "plt.figure()\n",
        "\n",
        "\n",
        "\n",
        "zoomed_loss = loss[200:]\n",
        "zoomed_epochs = range(200,500)\n",
        "\n",
        "\n",
        "#------------------------------------------------\n",
        "# Plot training and validation loss per epoch\n",
        "#------------------------------------------------\n",
        "plt.plot(zoomed_epochs, zoomed_loss, 'r')\n",
        "plt.title('Training loss')\n",
        "plt.xlabel(\"Epochs\")\n",
        "plt.ylabel(\"Loss\")\n",
        "plt.legend([\"Loss\"])\n",
        "\n",
        "plt.figure()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 42
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEWCAYAAABrDZDcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nZlZwDgIzs4JzEJiZFZyDwMys4P4/yTzCS0vyvfoAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYwAAAEWCAYAAAB1xKBvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzt3Xd4VGXax/HvHXoTkKYSJIgNRYga\nEcWC3cW1oK69rIq4viqWxS6666prWextBREV7JW1916jIqLYRQUpAaQq/X7/eM44k5AySWYyM8nv\nc11zzTlnzpm5j4Nz5+nm7oiIiFQlL9MBiIhIblDCEBGRpChhiIhIUpQwREQkKUoYIiKSFCUMERFJ\nihKGSCXMrJGZLTaz9VN5bg3iuMzMxqb6fUWqo3GmAxBJJTNbnLDbElgGrIr2T3L38dV5P3dfBbRO\n9bkiuUgJQ+oVd//jB9vMpgJD3P2lis43s8buvrIuYhPJdaqSkgYlqtp50MzuN7NFwFFmtp2ZvWdm\n881shpndaGZNovMbm5mbWUG0Py56/VkzW2Rm75pZj+qeG73+JzP72swWmNlNZva2mf01yfsYbGaf\nRzG/YmabJLx2gZn9YmYLzexLMxsYHe9vZh9Hx2eZ2TUp+E8qDYgShjREg4H7gLbAg8BK4HSgIzAA\n2Bs4qZLrjwBGAGsDPwH/qu65ZtYZeAg4O/rcH4B+yQRvZr2Ae4HTgE7AS8AEM2tiZptHsW/l7msB\nf4o+F+Am4Jro+IbAI8l8nkiMEoY0RG+5+//cfbW7/+7uH7r7++6+0t2/B+4Adq7k+kfcvdjdVwDj\ngcIanPtnYKK7Pxm9dh0wJ8n4DwMmuPsr0bVXEpLftoTk1xzYPKpu+yG6J4AVwEZm1sHdF7n7+0l+\nngighCEN08+JO2a2qZk9bWYzzWwhcCnhr/6KzEzY/o3KG7orOne9xDg8zAI6LYnYY9f+mHDt6uja\nru7+FfB3wj3Mjqre1olOPQ7YDPjKzD4ws0FJfp4IoIQhDVPZKZr/C0wGNoyqay4GLM0xzADyYztm\nZkDXJK/9BeiecG1e9F7TAdx9nLsPAHoAjYB/R8e/cvfDgM7ASOBRM2te+1uRhkIJQwTaAAuAJVH7\nQGXtF6nyFLCVme1rZo0JbSidkrz2IWA/MxsYNc6fDSwC3jezXma2i5k1A36PHqsBzOxoM+sYlUgW\nEBLn6tTeltRnShgioQrnWMKP7n8JDeFp5e6zgEOBa4G5QE/gE8K4kaqu/ZwQ721ACaGRfr+oPaMZ\ncDWhPWQm0B64MLp0EDAl6h32H+BQd1+ewtuSes60gJJI5plZI0JV08Hu/mam4xEpj0oYIhliZnub\nWbuo+mgEoRfTBxkOS6RCShgimbMD8D2hWmkvYLC7V1klJZIpqpISEZGkqIQhIiJJqVeTD3bs2NEL\nCgoyHYaISM746KOP5rh7Ul2661XCKCgooLi4ONNhiIjkDDP7seqzAlVJiYhIUpQwREQkKUoYIiKS\nlHrVhiEiUhMrVqxg2rRpLF26NNOhpE3z5s3Jz8+nSZMmNX4PJQwRafCmTZtGmzZtKCgoIEwcXL+4\nO3PnzmXatGn06NGj6gsqoCopEWnwli5dSocOHeplsgAwMzp06FDrEpQShogI1NtkEZOK+1PC+P13\nGDkSXnst05GIiGQ1tWE0bhwSxlZbQZMmsP760K1bpqMSkQamdevWLF68ONNhVCptJQwzG2Nms81s\ncsKxf5jZdDObGD3KXVPYzM40s8/NbHK0JnH6lpFs0gROOAGeeQZ23hm22Qa++CJtHycikqvSWSU1\nlrASWFnXuXth9Him7Itm1hUYBhS5e2/CmsSHpTFOGDIkPOdHSywffDDUp+51jz4KU6dmOgoRqaap\nU6ey66670qdPH3bbbTd++uknAB5++GF69+5N37592WmnnQD4/PPP6devH4WFhfTp04dvvvkm5fGk\nrUrK3d8ws4IaXt4YaGFmK4CWhJXI0qd7d3jySdh0U/j+e9h7bzjxRBg+HHr0gLXWSuvHp5U7HHYY\nnHUWXHVVpqMRyX5nnAETJ6b2PQsL4frrq33ZaaedxrHHHsuxxx7LmDFjGDZsGE888QSXXnopzz//\nPF27dmX+/PkA3H777Zx++ukceeSRLF++nFWrVqX2HshMo/epZjYpqrJqX/ZFd59OWG/4J2AGsMDd\nX6jozcxsqJkVm1lxSUlJzaPad1/YaCPYay+49FIYNy58yUcdVfP3zAZLlsDKlbBoUaYjEZFqevfd\ndzniiCMAOProo3nrrbcAGDBgAH/9618ZNWrUH4lhu+2244orruCqq67ixx9/pEWLFqkPyN3T9gAK\ngMkJ+10IVUx5wOXAmHKuaQ+8AnQCmgBPAEcl83lbb721p8xbb7kfeaS7mfs997h/8UXq3rsuTZvm\nDu7HHJPpSESy1hdZ8P93q1at1jjWoUMHX758ubu7L1++3Dt06PDHa++9956PGDHCu3fv7nPmzHF3\n92+//dZvuOEG33DDDf3ll19e4/3Ku0+g2JP8Ta/TEoa7z3L3Ve6+GhgF9CvntN2BH9y9xN1XAI8B\n29dlnAAMGBCqcPLy4Jhj4KCDQvXO6tV1HkqtLFgQnpcsyWwcIlJt22+/PQ888AAA48ePZ8cddwTg\nu+++Y9ttt+XSSy+lU6dO/Pzzz3z//fdssMEGDBs2jP33359JkyalPJ46TRhmtm7C7mBgcjmn/QT0\nN7OWFkaa7AZMqYv41tC1K1x7LRxyCEyZAuutB8cdl5FQamzhwvCshCGS1X777Tfy8/P/eFx77bXc\ndNNN3HXXXfTp04d7772XG264AYCzzz6bLbbYgt69e7P99tvTt29fHnroIXr37k1hYSGTJ0/mmGOO\nSXmMaVvT28zuBwYCHYFZwCXRfiHgwFTgJHefYWbrAaPdfVB07T+BQ4GVwCfAEHdfVtVnFhUVeVoW\nUFq2DAoKYObMsP/ii9C6NfTvn/rPSrXnnw+N+DvsAG++meloRLLSlClT6NWrV6bDSLvy7tPMPnL3\nomSuT2cvqcPLOXxnBef+AgxK2L+EkGCyQ7Nm8Mor4Qf3pJNgzz1h443hyy8zHVnVVMIQkRTR1CDJ\n6tUrVEe1bRvaMr76CsaOhUceyXRklVMbhoikiBJGdTRpAiNGwJFHhv3jjgujxH//PbNxVUYJQyQp\n6aqezxapuD8ljOr6+99DyaJ167C/cGEYt/HKKxkNq0KqkhKpUvPmzZk7d269TRoerYfRvHntZlnS\n5IM10bgxDBoUGsG//x6uvBKuuw4mT4YNN8x0dKXFShhZPqmZSCbl5+czbdo0ajX4N8vFVtyrDSWM\nmrrvvjAm4+mn4aOP4IYbQlXVffdBz56Zji4uljBWroTly6Fp08zGI5KFmjRpUquV6BoKVUnVVKNG\noU3jgAPgX/+CO+8Ms9xutBEcfXRoGM8GsSopULWUiNSKEkaq/OUvoZvtKaeEeahuvTX8RZ8Kv/8O\n558Pv/1W/WtjJQxQwhCRWlHCSKWuXUPV1A47wKmnQt++8R/5hx6CXXcNJY/58+G555J/37ffDu0k\nb7xR/ZhUwhCRFFEbRqrl5YV2jbFj4fTTYbvtoHNnaNUKXn0VZsyAO+4IPavmzoX2a0zYu6ba9HRa\nsACaNw/re6jhW0RqQSWMdFhrLRg2LMxBNWkSvPRSWNEPwoC/jz4KJY1fklzmo6qEsWRJmOfqhTKz\nwI8fD19/HV6r7HoRkSQoYaTTPffAhx+G7RUrwvPXX8PHH4ftWMKYPx/GjKm4obyqhDFrVii5TE6Y\ny3HlyvhaHmuvXfn1IiJJUMJIp2bNoKgoLMQU8+ab8URx992w2WZw221hxPi338bPW7AAzj03THwY\nSxgVVSnFEkFie0ViA3lsZLoShojUghJGXTj00FAttPnm8OCD8eMPPRSmTY/NR5VYRfXcc3D11fD+\n+2uWMF5/HRKXX4wdT1xVLzZdya23wn77he2//CUsPZuYWEREkqSEURfOOQe++y6MAl+5MjSMN2sW\nr6YqW0WVuD1nTumEMXkyDBwYpi2PqayE0aJFaHCPGT06rFUuIlJNShh1IS8v9FTaYouwf911UN6o\n0u++g8ceC20ZM2aEY2UTxs8/h+3Zs+PXlZcwYiWMli1LJ4wdd4SHH07dGBERaTCUMOrSOedAcXHo\nQRXruZRoxIiwFOxDD1VcwoglivLGV8SqpJYtK13CaNkyfu6554ZG9pdeSt19iUiDoIRRl9q0ga23\nDtuxhFHe7JHjx8cTRklJ6QkEYwkjsb0isYSxbFkYQDh6dDjWokUo4QBsuy3ssQe0axc+Q0SkGpQw\nMmXdaHnzvfYKzx07xl975hn45JOwXbaEEZtNs6ISxrx5YUBgbAH4WOli2jR47bUw+eCRR8Kjj8Kv\nv6b8tkSk/lLCyJTNNw9tC//+N1x+OeyyS/z4qlWh2giqVyW1cGH8uth5LVqE565d46WZIUNCSeTe\ne9NzbyJSLylhZMpRR8GPP4alXy+4IPygAxxxROnzSkqqThix9oqFC+PVV7GSSCxhJCoshAEDwvxU\nmi5ERJKkhJEpjRpBhw7x/VibRmEhdOsWtps0WTNhJFZJLVoUxlgUF4djixbFSxixaxIbvBNdc03o\niTVyZOruSUTqtbQlDDMbY2azzWxywrF/mNl0M5sYPQZVcG07M3vEzL40sylmtl264swaW24Zqqi2\n3jre5bZbt9D2sHJl2E9s9F64MExw+L//xcdkrFhRurstlF/CgDAp4r77pnYadhGp19JZwhgL7F3O\n8evcvTB6PFPBtTcAz7n7pkBfYEqaYsweu+8eGqG7dIF//jMc23vvsKofhGVhE6ukFi2C995b831i\n4zRiKkoYACefHN7viSdqH7+I1HtpSxju/gYwr7rXmVlbYCfgzuh9lrv7/BSHl52aNAnPAweGwXuJ\nc1Ctu25on1i6NOz/+iu8+OKa7/HTT6X3K0sYe+0VlpMdPjyUZEREKpGJNoxTzWxSVGVV3mIQPYAS\n4C4z+8TMRptZq3LOA8DMhppZsZkV17sF3Dt1im/HuuFCaJeYOjXenpEosYTRtGloK6lIXl4Y9T1/\nPhx+ePYsKysiWamuE8ZtQE+gEJgBlNfi2hjYCrjN3bcElgDnVfSG7n6Huxe5e1GnxB/Y+mDgQGjd\nOmwnTiWy8cbx7cbRGlixxJCYMCorXcRsuSVcey289VbpiRFFRMqo04Th7rPcfZW7rwZGAf3KOW0a\nMM3d34/2HyEkkIanXbtQinj1Vdhtt/jxnXeOb/eL/hN27hyeE6uWKuohVdZxx4XqrwsvjDewi4iU\nUacJw8wS6lUYDEwue467zwR+NrNNokO7AV/UQXjZqXnz0iUNCO0OMZtuGp5jVVbzE5p7kilhQCid\nXHIJfP893H9/rcIVkforbWt6m9n9wECgo5lNAy4BBppZIeDAVOCk6Nz1gNHuHutmexow3syaAt8D\nx6UrzpwRm3F2gw3CErCxY7FquHXWWfOaZBMGhPEchYVw6qkhIW2/fe3iFZF6J20Jw90PL+fwnRWc\n+wswKGF/IlCUptByU6x31GabhUkMIcw/FUsYPXqEUkjiyO3qJIy8PJgwIVR97bEHPPVUfLoSERE0\n0jt39OkTns84I54wOnSIT1q41lqwzTZh2yw8J9uGEdOtG7zxBnTvDkcfHU9SIiIoYeSOTTcNg/h2\n2y0+MrtDh3gJo1Ur6N8/bMeqrKpTwohZZx24+WaYPj1MH6JR4CISUcLIJbGSQ1FR6E570UXxEkar\nVvESRmwCwpokDAhVUbvvDhdfHNo1fvihdnGLSL2ghJGLunQJ80bttFO8d1T79vEutrHEUd0qqRiz\nMEfVgw/CzJmh5PLBB7WPW0RymhJGruvWLUwRcsghYYr0b76BG24Ir9W0hAGhO+8hh8A774TSy8CB\nMGpUWKAp0eTJcMstGr8h0gAoYdQHu+8eTw4bbghrrx22a5MwYjbdFN59NzS6Dx0aktI554QE4Q7H\nHhu64h50UO0/S0SyWtq61UoGxQb51bRKqqwuXeDtt+HDD+H220Nj+CefhO63H38cksmECWFp2MQ1\nPkSkXlEJoz6KJYxUlDBiGjUKbRljx4aqqfffh3PPhS22CCv3AUycmLrPE5Gso4RRH7VpAwceWHrO\nqVQaMgS++y40hH/8cbyR/ZNP0vN5IpIVVCVVH+XlwaOPpvczOnWKjwHp2BHy85UwROo5lTAkNbbc\nUglDpJ5TwpDU2Hpr+PLLeLfbJ54IbR4LF2Y2LhFJGSUMSY099gjdbF96KXS5HT48NIxfd12mIxOR\nFFHCkNTo1y8s+PTcc3DffaFRvEcPGDlSpQyRekIJQ1KjceMwgPCZZ+Bf/4K+feHuu2HRInj66UxH\nJyIpoIQhqXPKKfDrr/Dtt2HiwgEDwlxXDz8cXl+1SmM1RHKYEoakzsCBoUrqwgvhgANC996DDoJn\nnw1Lx95xR+hN9cYbmY5URGpACUNSa5dd4LLLQrKAMMhv6VK49loYPz4cu+KKzMUnIjWmhCHp1bdv\nKGX8619hPqoePeD55+GttzIdmYhUkxKGpN9//hMSR7NmYZLC/Pwww+3y5WEVwS++CF1yRSSrKWFI\n+hUUhFHgs2ZB795hvY5PP4V99glTpm++OYwZk+koRaQKaUsYZjbGzGab2eSEY/8ws+lmNjF6DKrk\n+kZm9omZPZWuGKUOmUHbtmH7wAPhzjvhzTfDc9u2cPbZYeyGiGStdJYwxgJ7l3P8OncvjB7PVHL9\n6cCUtEQmmXf88fDLL6FX1bvvhoTSrx+8+mqmIxORCqQtYbj7G8C8Kk8sh5nlA/sAo1MalGSXtdeG\nvfaCXr3CVOldusCee4ZFmkQk62SiDeNUM5sUVVm1r+Cc64FzgNVVvZmZDTWzYjMrLikpSWmgUod6\n9gwljT33hJNPhhNPhGXLMh2ViCSo64RxG9ATKARmACPLnmBmfwZmu/tHybyhu9/h7kXuXtQptj6D\n5Ka2bUMvqgsugNGjQ1vH+efDFNVMimSDOk0Y7j7L3Ve5+2pgFNCvnNMGAPuZ2VTgAWBXMxtXh2FK\nJjVqBJdfDtdfH+aluvJK2G67sJ64iGRUnSYMM1s3YXcwMLnsOe5+vrvnu3sBcBjwirsfVUchSrY4\n/fQwPuOrr6B9exg8GG66Ce69V7PfimRI2pZoNbP7gYFARzObBlwCDDSzQsCBqcBJ0bnrAaPdvcJu\nttIA9eoVnh9/HHbdFYYNC/tbbQUvvBB6VO20E3TunLkYRRoQ83o0wraoqMiLi4szHYakw/LloWTx\n5ptw+OHQpg3MmRMmOXz88UxHJ5KzzOwjdy9K5lyN9Jbc0LQpdOwYqqb+9z/47TfYYIOwFOwHH8Dk\nybBkSaajFKnXlDAk9+yxRyhdfPxxGLux006wxRZwwgmZjkykXlPCkNzUokXohvvBB2FOqp12ggcf\nDFVWIpIWShiS29ZfHx59NIzfWGedkDhOOw1Wrsx0ZCL1jhKG1A9t24YZcIcNg5tvhvXWC4s3ff11\npiMTqTeUMKT+6Nw5TJ0+YUKYYuSBB6B//1BtJSK1poQh9c+++8K4caHnVLt2cPDBsGBBpqMSyXlK\nGFJ/FRTAfffB9OlhfioRqRUlDKnf+vcPs9/+97/wzTfx4199FRrHr7kmc7GJ5Ji0TQ0ikjVGjICx\nY+Gss0L7RklJ6E01Z05YU7x9+9BALiKVUglD6r8uXeCyy+Cpp+Cf/4QjjghtGh9/HBLHiBGwalWm\noxTJekoY0jCcfjocemhIGK+8AnfcAX37wv/9H8ycCW+/nekIRbKeEoY0DGZw//1w551hssJjjgnH\n99kHmjeHhx+OnztyZHh9dZULPoo0KGrDkIbDDI4/vvSx1q1DN9zx4+Hf/4bXXoPhw8Nr228Pf/tb\nnYcpkq1UwhA56yz49Vc46ig46CAoLISBA8PysBq/IfIHJQyR/v3DAk1PPhlKFS+/DP/5D8yfH1b5\nExFACUMkePxx+PHHsIrf2mvD1lvDn/8M114LixZlOjqRrKCEIQKw1lph5ttEl1wSqqpuvjkzMYlk\nGSUMkYoUFcGgQXDRRbDllrDddqErbvv2cMUVUI+WNxZJhhKGSGXuvhvOPRfWXTd0v23dOrRzXHgh\nXH55pqMTqVPqVitSmY4dQ2kikXvoUTViBHz+OYwZE1YAFKnnkkoYZtYTmObuy8xsINAHuMfd51dy\nzRjgz8Bsd+8dHfsHcCJQEp12gbs/U+a6bsA9QBfAgTvc/Ybq3JRIWpmFuak23TS0c/z4I2y7bZhe\nZJdd4IADwjki9UyyVVKPAqvMbEPgDqAbcF8V14wF9i7n+HXuXhg9ninn9ZXA3919M6A/cIqZbZZk\nnCJ1o0mTUMIYMwZ++glGjw7bBx4Yph0RqYeSTRir3X0lMBi4yd3PBtat7AJ3fwOYV92A3H2Gu38c\nbS8CpgBdq/s+InXir3+FadNC19sFC2CvvcK8VV99lenIRFIu2YSxwswOB44FnoqONanhZ55qZpPM\nbIyZta/sRDMrALYE3q/knKFmVmxmxSUlJRWdJpJ+jRqFRvJmzeCMM0Jbx/TpmY5KJGWSTRjHAdsB\nl7v7D2bWA7i3Bp93G9ATKARmACMrOtHMWhOqws5w94UVnefud7h7kbsXderUqQYhiaRQly5hRtzn\nnoPu3SE/P0xkeMst8OijsHJlpiMUqbGkGr3d/QtgGEBUKmjj7ldV98PcfVZs28xGES+tlGJmTQjJ\nYry7P1bdzxHJqGHDwvN//xt6U913H9wb/X3197+HaUdEclBSJQwze83M1jKztYGPgVFmdm11P8zM\nEts9BgOTyznHgDuBKe5e7c8Qybi8vFAlNWVKSBSLFsGsWWGp2JEj4QZ1+pPclGyVVNuoWuhAQnfa\nbYHdK7vAzO4H3gU2MbNpZnYCcLWZfWZmk4BdgDOjc9czs1iPqQHA0cCuZjYxegyq/q2JZIkWLaBz\nZ7j+ehg8OCSTxPU3RHKEeRLTG5jZZ8CewN3Ahe7+oZlNcvc+6Q6wOoqKiry4uDjTYYhUbOVK6Ncv\nrCs+ZUoYOS6SQWb2kbsXJXNusiWMS4Hnge+iZLEB8E1NAxRpsBo3DpMZTp8eGsNXrQpLxu64I+y3\nHzzzjOaokqyVVAkjV6iEITnjhhtC1dRmm8EXX0BBQUgeP/8Mt94a2jtE6kDKSxhmlm9mj5vZ7Ojx\nqJnl1y5MkQbs9NNh1Cj45ZfQq+rzz+Hbb8NCThddFKZVF8kyyVZJ3QVMANaLHv+LjolITQ0ZAvPm\nhdJGy5bQtGlYsOnXX+G88zIdncgakk0Yndz9LndfGT3GAholJ1JbZScp7NsXhg8P81Hdey98+CFc\ndhnssQecdlookYhkSLLTm881s6OA+6P9w4G56QlJpIG79FJ4993QKB7Tuze88Qa89VZ4rXnzzMUn\nDVayJYzjgUOAmYQpPQ4G/pqmmEQatubN4eWXw3Qi48aFQX+ffQaPPAITJ8LFF2c6QmmgatxLyszO\ncPfrUxxPraiXlNR7Q4bAPfeEBLLJJpmORuqBdIzDKM9ZtbhWRGriiitCA/nBB0NxMUydGrrjitSB\n2iQMLSkmUtc6d4bHHgtdcLfZBnr0gC22CO0aImlWmzW968+IP5FcsuuuMHkyTJoEM2bANdfAnnuG\n47/8AoMGwd/+FtblWHvtTEcr9UilbRhmtojyE4MBLdy9Ngkn5dSGIQ3S9Omwww7w22+hXePNN8Px\nvDw46CDo2TMs7nTJJWFpWZEE1WnDqPQH393bpCYkEUmbrl3DRIaNGoWE8OCD8OOPMHduGBS4fHmY\nn+qBB8KSshdeuOb4D5EkZFUJQURqKHFcxqGHxreHDg0JY9IkuP12GDEC2rQJU5OIVJMShkh91rNn\neO7VC/7yFzjwwDDp4ZIlcP75KmlItdSml5SI5JK8PHjoITjyyFAtddhh8P77mY5KcogShkhD0rRp\nGPh38cXw1FPQvz+ccILW4JCkKGGINDR5efDPf4YpR047DcaMCYs4iVRBCUOkoWrdGq6+GvLzw7iN\nUaNg9WpYvDjTkUmWUsIQaciaN4f//jeUOoYODY3knTuHKdafeirT0UmWUcIQaegGDQrjOE48MfSe\n2mUXGDkS9t03rMkhEklbwjCzMdFyrpMTjv3DzKab2cToMaiCa/c2s6/M7Fsz09JjIumWlxcWbZo5\nE55+OowaHzgwlDp+/DHT0UmWSOc4jLHAzcA9ZY5f5+7/qegiM2sE3ALsAUwDPjSzCe7+RboCFZFI\nXvQ3ZIsWoTfVxhuHksfuu4cBf926wZ/+FEaVS4OTtoTh7m+YWUENLu0HfOvu3wOY2QPA/oAShkhd\n6tYNzjknrAD44ovx40VF8Oyz0LFj5mKTjMhEG8apZjYpqrJqX87rXYGfE/anRcdEpK5dfDF89x0s\nXBhmwr333rB401/+AitWZDo6qWN1nTBuA3oChYSlXkfW9g3NbKiZFZtZcUlJSW3fTkQSNWoEG2wQ\nqqPWXReOOgpGj4bXXoMzz8x0dFLH6jRhuPssd1/l7quBUYTqp7KmA90S9vOjYxW95x3uXuTuRZ06\ndUptwCKypqOOCt1ub7kl9KaSBqNOE4aZrZuwOxiYXM5pHwIbmVkPM2sKHAZMqIv4RCRJV14ZqqWG\nD4d+/eDOOzXgrwFIZ7fa+4F3gU3MbJqZnQBcbWafmdkkYBfgzOjc9czsGQB3XwmcCjwPTAEecvfP\n0xWniNRAo0YwbhzceGPogjtkSBgxrsF+9VqlK+7lGq24J5IB7vDOO2GNjYkTwxQjxx2X6agkSdVZ\ncU8jvUWkdsxgwAB49dWwrvjxx0P37mHKkZ9+Kn3u3Lmh0fyrrzITq9SKFlASkdRo0yZUSY0ZA+PH\nhwkNIbR1FBaGEeTFxWEFwMaNQ4lk+HBYZ53Mxi1JU5WUiKSeO7z+Orz8Mlx1VRizUVgIe+wB++0H\nY8eGxAKhCuuEEzIabkNWnSopJQwRSa/ffgulinbtSh//8sswV9Wnn4a2jx49MhNfA6c2DBHJHi1b\nrpksADbdNLRnLFsWBgeecopGj2c5JQwRyZyNN4ZJk+DUU+HWW2H77eHrrzMdlVRACUNEMmvjjeGm\nm+Dhh+GHH0LS+OCDTEcl5VBsA+z2AAATEElEQVTCEJHscPDB8N570LZt6J6rdcazjhKGiGSPDTeE\nt9+GggI46CD49ttMRyQJlDBEJLussw5MmBAWc9prrzC9umQFJQwRyT4bbBAWaSopCaWOiy7KdESC\nEoaIZKt+/cJiTYcdBpdfHgYCSkYpYYhI9urePYzVWH/9sFSsZJQShohkt1atwqJNr78eJi+UjFHC\nEJHsN3gwrFoF//tfpiNp0JQwRCT7bb01dOsGjz+e6UjS46ij4LrrMh1FlZQwRCT7mYVSxgsv1M+l\nYF9/PSxCleWUMEQkNwweDEuXwnPPZTqS1Fu6NMzqm+WUMEQkN+ywA3TsCI89lulIUm/ZMiUMEZGU\nadwYDjggNHz//numo0mtpUthyZJMR1ElJQwRyR2HHBLaMJ5/PtORpM7q1WEdkIZewjCzMWY228wm\nl/Pa383MzaxjBddebWafm9kUM7vRzCydsYpIDthlF+jQIUyFXl8sWxaeExPG4YfDaadlJp5KpLuE\nMRbYu+xBM+sG7An8VN5FZrY9MADoA/QGtgF2TluUIpIbGjcOjd8TJtSfaqmlS8NzYpXUF1/AlCmZ\niacSaU0Y7v4GMK+cl64DzgEqWlDcgeZAU6AZ0ASYlY4YRSTH1LdqqfJKGMuWxRNJFqnzNgwz2x+Y\n7u6fVnSOu78LvArMiB7Pu3v2pVsRqXuxaqmxYzMdSWrEEsNvv4FHf0MvWxZPJFmkThOGmbUELgAu\nruK8DYFeQD7QFdjVzHas4NyhZlZsZsUlJSWpDllEsk3jxjBsGDz5ZHhU1333hZlw3WHWLLj//tTH\nWB2xxLB6dXx76VKVMICeQA/gUzObSkgIH5vZOmXOGwy85+6L3X0x8CywXXlv6O53uHuRuxd16tQp\njaGLSNY47zzo1QuuuKL61778Mnz4ISxYALfdBkccAfPnpz7GZCUmhli1lEoY4O6fuXtndy9w9wJg\nGrCVu88sc+pPwM5m1tjMmhAavFUlJSJB06aw337w8cfVb/z+4YfwPGsWfPNN2J4zJ7XxVUdiYog1\nfDfEhGFm9wPvApuY2TQzO6GSc4vMbHS0+wjwHfAZ8CnwqbtrmkoRidt+e1i5Ej76qHrXxRLGzJnx\nhDGvvL45daSiEkYWVkk1Tuebu/vhVbxekLBdDAyJtlcBJ6UzNhHJcf37h+d33w3ThiRjxQr4KerN\nn5gw5s4Nk//95z8wfjy0aJH6eCtSNmGsXBmmcm9oJQwRkbTp3Bl69gxThaxYUfF5sZHUAD//HPYh\njHWItV3MmAHHHRemT3/1VfjlFzj33LoZfV22Siqx4TvLKGGISO4680x480048cTSx++8M94gfuaZ\n0LYtnHUWfP99/Jy3345v33ILfP015OXBs8+G/auvhhEjwl/8Tz4ZTzSpVraEEUsYy5bFu9lmibRW\nSYmIpNUpp4Qf+ltvDT/wnTuHZDFkSFhDY8CA8Fp+fligKDZ6ulGjkGhiPvkEmjSB3XYLCWPTTcPx\nm2+Gdu3g4ovh9tvhpDTUlCeWMBITBsDy5dCsWeo/s4ZUwhCR3DZ0aCgF3H9/qHq6+GLYYovw1/l+\n+4VE8Oab0KdPWEujWTPYbLPwY9y0KbRuHc7t1g323Re++w5efBF69w7nxOatuvFG2HvvkHQOPjhU\nY6VCYgljyZLS+1nWjqGEISK5bfPNoagILr0UTj45tD9cdlkoXSxcGEoG+fnw4INw5ZXw0kvQtWu4\nduedoUuXsL3++nDMMWF/+fLwXgCffRaev/giTEcyYgQ8+miotmrdOpROauKll0IVWWUljCxrx1DC\nEJHcN25cqEa6804oKIB99gnbTz0VkgCE1889N/Soio3dGDQoTDMCIWG0bg2XXx5KHvvuGxrVAbbZ\nBg47LGw/80x4vvzyUCK4++6QPKrbNfeww2DkyDVLGIkJQyUMEZEU22QTeOut8Bf7O++ENopNNgmJ\nozyx3k9/+hOsvXbY7t49PJ9wApSUhCqqPn3Csd13D1Vem2yy5kDBH3+EU0+FgQPD/tNPV73uuDv8\n+msoAVXU6A0qYYiIpIUZ9OgB665b9bl33w3XXAMbb1y6hBGz1lrhOZYwNtus9HOiDz8Mz599Fuap\n+vOf4Ywz4MsvK+7uu3hx6HWVmCDy8tZMGCphiIhkWK9eMHx4SDLlJYyY7bYL52y9ddjffPPwHGsD\nAZg+Pb59883hecqU8BnHH7/me86fH5+KJNbI3bw5tGq1ZqO3ShgiIlkkViVVXsLYc0+YOjX8+EO8\nhPG3v0GnTtAxWjC0adOQRN57r/T148aV3ncPPbguuijsx9osmjeHli0rL2G4p28sSJKUMESkYevf\nH/r2DdVZZZmVTiS77w777x8GCs6eHW8jyc8P1VexgXZz5675Xl9+GUoj06ZBcXE49ttvoRTRrFlI\nGGUbvRNLGP/3f3DQQbW711pSwhCRhm2vvWDixOQGyHXqBE88Ee+KG2svWX/9UHKISRxRvtVWYbGn\nXr3ggAPCsalTw3NiCaN168p7SX35ZXzuKwjtI3W8TK0ShohITa23Xnju1q10wkhs7P7kkzBPFcRn\n1l2+PDzH2iyaNQsJY/HiitswFi8uPbfVBReE1QfrkBKGiEhNJZYwYj2qEvXsCYceWvH1yZQwtt02\nVJuVTRhffRVKKp9+Gkam1wHNJSUiUlOxhBErYdx6K7zyCjzySBgL8s03oV2ja9cw2eH775e+PtaG\nEUsYM2eu2YbxwQdhu2vXkFBuvz1MXzJ/fti/9VaYMCF1U5VUQiUMEZGaKiyEwYNDbyqzMJ3IhhuG\n1zp0CMfy8sKI7iOPXPP6ZcvCj35ilVTZqUJiYiWMp58O05z8+mvYX7IkdMmtAyphiIjUVKtW8Nhj\npY+1bx+eY+M7YjbaqPz3mDcvtIWU14YRWx0QYNGi0K123rzQCyvWzXbevDpLGCphiIikUmxcR9mE\nESt5NGpU+vi8eRWXMGLTsUN8DMbs2aFkMXt2fF8JQ0QkB8USRmxQX0z37iExlG0cnzcv3obx++8h\nGcSWiE1MGDGxRBHraVVSooQhIpKTKiphNGkCr70WplhPlNhLCkICadkSGjcuXSUVs3Bh6X2VMERE\nclRFCQNC99jYan6NE5qQW7eOJ4w5c0JJpHnz5D5v6VIlDBGRnFRZwkg8njirbn5+PGHMnRsSRnWW\nZs31hGFmY8xstplNLue1v5uZm1nHCq5d38xeMLMpZvaFmRWkK04RkZRabz047zw48MDyX2/ZMrRR\nxEaJQxhjUTZhxBrHYzPlVibXEwYwFti77EEz6wbsCfxUybX3ANe4ey+gHzA7HQGKiKRcXh78+9/x\nXlFlmcFdd8E558SPlS1hNG8eb9zebbeqPzN2bZqlLWG4+xtAeWsWXgecA3h515nZZkBjd38xep/F\n7v5beeeKiOSkQw8N65DHdO0aLyXMn1+6OiqZhFEPShhrMLP9genu/mklp20MzDezx8zsEzO7xswa\nVXSymQ01s2IzKy4pKUl5zCIiaZH4I59YwoDSCWPLLav3XmlUZwnDzFoCFwAXV3FqY2BHYDiwDbAB\n8NeKTnb3O9y9yN2LOnXqlKJoRUTSrGXL+HabNmsmjDZtwnPbtlW/V31LGEBPoAfwqZlNBfKBj81s\nnTLnTQMmuvv37r4SeALYqg7jFBFJv7LdZhMTRosW8MsvoYtt06ZrntuiRUgoMfVtLil3/wzoHNuP\nkkaRu88pc+qHQDsz6+TuJcCuQHFdxSkiUifMSu8nJoxNNim9v9ZaYdzG4sVhv6gozCX19tvhOddL\nGGZ2P/AusImZTTOzEyo5t8jMRgO4+ypCddTLZvYZYMCodMUpIpJRsSqnpk3jx8p2pW3bNjxiJY3b\nboOXXoonilwvYbj74VW8XpCwXQwMSdh/EShnNRIRkXrks8+gc1TxkljiSOxBBaGEkZcX5ppauhTa\ntYuvA754ce4nDBERqULv3uUf32CD0vvt2oWEsWRJmGsqVl0VazhXwhARaaDKtm9cfnmYpPDEE8N+\nLEHEEkYdDdxTwhARyRbvvFN+N9pttw3PLVuGdozYxIX1pQ1DRESqabvtKn+9RYvSpYk6rpLSbLUi\nIrmiZcs1E4ZZ8lOh15IShohIrmjXLr5mOISSRatWa7Z5pImqpEREcsVVV4WeUjEtW9ZZdRQoYYiI\n5I6ePUvvDx0KO+1UZx+vhCEikqsGDAiPOqI2DBERSYoShoiIJEUJQ0REkqKEISIiSVHCEBGRpChh\niIhIUpQwREQkKUoYIiKSFHP3TMeQMmZWAvxYw8s7AmXXF89VupfsU1/uA3Qv2aqm99Ld3Tslc2K9\nShi1YWbF7l5U9ZnZT/eSferLfYDuJVvVxb2oSkpERJKihCEiIklRwoi7I9MBpJDuJfvUl/sA3Uu2\nSvu9qA1DRESSohKGiIgkRQlDRESS0iAShpl1M7NXzewLM/vczE6Pjq9tZi+a2TfRc/vouJnZjWb2\nrZlNMrOtMnsHcZXcyz/MbLqZTYwegxKuOT+6l6/MbK/MRV+amTU3sw/M7NPoXv4ZHe9hZu9HMT9o\nZk2j482i/W+j1wsyGX+iSu5lrJn9kPC9FEbHs/bfGICZNTKzT8zsqWg/576TmHLuJSe/EwAzm2pm\nn0VxF0fH6u53zN3r/QNYF9gq2m4DfA1sBlwNnBcdPw+4KtoeBDwLGNAfeD/T95DEvfwDGF7O+ZsB\nnwLNgB7Ad0CjTN9HFJsBraPtJsD70X/vh4DDouO3AydH2/8H3B5tHwY8mOl7SOJexgIHl3N+1v4b\ni+I7C7gPeCraz7nvpJJ7ycnvJIpxKtCxzLE6+x1rECUMd5/h7h9H24uAKUBXYH/g7ui0u4EDou39\ngXs8eA9oZ2br1nHY5arkXiqyP/CAuy9z9x+Ab4F+6Y+0atF/38XRbpPo4cCuwCPR8bLfS+z7egTY\nzcysjsKtVCX3UpGs/TdmZvnAPsDoaN/Iwe8E1ryXKmTtd1KFOvsdaxAJI1FUZN6S8BdgF3efEb00\nE+gSbXcFfk64bBqV/yhnRJl7ATg1KnqOiRVLyfJ7iaoLJgKzgRcJJaD57r4yOiUx3j/uJXp9AdCh\nbiOuWNl7cffY93J59L1cZ2bNomPZ/L1cD5wDrI72O5Cj3wlr3ktMrn0nMQ68YGYfmdnQ6Fid/Y41\nqIRhZq2BR4Ez3H1h4mseynA508e4nHu5DegJFAIzgJEZDC9p7r7K3QuBfELJZ9MMh1RjZe/FzHoD\n5xPuaRtgbeDcDIZYJTP7MzDb3T/KdCy1Vcm95NR3UsYO7r4V8CfgFDPbKfHFdP+ONZiEYWZNCD+w\n4939sejwrFgRLXqeHR2fDnRLuDw/OpYVyrsXd58V/WCtBkYRr3bK6nuJcff5wKvAdoSic+PopcR4\n/7iX6PW2wNw6DrVKCfeyd1SF6O6+DLiL7P9eBgD7mdlU4AFCVdQN5OZ3ssa9mNm4HPxO/uDu06Pn\n2cDjhNjr7HesQSSMqE71TmCKu1+b8NIE4Nho+1jgyYTjx0S9DPoDCxKKfBlV0b2UqZscDEyOticA\nh0W9WXoAGwEf1FW8lTGzTmbWLtpuAexBaJN5FTg4Oq3s9xL7vg4GXon+osq4Cu7ly4T/kY1Qt5z4\nvWTdvzF3P9/d8929gNCI/Yq7H0kOficV3MtRufadxJhZKzNrE9sG9iTEXne/Y7VtNc+FB7ADoZg2\nCZgYPQYR6lpfBr4BXgLWjs434BZCffpnQFGm7yGJe7k3inVS9A9l3YRrLozu5SvgT5m+h4S4+gCf\nRDFPBi6Ojm9ASGrfAg8DzaLjzaP9b6PXN8j0PSRxL69E38tkYBzxnlRZ+28s4Z4GEu9ZlHPfSSX3\nkpPfSfQdfBo9PgcujI7X2e+YpgYREZGkNIgqKRERqT0lDBERSYoShoiIJEUJQ0REkqKEISIiSVHC\nEKmCma1KmNl0opmdl8L3LjCzyVWfKZJ5jas+RaTB+93DlB8iDZpKGCI1FK1NcHW0PsEHZrZhdLzA\nzF6JJrd72czWj453MbPHLayZ8amZbR+9VSMzG2VhHY0XopHimNkwC+ueTDKzBzJ0myJ/UMIQqVqL\nMlVShya8tsDdtwBuJsyMCnATcLe79wHGAzdGx28EXnf3vsBWhNG6EKZrucXdNwfmAwdFx88Dtoze\n52/pujmRZGmkt0gVzGyxu7cu5/hUYFd3/z6aEHKmu3cwszmEqVlWRMdnuHtHMysB8j1Mehd7jwLC\nVOgbRfvnAk3c/TIzew5YDDwBPOHx9TZEMkIlDJHa8Qq2q2NZwvYq4m2L+xDmAtoK+DBhtliRjFDC\nEKmdQxOe34223yHMjgpwJPBmtP0ycDL8sdhS24re1MzygG7u/iphvYa2wBqlHJG6pL9YRKrWIlpJ\nL+Y5d491rW1vZpMIpYTDo2OnAXeZ2dlACXBcdPx04A4zO4FQkjiZsNhVeRoB46KkYsCNHtbZEMkY\ntWGI1FDUhlHk7nMyHYtIXVCVlIiIJEUlDBERSYpKGCIikhQlDBERSYoShoiIJEUJQ0REkqKEISIi\nSfl/ajIimPko3MQAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AOVzQXxCwkzP",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "a68bf3ce-bb7d-4759-b51e-8eea04dc138d"
      },
      "source": [
        "print(rnn_forecast)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[1.80334488e+02 1.65932495e+02 2.21385651e+02 2.02602341e+02\n",
            " 1.95812378e+02 1.66568024e+02 1.72221893e+02 1.72351089e+02\n",
            " 1.71359238e+02 1.92315262e+02 1.79430908e+02 1.90465561e+02\n",
            " 1.70150406e+02 1.54548752e+02 1.45760834e+02 1.52615173e+02\n",
            " 1.56858536e+02 1.52650833e+02 1.31245773e+02 1.35252060e+02\n",
            " 1.27649216e+02 9.58357544e+01 9.90672379e+01 8.83868637e+01\n",
            " 8.30485153e+01 8.85051575e+01 8.65534286e+01 8.34583511e+01\n",
            " 1.14608894e+02 1.31056335e+02 1.32336349e+02 1.03073143e+02\n",
            " 9.93796616e+01 1.11037933e+02 8.51060486e+01 8.48528519e+01\n",
            " 7.69715576e+01 6.63921509e+01 5.09542046e+01 4.30102539e+01\n",
            " 4.26469612e+01 3.53892822e+01 4.97657166e+01 5.25579033e+01\n",
            " 6.69459915e+01 4.63514252e+01 4.48055305e+01 4.08606110e+01\n",
            " 4.48441048e+01 3.97551613e+01 2.68832245e+01 2.22263908e+01\n",
            " 3.32603378e+01 1.98383331e+01 3.29938431e+01 2.01695232e+01\n",
            " 3.07538300e+01 2.96630077e+01 1.91891308e+01 1.61151085e+01\n",
            " 1.49790630e+01 1.13105726e+01 6.95184612e+00 1.25114498e+01\n",
            " 8.35816765e+00 8.35663891e+00 5.78935814e+00 5.79682589e+00\n",
            " 1.16962347e+01 8.03432274e+00 1.20681610e+01 1.44447117e+01\n",
            " 1.35417757e+01 2.94974022e+01 3.03362312e+01 1.65208931e+01\n",
            " 2.06109409e+01 3.65144157e+01 4.15551453e+01 4.04895859e+01\n",
            " 5.55136185e+01 5.71473541e+01 7.38654099e+01 1.04196785e+02\n",
            " 9.85529556e+01 1.01231300e+02 1.45900040e+02 1.49653046e+02\n",
            " 1.48674088e+02 1.78597641e+02 1.60793045e+02 1.76005417e+02\n",
            " 2.15345978e+02 2.18692657e+02 2.04921768e+02 2.43032333e+02\n",
            " 2.29149796e+02 2.17396591e+02 1.98263641e+02 2.14212067e+02\n",
            " 2.20654266e+02 2.16180878e+02 2.44907959e+02 2.26471115e+02\n",
            " 2.09466293e+02 2.62323364e+02 2.63122223e+02 2.32738235e+02\n",
            " 2.47076920e+02 2.25964447e+02 2.12129639e+02 2.24748581e+02\n",
            " 2.24300079e+02 2.14740845e+02 2.13313644e+02 2.24360916e+02\n",
            " 2.29402786e+02 2.31637024e+02 2.16838806e+02 1.99786560e+02\n",
            " 2.19870972e+02 2.33072495e+02 1.94775742e+02 2.21829193e+02\n",
            " 2.00144699e+02 2.08412048e+02 2.09843536e+02 1.94290909e+02\n",
            " 2.28901260e+02 1.92416458e+02 1.65460159e+02 1.71025986e+02\n",
            " 1.66090714e+02 1.84573349e+02 1.55505737e+02 1.50474411e+02\n",
            " 1.63159149e+02 1.60443939e+02 1.57341064e+02 1.67280853e+02\n",
            " 1.73591751e+02 1.73301941e+02 1.35892731e+02 1.36431610e+02\n",
            " 1.28136902e+02 1.00352776e+02 8.78241730e+01 8.31630249e+01\n",
            " 8.49211044e+01 7.78040009e+01 1.02881683e+02 9.54657288e+01\n",
            " 8.53281555e+01 9.38234711e+01 7.08486710e+01 6.14621582e+01\n",
            " 6.10434227e+01 5.56720924e+01 6.86077499e+01 6.48660736e+01\n",
            " 6.59970856e+01 6.47290726e+01 6.22114944e+01 4.65568886e+01\n",
            " 4.21045647e+01 6.45580597e+01 5.24081154e+01 4.62037430e+01\n",
            " 4.42243958e+01 3.66657181e+01 3.61943703e+01 2.95667744e+01\n",
            " 4.08017769e+01 5.41507797e+01 4.84194908e+01 3.74645729e+01\n",
            " 4.81898193e+01 5.19782829e+01 4.97775040e+01 4.16737480e+01\n",
            " 3.44632759e+01 2.95142746e+01 2.84039707e+01 2.71609306e+01\n",
            " 2.07774258e+01 2.04812775e+01 1.92006016e+01 1.43246059e+01\n",
            " 1.73267021e+01 1.34946404e+01 1.32155418e+01 1.45473328e+01\n",
            " 1.95044460e+01 2.43999100e+01 2.37942066e+01 2.13729115e+01\n",
            " 1.50702238e+01 3.09054050e+01 2.45510693e+01 2.24067688e+01\n",
            " 2.06282139e+01 2.46833229e+01 2.95198078e+01 2.66795120e+01\n",
            " 2.86514931e+01 3.88216896e+01 3.54653816e+01 3.85191727e+01\n",
            " 6.10737991e+01 5.90192719e+01 6.47911224e+01 7.58662643e+01\n",
            " 7.11807938e+01 7.28578415e+01 8.01933594e+01 8.03466644e+01\n",
            " 9.44169617e+01 1.27368805e+02 1.22662178e+02 1.45626099e+02\n",
            " 1.11022209e+02 1.24579285e+02 1.07586777e+02 1.22594559e+02\n",
            " 1.44712280e+02 1.16514267e+02 1.29475647e+02 1.32039581e+02\n",
            " 1.63347763e+02 1.65466949e+02 1.57472092e+02 1.41724915e+02\n",
            " 1.29263138e+02 1.59658417e+02 1.53026062e+02 1.45310257e+02\n",
            " 1.57103409e+02 1.60838135e+02 1.55957870e+02 1.36826782e+02\n",
            " 1.53627792e+02 1.48487869e+02 1.64266983e+02 1.82798874e+02\n",
            " 1.55618439e+02 1.68804001e+02 1.55359619e+02 1.46372482e+02\n",
            " 1.46720581e+02 1.37912399e+02 1.40340729e+02 1.37690002e+02\n",
            " 1.40409103e+02 1.51524551e+02 1.70487900e+02 1.48999390e+02\n",
            " 1.56667755e+02 1.69161911e+02 1.54085800e+02 1.61518280e+02\n",
            " 1.42558731e+02 1.46152420e+02 1.33228745e+02 1.38533051e+02\n",
            " 1.27636490e+02 1.32584366e+02 1.21180977e+02 1.02187469e+02\n",
            " 1.08934128e+02 9.35607224e+01 8.34286499e+01 1.09131691e+02\n",
            " 9.04194183e+01 8.18922348e+01 8.28097763e+01 8.71070557e+01\n",
            " 1.06209503e+02 9.04404297e+01 1.18110535e+02 1.09778366e+02\n",
            " 9.71743469e+01 1.14202858e+02 1.17811287e+02 1.11850777e+02\n",
            " 1.14708366e+02 1.01724861e+02 9.69455566e+01 7.77068405e+01\n",
            " 7.41204224e+01 6.82534637e+01 6.36085320e+01 6.70533142e+01\n",
            " 7.78339386e+01 6.45392838e+01 6.25613327e+01 4.96888351e+01\n",
            " 4.44300079e+01 7.23760147e+01 4.58792725e+01 4.47138252e+01\n",
            " 4.36695747e+01 3.98090401e+01 3.86550903e+01 3.47973213e+01\n",
            " 5.28686638e+01 5.18004646e+01 5.16592484e+01 7.15002136e+01\n",
            " 5.13423462e+01 5.90771980e+01 6.61558838e+01 4.60376434e+01\n",
            " 4.23186226e+01 3.65426102e+01 2.59141750e+01 2.36728363e+01\n",
            " 1.79407597e+01 1.81218376e+01 2.05800419e+01 3.26580505e+01\n",
            " 4.96932297e+01 2.58307934e+01 2.37189503e+01 2.75787201e+01\n",
            " 1.59655275e+01 1.79281807e+01 1.54195366e+01 2.69255600e+01\n",
            " 2.71711121e+01 2.34288063e+01 2.38378830e+01 1.36002750e+01\n",
            " 2.15883160e+01 2.14146500e+01 2.87655106e+01 1.67184448e+01\n",
            " 2.31876316e+01 2.59262447e+01 3.13802509e+01 2.02937450e+01\n",
            " 2.30752945e+01 2.86415310e+01 4.65805931e+01 3.40712738e+01\n",
            " 4.51508255e+01 5.84258728e+01 5.73884315e+01 4.77170753e+01\n",
            " 6.17267189e+01 6.93778839e+01 1.06798279e+02 1.01173576e+02\n",
            " 1.25798859e+02 1.14290581e+02 1.26883446e+02 1.08950104e+02\n",
            " 9.63511734e+01 1.57578827e+02 1.64986588e+02 1.45776459e+02\n",
            " 1.71711548e+02 2.07875031e+02 1.89406433e+02 1.97116501e+02\n",
            " 1.66152267e+02 1.86906891e+02 1.96798569e+02 2.13821106e+02\n",
            " 2.00795670e+02 2.37212585e+02 2.25038651e+02 2.27328903e+02\n",
            " 2.24799301e+02 2.15439728e+02 2.12535004e+02 1.90247864e+02\n",
            " 2.12942429e+02 2.18225601e+02 2.07538895e+02 1.93824875e+02\n",
            " 1.89976395e+02 2.01728226e+02 2.07946136e+02 1.98870071e+02\n",
            " 2.17024399e+02 1.68301407e+02 1.81981216e+02 1.82457962e+02\n",
            " 1.97727020e+02 1.88343185e+02 1.49431442e+02 1.82772308e+02\n",
            " 2.01412781e+02 2.17511108e+02 2.00755905e+02 1.84940018e+02\n",
            " 1.90339188e+02 1.62331955e+02 2.03049973e+02 1.98622971e+02\n",
            " 1.77716980e+02 1.46634521e+02 1.60829437e+02 1.43446655e+02\n",
            " 1.58258118e+02 1.63973007e+02 1.42714890e+02 1.38602463e+02\n",
            " 1.56820694e+02 1.27004326e+02 1.01986191e+02 1.01265823e+02\n",
            " 1.04583130e+02 1.24426971e+02 1.29546646e+02 1.15099701e+02\n",
            " 1.07097542e+02 8.85315323e+01 8.25859985e+01 6.30382462e+01\n",
            " 5.56807175e+01 6.91187668e+01 9.33850861e+01 1.01377014e+02\n",
            " 9.13077698e+01 9.78228073e+01 7.85954819e+01 6.81543961e+01\n",
            " 5.30293350e+01 3.94819031e+01 3.00879307e+01 2.83126087e+01\n",
            " 2.51978912e+01 2.24993820e+01 2.48098640e+01 2.01837215e+01\n",
            " 2.23874874e+01 3.07724266e+01 2.86077805e+01 3.76808357e+01\n",
            " 1.98459835e+01 1.63331356e+01 1.79056911e+01 1.83431244e+01\n",
            " 2.05366077e+01 1.43484707e+01 2.42645588e+01 1.71379471e+01\n",
            " 2.40714989e+01 2.12402115e+01 1.17511625e+01 1.65559921e+01\n",
            " 1.34101753e+01 1.15197372e+01 3.05464115e+01 1.81261215e+01\n",
            " 1.57849283e+01 1.54593706e+01 9.57685375e+00 1.46957550e+01\n",
            " 3.49923592e+01 3.83278351e+01 2.71602192e+01 4.03216286e+01\n",
            " 4.55052719e+01 4.24415779e+01 5.93745422e+01 5.05955048e+01\n",
            " 3.88134689e+01 6.61922760e+01 4.73455391e+01 7.82362366e+01\n",
            " 9.58514404e+01 7.42624207e+01 1.10761414e+02 1.15014664e+02\n",
            " 1.26037346e+02 1.35335175e+02 1.45298035e+02 1.50355728e+02\n",
            " 1.97872986e+02 1.92686111e+02 1.94778580e+02 1.77710831e+02\n",
            " 1.75571198e+02 1.89388428e+02 2.41121033e+02 1.92695236e+02\n",
            " 2.20222687e+02 2.13181564e+02 2.07627792e+02 2.23591064e+02\n",
            " 2.06799774e+02 2.15947510e+02 1.83855606e+02 1.90535431e+02\n",
            " 1.84993607e+02 1.74737930e+02 1.60824341e+02 1.78468948e+02\n",
            " 2.17662018e+02 1.71456802e+02 1.89266907e+02 1.74205109e+02\n",
            " 1.82910568e+02 1.85110886e+02 2.12517929e+02 1.88914459e+02\n",
            " 1.90675659e+02 1.72681061e+02 1.98672424e+02 2.10537552e+02\n",
            " 2.13916534e+02 1.80722076e+02 1.92636520e+02 1.68750626e+02\n",
            " 1.92674103e+02 1.90199127e+02 2.09849045e+02 1.65790207e+02\n",
            " 1.54409256e+02 1.17261490e+02 1.04469048e+02 1.14816956e+02\n",
            " 1.00872490e+02 1.01885056e+02 1.26273621e+02 1.23363533e+02\n",
            " 1.23726036e+02 9.97467499e+01 1.21678078e+02 1.07984688e+02\n",
            " 1.03027596e+02 9.32366333e+01 8.26887207e+01 8.41713028e+01\n",
            " 7.15002060e+01 5.34836578e+01 6.76119995e+01 5.22423401e+01\n",
            " 6.56422577e+01 7.88098297e+01 5.84585075e+01 5.73587303e+01\n",
            " 4.58240585e+01 3.75269508e+01 3.98293457e+01 4.76420822e+01\n",
            " 3.83973694e+01 4.27290077e+01 6.01959763e+01 3.54597549e+01\n",
            " 3.89191437e+01 3.82567596e+01 4.20302887e+01 4.56628342e+01\n",
            " 3.06837997e+01 2.92158928e+01 2.70421333e+01 2.34738522e+01\n",
            " 2.30245323e+01 2.13065662e+01 3.05301132e+01 1.89811268e+01\n",
            " 1.97886353e+01 1.88074112e+01 1.33204441e+01 1.63984547e+01\n",
            " 1.32847338e+01 1.29002094e+01 1.76902390e+01 1.44467535e+01\n",
            " 2.02372074e+01 1.09312105e+01 9.21892834e+00 1.56664429e+01\n",
            " 1.47100868e+01 1.41152143e+01 1.59936581e+01 1.34268703e+01\n",
            " 2.20998230e+01 2.59100056e+01 2.45539494e+01 1.88398476e+01\n",
            " 3.37079315e+01 5.19081993e+01 3.59769440e+01 5.18979683e+01\n",
            " 5.42667580e+01 4.58354340e+01 5.31654663e+01 7.44985199e+01\n",
            " 6.71019135e+01 7.36556931e+01 8.62165298e+01 8.96582336e+01\n",
            " 1.09621918e+02 1.21192017e+02 9.00263596e+01 1.03268692e+02\n",
            " 1.12265457e+02 9.31935501e+01 1.03553139e+02 1.05156731e+02\n",
            " 9.78892365e+01 1.34182327e+02 1.63898529e+02 1.64174423e+02\n",
            " 1.47922699e+02 1.25164047e+02 1.50472061e+02 1.68578812e+02\n",
            " 1.32656494e+02 1.43135483e+02 1.54517120e+02 1.91370972e+02\n",
            " 1.84853775e+02 1.70453476e+02 1.82723846e+02 2.11843750e+02\n",
            " 1.83936813e+02 1.72160034e+02 1.57568436e+02 1.59526108e+02\n",
            " 1.49604919e+02 1.48913162e+02 1.33925598e+02 1.54804199e+02\n",
            " 1.54612732e+02 1.46424118e+02 1.87063766e+02 1.36210068e+02\n",
            " 1.58797394e+02 2.05637817e+02 1.81334396e+02 1.79982254e+02\n",
            " 1.99379791e+02 1.82158401e+02 1.77415649e+02 1.62385971e+02\n",
            " 1.79342896e+02 1.81226486e+02 1.47111267e+02 1.63654755e+02\n",
            " 1.66628510e+02 1.75647537e+02 1.58465546e+02 1.54251923e+02\n",
            " 1.44591171e+02 1.39448059e+02 1.03712502e+02 1.09580147e+02\n",
            " 1.04321800e+02 9.44769287e+01 1.18566704e+02 1.21424782e+02\n",
            " 1.15656456e+02 9.52442932e+01 1.04032982e+02 9.21928253e+01\n",
            " 8.18520508e+01 7.40205688e+01 7.67434921e+01 7.56975021e+01\n",
            " 6.56690903e+01 7.41820068e+01 6.89404755e+01 7.92917175e+01\n",
            " 7.21156921e+01 6.00307045e+01 7.31430817e+01 6.95355606e+01\n",
            " 4.75346947e+01 5.42175293e+01 4.76590500e+01 4.11143684e+01\n",
            " 4.28007965e+01 5.71633034e+01 5.40559845e+01 6.05606003e+01\n",
            " 6.10982285e+01 4.68347092e+01 3.66626434e+01 3.09717350e+01\n",
            " 5.14881783e+01 2.61044807e+01 2.35336952e+01 2.05317917e+01\n",
            " 3.91963463e+01 3.40886078e+01 3.12679329e+01 2.97338715e+01\n",
            " 2.49889164e+01 2.58229618e+01 2.12815704e+01 3.47120247e+01\n",
            " 2.53222256e+01 3.07066555e+01 2.39503212e+01 1.58125010e+01\n",
            " 1.35190907e+01 1.71051617e+01 2.06625671e+01 1.93207779e+01\n",
            " 1.66601009e+01 1.09768848e+01 9.81029224e+00 6.65819788e+00\n",
            " 1.25605974e+01 8.19422722e+00 7.91126776e+00 1.15781851e+01\n",
            " 7.00577068e+00 8.57801247e+00 8.43622017e+00 4.41735744e+00\n",
            " 3.18398142e+00 1.50581449e-01 1.40160990e+00 4.90350819e+00\n",
            " 3.67296886e+00 3.25497532e+00 2.01033282e+00 1.16522419e+00\n",
            " 1.44637966e+00 2.53149462e+00 6.23854780e+00 7.88327742e+00\n",
            " 5.30345678e+00 7.41738462e+00 9.50735378e+00 9.28595543e+00\n",
            " 1.61396351e+01 1.96733265e+01 2.68694553e+01 2.54292641e+01\n",
            " 1.62891922e+01 1.69387188e+01 1.89826660e+01 2.42581692e+01\n",
            " 2.89116497e+01 3.50068130e+01 3.46775017e+01 3.57827148e+01\n",
            " 2.89749832e+01 2.99346275e+01 4.53510017e+01 6.74130096e+01\n",
            " 6.88268204e+01 6.17183037e+01 6.09583359e+01 6.45234375e+01\n",
            " 6.50637360e+01 1.00573746e+02 1.10612930e+02 1.22751320e+02\n",
            " 1.09334290e+02 1.01776459e+02 7.49586792e+01 8.49487152e+01\n",
            " 8.35407028e+01 9.21721115e+01 9.36255646e+01 9.82295227e+01\n",
            " 9.70084000e+01 9.68032608e+01 8.61651688e+01 9.08066864e+01\n",
            " 7.13037186e+01 9.02029114e+01 6.99358292e+01 7.86336365e+01\n",
            " 9.93824615e+01 1.06921150e+02 8.48748245e+01 9.21117172e+01\n",
            " 9.33974838e+01 6.95344467e+01 1.05391159e+02 1.03869339e+02\n",
            " 1.15744820e+02 1.16837944e+02 1.35267197e+02 1.28180435e+02\n",
            " 1.20749458e+02 1.21289268e+02 1.12901413e+02 1.09156021e+02\n",
            " 1.11610748e+02 1.25188507e+02 1.01079384e+02 1.09566162e+02\n",
            " 1.13436142e+02 1.00501106e+02 8.44188538e+01 7.29109955e+01\n",
            " 7.62326584e+01 8.32038193e+01 7.17094879e+01 7.25647659e+01\n",
            " 6.97003098e+01 7.51444550e+01 6.75707550e+01 6.66218491e+01\n",
            " 6.33238602e+01 6.02824745e+01 5.91588631e+01 5.68944321e+01\n",
            " 4.65961800e+01 5.25061111e+01 3.51774902e+01 3.57129478e+01\n",
            " 4.79662704e+01 4.27065353e+01 3.88280678e+01 3.17364311e+01\n",
            " 2.59528580e+01 2.72865810e+01 2.77640705e+01 2.25494385e+01\n",
            " 3.29074783e+01 2.21985989e+01 2.23227921e+01 2.22134590e+01\n",
            " 3.09352245e+01 3.89845276e+01 2.08309975e+01 1.81112862e+01\n",
            " 1.52645435e+01 1.02301559e+01 1.22047253e+01 9.86316109e+00\n",
            " 1.05511665e+01 1.33266296e+01 1.59221144e+01]\n"
          ],
          "name": "stdout"
        }
      ]
    }
  ]
}