linear_regression_with_synthetic_data.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "cellView": "form",
    "colab": {},
    "colab_type": "code",
    "id": "wDlWLbfkJtvu"
   },
   "outputs": [],
   "source": [
    "#@title Copyright 2020 Google LLC. Double-click here for license information.\n",
    "# Licensed under the Apache License, Version 2.0 (the \"License\");\n",
    "# you may not use this file except in compliance with the License.\n",
    "# You may obtain a copy of the License at\n",
    "#\n",
    "# https://www.apache.org/licenses/LICENSE-2.0\n",
    "#\n",
    "# Unless required by applicable law or agreed to in writing, software\n",
    "# distributed under the License is distributed on an \"AS IS\" BASIS,\n",
    "# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
    "# See the License for the specific language governing permissions and\n",
    "# limitations under the License."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "TL5y5fY9Jy_x"
   },
   "source": [
    "# Simple Linear Regression with Synthetic Data\n",
    "\n",
    "In this first Colab, you'll explore linear regression with a simple database. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "DZz29MwGgE2Y"
   },
   "source": [
    "## Learning objectives:\n",
    "\n",
    "After doing this exercise, you'll know how to do the following:\n",
    "\n",
    "  * Run Colabs.\n",
    "  * Tune the following [hyperparameters](https://developers.google.com/machine-learning/glossary/#hyperparameter):\n",
    "    * [learning rate](https://developers.google.com/machine-learning/glossary/#learning_rate)\n",
    "    * number of [epochs](https://developers.google.com/machine-learning/glossary/#epoch)\n",
    "    * [batch size](https://developers.google.com/machine-learning/glossary/#batch_size)\n",
    "  * Interpret different kinds of [loss curves](https://developers.google.com/machine-learning/glossary/#loss_curve)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "wm3K3dHlf_1n"
   },
   "source": [
    "## About Colabs\n",
    "\n",
    "Machine Learning Crash Course uses Colaboratories (**Colabs**) for all programming exercises. Colab is Google's implementation of [Jupyter Notebook](https://jupyter.org/). Like all Jupyter Notebooks, a Colab consists of two kinds of components:\n",
    "\n",
    "  * **Text cells**, which contain explanations. You are currently reading a text cell.\n",
    "  * **Code cells**, which contain Python code for you to run. Code cells have a light gray background.\n",
    "\n",
    "You *read* the text cells and *run* the code cells."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "qc3e5Jbbf5aA"
   },
   "source": [
    "### Running code cells\n",
    "\n",
    "You must run code cells in order. In other words, you may only run a code cell once all the code cells preceding it have already been run. \n",
    "\n",
    "To run a code cell:\n",
    "\n",
    "  1. Place the cursor anywhere inside the [ ] area at the top left of a code cell. The area inside the [ ] will display an arrow.\n",
    "  2. Click the arrow.\n",
    "\n",
    "Alternatively, you may invoke **Runtime->Run all**. Note, though, that some of the code cells will fail because not all the coding is complete. (You'll complete the coding as part of the exercise.)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Sa35idJHf0__"
   },
   "source": [
    "### Understanding hidden code cells\n",
    "\n",
    "We've **hidden** the code in code cells that don't advance the learning objectives. For example, we've hidden the code that plots graphs. However, **you must still run code cells containing hidden code**. You'll know that the code is hidden because you'll see a title (for example, \"Load the functions that build and train a model\") without seeing the code.\n",
    "\n",
    "To view the hidden code, just double click the header."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "8H508dWUfvY1"
   },
   "source": [
    "### Why did you see an error?\n",
    "\n",
    "If a code cell returns an error when you run it, consider two common problems:\n",
    "\n",
    "  * You didn't run *all* of the code cells preceding the current code cell.\n",
    "  * If the code cell is labeled as a **Task**, then you haven't written the necessary code. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "bMr7MPVmoiHf"
   },
   "source": [
    "## Use the right version of TensorFlow\n",
    "\n",
    "The following hidden code cell ensures that the Colab will run on TensorFlow 2.X, which is the most recent version of TensorFlow:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "cellView": "form",
    "colab": {},
    "colab_type": "code",
    "id": "Z1pOWL7eevO8"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "UsageError: Line magic function `%tensorflow_version` not found.\n"
     ]
    }
   ],
   "source": [
    "#@title Run this Colab on TensorFlow 2.x\n",
    "%tensorflow_version 2.x"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "xchnxAsaKKqO"
   },
   "source": [
    "## Import relevant modules\n",
    "\n",
    "The following cell imports the packages that the program requires:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "9n9_cTveKmse"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import tensorflow as tf\n",
    "from matplotlib import pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "SIpsyJITPcbG"
   },
   "source": [
    "## Define functions that build and train a model\n",
    "\n",
    "The following code defines two functions:\n",
    "\n",
    "  * `build_model(my_learning_rate)`, which builds an empty model.\n",
    "  * `train_model(model, feature, label, epochs)`, which trains the model from the examples (feature and label) you pass. \n",
    "\n",
    "Since you don't need to understand model building code right now, we've hidden this code cell.  You may optionally double-click the headline to explore this code."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "cellView": "form",
    "colab": {},
    "colab_type": "code",
    "id": "xvO_beKVP1Ke"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Defined create_model and train_model\n"
     ]
    }
   ],
   "source": [
    "#@title Define the functions that build and train a model\n",
    "def build_model(my_learning_rate):\n",
    "  \"\"\"Create and compile a simple linear regression model.\"\"\"\n",
    "  # Most simple tf.keras models are sequential. \n",
    "  # A sequential model contains one or more layers.\n",
    "  model = tf.keras.models.Sequential()\n",
    "\n",
    "  # Describe the topography of the model.\n",
    "  # The topography of a simple linear regression model\n",
    "  # is a single node in a single layer. \n",
    "  model.add(tf.keras.layers.Dense(units=1, \n",
    "                                  input_shape=(1,)))\n",
    "\n",
    "  # Compile the model topography into code that \n",
    "  # TensorFlow can efficiently execute. Configure \n",
    "  # training to minimize the model's mean squared error. \n",
    "  model.compile(optimizer=tf.keras.optimizers.RMSprop(lr=my_learning_rate),\n",
    "                loss=\"mean_squared_error\",\n",
    "                metrics=[tf.keras.metrics.RootMeanSquaredError()])\n",
    "\n",
    "  return model           \n",
    "\n",
    "\n",
    "def train_model(model, feature, label, epochs, batch_size):\n",
    "  \"\"\"Train the model by feeding it data.\"\"\"\n",
    "\n",
    "  # Feed the feature values and the label values to the \n",
    "  # model. The model will train for the specified number \n",
    "  # of epochs, gradually learning how the feature values\n",
    "  # relate to the label values. \n",
    "  history = model.fit(x=feature,\n",
    "                      y=label,\n",
    "                      batch_size=None,\n",
    "                      epochs=epochs)\n",
    "\n",
    "  # Gather the trained model's weight and bias.\n",
    "  trained_weight = model.get_weights()[0]\n",
    "  trained_bias = model.get_weights()[1]\n",
    "\n",
    "  # The list of epochs is stored separately from the \n",
    "  # rest of history.\n",
    "  epochs = history.epoch\n",
    "  \n",
    "  # Gather the history (a snapshot) of each epoch.\n",
    "  hist = pd.DataFrame(history.history)\n",
    "\n",
    "  # Specifically gather the model's root mean \n",
    "  #squared error at each epoch. \n",
    "  rmse = hist[\"root_mean_squared_error\"]\n",
    "\n",
    "  return trained_weight, trained_bias, epochs, rmse\n",
    "\n",
    "print(\"Defined create_model and train_model\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Ak_TMAzGOIFq"
   },
   "source": [
    "## Define plotting functions\n",
    "\n",
    "We're using a popular Python library called [Matplotlib](https://developers.google.com/machine-learning/glossary/#matplotlib) to create the following two plots:\n",
    "\n",
    "*  a plot of the feature values vs. the label values, and a line showing the output of the trained model.\n",
    "*  a [loss curve](https://developers.google.com/machine-learning/glossary/#loss_curve).\n",
    "\n",
    "We hid the following code cell because learning Matplotlib is not relevant to the learning objectives. Regardless, you must still run all hidden code cells."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "cellView": "form",
    "colab": {},
    "colab_type": "code",
    "id": "QF0BFRXTOeR3"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Defined the plot_the_model and plot_the_loss_curve functions.\n"
     ]
    }
   ],
   "source": [
    "#@title Define the plotting functions\n",
    "def plot_the_model(trained_weight, trained_bias, feature, label):\n",
    "  \"\"\"Plot the trained model against the training feature and label.\"\"\"\n",
    "\n",
    "  # Label the axes.\n",
    "  plt.xlabel(\"feature\")\n",
    "  plt.ylabel(\"label\")\n",
    "\n",
    "  # Plot the feature values vs. label values.\n",
    "  plt.scatter(feature, label)\n",
    "\n",
    "  # Create a red line representing the model. The red line starts\n",
    "  # at coordinates (x0, y0) and ends at coordinates (x1, y1).\n",
    "  x0 = 0\n",
    "  y0 = trained_bias\n",
    "  x1 = my_feature[-1]\n",
    "  y1 = trained_bias + (trained_weight * x1)\n",
    "  plt.plot([x0, x1], [y0, y1], c='r')\n",
    "\n",
    "  # Render the scatter plot and the red line.\n",
    "  plt.show()\n",
    "\n",
    "def plot_the_loss_curve(epochs, rmse):\n",
    "  \"\"\"Plot the loss curve, which shows loss vs. epoch.\"\"\"\n",
    "\n",
    "  plt.figure()\n",
    "  plt.xlabel(\"Epoch\")\n",
    "  plt.ylabel(\"Root Mean Squared Error\")\n",
    "\n",
    "  plt.plot(epochs, rmse, label=\"Loss\")\n",
    "  plt.legend()\n",
    "  plt.ylim([rmse.min()*0.97, rmse.max()])\n",
    "  plt.show()\n",
    "\n",
    "print(\"Defined the plot_the_model and plot_the_loss_curve functions.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "LVSDPusELEZ5"
   },
   "source": [
    "## Define the dataset\n",
    "\n",
    "The dataset consists of 12 [examples](https://developers.google.com/machine-learning/glossary/#example). Each example consists of one [feature](https://developers.google.com/machine-learning/glossary/#feature) and one [label](https://developers.google.com/machine-learning/glossary/#label).\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "rnUSYKw4LUuh"
   },
   "outputs": [],
   "source": [
    "my_feature = ([1.0, 2.0,  3.0,  4.0,  5.0,  6.0,  7.0,  8.0,  9.0, 10.0, 11.0, 12.0])\n",
    "my_label   = ([5.0, 8.8,  9.6, 14.2, 18.8, 19.5, 21.4, 26.8, 28.9, 32.0, 33.8, 38.2])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "K24afla-4s2x"
   },
   "source": [
    "## Specify the hyperparameters\n",
    "\n",
    "The hyperparameters in this Colab are as follows:\n",
    "\n",
    "  * [learning rate](https://developers.google.com/machine-learning/glossary/#learning_rate)\n",
    "  * [epochs](https://developers.google.com/machine-learning/glossary/#epoch)\n",
    "  * [batch_size](https://developers.google.com/machine-learning/glossary/#batch_size)\n",
    "\n",
    "The following code cell initializes these hyperparameters and then invokes the functions that build and train the model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "Ye730h13CQ97"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "1/1 [==============================] - 0s 3ms/step - loss: 280.0426 - root_mean_squared_error: 16.7345\n",
      "Epoch 2/10\n",
      "1/1 [==============================] - 0s 996us/step - loss: 271.3862 - root_mean_squared_error: 16.4738\n",
      "Epoch 3/10\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 265.2376 - root_mean_squared_error: 16.2861\n",
      "Epoch 4/10\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 260.1668 - root_mean_squared_error: 16.1297\n",
      "Epoch 5/10\n",
      "1/1 [==============================] - 0s 956us/step - loss: 255.7218 - root_mean_squared_error: 15.9913\n",
      "Epoch 6/10\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 251.6936 - root_mean_squared_error: 15.8649\n",
      "Epoch 7/10\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 247.9660 - root_mean_squared_error: 15.7469\n",
      "Epoch 8/10\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 244.4667 - root_mean_squared_error: 15.6354\n",
      "Epoch 9/10\n",
      "1/1 [==============================] - 0s 995us/step - loss: 241.1474 - root_mean_squared_error: 15.5289\n",
      "Epoch 10/10\n",
      "1/1 [==============================] - 0s 673us/step - loss: 237.9741 - root_mean_squared_error: 15.4264\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/eric/.conda/envs/mlbase/lib/python3.8/site-packages/numpy/core/_asarray.py:136: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray\n",
      "  return array(a, dtype, copy=False, order=order, subok=True)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "learning_rate=0.01\n",
    "epochs=10\n",
    "my_batch_size=12\n",
    "\n",
    "my_model = build_model(learning_rate)\n",
    "trained_weight, trained_bias, epochs, rmse = train_model(my_model, my_feature, \n",
    "                                                         my_label, epochs,\n",
    "                                                         my_batch_size)\n",
    "plot_the_model(trained_weight, trained_bias, my_feature, my_label)\n",
    "plot_the_loss_curve(epochs, rmse)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "QwSm60H6pQjJ"
   },
   "source": [
    "## Task 1: Examine the graphs\n",
    "\n",
    "Examine the top graph. The blue dots identify the actual data; the red line identifies the output of the trained model. Ideally, the red line should align nicely with the blue dots.  Does it?  Probably not.\n",
    "\n",
    "A certain amount of randomness plays into training a model, so you'll get somewhat different results every time you train.  That said, unless you are an extremely lucky person, the red line probably *doesn't* align nicely with the blue dots.  \n",
    "\n",
    "Examine the bottom graph, which shows the loss curve. Notice that the loss curve decreases but doesn't flatten out, which is a sign that the model hasn't trained sufficiently."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "lLXPvqCRvgI4"
   },
   "source": [
    "## Task 2: Increase the number of epochs\n",
    "\n",
    "Training loss should steadily decrease, steeply at first, and then more slowly. Eventually, training loss should eventually stay steady (zero slope or nearly zero slope), which indicates that training has [converged](http://developers.google.com/machine-learning/glossary/#convergence).\n",
    "\n",
    "In Task 1, the training loss did not converge. One possible solution is to train for more epochs.  Your task is to increase the number of epochs sufficiently to get the model to converge. However, it is inefficient to train past convergence, so don't just set the number of epochs to an arbitrarily high value.\n",
    "\n",
    "Examine the loss curve. Does the model converge?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "uXuJH3h6t5qs"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/400\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 184.9640 - root_mean_squared_error: 13.6001\n",
      "Epoch 2/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 177.9486 - root_mean_squared_error: 13.3397\n",
      "Epoch 3/400\n",
      "1/1 [==============================] - 0s 935us/step - loss: 172.9904 - root_mean_squared_error: 13.1526\n",
      "Epoch 4/400\n",
      "1/1 [==============================] - 0s 833us/step - loss: 168.9159 - root_mean_squared_error: 12.9968\n",
      "Epoch 5/400\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 165.3552 - root_mean_squared_error: 12.8591\n",
      "Epoch 6/400\n",
      "1/1 [==============================] - 0s 839us/step - loss: 162.1370 - root_mean_squared_error: 12.7333\n",
      "Epoch 7/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 159.1664 - root_mean_squared_error: 12.6161\n",
      "Epoch 8/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 156.3842 - root_mean_squared_error: 12.5054\n",
      "Epoch 9/400\n",
      "1/1 [==============================] - 0s 743us/step - loss: 153.7511 - root_mean_squared_error: 12.3996\n",
      "Epoch 10/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 151.2391 - root_mean_squared_error: 12.2979\n",
      "Epoch 11/400\n",
      "1/1 [==============================] - 0s 795us/step - loss: 148.8279 - root_mean_squared_error: 12.1995\n",
      "Epoch 12/400\n",
      "1/1 [==============================] - 0s 882us/step - loss: 146.5022 - root_mean_squared_error: 12.1038\n",
      "Epoch 13/400\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 144.2498 - root_mean_squared_error: 12.0104\n",
      "Epoch 14/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 142.0612 - root_mean_squared_error: 11.9189\n",
      "Epoch 15/400\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 139.9290 - root_mean_squared_error: 11.8292\n",
      "Epoch 16/400\n",
      "1/1 [==============================] - 0s 917us/step - loss: 137.8468 - root_mean_squared_error: 11.7408\n",
      "Epoch 17/400\n",
      "1/1 [==============================] - 0s 809us/step - loss: 135.8095 - root_mean_squared_error: 11.6537\n",
      "Epoch 18/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 133.8128 - root_mean_squared_error: 11.5677\n",
      "Epoch 19/400\n",
      "1/1 [==============================] - 0s 850us/step - loss: 131.8530 - root_mean_squared_error: 11.4827\n",
      "Epoch 20/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 129.9272 - root_mean_squared_error: 11.3986\n",
      "Epoch 21/400\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 128.0326 - root_mean_squared_error: 11.3152\n",
      "Epoch 22/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 126.1671 - root_mean_squared_error: 11.2324\n",
      "Epoch 23/400\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 124.3287 - root_mean_squared_error: 11.1503\n",
      "Epoch 24/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 122.5156 - root_mean_squared_error: 11.0687\n",
      "Epoch 25/400\n",
      "1/1 [==============================] - 0s 732us/step - loss: 120.7266 - root_mean_squared_error: 10.9876\n",
      "Epoch 26/400\n",
      "1/1 [==============================] - 0s 932us/step - loss: 118.9602 - root_mean_squared_error: 10.9069\n",
      "Epoch 27/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 117.2153 - root_mean_squared_error: 10.8266\n",
      "Epoch 28/400\n",
      "1/1 [==============================] - 0s 773us/step - loss: 115.4910 - root_mean_squared_error: 10.7467\n",
      "Epoch 29/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 113.7864 - root_mean_squared_error: 10.6671\n",
      "Epoch 30/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 112.1006 - root_mean_squared_error: 10.5878\n",
      "Epoch 31/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 110.4332 - root_mean_squared_error: 10.5087\n",
      "Epoch 32/400\n",
      "1/1 [==============================] - 0s 816us/step - loss: 108.7833 - root_mean_squared_error: 10.4299\n",
      "Epoch 33/400\n",
      "1/1 [==============================] - 0s 766us/step - loss: 107.1506 - root_mean_squared_error: 10.3514\n",
      "Epoch 34/400\n",
      "1/1 [==============================] - 0s 790us/step - loss: 105.5345 - root_mean_squared_error: 10.2730\n",
      "Epoch 35/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 103.9346 - root_mean_squared_error: 10.1948\n",
      "Epoch 36/400\n",
      "1/1 [==============================] - 0s 958us/step - loss: 102.3505 - root_mean_squared_error: 10.1168\n",
      "Epoch 37/400\n",
      "1/1 [==============================] - 0s 961us/step - loss: 100.7819 - root_mean_squared_error: 10.0390\n",
      "Epoch 38/400\n",
      "1/1 [==============================] - 0s 790us/step - loss: 99.2284 - root_mean_squared_error: 9.9613\n",
      "Epoch 39/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 97.6899 - root_mean_squared_error: 9.8838\n",
      "Epoch 40/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 96.1660 - root_mean_squared_error: 9.8064\n",
      "Epoch 41/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 94.6566 - root_mean_squared_error: 9.7292\n",
      "Epoch 42/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 93.1614 - root_mean_squared_error: 9.6520\n",
      "Epoch 43/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 91.6803 - root_mean_squared_error: 9.5750\n",
      "Epoch 44/400\n",
      "1/1 [==============================] - 0s 724us/step - loss: 90.2131 - root_mean_squared_error: 9.4981\n",
      "Epoch 45/400\n",
      "1/1 [==============================] - 0s 753us/step - loss: 88.7597 - root_mean_squared_error: 9.4212\n",
      "Epoch 46/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 87.3199 - root_mean_squared_error: 9.3445\n",
      "Epoch 47/400\n",
      "1/1 [==============================] - 0s 824us/step - loss: 85.8936 - root_mean_squared_error: 9.2679\n",
      "Epoch 48/400\n",
      "1/1 [==============================] - 0s 916us/step - loss: 84.4808 - root_mean_squared_error: 9.1913\n",
      "Epoch 49/400\n",
      "1/1 [==============================] - 0s 870us/step - loss: 83.0813 - root_mean_squared_error: 9.1149\n",
      "Epoch 50/400\n",
      "1/1 [==============================] - 0s 997us/step - loss: 81.6950 - root_mean_squared_error: 9.0385\n",
      "Epoch 51/400\n",
      "1/1 [==============================] - 0s 947us/step - loss: 80.3219 - root_mean_squared_error: 8.9622\n",
      "Epoch 52/400\n",
      "1/1 [==============================] - 0s 713us/step - loss: 78.9618 - root_mean_squared_error: 8.8860\n",
      "Epoch 53/400\n",
      "1/1 [==============================] - 0s 914us/step - loss: 77.6148 - root_mean_squared_error: 8.8099\n",
      "Epoch 54/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 76.2807 - root_mean_squared_error: 8.7339\n",
      "Epoch 55/400\n",
      "1/1 [==============================] - 0s 648us/step - loss: 74.9595 - root_mean_squared_error: 8.6579\n",
      "Epoch 56/400\n",
      "1/1 [==============================] - 0s 697us/step - loss: 73.6512 - root_mean_squared_error: 8.5820\n",
      "Epoch 57/400\n",
      "1/1 [==============================] - 0s 645us/step - loss: 72.3557 - root_mean_squared_error: 8.5062\n",
      "Epoch 58/400\n",
      "1/1 [==============================] - 0s 645us/step - loss: 71.0729 - root_mean_squared_error: 8.4305\n",
      "Epoch 59/400\n",
      "1/1 [==============================] - 0s 732us/step - loss: 69.8029 - root_mean_squared_error: 8.3548\n",
      "Epoch 60/400\n",
      "1/1 [==============================] - 0s 928us/step - loss: 68.5455 - root_mean_squared_error: 8.2792\n",
      "Epoch 61/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 67.3008 - root_mean_squared_error: 8.2037\n",
      "Epoch 62/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 66.0687 - root_mean_squared_error: 8.1283\n",
      "Epoch 63/400\n",
      "1/1 [==============================] - 0s 867us/step - loss: 64.8491 - root_mean_squared_error: 8.0529\n",
      "Epoch 64/400\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 63.6422 - root_mean_squared_error: 7.9776\n",
      "Epoch 65/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 62.4477 - root_mean_squared_error: 7.9024\n",
      "Epoch 66/400\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 61.2658 - root_mean_squared_error: 7.8272\n",
      "Epoch 67/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 60.0963 - root_mean_squared_error: 7.7522\n",
      "Epoch 68/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 58.9393 - root_mean_squared_error: 7.6772\n",
      "Epoch 69/400\n",
      "1/1 [==============================] - 0s 959us/step - loss: 57.7947 - root_mean_squared_error: 7.6023\n",
      "Epoch 70/400\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 56.6625 - root_mean_squared_error: 7.5275\n",
      "Epoch 71/400\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 1ms/step - loss: 55.5428 - root_mean_squared_error: 7.4527\n",
      "Epoch 72/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 54.4354 - root_mean_squared_error: 7.3780\n",
      "Epoch 73/400\n",
      "1/1 [==============================] - 0s 730us/step - loss: 53.3403 - root_mean_squared_error: 7.3034\n",
      "Epoch 74/400\n",
      "1/1 [==============================] - 0s 719us/step - loss: 52.2576 - root_mean_squared_error: 7.2289\n",
      "Epoch 75/400\n",
      "1/1 [==============================] - 0s 824us/step - loss: 51.1872 - root_mean_squared_error: 7.1545\n",
      "Epoch 76/400\n",
      "1/1 [==============================] - 0s 951us/step - loss: 50.1290 - root_mean_squared_error: 7.0802\n",
      "Epoch 77/400\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 49.0832 - root_mean_squared_error: 7.0059\n",
      "Epoch 78/400\n",
      "1/1 [==============================] - 0s 795us/step - loss: 48.0496 - root_mean_squared_error: 6.9318\n",
      "Epoch 79/400\n",
      "1/1 [==============================] - 0s 739us/step - loss: 47.0282 - root_mean_squared_error: 6.8577\n",
      "Epoch 80/400\n",
      "1/1 [==============================] - 0s 720us/step - loss: 46.0191 - root_mean_squared_error: 6.7837\n",
      "Epoch 81/400\n",
      "1/1 [==============================] - 0s 788us/step - loss: 45.0221 - root_mean_squared_error: 6.7099\n",
      "Epoch 82/400\n",
      "1/1 [==============================] - 0s 982us/step - loss: 44.0374 - root_mean_squared_error: 6.6361\n",
      "Epoch 83/400\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 43.0648 - root_mean_squared_error: 6.5624\n",
      "Epoch 84/400\n",
      "1/1 [==============================] - 0s 937us/step - loss: 42.1044 - root_mean_squared_error: 6.4888\n",
      "Epoch 85/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 41.1560 - root_mean_squared_error: 6.4153\n",
      "Epoch 86/400\n",
      "1/1 [==============================] - 0s 724us/step - loss: 40.2198 - root_mean_squared_error: 6.3419\n",
      "Epoch 87/400\n",
      "1/1 [==============================] - 0s 701us/step - loss: 39.2957 - root_mean_squared_error: 6.2686\n",
      "Epoch 88/400\n",
      "1/1 [==============================] - 0s 795us/step - loss: 38.3837 - root_mean_squared_error: 6.1955\n",
      "Epoch 89/400\n",
      "1/1 [==============================] - 0s 938us/step - loss: 37.4837 - root_mean_squared_error: 6.1224\n",
      "Epoch 90/400\n",
      "1/1 [==============================] - 0s 683us/step - loss: 36.5957 - root_mean_squared_error: 6.0494\n",
      "Epoch 91/400\n",
      "1/1 [==============================] - 0s 895us/step - loss: 35.7198 - root_mean_squared_error: 5.9766\n",
      "Epoch 92/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 34.8558 - root_mean_squared_error: 5.9039\n",
      "Epoch 93/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 34.0039 - root_mean_squared_error: 5.8313\n",
      "Epoch 94/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 33.1639 - root_mean_squared_error: 5.7588\n",
      "Epoch 95/400\n",
      "1/1 [==============================] - 0s 997us/step - loss: 32.3358 - root_mean_squared_error: 5.6865\n",
      "Epoch 96/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 31.5196 - root_mean_squared_error: 5.6142\n",
      "Epoch 97/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 30.7153 - root_mean_squared_error: 5.5421\n",
      "Epoch 98/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 29.9229 - root_mean_squared_error: 5.4702\n",
      "Epoch 99/400\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 29.1423 - root_mean_squared_error: 5.3984\n",
      "Epoch 100/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 28.3736 - root_mean_squared_error: 5.3267\n",
      "Epoch 101/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 27.6166 - root_mean_squared_error: 5.2552\n",
      "Epoch 102/400\n",
      "1/1 [==============================] - 0s 818us/step - loss: 26.8714 - root_mean_squared_error: 5.1838\n",
      "Epoch 103/400\n",
      "1/1 [==============================] - 0s 766us/step - loss: 26.1380 - root_mean_squared_error: 5.1125\n",
      "Epoch 104/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 25.4163 - root_mean_squared_error: 5.0415\n",
      "Epoch 105/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 24.7063 - root_mean_squared_error: 4.9705\n",
      "Epoch 106/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 24.0080 - root_mean_squared_error: 4.8998\n",
      "Epoch 107/400\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 23.3213 - root_mean_squared_error: 4.8292\n",
      "Epoch 108/400\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 22.6462 - root_mean_squared_error: 4.7588\n",
      "Epoch 109/400\n",
      "1/1 [==============================] - 0s 850us/step - loss: 21.9827 - root_mean_squared_error: 4.6886\n",
      "Epoch 110/400\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 21.3308 - root_mean_squared_error: 4.6185\n",
      "Epoch 111/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 20.6904 - root_mean_squared_error: 4.5487\n",
      "Epoch 112/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 20.0615 - root_mean_squared_error: 4.4790\n",
      "Epoch 113/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 19.4441 - root_mean_squared_error: 4.4095\n",
      "Epoch 114/400\n",
      "1/1 [==============================] - 0s 912us/step - loss: 18.8381 - root_mean_squared_error: 4.3403\n",
      "Epoch 115/400\n",
      "1/1 [==============================] - 0s 990us/step - loss: 18.2435 - root_mean_squared_error: 4.2712\n",
      "Epoch 116/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 17.6603 - root_mean_squared_error: 4.2024\n",
      "Epoch 117/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 17.0884 - root_mean_squared_error: 4.1338\n",
      "Epoch 118/400\n",
      "1/1 [==============================] - 0s 688us/step - loss: 16.5278 - root_mean_squared_error: 4.0654\n",
      "Epoch 119/400\n",
      "1/1 [==============================] - 0s 722us/step - loss: 15.9784 - root_mean_squared_error: 3.9973\n",
      "Epoch 120/400\n",
      "1/1 [==============================] - 0s 656us/step - loss: 15.4403 - root_mean_squared_error: 3.9294\n",
      "Epoch 121/400\n",
      "1/1 [==============================] - 0s 658us/step - loss: 14.9134 - root_mean_squared_error: 3.8618\n",
      "Epoch 122/400\n",
      "1/1 [==============================] - 0s 681us/step - loss: 14.3976 - root_mean_squared_error: 3.7944\n",
      "Epoch 123/400\n",
      "1/1 [==============================] - 0s 683us/step - loss: 13.8929 - root_mean_squared_error: 3.7273\n",
      "Epoch 124/400\n",
      "1/1 [==============================] - 0s 676us/step - loss: 13.3993 - root_mean_squared_error: 3.6605\n",
      "Epoch 125/400\n",
      "1/1 [==============================] - 0s 679us/step - loss: 12.9167 - root_mean_squared_error: 3.5940\n",
      "Epoch 126/400\n",
      "1/1 [==============================] - 0s 735us/step - loss: 12.4451 - root_mean_squared_error: 3.5278\n",
      "Epoch 127/400\n",
      "1/1 [==============================] - 0s 638us/step - loss: 11.9843 - root_mean_squared_error: 3.4618\n",
      "Epoch 128/400\n",
      "1/1 [==============================] - 0s 960us/step - loss: 11.5345 - root_mean_squared_error: 3.3962\n",
      "Epoch 129/400\n",
      "1/1 [==============================] - 0s 916us/step - loss: 11.0955 - root_mean_squared_error: 3.3310\n",
      "Epoch 130/400\n",
      "1/1 [==============================] - 0s 911us/step - loss: 10.6672 - root_mean_squared_error: 3.2661\n",
      "Epoch 131/400\n",
      "1/1 [==============================] - 0s 691us/step - loss: 10.2497 - root_mean_squared_error: 3.2015\n",
      "Epoch 132/400\n",
      "1/1 [==============================] - 0s 672us/step - loss: 9.8428 - root_mean_squared_error: 3.1373\n",
      "Epoch 133/400\n",
      "1/1 [==============================] - 0s 731us/step - loss: 9.4466 - root_mean_squared_error: 3.0735\n",
      "Epoch 134/400\n",
      "1/1 [==============================] - 0s 838us/step - loss: 9.0609 - root_mean_squared_error: 3.0101\n",
      "Epoch 135/400\n",
      "1/1 [==============================] - 0s 803us/step - loss: 8.6857 - root_mean_squared_error: 2.9471\n",
      "Epoch 136/400\n",
      "1/1 [==============================] - 0s 712us/step - loss: 8.3209 - root_mean_squared_error: 2.8846\n",
      "Epoch 137/400\n",
      "1/1 [==============================] - 0s 767us/step - loss: 7.9665 - root_mean_squared_error: 2.8225\n",
      "Epoch 138/400\n",
      "1/1 [==============================] - 0s 740us/step - loss: 7.6224 - root_mean_squared_error: 2.7609\n",
      "Epoch 139/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 7.2885 - root_mean_squared_error: 2.6997\n",
      "Epoch 140/400\n",
      "1/1 [==============================] - 0s 818us/step - loss: 6.9648 - root_mean_squared_error: 2.6391\n",
      "Epoch 141/400\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 778us/step - loss: 6.6511 - root_mean_squared_error: 2.5790\n",
      "Epoch 142/400\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 6.3475 - root_mean_squared_error: 2.5194\n",
      "Epoch 143/400\n",
      "1/1 [==============================] - 0s 934us/step - loss: 6.0538 - root_mean_squared_error: 2.4604\n",
      "Epoch 144/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 5.7699 - root_mean_squared_error: 2.4021\n",
      "Epoch 145/400\n",
      "1/1 [==============================] - 0s 946us/step - loss: 5.4958 - root_mean_squared_error: 2.3443\n",
      "Epoch 146/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 5.2313 - root_mean_squared_error: 2.2872\n",
      "Epoch 147/400\n",
      "1/1 [==============================] - 0s 874us/step - loss: 4.9764 - root_mean_squared_error: 2.2308\n",
      "Epoch 148/400\n",
      "1/1 [==============================] - 0s 854us/step - loss: 4.7311 - root_mean_squared_error: 2.1751\n",
      "Epoch 149/400\n",
      "1/1 [==============================] - 0s 828us/step - loss: 4.4950 - root_mean_squared_error: 2.1202\n",
      "Epoch 150/400\n",
      "1/1 [==============================] - 0s 855us/step - loss: 4.2683 - root_mean_squared_error: 2.0660\n",
      "Epoch 151/400\n",
      "1/1 [==============================] - 0s 874us/step - loss: 4.0508 - root_mean_squared_error: 2.0126\n",
      "Epoch 152/400\n",
      "1/1 [==============================] - 0s 752us/step - loss: 3.8423 - root_mean_squared_error: 1.9602\n",
      "Epoch 153/400\n",
      "1/1 [==============================] - 0s 839us/step - loss: 3.6427 - root_mean_squared_error: 1.9086\n",
      "Epoch 154/400\n",
      "1/1 [==============================] - 0s 919us/step - loss: 3.4520 - root_mean_squared_error: 1.8579\n",
      "Epoch 155/400\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 3.2699 - root_mean_squared_error: 1.8083\n",
      "Epoch 156/400\n",
      "1/1 [==============================] - 0s 711us/step - loss: 3.0964 - root_mean_squared_error: 1.7597\n",
      "Epoch 157/400\n",
      "1/1 [==============================] - 0s 739us/step - loss: 2.9314 - root_mean_squared_error: 1.7121\n",
      "Epoch 158/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 2.7746 - root_mean_squared_error: 1.6657\n",
      "Epoch 159/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 2.6260 - root_mean_squared_error: 1.6205\n",
      "Epoch 160/400\n",
      "1/1 [==============================] - 0s 933us/step - loss: 2.4854 - root_mean_squared_error: 1.5765\n",
      "Epoch 161/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 2.3525 - root_mean_squared_error: 1.5338\n",
      "Epoch 162/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 2.2274 - root_mean_squared_error: 1.4924\n",
      "Epoch 163/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 2.1097 - root_mean_squared_error: 1.4525\n",
      "Epoch 164/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.9993 - root_mean_squared_error: 1.4140\n",
      "Epoch 165/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.8961 - root_mean_squared_error: 1.3770\n",
      "Epoch 166/400\n",
      "1/1 [==============================] - 0s 926us/step - loss: 1.7998 - root_mean_squared_error: 1.3416\n",
      "Epoch 167/400\n",
      "1/1 [==============================] - 0s 691us/step - loss: 1.7102 - root_mean_squared_error: 1.3077\n",
      "Epoch 168/400\n",
      "1/1 [==============================] - 0s 843us/step - loss: 1.6271 - root_mean_squared_error: 1.2756\n",
      "Epoch 169/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.5503 - root_mean_squared_error: 1.2451\n",
      "Epoch 170/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.4795 - root_mean_squared_error: 1.2164\n",
      "Epoch 171/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.4147 - root_mean_squared_error: 1.1894\n",
      "Epoch 172/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.3553 - root_mean_squared_error: 1.1642\n",
      "Epoch 173/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.3014 - root_mean_squared_error: 1.1408\n",
      "Epoch 174/400\n",
      "1/1 [==============================] - 0s 906us/step - loss: 1.2525 - root_mean_squared_error: 1.1192\n",
      "Epoch 175/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.2085 - root_mean_squared_error: 1.0993\n",
      "Epoch 176/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.1691 - root_mean_squared_error: 1.0812\n",
      "Epoch 177/400\n",
      "1/1 [==============================] - 0s 894us/step - loss: 1.1339 - root_mean_squared_error: 1.0648\n",
      "Epoch 178/400\n",
      "1/1 [==============================] - 0s 913us/step - loss: 1.1028 - root_mean_squared_error: 1.0501\n",
      "Epoch 179/400\n",
      "1/1 [==============================] - 0s 944us/step - loss: 1.0753 - root_mean_squared_error: 1.0370\n",
      "Epoch 180/400\n",
      "1/1 [==============================] - 0s 875us/step - loss: 1.0514 - root_mean_squared_error: 1.0254\n",
      "Epoch 181/400\n",
      "1/1 [==============================] - 0s 865us/step - loss: 1.0306 - root_mean_squared_error: 1.0152\n",
      "Epoch 182/400\n",
      "1/1 [==============================] - 0s 852us/step - loss: 1.0128 - root_mean_squared_error: 1.0064\n",
      "Epoch 183/400\n",
      "1/1 [==============================] - 0s 872us/step - loss: 0.9975 - root_mean_squared_error: 0.9988\n",
      "Epoch 184/400\n",
      "1/1 [==============================] - 0s 824us/step - loss: 0.9846 - root_mean_squared_error: 0.9923\n",
      "Epoch 185/400\n",
      "1/1 [==============================] - 0s 775us/step - loss: 0.9738 - root_mean_squared_error: 0.9868\n",
      "Epoch 186/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.9649 - root_mean_squared_error: 0.9823\n",
      "Epoch 187/400\n",
      "1/1 [==============================] - 0s 766us/step - loss: 0.9575 - root_mean_squared_error: 0.9785\n",
      "Epoch 188/400\n",
      "1/1 [==============================] - 0s 714us/step - loss: 0.9516 - root_mean_squared_error: 0.9755\n",
      "Epoch 189/400\n",
      "1/1 [==============================] - 0s 727us/step - loss: 0.9467 - root_mean_squared_error: 0.9730\n",
      "Epoch 190/400\n",
      "1/1 [==============================] - 0s 747us/step - loss: 0.9429 - root_mean_squared_error: 0.9710\n",
      "Epoch 191/400\n",
      "1/1 [==============================] - 0s 930us/step - loss: 0.9398 - root_mean_squared_error: 0.9694\n",
      "Epoch 192/400\n",
      "1/1 [==============================] - 0s 882us/step - loss: 0.9374 - root_mean_squared_error: 0.9682\n",
      "Epoch 193/400\n",
      "1/1 [==============================] - 0s 767us/step - loss: 0.9355 - root_mean_squared_error: 0.9672\n",
      "Epoch 194/400\n",
      "1/1 [==============================] - 0s 743us/step - loss: 0.9340 - root_mean_squared_error: 0.9664\n",
      "Epoch 195/400\n",
      "1/1 [==============================] - 0s 701us/step - loss: 0.9327 - root_mean_squared_error: 0.9658\n",
      "Epoch 196/400\n",
      "1/1 [==============================] - 0s 921us/step - loss: 0.9317 - root_mean_squared_error: 0.9652\n",
      "Epoch 197/400\n",
      "1/1 [==============================] - 0s 855us/step - loss: 0.9308 - root_mean_squared_error: 0.9648\n",
      "Epoch 198/400\n",
      "1/1 [==============================] - 0s 704us/step - loss: 0.9301 - root_mean_squared_error: 0.9644\n",
      "Epoch 199/400\n",
      "1/1 [==============================] - 0s 754us/step - loss: 0.9294 - root_mean_squared_error: 0.9640\n",
      "Epoch 200/400\n",
      "1/1 [==============================] - 0s 653us/step - loss: 0.9287 - root_mean_squared_error: 0.9637\n",
      "Epoch 201/400\n",
      "1/1 [==============================] - 0s 726us/step - loss: 0.9280 - root_mean_squared_error: 0.9633\n",
      "Epoch 202/400\n",
      "1/1 [==============================] - 0s 698us/step - loss: 0.9274 - root_mean_squared_error: 0.9630\n",
      "Epoch 203/400\n",
      "1/1 [==============================] - 0s 734us/step - loss: 0.9267 - root_mean_squared_error: 0.9626\n",
      "Epoch 204/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.9260 - root_mean_squared_error: 0.9623\n",
      "Epoch 205/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.9253 - root_mean_squared_error: 0.9619\n",
      "Epoch 206/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.9246 - root_mean_squared_error: 0.9615\n",
      "Epoch 207/400\n",
      "1/1 [==============================] - 0s 869us/step - loss: 0.9238 - root_mean_squared_error: 0.9612\n",
      "Epoch 208/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.9231 - root_mean_squared_error: 0.9608\n",
      "Epoch 209/400\n",
      "1/1 [==============================] - 0s 752us/step - loss: 0.9223 - root_mean_squared_error: 0.9603\n",
      "Epoch 210/400\n",
      "1/1 [==============================] - 0s 736us/step - loss: 0.9215 - root_mean_squared_error: 0.9599\n",
      "Epoch 211/400\n",
      "1/1 [==============================] - 0s 867us/step - loss: 0.9206 - root_mean_squared_error: 0.9595\n",
      "Epoch 212/400\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 797us/step - loss: 0.9198 - root_mean_squared_error: 0.9590\n",
      "Epoch 213/400\n",
      "1/1 [==============================] - 0s 812us/step - loss: 0.9189 - root_mean_squared_error: 0.9586\n",
      "Epoch 214/400\n",
      "1/1 [==============================] - 0s 767us/step - loss: 0.9180 - root_mean_squared_error: 0.9581\n",
      "Epoch 215/400\n",
      "1/1 [==============================] - 0s 758us/step - loss: 0.9170 - root_mean_squared_error: 0.9576\n",
      "Epoch 216/400\n",
      "1/1 [==============================] - 0s 758us/step - loss: 0.9161 - root_mean_squared_error: 0.9571\n",
      "Epoch 217/400\n",
      "1/1 [==============================] - 0s 733us/step - loss: 0.9152 - root_mean_squared_error: 0.9566\n",
      "Epoch 218/400\n",
      "1/1 [==============================] - 0s 859us/step - loss: 0.9142 - root_mean_squared_error: 0.9561\n",
      "Epoch 219/400\n",
      "1/1 [==============================] - 0s 762us/step - loss: 0.9132 - root_mean_squared_error: 0.9556\n",
      "Epoch 220/400\n",
      "1/1 [==============================] - 0s 815us/step - loss: 0.9122 - root_mean_squared_error: 0.9551\n",
      "Epoch 221/400\n",
      "1/1 [==============================] - 0s 828us/step - loss: 0.9112 - root_mean_squared_error: 0.9546\n",
      "Epoch 222/400\n",
      "1/1 [==============================] - 0s 874us/step - loss: 0.9102 - root_mean_squared_error: 0.9540\n",
      "Epoch 223/400\n",
      "1/1 [==============================] - 0s 804us/step - loss: 0.9091 - root_mean_squared_error: 0.9535\n",
      "Epoch 224/400\n",
      "1/1 [==============================] - 0s 760us/step - loss: 0.9081 - root_mean_squared_error: 0.9529\n",
      "Epoch 225/400\n",
      "1/1 [==============================] - 0s 789us/step - loss: 0.9071 - root_mean_squared_error: 0.9524\n",
      "Epoch 226/400\n",
      "1/1 [==============================] - 0s 952us/step - loss: 0.9060 - root_mean_squared_error: 0.9519\n",
      "Epoch 227/400\n",
      "1/1 [==============================] - 0s 755us/step - loss: 0.9050 - root_mean_squared_error: 0.9513\n",
      "Epoch 228/400\n",
      "1/1 [==============================] - 0s 757us/step - loss: 0.9040 - root_mean_squared_error: 0.9508\n",
      "Epoch 229/400\n",
      "1/1 [==============================] - 0s 778us/step - loss: 0.9029 - root_mean_squared_error: 0.9502\n",
      "Epoch 230/400\n",
      "1/1 [==============================] - 0s 693us/step - loss: 0.9019 - root_mean_squared_error: 0.9497\n",
      "Epoch 231/400\n",
      "1/1 [==============================] - 0s 691us/step - loss: 0.9009 - root_mean_squared_error: 0.9491\n",
      "Epoch 232/400\n",
      "1/1 [==============================] - 0s 734us/step - loss: 0.8999 - root_mean_squared_error: 0.9486\n",
      "Epoch 233/400\n",
      "1/1 [==============================] - 0s 720us/step - loss: 0.8989 - root_mean_squared_error: 0.9481\n",
      "Epoch 234/400\n",
      "1/1 [==============================] - 0s 669us/step - loss: 0.8979 - root_mean_squared_error: 0.9476\n",
      "Epoch 235/400\n",
      "1/1 [==============================] - 0s 649us/step - loss: 0.8969 - root_mean_squared_error: 0.9470\n",
      "Epoch 236/400\n",
      "1/1 [==============================] - 0s 668us/step - loss: 0.8959 - root_mean_squared_error: 0.9465\n",
      "Epoch 237/400\n",
      "1/1 [==============================] - 0s 822us/step - loss: 0.8950 - root_mean_squared_error: 0.9460\n",
      "Epoch 238/400\n",
      "1/1 [==============================] - 0s 809us/step - loss: 0.8941 - root_mean_squared_error: 0.9455\n",
      "Epoch 239/400\n",
      "1/1 [==============================] - 0s 747us/step - loss: 0.8931 - root_mean_squared_error: 0.9451\n",
      "Epoch 240/400\n",
      "1/1 [==============================] - 0s 865us/step - loss: 0.8923 - root_mean_squared_error: 0.9446\n",
      "Epoch 241/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8914 - root_mean_squared_error: 0.9441\n",
      "Epoch 242/400\n",
      "1/1 [==============================] - 0s 987us/step - loss: 0.8905 - root_mean_squared_error: 0.9437\n",
      "Epoch 243/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8897 - root_mean_squared_error: 0.9432\n",
      "Epoch 244/400\n",
      "1/1 [==============================] - 0s 795us/step - loss: 0.8889 - root_mean_squared_error: 0.9428\n",
      "Epoch 245/400\n",
      "1/1 [==============================] - 0s 691us/step - loss: 0.8881 - root_mean_squared_error: 0.9424\n",
      "Epoch 246/400\n",
      "1/1 [==============================] - 0s 757us/step - loss: 0.8874 - root_mean_squared_error: 0.9420\n",
      "Epoch 247/400\n",
      "1/1 [==============================] - 0s 709us/step - loss: 0.8866 - root_mean_squared_error: 0.9416\n",
      "Epoch 248/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8859 - root_mean_squared_error: 0.9412\n",
      "Epoch 249/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8853 - root_mean_squared_error: 0.9409\n",
      "Epoch 250/400\n",
      "1/1 [==============================] - 0s 865us/step - loss: 0.8846 - root_mean_squared_error: 0.9405\n",
      "Epoch 251/400\n",
      "1/1 [==============================] - 0s 696us/step - loss: 0.8840 - root_mean_squared_error: 0.9402\n",
      "Epoch 252/400\n",
      "1/1 [==============================] - 0s 947us/step - loss: 0.8834 - root_mean_squared_error: 0.9399\n",
      "Epoch 253/400\n",
      "1/1 [==============================] - 0s 913us/step - loss: 0.8828 - root_mean_squared_error: 0.9396\n",
      "Epoch 254/400\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 0.8822 - root_mean_squared_error: 0.9393\n",
      "Epoch 255/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8817 - root_mean_squared_error: 0.9390\n",
      "Epoch 256/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8816 - root_mean_squared_error: 0.9389\n",
      "Epoch 257/400\n",
      "1/1 [==============================] - 0s 923us/step - loss: 0.8852 - root_mean_squared_error: 0.9408\n",
      "Epoch 258/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8950 - root_mean_squared_error: 0.9460\n",
      "Epoch 259/400\n",
      "1/1 [==============================] - 0s 912us/step - loss: 0.8878 - root_mean_squared_error: 0.9422\n",
      "Epoch 260/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8819 - root_mean_squared_error: 0.9391\n",
      "Epoch 261/400\n",
      "1/1 [==============================] - 0s 943us/step - loss: 0.8801 - root_mean_squared_error: 0.9381\n",
      "Epoch 262/400\n",
      "1/1 [==============================] - 0s 912us/step - loss: 0.8794 - root_mean_squared_error: 0.9378\n",
      "Epoch 263/400\n",
      "1/1 [==============================] - 0s 845us/step - loss: 0.8790 - root_mean_squared_error: 0.9375\n",
      "Epoch 264/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8787 - root_mean_squared_error: 0.9374\n",
      "Epoch 265/400\n",
      "1/1 [==============================] - 0s 985us/step - loss: 0.8784 - root_mean_squared_error: 0.9372\n",
      "Epoch 266/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8782 - root_mean_squared_error: 0.9371\n",
      "Epoch 267/400\n",
      "1/1 [==============================] - 0s 961us/step - loss: 0.8779 - root_mean_squared_error: 0.9370\n",
      "Epoch 268/400\n",
      "1/1 [==============================] - 0s 974us/step - loss: 0.8777 - root_mean_squared_error: 0.9369\n",
      "Epoch 269/400\n",
      "1/1 [==============================] - 0s 976us/step - loss: 0.8776 - root_mean_squared_error: 0.9368\n",
      "Epoch 270/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8775 - root_mean_squared_error: 0.9367\n",
      "Epoch 271/400\n",
      "1/1 [==============================] - 0s 700us/step - loss: 0.8776 - root_mean_squared_error: 0.9368\n",
      "Epoch 272/400\n",
      "1/1 [==============================] - 0s 872us/step - loss: 0.8781 - root_mean_squared_error: 0.9371\n",
      "Epoch 273/400\n",
      "1/1 [==============================] - 0s 802us/step - loss: 0.8795 - root_mean_squared_error: 0.9378\n",
      "Epoch 274/400\n",
      "1/1 [==============================] - 0s 991us/step - loss: 0.8815 - root_mean_squared_error: 0.9389\n",
      "Epoch 275/400\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 0.8819 - root_mean_squared_error: 0.9391\n",
      "Epoch 276/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8801 - root_mean_squared_error: 0.9382\n",
      "Epoch 277/400\n",
      "1/1 [==============================] - 0s 879us/step - loss: 0.8785 - root_mean_squared_error: 0.9373\n",
      "Epoch 278/400\n",
      "1/1 [==============================] - 0s 860us/step - loss: 0.8774 - root_mean_squared_error: 0.9367\n",
      "Epoch 279/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8769 - root_mean_squared_error: 0.9364\n",
      "Epoch 280/400\n",
      "1/1 [==============================] - 0s 796us/step - loss: 0.8766 - root_mean_squared_error: 0.9363\n",
      "Epoch 281/400\n",
      "1/1 [==============================] - 0s 730us/step - loss: 0.8764 - root_mean_squared_error: 0.9362\n",
      "Epoch 282/400\n",
      "1/1 [==============================] - 0s 638us/step - loss: 0.8764 - root_mean_squared_error: 0.9361\n",
      "Epoch 283/400\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 851us/step - loss: 0.8764 - root_mean_squared_error: 0.9361\n",
      "Epoch 284/400\n",
      "1/1 [==============================] - 0s 661us/step - loss: 0.8765 - root_mean_squared_error: 0.9362\n",
      "Epoch 285/400\n",
      "1/1 [==============================] - 0s 667us/step - loss: 0.8767 - root_mean_squared_error: 0.9363\n",
      "Epoch 286/400\n",
      "1/1 [==============================] - 0s 822us/step - loss: 0.8772 - root_mean_squared_error: 0.9366\n",
      "Epoch 287/400\n",
      "1/1 [==============================] - 0s 794us/step - loss: 0.8778 - root_mean_squared_error: 0.9369\n",
      "Epoch 288/400\n",
      "1/1 [==============================] - 0s 763us/step - loss: 0.8785 - root_mean_squared_error: 0.9373\n",
      "Epoch 289/400\n",
      "1/1 [==============================] - 0s 973us/step - loss: 0.8787 - root_mean_squared_error: 0.9374\n",
      "Epoch 290/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8783 - root_mean_squared_error: 0.9372\n",
      "Epoch 291/400\n",
      "1/1 [==============================] - 0s 771us/step - loss: 0.8777 - root_mean_squared_error: 0.9369\n",
      "Epoch 292/400\n",
      "1/1 [==============================] - 0s 676us/step - loss: 0.8771 - root_mean_squared_error: 0.9366\n",
      "Epoch 293/400\n",
      "1/1 [==============================] - 0s 665us/step - loss: 0.8767 - root_mean_squared_error: 0.9363\n",
      "Epoch 294/400\n",
      "1/1 [==============================] - 0s 770us/step - loss: 0.8765 - root_mean_squared_error: 0.9362\n",
      "Epoch 295/400\n",
      "1/1 [==============================] - 0s 848us/step - loss: 0.8763 - root_mean_squared_error: 0.9361\n",
      "Epoch 296/400\n",
      "1/1 [==============================] - 0s 767us/step - loss: 0.8763 - root_mean_squared_error: 0.9361\n",
      "Epoch 297/400\n",
      "1/1 [==============================] - 0s 696us/step - loss: 0.8764 - root_mean_squared_error: 0.9361\n",
      "Epoch 298/400\n",
      "1/1 [==============================] - 0s 880us/step - loss: 0.8765 - root_mean_squared_error: 0.9362\n",
      "Epoch 299/400\n",
      "1/1 [==============================] - 0s 977us/step - loss: 0.8767 - root_mean_squared_error: 0.9363\n",
      "Epoch 300/400\n",
      "1/1 [==============================] - 0s 807us/step - loss: 0.8770 - root_mean_squared_error: 0.9365\n",
      "Epoch 301/400\n",
      "1/1 [==============================] - 0s 990us/step - loss: 0.8773 - root_mean_squared_error: 0.9367\n",
      "Epoch 302/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8775 - root_mean_squared_error: 0.9368\n",
      "Epoch 303/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8776 - root_mean_squared_error: 0.9368\n",
      "Epoch 304/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8774 - root_mean_squared_error: 0.9367\n",
      "Epoch 305/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8772 - root_mean_squared_error: 0.9366\n",
      "Epoch 306/400\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 0.8769 - root_mean_squared_error: 0.9364\n",
      "Epoch 307/400\n",
      "1/1 [==============================] - 0s 802us/step - loss: 0.8767 - root_mean_squared_error: 0.9363\n",
      "Epoch 308/400\n",
      "1/1 [==============================] - 0s 717us/step - loss: 0.8766 - root_mean_squared_error: 0.9363\n",
      "Epoch 309/400\n",
      "1/1 [==============================] - 0s 665us/step - loss: 0.8765 - root_mean_squared_error: 0.9362\n",
      "Epoch 310/400\n",
      "1/1 [==============================] - 0s 689us/step - loss: 0.8765 - root_mean_squared_error: 0.9362\n",
      "Epoch 311/400\n",
      "1/1 [==============================] - 0s 643us/step - loss: 0.8766 - root_mean_squared_error: 0.9362\n",
      "Epoch 312/400\n",
      "1/1 [==============================] - 0s 697us/step - loss: 0.8767 - root_mean_squared_error: 0.9363\n",
      "Epoch 313/400\n",
      "1/1 [==============================] - 0s 633us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 314/400\n",
      "1/1 [==============================] - 0s 690us/step - loss: 0.8770 - root_mean_squared_error: 0.9365\n",
      "Epoch 315/400\n",
      "1/1 [==============================] - 0s 785us/step - loss: 0.8771 - root_mean_squared_error: 0.9365\n",
      "Epoch 316/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8771 - root_mean_squared_error: 0.9366\n",
      "Epoch 317/400\n",
      "1/1 [==============================] - 0s 653us/step - loss: 0.8771 - root_mean_squared_error: 0.9366\n",
      "Epoch 318/400\n",
      "1/1 [==============================] - 0s 938us/step - loss: 0.8771 - root_mean_squared_error: 0.9365\n",
      "Epoch 319/400\n",
      "1/1 [==============================] - 0s 871us/step - loss: 0.8769 - root_mean_squared_error: 0.9365\n",
      "Epoch 320/400\n",
      "1/1 [==============================] - 0s 755us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 321/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8767 - root_mean_squared_error: 0.9363\n",
      "Epoch 322/400\n",
      "1/1 [==============================] - 0s 763us/step - loss: 0.8767 - root_mean_squared_error: 0.9363\n",
      "Epoch 323/400\n",
      "1/1 [==============================] - 0s 715us/step - loss: 0.8766 - root_mean_squared_error: 0.9363\n",
      "Epoch 324/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8767 - root_mean_squared_error: 0.9363\n",
      "Epoch 325/400\n",
      "1/1 [==============================] - 0s 971us/step - loss: 0.8767 - root_mean_squared_error: 0.9363\n",
      "Epoch 326/400\n",
      "1/1 [==============================] - 0s 751us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 327/400\n",
      "1/1 [==============================] - 0s 783us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 328/400\n",
      "1/1 [==============================] - 0s 678us/step - loss: 0.8769 - root_mean_squared_error: 0.9364\n",
      "Epoch 329/400\n",
      "1/1 [==============================] - 0s 639us/step - loss: 0.8770 - root_mean_squared_error: 0.9365\n",
      "Epoch 330/400\n",
      "1/1 [==============================] - 0s 641us/step - loss: 0.8770 - root_mean_squared_error: 0.9365\n",
      "Epoch 331/400\n",
      "1/1 [==============================] - 0s 708us/step - loss: 0.8770 - root_mean_squared_error: 0.9365\n",
      "Epoch 332/400\n",
      "1/1 [==============================] - 0s 708us/step - loss: 0.8769 - root_mean_squared_error: 0.9364\n",
      "Epoch 333/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8769 - root_mean_squared_error: 0.9364\n",
      "Epoch 334/400\n",
      "1/1 [==============================] - 0s 720us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 335/400\n",
      "1/1 [==============================] - 0s 649us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 336/400\n",
      "1/1 [==============================] - 0s 713us/step - loss: 0.8767 - root_mean_squared_error: 0.9363\n",
      "Epoch 337/400\n",
      "1/1 [==============================] - 0s 714us/step - loss: 0.8767 - root_mean_squared_error: 0.9363\n",
      "Epoch 338/400\n",
      "1/1 [==============================] - 0s 676us/step - loss: 0.8767 - root_mean_squared_error: 0.9363\n",
      "Epoch 339/400\n",
      "1/1 [==============================] - 0s 699us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 340/400\n",
      "1/1 [==============================] - 0s 932us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 341/400\n",
      "1/1 [==============================] - 0s 994us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 342/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8769 - root_mean_squared_error: 0.9364\n",
      "Epoch 343/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8769 - root_mean_squared_error: 0.9364\n",
      "Epoch 344/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8769 - root_mean_squared_error: 0.9364\n",
      "Epoch 345/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8769 - root_mean_squared_error: 0.9364\n",
      "Epoch 346/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8769 - root_mean_squared_error: 0.9364\n",
      "Epoch 347/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 348/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 349/400\n",
      "1/1 [==============================] - 0s 922us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 350/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 351/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 352/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 353/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 354/400\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 2ms/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 355/400\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 356/400\n",
      "1/1 [==============================] - 0s 858us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 357/400\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 0.8769 - root_mean_squared_error: 0.9364\n",
      "Epoch 358/400\n",
      "1/1 [==============================] - 0s 929us/step - loss: 0.8769 - root_mean_squared_error: 0.9364\n",
      "Epoch 359/400\n",
      "1/1 [==============================] - 0s 939us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 360/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 361/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 362/400\n",
      "1/1 [==============================] - 0s 689us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 363/400\n",
      "1/1 [==============================] - 0s 784us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 364/400\n",
      "1/1 [==============================] - 0s 680us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 365/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 366/400\n",
      "1/1 [==============================] - 0s 901us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 367/400\n",
      "1/1 [==============================] - 0s 959us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 368/400\n",
      "1/1 [==============================] - 0s 686us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 369/400\n",
      "1/1 [==============================] - 0s 861us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 370/400\n",
      "1/1 [==============================] - 0s 814us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 371/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 372/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 373/400\n",
      "1/1 [==============================] - 0s 778us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 374/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 375/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 376/400\n",
      "1/1 [==============================] - 0s 950us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 377/400\n",
      "1/1 [==============================] - 0s 989us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 378/400\n",
      "1/1 [==============================] - 0s 774us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 379/400\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 380/400\n",
      "1/1 [==============================] - 0s 903us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 381/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 382/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 383/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 384/400\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 385/400\n",
      "1/1 [==============================] - 0s 865us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 386/400\n",
      "1/1 [==============================] - 0s 812us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 387/400\n",
      "1/1 [==============================] - 0s 941us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 388/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 389/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 390/400\n",
      "1/1 [==============================] - 0s 853us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 391/400\n",
      "1/1 [==============================] - 0s 759us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 392/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 393/400\n",
      "1/1 [==============================] - 0s 697us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 394/400\n",
      "1/1 [==============================] - 0s 820us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 395/400\n",
      "1/1 [==============================] - 0s 794us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 396/400\n",
      "1/1 [==============================] - 0s 840us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 397/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 398/400\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 399/400\n",
      "1/1 [==============================] - 0s 803us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 400/400\n",
      "1/1 [==============================] - 0s 935us/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/eric/.conda/envs/mlbase/lib/python3.8/site-packages/numpy/core/_asarray.py:136: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray\n",
      "  return array(a, dtype, copy=False, order=order, subok=True)\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAEKCAYAAAAVaT4rAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/d3fzzAAAACXBIWXMAAAsTAAALEwEAmpwYAAAlVElEQVR4nO3de3hV1Z3/8fdXTGtEaqSCApZqq4NCocBEpVjbolUs2pJa+0NmvFU0UnFUbGFAa6W2CgxeqNQiUaQ4YqdQuQ0OpjSKNlbESLjfvFExYSBYU7yEAcL398c6aMAgue2zz+Xzep485+x9zj77ex7xk5W1117L3B0REckeh8VdgIiIJJeCX0Qkyyj4RUSyjIJfRCTLKPhFRLKMgl9EJMtEHvxm1srMys1sQWK7rZktMrNXE4/HRF2DiIh8LBkt/puAdXW2RwEl7n4KUJLYFhGRJIk0+M3sBOBC4JE6uwcC0xPPpwMFUdYgIiL7Ozziz58IjATa1Nl3nLtvAXD3LWbWvr4DzawQKARo3br1P5966qkRlyoiklleeeWV7e7e7sD9kQW/mV0EbHP3V8zsW4093t2LgCKA/Px8Lysra9kCRUQynJn9rb79Ubb4zwK+Z2YDgCOAz5nZ48BWM+uQaO13ALZFWIOIiBwgsj5+dx/t7ie4+4nApcAz7n4ZMB+4MvG2K4F5UdUgIiKfFMc4/nHAeWb2KnBeYltERJIk6ou7ALj7YmBx4vk7wLnJOK+IiHyS7twVEckySWnxi4hI48wtr2BC8QYqq2vomJfLiP5dKOjVqUU+W8EvIpJi5pZXMHr2Kmp21wJQUV3D6NmrAFok/NXVIyKSYiYUb/go9Pep2V3LhOINLfL5Cn4RkRRTWV3TqP2NpeAXEUkxHfNyAfjSO2/Xu7+5FPwiIilm9Nc78ctnivjzIz/m26++BEBuTitG9O/SIp+vi7siIqlk4UIuGjoU37yZJ782kCWdu9NJo3pERDJQVRXcfDM88QScdhpWWsolfftySQSnUlePiEic3OHxx+G002DWLBgzBsrLoW/fyE6pFr+ISFw2bYKhQ6G4GPr0gUcegW7dIj+tWvwiIslWWwsTJ4aQf+EFmDQJSkuTEvqgFr+ISHKtWgXXXANLl8KAATB5MnTunNQS1OIXEUmGnTvh9tuhd294881wEXfBgqSHPqjFLyISvdJSuPZaWL8errgC7rsPPv/52MpRi19EJCo7dsD118PZZ4cWf3ExTJ8ea+iDgl9EJBrz50PXrjBlCtxyC6xeDeefH3dVgIJfRKRlbd0KgwbBwIHQti28+CLcey+0bh13ZR9R8IuItAR3mDYt3Ig1bx7cdRe88gqccUbclX1CZMFvZkeY2VIzW2Fma8zsF4n9Y8yswsyWJ34GRFWDiEhSvP46nHceXH01fOUrsGIF3Hor5OTEXVm9ohzV83/AOe7+vpnlAKVmtjDx2v3ufk+E5xYRid6ePeFGrJ//PIT8Qw+F0TuHpXZnSmTB7+4OvJ/YzEn8eFTnExFJquXLYcgQWLYs9Oc/+CB0apnZM6MW6a8lM2tlZsuBbcAid38p8dINZrbSzB41s2MOcmyhmZWZWVlVVVWUZYqINFxNDYwaBfn5UFERJlabMydtQh8iDn53r3X3nsAJwBlm9hVgMvBloCewBbj3IMcWuXu+u+e3a9cuyjJFRBrm2WehRw8YPx6uugrWrYNLLgGzuCtrlKR0RLl7NbAYuMDdtyZ+IewFHgZS75K3iEhd774b+u7POSeM3ikpCTNpHlNvh0XKi3JUTzszy0s8zwW+Daw3sw513vZ9YHVUNYiINNuTT4YbsaZNg5Ej+e/HnuaspXDSqKc4a9wzzC2viLvCRotyVE8HYLqZtSL8gpnp7gvM7D/NrCfhQu8m4LoIaxARaZrKSrjhhtB/36sXPPUUc+04Rs9eRc3uWgAqqmsYPXsVQIsti5gMUY7qWQn0qmf/5VGdU0Sk2fbuDd04I0bArl2hP/+WW+Dww5kw7pmPQn+fmt21TCjeoOAXEUlLGzdCYSE89xz06wdFRXDyyR+9XFldU+9hB9ufqlL7LgMRkWTYvRvGjg0jdlasgKlTwwXcOqEP0DEvt97DD7Y/VSn4RSS7lZWFMfm33grf+14Yonn11fUO0RzRvwu5Oa3225eb04oR/bskq9oWoeAXkez0wQfw05/CmWfC9u0wdy7MnAnHH3/QQwp6dWLsxd3plJeLAZ3ychl7cfe06t8H9fGLSDZatAiuuy4sgTh0KIwbB0cf3aBDC3p1SrugP5Ba/CKSPd55J9xxe/75YVK1558Pi503MPQzhYJfRDKfO/zhD+FGrBkz4LbbwkXcs8+Ou7JYqKtHRDLG3PIKJhRvoLK6ho55uYzo34WCY/eGdW8XLIDTTw/dPD16xF1qrNTiF5GMMLe8gtGzV1FRXYMDle9+wMpbx7L7tK7wzDNw331hGcQsD31Qi19EMsSE4g0f3VX75e2bGf/0A+RXrOOlU/I5s3gmnHRSzBWmDgW/iGSEyuoacmp38+Mlf2TYi3/gw5xchl94C3O79eNNhf5+FPwikhHO+8cb/OSP99Bl+1vMO+2b3HnutbzTOo9OaXZXbTIo+EUkvb33Htx2G1Om/IYtbY7lR5fcwbNfPh1Iz7tqk0HBLyLpa+HCcAPW5s3YsGEsG3QDG0srsLqjetL8ZqsoKPhFJP1UVcHNN8MTT8Bpp0FpKfTty0XARV9XC/9QNJxTRNKHOzz+eAj7WbNgzBgoL4e+feOuLK2oxS8i6WHTptCtU1wMffqExVK6dYu7qrSkFr+IpLbaWpg4MYT8Cy/ApEmha0eh32RRLrZ+hJktNbMVZrbGzH6R2N/WzBaZ2auJx/Rcpl5EordqVejGGT4cvvUtWLMmrIPbqtUhD5WDi7LF/3/AOe7+VaAncIGZ9QFGASXufgpQktgWEfnYzp1w++3Quze88Ua4iLtgAXTuHHdlGSGy4Pfg/cRmTuLHgYHA9MT+6UBBVDWISBoqLYVeveBXv4J/+ZewItbgwfWuiCVNE2kfv5m1MrPlwDZgkbu/BBzn7lsAEo/tD3JsoZmVmVlZVVVVlGWKSCrYsSPMonn22aHFX1wM06fDscfGXVnGiTT43b3W3XsCJwBnmNlXGnFskbvnu3t+u3btIqtRRFLA/PlhrvwpU+CWW2D16rBYikQiKaN63L0aWAxcAGw1sw4AicdtyahBRFLQ1q0waBAMHAht24Zpk++9F1q3jruyjBbZOH4zawfsdvdqM8sFvg2MB+YDVwLjEo/zoqpBRFLDJxZIOf+fKFixCH7yE/jwQ7jrLhgxIiyHKJGL8gauDsB0M2tF+MtiprsvMLMXgZlmNgR4C/hhhDWISMz2LZCyb678Vm++wXE/+AlsWh768x9+GLpomoVkiiz43X0l0Kue/e8A50Z1XhFJLfsWSGm1t5arX57HLaUz2H1YK/6j4GZGPnkvHKb7SJNNUzaISKQqq2vouvUNxi/8Nd23vs6fTunD7ecNZVubYxmp0I+Fgl9EolNTw51LZjD4+T/w7pGf48cDR7Gwy1lgpgVSYqTgF5FoPPssFBZy+Wuv8cee/bnzmz9ixxFHAVogJW76O0tEWta778K118I554RplEtKOPzRqbQ5vh0GdMrLZezF3bVASozU4heRlvPkk2EStaoqGDkS7rgDjjySAlDQpxAFv4g0X2VlCPw5c8I8O089FSZYk5Skrh4Rabq9e6GoKKyItXAhjB8PS5cq9FOcWvwi0jQbN0JhITz3HPTrF34BnHxy3FVJA6jFLyKNs3s3jB0LPXrAihUwdSqUlCj004ha/CJZ7hPz6PTvcvALsWVlMGQIrFwJl1wCDzwAHTokt2BpNrX4RbLYvnl0KqprcKCiuobRs1cxt7xi/zd+8AH89Kdw5pmwfXu4iDtrlkI/TSn4RbLYvnl06qrZXcuE4g0f71i0CLp3D9MlFxbC2rVQUJDcQqVFKfhFslhldc3B97/zDlx1VVgQJScHnn8eJk+Go49ObpHS4hT8IlmsY33z5bhz+VtLwopYM2bAbbeFi7hnn538AiUSurgrksVG9O+y31z5HXZUcfefJ9Pv1aVw+umhm6dHj5irlJam4BfJYvtG79yzcB39Fs9m1PPT+aw53Hcf3HgjtGoVc4USBQW/SJYrOGIHBU/dAX/9a+jPf+ghOOmkuMuSCKmPXyRb7doFd94JPXvC+vXw2GPw9NMK/SygFr9INlqyBK65BtasgcGDYeJEaN8+7qokSSJr8ZvZF8zsWTNbZ2ZrzOymxP4xZlZhZssTPwOiqkFEDvD++3DTTdC3L+zYAQsWwBNPKPSzTJQt/j3AT9x9mZm1AV4xs0WJ1+5393siPLeIHGjhQhg6FDZvhmHD4O67oU2buKuSGEQW/O6+BdiSeP6ema0DtBKDSLJVVcHNN4eW/WmnQWlpaPFL1krKxV0zOxHoBbyU2HWDma00s0fN7JiDHFNoZmVmVlZVVZWMMkUyizs8/ngI+1mzYMwYKC9X6Ev0wW9mRwFPAje7+w5gMvBloCfhL4J76zvO3YvcPd/d89u1axd1mSKZZdMm+M534PLL4ZRTQuDfcQd89rNxVyYpINLgN7McQujPcPfZAO6+1d1r3X0v8DBwRpQ1iGSV2towQqdbN3jhBZg0KXTtdOsWd2WSQiLr4zczA6YC69z9vjr7OyT6/wG+D6yOqgaRrLJqVRiiuXQpDBgQJlTr3DnuqiQFRTmq5yzgcmCVmS1P7LsVGGxmPQEHNgHXRViDSObbuRPuugvGjYO8vHAR99JLwSzuyiRFRTmqpxSo71/e/0R1TpGsU1oK114b7ry94oowZ/6xx8ZdlaQ4Tdkgko527IDrrw9TJdfUhKkWpk9X6EuDKPhF0s38+WGu/ClTYPhwWL0a+vePuypJIwp+kXSxdSsMGgQDB0LbtvDii2H65KOOirsySTMKfpFU5w7TpoUbsebNCxdyX3kFztBIaGkazc4pkspefx2uuw5KSkJ//sMPQ5cucVclaU4tfpFUtGcP3HMPdO8OL78cFkdZvFihLy1CLX6RVLN8OQwZAsuWhf78Bx+ETprfUFqOWvwiqaKmBkaPhvx8qKgIE6vNmaPQlxanFr9IKli8ONyI9dprobU/YQIcU+/EtSLNpha/SJyqq0Pg9+sXRu+UlMAjjyj0JVJq8YvEZfbssBJWVRWMHBmmTT7yyP3eMre8ggnFG6isrqFjXi4j+nehoJe6fqR5PjX4zeziT3t931TLItIIlZVwww2h/75XL3jqKejd+xNvm1tewejZq6jZXQtARXUNo2evAlD4S7McqsX/3U95zQEFv0hD7d0bunFGjIBdu2D8eLjlFji8/v8NJxRv+Cj096nZXcuE4g0KfmmWTw1+d/9RsgoRyWgbN0JhITz3XOjPLyqCk0/+1EMqq2satV+koRp0cdfMjjOzqWa2MLHd1cyGRFuaSAbYvRvGjoUePWDFCpg6NVzAPUToA3TMy23UfpGGauiont8BxUDHxPZG4OYI6hHJHGVlYUz+rbfCd78La9fC1Vc3eIGUEf27kJvTar99uTmtGNFfd+9K8zQ0+I9195nAXgB33wPUfvohIlnqgw/gpz+FM8+E7dvDRdxZs6BDh0Z9TEGvToy9uDud8nIxoFNeLmMv7q7+fWm2hg7n/MDMPk+4oIuZ9QH+EVlVIulq0aIwqdqbb4bH8ePh6KOb/HEFvTop6KXFNbTFfwswH/iymb0APAb826cdYGZfMLNnzWydma0xs5sS+9ua2SIzezXxqDtVJP298w5cdRWcfz7k5ISLuA891KzQF4lKg4Lf3ZcB3wT6EhZH7+buKw9x2B7gJ+5+GtAHGGZmXYFRQIm7nwKUJLZF0pM7/OEPYUWsGTPgttvCRdxvfCPuykQOqkFdPWZ2BHA98HVCd89fzOwhd995sGPcfQuwJfH8PTNbB3QCBgLfSrxtOrAY+Pcm1i8Sn82bw7q3CxbA6aeHbp4ePeKuSuSQGtrV8xjQDZgE/AboCvxnQ09iZicCvYCXgOMSvxT2/XJof5BjCs2szMzKqqqqGnoqkejt3cuK0XfzwSmn8mHxIh648MfM/e0fFfqSNhp6cbeLu3+1zvazZraiIQea2VHAk8DN7r7DGjiUzd2LgCKA/Px8b2CdItFat453Bl/BV1eU8fyJvbi1/zDezjue3HlroVUrXYiVtNDQFn95YiQPAGZ2JvDCoQ4ysxxC6M+oM6/PVjPrkHi9A7CtcSWLxGDXLrjzTujZk8M3bmD4hbdwxf+7k7fzjgc+nkpBJB0capK2VYQ+/RzgCjN7K7H9RWDtIY41YCqwzt3vq/PSfOBKYFzicV6TqxdJhiVL4JprYM0aGDyYcz9/Idtb533ibZpKQdLFobp6LmrGZ58FXA6sMrPliX23EgJ/ZmLKh7eAHzbjHCKf0GJTGb//fhilM2kSnHBCuIh74YV8dtwzUE/IayoFSReHmqTtb3W3zaw9cERDPtjdS4GDdeif26DqRBqpxaYyXrgQhg4NI3eGDYO774Y2bYAwlULdc4CmUpD00tBJ2r5nZq8CbwLPAZuAhRHWJdIknzaVcYNUVcFll8GAAdC6NZSWhhZ/IvRBUylI+mvoqJ5fEm7C+rO79zKzfsDg6MoSaZomT2XsHm7Auvlm2LEDxoyBUaPgs5+t9+2aSkHSWUNH9ex293eAw8zsMHd/FugZXVkiTdOkqYw3bYLvfAcuvxxOOQXKy8MyiAcJfZF019Dgr06Mx38emGFmvyZMySCSUho1lXFtLUycCN26wQsvhC6d0tKwLZLBGtrVMxDYCQwH/hU4GrgzqqJEmmpf98shR/WsWhWGaC5dGvrzJ0+Gzp1jqFgk+RoU/O7+QZ3N6RHVItIiPrX/fedOuOsuGDcO8vLgiSfg0ksbvDiKSCY41A1c75GYg//AlwB3989FUpVIFEpL4dprYf16uOIKuPdeOPbYuKsSSbpDjeNv82mvi6SFHTvCCJ3Jk+GLX4Snn4b+/eOuSiQ2Db24K5Ke5s8Pc+VPmQLDh8Pq1Qp9yXoKfslMW7fCoEEwcCC0bQsvvgj33QdHHRV3ZSKxU/BLZnGHadPgtNNg3rxwIfeVV+CMM+KuTCRlNHQ4p0jqe/31sMB5SQmcfTY8/DB00fw5IgdSi1/S3549cM890L07vPxyWOR88WKFvshBqMUv6W35chgyBJYtC/35Dz4InTSHjsinUYtf0lNNDYweDfn5UFEBs2bBnDkKfZEGUItf0s/ixeFGrNdeC639CRPgmGPirkokbajFL+mjujoEfr9+YfROSQk88ohCX6SRFPySHmbPDkM0p02DkSNh5Uo455y4qxJJS5EFv5k9ambbzGx1nX1jzKzCzJYnfgZEdX7JEJWVcPHF8IMfQIcOYTbN8ePhyCPjrkwkbUXZ4v8dcEE9++93956Jn/+J8PySzvbuhaKi0MpfuDCE/dKl0Lt33JWJpL3ILu66+/NmdmJUny8ZbONGKCyE554L/flFRXDyyXFXJZIx4ujjv8HMVia6gnRVTj62ezeMHQs9esCKFTB1ariAq9AXaVHJDv7JwJcJ6/VuAe492BvNrNDMysysrKqqKknlSWzKysKY/Ftvhe9+F9auhauv1gIpIhFI6jh+d9+677mZPQws+JT3FgFFAPn5+fUtBiNpaG55xX7LIo76xhf47uwpcP/9cPzx4SasgoK4yxTJaEkNfjPr4O5bEpvfB1Z/2vsls8wtr2D07FXU7K4F4KTyv9Jz/INQ/b9hcrXx4+Hoo2OuUiTzRRb8ZvZ74FvAsWb2NnAH8C0z60lYznETcF1U55fUM6F4AzW7a8mr2cHPnpnKJatLeL1tJ64vvJ/fPnRz3OWJZI0oR/UMrmf31KjOJ6mv8t0PuWj9X7jjz0Xk7XyPSV8bxG/6DmLX4Z+JuzSRrKK5eiQ5Nm/mP+ffxdfXL2F5h1O4fNAvWd/+JAA65eXGXJxIdlHwS7T27g3z448aRZ/dexh3XiFFPS9k72GtAMjNacWI/po3XySZNFePRGfdurAS1rBh8LWvcfjaNZw6/ud0aHsURmjpj724OwW9NJWySDKpxS8tb9cuGDcurHd71FHw2GNw2WVgRgEo6EVipuCXlrVkCVxzDaxZA4MHw8SJ0L593FWJSB3q6pGW8f77cNNN0Lcv7NgBCxbAE08o9EVSkFr80nwLF8LQobB5c+jPv/tuaNMm7qpE5CDU4pemq6oKffcDBkDr1lBaCpMmKfRFUpyCXxrPHR5/PMyVP3MmjBkD5eWhm0dEUp66eqRxNm0K3TrFxdCnT1jztlu3uKsSkUZQi18aprY2jNDp1g1eeCF06ZSWKvRF0pBa/HJoq1aFIZpLl4b+/MmToXPnuKsSkSZSi18ObudOuP32sM7tG2+E4ZkLFij0RdKcWvxSv9JSuPZaWL8errgC7r0Xjj027qpEpAWoxS/727EDrr8+zLFTUwNPPw3Tpyv0RTKIgl8+Nn8+dO0KU6bA8OGwejX07x93VSLSwhT8Alu3wqBBMHAgtG0LL74I990XJlgTkYyj4M9m7jBtWrgRa+5c+NWvoKwMzjgj7spEJEK6uJutXn89LHBeUhL684uK4NRT465KRJIgsha/mT1qZtvMbHWdfW3NbJGZvZp4PCaq88tB7NkD99wD3bvDyy+H1bEWL1boi2SRKLt6fgdccMC+UUCJu58ClCS2JVmWL4czz4QRI+D882Ht2tDqP0w9fiLZJLL/4939eeDvB+weCExPPJ8OFER1fqmjpgZGj4b8fKiogFmzYM4c6PTxSlhzyys4a9wznDTqKc4a9wxzyytiLFhEopTsPv7j3H0LgLtvMbODrtJhZoVAIUBn3SnadIsXhxuxXnsNhgyBCRPgmP172OaWVzB69ipqdtcCUFFdw+jZqwAtkyiSiVL2b3x3L3L3fHfPb9euXdzlpJ/q6hD4/fqF0TslJWEmzWM+eVllQvGGj0J/n5rdtUwo3pCkYkUkmZId/FvNrANA4nFbks+fHWbPDkM0p02DkSNh5Uo455yDvr2yuqZR+0UkvSU7+OcDVyaeXwnMS/L5M1tlJVx8MfzgB9ChQ5hNc/x4OPLITz2sY15uo/aLSHqLcjjn74EXgS5m9raZDQHGAeeZ2avAeYltOYRDXnjduzeMw+/aNax/O358CP3evRv0+SP6dyE3p9V++3JzWjGif5eW+goikkIiu7jr7oMP8tK5UZ0zEx3ywuvGjVBYCM89F/rzi4rg5JMbdY59F3AnFG+gsrqGjnm5jOjfRRd2RTKUuXvcNRxSfn6+l5WVxV1GLM4a9wwV9fS1d26Tw/OHvQK/+AXk5oZpk3/0IzCLoUoRSUVm9oq75x+4X1M2pLj6LrB23/Iq//Hor6FqE1xyCTzwQOjTFxFpAAV/iuuYl/tRiz93106Gl85gSNk8/t7mmHATVkFBvAWKSNpJ2XH8Euy78Pr1N8spfnQYhS/PYVavC3jpv/+i0BeRJlGLP8UVdD6C3isepfN/z+KNtp24vvB+zh/6Qy7ShVcRaSIFf6pyh5kz4cYb6fz3v8Ntt/Gln/2M3x5xRNyViUiaU/Cnos2bw7q3CxaEidX+9Cf46lfjrkpEMoT6+FPJ3r3w299Ct27wzDNh+cMlSxT6ItKi1OJPFevWwTXXwF//GubKf+ghOOmkuKsSkQykFn/cdu2CO++Enj1h/Xp47DF4+mmFvohERi3+OC1ZElr5a9bA4MEwcSK0P+gSBSIiLUIt/ji8/z7cdBP07Qs7doSLuE88odAXkaRQiz/ZFi6EoUPDyJ1hw+Duu6FNm7irEpEsohZ/slRVwWWXwYAB0Lo1lJbCpEkKfRFJOgV/1Nzh8cfDilgzZ8KYMVBeHrp5RERioK6eKP3tb6Fb5+mnoU+fsOZtt25xVyUiWU4t/ijU1sKvfx1Cfl+XTmmpQl9EUoJa/C1t1aowRHPp0tCfP3kydO4cd1UiIh+JJfjNbBPwHlAL7KlvhZh0Mbe8ggnFG9i+/R+MXjabK/7yXxyWlxeGZ156qVbEEpGUE2eLv5+7b4/x/M22bz3cbm+uZPrCSZz897eZ2/1cPjPxfgac0z3u8kRE6qWunmZ4cN4ybn3qIS4v/x/e/lx7rvjhL3j+S/9Mp6VVDDgn7upEROoXV/A78Cczc2CKuxcd+AYzKwQKATqnYh/5/Pk8du/VtP/gXR7JH8h9Z1/Gh5/JBepfJ1dEJFXEFfxnuXulmbUHFpnZend/vu4bEr8MigDy8/M9jiLrtXUr3HgjzJzJB8efxMXfv5UVHbvs95aOebkxFScicmixDOd098rE4zZgDnBGHHU0ijtMmxZuxJo7F371K9bMK2HjF7vu97bcnFaM6N+l/s8QEUkBSW/xm1lr4DB3fy/x/HzgzmTX0Sivvw7XXQclJXD22VBUBKeeykDAcz7DhOINVFbX0DEvlxH9u1Cg9XBFJIXF0dVzHDDHwjDHw4En3P3pGOo4tD17wlTJP/855OSExVGuvRYO+/gPpYJenRT0IpJWkh787v4GkPprCS5fDkOGwLJlMHAgPPggdFLAi0j605QNB6qpgdGjwyLnFRUwaxbMmaPQF5GMoXH8dS1eHLpyXnsttPYnTIBjjom7KhGRFqUWP0B1dQj8fv3C6J2SkjCTpkJfRDKQgn/27DBEc9o0GDkSVq6Ec3TbrYhkruzt6qmshBtuCP33vXrBU09B795xVyUiErnsa/Hv3QsPPwxdu4b1b8ePD1MoK/RFJEtkV4t/40YoLITnngv9+UVFcPLJcVclIpJU2dHi370bxo6FHj1gxQqYOjVcwFXoi0gWyvwWf1lZGJq5ciVccgk88AB06BB3VSIiscnsFv9dd8GZZ8L27eEi7qxZCn0RyXqZHfxf+lIYn792LRQUxF2NiEhKyOyunsGDw4+IiHwks1v8IiLyCQp+EZEso+AXEckyCn4RkSyj4BcRyTIKfhGRLKPgFxHJMgp+EZEsY+4edw2HZGZVwN+aePixwPYWLCdO+i6pJ1O+B+i7pKrmfJcvunu7A3emRfA3h5mVuXt+3HW0BH2X1JMp3wP0XVJVFN9FXT0iIllGwS8ikmWyIfiL4i6gBem7pJ5M+R6g75KqWvy7ZHwfv4iI7C8bWvwiIlKHgl9EJMtkdPCb2QVmtsHMXjOzUXHX0xRm9gUze9bM1pnZGjO7Ke6amsvMWplZuZktiLuW5jCzPDP7o5mtT/z3+VrcNTWVmQ1P/PtabWa/N7Mj4q6poczsUTPbZmar6+xra2aLzOzVxOMxcdbYEAf5HhMS/75WmtkcM8triXNlbPCbWSvgQeA7QFdgsJl1jbeqJtkD/MTdTwP6AMPS9HvUdROwLu4iWsCvgafd/VTgq6TpdzKzTsCNQL67fwVoBVwab1WN8jvgggP2jQJK3P0UoCSxnep+xye/xyLgK+7eA9gIjG6JE2Vs8ANnAK+5+xvuvgv4L2BgzDU1mrtvcfdliefvEcKlU7xVNZ2ZnQBcCDwSdy3NYWafA74BTAVw913uXh1rUc1zOJBrZocDRwKVMdfTYO7+PPD3A3YPBKYnnk8HCpJZU1PU9z3c/U/uviexuQQ4oSXOlcnB3wnYXGf7bdI4MAHM7ESgF/BSzKU0x0RgJLA35jqa60tAFTAt0W31iJm1jruopnD3CuAe4C1gC/APd/9TvFU123HuvgVC4wloH3M9LeFqYGFLfFAmB7/Vsy9tx66a2VHAk8DN7r4j7nqawswuAra5+ytx19ICDgd6A5PdvRfwAenRnfAJif7vgcBJQEegtZldFm9VUpeZ3Ubo9p3REp+XycH/NvCFOtsnkEZ/vtZlZjmE0J/h7rPjrqcZzgK+Z2abCF1v55jZ4/GW1GRvA2+7+76/vv5I+EWQjr4NvOnuVe6+G5gN9I25pubaamYdABKP22Kup8nM7ErgIuBfvYVuvMrk4H8ZOMXMTjKzzxAuVs2PuaZGMzMj9COvc/f74q6nOdx9tLuf4O4nEv57POPuadmydPf/BTabWZfErnOBtTGW1BxvAX3M7MjEv7dzSdML1XXMB65MPL8SmBdjLU1mZhcA/w58z90/bKnPzdjgT1wQuQEoJvwjnunua+KtqknOAi4ntI6XJ34GxF2UAPBvwAwzWwn0BO6Ot5ymSfzV8kdgGbCKkAtpM+WBmf0eeBHoYmZvm9kQYBxwnpm9CpyX2E5pB/kevwHaAIsS/+8/1CLn0pQNIiLZJWNb/CIiUj8Fv4hIllHwi4hkGQW/iEiWUfCLiGQZBb9kNTO7MTGzZqPuiDSzE83sX6KqSyRKCn7JdtcDA9z9Xxt53IlAo4M/MWusSKwU/JK1EjfDfAmYb2a3JeZDfzkx6drAxHtONLO/mNmyxM++qQzGAWcnbqoZbmZXmdlv6nz2AjP7VuL5+2Z2p5m9BHzNzC4zs6WJY6fol4Ekm4Jfspa7DyXM39QPaE2YQuL0xPaExGyb24Dz3L03MAh4IHH4KOAv7t7T3e8/xKlaA6vd/UzgncTnnOXuPYFaoLF/bYg0y+FxFyCSIs4nTCD308T2EUBnwi+G35hZT0JI/1MTPruWMMkehHlw/hl4OUyLQy5pPIGYpCcFv0hgwA/cfcN+O83GAFsJK2wdBuw8yPF72P8v6LpLF+5099o655nu7i2ykpJIU6irRyQoBv4tMTslZtYrsf9oYIu77yVMlrevP/49wuRZ+2wCeprZYWb2BcIKcPUpAS4xs/aJ87Q1sy+26DcROQQFv0jwSyAHWJlY7PqXif2/Ba40syWEbp4PEvtXAnvMbIWZDQdeAN4kzG55D2Gmy09w97XAz4A/JWb1XAR0iOYridRPs3OKiGQZtfhFRLKMgl9EJMso+EVEsoyCX0Qkyyj4RUSyjIJfRCTLKPhFRLLM/wdOO6MbxlIUwgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "learning_rate=0.01\n",
    "epochs= 400   # Replace ? with an integer.\n",
    "my_batch_size=12\n",
    "\n",
    "my_model = build_model(learning_rate)\n",
    "trained_weight, trained_bias, epochs, rmse = train_model(my_model, my_feature, \n",
    "                                                        my_label, epochs,\n",
    "                                                        my_batch_size)\n",
    "plot_the_model(trained_weight, trained_bias, my_feature, my_label)\n",
    "plot_the_loss_curve(epochs, rmse)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "cellView": "form",
    "colab": {},
    "colab_type": "code",
    "id": "1tWrzP4Ww7sD"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/450\n",
      "1/1 [==============================] - 0s 871us/step - loss: 1328.2345 - root_mean_squared_error: 36.4450\n",
      "Epoch 2/450\n",
      "1/1 [==============================] - 0s 779us/step - loss: 1309.2863 - root_mean_squared_error: 36.1841\n",
      "Epoch 3/450\n",
      "1/1 [==============================] - 0s 808us/step - loss: 1295.6715 - root_mean_squared_error: 35.9954\n",
      "Epoch 4/450\n",
      "1/1 [==============================] - 0s 890us/step - loss: 1284.3497 - root_mean_squared_error: 35.8378\n",
      "Epoch 5/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1274.3564 - root_mean_squared_error: 35.6981\n",
      "Epoch 6/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1265.2445 - root_mean_squared_error: 35.5703\n",
      "Epoch 7/450\n",
      "1/1 [==============================] - 0s 755us/step - loss: 1256.7648 - root_mean_squared_error: 35.4509\n",
      "Epoch 8/450\n",
      "1/1 [==============================] - 0s 718us/step - loss: 1248.7623 - root_mean_squared_error: 35.3378\n",
      "Epoch 9/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1241.1339 - root_mean_squared_error: 35.2297\n",
      "Epoch 10/450\n",
      "1/1 [==============================] - 0s 818us/step - loss: 1233.8062 - root_mean_squared_error: 35.1256\n",
      "Epoch 11/450\n",
      "1/1 [==============================] - 0s 728us/step - loss: 1226.7255 - root_mean_squared_error: 35.0246\n",
      "Epoch 12/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1219.8511 - root_mean_squared_error: 34.9264\n",
      "Epoch 13/450\n",
      "1/1 [==============================] - 0s 805us/step - loss: 1213.1515 - root_mean_squared_error: 34.8303\n",
      "Epoch 14/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1206.6016 - root_mean_squared_error: 34.7362\n",
      "Epoch 15/450\n",
      "1/1 [==============================] - 0s 860us/step - loss: 1200.1810 - root_mean_squared_error: 34.6436\n",
      "Epoch 16/450\n",
      "1/1 [==============================] - 0s 706us/step - loss: 1193.8735 - root_mean_squared_error: 34.5525\n",
      "Epoch 17/450\n",
      "1/1 [==============================] - 0s 801us/step - loss: 1187.6654 - root_mean_squared_error: 34.4625\n",
      "Epoch 18/450\n",
      "1/1 [==============================] - 0s 734us/step - loss: 1181.5453 - root_mean_squared_error: 34.3736\n",
      "Epoch 19/450\n",
      "1/1 [==============================] - 0s 845us/step - loss: 1175.5034 - root_mean_squared_error: 34.2856\n",
      "Epoch 20/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1169.5317 - root_mean_squared_error: 34.1984\n",
      "Epoch 21/450\n",
      "1/1 [==============================] - 0s 820us/step - loss: 1163.6234 - root_mean_squared_error: 34.1119\n",
      "Epoch 22/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1157.7721 - root_mean_squared_error: 34.0261\n",
      "Epoch 23/450\n",
      "1/1 [==============================] - 0s 803us/step - loss: 1151.9730 - root_mean_squared_error: 33.9407\n",
      "Epoch 24/450\n",
      "1/1 [==============================] - 0s 715us/step - loss: 1146.2216 - root_mean_squared_error: 33.8559\n",
      "Epoch 25/450\n",
      "1/1 [==============================] - 0s 872us/step - loss: 1140.5135 - root_mean_squared_error: 33.7715\n",
      "Epoch 26/450\n",
      "1/1 [==============================] - 0s 736us/step - loss: 1134.8458 - root_mean_squared_error: 33.6875\n",
      "Epoch 27/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1129.2155 - root_mean_squared_error: 33.6038\n",
      "Epoch 28/450\n",
      "1/1 [==============================] - 0s 844us/step - loss: 1123.6198 - root_mean_squared_error: 33.5204\n",
      "Epoch 29/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1118.0562 - root_mean_squared_error: 33.4373\n",
      "Epoch 30/450\n",
      "1/1 [==============================] - 0s 862us/step - loss: 1112.5228 - root_mean_squared_error: 33.3545\n",
      "Epoch 31/450\n",
      "1/1 [==============================] - 0s 711us/step - loss: 1107.0181 - root_mean_squared_error: 33.2719\n",
      "Epoch 32/450\n",
      "1/1 [==============================] - 0s 843us/step - loss: 1101.5400 - root_mean_squared_error: 33.1895\n",
      "Epoch 33/450\n",
      "1/1 [==============================] - 0s 765us/step - loss: 1096.0873 - root_mean_squared_error: 33.1072\n",
      "Epoch 34/450\n",
      "1/1 [==============================] - 0s 860us/step - loss: 1090.6587 - root_mean_squared_error: 33.0251\n",
      "Epoch 35/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1085.2532 - root_mean_squared_error: 32.9432\n",
      "Epoch 36/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1079.8695 - root_mean_squared_error: 32.8614\n",
      "Epoch 37/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1074.5071 - root_mean_squared_error: 32.7797\n",
      "Epoch 38/450\n",
      "1/1 [==============================] - 0s 691us/step - loss: 1069.1648 - root_mean_squared_error: 32.6981\n",
      "Epoch 39/450\n",
      "1/1 [==============================] - 0s 756us/step - loss: 1063.8419 - root_mean_squared_error: 32.6166\n",
      "Epoch 40/450\n",
      "1/1 [==============================] - 0s 755us/step - loss: 1058.5382 - root_mean_squared_error: 32.5352\n",
      "Epoch 41/450\n",
      "1/1 [==============================] - 0s 723us/step - loss: 1053.2528 - root_mean_squared_error: 32.4539\n",
      "Epoch 42/450\n",
      "1/1 [==============================] - 0s 896us/step - loss: 1047.9852 - root_mean_squared_error: 32.3726\n",
      "Epoch 43/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1042.7350 - root_mean_squared_error: 32.2914\n",
      "Epoch 44/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1037.5018 - root_mean_squared_error: 32.2103\n",
      "Epoch 45/450\n",
      "1/1 [==============================] - 0s 883us/step - loss: 1032.2850 - root_mean_squared_error: 32.1292\n",
      "Epoch 46/450\n",
      "1/1 [==============================] - 0s 761us/step - loss: 1027.0845 - root_mean_squared_error: 32.0482\n",
      "Epoch 47/450\n",
      "1/1 [==============================] - 0s 867us/step - loss: 1021.8999 - root_mean_squared_error: 31.9672\n",
      "Epoch 48/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1016.7310 - root_mean_squared_error: 31.8862\n",
      "Epoch 49/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1011.5776 - root_mean_squared_error: 31.8053\n",
      "Epoch 50/450\n",
      "1/1 [==============================] - 0s 743us/step - loss: 1006.4395 - root_mean_squared_error: 31.7244\n",
      "Epoch 51/450\n",
      "1/1 [==============================] - 0s 729us/step - loss: 1001.3162 - root_mean_squared_error: 31.6436\n",
      "Epoch 52/450\n",
      "1/1 [==============================] - 0s 715us/step - loss: 996.2080 - root_mean_squared_error: 31.5628\n",
      "Epoch 53/450\n",
      "1/1 [==============================] - 0s 804us/step - loss: 991.1144 - root_mean_squared_error: 31.4820\n",
      "Epoch 54/450\n",
      "1/1 [==============================] - 0s 839us/step - loss: 986.0354 - root_mean_squared_error: 31.4012\n",
      "Epoch 55/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 980.9709 - root_mean_squared_error: 31.3205\n",
      "Epoch 56/450\n",
      "1/1 [==============================] - 0s 688us/step - loss: 975.9205 - root_mean_squared_error: 31.2397\n",
      "Epoch 57/450\n",
      "1/1 [==============================] - 0s 744us/step - loss: 970.8845 - root_mean_squared_error: 31.1590\n",
      "Epoch 58/450\n",
      "1/1 [==============================] - 0s 691us/step - loss: 965.8625 - root_mean_squared_error: 31.0783\n",
      "Epoch 59/450\n",
      "1/1 [==============================] - 0s 700us/step - loss: 960.8547 - root_mean_squared_error: 30.9977\n",
      "Epoch 60/450\n",
      "1/1 [==============================] - 0s 669us/step - loss: 955.8607 - root_mean_squared_error: 30.9170\n",
      "Epoch 61/450\n",
      "1/1 [==============================] - 0s 682us/step - loss: 950.8808 - root_mean_squared_error: 30.8364\n",
      "Epoch 62/450\n",
      "1/1 [==============================] - 0s 725us/step - loss: 945.9146 - root_mean_squared_error: 30.7557\n",
      "Epoch 63/450\n",
      "1/1 [==============================] - 0s 663us/step - loss: 940.9622 - root_mean_squared_error: 30.6751\n",
      "Epoch 64/450\n",
      "1/1 [==============================] - 0s 687us/step - loss: 936.0234 - root_mean_squared_error: 30.5945\n",
      "Epoch 65/450\n",
      "1/1 [==============================] - 0s 629us/step - loss: 931.0986 - root_mean_squared_error: 30.5139\n",
      "Epoch 66/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 926.1872 - root_mean_squared_error: 30.4333\n",
      "Epoch 67/450\n",
      "1/1 [==============================] - 0s 671us/step - loss: 921.2895 - root_mean_squared_error: 30.3528\n",
      "Epoch 68/450\n",
      "1/1 [==============================] - 0s 642us/step - loss: 916.4055 - root_mean_squared_error: 30.2722\n",
      "Epoch 69/450\n",
      "1/1 [==============================] - 0s 701us/step - loss: 911.5349 - root_mean_squared_error: 30.1916\n",
      "Epoch 70/450\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 1ms/step - loss: 906.6779 - root_mean_squared_error: 30.1111\n",
      "Epoch 71/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 901.8343 - root_mean_squared_error: 30.0306\n",
      "Epoch 72/450\n",
      "1/1 [==============================] - 0s 940us/step - loss: 897.0042 - root_mean_squared_error: 29.9500\n",
      "Epoch 73/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 892.1878 - root_mean_squared_error: 29.8695\n",
      "Epoch 74/450\n",
      "1/1 [==============================] - 0s 945us/step - loss: 887.3846 - root_mean_squared_error: 29.7890\n",
      "Epoch 75/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 882.5949 - root_mean_squared_error: 29.7085\n",
      "Epoch 76/450\n",
      "1/1 [==============================] - 0s 977us/step - loss: 877.8186 - root_mean_squared_error: 29.6280\n",
      "Epoch 77/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 873.0557 - root_mean_squared_error: 29.5475\n",
      "Epoch 78/450\n",
      "1/1 [==============================] - 0s 956us/step - loss: 868.3062 - root_mean_squared_error: 29.4670\n",
      "Epoch 79/450\n",
      "1/1 [==============================] - 0s 845us/step - loss: 863.5701 - root_mean_squared_error: 29.3866\n",
      "Epoch 80/450\n",
      "1/1 [==============================] - 0s 947us/step - loss: 858.8474 - root_mean_squared_error: 29.3061\n",
      "Epoch 81/450\n",
      "1/1 [==============================] - 0s 808us/step - loss: 854.1380 - root_mean_squared_error: 29.2256\n",
      "Epoch 82/450\n",
      "1/1 [==============================] - 0s 907us/step - loss: 849.4421 - root_mean_squared_error: 29.1452\n",
      "Epoch 83/450\n",
      "1/1 [==============================] - 0s 838us/step - loss: 844.7593 - root_mean_squared_error: 29.0647\n",
      "Epoch 84/450\n",
      "1/1 [==============================] - 0s 829us/step - loss: 840.0899 - root_mean_squared_error: 28.9843\n",
      "Epoch 85/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 835.4339 - root_mean_squared_error: 28.9039\n",
      "Epoch 86/450\n",
      "1/1 [==============================] - 0s 845us/step - loss: 830.7912 - root_mean_squared_error: 28.8234\n",
      "Epoch 87/450\n",
      "1/1 [==============================] - 0s 941us/step - loss: 826.1619 - root_mean_squared_error: 28.7430\n",
      "Epoch 88/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 821.5459 - root_mean_squared_error: 28.6626\n",
      "Epoch 89/450\n",
      "1/1 [==============================] - 0s 900us/step - loss: 816.9431 - root_mean_squared_error: 28.5822\n",
      "Epoch 90/450\n",
      "1/1 [==============================] - 0s 957us/step - loss: 812.3537 - root_mean_squared_error: 28.5018\n",
      "Epoch 91/450\n",
      "1/1 [==============================] - 0s 917us/step - loss: 807.7775 - root_mean_squared_error: 28.4214\n",
      "Epoch 92/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 803.2146 - root_mean_squared_error: 28.3410\n",
      "Epoch 93/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 798.6652 - root_mean_squared_error: 28.2607\n",
      "Epoch 94/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 794.1289 - root_mean_squared_error: 28.1803\n",
      "Epoch 95/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 789.6060 - root_mean_squared_error: 28.0999\n",
      "Epoch 96/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 785.0963 - root_mean_squared_error: 28.0196\n",
      "Epoch 97/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 780.5999 - root_mean_squared_error: 27.9392\n",
      "Epoch 98/450\n",
      "1/1 [==============================] - 0s 856us/step - loss: 776.1168 - root_mean_squared_error: 27.8589\n",
      "Epoch 99/450\n",
      "1/1 [==============================] - 0s 873us/step - loss: 771.6470 - root_mean_squared_error: 27.7785\n",
      "Epoch 100/450\n",
      "1/1 [==============================] - 0s 989us/step - loss: 767.1904 - root_mean_squared_error: 27.6982\n",
      "Epoch 101/450\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 762.7473 - root_mean_squared_error: 27.6179\n",
      "Epoch 102/450\n",
      "1/1 [==============================] - 0s 808us/step - loss: 758.3173 - root_mean_squared_error: 27.5376\n",
      "Epoch 103/450\n",
      "1/1 [==============================] - 0s 775us/step - loss: 753.9006 - root_mean_squared_error: 27.4573\n",
      "Epoch 104/450\n",
      "1/1 [==============================] - 0s 856us/step - loss: 749.4971 - root_mean_squared_error: 27.3769\n",
      "Epoch 105/450\n",
      "1/1 [==============================] - 0s 756us/step - loss: 745.1069 - root_mean_squared_error: 27.2966\n",
      "Epoch 106/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 740.7300 - root_mean_squared_error: 27.2164\n",
      "Epoch 107/450\n",
      "1/1 [==============================] - 0s 933us/step - loss: 736.3665 - root_mean_squared_error: 27.1361\n",
      "Epoch 108/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 732.0161 - root_mean_squared_error: 27.0558\n",
      "Epoch 109/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 727.6790 - root_mean_squared_error: 26.9755\n",
      "Epoch 110/450\n",
      "1/1 [==============================] - 0s 874us/step - loss: 723.3552 - root_mean_squared_error: 26.8953\n",
      "Epoch 111/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 719.0446 - root_mean_squared_error: 26.8150\n",
      "Epoch 112/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 714.7473 - root_mean_squared_error: 26.7348\n",
      "Epoch 113/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 710.4632 - root_mean_squared_error: 26.6545\n",
      "Epoch 114/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 706.1924 - root_mean_squared_error: 26.5743\n",
      "Epoch 115/450\n",
      "1/1 [==============================] - 0s 937us/step - loss: 701.9349 - root_mean_squared_error: 26.4941\n",
      "Epoch 116/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 697.6906 - root_mean_squared_error: 26.4138\n",
      "Epoch 117/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 693.4595 - root_mean_squared_error: 26.3336\n",
      "Epoch 118/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 689.2417 - root_mean_squared_error: 26.2534\n",
      "Epoch 119/450\n",
      "1/1 [==============================] - 0s 863us/step - loss: 685.0371 - root_mean_squared_error: 26.1732\n",
      "Epoch 120/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 680.8458 - root_mean_squared_error: 26.0930\n",
      "Epoch 121/450\n",
      "1/1 [==============================] - 0s 752us/step - loss: 676.6678 - root_mean_squared_error: 26.0128\n",
      "Epoch 122/450\n",
      "1/1 [==============================] - 0s 936us/step - loss: 672.5029 - root_mean_squared_error: 25.9327\n",
      "Epoch 123/450\n",
      "1/1 [==============================] - 0s 779us/step - loss: 668.3514 - root_mean_squared_error: 25.8525\n",
      "Epoch 124/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 664.2131 - root_mean_squared_error: 25.7723\n",
      "Epoch 125/450\n",
      "1/1 [==============================] - 0s 867us/step - loss: 660.0880 - root_mean_squared_error: 25.6922\n",
      "Epoch 126/450\n",
      "1/1 [==============================] - 0s 777us/step - loss: 655.9761 - root_mean_squared_error: 25.6120\n",
      "Epoch 127/450\n",
      "1/1 [==============================] - 0s 789us/step - loss: 651.8776 - root_mean_squared_error: 25.5319\n",
      "Epoch 128/450\n",
      "1/1 [==============================] - 0s 696us/step - loss: 647.7922 - root_mean_squared_error: 25.4518\n",
      "Epoch 129/450\n",
      "1/1 [==============================] - 0s 734us/step - loss: 643.7202 - root_mean_squared_error: 25.3716\n",
      "Epoch 130/450\n",
      "1/1 [==============================] - 0s 716us/step - loss: 639.6612 - root_mean_squared_error: 25.2915\n",
      "Epoch 131/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 635.6155 - root_mean_squared_error: 25.2114\n",
      "Epoch 132/450\n",
      "1/1 [==============================] - 0s 727us/step - loss: 631.5831 - root_mean_squared_error: 25.1313\n",
      "Epoch 133/450\n",
      "1/1 [==============================] - 0s 714us/step - loss: 627.5640 - root_mean_squared_error: 25.0512\n",
      "Epoch 134/450\n",
      "1/1 [==============================] - 0s 708us/step - loss: 623.5580 - root_mean_squared_error: 24.9711\n",
      "Epoch 135/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 619.5652 - root_mean_squared_error: 24.8911\n",
      "Epoch 136/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 615.5858 - root_mean_squared_error: 24.8110\n",
      "Epoch 137/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 611.6194 - root_mean_squared_error: 24.7309\n",
      "Epoch 138/450\n",
      "1/1 [==============================] - 0s 907us/step - loss: 607.6664 - root_mean_squared_error: 24.6509\n",
      "Epoch 139/450\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 729us/step - loss: 603.7266 - root_mean_squared_error: 24.5708\n",
      "Epoch 140/450\n",
      "1/1 [==============================] - 0s 673us/step - loss: 599.8000 - root_mean_squared_error: 24.4908\n",
      "Epoch 141/450\n",
      "1/1 [==============================] - 0s 947us/step - loss: 595.8866 - root_mean_squared_error: 24.4108\n",
      "Epoch 142/450\n",
      "1/1 [==============================] - 0s 923us/step - loss: 591.9864 - root_mean_squared_error: 24.3308\n",
      "Epoch 143/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 588.0994 - root_mean_squared_error: 24.2508\n",
      "Epoch 144/450\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 584.2258 - root_mean_squared_error: 24.1708\n",
      "Epoch 145/450\n",
      "1/1 [==============================] - 0s 960us/step - loss: 580.3652 - root_mean_squared_error: 24.0908\n",
      "Epoch 146/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 576.5179 - root_mean_squared_error: 24.0108\n",
      "Epoch 147/450\n",
      "1/1 [==============================] - 0s 875us/step - loss: 572.6838 - root_mean_squared_error: 23.9308\n",
      "Epoch 148/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 568.8629 - root_mean_squared_error: 23.8508\n",
      "Epoch 149/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 565.0552 - root_mean_squared_error: 23.7709\n",
      "Epoch 150/450\n",
      "1/1 [==============================] - 0s 975us/step - loss: 561.2608 - root_mean_squared_error: 23.6909\n",
      "Epoch 151/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 557.4795 - root_mean_squared_error: 23.6110\n",
      "Epoch 152/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 553.7114 - root_mean_squared_error: 23.5311\n",
      "Epoch 153/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 549.9566 - root_mean_squared_error: 23.4512\n",
      "Epoch 154/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 546.2150 - root_mean_squared_error: 23.3712\n",
      "Epoch 155/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 542.4865 - root_mean_squared_error: 23.2913\n",
      "Epoch 156/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 538.7713 - root_mean_squared_error: 23.2114\n",
      "Epoch 157/450\n",
      "1/1 [==============================] - 0s 856us/step - loss: 535.0693 - root_mean_squared_error: 23.1316\n",
      "Epoch 158/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 531.3803 - root_mean_squared_error: 23.0517\n",
      "Epoch 159/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 527.7047 - root_mean_squared_error: 22.9718\n",
      "Epoch 160/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 524.0422 - root_mean_squared_error: 22.8920\n",
      "Epoch 161/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 520.3929 - root_mean_squared_error: 22.8121\n",
      "Epoch 162/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 516.7568 - root_mean_squared_error: 22.7323\n",
      "Epoch 163/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 513.1340 - root_mean_squared_error: 22.6525\n",
      "Epoch 164/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 509.5243 - root_mean_squared_error: 22.5726\n",
      "Epoch 165/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 505.9277 - root_mean_squared_error: 22.4928\n",
      "Epoch 166/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 502.3444 - root_mean_squared_error: 22.4130\n",
      "Epoch 167/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 498.7743 - root_mean_squared_error: 22.3333\n",
      "Epoch 168/450\n",
      "1/1 [==============================] - 0s 963us/step - loss: 495.2173 - root_mean_squared_error: 22.2535\n",
      "Epoch 169/450\n",
      "1/1 [==============================] - 0s 933us/step - loss: 491.6735 - root_mean_squared_error: 22.1737\n",
      "Epoch 170/450\n",
      "1/1 [==============================] - 0s 985us/step - loss: 488.1429 - root_mean_squared_error: 22.0940\n",
      "Epoch 171/450\n",
      "1/1 [==============================] - 0s 749us/step - loss: 484.6254 - root_mean_squared_error: 22.0142\n",
      "Epoch 172/450\n",
      "1/1 [==============================] - 0s 969us/step - loss: 481.1211 - root_mean_squared_error: 21.9345\n",
      "Epoch 173/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 477.6300 - root_mean_squared_error: 21.8547\n",
      "Epoch 174/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 474.1521 - root_mean_squared_error: 21.7750\n",
      "Epoch 175/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 470.6874 - root_mean_squared_error: 21.6953\n",
      "Epoch 176/450\n",
      "1/1 [==============================] - 0s 815us/step - loss: 467.2357 - root_mean_squared_error: 21.6156\n",
      "Epoch 177/450\n",
      "1/1 [==============================] - 0s 751us/step - loss: 463.7973 - root_mean_squared_error: 21.5360\n",
      "Epoch 178/450\n",
      "1/1 [==============================] - 0s 865us/step - loss: 460.3721 - root_mean_squared_error: 21.4563\n",
      "Epoch 179/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 456.9600 - root_mean_squared_error: 21.3766\n",
      "Epoch 180/450\n",
      "1/1 [==============================] - 0s 947us/step - loss: 453.5610 - root_mean_squared_error: 21.2970\n",
      "Epoch 181/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 450.1753 - root_mean_squared_error: 21.2173\n",
      "Epoch 182/450\n",
      "1/1 [==============================] - 0s 816us/step - loss: 446.8026 - root_mean_squared_error: 21.1377\n",
      "Epoch 183/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 443.4432 - root_mean_squared_error: 21.0581\n",
      "Epoch 184/450\n",
      "1/1 [==============================] - 0s 960us/step - loss: 440.0968 - root_mean_squared_error: 20.9785\n",
      "Epoch 185/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 436.7638 - root_mean_squared_error: 20.8989\n",
      "Epoch 186/450\n",
      "1/1 [==============================] - 0s 710us/step - loss: 433.4437 - root_mean_squared_error: 20.8193\n",
      "Epoch 187/450\n",
      "1/1 [==============================] - 0s 668us/step - loss: 430.1368 - root_mean_squared_error: 20.7397\n",
      "Epoch 188/450\n",
      "1/1 [==============================] - 0s 827us/step - loss: 426.8431 - root_mean_squared_error: 20.6602\n",
      "Epoch 189/450\n",
      "1/1 [==============================] - 0s 769us/step - loss: 423.5625 - root_mean_squared_error: 20.5806\n",
      "Epoch 190/450\n",
      "1/1 [==============================] - 0s 880us/step - loss: 420.2951 - root_mean_squared_error: 20.5011\n",
      "Epoch 191/450\n",
      "1/1 [==============================] - 0s 769us/step - loss: 417.0409 - root_mean_squared_error: 20.4216\n",
      "Epoch 192/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 413.7997 - root_mean_squared_error: 20.3421\n",
      "Epoch 193/450\n",
      "1/1 [==============================] - 0s 687us/step - loss: 410.5717 - root_mean_squared_error: 20.2626\n",
      "Epoch 194/450\n",
      "1/1 [==============================] - 0s 803us/step - loss: 407.3568 - root_mean_squared_error: 20.1831\n",
      "Epoch 195/450\n",
      "1/1 [==============================] - 0s 816us/step - loss: 404.1550 - root_mean_squared_error: 20.1036\n",
      "Epoch 196/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 400.9664 - root_mean_squared_error: 20.0241\n",
      "Epoch 197/450\n",
      "1/1 [==============================] - 0s 976us/step - loss: 397.7909 - root_mean_squared_error: 19.9447\n",
      "Epoch 198/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 394.6286 - root_mean_squared_error: 19.8653\n",
      "Epoch 199/450\n",
      "1/1 [==============================] - 0s 719us/step - loss: 391.4793 - root_mean_squared_error: 19.7858\n",
      "Epoch 200/450\n",
      "1/1 [==============================] - 0s 725us/step - loss: 388.3433 - root_mean_squared_error: 19.7064\n",
      "Epoch 201/450\n",
      "1/1 [==============================] - 0s 675us/step - loss: 385.2202 - root_mean_squared_error: 19.6270\n",
      "Epoch 202/450\n",
      "1/1 [==============================] - 0s 682us/step - loss: 382.1104 - root_mean_squared_error: 19.5476\n",
      "Epoch 203/450\n",
      "1/1 [==============================] - 0s 683us/step - loss: 379.0136 - root_mean_squared_error: 19.4683\n",
      "Epoch 204/450\n",
      "1/1 [==============================] - 0s 801us/step - loss: 375.9300 - root_mean_squared_error: 19.3889\n",
      "Epoch 205/450\n",
      "1/1 [==============================] - 0s 721us/step - loss: 372.8594 - root_mean_squared_error: 19.3096\n",
      "Epoch 206/450\n",
      "1/1 [==============================] - 0s 755us/step - loss: 369.8020 - root_mean_squared_error: 19.2302\n",
      "Epoch 207/450\n",
      "1/1 [==============================] - 0s 861us/step - loss: 366.7577 - root_mean_squared_error: 19.1509\n",
      "Epoch 208/450\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 863us/step - loss: 363.7265 - root_mean_squared_error: 19.0716\n",
      "Epoch 209/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 360.7084 - root_mean_squared_error: 18.9923\n",
      "Epoch 210/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 357.7034 - root_mean_squared_error: 18.9130\n",
      "Epoch 211/450\n",
      "1/1 [==============================] - 0s 958us/step - loss: 354.7115 - root_mean_squared_error: 18.8338\n",
      "Epoch 212/450\n",
      "1/1 [==============================] - 0s 945us/step - loss: 351.7327 - root_mean_squared_error: 18.7545\n",
      "Epoch 213/450\n",
      "1/1 [==============================] - 0s 972us/step - loss: 348.7670 - root_mean_squared_error: 18.6753\n",
      "Epoch 214/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 345.8144 - root_mean_squared_error: 18.5961\n",
      "Epoch 215/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 342.8748 - root_mean_squared_error: 18.5169\n",
      "Epoch 216/450\n",
      "1/1 [==============================] - 0s 868us/step - loss: 339.9484 - root_mean_squared_error: 18.4377\n",
      "Epoch 217/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 337.0350 - root_mean_squared_error: 18.3585\n",
      "Epoch 218/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 334.1348 - root_mean_squared_error: 18.2794\n",
      "Epoch 219/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 331.2476 - root_mean_squared_error: 18.2002\n",
      "Epoch 220/450\n",
      "1/1 [==============================] - 0s 822us/step - loss: 328.3735 - root_mean_squared_error: 18.1211\n",
      "Epoch 221/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 325.5124 - root_mean_squared_error: 18.0420\n",
      "Epoch 222/450\n",
      "1/1 [==============================] - 0s 852us/step - loss: 322.6645 - root_mean_squared_error: 17.9629\n",
      "Epoch 223/450\n",
      "1/1 [==============================] - 0s 851us/step - loss: 319.8296 - root_mean_squared_error: 17.8838\n",
      "Epoch 224/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 317.0078 - root_mean_squared_error: 17.8047\n",
      "Epoch 225/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 314.1990 - root_mean_squared_error: 17.7257\n",
      "Epoch 226/450\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 311.4033 - root_mean_squared_error: 17.6466\n",
      "Epoch 227/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 308.6207 - root_mean_squared_error: 17.5676\n",
      "Epoch 228/450\n",
      "1/1 [==============================] - 0s 835us/step - loss: 305.8511 - root_mean_squared_error: 17.4886\n",
      "Epoch 229/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 303.0946 - root_mean_squared_error: 17.4096\n",
      "Epoch 230/450\n",
      "1/1 [==============================] - 0s 764us/step - loss: 300.3511 - root_mean_squared_error: 17.3306\n",
      "Epoch 231/450\n",
      "1/1 [==============================] - 0s 716us/step - loss: 297.6207 - root_mean_squared_error: 17.2517\n",
      "Epoch 232/450\n",
      "1/1 [==============================] - 0s 655us/step - loss: 294.9034 - root_mean_squared_error: 17.1728\n",
      "Epoch 233/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 292.1990 - root_mean_squared_error: 17.0938\n",
      "Epoch 234/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 289.5077 - root_mean_squared_error: 17.0149\n",
      "Epoch 235/450\n",
      "1/1 [==============================] - 0s 846us/step - loss: 286.8295 - root_mean_squared_error: 16.9360\n",
      "Epoch 236/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 284.1643 - root_mean_squared_error: 16.8572\n",
      "Epoch 237/450\n",
      "1/1 [==============================] - 0s 956us/step - loss: 281.5122 - root_mean_squared_error: 16.7783\n",
      "Epoch 238/450\n",
      "1/1 [==============================] - 0s 738us/step - loss: 278.8730 - root_mean_squared_error: 16.6995\n",
      "Epoch 239/450\n",
      "1/1 [==============================] - 0s 868us/step - loss: 276.2469 - root_mean_squared_error: 16.6207\n",
      "Epoch 240/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 273.6338 - root_mean_squared_error: 16.5419\n",
      "Epoch 241/450\n",
      "1/1 [==============================] - 0s 932us/step - loss: 271.0338 - root_mean_squared_error: 16.4631\n",
      "Epoch 242/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 268.4467 - root_mean_squared_error: 16.3843\n",
      "Epoch 243/450\n",
      "1/1 [==============================] - 0s 849us/step - loss: 265.8727 - root_mean_squared_error: 16.3056\n",
      "Epoch 244/450\n",
      "1/1 [==============================] - 0s 847us/step - loss: 263.3117 - root_mean_squared_error: 16.2269\n",
      "Epoch 245/450\n",
      "1/1 [==============================] - 0s 819us/step - loss: 260.7638 - root_mean_squared_error: 16.1482\n",
      "Epoch 246/450\n",
      "1/1 [==============================] - 0s 853us/step - loss: 258.2287 - root_mean_squared_error: 16.0695\n",
      "Epoch 247/450\n",
      "1/1 [==============================] - 0s 868us/step - loss: 255.7068 - root_mean_squared_error: 15.9908\n",
      "Epoch 248/450\n",
      "1/1 [==============================] - 0s 928us/step - loss: 253.1978 - root_mean_squared_error: 15.9122\n",
      "Epoch 249/450\n",
      "1/1 [==============================] - 0s 745us/step - loss: 250.7018 - root_mean_squared_error: 15.8336\n",
      "Epoch 250/450\n",
      "1/1 [==============================] - 0s 952us/step - loss: 248.2188 - root_mean_squared_error: 15.7550\n",
      "Epoch 251/450\n",
      "1/1 [==============================] - 0s 802us/step - loss: 245.7489 - root_mean_squared_error: 15.6764\n",
      "Epoch 252/450\n",
      "1/1 [==============================] - 0s 788us/step - loss: 243.2919 - root_mean_squared_error: 15.5978\n",
      "Epoch 253/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 240.8479 - root_mean_squared_error: 15.5193\n",
      "Epoch 254/450\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 238.4168 - root_mean_squared_error: 15.4408\n",
      "Epoch 255/450\n",
      "1/1 [==============================] - 0s 792us/step - loss: 235.9988 - root_mean_squared_error: 15.3623\n",
      "Epoch 256/450\n",
      "1/1 [==============================] - 0s 905us/step - loss: 233.5937 - root_mean_squared_error: 15.2838\n",
      "Epoch 257/450\n",
      "1/1 [==============================] - 0s 889us/step - loss: 231.2016 - root_mean_squared_error: 15.2053\n",
      "Epoch 258/450\n",
      "1/1 [==============================] - 0s 715us/step - loss: 228.8225 - root_mean_squared_error: 15.1269\n",
      "Epoch 259/450\n",
      "1/1 [==============================] - 0s 740us/step - loss: 226.4563 - root_mean_squared_error: 15.0485\n",
      "Epoch 260/450\n",
      "1/1 [==============================] - 0s 853us/step - loss: 224.1031 - root_mean_squared_error: 14.9701\n",
      "Epoch 261/450\n",
      "1/1 [==============================] - 0s 869us/step - loss: 221.7629 - root_mean_squared_error: 14.8917\n",
      "Epoch 262/450\n",
      "1/1 [==============================] - 0s 774us/step - loss: 219.4356 - root_mean_squared_error: 14.8134\n",
      "Epoch 263/450\n",
      "1/1 [==============================] - 0s 763us/step - loss: 217.1213 - root_mean_squared_error: 14.7350\n",
      "Epoch 264/450\n",
      "1/1 [==============================] - 0s 777us/step - loss: 214.8199 - root_mean_squared_error: 14.6567\n",
      "Epoch 265/450\n",
      "1/1 [==============================] - 0s 817us/step - loss: 212.5314 - root_mean_squared_error: 14.5785\n",
      "Epoch 266/450\n",
      "1/1 [==============================] - 0s 798us/step - loss: 210.2560 - root_mean_squared_error: 14.5002\n",
      "Epoch 267/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 207.9934 - root_mean_squared_error: 14.4220\n",
      "Epoch 268/450\n",
      "1/1 [==============================] - 0s 907us/step - loss: 205.7438 - root_mean_squared_error: 14.3438\n",
      "Epoch 269/450\n",
      "1/1 [==============================] - 0s 831us/step - loss: 203.5071 - root_mean_squared_error: 14.2656\n",
      "Epoch 270/450\n",
      "1/1 [==============================] - 0s 961us/step - loss: 201.2833 - root_mean_squared_error: 14.1874\n",
      "Epoch 271/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 199.0724 - root_mean_squared_error: 14.1093\n",
      "Epoch 272/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 196.8745 - root_mean_squared_error: 14.0312\n",
      "Epoch 273/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 194.6895 - root_mean_squared_error: 13.9531\n",
      "Epoch 274/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 192.5174 - root_mean_squared_error: 13.8751\n",
      "Epoch 275/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 190.3582 - root_mean_squared_error: 13.7970\n",
      "Epoch 276/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 188.2119 - root_mean_squared_error: 13.7190\n",
      "Epoch 277/450\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 1ms/step - loss: 186.0785 - root_mean_squared_error: 13.6411\n",
      "Epoch 278/450\n",
      "1/1 [==============================] - 0s 873us/step - loss: 183.9580 - root_mean_squared_error: 13.5631\n",
      "Epoch 279/450\n",
      "1/1 [==============================] - 0s 818us/step - loss: 181.8504 - root_mean_squared_error: 13.4852\n",
      "Epoch 280/450\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 179.7557 - root_mean_squared_error: 13.4073\n",
      "Epoch 281/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 177.6738 - root_mean_squared_error: 13.3294\n",
      "Epoch 282/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 175.6049 - root_mean_squared_error: 13.2516\n",
      "Epoch 283/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 173.5488 - root_mean_squared_error: 13.1738\n",
      "Epoch 284/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 171.5056 - root_mean_squared_error: 13.0960\n",
      "Epoch 285/450\n",
      "1/1 [==============================] - 0s 922us/step - loss: 169.4752 - root_mean_squared_error: 13.0183\n",
      "Epoch 286/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 167.4577 - root_mean_squared_error: 12.9405\n",
      "Epoch 287/450\n",
      "1/1 [==============================] - 0s 916us/step - loss: 165.4531 - root_mean_squared_error: 12.8629\n",
      "Epoch 288/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 163.4613 - root_mean_squared_error: 12.7852\n",
      "Epoch 289/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 161.4823 - root_mean_squared_error: 12.7076\n",
      "Epoch 290/450\n",
      "1/1 [==============================] - 0s 770us/step - loss: 159.5162 - root_mean_squared_error: 12.6300\n",
      "Epoch 291/450\n",
      "1/1 [==============================] - 0s 979us/step - loss: 157.5630 - root_mean_squared_error: 12.5524\n",
      "Epoch 292/450\n",
      "1/1 [==============================] - 0s 871us/step - loss: 155.6225 - root_mean_squared_error: 12.4749\n",
      "Epoch 293/450\n",
      "1/1 [==============================] - 0s 862us/step - loss: 153.6949 - root_mean_squared_error: 12.3974\n",
      "Epoch 294/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 151.7801 - root_mean_squared_error: 12.3199\n",
      "Epoch 295/450\n",
      "1/1 [==============================] - 0s 756us/step - loss: 149.8781 - root_mean_squared_error: 12.2425\n",
      "Epoch 296/450\n",
      "1/1 [==============================] - 0s 736us/step - loss: 147.9889 - root_mean_squared_error: 12.1651\n",
      "Epoch 297/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 146.1126 - root_mean_squared_error: 12.0877\n",
      "Epoch 298/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 144.2490 - root_mean_squared_error: 12.0104\n",
      "Epoch 299/450\n",
      "1/1 [==============================] - 0s 824us/step - loss: 142.3983 - root_mean_squared_error: 11.9331\n",
      "Epoch 300/450\n",
      "1/1 [==============================] - 0s 902us/step - loss: 140.5603 - root_mean_squared_error: 11.8558\n",
      "Epoch 301/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 138.7351 - root_mean_squared_error: 11.7786\n",
      "Epoch 302/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 136.9227 - root_mean_squared_error: 11.7014\n",
      "Epoch 303/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 135.1230 - root_mean_squared_error: 11.6242\n",
      "Epoch 304/450\n",
      "1/1 [==============================] - 0s 842us/step - loss: 133.3361 - root_mean_squared_error: 11.5471\n",
      "Epoch 305/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 131.5620 - root_mean_squared_error: 11.4700\n",
      "Epoch 306/450\n",
      "1/1 [==============================] - 0s 697us/step - loss: 129.8007 - root_mean_squared_error: 11.3930\n",
      "Epoch 307/450\n",
      "1/1 [==============================] - 0s 877us/step - loss: 128.0520 - root_mean_squared_error: 11.3160\n",
      "Epoch 308/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 126.3162 - root_mean_squared_error: 11.2390\n",
      "Epoch 309/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 124.5931 - root_mean_squared_error: 11.1621\n",
      "Epoch 310/450\n",
      "1/1 [==============================] - 0s 761us/step - loss: 122.8827 - root_mean_squared_error: 11.0852\n",
      "Epoch 311/450\n",
      "1/1 [==============================] - 0s 751us/step - loss: 121.1850 - root_mean_squared_error: 11.0084\n",
      "Epoch 312/450\n",
      "1/1 [==============================] - 0s 768us/step - loss: 119.5001 - root_mean_squared_error: 10.9316\n",
      "Epoch 313/450\n",
      "1/1 [==============================] - 0s 946us/step - loss: 117.8278 - root_mean_squared_error: 10.8549\n",
      "Epoch 314/450\n",
      "1/1 [==============================] - 0s 670us/step - loss: 116.1683 - root_mean_squared_error: 10.7781\n",
      "Epoch 315/450\n",
      "1/1 [==============================] - 0s 728us/step - loss: 114.5214 - root_mean_squared_error: 10.7015\n",
      "Epoch 316/450\n",
      "1/1 [==============================] - 0s 716us/step - loss: 112.8873 - root_mean_squared_error: 10.6248\n",
      "Epoch 317/450\n",
      "1/1 [==============================] - 0s 976us/step - loss: 111.2658 - root_mean_squared_error: 10.5483\n",
      "Epoch 318/450\n",
      "1/1 [==============================] - 0s 810us/step - loss: 109.6571 - root_mean_squared_error: 10.4717\n",
      "Epoch 319/450\n",
      "1/1 [==============================] - 0s 701us/step - loss: 108.0610 - root_mean_squared_error: 10.3952\n",
      "Epoch 320/450\n",
      "1/1 [==============================] - 0s 682us/step - loss: 106.4776 - root_mean_squared_error: 10.3188\n",
      "Epoch 321/450\n",
      "1/1 [==============================] - 0s 681us/step - loss: 104.9068 - root_mean_squared_error: 10.2424\n",
      "Epoch 322/450\n",
      "1/1 [==============================] - 0s 956us/step - loss: 103.3487 - root_mean_squared_error: 10.1661\n",
      "Epoch 323/450\n",
      "1/1 [==============================] - 0s 666us/step - loss: 101.8032 - root_mean_squared_error: 10.0898\n",
      "Epoch 324/450\n",
      "1/1 [==============================] - 0s 633us/step - loss: 100.2703 - root_mean_squared_error: 10.0135\n",
      "Epoch 325/450\n",
      "1/1 [==============================] - 0s 656us/step - loss: 98.7501 - root_mean_squared_error: 9.9373\n",
      "Epoch 326/450\n",
      "1/1 [==============================] - 0s 791us/step - loss: 97.2425 - root_mean_squared_error: 9.8612\n",
      "Epoch 327/450\n",
      "1/1 [==============================] - 0s 747us/step - loss: 95.7476 - root_mean_squared_error: 9.7851\n",
      "Epoch 328/450\n",
      "1/1 [==============================] - 0s 936us/step - loss: 94.2652 - root_mean_squared_error: 9.7090\n",
      "Epoch 329/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 92.7954 - root_mean_squared_error: 9.6330\n",
      "Epoch 330/450\n",
      "1/1 [==============================] - 0s 883us/step - loss: 91.3382 - root_mean_squared_error: 9.5571\n",
      "Epoch 331/450\n",
      "1/1 [==============================] - 0s 726us/step - loss: 89.8936 - root_mean_squared_error: 9.4812\n",
      "Epoch 332/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 88.4616 - root_mean_squared_error: 9.4054\n",
      "Epoch 333/450\n",
      "1/1 [==============================] - 0s 696us/step - loss: 87.0421 - root_mean_squared_error: 9.3296\n",
      "Epoch 334/450\n",
      "1/1 [==============================] - 0s 894us/step - loss: 85.6352 - root_mean_squared_error: 9.2539\n",
      "Epoch 335/450\n",
      "1/1 [==============================] - 0s 988us/step - loss: 84.2409 - root_mean_squared_error: 9.1783\n",
      "Epoch 336/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 82.8590 - root_mean_squared_error: 9.1027\n",
      "Epoch 337/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 81.4898 - root_mean_squared_error: 9.0272\n",
      "Epoch 338/450\n",
      "1/1 [==============================] - 0s 919us/step - loss: 80.1330 - root_mean_squared_error: 8.9517\n",
      "Epoch 339/450\n",
      "1/1 [==============================] - 0s 773us/step - loss: 78.7888 - root_mean_squared_error: 8.8763\n",
      "Epoch 340/450\n",
      "1/1 [==============================] - 0s 757us/step - loss: 77.4570 - root_mean_squared_error: 8.8010\n",
      "Epoch 341/450\n",
      "1/1 [==============================] - 0s 772us/step - loss: 76.1377 - root_mean_squared_error: 8.7257\n",
      "Epoch 342/450\n",
      "1/1 [==============================] - 0s 745us/step - loss: 74.8310 - root_mean_squared_error: 8.6505\n",
      "Epoch 343/450\n",
      "1/1 [==============================] - 0s 724us/step - loss: 73.5367 - root_mean_squared_error: 8.5754\n",
      "Epoch 344/450\n",
      "1/1 [==============================] - 0s 708us/step - loss: 72.2548 - root_mean_squared_error: 8.5003\n",
      "Epoch 345/450\n",
      "1/1 [==============================] - 0s 749us/step - loss: 70.9854 - root_mean_squared_error: 8.4253\n",
      "Epoch 346/450\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 786us/step - loss: 69.7285 - root_mean_squared_error: 8.3504\n",
      "Epoch 347/450\n",
      "1/1 [==============================] - 0s 879us/step - loss: 68.4840 - root_mean_squared_error: 8.2755\n",
      "Epoch 348/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 67.2519 - root_mean_squared_error: 8.2007\n",
      "Epoch 349/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 66.0322 - root_mean_squared_error: 8.1260\n",
      "Epoch 350/450\n",
      "1/1 [==============================] - 0s 909us/step - loss: 64.8249 - root_mean_squared_error: 8.0514\n",
      "Epoch 351/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 63.6300 - root_mean_squared_error: 7.9768\n",
      "Epoch 352/450\n",
      "1/1 [==============================] - 0s 987us/step - loss: 62.4475 - root_mean_squared_error: 7.9024\n",
      "Epoch 353/450\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 61.2774 - root_mean_squared_error: 7.8280\n",
      "Epoch 354/450\n",
      "1/1 [==============================] - 0s 985us/step - loss: 60.1196 - root_mean_squared_error: 7.7537\n",
      "Epoch 355/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 58.9741 - root_mean_squared_error: 7.6795\n",
      "Epoch 356/450\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 57.8410 - root_mean_squared_error: 7.6053\n",
      "Epoch 357/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 56.7202 - root_mean_squared_error: 7.5313\n",
      "Epoch 358/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 55.6117 - root_mean_squared_error: 7.4573\n",
      "Epoch 359/450\n",
      "1/1 [==============================] - 0s 925us/step - loss: 54.5154 - root_mean_squared_error: 7.3835\n",
      "Epoch 360/450\n",
      "1/1 [==============================] - 0s 752us/step - loss: 53.4315 - root_mean_squared_error: 7.3097\n",
      "Epoch 361/450\n",
      "1/1 [==============================] - 0s 842us/step - loss: 52.3598 - root_mean_squared_error: 7.2360\n",
      "Epoch 362/450\n",
      "1/1 [==============================] - 0s 845us/step - loss: 51.3004 - root_mean_squared_error: 7.1624\n",
      "Epoch 363/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 50.2532 - root_mean_squared_error: 7.0889\n",
      "Epoch 364/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 49.2182 - root_mean_squared_error: 7.0156\n",
      "Epoch 365/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 48.1954 - root_mean_squared_error: 6.9423\n",
      "Epoch 366/450\n",
      "1/1 [==============================] - 0s 908us/step - loss: 47.1849 - root_mean_squared_error: 6.8691\n",
      "Epoch 367/450\n",
      "1/1 [==============================] - 0s 955us/step - loss: 46.1865 - root_mean_squared_error: 6.7961\n",
      "Epoch 368/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 45.2002 - root_mean_squared_error: 6.7231\n",
      "Epoch 369/450\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 44.2261 - root_mean_squared_error: 6.6503\n",
      "Epoch 370/450\n",
      "1/1 [==============================] - 0s 944us/step - loss: 43.2642 - root_mean_squared_error: 6.5776\n",
      "Epoch 371/450\n",
      "1/1 [==============================] - 0s 924us/step - loss: 42.3143 - root_mean_squared_error: 6.5049\n",
      "Epoch 372/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 41.3766 - root_mean_squared_error: 6.4325\n",
      "Epoch 373/450\n",
      "1/1 [==============================] - 0s 857us/step - loss: 40.4509 - root_mean_squared_error: 6.3601\n",
      "Epoch 374/450\n",
      "1/1 [==============================] - 0s 956us/step - loss: 39.5373 - root_mean_squared_error: 6.2879\n",
      "Epoch 375/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 38.6357 - root_mean_squared_error: 6.2158\n",
      "Epoch 376/450\n",
      "1/1 [==============================] - 0s 940us/step - loss: 37.7462 - root_mean_squared_error: 6.1438\n",
      "Epoch 377/450\n",
      "1/1 [==============================] - 0s 959us/step - loss: 36.8686 - root_mean_squared_error: 6.0720\n",
      "Epoch 378/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 36.0031 - root_mean_squared_error: 6.0003\n",
      "Epoch 379/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 35.1495 - root_mean_squared_error: 5.9287\n",
      "Epoch 380/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 34.3079 - root_mean_squared_error: 5.8573\n",
      "Epoch 381/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 33.4782 - root_mean_squared_error: 5.7860\n",
      "Epoch 382/450\n",
      "1/1 [==============================] - 0s 826us/step - loss: 32.6604 - root_mean_squared_error: 5.7149\n",
      "Epoch 383/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 31.8545 - root_mean_squared_error: 5.6440\n",
      "Epoch 384/450\n",
      "1/1 [==============================] - 0s 978us/step - loss: 31.0604 - root_mean_squared_error: 5.5732\n",
      "Epoch 385/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 30.2782 - root_mean_squared_error: 5.5026\n",
      "Epoch 386/450\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 29.5078 - root_mean_squared_error: 5.4321\n",
      "Epoch 387/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 28.7492 - root_mean_squared_error: 5.3618\n",
      "Epoch 388/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 28.0024 - root_mean_squared_error: 5.2917\n",
      "Epoch 389/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 27.2674 - root_mean_squared_error: 5.2218\n",
      "Epoch 390/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 26.5440 - root_mean_squared_error: 5.1521\n",
      "Epoch 391/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 25.8324 - root_mean_squared_error: 5.0826\n",
      "Epoch 392/450\n",
      "1/1 [==============================] - 0s 727us/step - loss: 25.1324 - root_mean_squared_error: 5.0132\n",
      "Epoch 393/450\n",
      "1/1 [==============================] - 0s 766us/step - loss: 24.4441 - root_mean_squared_error: 4.9441\n",
      "Epoch 394/450\n",
      "1/1 [==============================] - 0s 716us/step - loss: 23.7674 - root_mean_squared_error: 4.8752\n",
      "Epoch 395/450\n",
      "1/1 [==============================] - 0s 711us/step - loss: 23.1023 - root_mean_squared_error: 4.8065\n",
      "Epoch 396/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 22.4487 - root_mean_squared_error: 4.7380\n",
      "Epoch 397/450\n",
      "1/1 [==============================] - 0s 745us/step - loss: 21.8067 - root_mean_squared_error: 4.6698\n",
      "Epoch 398/450\n",
      "1/1 [==============================] - 0s 723us/step - loss: 21.1761 - root_mean_squared_error: 4.6018\n",
      "Epoch 399/450\n",
      "1/1 [==============================] - 0s 703us/step - loss: 20.5571 - root_mean_squared_error: 4.5340\n",
      "Epoch 400/450\n",
      "1/1 [==============================] - 0s 665us/step - loss: 19.9495 - root_mean_squared_error: 4.4665\n",
      "Epoch 401/450\n",
      "1/1 [==============================] - 0s 702us/step - loss: 19.3532 - root_mean_squared_error: 4.3992\n",
      "Epoch 402/450\n",
      "1/1 [==============================] - 0s 677us/step - loss: 18.7684 - root_mean_squared_error: 4.3322\n",
      "Epoch 403/450\n",
      "1/1 [==============================] - 0s 848us/step - loss: 18.1948 - root_mean_squared_error: 4.2655\n",
      "Epoch 404/450\n",
      "1/1 [==============================] - 0s 897us/step - loss: 17.6326 - root_mean_squared_error: 4.1991\n",
      "Epoch 405/450\n",
      "1/1 [==============================] - 0s 728us/step - loss: 17.0816 - root_mean_squared_error: 4.1330\n",
      "Epoch 406/450\n",
      "1/1 [==============================] - 0s 800us/step - loss: 16.5419 - root_mean_squared_error: 4.0672\n",
      "Epoch 407/450\n",
      "1/1 [==============================] - 0s 735us/step - loss: 16.0133 - root_mean_squared_error: 4.0017\n",
      "Epoch 408/450\n",
      "1/1 [==============================] - 0s 898us/step - loss: 15.4959 - root_mean_squared_error: 3.9365\n",
      "Epoch 409/450\n",
      "1/1 [==============================] - 0s 745us/step - loss: 14.9896 - root_mean_squared_error: 3.8716\n",
      "Epoch 410/450\n",
      "1/1 [==============================] - 0s 972us/step - loss: 14.4944 - root_mean_squared_error: 3.8071\n",
      "Epoch 411/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 14.0101 - root_mean_squared_error: 3.7430\n",
      "Epoch 412/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 13.5368 - root_mean_squared_error: 3.6792\n",
      "Epoch 413/450\n",
      "1/1 [==============================] - 0s 988us/step - loss: 13.0745 - root_mean_squared_error: 3.6159\n",
      "Epoch 414/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 12.6230 - root_mean_squared_error: 3.5529\n",
      "Epoch 415/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 12.1824 - root_mean_squared_error: 3.4903\n",
      "Epoch 416/450\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 1ms/step - loss: 11.7525 - root_mean_squared_error: 3.4282\n",
      "Epoch 417/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 11.3334 - root_mean_squared_error: 3.3665\n",
      "Epoch 418/450\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 10.9249 - root_mean_squared_error: 3.3053\n",
      "Epoch 419/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 10.5270 - root_mean_squared_error: 3.2445\n",
      "Epoch 420/450\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 10.1397 - root_mean_squared_error: 3.1843\n",
      "Epoch 421/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 9.7628 - root_mean_squared_error: 3.1246\n",
      "Epoch 422/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 9.3964 - root_mean_squared_error: 3.0654\n",
      "Epoch 423/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 9.0404 - root_mean_squared_error: 3.0067\n",
      "Epoch 424/450\n",
      "1/1 [==============================] - 0s 936us/step - loss: 8.6946 - root_mean_squared_error: 2.9487\n",
      "Epoch 425/450\n",
      "1/1 [==============================] - 0s 856us/step - loss: 8.3591 - root_mean_squared_error: 2.8912\n",
      "Epoch 426/450\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 8.0337 - root_mean_squared_error: 2.8344\n",
      "Epoch 427/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 7.7184 - root_mean_squared_error: 2.7782\n",
      "Epoch 428/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 7.4131 - root_mean_squared_error: 2.7227\n",
      "Epoch 429/450\n",
      "1/1 [==============================] - 0s 699us/step - loss: 7.1178 - root_mean_squared_error: 2.6679\n",
      "Epoch 430/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 6.8322 - root_mean_squared_error: 2.6139\n",
      "Epoch 431/450\n",
      "1/1 [==============================] - 0s 937us/step - loss: 6.5565 - root_mean_squared_error: 2.5606\n",
      "Epoch 432/450\n",
      "1/1 [==============================] - 0s 658us/step - loss: 6.2903 - root_mean_squared_error: 2.5081\n",
      "Epoch 433/450\n",
      "1/1 [==============================] - 0s 779us/step - loss: 6.0338 - root_mean_squared_error: 2.4564\n",
      "Epoch 434/450\n",
      "1/1 [==============================] - 0s 922us/step - loss: 5.7867 - root_mean_squared_error: 2.4056\n",
      "Epoch 435/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 5.5490 - root_mean_squared_error: 2.3556\n",
      "Epoch 436/450\n",
      "1/1 [==============================] - 0s 718us/step - loss: 5.3205 - root_mean_squared_error: 2.3066\n",
      "Epoch 437/450\n",
      "1/1 [==============================] - 0s 705us/step - loss: 5.1013 - root_mean_squared_error: 2.2586\n",
      "Epoch 438/450\n",
      "1/1 [==============================] - 0s 781us/step - loss: 4.8910 - root_mean_squared_error: 2.2116\n",
      "Epoch 439/450\n",
      "1/1 [==============================] - 0s 868us/step - loss: 4.6897 - root_mean_squared_error: 2.1656\n",
      "Epoch 440/450\n",
      "1/1 [==============================] - 0s 665us/step - loss: 4.4972 - root_mean_squared_error: 2.1207\n",
      "Epoch 441/450\n",
      "1/1 [==============================] - 0s 921us/step - loss: 4.3133 - root_mean_squared_error: 2.0769\n",
      "Epoch 442/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 4.1380 - root_mean_squared_error: 2.0342\n",
      "Epoch 443/450\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 3.9711 - root_mean_squared_error: 1.9928\n",
      "Epoch 444/450\n",
      "1/1 [==============================] - 0s 652us/step - loss: 3.8125 - root_mean_squared_error: 1.9526\n",
      "Epoch 445/450\n",
      "1/1 [==============================] - 0s 770us/step - loss: 3.6619 - root_mean_squared_error: 1.9136\n",
      "Epoch 446/450\n",
      "1/1 [==============================] - 0s 848us/step - loss: 3.5193 - root_mean_squared_error: 1.8760\n",
      "Epoch 447/450\n",
      "1/1 [==============================] - 0s 836us/step - loss: 3.3845 - root_mean_squared_error: 1.8397\n",
      "Epoch 448/450\n",
      "1/1 [==============================] - 0s 873us/step - loss: 3.2573 - root_mean_squared_error: 1.8048\n",
      "Epoch 449/450\n",
      "1/1 [==============================] - 0s 675us/step - loss: 3.1375 - root_mean_squared_error: 1.7713\n",
      "Epoch 450/450\n",
      "1/1 [==============================] - 0s 692us/step - loss: 3.0249 - root_mean_squared_error: 1.7392\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/eric/.conda/envs/mlbase/lib/python3.8/site-packages/numpy/core/_asarray.py:136: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray\n",
      "  return array(a, dtype, copy=False, order=order, subok=True)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#@title Double-click to view a possible s \n",
    "epochs=450\n",
    "my_batch_size=12 \n",
    "\n",
    "my_model = build_model(learning_rate)\n",
    "trained_weight, trained_bias, epochs, rmse = train_model(my_model, my_feature, \n",
    "                                                         my_label, epochs,\n",
    "                                                         my_batch_size)\n",
    "plot_the_model(trained_weight, trained_bias, my_feature, my_label)\n",
    "plot_the_loss_curve(epochs, rmse)\n",
    "\n",
    "# The loss curve suggests that the model does converge."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "0KmzfFB5zwvd"
   },
   "source": [
    "## Task 3: Increase the learning rate\n",
    "\n",
    "In Task 2, you increased the number of epochs to get the model to converge. Sometimes, you can get the model to converge more quickly by increasing the learning rate. However, setting the learning rate too high often makes it impossible for a model to converge. In Task 3, we've intentionally set the learning rate too high. Run the following code cell and see what happens."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "eD1hTmdd0uCo"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/500\n",
      "1/1 [==============================] - 0s 854us/step - loss: 1142.0425 - root_mean_squared_error: 33.7941\n",
      "Epoch 2/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 6641472.0000 - root_mean_squared_error: 2577.1055\n",
      "Epoch 3/500\n",
      "1/1 [==============================] - 0s 938us/step - loss: 1128.4662 - root_mean_squared_error: 33.5927\n",
      "Epoch 4/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 6.8025 - root_mean_squared_error: 2.6082\n",
      "Epoch 5/500\n",
      "1/1 [==============================] - 0s 743us/step - loss: 1.6720 - root_mean_squared_error: 1.2931\n",
      "Epoch 6/500\n",
      "1/1 [==============================] - 0s 724us/step - loss: 1.5497 - root_mean_squared_error: 1.2449\n",
      "Epoch 7/500\n",
      "1/1 [==============================] - 0s 810us/step - loss: 1.5038 - root_mean_squared_error: 1.2263\n",
      "Epoch 8/500\n",
      "1/1 [==============================] - 0s 810us/step - loss: 1.4610 - root_mean_squared_error: 1.2087\n",
      "Epoch 9/500\n",
      "1/1 [==============================] - 0s 909us/step - loss: 1.4191 - root_mean_squared_error: 1.1913\n",
      "Epoch 10/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.3782 - root_mean_squared_error: 1.1740\n",
      "Epoch 11/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.3383 - root_mean_squared_error: 1.1569\n",
      "Epoch 12/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.2997 - root_mean_squared_error: 1.1400\n",
      "Epoch 13/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.2624 - root_mean_squared_error: 1.1236\n",
      "Epoch 14/500\n",
      "1/1 [==============================] - 0s 730us/step - loss: 1.2266 - root_mean_squared_error: 1.1075\n",
      "Epoch 15/500\n",
      "1/1 [==============================] - 0s 742us/step - loss: 1.1924 - root_mean_squared_error: 1.0920\n",
      "Epoch 16/500\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 1.1599 - root_mean_squared_error: 1.0770\n",
      "Epoch 17/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.1292 - root_mean_squared_error: 1.0626\n",
      "Epoch 18/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.1003 - root_mean_squared_error: 1.0490\n",
      "Epoch 19/500\n",
      "1/1 [==============================] - 0s 947us/step - loss: 1.0734 - root_mean_squared_error: 1.0361\n",
      "Epoch 20/500\n",
      "1/1 [==============================] - 0s 731us/step - loss: 1.0485 - root_mean_squared_error: 1.0240\n",
      "Epoch 21/500\n",
      "1/1 [==============================] - 0s 743us/step - loss: 1.0256 - root_mean_squared_error: 1.0127\n",
      "Epoch 22/500\n",
      "1/1 [==============================] - 0s 899us/step - loss: 1.0046 - root_mean_squared_error: 1.0023\n",
      "Epoch 23/500\n",
      "1/1 [==============================] - 0s 804us/step - loss: 0.9857 - root_mean_squared_error: 0.9928\n",
      "Epoch 24/500\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 0.9687 - root_mean_squared_error: 0.9842\n",
      "Epoch 25/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.9536 - root_mean_squared_error: 0.9765\n",
      "Epoch 26/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.9407 - root_mean_squared_error: 0.9699\n",
      "Epoch 27/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.9319 - root_mean_squared_error: 0.9653\n",
      "Epoch 28/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.9442 - root_mean_squared_error: 0.9717\n",
      "Epoch 29/500\n",
      "1/1 [==============================] - 0s 3ms/step - loss: 1.1394 - root_mean_squared_error: 1.0674\n",
      "Epoch 30/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 3.2535 - root_mean_squared_error: 1.8037\n",
      "Epoch 31/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 28.5171 - root_mean_squared_error: 5.3401\n",
      "Epoch 32/500\n",
      "1/1 [==============================] - 0s 777us/step - loss: 372.3106 - root_mean_squared_error: 19.2954\n",
      "Epoch 33/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 5697.3716 - root_mean_squared_error: 75.4809\n",
      "Epoch 34/500\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 96930.3047 - root_mean_squared_error: 311.3363\n",
      "Epoch 35/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1215697.3750 - root_mean_squared_error: 1102.5867\n",
      "Epoch 36/500\n",
      "1/1 [==============================] - 0s 815us/step - loss: 1531748.8750 - root_mean_squared_error: 1237.6384\n",
      "Epoch 37/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 438191.4688 - root_mean_squared_error: 661.9603\n",
      "Epoch 38/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 108170.2578 - root_mean_squared_error: 328.8925\n",
      "Epoch 39/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 32430.2754 - root_mean_squared_error: 180.0841\n",
      "Epoch 40/500\n",
      "1/1 [==============================] - 0s 987us/step - loss: 12507.9482 - root_mean_squared_error: 111.8389\n",
      "Epoch 41/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 6202.2422 - root_mean_squared_error: 78.7543\n",
      "Epoch 42/500\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 3907.5085 - root_mean_squared_error: 62.5101\n",
      "Epoch 43/500\n",
      "1/1 [==============================] - 0s 965us/step - loss: 3086.9395 - root_mean_squared_error: 55.5602\n",
      "Epoch 44/500\n",
      "1/1 [==============================] - 0s 992us/step - loss: 3020.5139 - root_mean_squared_error: 54.9592\n",
      "Epoch 45/500\n",
      "1/1 [==============================] - 0s 3ms/step - loss: 3619.7830 - root_mean_squared_error: 60.1646\n",
      "Epoch 46/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 5257.3481 - root_mean_squared_error: 72.5076\n",
      "Epoch 47/500\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 9154.9795 - root_mean_squared_error: 95.6817\n",
      "Epoch 48/500\n",
      "1/1 [==============================] - 0s 822us/step - loss: 18861.0859 - root_mean_squared_error: 137.3357\n",
      "Epoch 49/500\n",
      "1/1 [==============================] - 0s 984us/step - loss: 44963.7188 - root_mean_squared_error: 212.0465\n",
      "Epoch 50/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 118032.3750 - root_mean_squared_error: 343.5584\n",
      "Epoch 51/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 300587.2188 - root_mean_squared_error: 548.2584\n",
      "Epoch 52/500\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 566225.5625 - root_mean_squared_error: 752.4796\n",
      "Epoch 53/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 599278.1875 - root_mean_squared_error: 774.1306\n",
      "Epoch 54/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 389363.8438 - root_mean_squared_error: 623.9902\n",
      "Epoch 55/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 209097.2031 - root_mean_squared_error: 457.2715\n",
      "Epoch 56/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 114408.9609 - root_mean_squared_error: 338.2439\n",
      "Epoch 57/500\n",
      "1/1 [==============================] - 0s 960us/step - loss: 70348.6953 - root_mean_squared_error: 265.2333\n",
      "Epoch 58/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 50498.2070 - root_mean_squared_error: 224.7181\n",
      "Epoch 59/500\n",
      "1/1 [==============================] - 0s 974us/step - loss: 42777.5977 - root_mean_squared_error: 206.8275\n",
      "Epoch 60/500\n",
      "1/1 [==============================] - 0s 915us/step - loss: 42668.9258 - root_mean_squared_error: 206.5646\n",
      "Epoch 61/500\n",
      "1/1 [==============================] - 0s 814us/step - loss: 49600.6562 - root_mean_squared_error: 222.7121\n",
      "Epoch 62/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 65975.4844 - root_mean_squared_error: 256.8569\n",
      "Epoch 63/500\n",
      "1/1 [==============================] - 0s 878us/step - loss: 97432.3125 - root_mean_squared_error: 312.1415\n",
      "Epoch 64/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 152011.1562 - root_mean_squared_error: 389.8861\n",
      "Epoch 65/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 231780.7969 - root_mean_squared_error: 481.4362\n",
      "Epoch 66/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 312238.2500 - root_mean_squared_error: 558.7828\n",
      "Epoch 67/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 342454.7812 - root_mean_squared_error: 585.1964\n",
      "Epoch 68/500\n",
      "1/1 [==============================] - 0s 797us/step - loss: 304327.9062 - root_mean_squared_error: 551.6592\n",
      "Epoch 69/500\n",
      "1/1 [==============================] - 0s 905us/step - loss: 235348.2969 - root_mean_squared_error: 485.1271\n",
      "Epoch 70/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 661us/step - loss: 173701.7031 - root_mean_squared_error: 416.7754\n",
      "Epoch 71/500\n",
      "1/1 [==============================] - 0s 810us/step - loss: 131563.6719 - root_mean_squared_error: 362.7171\n",
      "Epoch 72/500\n",
      "1/1 [==============================] - 0s 868us/step - loss: 106906.8516 - root_mean_squared_error: 326.9661\n",
      "Epoch 73/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 95266.0000 - root_mean_squared_error: 308.6519\n",
      "Epoch 74/500\n",
      "1/1 [==============================] - 0s 763us/step - loss: 93661.1953 - root_mean_squared_error: 306.0412\n",
      "Epoch 75/500\n",
      "1/1 [==============================] - 0s 867us/step - loss: 101004.1641 - root_mean_squared_error: 317.8115\n",
      "Epoch 76/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 117548.6016 - root_mean_squared_error: 342.8536\n",
      "Epoch 77/500\n",
      "1/1 [==============================] - 0s 942us/step - loss: 143737.0781 - root_mean_squared_error: 379.1267\n",
      "Epoch 78/500\n",
      "1/1 [==============================] - 0s 752us/step - loss: 177951.9219 - root_mean_squared_error: 421.8435\n",
      "Epoch 79/500\n",
      "1/1 [==============================] - 0s 745us/step - loss: 213538.8281 - root_mean_squared_error: 462.1026\n",
      "Epoch 80/500\n",
      "1/1 [==============================] - 0s 754us/step - loss: 238653.6250 - root_mean_squared_error: 488.5219\n",
      "Epoch 81/500\n",
      "1/1 [==============================] - 0s 986us/step - loss: 243139.3281 - root_mean_squared_error: 493.0916\n",
      "Epoch 82/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 227143.4219 - root_mean_squared_error: 476.5957\n",
      "Epoch 83/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 200176.4375 - root_mean_squared_error: 447.4108\n",
      "Epoch 84/500\n",
      "1/1 [==============================] - 0s 750us/step - loss: 172678.6406 - root_mean_squared_error: 415.5462\n",
      "Epoch 85/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 150753.4219 - root_mean_squared_error: 388.2698\n",
      "Epoch 86/500\n",
      "1/1 [==============================] - 0s 836us/step - loss: 136416.9844 - root_mean_squared_error: 369.3467\n",
      "Epoch 87/500\n",
      "1/1 [==============================] - 0s 723us/step - loss: 129657.7422 - root_mean_squared_error: 360.0802\n",
      "Epoch 88/500\n",
      "1/1 [==============================] - 0s 807us/step - loss: 129892.7812 - root_mean_squared_error: 360.4064\n",
      "Epoch 89/500\n",
      "1/1 [==============================] - 0s 700us/step - loss: 136483.9219 - root_mean_squared_error: 369.4373\n",
      "Epoch 90/500\n",
      "1/1 [==============================] - 0s 699us/step - loss: 148593.8906 - root_mean_squared_error: 385.4788\n",
      "Epoch 91/500\n",
      "1/1 [==============================] - 0s 834us/step - loss: 164655.1094 - root_mean_squared_error: 405.7772\n",
      "Epoch 92/500\n",
      "1/1 [==============================] - 0s 687us/step - loss: 181822.7656 - root_mean_squared_error: 426.4068\n",
      "Epoch 93/500\n",
      "1/1 [==============================] - 0s 726us/step - loss: 196076.5625 - root_mean_squared_error: 442.8053\n",
      "Epoch 94/500\n",
      "1/1 [==============================] - 0s 665us/step - loss: 203536.2656 - root_mean_squared_error: 451.1499\n",
      "Epoch 95/500\n",
      "1/1 [==============================] - 0s 656us/step - loss: 202436.8281 - root_mean_squared_error: 449.9298\n",
      "Epoch 96/500\n",
      "1/1 [==============================] - 0s 682us/step - loss: 194065.0469 - root_mean_squared_error: 440.5281\n",
      "Epoch 97/500\n",
      "1/1 [==============================] - 0s 698us/step - loss: 181773.2969 - root_mean_squared_error: 426.3488\n",
      "Epoch 98/500\n",
      "1/1 [==============================] - 0s 706us/step - loss: 169136.7656 - root_mean_squared_error: 411.2624\n",
      "Epoch 99/500\n",
      "1/1 [==============================] - 0s 808us/step - loss: 158742.5781 - root_mean_squared_error: 398.4251\n",
      "Epoch 100/500\n",
      "1/1 [==============================] - 0s 706us/step - loss: 151971.4062 - root_mean_squared_error: 389.8351\n",
      "Epoch 101/500\n",
      "1/1 [==============================] - 0s 682us/step - loss: 149296.6719 - root_mean_squared_error: 386.3893\n",
      "Epoch 102/500\n",
      "1/1 [==============================] - 0s 790us/step - loss: 150613.2969 - root_mean_squared_error: 388.0893\n",
      "Epoch 103/500\n",
      "1/1 [==============================] - 0s 683us/step - loss: 155395.5625 - root_mean_squared_error: 394.2025\n",
      "Epoch 104/500\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 162697.2500 - root_mean_squared_error: 403.3575\n",
      "Epoch 105/500\n",
      "1/1 [==============================] - 0s 680us/step - loss: 171118.8750 - root_mean_squared_error: 413.6652\n",
      "Epoch 106/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 178916.2344 - root_mean_squared_error: 422.9849\n",
      "Epoch 107/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 184368.2969 - root_mean_squared_error: 429.3813\n",
      "Epoch 108/500\n",
      "1/1 [==============================] - 0s 725us/step - loss: 186324.7031 - root_mean_squared_error: 431.6534\n",
      "Epoch 109/500\n",
      "1/1 [==============================] - 0s 890us/step - loss: 184623.8125 - root_mean_squared_error: 429.6787\n",
      "Epoch 110/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 180090.9531 - root_mean_squared_error: 424.3712\n",
      "Epoch 111/500\n",
      "1/1 [==============================] - 0s 725us/step - loss: 174131.0938 - root_mean_squared_error: 417.2902\n",
      "Epoch 112/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 168205.1406 - root_mean_squared_error: 410.1282\n",
      "Epoch 113/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 163463.0469 - root_mean_squared_error: 404.3056\n",
      "Epoch 114/500\n",
      "1/1 [==============================] - 0s 748us/step - loss: 160606.6406 - root_mean_squared_error: 400.7576\n",
      "Epoch 115/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 159903.9844 - root_mean_squared_error: 399.8800\n",
      "Epoch 116/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 161252.7031 - root_mean_squared_error: 401.5628\n",
      "Epoch 117/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 164238.8906 - root_mean_squared_error: 405.2640\n",
      "Epoch 118/500\n",
      "1/1 [==============================] - 0s 790us/step - loss: 168194.3594 - root_mean_squared_error: 410.1151\n",
      "Epoch 119/500\n",
      "1/1 [==============================] - 0s 888us/step - loss: 172286.0469 - root_mean_squared_error: 415.0735\n",
      "Epoch 120/500\n",
      "1/1 [==============================] - 0s 790us/step - loss: 175664.8906 - root_mean_squared_error: 419.1240\n",
      "Epoch 121/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 177660.2344 - root_mean_squared_error: 421.4976\n",
      "Epoch 122/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 177953.7500 - root_mean_squared_error: 421.8456\n",
      "Epoch 123/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 176650.2500 - root_mean_squared_error: 420.2978\n",
      "Epoch 124/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 174209.4375 - root_mean_squared_error: 417.3840\n",
      "Epoch 125/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 171280.1094 - root_mean_squared_error: 413.8600\n",
      "Epoch 126/500\n",
      "1/1 [==============================] - 0s 821us/step - loss: 168515.2656 - root_mean_squared_error: 410.5061\n",
      "Epoch 127/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 166434.4531 - root_mean_squared_error: 407.9638\n",
      "Epoch 128/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 165352.8750 - root_mean_squared_error: 406.6360\n",
      "Epoch 129/500\n",
      "1/1 [==============================] - 0s 950us/step - loss: 165364.7031 - root_mean_squared_error: 406.6506\n",
      "Epoch 130/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 166359.6250 - root_mean_squared_error: 407.8721\n",
      "Epoch 131/500\n",
      "1/1 [==============================] - 0s 850us/step - loss: 168060.1875 - root_mean_squared_error: 409.9514\n",
      "Epoch 132/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170076.5000 - root_mean_squared_error: 412.4033\n",
      "Epoch 133/500\n",
      "1/1 [==============================] - 0s 762us/step - loss: 171980.6406 - root_mean_squared_error: 414.7055\n",
      "Epoch 134/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 173393.1094 - root_mean_squared_error: 416.4050\n",
      "Epoch 135/500\n",
      "1/1 [==============================] - 0s 804us/step - loss: 174064.3125 - root_mean_squared_error: 417.2101\n",
      "Epoch 136/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 173924.3906 - root_mean_squared_error: 417.0424\n",
      "Epoch 137/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 2ms/step - loss: 173085.2500 - root_mean_squared_error: 416.0352\n",
      "Epoch 138/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 171795.5000 - root_mean_squared_error: 414.4822\n",
      "Epoch 139/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170367.6562 - root_mean_squared_error: 412.7562\n",
      "Epoch 140/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 169102.5000 - root_mean_squared_error: 411.2207\n",
      "Epoch 141/500\n",
      "1/1 [==============================] - 0s 896us/step - loss: 168230.8281 - root_mean_squared_error: 410.1595\n",
      "Epoch 142/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 167880.0781 - root_mean_squared_error: 409.7317\n",
      "Epoch 143/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 168064.7656 - root_mean_squared_error: 409.9570\n",
      "Epoch 144/500\n",
      "1/1 [==============================] - 0s 651us/step - loss: 168696.4844 - root_mean_squared_error: 410.7268\n",
      "Epoch 145/500\n",
      "1/1 [==============================] - 0s 675us/step - loss: 169609.6406 - root_mean_squared_error: 411.8369\n",
      "Epoch 146/500\n",
      "1/1 [==============================] - 0s 659us/step - loss: 170597.5469 - root_mean_squared_error: 413.0345\n",
      "Epoch 147/500\n",
      "1/1 [==============================] - 0s 632us/step - loss: 171455.2344 - root_mean_squared_error: 414.0715\n",
      "Epoch 148/500\n",
      "1/1 [==============================] - 0s 789us/step - loss: 172020.9844 - root_mean_squared_error: 414.7541\n",
      "Epoch 149/500\n",
      "1/1 [==============================] - 0s 685us/step - loss: 172207.1719 - root_mean_squared_error: 414.9785\n",
      "Epoch 150/500\n",
      "1/1 [==============================] - 0s 743us/step - loss: 172013.3281 - root_mean_squared_error: 414.7449\n",
      "Epoch 151/500\n",
      "1/1 [==============================] - 0s 649us/step - loss: 171517.6406 - root_mean_squared_error: 414.1469\n",
      "Epoch 152/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170852.1406 - root_mean_squared_error: 413.3427\n",
      "Epoch 153/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170168.6875 - root_mean_squared_error: 412.5151\n",
      "Epoch 154/500\n",
      "1/1 [==============================] - 0s 765us/step - loss: 169605.5000 - root_mean_squared_error: 411.8319\n",
      "Epoch 155/500\n",
      "1/1 [==============================] - 0s 701us/step - loss: 169261.2656 - root_mean_squared_error: 411.4137\n",
      "Epoch 156/500\n",
      "1/1 [==============================] - 0s 689us/step - loss: 169181.4219 - root_mean_squared_error: 411.3167\n",
      "Epoch 157/500\n",
      "1/1 [==============================] - 0s 703us/step - loss: 169355.1094 - root_mean_squared_error: 411.5278\n",
      "Epoch 158/500\n",
      "1/1 [==============================] - 0s 944us/step - loss: 169723.0625 - root_mean_squared_error: 411.9746\n",
      "Epoch 159/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170193.7031 - root_mean_squared_error: 412.5454\n",
      "Epoch 160/500\n",
      "1/1 [==============================] - 0s 699us/step - loss: 170663.2344 - root_mean_squared_error: 413.1141\n",
      "Epoch 161/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 171037.5625 - root_mean_squared_error: 413.5669\n",
      "Epoch 162/500\n",
      "1/1 [==============================] - 0s 964us/step - loss: 171250.3906 - root_mean_squared_error: 413.8241\n",
      "Epoch 163/500\n",
      "1/1 [==============================] - 0s 885us/step - loss: 171274.6250 - root_mean_squared_error: 413.8534\n",
      "Epoch 164/500\n",
      "1/1 [==============================] - 0s 823us/step - loss: 171124.2812 - root_mean_squared_error: 413.6717\n",
      "Epoch 165/500\n",
      "1/1 [==============================] - 0s 677us/step - loss: 170847.4375 - root_mean_squared_error: 413.3369\n",
      "Epoch 166/500\n",
      "1/1 [==============================] - 0s 903us/step - loss: 170512.5156 - root_mean_squared_error: 412.9316\n",
      "Epoch 167/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170192.7031 - root_mean_squared_error: 412.5442\n",
      "Epoch 168/500\n",
      "1/1 [==============================] - 0s 678us/step - loss: 169950.2031 - root_mean_squared_error: 412.2502\n",
      "Epoch 169/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 169825.3125 - root_mean_squared_error: 412.0987\n",
      "Epoch 170/500\n",
      "1/1 [==============================] - 0s 713us/step - loss: 169830.9375 - root_mean_squared_error: 412.1055\n",
      "Epoch 171/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 169952.9531 - root_mean_squared_error: 412.2535\n",
      "Epoch 172/500\n",
      "1/1 [==============================] - 0s 922us/step - loss: 170155.6719 - root_mean_squared_error: 412.4993\n",
      "Epoch 173/500\n",
      "1/1 [==============================] - 0s 701us/step - loss: 170390.7031 - root_mean_squared_error: 412.7841\n",
      "Epoch 174/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170607.9219 - root_mean_squared_error: 413.0471\n",
      "Epoch 175/500\n",
      "1/1 [==============================] - 0s 850us/step - loss: 170765.4219 - root_mean_squared_error: 413.2377\n",
      "Epoch 176/500\n",
      "1/1 [==============================] - 0s 722us/step - loss: 170837.5625 - root_mean_squared_error: 413.3250\n",
      "Epoch 177/500\n",
      "1/1 [==============================] - 0s 711us/step - loss: 170818.2969 - root_mean_squared_error: 413.3017\n",
      "Epoch 178/500\n",
      "1/1 [==============================] - 0s 657us/step - loss: 170720.6719 - root_mean_squared_error: 413.1836\n",
      "Epoch 179/500\n",
      "1/1 [==============================] - 0s 732us/step - loss: 170572.1406 - root_mean_squared_error: 413.0038\n",
      "Epoch 180/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170407.7031 - root_mean_squared_error: 412.8047\n",
      "Epoch 181/500\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 170261.8125 - root_mean_squared_error: 412.6279\n",
      "Epoch 182/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170161.6719 - root_mean_squared_error: 412.5066\n",
      "Epoch 183/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170122.7344 - root_mean_squared_error: 412.4594\n",
      "Epoch 184/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170146.4375 - root_mean_squared_error: 412.4881\n",
      "Epoch 185/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170222.0469 - root_mean_squared_error: 412.5797\n",
      "Epoch 186/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170329.1250 - root_mean_squared_error: 412.7095\n",
      "Epoch 187/500\n",
      "1/1 [==============================] - 0s 713us/step - loss: 170443.2031 - root_mean_squared_error: 412.8477\n",
      "Epoch 188/500\n",
      "1/1 [==============================] - 0s 938us/step - loss: 170540.8594 - root_mean_squared_error: 412.9659\n",
      "Epoch 189/500\n",
      "1/1 [==============================] - 0s 962us/step - loss: 170604.1719 - root_mean_squared_error: 413.0426\n",
      "Epoch 190/500\n",
      "1/1 [==============================] - 0s 663us/step - loss: 170623.7500 - root_mean_squared_error: 413.0663\n",
      "Epoch 191/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170600.1094 - root_mean_squared_error: 413.0377\n",
      "Epoch 192/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170542.2031 - root_mean_squared_error: 412.9676\n",
      "Epoch 193/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170465.2500 - root_mean_squared_error: 412.8744\n",
      "Epoch 194/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170386.4375 - root_mean_squared_error: 412.7789\n",
      "Epoch 195/500\n",
      "1/1 [==============================] - 0s 817us/step - loss: 170321.7656 - root_mean_squared_error: 412.7006\n",
      "Epoch 196/500\n",
      "1/1 [==============================] - 0s 816us/step - loss: 170282.7031 - root_mean_squared_error: 412.6533\n",
      "Epoch 197/500\n",
      "1/1 [==============================] - 0s 711us/step - loss: 170274.5469 - root_mean_squared_error: 412.6434\n",
      "Epoch 198/500\n",
      "1/1 [==============================] - 0s 852us/step - loss: 170295.8125 - root_mean_squared_error: 412.6691\n",
      "Epoch 199/500\n",
      "1/1 [==============================] - 0s 721us/step - loss: 170339.3281 - root_mean_squared_error: 412.7219\n",
      "Epoch 200/500\n",
      "1/1 [==============================] - 0s 758us/step - loss: 170394.0625 - root_mean_squared_error: 412.7881\n",
      "Epoch 201/500\n",
      "1/1 [==============================] - 0s 870us/step - loss: 170448.0000 - root_mean_squared_error: 412.8535\n",
      "Epoch 202/500\n",
      "1/1 [==============================] - 0s 900us/step - loss: 170490.4844 - root_mean_squared_error: 412.9049\n",
      "Epoch 203/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170514.1094 - root_mean_squared_error: 412.9335\n",
      "Epoch 204/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 728us/step - loss: 170516.0156 - root_mean_squared_error: 412.9359\n",
      "Epoch 205/500\n",
      "1/1 [==============================] - 0s 739us/step - loss: 170498.0781 - root_mean_squared_error: 412.9141\n",
      "Epoch 206/500\n",
      "1/1 [==============================] - 0s 753us/step - loss: 170465.7656 - root_mean_squared_error: 412.8750\n",
      "Epoch 207/500\n",
      "1/1 [==============================] - 0s 733us/step - loss: 170427.1094 - root_mean_squared_error: 412.8282\n",
      "Epoch 208/500\n",
      "1/1 [==============================] - 0s 768us/step - loss: 170390.4531 - root_mean_squared_error: 412.7838\n",
      "Epoch 209/500\n",
      "1/1 [==============================] - 0s 689us/step - loss: 170362.8594 - root_mean_squared_error: 412.7504\n",
      "Epoch 210/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170348.8906 - root_mean_squared_error: 412.7334\n",
      "Epoch 211/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170350.1094 - root_mean_squared_error: 412.7349\n",
      "Epoch 212/500\n",
      "1/1 [==============================] - 0s 716us/step - loss: 170364.6875 - root_mean_squared_error: 412.7526\n",
      "Epoch 213/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170388.3906 - root_mean_squared_error: 412.7813\n",
      "Epoch 214/500\n",
      "1/1 [==============================] - 0s 997us/step - loss: 170415.5000 - root_mean_squared_error: 412.8141\n",
      "Epoch 215/500\n",
      "1/1 [==============================] - 0s 849us/step - loss: 170440.2969 - root_mean_squared_error: 412.8441\n",
      "Epoch 216/500\n",
      "1/1 [==============================] - 0s 701us/step - loss: 170458.0938 - root_mean_squared_error: 412.8657\n",
      "Epoch 217/500\n",
      "1/1 [==============================] - 0s 734us/step - loss: 170466.0000 - root_mean_squared_error: 412.8753\n",
      "Epoch 218/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170463.3438 - root_mean_squared_error: 412.8721\n",
      "Epoch 219/500\n",
      "1/1 [==============================] - 0s 750us/step - loss: 170451.8281 - root_mean_squared_error: 412.8581\n",
      "Epoch 220/500\n",
      "1/1 [==============================] - 0s 648us/step - loss: 170434.5625 - root_mean_squared_error: 412.8372\n",
      "Epoch 221/500\n",
      "1/1 [==============================] - 0s 864us/step - loss: 170415.6875 - root_mean_squared_error: 412.8144\n",
      "Epoch 222/500\n",
      "1/1 [==============================] - 0s 848us/step - loss: 170399.1094 - root_mean_squared_error: 412.7943\n",
      "Epoch 223/500\n",
      "1/1 [==============================] - 0s 710us/step - loss: 170387.9219 - root_mean_squared_error: 412.7807\n",
      "Epoch 224/500\n",
      "1/1 [==============================] - 0s 766us/step - loss: 170383.7344 - root_mean_squared_error: 412.7756\n",
      "Epoch 225/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170386.6406 - root_mean_squared_error: 412.7792\n",
      "Epoch 226/500\n",
      "1/1 [==============================] - 0s 730us/step - loss: 170395.4531 - root_mean_squared_error: 412.7899\n",
      "Epoch 227/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170407.8594 - root_mean_squared_error: 412.8049\n",
      "Epoch 228/500\n",
      "1/1 [==============================] - 0s 739us/step - loss: 170420.8750 - root_mean_squared_error: 412.8206\n",
      "Epoch 229/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170432.0000 - root_mean_squared_error: 412.8341\n",
      "Epoch 230/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170439.0781 - root_mean_squared_error: 412.8427\n",
      "Epoch 231/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170441.1406 - root_mean_squared_error: 412.8452\n",
      "Epoch 232/500\n",
      "1/1 [==============================] - 0s 974us/step - loss: 170438.3594 - root_mean_squared_error: 412.8418\n",
      "Epoch 233/500\n",
      "1/1 [==============================] - 0s 930us/step - loss: 170431.5469 - root_mean_squared_error: 412.8336\n",
      "Epoch 234/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170422.6875 - root_mean_squared_error: 412.8228\n",
      "Epoch 235/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170413.6094 - root_mean_squared_error: 412.8118\n",
      "Epoch 236/500\n",
      "1/1 [==============================] - 0s 776us/step - loss: 170406.2344 - root_mean_squared_error: 412.8029\n",
      "Epoch 237/500\n",
      "1/1 [==============================] - 0s 818us/step - loss: 170401.9219 - root_mean_squared_error: 412.7977\n",
      "Epoch 238/500\n",
      "1/1 [==============================] - 0s 748us/step - loss: 170401.0938 - root_mean_squared_error: 412.7967\n",
      "Epoch 239/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170403.6719 - root_mean_squared_error: 412.7998\n",
      "Epoch 240/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170408.7656 - root_mean_squared_error: 412.8060\n",
      "Epoch 241/500\n",
      "1/1 [==============================] - 0s 734us/step - loss: 170415.0469 - root_mean_squared_error: 412.8136\n",
      "Epoch 242/500\n",
      "1/1 [==============================] - 0s 711us/step - loss: 170421.2344 - root_mean_squared_error: 412.8211\n",
      "Epoch 243/500\n",
      "1/1 [==============================] - 0s 744us/step - loss: 170426.1875 - root_mean_squared_error: 412.8271\n",
      "Epoch 244/500\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 170428.7812 - root_mean_squared_error: 412.8302\n",
      "Epoch 245/500\n",
      "1/1 [==============================] - 0s 661us/step - loss: 170428.8750 - root_mean_squared_error: 412.8303\n",
      "Epoch 246/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170426.7500 - root_mean_squared_error: 412.8278\n",
      "Epoch 247/500\n",
      "1/1 [==============================] - 0s 724us/step - loss: 170422.9531 - root_mean_squared_error: 412.8232\n",
      "Epoch 248/500\n",
      "1/1 [==============================] - 0s 979us/step - loss: 170418.4219 - root_mean_squared_error: 412.8177\n",
      "Epoch 249/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170414.1406 - root_mean_squared_error: 412.8125\n",
      "Epoch 250/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170411.0000 - root_mean_squared_error: 412.8087\n",
      "Epoch 251/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170409.4844 - root_mean_squared_error: 412.8068\n",
      "Epoch 252/500\n",
      "1/1 [==============================] - 0s 687us/step - loss: 170409.7812 - root_mean_squared_error: 412.8072\n",
      "Epoch 253/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170411.5781 - root_mean_squared_error: 412.8094\n",
      "Epoch 254/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170414.3281 - root_mean_squared_error: 412.8127\n",
      "Epoch 255/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.5469 - root_mean_squared_error: 412.8166\n",
      "Epoch 256/500\n",
      "1/1 [==============================] - 0s 711us/step - loss: 170420.3594 - root_mean_squared_error: 412.8200\n",
      "Epoch 257/500\n",
      "1/1 [==============================] - 0s 994us/step - loss: 170422.3438 - root_mean_squared_error: 412.8224\n",
      "Epoch 258/500\n",
      "1/1 [==============================] - 0s 767us/step - loss: 170423.2188 - root_mean_squared_error: 412.8235\n",
      "Epoch 259/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170422.9531 - root_mean_squared_error: 412.8232\n",
      "Epoch 260/500\n",
      "1/1 [==============================] - 0s 954us/step - loss: 170421.6250 - root_mean_squared_error: 412.8215\n",
      "Epoch 261/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170419.5469 - root_mean_squared_error: 412.8190\n",
      "Epoch 262/500\n",
      "1/1 [==============================] - 0s 808us/step - loss: 170417.2344 - root_mean_squared_error: 412.8162\n",
      "Epoch 263/500\n",
      "1/1 [==============================] - 0s 866us/step - loss: 170415.3281 - root_mean_squared_error: 412.8139\n",
      "Epoch 264/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170414.0625 - root_mean_squared_error: 412.8124\n",
      "Epoch 265/500\n",
      "1/1 [==============================] - 0s 708us/step - loss: 170413.5781 - root_mean_squared_error: 412.8118\n",
      "Epoch 266/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170413.9531 - root_mean_squared_error: 412.8123\n",
      "Epoch 267/500\n",
      "1/1 [==============================] - 0s 754us/step - loss: 170414.9531 - root_mean_squared_error: 412.8134\n",
      "Epoch 268/500\n",
      "1/1 [==============================] - 0s 703us/step - loss: 170416.2969 - root_mean_squared_error: 412.8151\n",
      "Epoch 269/500\n",
      "1/1 [==============================] - 0s 765us/step - loss: 170417.7969 - root_mean_squared_error: 412.8169\n",
      "Epoch 270/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170419.1406 - root_mean_squared_error: 412.8185\n",
      "Epoch 271/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 2ms/step - loss: 170419.9375 - root_mean_squared_error: 412.8195\n",
      "Epoch 272/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170420.2344 - root_mean_squared_error: 412.8199\n",
      "Epoch 273/500\n",
      "1/1 [==============================] - 0s 656us/step - loss: 170420.0625 - root_mean_squared_error: 412.8196\n",
      "Epoch 274/500\n",
      "1/1 [==============================] - 0s 672us/step - loss: 170419.3125 - root_mean_squared_error: 412.8187\n",
      "Epoch 275/500\n",
      "1/1 [==============================] - 0s 704us/step - loss: 170418.3281 - root_mean_squared_error: 412.8175\n",
      "Epoch 276/500\n",
      "1/1 [==============================] - 0s 700us/step - loss: 170417.2344 - root_mean_squared_error: 412.8162\n",
      "Epoch 277/500\n",
      "1/1 [==============================] - 0s 941us/step - loss: 170416.3438 - root_mean_squared_error: 412.8152\n",
      "Epoch 278/500\n",
      "1/1 [==============================] - 0s 861us/step - loss: 170415.7812 - root_mean_squared_error: 412.8145\n",
      "Epoch 279/500\n",
      "1/1 [==============================] - 0s 779us/step - loss: 170415.6719 - root_mean_squared_error: 412.8143\n",
      "Epoch 280/500\n",
      "1/1 [==============================] - 0s 877us/step - loss: 170415.9375 - root_mean_squared_error: 412.8147\n",
      "Epoch 281/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170416.4375 - root_mean_squared_error: 412.8152\n",
      "Epoch 282/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.2344 - root_mean_squared_error: 412.8162\n",
      "Epoch 283/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.9375 - root_mean_squared_error: 412.8171\n",
      "Epoch 284/500\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 170418.4844 - root_mean_squared_error: 412.8177\n",
      "Epoch 285/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170418.8750 - root_mean_squared_error: 412.8182\n",
      "Epoch 286/500\n",
      "1/1 [==============================] - 0s 993us/step - loss: 170418.9062 - root_mean_squared_error: 412.8182\n",
      "Epoch 287/500\n",
      "1/1 [==============================] - 0s 825us/step - loss: 170418.6719 - root_mean_squared_error: 412.8180\n",
      "Epoch 288/500\n",
      "1/1 [==============================] - 0s 795us/step - loss: 170418.2656 - root_mean_squared_error: 412.8175\n",
      "Epoch 289/500\n",
      "1/1 [==============================] - 0s 848us/step - loss: 170417.7500 - root_mean_squared_error: 412.8168\n",
      "Epoch 290/500\n",
      "1/1 [==============================] - 0s 819us/step - loss: 170417.1719 - root_mean_squared_error: 412.8162\n",
      "Epoch 291/500\n",
      "1/1 [==============================] - 0s 882us/step - loss: 170416.7656 - root_mean_squared_error: 412.8156\n",
      "Epoch 292/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170416.5469 - root_mean_squared_error: 412.8154\n",
      "Epoch 293/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170416.5000 - root_mean_squared_error: 412.8153\n",
      "Epoch 294/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170416.7500 - root_mean_squared_error: 412.8156\n",
      "Epoch 295/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.0781 - root_mean_squared_error: 412.8160\n",
      "Epoch 296/500\n",
      "1/1 [==============================] - 0s 915us/step - loss: 170417.4531 - root_mean_squared_error: 412.8165\n",
      "Epoch 297/500\n",
      "1/1 [==============================] - 0s 999us/step - loss: 170417.8125 - root_mean_squared_error: 412.8169\n",
      "Epoch 298/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170418.1719 - root_mean_squared_error: 412.8174\n",
      "Epoch 299/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170418.2656 - root_mean_squared_error: 412.8175\n",
      "Epoch 300/500\n",
      "1/1 [==============================] - 0s 982us/step - loss: 170418.3281 - root_mean_squared_error: 412.8175\n",
      "Epoch 301/500\n",
      "1/1 [==============================] - 0s 872us/step - loss: 170418.2656 - root_mean_squared_error: 412.8175\n",
      "Epoch 302/500\n",
      "1/1 [==============================] - 0s 941us/step - loss: 170418.0156 - root_mean_squared_error: 412.8172\n",
      "Epoch 303/500\n",
      "1/1 [==============================] - 0s 962us/step - loss: 170417.6719 - root_mean_squared_error: 412.8167\n",
      "Epoch 304/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.3594 - root_mean_squared_error: 412.8164\n",
      "Epoch 305/500\n",
      "1/1 [==============================] - 0s 803us/step - loss: 170417.1719 - root_mean_squared_error: 412.8162\n",
      "Epoch 306/500\n",
      "1/1 [==============================] - 0s 734us/step - loss: 170417.0469 - root_mean_squared_error: 412.8160\n",
      "Epoch 307/500\n",
      "1/1 [==============================] - 0s 697us/step - loss: 170416.9531 - root_mean_squared_error: 412.8159\n",
      "Epoch 308/500\n",
      "1/1 [==============================] - 0s 734us/step - loss: 170417.0312 - root_mean_squared_error: 412.8160\n",
      "Epoch 309/500\n",
      "1/1 [==============================] - 0s 690us/step - loss: 170417.2031 - root_mean_squared_error: 412.8162\n",
      "Epoch 310/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.4531 - root_mean_squared_error: 412.8165\n",
      "Epoch 311/500\n",
      "1/1 [==============================] - 0s 690us/step - loss: 170417.6719 - root_mean_squared_error: 412.8167\n",
      "Epoch 312/500\n",
      "1/1 [==============================] - 0s 765us/step - loss: 170417.8750 - root_mean_squared_error: 412.8170\n",
      "Epoch 313/500\n",
      "1/1 [==============================] - 0s 649us/step - loss: 170418.0312 - root_mean_squared_error: 412.8172\n",
      "Epoch 314/500\n",
      "1/1 [==============================] - 0s 666us/step - loss: 170417.9844 - root_mean_squared_error: 412.8171\n",
      "Epoch 315/500\n",
      "1/1 [==============================] - 0s 716us/step - loss: 170417.8750 - root_mean_squared_error: 412.8170\n",
      "Epoch 316/500\n",
      "1/1 [==============================] - 0s 738us/step - loss: 170417.7656 - root_mean_squared_error: 412.8169\n",
      "Epoch 317/500\n",
      "1/1 [==============================] - 0s 844us/step - loss: 170417.5781 - root_mean_squared_error: 412.8167\n",
      "Epoch 318/500\n",
      "1/1 [==============================] - 0s 926us/step - loss: 170417.4219 - root_mean_squared_error: 412.8164\n",
      "Epoch 319/500\n",
      "1/1 [==============================] - 0s 757us/step - loss: 170417.2969 - root_mean_squared_error: 412.8163\n",
      "Epoch 320/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.2031 - root_mean_squared_error: 412.8162\n",
      "Epoch 321/500\n",
      "1/1 [==============================] - 0s 893us/step - loss: 170417.2969 - root_mean_squared_error: 412.8163\n",
      "Epoch 322/500\n",
      "1/1 [==============================] - 0s 975us/step - loss: 170417.3594 - root_mean_squared_error: 412.8164\n",
      "Epoch 323/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.5000 - root_mean_squared_error: 412.8166\n",
      "Epoch 324/500\n",
      "1/1 [==============================] - 0s 658us/step - loss: 170417.6875 - root_mean_squared_error: 412.8168\n",
      "Epoch 325/500\n",
      "1/1 [==============================] - 0s 712us/step - loss: 170417.7656 - root_mean_squared_error: 412.8169\n",
      "Epoch 326/500\n",
      "1/1 [==============================] - 0s 937us/step - loss: 170417.7969 - root_mean_squared_error: 412.8169\n",
      "Epoch 327/500\n",
      "1/1 [==============================] - 0s 684us/step - loss: 170417.7344 - root_mean_squared_error: 412.8168\n",
      "Epoch 328/500\n",
      "1/1 [==============================] - 0s 963us/step - loss: 170417.7031 - root_mean_squared_error: 412.8168\n",
      "Epoch 329/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.5781 - root_mean_squared_error: 412.8167\n",
      "Epoch 330/500\n",
      "1/1 [==============================] - 0s 701us/step - loss: 170417.5312 - root_mean_squared_error: 412.8166\n",
      "Epoch 331/500\n",
      "1/1 [==============================] - 0s 711us/step - loss: 170417.5156 - root_mean_squared_error: 412.8166\n",
      "Epoch 332/500\n",
      "1/1 [==============================] - 0s 794us/step - loss: 170417.4062 - root_mean_squared_error: 412.8164\n",
      "Epoch 333/500\n",
      "1/1 [==============================] - 0s 753us/step - loss: 170417.3125 - root_mean_squared_error: 412.8163\n",
      "Epoch 334/500\n",
      "1/1 [==============================] - 0s 772us/step - loss: 170417.3438 - root_mean_squared_error: 412.8163\n",
      "Epoch 335/500\n",
      "1/1 [==============================] - 0s 758us/step - loss: 170417.4219 - root_mean_squared_error: 412.8164\n",
      "Epoch 336/500\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 170417.5156 - root_mean_squared_error: 412.8166\n",
      "Epoch 337/500\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 170417.5156 - root_mean_squared_error: 412.8166\n",
      "Epoch 338/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.6094 - root_mean_squared_error: 412.8167\n",
      "Epoch 339/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.7031 - root_mean_squared_error: 412.8168\n",
      "Epoch 340/500\n",
      "1/1 [==============================] - 0s 683us/step - loss: 170417.7656 - root_mean_squared_error: 412.8169\n",
      "Epoch 341/500\n",
      "1/1 [==============================] - 0s 903us/step - loss: 170417.6719 - root_mean_squared_error: 412.8167\n",
      "Epoch 342/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.5625 - root_mean_squared_error: 412.8166\n",
      "Epoch 343/500\n",
      "1/1 [==============================] - 0s 748us/step - loss: 170417.4531 - root_mean_squared_error: 412.8165\n",
      "Epoch 344/500\n",
      "1/1 [==============================] - 0s 710us/step - loss: 170417.4531 - root_mean_squared_error: 412.8165\n",
      "Epoch 345/500\n",
      "1/1 [==============================] - 0s 733us/step - loss: 170417.4062 - root_mean_squared_error: 412.8164\n",
      "Epoch 346/500\n",
      "1/1 [==============================] - 0s 789us/step - loss: 170417.5000 - root_mean_squared_error: 412.8166\n",
      "Epoch 347/500\n",
      "1/1 [==============================] - 0s 771us/step - loss: 170417.6094 - root_mean_squared_error: 412.8167\n",
      "Epoch 348/500\n",
      "1/1 [==============================] - 0s 830us/step - loss: 170417.5781 - root_mean_squared_error: 412.8167\n",
      "Epoch 349/500\n",
      "1/1 [==============================] - 0s 905us/step - loss: 170417.5781 - root_mean_squared_error: 412.8167\n",
      "Epoch 350/500\n",
      "1/1 [==============================] - 0s 922us/step - loss: 170417.6875 - root_mean_squared_error: 412.8168\n",
      "Epoch 351/500\n",
      "1/1 [==============================] - 0s 700us/step - loss: 170417.6094 - root_mean_squared_error: 412.8167\n",
      "Epoch 352/500\n",
      "1/1 [==============================] - 0s 744us/step - loss: 170417.5938 - root_mean_squared_error: 412.8167\n",
      "Epoch 353/500\n",
      "1/1 [==============================] - 0s 690us/step - loss: 170417.6094 - root_mean_squared_error: 412.8167\n",
      "Epoch 354/500\n",
      "1/1 [==============================] - 0s 623us/step - loss: 170417.6875 - root_mean_squared_error: 412.8168\n",
      "Epoch 355/500\n",
      "1/1 [==============================] - 0s 716us/step - loss: 170417.6250 - root_mean_squared_error: 412.8167\n",
      "Epoch 356/500\n",
      "1/1 [==============================] - 0s 750us/step - loss: 170417.5156 - root_mean_squared_error: 412.8166\n",
      "Epoch 357/500\n",
      "1/1 [==============================] - 0s 712us/step - loss: 170417.4219 - root_mean_squared_error: 412.8164\n",
      "Epoch 358/500\n",
      "1/1 [==============================] - 0s 951us/step - loss: 170417.4531 - root_mean_squared_error: 412.8165\n",
      "Epoch 359/500\n",
      "1/1 [==============================] - 0s 656us/step - loss: 170417.5156 - root_mean_squared_error: 412.8166\n",
      "Epoch 360/500\n",
      "1/1 [==============================] - 0s 669us/step - loss: 170417.4531 - root_mean_squared_error: 412.8165\n",
      "Epoch 361/500\n",
      "1/1 [==============================] - 0s 672us/step - loss: 170417.4844 - root_mean_squared_error: 412.8165\n",
      "Epoch 362/500\n",
      "1/1 [==============================] - 0s 661us/step - loss: 170417.4531 - root_mean_squared_error: 412.8165\n",
      "Epoch 363/500\n",
      "1/1 [==============================] - 0s 694us/step - loss: 170417.5156 - root_mean_squared_error: 412.8166\n",
      "Epoch 364/500\n",
      "1/1 [==============================] - 0s 768us/step - loss: 170417.4531 - root_mean_squared_error: 412.8165\n",
      "Epoch 365/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.4844 - root_mean_squared_error: 412.8165\n",
      "Epoch 366/500\n",
      "1/1 [==============================] - 0s 872us/step - loss: 170417.4844 - root_mean_squared_error: 412.8165\n",
      "Epoch 367/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.5000 - root_mean_squared_error: 412.8166\n",
      "Epoch 368/500\n",
      "1/1 [==============================] - 0s 858us/step - loss: 170417.4844 - root_mean_squared_error: 412.8165\n",
      "Epoch 369/500\n",
      "1/1 [==============================] - 0s 812us/step - loss: 170417.5000 - root_mean_squared_error: 412.8166\n",
      "Epoch 370/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.5000 - root_mean_squared_error: 412.8166\n",
      "Epoch 371/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.4844 - root_mean_squared_error: 412.8165\n",
      "Epoch 372/500\n",
      "1/1 [==============================] - 0s 967us/step - loss: 170417.4531 - root_mean_squared_error: 412.8165\n",
      "Epoch 373/500\n",
      "1/1 [==============================] - 0s 994us/step - loss: 170417.4375 - root_mean_squared_error: 412.8165\n",
      "Epoch 374/500\n",
      "1/1 [==============================] - 0s 704us/step - loss: 170417.4531 - root_mean_squared_error: 412.8165\n",
      "Epoch 375/500\n",
      "1/1 [==============================] - 0s 682us/step - loss: 170417.5469 - root_mean_squared_error: 412.8166\n",
      "Epoch 376/500\n",
      "1/1 [==============================] - 0s 731us/step - loss: 170417.5938 - root_mean_squared_error: 412.8167\n",
      "Epoch 377/500\n",
      "1/1 [==============================] - 0s 586us/step - loss: 170417.6094 - root_mean_squared_error: 412.8167\n",
      "Epoch 378/500\n",
      "1/1 [==============================] - 0s 680us/step - loss: 170417.6719 - root_mean_squared_error: 412.8167\n",
      "Epoch 379/500\n",
      "1/1 [==============================] - 0s 934us/step - loss: 170417.6719 - root_mean_squared_error: 412.8167\n",
      "Epoch 380/500\n",
      "1/1 [==============================] - 0s 700us/step - loss: 170417.6094 - root_mean_squared_error: 412.8167\n",
      "Epoch 381/500\n",
      "1/1 [==============================] - 0s 685us/step - loss: 170417.6094 - root_mean_squared_error: 412.8167\n",
      "Epoch 382/500\n",
      "1/1 [==============================] - 0s 724us/step - loss: 170417.4844 - root_mean_squared_error: 412.8165\n",
      "Epoch 383/500\n",
      "1/1 [==============================] - 0s 822us/step - loss: 170417.4531 - root_mean_squared_error: 412.8165\n",
      "Epoch 384/500\n",
      "1/1 [==============================] - 0s 655us/step - loss: 170417.4531 - root_mean_squared_error: 412.8165\n",
      "Epoch 385/500\n",
      "1/1 [==============================] - 0s 714us/step - loss: 170417.5156 - root_mean_squared_error: 412.8166\n",
      "Epoch 386/500\n",
      "1/1 [==============================] - 0s 725us/step - loss: 170417.4531 - root_mean_squared_error: 412.8165\n",
      "Epoch 387/500\n",
      "1/1 [==============================] - 0s 872us/step - loss: 170417.4375 - root_mean_squared_error: 412.8165\n",
      "Epoch 388/500\n",
      "1/1 [==============================] - 0s 855us/step - loss: 170417.3750 - root_mean_squared_error: 412.8164\n",
      "Epoch 389/500\n",
      "1/1 [==============================] - 0s 856us/step - loss: 170417.4062 - root_mean_squared_error: 412.8164\n",
      "Epoch 390/500\n",
      "1/1 [==============================] - 0s 945us/step - loss: 170417.5156 - root_mean_squared_error: 412.8166\n",
      "Epoch 391/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.5156 - root_mean_squared_error: 412.8166\n",
      "Epoch 392/500\n",
      "1/1 [==============================] - 0s 698us/step - loss: 170417.5156 - root_mean_squared_error: 412.8166\n",
      "Epoch 393/500\n",
      "1/1 [==============================] - 0s 711us/step - loss: 170417.5156 - root_mean_squared_error: 412.8166\n",
      "Epoch 394/500\n",
      "1/1 [==============================] - 0s 798us/step - loss: 170417.5156 - root_mean_squared_error: 412.8166\n",
      "Epoch 395/500\n",
      "1/1 [==============================] - 0s 717us/step - loss: 170417.5156 - root_mean_squared_error: 412.8166\n",
      "Epoch 396/500\n",
      "1/1 [==============================] - 0s 678us/step - loss: 170417.5156 - root_mean_squared_error: 412.8166\n",
      "Epoch 397/500\n",
      "1/1 [==============================] - 0s 995us/step - loss: 170417.5156 - root_mean_squared_error: 412.8166\n",
      "Epoch 398/500\n",
      "1/1 [==============================] - 0s 774us/step - loss: 170417.5156 - root_mean_squared_error: 412.8166\n",
      "Epoch 399/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.5938 - root_mean_squared_error: 412.8167\n",
      "Epoch 400/500\n",
      "1/1 [==============================] - 0s 821us/step - loss: 170417.6875 - root_mean_squared_error: 412.8168\n",
      "Epoch 401/500\n",
      "1/1 [==============================] - 0s 846us/step - loss: 170417.6719 - root_mean_squared_error: 412.8167\n",
      "Epoch 402/500\n",
      "1/1 [==============================] - 0s 721us/step - loss: 170417.6406 - root_mean_squared_error: 412.8167\n",
      "Epoch 403/500\n",
      "1/1 [==============================] - 0s 694us/step - loss: 170417.6562 - root_mean_squared_error: 412.8167\n",
      "Epoch 404/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 702us/step - loss: 170417.6250 - root_mean_squared_error: 412.8167\n",
      "Epoch 405/500\n",
      "1/1 [==============================] - 0s 693us/step - loss: 170417.6406 - root_mean_squared_error: 412.8167\n",
      "Epoch 406/500\n",
      "1/1 [==============================] - 0s 681us/step - loss: 170417.6406 - root_mean_squared_error: 412.8167\n",
      "Epoch 407/500\n",
      "1/1 [==============================] - 0s 715us/step - loss: 170417.5781 - root_mean_squared_error: 412.8167\n",
      "Epoch 408/500\n",
      "1/1 [==============================] - 0s 669us/step - loss: 170417.5469 - root_mean_squared_error: 412.8166\n",
      "Epoch 409/500\n",
      "1/1 [==============================] - 0s 796us/step - loss: 170417.4844 - root_mean_squared_error: 412.8165\n",
      "Epoch 410/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.4219 - root_mean_squared_error: 412.8164\n",
      "Epoch 411/500\n",
      "1/1 [==============================] - 0s 788us/step - loss: 170417.3906 - root_mean_squared_error: 412.8164\n",
      "Epoch 412/500\n",
      "1/1 [==============================] - 0s 933us/step - loss: 170417.3750 - root_mean_squared_error: 412.8164\n",
      "Epoch 413/500\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 170417.4062 - root_mean_squared_error: 412.8164\n",
      "Epoch 414/500\n",
      "1/1 [==============================] - 0s 698us/step - loss: 170417.3750 - root_mean_squared_error: 412.8164\n",
      "Epoch 415/500\n",
      "1/1 [==============================] - 0s 656us/step - loss: 170417.4375 - root_mean_squared_error: 412.8165\n",
      "Epoch 416/500\n",
      "1/1 [==============================] - 0s 769us/step - loss: 170417.5156 - root_mean_squared_error: 412.8166\n",
      "Epoch 417/500\n",
      "1/1 [==============================] - 0s 703us/step - loss: 170417.6406 - root_mean_squared_error: 412.8167\n",
      "Epoch 418/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.7031 - root_mean_squared_error: 412.8168\n",
      "Epoch 419/500\n",
      "1/1 [==============================] - 0s 669us/step - loss: 170417.7969 - root_mean_squared_error: 412.8169\n",
      "Epoch 420/500\n",
      "1/1 [==============================] - 0s 859us/step - loss: 170417.7500 - root_mean_squared_error: 412.8168\n",
      "Epoch 421/500\n",
      "1/1 [==============================] - 0s 1000us/step - loss: 170417.6719 - root_mean_squared_error: 412.8167\n",
      "Epoch 422/500\n",
      "1/1 [==============================] - 0s 667us/step - loss: 170417.5469 - root_mean_squared_error: 412.8166\n",
      "Epoch 423/500\n",
      "1/1 [==============================] - 0s 984us/step - loss: 170417.4844 - root_mean_squared_error: 412.8165\n",
      "Epoch 424/500\n",
      "1/1 [==============================] - 0s 809us/step - loss: 170417.4531 - root_mean_squared_error: 412.8165\n",
      "Epoch 425/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.4844 - root_mean_squared_error: 412.8165\n",
      "Epoch 426/500\n",
      "1/1 [==============================] - 0s 995us/step - loss: 170417.4062 - root_mean_squared_error: 412.8164\n",
      "Epoch 427/500\n",
      "1/1 [==============================] - 0s 925us/step - loss: 170417.4219 - root_mean_squared_error: 412.8164\n",
      "Epoch 428/500\n",
      "1/1 [==============================] - 0s 978us/step - loss: 170417.3594 - root_mean_squared_error: 412.8164\n",
      "Epoch 429/500\n",
      "1/1 [==============================] - 0s 950us/step - loss: 170417.4062 - root_mean_squared_error: 412.8164\n",
      "Epoch 430/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.4219 - root_mean_squared_error: 412.8164\n",
      "Epoch 431/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.4062 - root_mean_squared_error: 412.8164\n",
      "Epoch 432/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.5000 - root_mean_squared_error: 412.8166\n",
      "Epoch 433/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.5469 - root_mean_squared_error: 412.8166\n",
      "Epoch 434/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.6094 - root_mean_squared_error: 412.8167\n",
      "Epoch 435/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.6094 - root_mean_squared_error: 412.8167\n",
      "Epoch 436/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.6875 - root_mean_squared_error: 412.8168\n",
      "Epoch 437/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.6406 - root_mean_squared_error: 412.8167\n",
      "Epoch 438/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.6406 - root_mean_squared_error: 412.8167\n",
      "Epoch 439/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.6250 - root_mean_squared_error: 412.8167\n",
      "Epoch 440/500\n",
      "1/1 [==============================] - 0s 989us/step - loss: 170417.5625 - root_mean_squared_error: 412.8166\n",
      "Epoch 441/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.5625 - root_mean_squared_error: 412.8166\n",
      "Epoch 442/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.5469 - root_mean_squared_error: 412.8166\n",
      "Epoch 443/500\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 170417.4219 - root_mean_squared_error: 412.8164\n",
      "Epoch 444/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.4844 - root_mean_squared_error: 412.8165\n",
      "Epoch 445/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.5156 - root_mean_squared_error: 412.8166\n",
      "Epoch 446/500\n",
      "1/1 [==============================] - 0s 672us/step - loss: 170417.5781 - root_mean_squared_error: 412.8167\n",
      "Epoch 447/500\n",
      "1/1 [==============================] - 0s 739us/step - loss: 170417.6094 - root_mean_squared_error: 412.8167\n",
      "Epoch 448/500\n",
      "1/1 [==============================] - 0s 969us/step - loss: 170417.6875 - root_mean_squared_error: 412.8168\n",
      "Epoch 449/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.6406 - root_mean_squared_error: 412.8167\n",
      "Epoch 450/500\n",
      "1/1 [==============================] - 0s 967us/step - loss: 170417.6406 - root_mean_squared_error: 412.8167\n",
      "Epoch 451/500\n",
      "1/1 [==============================] - 0s 966us/step - loss: 170417.6719 - root_mean_squared_error: 412.8167\n",
      "Epoch 452/500\n",
      "1/1 [==============================] - 0s 751us/step - loss: 170417.6406 - root_mean_squared_error: 412.8167\n",
      "Epoch 453/500\n",
      "1/1 [==============================] - 0s 730us/step - loss: 170417.6406 - root_mean_squared_error: 412.8167\n",
      "Epoch 454/500\n",
      "1/1 [==============================] - 0s 806us/step - loss: 170417.6406 - root_mean_squared_error: 412.8167\n",
      "Epoch 455/500\n",
      "1/1 [==============================] - 0s 947us/step - loss: 170417.6719 - root_mean_squared_error: 412.8167\n",
      "Epoch 456/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.6406 - root_mean_squared_error: 412.8167\n",
      "Epoch 457/500\n",
      "1/1 [==============================] - 0s 692us/step - loss: 170417.6094 - root_mean_squared_error: 412.8167\n",
      "Epoch 458/500\n",
      "1/1 [==============================] - 0s 904us/step - loss: 170417.4844 - root_mean_squared_error: 412.8165\n",
      "Epoch 459/500\n",
      "1/1 [==============================] - 0s 704us/step - loss: 170417.4375 - root_mean_squared_error: 412.8165\n",
      "Epoch 460/500\n",
      "1/1 [==============================] - 0s 720us/step - loss: 170417.4531 - root_mean_squared_error: 412.8165\n",
      "Epoch 461/500\n",
      "1/1 [==============================] - 0s 747us/step - loss: 170417.4375 - root_mean_squared_error: 412.8165\n",
      "Epoch 462/500\n",
      "1/1 [==============================] - 0s 785us/step - loss: 170417.3906 - root_mean_squared_error: 412.8164\n",
      "Epoch 463/500\n",
      "1/1 [==============================] - 0s 871us/step - loss: 170417.5000 - root_mean_squared_error: 412.8166\n",
      "Epoch 464/500\n",
      "1/1 [==============================] - 0s 727us/step - loss: 170417.5000 - root_mean_squared_error: 412.8166\n",
      "Epoch 465/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.5781 - root_mean_squared_error: 412.8167\n",
      "Epoch 466/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.6094 - root_mean_squared_error: 412.8167\n",
      "Epoch 467/500\n",
      "1/1 [==============================] - 0s 698us/step - loss: 170417.5781 - root_mean_squared_error: 412.8167\n",
      "Epoch 468/500\n",
      "1/1 [==============================] - 0s 678us/step - loss: 170417.5469 - root_mean_squared_error: 412.8166\n",
      "Epoch 469/500\n",
      "1/1 [==============================] - 0s 710us/step - loss: 170417.5781 - root_mean_squared_error: 412.8167\n",
      "Epoch 470/500\n",
      "1/1 [==============================] - 0s 832us/step - loss: 170417.5625 - root_mean_squared_error: 412.8166\n",
      "Epoch 471/500\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 798us/step - loss: 170417.5781 - root_mean_squared_error: 412.8167\n",
      "Epoch 472/500\n",
      "1/1 [==============================] - 0s 890us/step - loss: 170417.5938 - root_mean_squared_error: 412.8167\n",
      "Epoch 473/500\n",
      "1/1 [==============================] - 0s 732us/step - loss: 170417.5469 - root_mean_squared_error: 412.8166\n",
      "Epoch 474/500\n",
      "1/1 [==============================] - 0s 672us/step - loss: 170417.4375 - root_mean_squared_error: 412.8165\n",
      "Epoch 475/500\n",
      "1/1 [==============================] - 0s 859us/step - loss: 170417.4844 - root_mean_squared_error: 412.8165\n",
      "Epoch 476/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.5156 - root_mean_squared_error: 412.8166\n",
      "Epoch 477/500\n",
      "1/1 [==============================] - 0s 688us/step - loss: 170417.5781 - root_mean_squared_error: 412.8167\n",
      "Epoch 478/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.6875 - root_mean_squared_error: 412.8168\n",
      "Epoch 479/500\n",
      "1/1 [==============================] - 0s 731us/step - loss: 170417.7344 - root_mean_squared_error: 412.8168\n",
      "Epoch 480/500\n",
      "1/1 [==============================] - 0s 754us/step - loss: 170417.7344 - root_mean_squared_error: 412.8168\n",
      "Epoch 481/500\n",
      "1/1 [==============================] - 0s 744us/step - loss: 170417.6094 - root_mean_squared_error: 412.8167\n",
      "Epoch 482/500\n",
      "1/1 [==============================] - 0s 785us/step - loss: 170417.5156 - root_mean_squared_error: 412.8166\n",
      "Epoch 483/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.4844 - root_mean_squared_error: 412.8165\n",
      "Epoch 484/500\n",
      "1/1 [==============================] - 0s 786us/step - loss: 170417.4219 - root_mean_squared_error: 412.8164\n",
      "Epoch 485/500\n",
      "1/1 [==============================] - 0s 877us/step - loss: 170417.3906 - root_mean_squared_error: 412.8164\n",
      "Epoch 486/500\n",
      "1/1 [==============================] - 0s 758us/step - loss: 170417.4531 - root_mean_squared_error: 412.8165\n",
      "Epoch 487/500\n",
      "1/1 [==============================] - 0s 754us/step - loss: 170417.4531 - root_mean_squared_error: 412.8165\n",
      "Epoch 488/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.4531 - root_mean_squared_error: 412.8165\n",
      "Epoch 489/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.4531 - root_mean_squared_error: 412.8165\n",
      "Epoch 490/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.4531 - root_mean_squared_error: 412.8165\n",
      "Epoch 491/500\n",
      "1/1 [==============================] - 0s 909us/step - loss: 170417.4375 - root_mean_squared_error: 412.8165\n",
      "Epoch 492/500\n",
      "1/1 [==============================] - 0s 771us/step - loss: 170417.5000 - root_mean_squared_error: 412.8166\n",
      "Epoch 493/500\n",
      "1/1 [==============================] - 0s 782us/step - loss: 170417.6250 - root_mean_squared_error: 412.8167\n",
      "Epoch 494/500\n",
      "1/1 [==============================] - 0s 840us/step - loss: 170417.5625 - root_mean_squared_error: 412.8166\n",
      "Epoch 495/500\n",
      "1/1 [==============================] - 0s 833us/step - loss: 170417.6250 - root_mean_squared_error: 412.8167\n",
      "Epoch 496/500\n",
      "1/1 [==============================] - 0s 823us/step - loss: 170417.6719 - root_mean_squared_error: 412.8167\n",
      "Epoch 497/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.6250 - root_mean_squared_error: 412.8167\n",
      "Epoch 498/500\n",
      "1/1 [==============================] - 0s 782us/step - loss: 170417.5625 - root_mean_squared_error: 412.8166\n",
      "Epoch 499/500\n",
      "1/1 [==============================] - 0s 816us/step - loss: 170417.5781 - root_mean_squared_error: 412.8167\n",
      "Epoch 500/500\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 170417.5156 - root_mean_squared_error: 412.8166\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/eric/.conda/envs/mlbase/lib/python3.8/site-packages/numpy/core/_asarray.py:136: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray\n",
      "  return array(a, dtype, copy=False, order=order, subok=True)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Increase the learning rate and decrease the number of epochs.\n",
    "learning_rate=100 \n",
    "epochs=500 \n",
    "\n",
    "my_model = build_model(learning_rate)\n",
    "trained_weight, trained_bias, epochs, rmse = train_model(my_model, my_feature, \n",
    "                                                         my_label, epochs,\n",
    "                                                         my_batch_size)\n",
    "plot_the_model(trained_weight, trained_bias, my_feature, my_label)\n",
    "plot_the_loss_curve(epochs, rmse)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "c96ITm021NEV"
   },
   "source": [
    "The resulting model is terrible; the red line doesn't align with the blue dots. Furthermore, the loss curve oscillates like a [roller coaster](https://www.wikipedia.org/wiki/Roller_coaster).  An oscillating loss curve strongly suggests that the learning rate is too high. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "r63YkMx82WVr"
   },
   "source": [
    "## Task 4: Find the ideal combination of epochs and learning rate\n",
    "\n",
    "Assign values to the following two hyperparameters to make training converge as efficiently as possible: \n",
    "\n",
    "*  learning_rate\n",
    "*  epochs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "ZYC8eR5x5n4m"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/65\n",
      "1/1 [==============================] - 0s 908us/step - loss: 367.3199 - root_mean_squared_error: 19.1656\n",
      "Epoch 2/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 248.8947 - root_mean_squared_error: 15.7764\n",
      "Epoch 3/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 183.7997 - root_mean_squared_error: 13.5573\n",
      "Epoch 4/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 139.9637 - root_mean_squared_error: 11.8306\n",
      "Epoch 5/65\n",
      "1/1 [==============================] - 0s 824us/step - loss: 108.0538 - root_mean_squared_error: 10.3949\n",
      "Epoch 6/65\n",
      "1/1 [==============================] - 0s 680us/step - loss: 83.8996 - root_mean_squared_error: 9.1597\n",
      "Epoch 7/65\n",
      "1/1 [==============================] - 0s 703us/step - loss: 65.2202 - root_mean_squared_error: 8.0759\n",
      "Epoch 8/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 50.6058 - root_mean_squared_error: 7.1138\n",
      "Epoch 9/65\n",
      "1/1 [==============================] - 0s 866us/step - loss: 39.1105 - root_mean_squared_error: 6.2538\n",
      "Epoch 10/65\n",
      "1/1 [==============================] - 0s 847us/step - loss: 30.0600 - root_mean_squared_error: 5.4827\n",
      "Epoch 11/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 22.9515 - root_mean_squared_error: 4.7908\n",
      "Epoch 12/65\n",
      "1/1 [==============================] - 0s 800us/step - loss: 17.3974 - root_mean_squared_error: 4.1710\n",
      "Epoch 13/65\n",
      "1/1 [==============================] - 0s 819us/step - loss: 13.0904 - root_mean_squared_error: 3.6181\n",
      "Epoch 14/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 9.7829 - root_mean_squared_error: 3.1278\n",
      "Epoch 15/65\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 7.2726 - root_mean_squared_error: 2.6968\n",
      "Epoch 16/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 5.3934 - root_mean_squared_error: 2.3224\n",
      "Epoch 17/65\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 4.0082 - root_mean_squared_error: 2.0021\n",
      "Epoch 18/65\n",
      "1/1 [==============================] - 0s 828us/step - loss: 3.0050 - root_mean_squared_error: 1.7335\n",
      "Epoch 19/65\n",
      "1/1 [==============================] - 0s 953us/step - loss: 2.2923 - root_mean_squared_error: 1.5140\n",
      "Epoch 20/65\n",
      "1/1 [==============================] - 0s 882us/step - loss: 1.7966 - root_mean_squared_error: 1.3404\n",
      "Epoch 21/65\n",
      "1/1 [==============================] - 0s 736us/step - loss: 1.4598 - root_mean_squared_error: 1.2082\n",
      "Epoch 22/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.2367 - root_mean_squared_error: 1.1121\n",
      "Epoch 23/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.0929 - root_mean_squared_error: 1.0454\n",
      "Epoch 24/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.0029 - root_mean_squared_error: 1.0014\n",
      "Epoch 25/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.9483 - root_mean_squared_error: 0.9738\n",
      "Epoch 26/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.9163 - root_mean_squared_error: 0.9572\n",
      "Epoch 27/65\n",
      "1/1 [==============================] - 0s 772us/step - loss: 0.8982 - root_mean_squared_error: 0.9477\n",
      "Epoch 28/65\n",
      "1/1 [==============================] - 0s 840us/step - loss: 0.8884 - root_mean_squared_error: 0.9425\n",
      "Epoch 29/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8832 - root_mean_squared_error: 0.9398\n",
      "Epoch 30/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8806 - root_mean_squared_error: 0.9384\n",
      "Epoch 31/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8794 - root_mean_squared_error: 0.9378\n",
      "Epoch 32/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8788 - root_mean_squared_error: 0.9374\n",
      "Epoch 33/65\n",
      "1/1 [==============================] - 0s 756us/step - loss: 0.8785 - root_mean_squared_error: 0.9373\n",
      "Epoch 34/65\n",
      "1/1 [==============================] - 0s 854us/step - loss: 0.8783 - root_mean_squared_error: 0.9372\n",
      "Epoch 35/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8782 - root_mean_squared_error: 0.9371\n",
      "Epoch 36/65\n",
      "1/1 [==============================] - 0s 853us/step - loss: 0.8782 - root_mean_squared_error: 0.9371\n",
      "Epoch 37/65\n",
      "1/1 [==============================] - 0s 820us/step - loss: 0.8781 - root_mean_squared_error: 0.9371\n",
      "Epoch 38/65\n",
      "1/1 [==============================] - 0s 934us/step - loss: 0.8780 - root_mean_squared_error: 0.9370\n",
      "Epoch 39/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8779 - root_mean_squared_error: 0.9370\n",
      "Epoch 40/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8778 - root_mean_squared_error: 0.9369\n",
      "Epoch 41/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8778 - root_mean_squared_error: 0.9369\n",
      "Epoch 42/65\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 0.8777 - root_mean_squared_error: 0.9368\n",
      "Epoch 43/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8776 - root_mean_squared_error: 0.9368\n",
      "Epoch 44/65\n",
      "1/1 [==============================] - 0s 994us/step - loss: 0.8775 - root_mean_squared_error: 0.9367\n",
      "Epoch 45/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8774 - root_mean_squared_error: 0.9367\n",
      "Epoch 46/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8773 - root_mean_squared_error: 0.9367\n",
      "Epoch 47/65\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 0.8772 - root_mean_squared_error: 0.9366\n",
      "Epoch 48/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8771 - root_mean_squared_error: 0.9366\n",
      "Epoch 49/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8770 - root_mean_squared_error: 0.9365\n",
      "Epoch 50/65\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 0.8769 - root_mean_squared_error: 0.9365\n",
      "Epoch 51/65\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 0.8769 - root_mean_squared_error: 0.9364\n",
      "Epoch 52/65\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 0.8768 - root_mean_squared_error: 0.9364\n",
      "Epoch 53/65\n",
      "1/1 [==============================] - 0s 961us/step - loss: 0.8767 - root_mean_squared_error: 0.9363\n",
      "Epoch 54/65\n",
      "1/1 [==============================] - 0s 998us/step - loss: 0.8766 - root_mean_squared_error: 0.9363\n",
      "Epoch 55/65\n",
      "1/1 [==============================] - 0s 990us/step - loss: 0.8765 - root_mean_squared_error: 0.9362\n",
      "Epoch 56/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8764 - root_mean_squared_error: 0.9362\n",
      "Epoch 57/65\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 0.8763 - root_mean_squared_error: 0.9361\n",
      "Epoch 58/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8762 - root_mean_squared_error: 0.9361\n",
      "Epoch 59/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8761 - root_mean_squared_error: 0.9360\n",
      "Epoch 60/65\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 0.8760 - root_mean_squared_error: 0.9360\n",
      "Epoch 61/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8760 - root_mean_squared_error: 0.9359\n",
      "Epoch 62/65\n",
      "1/1 [==============================] - 0s 935us/step - loss: 0.8759 - root_mean_squared_error: 0.9359\n",
      "Epoch 63/65\n",
      "1/1 [==============================] - 0s 792us/step - loss: 0.8758 - root_mean_squared_error: 0.9358\n",
      "Epoch 64/65\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8757 - root_mean_squared_error: 0.9358\n",
      "Epoch 65/65\n",
      "1/1 [==============================] - 0s 970us/step - loss: 0.8757 - root_mean_squared_error: 0.9358\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/eric/.conda/envs/mlbase/lib/python3.8/site-packages/numpy/core/_asarray.py:136: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray\n",
      "  return array(a, dtype, copy=False, order=order, subok=True)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Set the learning rate and number of epochs\n",
    "learning_rate= 0.13  # Replace ? with a floating-point number\n",
    "epochs= 65   # Replace ? with an integer\n",
    "\n",
    "my_model = build_model(learning_rate)\n",
    "trained_weight, trained_bias, epochs, rmse = train_model(my_model, my_feature, \n",
    "                                                         my_label, epochs,\n",
    "                                                         my_batch_size)\n",
    "plot_the_model(trained_weight, trained_bias, my_feature, my_label)\n",
    "plot_the_loss_curve(epochs, rmse)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "cellView": "form",
    "colab": {},
    "colab_type": "code",
    "id": "_GMGgR6O54IN"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/70\n",
      "1/1 [==============================] - 0s 990us/step - loss: 1034.5702 - root_mean_squared_error: 32.1647\n",
      "Epoch 2/70\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 812.9653 - root_mean_squared_error: 28.5125\n",
      "Epoch 3/70\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 677.0311 - root_mean_squared_error: 26.0198\n",
      "Epoch 4/70\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 576.7755 - root_mean_squared_error: 24.0161\n",
      "Epoch 5/70\n",
      "1/1 [==============================] - 0s 729us/step - loss: 497.1628 - root_mean_squared_error: 22.2971\n",
      "Epoch 6/70\n",
      "1/1 [==============================] - 0s 743us/step - loss: 431.3899 - root_mean_squared_error: 20.7699\n",
      "Epoch 7/70\n",
      "1/1 [==============================] - 0s 691us/step - loss: 375.7209 - root_mean_squared_error: 19.3835\n",
      "Epoch 8/70\n",
      "1/1 [==============================] - 0s 810us/step - loss: 327.8513 - root_mean_squared_error: 18.1067\n",
      "Epoch 9/70\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 286.2434 - root_mean_squared_error: 16.9187\n",
      "Epoch 10/70\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 249.8103 - root_mean_squared_error: 15.8054\n",
      "Epoch 11/70\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 217.7491 - root_mean_squared_error: 14.7563\n",
      "Epoch 12/70\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 189.4454 - root_mean_squared_error: 13.7639\n",
      "Epoch 13/70\n",
      "1/1 [==============================] - 0s 911us/step - loss: 164.4144 - root_mean_squared_error: 12.8224\n",
      "Epoch 14/70\n",
      "1/1 [==============================] - 0s 895us/step - loss: 142.2639 - root_mean_squared_error: 11.9274\n",
      "Epoch 15/70\n",
      "1/1 [==============================] - 0s 731us/step - loss: 122.6696 - root_mean_squared_error: 11.0756\n",
      "Epoch 16/70\n",
      "1/1 [==============================] - 0s 851us/step - loss: 105.3578 - root_mean_squared_error: 10.2644\n",
      "Epoch 17/70\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 90.0935 - root_mean_squared_error: 9.4918\n",
      "Epoch 18/70\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 76.6714 - root_mean_squared_error: 8.7562\n",
      "Epoch 19/70\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 64.9101 - root_mean_squared_error: 8.0567\n",
      "Epoch 20/70\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 54.6467 - root_mean_squared_error: 7.3923\n",
      "Epoch 21/70\n",
      "1/1 [==============================] - 0s 705us/step - loss: 45.7337 - root_mean_squared_error: 6.7627\n",
      "Epoch 22/70\n",
      "1/1 [==============================] - 0s 763us/step - loss: 38.0363 - root_mean_squared_error: 6.1674\n",
      "Epoch 23/70\n",
      "1/1 [==============================] - 0s 897us/step - loss: 31.4301 - root_mean_squared_error: 5.6063\n",
      "Epoch 24/70\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 25.8001 - root_mean_squared_error: 5.0794\n",
      "Epoch 25/70\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 21.0390 - root_mean_squared_error: 4.5868\n",
      "Epoch 26/70\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 17.0473 - root_mean_squared_error: 4.1288\n",
      "Epoch 27/70\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 13.7320 - root_mean_squared_error: 3.7057\n",
      "Epoch 28/70\n",
      "1/1 [==============================] - 0s 968us/step - loss: 11.0067 - root_mean_squared_error: 3.3176\n",
      "Epoch 29/70\n",
      "1/1 [==============================] - 0s 737us/step - loss: 8.7915 - root_mean_squared_error: 2.9651\n",
      "Epoch 30/70\n",
      "1/1 [==============================] - 0s 706us/step - loss: 7.0128 - root_mean_squared_error: 2.6482\n",
      "Epoch 31/70\n",
      "1/1 [==============================] - 0s 789us/step - loss: 5.6033 - root_mean_squared_error: 2.3671\n",
      "Epoch 32/70\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 4.5023 - root_mean_squared_error: 2.1219\n",
      "Epoch 33/70\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 3.6552 - root_mean_squared_error: 1.9119\n",
      "Epoch 34/70\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 3.0141 - root_mean_squared_error: 1.7361\n",
      "Epoch 35/70\n",
      "1/1 [==============================] - 0s 672us/step - loss: 2.5372 - root_mean_squared_error: 1.5929\n",
      "Epoch 36/70\n",
      "1/1 [==============================] - 0s 753us/step - loss: 2.1889 - root_mean_squared_error: 1.4795\n",
      "Epoch 37/70\n",
      "1/1 [==============================] - 0s 992us/step - loss: 1.9392 - root_mean_squared_error: 1.3926\n",
      "Epoch 38/70\n",
      "1/1 [==============================] - 0s 849us/step - loss: 1.7635 - root_mean_squared_error: 1.3280\n",
      "Epoch 39/70\n",
      "1/1 [==============================] - 0s 930us/step - loss: 1.6420 - root_mean_squared_error: 1.2814\n",
      "Epoch 40/70\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 1.5593 - root_mean_squared_error: 1.2487\n",
      "Epoch 41/70\n",
      "1/1 [==============================] - 0s 851us/step - loss: 1.5035 - root_mean_squared_error: 1.2262\n",
      "Epoch 42/70\n",
      "1/1 [==============================] - 0s 870us/step - loss: 1.4658 - root_mean_squared_error: 1.2107\n",
      "Epoch 43/70\n",
      "1/1 [==============================] - 0s 745us/step - loss: 1.4399 - root_mean_squared_error: 1.2000\n",
      "Epoch 44/70\n",
      "1/1 [==============================] - 0s 955us/step - loss: 1.4214 - root_mean_squared_error: 1.1922\n",
      "Epoch 45/70\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.4072 - root_mean_squared_error: 1.1863\n",
      "Epoch 46/70\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 1.3956 - root_mean_squared_error: 1.1813\n",
      "Epoch 47/70\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 1.3851 - root_mean_squared_error: 1.1769\n",
      "Epoch 48/70\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.3753 - root_mean_squared_error: 1.1727\n",
      "Epoch 49/70\n",
      "1/1 [==============================] - 0s 823us/step - loss: 1.3655 - root_mean_squared_error: 1.1686\n",
      "Epoch 50/70\n",
      "1/1 [==============================] - 0s 817us/step - loss: 1.3557 - root_mean_squared_error: 1.1643\n",
      "Epoch 51/70\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.3456 - root_mean_squared_error: 1.1600\n",
      "Epoch 52/70\n",
      "1/1 [==============================] - 0s 700us/step - loss: 1.3353 - root_mean_squared_error: 1.1555\n",
      "Epoch 53/70\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 1.3247 - root_mean_squared_error: 1.1509\n",
      "Epoch 54/70\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.3137 - root_mean_squared_error: 1.1462\n",
      "Epoch 55/70\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.3025 - root_mean_squared_error: 1.1413\n",
      "Epoch 56/70\n",
      "1/1 [==============================] - 0s 823us/step - loss: 1.2910 - root_mean_squared_error: 1.1362\n",
      "Epoch 57/70\n",
      "1/1 [==============================] - 0s 912us/step - loss: 1.2792 - root_mean_squared_error: 1.1310\n",
      "Epoch 58/70\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.2672 - root_mean_squared_error: 1.1257\n",
      "Epoch 59/70\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 1.2549 - root_mean_squared_error: 1.1202\n",
      "Epoch 60/70\n",
      "1/1 [==============================] - 0s 864us/step - loss: 1.2423 - root_mean_squared_error: 1.1146\n",
      "Epoch 61/70\n",
      "1/1 [==============================] - 0s 726us/step - loss: 1.2295 - root_mean_squared_error: 1.1088\n",
      "Epoch 62/70\n",
      "1/1 [==============================] - 0s 711us/step - loss: 1.2165 - root_mean_squared_error: 1.1030\n",
      "Epoch 63/70\n",
      "1/1 [==============================] - 0s 790us/step - loss: 1.2034 - root_mean_squared_error: 1.0970\n",
      "Epoch 64/70\n",
      "1/1 [==============================] - 0s 664us/step - loss: 1.1901 - root_mean_squared_error: 1.0909\n",
      "Epoch 65/70\n",
      "1/1 [==============================] - 0s 671us/step - loss: 1.1767 - root_mean_squared_error: 1.0847\n",
      "Epoch 66/70\n",
      "1/1 [==============================] - 0s 828us/step - loss: 1.1631 - root_mean_squared_error: 1.0785\n",
      "Epoch 67/70\n",
      "1/1 [==============================] - 0s 661us/step - loss: 1.1496 - root_mean_squared_error: 1.0722\n",
      "Epoch 68/70\n",
      "1/1 [==============================] - 0s 690us/step - loss: 1.1359 - root_mean_squared_error: 1.0658\n",
      "Epoch 69/70\n",
      "1/1 [==============================] - 0s 664us/step - loss: 1.1223 - root_mean_squared_error: 1.0594\n",
      "Epoch 70/70\n",
      "1/1 [==============================] - 0s 716us/step - loss: 1.1088 - root_mean_squared_error: 1.0530\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/eric/.conda/envs/mlbase/lib/python3.8/site-packages/numpy/core/_asarray.py:136: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray\n",
      "  return array(a, dtype, copy=False, order=order, subok=True)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#@title Double-click to view a possible solution\n",
    "\n",
    "learning_rate=0.14\n",
    "epochs=70\n",
    "my_batch_size=12\n",
    "\n",
    "my_model = build_model(learning_rate)\n",
    "trained_weight, trained_bias, epochs, rmse = train_model(my_model, my_feature, \n",
    "                                                         my_label, epochs,\n",
    "                                                         my_batch_size)\n",
    "plot_the_model(trained_weight, trained_bias, my_feature, my_label)\n",
    "plot_the_loss_curve(epochs, rmse)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "0NDET9e6AAbA"
   },
   "source": [
    "## Task 5: Adjust the batch size\n",
    "\n",
    "The system recalculates the model's loss value and adjusts the model's weights and bias after each **iteration**.  Each iteration is the span in which the system processes one batch. For example, if the **batch size** is 6, then the system recalculates the model's loss value and adjusts the model's weights and bias after processing every 6 examples.  \n",
    "\n",
    "One **epoch** spans sufficient iterations to process every example in the dataset. For example, if the batch size is 12, then each epoch lasts one iteration. However, if the batch size is 6, then each epoch consumes two iterations.  \n",
    "\n",
    "It is tempting to simply set the batch size to the number of examples in the dataset (12, in this case). However, the model might actually train faster on smaller batches. Conversely, very small batches might not contain enough information to help the model converge. \n",
    "\n",
    "Experiment with `batch_size` in the following code cell. What's the smallest integer you can set for `batch_size` and still have the model converge in a hundred epochs?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "_vGx0lOodQrT"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "1/1 [==============================] - 0s 881us/step - loss: 186.5416 - root_mean_squared_error: 13.6580\n",
      "Epoch 2/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 152.6761 - root_mean_squared_error: 12.3562\n",
      "Epoch 3/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 131.2984 - root_mean_squared_error: 11.4586\n",
      "Epoch 4/100\n",
      "1/1 [==============================] - 0s 709us/step - loss: 115.1687 - root_mean_squared_error: 10.7317\n",
      "Epoch 5/100\n",
      "1/1 [==============================] - 0s 872us/step - loss: 102.0897 - root_mean_squared_error: 10.1039\n",
      "Epoch 6/100\n",
      "1/1 [==============================] - 0s 880us/step - loss: 91.0633 - root_mean_squared_error: 9.5427\n",
      "Epoch 7/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 81.5406 - root_mean_squared_error: 9.0300\n",
      "Epoch 8/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 73.1828 - root_mean_squared_error: 8.5547\n",
      "Epoch 9/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 65.7644 - root_mean_squared_error: 8.1095\n",
      "Epoch 10/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 59.1263 - root_mean_squared_error: 7.6894\n",
      "Epoch 11/100\n",
      "1/1 [==============================] - 0s 847us/step - loss: 53.1517 - root_mean_squared_error: 7.2905\n",
      "Epoch 12/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 47.7514 - root_mean_squared_error: 6.9102\n",
      "Epoch 13/100\n",
      "1/1 [==============================] - 0s 742us/step - loss: 42.8559 - root_mean_squared_error: 6.5464\n",
      "Epoch 14/100\n",
      "1/1 [==============================] - 0s 820us/step - loss: 38.4091 - root_mean_squared_error: 6.1975\n",
      "Epoch 15/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 34.3654 - root_mean_squared_error: 5.8622\n",
      "Epoch 16/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 30.6866 - root_mean_squared_error: 5.5395\n",
      "Epoch 17/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 27.3403 - root_mean_squared_error: 5.2288\n",
      "Epoch 18/100\n",
      "1/1 [==============================] - 0s 754us/step - loss: 24.2987 - root_mean_squared_error: 4.9294\n",
      "Epoch 19/100\n",
      "1/1 [==============================] - 0s 826us/step - loss: 21.5374 - root_mean_squared_error: 4.6408\n",
      "Epoch 20/100\n",
      "1/1 [==============================] - 0s 726us/step - loss: 19.0347 - root_mean_squared_error: 4.3629\n",
      "Epoch 21/100\n",
      "1/1 [==============================] - 0s 808us/step - loss: 16.7713 - root_mean_squared_error: 4.0953\n",
      "Epoch 22/100\n",
      "1/1 [==============================] - 0s 846us/step - loss: 14.7296 - root_mean_squared_error: 3.8379\n",
      "Epoch 23/100\n",
      "1/1 [==============================] - 0s 877us/step - loss: 12.8934 - root_mean_squared_error: 3.5907\n",
      "Epoch 24/100\n",
      "1/1 [==============================] - 0s 994us/step - loss: 11.2479 - root_mean_squared_error: 3.3538\n",
      "Epoch 25/100\n",
      "1/1 [==============================] - 0s 814us/step - loss: 9.7790 - root_mean_squared_error: 3.1271\n",
      "Epoch 26/100\n",
      "1/1 [==============================] - 0s 868us/step - loss: 8.4735 - root_mean_squared_error: 2.9109\n",
      "Epoch 27/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 7.3189 - root_mean_squared_error: 2.7054\n",
      "Epoch 28/100\n",
      "1/1 [==============================] - 0s 762us/step - loss: 6.3034 - root_mean_squared_error: 2.5107\n",
      "Epoch 29/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 5.4154 - root_mean_squared_error: 2.3271\n",
      "Epoch 30/100\n",
      "1/1 [==============================] - 0s 763us/step - loss: 4.6441 - root_mean_squared_error: 2.1550\n",
      "Epoch 31/100\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 3.9789 - root_mean_squared_error: 1.9947\n",
      "Epoch 32/100\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 3.4096 - root_mean_squared_error: 1.8465\n",
      "Epoch 33/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 2.9266 - root_mean_squared_error: 1.7107\n",
      "Epoch 34/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 2.5206 - root_mean_squared_error: 1.5876\n",
      "Epoch 35/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 2.1827 - root_mean_squared_error: 1.4774\n",
      "Epoch 36/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.9046 - root_mean_squared_error: 1.3801\n",
      "Epoch 37/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.6783 - root_mean_squared_error: 1.2955\n",
      "Epoch 38/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.4965 - root_mean_squared_error: 1.2233\n",
      "Epoch 39/100\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 1.3525 - root_mean_squared_error: 1.1630\n",
      "Epoch 40/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.2400 - root_mean_squared_error: 1.1136\n",
      "Epoch 41/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.1536 - root_mean_squared_error: 1.0741\n",
      "Epoch 42/100\n",
      "1/1 [==============================] - 0s 902us/step - loss: 1.0883 - root_mean_squared_error: 1.0432\n",
      "Epoch 43/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.0398 - root_mean_squared_error: 1.0197\n",
      "Epoch 44/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.0044 - root_mean_squared_error: 1.0022\n",
      "Epoch 45/100\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 0.9792 - root_mean_squared_error: 0.9895\n",
      "Epoch 46/100\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 0.9614 - root_mean_squared_error: 0.9805\n",
      "Epoch 47/100\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 0.9493 - root_mean_squared_error: 0.9743\n",
      "Epoch 48/100\n",
      "1/1 [==============================] - 0s 824us/step - loss: 0.9410 - root_mean_squared_error: 0.9700\n",
      "Epoch 49/100\n",
      "1/1 [==============================] - 0s 876us/step - loss: 0.9355 - root_mean_squared_error: 0.9672\n",
      "Epoch 50/100\n",
      "1/1 [==============================] - 0s 779us/step - loss: 0.9318 - root_mean_squared_error: 0.9653\n",
      "Epoch 51/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.9293 - root_mean_squared_error: 0.9640\n",
      "Epoch 52/100\n",
      "1/1 [==============================] - 0s 752us/step - loss: 0.9275 - root_mean_squared_error: 0.9631\n",
      "Epoch 53/100\n",
      "1/1 [==============================] - 0s 659us/step - loss: 0.9262 - root_mean_squared_error: 0.9624\n",
      "Epoch 54/100\n",
      "1/1 [==============================] - 0s 704us/step - loss: 0.9251 - root_mean_squared_error: 0.9618\n",
      "Epoch 55/100\n",
      "1/1 [==============================] - 0s 705us/step - loss: 0.9241 - root_mean_squared_error: 0.9613\n",
      "Epoch 56/100\n",
      "1/1 [==============================] - 0s 974us/step - loss: 0.9232 - root_mean_squared_error: 0.9609\n",
      "Epoch 57/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.9224 - root_mean_squared_error: 0.9604\n",
      "Epoch 58/100\n",
      "1/1 [==============================] - 0s 900us/step - loss: 0.9215 - root_mean_squared_error: 0.9599\n",
      "Epoch 59/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.9205 - root_mean_squared_error: 0.9595\n",
      "Epoch 60/100\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 0.9196 - root_mean_squared_error: 0.9590\n",
      "Epoch 61/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.9186 - root_mean_squared_error: 0.9585\n",
      "Epoch 62/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.9176 - root_mean_squared_error: 0.9579\n",
      "Epoch 63/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.9166 - root_mean_squared_error: 0.9574\n",
      "Epoch 64/100\n",
      "1/1 [==============================] - 0s 747us/step - loss: 0.9156 - root_mean_squared_error: 0.9569\n",
      "Epoch 65/100\n",
      "1/1 [==============================] - 0s 701us/step - loss: 0.9145 - root_mean_squared_error: 0.9563\n",
      "Epoch 66/100\n",
      "1/1 [==============================] - 0s 671us/step - loss: 0.9134 - root_mean_squared_error: 0.9557\n",
      "Epoch 67/100\n",
      "1/1 [==============================] - 0s 815us/step - loss: 0.9122 - root_mean_squared_error: 0.9551\n",
      "Epoch 68/100\n",
      "1/1 [==============================] - 0s 772us/step - loss: 0.9111 - root_mean_squared_error: 0.9545\n",
      "Epoch 69/100\n",
      "1/1 [==============================] - 0s 926us/step - loss: 0.9099 - root_mean_squared_error: 0.9539\n",
      "Epoch 70/100\n",
      "1/1 [==============================] - 0s 985us/step - loss: 0.9087 - root_mean_squared_error: 0.9533\n",
      "Epoch 71/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.9075 - root_mean_squared_error: 0.9526\n",
      "Epoch 72/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 736us/step - loss: 0.9062 - root_mean_squared_error: 0.9520\n",
      "Epoch 73/100\n",
      "1/1 [==============================] - 0s 883us/step - loss: 0.9050 - root_mean_squared_error: 0.9513\n",
      "Epoch 74/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.9037 - root_mean_squared_error: 0.9506\n",
      "Epoch 75/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.9024 - root_mean_squared_error: 0.9500\n",
      "Epoch 76/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.9012 - root_mean_squared_error: 0.9493\n",
      "Epoch 77/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8999 - root_mean_squared_error: 0.9486\n",
      "Epoch 78/100\n",
      "1/1 [==============================] - 0s 679us/step - loss: 0.8986 - root_mean_squared_error: 0.9479\n",
      "Epoch 79/100\n",
      "1/1 [==============================] - 0s 941us/step - loss: 0.8973 - root_mean_squared_error: 0.9473\n",
      "Epoch 80/100\n",
      "1/1 [==============================] - 0s 913us/step - loss: 0.8960 - root_mean_squared_error: 0.9466\n",
      "Epoch 81/100\n",
      "1/1 [==============================] - 0s 781us/step - loss: 0.8948 - root_mean_squared_error: 0.9459\n",
      "Epoch 82/100\n",
      "1/1 [==============================] - 0s 630us/step - loss: 0.8935 - root_mean_squared_error: 0.9453\n",
      "Epoch 83/100\n",
      "1/1 [==============================] - 0s 691us/step - loss: 0.8923 - root_mean_squared_error: 0.9446\n",
      "Epoch 84/100\n",
      "1/1 [==============================] - 0s 697us/step - loss: 0.8911 - root_mean_squared_error: 0.9440\n",
      "Epoch 85/100\n",
      "1/1 [==============================] - 0s 923us/step - loss: 0.8899 - root_mean_squared_error: 0.9433\n",
      "Epoch 86/100\n",
      "1/1 [==============================] - 0s 756us/step - loss: 0.8887 - root_mean_squared_error: 0.9427\n",
      "Epoch 87/100\n",
      "1/1 [==============================] - 0s 804us/step - loss: 0.8876 - root_mean_squared_error: 0.9421\n",
      "Epoch 88/100\n",
      "1/1 [==============================] - 0s 897us/step - loss: 0.8866 - root_mean_squared_error: 0.9416\n",
      "Epoch 89/100\n",
      "1/1 [==============================] - 0s 784us/step - loss: 0.8855 - root_mean_squared_error: 0.9410\n",
      "Epoch 90/100\n",
      "1/1 [==============================] - 0s 711us/step - loss: 0.8845 - root_mean_squared_error: 0.9405\n",
      "Epoch 91/100\n",
      "1/1 [==============================] - 0s 674us/step - loss: 0.8836 - root_mean_squared_error: 0.9400\n",
      "Epoch 92/100\n",
      "1/1 [==============================] - 0s 741us/step - loss: 0.8827 - root_mean_squared_error: 0.9395\n",
      "Epoch 93/100\n",
      "1/1 [==============================] - 0s 749us/step - loss: 0.8819 - root_mean_squared_error: 0.9391\n",
      "Epoch 94/100\n",
      "1/1 [==============================] - 0s 738us/step - loss: 0.8811 - root_mean_squared_error: 0.9387\n",
      "Epoch 95/100\n",
      "1/1 [==============================] - 0s 953us/step - loss: 0.8804 - root_mean_squared_error: 0.9383\n",
      "Epoch 96/100\n",
      "1/1 [==============================] - 0s 673us/step - loss: 0.8797 - root_mean_squared_error: 0.9379\n",
      "Epoch 97/100\n",
      "1/1 [==============================] - 0s 979us/step - loss: 0.8791 - root_mean_squared_error: 0.9376\n",
      "Epoch 98/100\n",
      "1/1 [==============================] - 0s 756us/step - loss: 0.8785 - root_mean_squared_error: 0.9373\n",
      "Epoch 99/100\n",
      "1/1 [==============================] - 0s 720us/step - loss: 0.8780 - root_mean_squared_error: 0.9370\n",
      "Epoch 100/100\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.8776 - root_mean_squared_error: 0.9368\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/eric/.conda/envs/mlbase/lib/python3.8/site-packages/numpy/core/_asarray.py:136: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray\n",
      "  return array(a, dtype, copy=False, order=order, subok=True)\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAEKCAYAAAAVaT4rAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/d3fzzAAAACXBIWXMAAAsTAAALEwEAmpwYAAAlg0lEQVR4nO3debzWc/7/8cerBeeX5URJCxomaUFxrDEipjRjOsz4zmQZhDL2rVGYsUuyyxZRBhEq2TqlpGUi7YskS+jUVOKQnFSn1++P9xUnTjrb5/pcy/N+u53bdV2f8/lcn9d1k9f1Ou/P+/N6m7sjIiLZo0bcAYiISHIp8YuIZBklfhGRLKPELyKSZZT4RUSyjBK/iEiWiTzxm1lNM5tpZq8mXu9sZmPMbFHisW7UMYiIyE+SUfFfCiwo9boXMNbdmwFjE69FRCRJIk38ZtYE+APweKnNXYDBieeDgfwoYxARkc3Vivj97wX+CexQalsDd18G4O7LzGzXsg40s+5Ad4A6deoctO+++0YcqohIZpk+ffqX7l7/59sjS/xm9kdghbtPN7P2FT3e3QcAAwDy8vJ82rRp1RugiEiGM7PPytoeZcXfDviTmXUGtgN2NLOngeVm1jBR7TcEVkQYg4iI/ExkY/zu3tvdm7h7U+BvwDh3Px0YCZyZ2O1M4OWoYhARkV+KYx7/7cDxZrYIOD7xWkREkiTqi7sAuPt4YHzi+SqgQzLOKyIiv6Q7d0VEskxSKn4REamYETML6VewkKVFxTTKzaFnx+bkt21cLe+txC8ikmJGzCyk97C5FK8vAaCwqJjew+YCVEvy11CPiEiK6Vew8Mekv0nx+hL6FSyslvdX4hcRSTFLi4ortL2ilPhFRFJMo9wcAHZZU1Tm9qpS4hcRSTH/PHYvLpn6IpMf6cbhn80GIKd2TXp2bF4t76+LuyIiqWTaNLp0Pxdmz+at1kfx8S6701izekREMtCaNXD99XDPPdCgAQwbxjEnncTUCE6lxC8iErcxY6BHD/j0U+jeHfr2hdzcyE6nMX4RkbisWgVnnQW//z3UqgXjx8Ojj0aa9EGJX0Qk+dzh+eehZUt45hm45hqYMweOPjopp9dQj4hIMn3xBVxwAbz6KuTlwejRcMABSQ1BFb+ISDJs3AgPPhiq/LFj4a67YMqUpCd9UMUvIhK999+H886D//4Xjj8eHnkE9tortnBU8YuIRGXdOrjpJmjbFj74AAYPhoKCWJM+qOIXEYnGlCmhyp8/H7p2hXvvhV13jTsqQBW/iEj1Wr0aLrkE2rWDb78NF3GffTZlkj6o4hcRqT6vvw7nnw9LlsCFF8Jtt8EOO8Qd1S9EVvGb2XZmNtXMZpvZfDO7MbH9BjMrNLNZiZ/OUcUgIpIUK1fCqafCH/4A228PkybBAw+kZNKHaCv+H4Bj3f07M6sNTDKzNxK/u8fd74zw3CIi0XOHp5+Gyy8Pwzo33AC9esG228Yd2a+KLPG7uwPfJV7WTvx4VOcTEUmqTz8NwzqjR8Phh8Njj0GrVnFHVS6RXtw1s5pmNgtYAYxx93cTv7rIzOaY2RNmVncLx3Y3s2lmNm3lypVRhikiUn4lJaGDZuvWYV5+//5haCdNkj5EnPjdvcTd2wBNgEPMrDXwMLA30AZYBty1hWMHuHueu+fVr18/yjBFRMpnzpxQ3V9xBbRvH6ZqXngh1EivCZJJidbdi4DxQCd3X574QtgIPAYckowYREQqbe1auO46OOggWLwYhgwJ0zT32CPuyColylk99c0sN/E8BzgO+MDMGpba7SRgXlQxiIhU2YQJoZ/OrbfCaafx2vPjaLd4V37T+3Xa3T6OETML446wwqKc1dMQGGxmNQlfMEPd/VUz+4+ZtSFc6F0M9IgwBhGRyvnmG7j66tAfv2lTGD2aEfVa0nvYXIrXlwBQWFRM72FzAaptWcRkiHJWzxygbRnbz4jqnCIi1WLEiDB2/7//hfH8m26COnXod/u4H5P+JsXrS+hXsDCtEn96XZEQEYnSsmVwyilw0klQrx68805on1ynDgBLi4rLPGxL21OVEr+IiDsMHBh65b/ySmi1MG0aHHzwZrs1ys0p8/AtbU9VSvwikt0++gg6dIBzz4X99w9TNnv3htq1f7Frz47Nyaldc7NtObVr0rNj82RFWy2U+EUkO23YAH37wn77wfTp4SLuW2/BPvts8ZD8to3pc/J+NM7NwYDGuTn0OXm/tBrfB3XnFJFsNGNGqPBnzoT8/LAkYqNG5To0v23jtEv0P6eKX0Syx/ffQ8+eYex+2TJ46SUYPrzcST9TqOIXkewwdix07w6ffBJWxrrjDsjNjTuqWCjxi0jGGDGzkH4FC1laVEyj3Bx6dmxO/p45cNVV8OST8NvfwrhxcMwxcYcaKw31iEhGGDGzkN7D5lJYVIwDhV9/z9u3PMjaZs3hqadCn/w5c7I+6YMqfhHJEP0KFv54V+1u337JzWMe4viPpvJB433Yd+wYaNMm3gBTiBK/iGSEpUXFmG/ktFmjuHr8k9TauJFbjunGoLwufKSkvxklfhHJCIetW8kVQ/txcOH7TNyzDdd0uogvcnejcZrdVZsMSvwikt7WrYO+fXm6/y2srrkNV3W+jBdbdwCztLyrNhmU+EUkfb3zTrgRa/58av71r0w5txdTpn2FlZ7Vk+Y3W0VBiV9E0s9334UVse6/Hxo3hpEj4cQTOQE44bi4g0t9Svwikl5GjYIePeDzz0PP/Ntugx13jDuqtKJ5/CKSHlauhNNPhxNOCP3xJ02C/v2V9CtBiV9EUps7PP00tGgBQ4fCv/8dmqu1axd3ZGkrysXWtzOzqWY228zmm9mNie07m9kYM1uUeKwbVQwikuY++ww6d4YzzgjtFmbMgBtvhG23jTuytBZlxf8DcKy7HwC0ATqZ2WFAL2CsuzcDxiZei4j8pKQE7rsPWrWCiRPDRdzJk6F167gjywiRJX4Pvku8rJ34caALMDixfTCQH1UMIpKG5s6FI46Ayy6D3/0O5s+Hiy+GmjW3eqiUT6Rj/GZW08xmASuAMe7+LtDA3ZcBJB533cKx3c1smplNW7lyZZRhikgqWLsW/vUvOPDA0Dr52Wfhtddgzz3jjizjRJr43b3E3dsATYBDzKzcf6e5+wB3z3P3vPr160cWo4ikgEmToG1buOUW6NoVFiwIj2ZxR5aRkjKrx92LgPFAJ2C5mTUESDyuSEYMIpKCvv0WLrgAjjoKiovDHP2nnoJ69eKOLKNFdgOXmdUH1rt7kZnlAMcBfYGRwJnA7YnHl6OKQURSQ5kLpHwxPST9Zcvg8svhpptg++3jDjUrRHnnbkNgsJnVJPxlMdTdXzWzKcBQMzsH+Bw4JcIYRCRmmxZI2dQrf92SpWx72g2wYCLstx8MGwaHHBJvkFkmssTv7nOAtmVsXwV0iOq8IpJaflwgxZ1T5o7hunED2W7DDzz6+270eOVh2GabuEPMOurVIyKRWlpUzB5fL6NPwQO0+2wO7zZpxTWdLuKTXXanh5J+LJT4RSQ6GzZw1ZyRdBszmPU1anJNxwsZckBH3GpogZQYKfGLSDRmzoRzzuHCmTN5c5/Dufa4HizfIczW0QIp8VKTNhGpXsXFcPXVcPDBsHQpvPAC3w0ZSq3dd8eAxrk59Dl5Py2QEiNV/CJSfd56C847Dz7+GM45B/r1g7p1yQfyD2wSd3SSoIpfRKru66/DEojHHhtejx0Ljz8OddV8NxUp8YtI5bnDiy+GXvmDBoUhnrlzf/oCkJSkoR4RqZzCwrD04csvhz47r78eGqxJylPFLyIVs3EjPPootGwJBQVwxx0wdaqSfhpRxS+S5crso7OlGTcLF4aLtxMnhuGcAQNg772TG7BUmSp+kSy2qY9OYVExDhQWFdN72FxGzCzcfMd16+DWW2H//cMY/hNPwJtvKumnKSV+kSz2Yx+dUorXl9CvYOFPG6ZOhbw8uO466NIl9Mo/+2z1yk9jSvwiWWxpUfGWt69ZA1dcAYcfDl99BSNGwNChsNtuyQ1Sqp3G+EWyWKPcHArLSP75K+ZB64tg8WL4xz+gTx/YaafkByiRUMUvksV6dmxOTu2fFjGv+/033Pf6PdzzZC/YdttwEfehh5T0M4wqfpEstmn2Tr9RH5A3ZRQ3jHuMnX5YE8bzr70Wttsu5gglCkr8Ilkuf+cN5E+8C954Aw49FB57LKyMJRlLQz0i2aqkBO6/H1q1ggkT4N57YfJkJf0sEFniN7PdzewtM1tgZvPN7NLE9hvMrNDMZiV+OkcVg4hswfz5cOSRcOml4XHevPC8Zs2tHytpL8qhng3Ale4+w8x2AKab2ZjE7+5x9zsjPLeIlOWHH+C228IsnR13hKefhlNP1Zz8LBPlYuvLgGWJ56vNbAGglRdE4jJ5cmi3sGABnH463H031K8fd1QSg6SM8ZtZU6At8G5i00VmNsfMnjCzMht2m1l3M5tmZtNWrlyZjDBFMtO334YumkceGW7KeuMN+M9/lPSzWOSJ38y2B14CLnP3b4GHgb2BNoS/CO4q6zh3H+Duee6eV1//QEUq59VXw8Xbhx8OY/jz50OnTnFHJTGLNPGbWW1C0n/G3YcBuPtydy9x943AY8AhUcYgkpWWL4e//Q1OPBFyc2HKlDBrZ/vt445MUkCUs3oMGAgscPe7S21vWGq3k4B5UcUgknXcw0pYLVrA8OFw880wfXqYny+SEOWsnnbAGcBcM5uV2HYN0NXM2gAOLAZ6RBiDSPb45BPo3j2sd3vkkeFGrH33jTsqSUFRzuqZBJQ1R+z1qM4pkpU2bID77oN//Qtq1Qq9dXr0gBq6P1PKppYNIuls1iw499wwnHPiiSHpN2kSd1SS4lQSiKSj4mLo3TsskPLFF6FP/ssvK+lLuajiF0k348eHG7E++gi6dYN+/WDnneOOStKIKn6RdFFUFBL+McfAxo1hzduBA5X0pcKU+EXSwbBhYYrmE09Az55hwfMOHeKOStKUhnpEUtnSpXDRRWFOfps28NprcOCBcUclaU4Vv0gq2rgRBgwIVf4bb0DfvjB1qpK+VAtV/CKp5sMPw41Yb78N7duHL4BmzeKOSjKIKn6RVLF+feiTv//+MHs2PP44jBunpC/VThW/SCp4771wI9acOfCXv4QlERs23PpxIpWgil8kTmvWwJVXwmGHwZdfhou4L7ygpC+RUsUvEpfRo0NPncWL4fzz4fbbYaedNttlxMxC+hUsZGlRMY1yc+jZsTn5bbWQnVTNryZ+Mzv5136/qce+iFTAqlWhyh88GPbZByZMgKOO+sVuI2YW0nvYXIrXlwBQWFRM72FzAZT8pUq2VvGf+Cu/c0CJX6S83OH55+GSS+Drr+Haa+G662C77crcvV/Bwh+T/ibF60voV7BQiV+q5FcTv7ufnaxARDLa55/DBReEG7AOPji0W9h//189ZGlRcYW2i5RXuS7umlkDMxtoZm8kXrc0s3OiDU0kA2zcCP37h3Vv33oL7rknLIO4laQP0Cg3p0LbRcqrvLN6BgEFQKPE6w+ByyKIRyRzvP9+WAnr4ovhiCPCQueXXQY1a5br8J4dm5NTe/N9c2rXpGfH5hEEK9mkvIm/nrsPBTYCuPsGoOTXDxHJUj/8ADfeGHrrLFwITz0Fo0ZB06YVepv8to3pc/J+NM7NwYDGuTn0OXk/je9LlZV3OucaM9uFcEEXMzsM+CayqETS1ZQp4Uas99+HU08NQzu77lrpt8tv21iJXqpdeSv+K4CRwN5mNhl4Crj41w4ws93N7C0zW2Bm883s0sT2nc1sjJktSjzWrdInEEkFq1eHIZ127cLz116DZ56pUtIXiUq5Er+7zwCOBo4AegCt3H3OVg7bAFzp7i2Aw4ALzawl0AsY6+7NgLGJ1yLp67XXwsXbBx8MyX/+fOjcOe6oRLaoXEM9ZrYdcAFwJGG4Z6KZPeLua7d0jLsvA5Ylnq82swVAY6AL0D6x22BgPHB1JeMXic+KFeFi7ZAh0LIlTJ4Mhx8ed1QiW1XeoZ6ngFbAA0B/oCXwn/KexMyaAm2Bd4EGiS+FTV8OZf4tbGbdzWyamU1buXJleU8lEj13pt98H980/S3rnn+Bx487k5effFVJX9JGeS/uNnf3A0q9fsvMZpfnQDPbHngJuMzdvzWzcp3Q3QcAAwDy8vK8nHGKROvTT1lx6lkc9M4EpjVuQa9OF/NRvT3IeWUhXnsbXYiVtFDein9mYiYPAGZ2KDB5aweZWW1C0n+mVF+f5WbWMPH7hsCKioUsEoOSErj7bmjdmu1nvMd1x/+DU07ry0f19gB+aqUgkg621qRtLmFMvzbwdzP7PPF6T+D9rRxrwEBggbvfXepXI4EzgdsTjy9XOnqRMlR7R8vZs+G880LP/D/+keN2/zNLd6z/i93USkHSxdaGev5YhfduB5wBzDWzWYlt1xAS/tBEy4fPgVOqcA6RzVRrR8u1a+Hmm+GOO6BuXXjuOfi//8P6vgVlJHm1UpB0sbUmbZ+Vfm1muwJltxL85bGTgC0N6HcoV3QiFVRtHS3ffjuse/vhh3DWWXDnnbDLLkBopVD6ywXUSkHSS3mbtP3JzBYBnwJvA4uBNyKMS6RSqtzR8ptvwuIo7duHNXDHjIEnn/wx6YNaKUj6K++snpsJN2G96e5tzewYoGt0YYlUTqPcHAorOwwzfDhceCEsXw5XXQU33AB16pS5q1opSDor76ye9e6+CqhhZjXc/S2gTXRhiVROpTpaLlsGf/4znHxyaLHw7rvQr98Wk75IuitvxV+UmI8/AXjGzFYQWjKIpJRNVXi5ZvW4w8CBobpfuxb69AlLItauneSoRZLL3Ld+b5SZ1QHWEi7WngbsRJibvyra8IK8vDyfNm1aMk4l2WLRonDxdvx4OPpoeOwxaNYs7qhEqpWZTXf3vJ9vL1fF7+5rSr0cXG1RiSTb+vVw112hX/6224aE360b1CjvqKdI+tvaDVyrSfTg//mvAHf3HSOJSiQK06eHXvmzZoUx/QcegIYN445KJOm2No9/h2QFIhKZ77+H668PLRcaNIBhw+Ckk+KOSiQ25b24K5Ke3nwzzMv/5JMwpt+3L+Tmxh2VSKw0sCmZ6auv4Oyz4fjjw+Lm48fDo48q6YugxC+Zxh2efx5atICnn4ZrrglN1o4+Ou7IRFKGhnokcyxZAhdcAK+8Anl5MHo0HHDA1o8TyTKq+CX9bdwIDz0Ulj98880wXXPKFCV9kS1QxS/pbcGCMEXzv/8N4/mPPAJ77RV3VCIpTRW/pKd16+Cmm6BNG/jgAxg8GAoKlPRFykEVv6Sfd94JVf78+dC1K9x7b2iuJiLloopf0sfq1XDppXDEEaFv/iuvwLPPKumLVJAqfkkPb7wB558PX3wReubfdhvsoBvLRSojsorfzJ4wsxVmNq/UthvMrNDMZiV+Okd1fskQK1fCaadB586hP/6kSaHHjpK+SKVFOdQzCOhUxvZ73L1N4uf1CM8v6cw93IDVogW88EJYDWvmzDDMIyJVEtlQj7tPMLOmUb2/ZLDFi8OwTkEBHHYYPP44tGoVd1QiGSOOi7sXmdmcxFBQ3S3tZGbdzWyamU1buXJlMuOTuJSUhBk6rVrB5MlhSGfSJCV9kWqW7MT/MLA3Yb3eZcBdW9rR3Qe4e56759WvXz9J4Uls5syBww+Hyy+H9u3DVM2LLgoN1kSkWiV1Vo+7L9/03MweA15N5vklfiNmFm62Hu7V7ffkT68+Gdol160LQ4bAX/8KZnGHKpKxkpr4zayhuy9LvDwJmPdr+0tmGTGzkN7D5lK8vgSARnPfo/WdZ8KqJfD3v4eFUnbZJeYoRTJfZInfzIYA7YF6ZrYEuB5ob2ZtCMs5LgZ6RHV+ST39ChZSvL6EHX5Yw9XjB3H6rDf4YqcGXH52X+554p9xhyeSNaKc1dO1jM0DozqfpL6lRcUcv+gdbh79EPXXFPHYwfncfeTprN1mO+6JOziRLKI7dyU5/vc/Br52B8fOm8CC+k3pfvJ1zGm4DwCNc3NiDk4kuyjxS7Tc4ckn4corOfr7Yu5pfyYP5p3Ehprhn15O7Zr07Ng85iBFsouatEl0PvoIOnSAc86B/fen5pzZ/ObuW2mwyw4YodLvc/J+5LdtHHekIllFFb9Uvw0bwgyd66+HbbYJi5yfey7UqEE+KNGLxEyJX6rXjBkhyc+cCfn58OCD0KhR3FGJSCka6pHq8f338M9/wiGHwLJl8NJLMHy4kr5IClLFL1U3bhx07w4ffwznnffTXbgikpJU8Uvlff11uHDboUNosTBuHAwYoKQvkuKU+KXi3EOP/BYtwiLnvXqFJmvHHBN3ZCJSDhrqkYpZsiQsfThyJBx4IIwaBW3axB2ViFSAKn4pn40b4eGHoWVLGDMG7rwT3n1XSV8kDanil6374INw0XbSJDjuuDAvf6+94o5KRCpJFb9s2bp1cMstcMABYWGUJ5+E0aOV9EXSnCp+Kdu774YbsebNCwuj3HcfNGgQd1QiUg1U8cvmvvsOLrssLINYVBQu4j73nJK+SAZRxS8/GTUKzj8fPvsszNy57TbYcce4oxKRaqaKX+DLL+GMM+CEEyAnJ1zE7d9fSV8kQynxZzN3eOaZcCPW88/Dv/8Ns2ZBu3ZxRyYiEYos8ZvZE2a2wszmldq2s5mNMbNFiUfd2x+Xzz6Dzp3h9NNh771DV80bb4Rtt407MhGJWJQV/yCg08+29QLGunszYGzitSRTSQncfz+0agUTJ4bnkydD69ZxRyYiSRJZ4nf3CcBXP9vcBRiceD4YyI/q/FKGefPCMM6ll8Lvfhfm5l98MdSsGXdkIpJEyZ7V08DdlwG4+zIz2zXJ589OP/wAt94Kt98OO+0UxvW7dg0dNRNGzCykX8FClhYV0yg3h54dm2ulLJEMlbLTOc2sO9AdYI899og5mjQ2aVJot/DBB2Hmzt13Q716m+0yYmYhvYfNpXh9CQCFRcX0HjYX0DKJIpko2bN6lptZQ4DE44ot7ejuA9w9z93z6tevn7QAM8a334a5+EcdBcXFYY7+U0/9IukD9CtY+GPS36R4fQn9ChYmK1oRSaJkJ/6RwJmJ52cCLyf5/NnhlVdCF82HHw534c6bBx07bnH3pUXFFdouIuktsqEeMxsCtAfqmdkS4HrgdmComZ0DfA6cEtX5M0m5x9+XL4dLLoGhQ2G//cK6t4ceutX3b5SbQ2EZSb5Rbk51hC8iKSayxO/uXbfwqw5RnTMTlWv83R0GDYIrr4Q1a0JHzZ49YZttynWOnh2bb3YOgJzaNenZsXm1fhYRSQ26czfFbXX8/eOP4fjjoVu3MBd/9my49tpyJ30IXyB9Tt6Pxrk5GNA4N4c+J++nC7siGSplZ/VIsKVx9uVffRdWwfr3v6FWrTCe37071Kjcd3l+28ZK9CJZQok/xZU1/t5q+cfcNbo/LF0EXbrAgw9CYyVtESkfDfWkuJ4dm5NTO9xZu+36H7h6/CBeHnw5TX8oghdegOHDlfRFpEJU8ae4TcMvYx96jiteuovffL2Mxfl/o+kTD0Fd9bgTkYpT4k91X39N/kM3kP/446GL5otjaXrssXFHJSJpTEM9qco9zMNv2TIscn711TB3Lijpi0gVqeJPRYWFcNFFMGIEtG0Lr70GBx4Yd1QikiFU8aeSjRvh0UdDlT9qFNxxB0ydqqQvItVKFX+qWLgwzMOfMCEM5zz6KPz2t3FHJSIZSBV/3NavD73yDzgA5syBJ56AN99U0heRyKjij9N778E554SLtqecEpZB3G23uKMSkQynij8Oa9bAFVfAYYfBqlXhIu7QoUr6IpIUqviTraAAzj8fFi+Gf/wD+vQJyyGKiCSJKv5kWbUK/v536NQJtt0WJk6Ehx5S0heRpFPij5o7DBkCLVqEx+uug1mz4Mgj445MRLKUhnqi9PnnYTjn9dfhkENg7NiwMpaISIxU8UehpAQeeABatYK334Z774X//ldJX0RSgir+6jZ/Ppx7LrzzTljg/JFHoGnTuKMSEflRLInfzBYDq4ESYIO758URR3XYtBD6l19+yz9nDefst4dQY6cd4emn4dRTwSzuEEVENhNnxX+Mu38Z4/mrbNNC6C0+ncegUffTbNUXjGx9DLXuu5fOx+4fd3giImXSUE8VPPjyDHq9/ihnzHidpTvW46y/3MD4vfNoPPVLOqt7soikqLgSvwOjzcyBR919wM93MLPuQHeAPfbYI8nhlcOrrzL47m7stnoVgw46kTt/dwbfb5MDbHmBdBGRVBBX4m/n7kvNbFdgjJl94O4TSu+Q+DIYAJCXl+dxBFmmFSvg0kvhuecobvAbTs7vzaxGzTfbpVFuTkzBiYhsXSzTOd19aeJxBTAcOCSOOCrEHQYPDjdiDRsGN9/MvJFjWbhny812y6ldk54dm2/hTURE4pf0it/M6gA13H114vnvgZuSHUeFfPIJ9OgR2iUfeSQ89hjsuy9dAK+9Df0KFrK0qJhGuTn07Nj8xwXSRURSURxDPQ2A4RamOdYCnnX3UTHEsXUbNsB998G//gW1aoXeOj16QI2f/lDKb9tYiV5E0krSE7+7fwIckOzzVtjs2eFGrGnT4MQTQ9Jv0iTuqEREqkwtG36uuBiuuQYOOij02hk6FF5+WUlfRDKG5vGX9vbbcN55sGgRdOsG/frBzjvHHZWISLVSxQ9QVBQWOm/fPjRYe/NNGDhQSV9EMpIS//Dh0LJlSPQ9e4b1bzt0iDsqEZHIZO9Qz9KlcPHFYU5+mzbwyithXF9EJMNlX8W/cWOYh9+yZVggpW9fmDpVSV9EskZ2VfwffhjG8t9+O4znDxgAzZrFHZWISFJlR8W/fj306QP77x/Wu338cRg3TklfRLJS5lf806aFG7Fmz4a//AXuvx8aNow7KhGR2GR2xX/rrXDoobByZZi988ILSvoikvUyO/HvtVe4Iev99yE/P+5oRERSQmYP9XTtGn5ERORHmV3xi4jILyjxi4hkGSV+EZEso8QvIpJllPhFRLKMEr+ISJZR4hcRyTJK/CIiWcbcPe4YtsrMVgKfVfLwesCX1RhOnPRZUk+mfA7QZ0lVVfkse7p7/Z9vTIvEXxVmNs3d8+KOozros6SeTPkcoM+SqqL4LBrqERHJMkr8IiJZJhsS/4C4A6hG+iypJ1M+B+izpKpq/ywZP8YvIiKby4aKX0RESlHiFxHJMhmd+M2sk5ktNLOPzKxX3PFUhpntbmZvmdkCM5tvZpfGHVNVmVlNM5tpZq/GHUtVmFmumb1oZh8k/vscHndMlWVmlyf+fc0zsyFmtl3cMZWXmT1hZivMbF6pbTub2RgzW5R4rBtnjOWxhc/RL/Hva46ZDTez3Oo4V8YmfjOrCTwInAC0BLqaWct4o6qUDcCV7t4COAy4ME0/R2mXAgviDqIa3AeMcvd9gQNI089kZo2BS4A8d28N1AT+Fm9UFTII6PSzbb2Ase7eDBibeJ3qBvHLzzEGaO3u+wMfAr2r40QZm/iBQ4CP3P0Td18HPAd0iTmmCnP3Ze4+I/F8NSG5NI43qsozsybAH4DH446lKsxsR+B3wEAAd1/n7kWxBlU1tYAcM6sF/D9gaczxlJu7TwC++tnmLsDgxPPBQH4yY6qMsj6Hu4929w2Jl+8ATarjXJmc+BsDX5R6vYQ0TpgAZtYUaAu8G3MoVXEv8E9gY8xxVNVewErgycSw1eNmVifuoCrD3QuBO4HPgWXAN+4+Ot6oqqyBuy+DUDwBu8YcT3XoBrxRHW+UyYnfytiWtnNXzWx74CXgMnf/Nu54KsPM/giscPfpccdSDWoBBwIPu3tbYA3pMZzwC4nx7y7Ab4BGQB0zOz3eqKQ0M7uWMOz7THW8XyYn/iXA7qVeNyGN/nwtzcxqE5L+M+4+LO54qqAd8CczW0wYejvWzJ6ON6RKWwIscfdNf329SPgiSEfHAZ+6+0p3Xw8MA46IOaaqWm5mDQESjytijqfSzOxM4I/AaV5NN15lcuJ/D2hmZr8xs20IF6tGxhxThZmZEcaRF7j73XHHUxXu3tvdm7h7U8J/j3HunpaVpbv/D/jCzJonNnUA3o8xpKr4HDjMzP5f4t9bB9L0QnUpI4EzE8/PBF6OMZZKM7NOwNXAn9z9++p634xN/IkLIhcBBYR/xEPdfX68UVVKO+AMQnU8K/HTOe6gBICLgWfMbA7QBrgt3nAqJ/FXy4vADGAuIS+kTcsDMxsCTAGam9kSMzsHuB043swWAccnXqe0LXyO/sAOwJjE//uPVMu51LJBRCS7ZGzFLyIiZVPiFxHJMkr8IiJZRolfRCTLKPGLiGQZJX7JamZ2SaKzZoXuiDSzpmZ2alRxiURJiV+y3QVAZ3c/rYLHNQUqnPgTXWNFYqXEL1krcTPMXsBIM7s20Q/9vUTTtS6JfZqa2UQzm5H42dTK4HbgqMRNNZeb2Vlm1r/Ue79qZu0Tz78zs5vM7F3gcDM73cymJo59VF8GkmxK/JK13P18Qv+mY4A6hBYSByde90t021wBHO/uBwJ/Be5PHN4LmOjubdz9nq2cqg4wz90PBVYl3qedu7cBSoCK/rUhUiW14g5AJEX8ntBA7qrE6+2APQhfDP3NrA0hSe9TifcuITTZg9AH5yDgvdAWhxzSuIGYpCclfpHAgD+7+8LNNprdACwnrLBVA1i7heM3sPlf0KWXLlzr7iWlzjPY3atlJSWRytBQj0hQAFyc6E6JmbVNbN8JWObuGwnN8jaNx68mNM/aZDHQxsxqmNnuhBXgyjIW+IuZ7Zo4z85mtme1fhKRrVDiFwluBmoDcxKLXd+c2P4QcKaZvUMY5lmT2D4H2GBms83scmAy8Cmhu+WdhE6Xv+Du7wPXAaMTXT3HAA2j+UgiZVN3ThGRLKOKX0Qkyyjxi4hkGSV+EZEso8QvIpJllPhFRLKMEr+ISJZR4hcRyTL/H6IbtWdRuXVnAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAEGCAYAAABiq/5QAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/d3fzzAAAACXBIWXMAAAsTAAALEwEAmpwYAAAmFklEQVR4nO3deXxU9b3/8dcnCwlZWBICAgHDjuxCVAr2Alor1rVVq1ZatLXWtrZe8WG1t4v11/beXmutdanWat1rb+tSrVqXqqi4oICIIKuAENawBMISyPL5/TETGkIShiQzJ5nzfj56zMyZmXPe34R+cvI93/M95u6IiEh4pAQdQEREEkuFX0QkZFT4RURCRoVfRCRkVPhFREImLegAsejWrZsXFRUd8efWl+1l+55Khvfq1PqhRETauLlz525x94L669tF4S8qKmLOnDlH/Ln7Zq3i589+zIs//hz5ORlxSCYi0naZ2acNrU/qrp6i/CwAPt22J+AkIiJtR1IX/qNrC//W3QEnERFpO5K68Bd2zcIMPt2qI34RkVrtoo+/uTLTU+nZKVOFXyTEKisrKSkpoaKiIugocZOZmUlhYSHp6ekxvT+pCz/A0fnZ6uoRCbGSkhJyc3MpKirCzIKO0+rcna1bt1JSUkK/fv1i+kxSd/VApJ9fR/wi4VVRUUF+fn5SFn0AMyM/P/+I/qIJQeHPZuvu/ezaVxV0FBEJSLIW/VpH2r4QFH6N7BERqStEhV/dPSISjJycnKAjHCQEhT8bUOEXEamV9IU/JyONbjkd1NUjIm3K/PnzGT9+PKNGjeKLX/wi27dvB+C2225j2LBhjBo1igsvvBCA119/nTFjxjBmzBiOPfZYysvLW7TvpB/OCdA3TyN7RARu/MciPl6/s1W3OaxXJ244c/gRf+5rX/sat99+O5MmTeKnP/0pN954I7feeiu/+tWvWLVqFRkZGZSVlQFw8803c+eddzJx4kR27dpFZmZmizIn/RE/QJHG8otIG7Jjxw7KysqYNGkSANOnT+eNN94AYNSoUVx88cU88sgjpKVFjs0nTpzIjBkzuO222ygrKzuwvrnCccSfn8VT89exr6qajLTUoOOISECac2SeaM899xxvvPEGzzzzDD//+c9ZtGgR119/PaeffjrPP/8848eP51//+hdDhw5t9j5Cc8TvDmu37Q06iogInTt3pmvXrrz55psAPPzww0yaNImamhrWrl3LlClTuOmmmygrK2PXrl188sknjBw5kuuuu47i4mKWLFnSov2H5ogfImP5B3ZvW8OqRCT57dmzh8LCwgPPZ8yYwYMPPsgVV1zBnj176N+/P/fffz/V1dVMmzaNHTt24O5cffXVdOnShZ/85Ce89tprpKamMmzYME477bQW5QlF4S/SkE4RCVBNTU2D6999991D1s2aNeuQdbfffnur5glFV0/XrHRyM9J0gldEhJAUfjPj6G5ZuhOXiAghKfwAR+dlq6tHJKTcPegIcXWk7QtP4c/PYu22PeyvarivTUSSU2ZmJlu3bk3a4l87H/+RXNQVipO7ELm6rqrGWbapnBG9OwcdR0QSpLCwkJKSEkpLS4OOEje1d+CKVWgK/8hosf9o3Q4VfpEQSU9Pj/nOVGERt64eM/uTmW02s4V11v3azJaY2QIze8rMusRr//X1zcuiU2YaH63bkahdioi0SfHs438AmFpv3cvACHcfBSwDfhjH/R/EzBhZ2JmPSlT4RSTc4lb43f0NYFu9dS+5e+09EN8FYu+UagUje3dhycad7KuqTuRuRUTalCBH9Xwd+Gcidziyd2cqq51lG3clcrciIm1KIIXfzH4EVAGPNvGey81sjpnNaa2z8XVP8IqIhFXCC7+ZTQfOAC72JgbWuvs97l7s7sUFBQWtsu8+eR3p3DGdj9aVtcr2RETao4QO5zSzqcB1wCR3T/hltGbGyN6ddcQvIqEWz+GcjwHvAEPMrMTMvgHcAeQCL5vZfDO7O177b8zIws4s3ViuE7wiElpxO+J394saWH1fvPYXq9oTvEs3ljOqsEvQcUREEi40c/XUqj3Bu0Dj+UUkpEJX+Au7dqRLVjoL1c8vIiEVusJfe4JXR/wiElahK/wQ6e5Ztqmcikqd4BWR8All4R/btytVNc78tWVBRxERSbhQFv7j+uVhBu+u3Bp0FBGRhAtl4e/cMZ1hPTsxe+W2w79ZRCTJhLLwA5zQL595a7brQi4RCZ3QFv7x/fPYV1XDh2s1ukdEwiW0hf/4aD//bPXzi0jIhLbwd8nqwNCjOvHuKhV+EQmX0BZ+gBP65TH30+3sr6oJOoqISMKEuvCP759PRWUNC0rKgo4iIpIwoS78x/fLA2D2Kg3rFJHwaLLwm1mqmf06UWESLS+7A0OPytWFXCISKk0WfnevBsaZmSUoT8Kd0C+POau3U1mtfn4RCYdYuno+AJ42s6+a2Zdql3gHS5TPDMhnb2U18z7dHnQUEZGEiOUOXHnAVuCkOusceDIuiRJswsBupKUYM5eVckL//KDjiIjE3WELv7tfmoggQemUmc64o7syc2kp100dGnQcEZG4O2xXj5kVmtlTZrbZzDaZ2RNmVpiIcIkyZWh3Fm/YycYdFUFHERGJu1j6+O8HngF6Ab2Bf0TXJY3JQwoAeH3Z5oCTiIjEXyyFv8Dd73f3qujyAFAQ51wJNaRHLj07ZzJzaWnQUURE4i6Wwr/FzKZFx/Snmtk0Iid7k4aZMXlIAbOWb9GwThFJerEU/q8DXwY2AhuA86LrksrkId0p31fFXA3rFJEk1+SoHjNLBf7b3c9KUJ7ATBzYjfRU47WlmxmvYZ0iksRiuXK3wMw6JChPYHIy0jiuKI/X1c8vIkkulq6e1cBbZvYTM5tRuxzuQ2b2p+gQ0IV11uWZ2ctmtjz6tWsLsre6yUMKWLKxnPVle4OOIiISN7EU/vXAs9H35tZZDucBYGq9ddcDr7j7IOCV6PM24+RjegDw8sebAk4iIhI/sfTxD3L3aUe6YXd/w8yK6q0+G5gcffwgMBO47ki3HS8DCnIY1D2HFxZuZPqEoqDjiIjERaL7+Hu4+4botjcA3Rt7o5ldbmZzzGxOaWni+t2njjiK2au2sm33/oTtU0QkkeLWx99S7n6Puxe7e3FBQeKuFzt1+FHUOPxL3T0ikqTi2cffkE1m1hMg+rXNzZEwvFcnCrt25IVFG4OOIiISF7HMznlj/XVmFst0zg15BpgO/Cr69elmbiduzIypw4/ioXc+pbyiktzM9KAjiYi0qkaP+M1sVp3HD9d7+b3DbdjMHgPeAYaYWYmZfYNIwT/FzJYDp0SftzmnjTyK/dU1vLqkzf1BIiLSYk0duWfXeTyi3muHvRWju1/UyEsnH+6zQTu2T1cKcjN4cdFGzh7TO+g4IiKtqqk+fm/kcUPPk0pKinHq8B68tqSUisrqoOOIiLSqpgp/FzP7opmdG31ce7/dc4HOCcoXmKnDe7K3spqZS9XdIyLJpamunteBs+o8PrPOa2/ELVEbMb5/Ht1yOvDMh+uZOqJn0HFERFpNo4U/2e+1ezhpqSmcMaoXf35vjUb3iEhSiWUcf2idNaYX+6tqeHGRLuYSkeShwt+EY/t0oU9eR56evy7oKCIirUaFvwlmxtmje/PWii2Ulu8LOo6ISKtotI/fzL7U1Afd/cnWj9P2nD2mF3e8toLnFqznkon9go4jItJiTY3qqR3F0x2YALwafT6FyHTKoSj8g3rkckzPTjz9oQq/iCSHRrt63P3S6MgeB4a5+7nufi4wPGHp2oizx/TigzVlrNm6J+goIiItFksff1HtHPpRm4DBccrTJp01uhdm8OQHJUFHERFpsVgK/0wze9HMLjGz6cBzwGtxztWm9OrSkYkDuvH43BJqapJ6tgoRCYHDFn53vxK4GxgNjAHucffvxTlXm3N+cSEl2/fy7qqtQUcREWmRWOfVnweUu/u/zCzLzHLdvTyewdqaU4cfRW5mGo/PKWHCgG5BxxERabbDHvGb2TeBx4E/RFf1Bv4ex0xtUmZ6KmeO7sXzCzdQXlEZdBwRkWaLpY//u8BEYCeAuy+niZukJ7PzxxVSUVnDcws2HP7NIiJtVCyFf5+77699Er3tYijPcI7p04WB3XP421yN7hGR9iuWwv+6mf0X0NHMTgH+BvwjvrHaJjPj/HGFzP10O5+U7go6johIs8RS+K8DSoGPgG8BzwM/jmeotuyLY3uTmmL89f21QUcREWmWJkf1mFkKsMDdRwB/TEyktq17bianHNODv80tYcbnB5ORlhp0JBGRI9LkEb+71wAfmlnfBOVpFy4e35dtu/fzwsKNQUcRETlisYzj7wksMrP3gN21K939rMY/ktwmDuhG37wsHp29hrPH9A46jojIEYml8N8Y9xTtTEqK8ZUT+vKrfy5h+aZyBvXIDTqSiEjMYpmy4fWGlkSEa8vOH1dIeqrx5/fWBB1FROSIxHLl7ngze9/MdpnZfjOrNrOdiQjXluXnZDB1RE+emFvC3v3VQccREYlZLMM57wAuApYDHYHLouuazcyuNrNFZrbQzB4zs8yWbC8oF5/Ql50VVfxjwfqgo4iIxCyme+66+wog1d2r3f1+YHJzd2hmvYHvA8XRYaKpwIXN3V6QTuiXx6DuOTz0zmrcQ3kxs4i0Q7EU/j1m1gGYb2Y3mdnVQHYL95tG5ErgNCALaJeHzGbG9AlFLFy3k3lrtgcdR0QkJrEU/q8SOSq/kshwzj7Auc3dobuvA24G1gAbgB3u/lL995nZ5WY2x8zmlJaWNnd3cfelsb3JzUzj/rdWBx1FRCQmsYzq+dTd97r7Tne/0d1nRLt+msXMugJnA/2AXkC2mU1rYL/3uHuxuxcXFBQ0d3dxl9UhjQuP68M/F25kw469QccRETmsWEb1rDKzlfWXFuzzc8Aqdy9190rgSWBCC7YXuK+OL6LGnUff1dBOEWn7YrmAq7jO40zgfCCvBftcA4w3syxgL3AyMKcF2wtc3/wsTh7ag8feW8OVJw0kM13z94hI2xVLV8/WOss6d78VOKm5O3T32UTu6DWPyIyfKcA9zd1eW3HpxCK27t7PPz5sl+epRSREDnvEb2Zj6zxNIfIXQIvmKHD3G4AbWrKNtmbCgHwG98jhvlmrOG9cIWYWdCQRkQbF0tXzmzqPq4DVwJfjkqYdMzMu+2x/fvD4At5asZUTB+mG7CLSNh228Lv7lEQESQZnj+nFr19cyj1vrlThF5E2K5aunhlNve7ut7RenPYtIy2VSyYU8esXl7J4w06O6dkp6EgiIoeI5QKuYuDbQO/ocgUwjEg/v+YjrufiE/rSMT2Ve99cFXQUEZEGxVL4uwFj3f0ad78GGAcURi/m0lz99XTJ6sCXiwt55sN1bNpZEXQcEZFDxFL4+wL76zzfDxTFJU2S+PqJ/aiucU3jICJtUiyF/2HgPTP7mZn9DJgNPBjXVO3c0fnZnDaiJ4+++yk7KyqDjiMicpBYLuD6JXApsB3YBlzq7v8T72Dt3bcnD6B8XxUPv/Np0FFERA7SaOE3sywzSwdw93nAC0Rm6eyXoGzt2ojenZk0uID731pFRaXu0CUibUdTR/wvEO3LN7OBwDtAf+C7Zvar+Edr/74zeQBbdu3nr3PWBh1FROSApgp/V3dfHn08HXjM3b8HnAacHvdkSeD4fnkUH92VP7y+ksrqmqDjiIgATRf+uvcSPAl4GcDd9wOqYjEwM74zZQDryvbyzHxN3iYibUNThX+Bmd0cvdXiQOAlADPrkohgyWLKkO4c07MTd85cQXWN7ssrIsFrqvB/E9hCpJ//8+6+J7p+GJFbJ0oMzIzvnTSQlaW7eXaBjvpFJHiNztXj7nuBQ07iuvvbwNvxDJVspg4/isE9crj91RWcOaoXKSmasllEghPLBVzSQikpxvdOGsSKzbt4fuGGoOOISMip8CfIF0b2ZGD3HG57ZTk16usXkQCp8CdIakqkr3/Zpl28sGhj0HFEJMQOW/jNbLCZ/dHMXjKzV2uXRIRLNmeM6kX/gmx+9y8d9YtIcGI54v8bkRuj/xi4ts4iRyg1xbjq5EEs3VTOsx+pr19EghFL4a9y97vc/T13n1u7xD1ZkjpzVC+G9Mjl1peXUaWreUUkALEU/n+Y2XfMrKeZ5dUucU+WpFJSjBmfH8zKLbt56oN1QccRkRA67D13iczTAwd37ziRCdukGT4/rAcje3fmd68s5+wxvemQpnPsIpI4sczH36+BRUW/BcyMaz4/mJLtezVzp4gkXCxH/JjZCCJTNWTWrnP3h+IVKgwmDS7guKKu3P7qcs4dW0jHDqlBRxKRkIhlOOcNwO3RZQpwE3BWS3ZqZl3M7HEzW2Jmi83sMy3ZXntkZlx76lA27dzH/W+vCjqOiIRILJ3L5wEnAxvd/VJgNJDRwv3+DnjB3YdGt7e4hdtrl47vl8fnjunOXTM/oWzP/sN/QESkFcRS+Pe6ew1QZWadgM204MRudBv/AdwHkfn93b2sudtr7649dSi79lXx+5mfBB1FREIilsI/JzoH/x+BuUQu5nqvBfvsD5QC95vZB2Z2r5ll13+TmV1uZnPMbE5paWkLdte2DTkql3PHFvLA26tZV7Y36DgiEgKxjOr5jruXufvdwCnA9GiXT3OlAWOBu9z9WGA3cH0D+73H3YvdvbigoKAFu2v7rj5lMAC/fXlZwElEJAxiOblrZjbNzH7q7quBMjM7vgX7LAFK3H129PnjRH4RhFbvLh2Z/pmjeWJeCR+v3xl0HBFJcrF09fwe+AxwUfR5OXBnc3fo7huBtWY2JLrqZODj5m4vWVw5ZRCdO6bz388vxl0TuIlI/MRS+E9w9+8CFQDuvh3o0ML9fg941MwWAGOA/27h9tq9zlnpfP+kQcxasYWZS5P3nIaIBC+Wwl9pZqlEpmnAzAqAFs0u5u7zo/33o9z9nOgvk9CbNv5oivKz+OXzizWBm4jETSyF/zbgKaC7mf0SmIWO0OOiQ1oK1592DCs27+Iv72sqBxGJj8NO2eDuj5rZXCJ98Qac4+6hvOAqEU4d3oPji/L47cvLOGtMLzplpgcdSUSSTKNH/PWmYN4MPAb8GdikaZnjx8z4yRnD2LZnP7e/sjzoOCKShJo64t9CZOhlVfS51XlN0zLH0cjCzpw/rpD731rNhcf3ZUBBTtCRRCSJNNXHfzuwHXiByJz8/TUtc+Jce+pQMtNT+cWzoR/pKiKtrNHC7+5XERlq+Tfgq8AHZnaTmfVLULZQK8jN4PsnD+S1paW8tmRz0HFEJIk0OarHI14DfgDcDVwKfC4RwQQumdCPft2y+fmzH7O/SsM7RaR1NHVyN9vMvmJmTwPPAznAWHf/Y8LShVyHtBR+esYwVm7ZzZ/e0pz9ItI6mjq5uxlYTmQ0zwoiJ3SPM7PjANz9yfjHkylDu/O5Y3pw2yvLOWt0L3p16Rh0JBFp55rq6vkb8AEwFDgDOLPOckb8o0mtG84cRnWN88vndPmEiLRco0f87n5JAnNIE/rkZfHdKQO55eVlXLi8lM8OSu5pqkUkvmKZskHagMv/oz9F+Vnc8PQi9lVVBx1HRNoxFf52IjM9lZ+dNZyVW3Zzz+srg44jIu1YLDdiOeTG6g2tk/ibPKQ7p4/qye2vrWDVlt1BxxGRdiqWI/53YlwnCXDDGcPISE3hx3//SDdsEZFmaWoc/1FmNg7oaGbHmtnY6DIZyEpUQDlY906Z/OC0oby1Yit/n78u6Dgi0g41NY7/VOASoBC4pc76cuC/4phJDuPi4/vyxNwSfv7sYiYP7k7X7JbeEE1EwqSpuXoedPcpwCXuPqXOcpYu3gpWSorxP18ayc69lfz8OU3iJiJHJpY+/lfM7BYzmxNdfmNmneOeTJp0TM9OfHvyAJ6ct46ZSzWJm4jELpbCfx+R7p0vR5edwP3xDCWxufKkgQwoyOZHTy1k176qw39ARITYCv8Ad7/B3VdGlxvRTVjahIy0VG46bxTrd+zl5heXBh1HRNqJWAr/XjM7sfaJmU0E9sYvkhyJcUfnMf0zRTz4zmreX70t6Dgi0g7EUvi/DdxpZqvN7FPgDuBb8Y0lR+LaU4dQ2LUj1/7tQ/bsV5ePiDTtsIXf3ee7+2hgFDDS3Y919wXxjyaxys5I49fnjWb11j3c9IK6fESkabFM2dDZzG4BXgVe1aietml8/3wumVDEA2+v5p1PtgYdR0TasFi6ev5EHEb1mFmqmX1gZs+2dFsScd3UoRTlZ3Ht4x9qlI+INCrIUT1XAbqzSCvq2CGVm88fzbqyvfziWV3YJSINC2RUj5kVAqcD97ZkO3Ko4qI8rpg0gL+8v5aXP94UdBwRaYOCGtVzK/ADoKaxN5jZ5bVXC5eWlrZwd+Fy9ecGM7xXJ65/YgGl5fuCjiMibcwRj+oBjot+bRYzOwPY7O5zD7Pfe9y92N2LCwp0q8Ej0SEthVsvGMOufVVc98QCTd8sIgdpalrmTmb2QzO7w8xOIXKC92vACiIneZtrInCWma0G/gKcZGaPtGB70oBBPXL54WlDeXXJZh6ZvSboOCLShjR1xP8wMAT4CPgm8BJwPnCOu5/d3B26+w/dvdDdi4ALgVfdfVpztyeNmz6hiEmDC/jFsx+zdGN50HFEpI1oqvD3d/dL3P0PwEVAMXCGu89PSDJpMTPj5vNHk5uZzpV/nsfe/bpJu4g0Xfgrax+4ezWwyt1b9bDR3We6+xmtuU05WEFuBr+9YDTLN+/i/2mIp4jQdOEfbWY7o0s5MKr2sZntTFRAabnPDirgikkDeOy9NTy3YEPQcUQkYI3eetHdUxMZROLrms8P5t2VW7nuiQUM69WJft2yg44kIgGJZRy/JIH01BTuvHgsaanGdx6dR0Wl+vtFwkqFP0R6d+nIb788hsUbdvKzZxYFHUdEAqLCHzJThnbnO5MjUzo8Prck6DgiEgAV/hCaccpgxvfP40dPfcTCdTuCjiMiCabCH0JpqSnc8ZWx5GV34FsPz2Xb7v1BRxKRBFLhD6luORncPW0cpbv28b3H5lFV3eh8eSKSZFT4Q2x0ny784pwRvLViK//7wpKg44hIgjQ6jl/C4cvFfVi0bgd/fHMVg3vkcn5xn6AjiUic6Yhf+MkZwzhxYDf+66mPeH/1tqDjiEicqfALaakp3PmVsfTpmsW3Hp7L2m17go4kInGkwi8AdM5K597pxVRV1/D1B95nx97Kw39IRNolFX45oH9BDnd/dRyrt+7mWw/PYV+VpnUQSUYq/HKQCQO6cdN5o3h35TZ+8Lhu2yiSjDSqRw7xxWMLWV9Wwa9fXEqvLh25burQoCOJSCtS4ZcGfWfyAEq27+WumZ+Qn92Byz7bP+hIItJKVPilQWbGL84ZQdme/fziucV0yerAeeMKg44lIq1AffzSqNQU49YLx3DiwG5c98QCXlq0MehIItIKVPilSRlpqfzhq+MY2bszV/75A15fVhp0JBFpIRV+OazsjDQeuPQ4BnbP4ZsPzWHW8i1BRxKRFlDhl5h0yerAo5edQP9u2Vz20Pu8/YmKv0h7pcIvMeua3YFHLjuBPl2z+PoD7/PmcnX7iLRHKvxyRLrlZPDnb46nKD+bbzwwh399vCnoSCJyhFT45YgV5Gbwl8vHc0zPXK54ZC7/+HB90JFE5AgkvPCbWR8ze83MFpvZIjO7KtEZpOW6ZEW6fcb27cr3//IBD72zOuhIIhKjII74q4Br3P0YYDzwXTMbFkAOaaHczHQe/PrxnDy0Bz99ehH/+8ISze0j0g4kvPC7+wZ3nxd9XA4sBnonOoe0jo4dUrl72li+ckJf7pr5CTP++qFm9RRp4wKdssHMioBjgdkNvHY5cDlA3759ExtMjkhaagq/PGcEPTtl8puXl7F22x7umjaOgtyMoKOJSAMCO7lrZjnAE8B/uvvO+q+7+z3uXuzuxQUFBYkPKEfEzPjeyYO4/aJjWbh+B2ffMYtF63cEHUtEGhBI4TezdCJF/1F3fzKIDBIfZ47uxeNXTMCB8+56hyfnlQQdSUTqCWJUjwH3AYvd/ZZE71/ib0Tvzjxz5YmMLOzMjL9+yHWPL6CiUv3+Im1FEEf8E4GvAieZ2fzo8oUAckgcFeRm8OfLTuDKKQP5vzlrOefOt1i6sTzoWCICWHsYfldcXOxz5swJOoY00+vLSpnxf/Mpr6jims8P5rLP9ic1xYKOJZL0zGyuuxfXX68rdyXuJg0u4MWr/4MpQwv4n38u4YI/vMMnpbuCjiUSWir8khDdcjK4e9o4bvnyaJZuKue0W9/kNy8tVd+/SABU+CVhzIwvjS3k1Wsmc/qontz+6gpO+e3rvLBwg674FUkgFX5JuILcDH57wRge++Z4MtNSueKReXzx928ze+XWoKOJhIIKvwTmMwPy+edVn+Wmc0excUcFF9zzLtPunc3bK7boLwCRONKoHmkT9u6v5qF3VnPvrFWUlu9jdGFnvn5iP6aOOIqMtNSg44m0S42N6lHhlzalorKaJ+et4543PmH11j3kZXfgvHGFnD+ukEE9coOOJ9KuqPBLu1JT48xasYXH3lvDyx9voqrGGdQ9h9NH9eTU4Ucx9KhcIheBi0hjVPil3Sot38c/F27guQUbeG/1NtwjJ4g/O6gbEwZ0Y2zfLvTrlq1fBCL1qPBLUti8s4KZy0p5c/kWZi0vZfueSgA6d0xnVGFnhvTIZfBRuQzsnkPfvCzyszvoF4KElgq/JJ2aGmdF6S4+WLOdD9aUsXD9DpZv2sW+qpoD78nqkErvLh3p3imD7rmZFORm0CUrnS4dO9C5YzrZGankZKSR1SGNzPQUMtNTyUxPpUNaCumpRnpKCimaXkLaqcYKf6A3YhFpiZQUY3CPXAb3yOWC4yI366mucdZu28MnpbtYu20Pa7btZV3ZHjaX7+O9Vdso3bWP/XV+McQiNcUii0W+phjRr4ZZ5LkZGLWP//2Lwuq8Fvn679ftwH8OXW8H1hv1/2AxswPvqd32gX3V2bHVWVd/24bV2/fB+6ubucH91nvOQe07dDs08Bmrt10O+T4dnKHJ/dRtY0PbqbOPg9rZUNvrfP/qbq/+97h2uwf/bBp/74H3NPAzrf0+/Ps9/3bKsB70ycs69AMtoMIvSSU1xSjqlk1Rt+xG31NRWU3ZnkrK9u5n974qdu+rZve+KiqqqqmorKGispqqamd/dQ37q2qornGqapzqmhqqa6DGneoap8YdJ/KXhzs4To1D7R/RjhP9Hx59779f48C1Cn7gP5HPHHhPdJsc8rnoO/3Axw5s/9+fq7P9etuu3e4h23Rwag79fJ1tUqcdB2/v4P15nXwHstV/3sB26ralfvsbavtB721kPw21lwa/Bwf/jNqK/gXZrV7420VXj5mVAp828+PdgC2tGKe9CGO7w9hmCGe7w9hmOPJ2H+3uh9zCsF0U/pYwszkN9XEluzC2O4xthnC2O4xthtZrt6ZsEBEJGRV+EZGQCUPhvyfoAAEJY7vD2GYIZ7vD2GZopXYnfR+/iIgcLAxH/CIiUocKv4hIyCR14TezqWa21MxWmNn1QeeJBzPrY2avmdliM1tkZldF1+eZ2ctmtjz6tWvQWVubmaWa2Qdm9mz0eRja3MXMHjezJdGf+WeSvd1mdnX03/ZCM3vMzDKTsc1m9icz22xmC+usa7SdZvbDaG1bamanHsm+krbwm1kqcCdwGjAMuMjMhgWbKi6qgGvc/RhgPPDdaDuvB15x90HAK9HnyeYqYHGd52Fo8++AF9x9KDCaSPuTtt1m1hv4PlDs7iOAVOBCkrPNDwBT661rsJ3R/49fCAyPfub30ZoXk6Qt/MDxwAp3X+nu+4G/AGcHnKnVufsGd58XfVxOpBD0JtLWB6NvexA4J5CAcWJmhcDpwL11Vid7mzsB/wHcB+Du+929jCRvN5GpZTqaWRqQBawnCdvs7m8A2+qtbqydZwN/cfd97r4KWEGk5sUkmQt/b2Btnecl0XVJy8yKgGOB2UAPd98AkV8OQPcAo8XDrcAPgLozriV7m/sDpcD90S6ue80smyRut7uvA24G1gAbgB3u/hJJ3OZ6Gmtni+pbMhf+hubSTdqxq2aWAzwB/Ke77ww6TzyZ2RnAZnefG3SWBEsDxgJ3ufuxwG6So4ujUdE+7bOBfkAvINvMpgWbqk1oUX1L5sJfAvSp87yQyJ+IScfM0okU/Ufd/cno6k1m1jP6ek9gc1D54mAicJaZrSbShXeSmT1CcrcZIv+mS9x9dvT540R+ESRzuz8HrHL3UnevBJ4EJpDcba6rsXa2qL4lc+F/HxhkZv3MrAOREyHPBJyp1VlkEu/7gMXufkudl54BpkcfTweeTnS2eHH3H7p7obsXEfm5vuru00jiNgO4+0ZgrZkNia46GfiY5G73GmC8mWVF/62fTOQ8VjK3ua7G2vkMcKGZZZhZP2AQ8F7MW3X3pF2ALwDLgE+AHwWdJ05tPJHIn3gLgPnR5QtAPpFRAMujX/OCzhqn9k8Gno0+Tvo2A2OAOdGf99+BrsnebuBGYAmwEHgYyEjGNgOPETmPUUnkiP4bTbUT+FG0ti0FTjuSfWnKBhGRkEnmrh4REWmACr+ISMio8IuIhIwKv4hIyKjwi4iEjAq/CGBm1WY2v87SalfEmllR3RkXRYKWFnQAkTZir7uPCTqESCLoiF+kCWa22sz+18zeiy4Do+uPNrNXzGxB9Gvf6PoeZvaUmX0YXSZEN5VqZn+Mziv/kpl1DKxREnoq/CIRHet19VxQ57Wd7n48cAeRWUGJPn7I3UcBjwK3RdffBrzu7qOJzKOzKLp+EHCnuw8HyoBz49oakSboyl0RwMx2uXtOA+tXAye5+8roZHgb3T3fzLYAPd29Mrp+g7t3M7NSoNDd99XZRhHwskdupoGZXQeku/svEtA0kUPoiF/k8LyRx429pyH76jyuRufXJEAq/CKHd0Gdr+9EH79NZGZQgIuBWdHHrwDfhgP3BO6UqJAisdJRh0hERzObX+f5C+5eO6Qzw8xmEzlQuii67vvAn8zsWiJ3xbo0uv4q4B4z+waRI/tvE5lxUaTNUB+/SBOiffzF7r4l6CwirUVdPSIiIaMjfhGRkNERv4hIyKjwi4iEjAq/iEjIqPCLiISMCr+ISMj8f59SteheqY8ZAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "learning_rate=0.05\n",
    "epochs=100\n",
    "my_batch_size=6  # Replace ? with an integer.\n",
    "\n",
    "my_model = build_model(learning_rate)\n",
    "trained_weight, trained_bias, epochs, rmse = train_model(my_model, my_feature, \n",
    "                                                        my_label, epochs,\n",
    "                                                        my_batch_size)\n",
    "plot_the_model(trained_weight, trained_bias, my_feature, my_label)\n",
    "plot_the_loss_curve(epochs, rmse)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "cellView": "form",
    "colab": {},
    "colab_type": "code",
    "id": "-mtVpoBrANAm"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 768.2532 - root_mean_squared_error: 27.7174\n",
      "Epoch 2/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 697.6463 - root_mean_squared_error: 26.4130\n",
      "Epoch 3/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 649.6736 - root_mean_squared_error: 25.4887\n",
      "Epoch 4/125\n",
      "1/1 [==============================] - 0s 847us/step - loss: 611.3787 - root_mean_squared_error: 24.7261\n",
      "Epoch 5/125\n",
      "1/1 [==============================] - 0s 971us/step - loss: 578.7313 - root_mean_squared_error: 24.0568\n",
      "Epoch 6/125\n",
      "1/1 [==============================] - 0s 765us/step - loss: 549.8802 - root_mean_squared_error: 23.4495\n",
      "Epoch 7/125\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 523.8010 - root_mean_squared_error: 22.8867\n",
      "Epoch 8/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 499.8600 - root_mean_squared_error: 22.3575\n",
      "Epoch 9/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 477.6344 - root_mean_squared_error: 21.8548\n",
      "Epoch 10/125\n",
      "1/1 [==============================] - 0s 869us/step - loss: 456.8272 - root_mean_squared_error: 21.3735\n",
      "Epoch 11/125\n",
      "1/1 [==============================] - 0s 755us/step - loss: 437.2206 - root_mean_squared_error: 20.9098\n",
      "Epoch 12/125\n",
      "1/1 [==============================] - 0s 944us/step - loss: 418.6501 - root_mean_squared_error: 20.4609\n",
      "Epoch 13/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 400.9880 - root_mean_squared_error: 20.0247\n",
      "Epoch 14/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 384.1337 - root_mean_squared_error: 19.5993\n",
      "Epoch 15/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 368.0058 - root_mean_squared_error: 19.1835\n",
      "Epoch 16/125\n",
      "1/1 [==============================] - 0s 926us/step - loss: 352.5381 - root_mean_squared_error: 18.7760\n",
      "Epoch 17/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 337.6756 - root_mean_squared_error: 18.3759\n",
      "Epoch 18/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 323.3727 - root_mean_squared_error: 17.9826\n",
      "Epoch 19/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 309.5905 - root_mean_squared_error: 17.5952\n",
      "Epoch 20/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 296.2964 - root_mean_squared_error: 17.2133\n",
      "Epoch 21/125\n",
      "1/1 [==============================] - 0s 898us/step - loss: 283.4619 - root_mean_squared_error: 16.8363\n",
      "Epoch 22/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 271.0628 - root_mean_squared_error: 16.4640\n",
      "Epoch 23/125\n",
      "1/1 [==============================] - 0s 960us/step - loss: 259.0779 - root_mean_squared_error: 16.0959\n",
      "Epoch 24/125\n",
      "1/1 [==============================] - 0s 833us/step - loss: 247.4887 - root_mean_squared_error: 15.7318\n",
      "Epoch 25/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 236.2790 - root_mean_squared_error: 15.3714\n",
      "Epoch 26/125\n",
      "1/1 [==============================] - 0s 997us/step - loss: 225.4343 - root_mean_squared_error: 15.0145\n",
      "Epoch 27/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 214.9419 - root_mean_squared_error: 14.6609\n",
      "Epoch 28/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 204.7903 - root_mean_squared_error: 14.3105\n",
      "Epoch 29/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 194.9694 - root_mean_squared_error: 13.9631\n",
      "Epoch 30/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 185.4698 - root_mean_squared_error: 13.6187\n",
      "Epoch 31/125\n",
      "1/1 [==============================] - 0s 767us/step - loss: 176.2831 - root_mean_squared_error: 13.2772\n",
      "Epoch 32/125\n",
      "1/1 [==============================] - 0s 762us/step - loss: 167.4016 - root_mean_squared_error: 12.9384\n",
      "Epoch 33/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 158.8183 - root_mean_squared_error: 12.6023\n",
      "Epoch 34/125\n",
      "1/1 [==============================] - 0s 3ms/step - loss: 150.5268 - root_mean_squared_error: 12.2689\n",
      "Epoch 35/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 142.5209 - root_mean_squared_error: 11.9382\n",
      "Epoch 36/125\n",
      "1/1 [==============================] - 0s 821us/step - loss: 134.7950 - root_mean_squared_error: 11.6101\n",
      "Epoch 37/125\n",
      "1/1 [==============================] - 0s 814us/step - loss: 127.3438 - root_mean_squared_error: 11.2847\n",
      "Epoch 38/125\n",
      "1/1 [==============================] - 0s 781us/step - loss: 120.1623 - root_mean_squared_error: 10.9619\n",
      "Epoch 39/125\n",
      "1/1 [==============================] - 0s 908us/step - loss: 113.2457 - root_mean_squared_error: 10.6417\n",
      "Epoch 40/125\n",
      "1/1 [==============================] - 0s 787us/step - loss: 106.5893 - root_mean_squared_error: 10.3242\n",
      "Epoch 41/125\n",
      "1/1 [==============================] - 0s 831us/step - loss: 100.1889 - root_mean_squared_error: 10.0094\n",
      "Epoch 42/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 94.0401 - root_mean_squared_error: 9.6974\n",
      "Epoch 43/125\n",
      "1/1 [==============================] - 0s 774us/step - loss: 88.1386 - root_mean_squared_error: 9.3882\n",
      "Epoch 44/125\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 82.4806 - root_mean_squared_error: 9.0819\n",
      "Epoch 45/125\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 77.0617 - root_mean_squared_error: 8.7785\n",
      "Epoch 46/125\n",
      "1/1 [==============================] - 0s 914us/step - loss: 71.8780 - root_mean_squared_error: 8.4781\n",
      "Epoch 47/125\n",
      "1/1 [==============================] - 0s 792us/step - loss: 66.9256 - root_mean_squared_error: 8.1808\n",
      "Epoch 48/125\n",
      "1/1 [==============================] - 0s 851us/step - loss: 62.2003 - root_mean_squared_error: 7.8867\n",
      "Epoch 49/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 57.6982 - root_mean_squared_error: 7.5959\n",
      "Epoch 50/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 53.4150 - root_mean_squared_error: 7.3086\n",
      "Epoch 51/125\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 49.3468 - root_mean_squared_error: 7.0247\n",
      "Epoch 52/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 45.4891 - root_mean_squared_error: 6.7446\n",
      "Epoch 53/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 41.8378 - root_mean_squared_error: 6.4682\n",
      "Epoch 54/125\n",
      "1/1 [==============================] - 0s 858us/step - loss: 38.3885 - root_mean_squared_error: 6.1958\n",
      "Epoch 55/125\n",
      "1/1 [==============================] - 0s 879us/step - loss: 35.1365 - root_mean_squared_error: 5.9276\n",
      "Epoch 56/125\n",
      "1/1 [==============================] - 0s 943us/step - loss: 32.0773 - root_mean_squared_error: 5.6637\n",
      "Epoch 57/125\n",
      "1/1 [==============================] - 0s 915us/step - loss: 29.2061 - root_mean_squared_error: 5.4043\n",
      "Epoch 58/125\n",
      "1/1 [==============================] - 0s 795us/step - loss: 26.5180 - root_mean_squared_error: 5.1496\n",
      "Epoch 59/125\n",
      "1/1 [==============================] - 0s 859us/step - loss: 24.0080 - root_mean_squared_error: 4.8998\n",
      "Epoch 60/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 21.6707 - root_mean_squared_error: 4.6552\n",
      "Epoch 61/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 19.5010 - root_mean_squared_error: 4.4160\n",
      "Epoch 62/125\n",
      "1/1 [==============================] - 0s 668us/step - loss: 17.4931 - root_mean_squared_error: 4.1825\n",
      "Epoch 63/125\n",
      "1/1 [==============================] - 0s 735us/step - loss: 15.6414 - root_mean_squared_error: 3.9549\n",
      "Epoch 64/125\n",
      "1/1 [==============================] - 0s 698us/step - loss: 13.9401 - root_mean_squared_error: 3.7336\n",
      "Epoch 65/125\n",
      "1/1 [==============================] - 0s 655us/step - loss: 12.3830 - root_mean_squared_error: 3.5190\n",
      "Epoch 66/125\n",
      "1/1 [==============================] - 0s 657us/step - loss: 10.9640 - root_mean_squared_error: 3.3112\n",
      "Epoch 67/125\n",
      "1/1 [==============================] - 0s 656us/step - loss: 9.6766 - root_mean_squared_error: 3.1107\n",
      "Epoch 68/125\n",
      "1/1 [==============================] - 0s 691us/step - loss: 8.5143 - root_mean_squared_error: 2.9179\n",
      "Epoch 69/125\n",
      "1/1 [==============================] - 0s 933us/step - loss: 7.4704 - root_mean_squared_error: 2.7332\n",
      "Epoch 70/125\n",
      "1/1 [==============================] - 0s 739us/step - loss: 6.5382 - root_mean_squared_error: 2.5570\n",
      "Epoch 71/125\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 651us/step - loss: 5.7106 - root_mean_squared_error: 2.3897\n",
      "Epoch 72/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 4.9808 - root_mean_squared_error: 2.2318\n",
      "Epoch 73/125\n",
      "1/1 [==============================] - 0s 689us/step - loss: 4.3416 - root_mean_squared_error: 2.0836\n",
      "Epoch 74/125\n",
      "1/1 [==============================] - 0s 737us/step - loss: 3.7860 - root_mean_squared_error: 1.9458\n",
      "Epoch 75/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 3.3069 - root_mean_squared_error: 1.8185\n",
      "Epoch 76/125\n",
      "1/1 [==============================] - 0s 676us/step - loss: 2.8974 - root_mean_squared_error: 1.7022\n",
      "Epoch 77/125\n",
      "1/1 [==============================] - 0s 886us/step - loss: 2.5506 - root_mean_squared_error: 1.5971\n",
      "Epoch 78/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 2.2598 - root_mean_squared_error: 1.5033\n",
      "Epoch 79/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 2.0186 - root_mean_squared_error: 1.4208\n",
      "Epoch 80/125\n",
      "1/1 [==============================] - 0s 867us/step - loss: 1.8208 - root_mean_squared_error: 1.3494\n",
      "Epoch 81/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.6606 - root_mean_squared_error: 1.2886\n",
      "Epoch 82/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.5324 - root_mean_squared_error: 1.2379\n",
      "Epoch 83/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.4312 - root_mean_squared_error: 1.1963\n",
      "Epoch 84/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.3524 - root_mean_squared_error: 1.1629\n",
      "Epoch 85/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.2919 - root_mean_squared_error: 1.1366\n",
      "Epoch 86/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.2461 - root_mean_squared_error: 1.1163\n",
      "Epoch 87/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.2119 - root_mean_squared_error: 1.1008\n",
      "Epoch 88/125\n",
      "1/1 [==============================] - 0s 816us/step - loss: 1.1865 - root_mean_squared_error: 1.0893\n",
      "Epoch 89/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.1678 - root_mean_squared_error: 1.0806\n",
      "Epoch 90/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.1540 - root_mean_squared_error: 1.0742\n",
      "Epoch 91/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.1437 - root_mean_squared_error: 1.0694\n",
      "Epoch 92/125\n",
      "1/1 [==============================] - 0s 951us/step - loss: 1.1358 - root_mean_squared_error: 1.0658\n",
      "Epoch 93/125\n",
      "1/1 [==============================] - 0s 988us/step - loss: 1.1296 - root_mean_squared_error: 1.0628\n",
      "Epoch 94/125\n",
      "1/1 [==============================] - 0s 868us/step - loss: 1.1243 - root_mean_squared_error: 1.0603\n",
      "Epoch 95/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.1197 - root_mean_squared_error: 1.0582\n",
      "Epoch 96/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.1154 - root_mean_squared_error: 1.0561\n",
      "Epoch 97/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.1112 - root_mean_squared_error: 1.0541\n",
      "Epoch 98/125\n",
      "1/1 [==============================] - 0s 734us/step - loss: 1.1070 - root_mean_squared_error: 1.0521\n",
      "Epoch 99/125\n",
      "1/1 [==============================] - 0s 823us/step - loss: 1.1028 - root_mean_squared_error: 1.0501\n",
      "Epoch 100/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.0984 - root_mean_squared_error: 1.0481\n",
      "Epoch 101/125\n",
      "1/1 [==============================] - 0s 2ms/step - loss: 1.0940 - root_mean_squared_error: 1.0459\n",
      "Epoch 102/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.0894 - root_mean_squared_error: 1.0438\n",
      "Epoch 103/125\n",
      "1/1 [==============================] - 0s 746us/step - loss: 1.0847 - root_mean_squared_error: 1.0415\n",
      "Epoch 104/125\n",
      "1/1 [==============================] - 0s 961us/step - loss: 1.0799 - root_mean_squared_error: 1.0392\n",
      "Epoch 105/125\n",
      "1/1 [==============================] - 0s 711us/step - loss: 1.0749 - root_mean_squared_error: 1.0368\n",
      "Epoch 106/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.0699 - root_mean_squared_error: 1.0343\n",
      "Epoch 107/125\n",
      "1/1 [==============================] - 0s 697us/step - loss: 1.0647 - root_mean_squared_error: 1.0318\n",
      "Epoch 108/125\n",
      "1/1 [==============================] - 0s 729us/step - loss: 1.0593 - root_mean_squared_error: 1.0292\n",
      "Epoch 109/125\n",
      "1/1 [==============================] - 0s 704us/step - loss: 1.0539 - root_mean_squared_error: 1.0266\n",
      "Epoch 110/125\n",
      "1/1 [==============================] - 0s 863us/step - loss: 1.0484 - root_mean_squared_error: 1.0239\n",
      "Epoch 111/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.0427 - root_mean_squared_error: 1.0211\n",
      "Epoch 112/125\n",
      "1/1 [==============================] - 0s 686us/step - loss: 1.0370 - root_mean_squared_error: 1.0183\n",
      "Epoch 113/125\n",
      "1/1 [==============================] - 0s 648us/step - loss: 1.0312 - root_mean_squared_error: 1.0155\n",
      "Epoch 114/125\n",
      "1/1 [==============================] - 0s 689us/step - loss: 1.0254 - root_mean_squared_error: 1.0126\n",
      "Epoch 115/125\n",
      "1/1 [==============================] - 0s 643us/step - loss: 1.0194 - root_mean_squared_error: 1.0097\n",
      "Epoch 116/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 1.0135 - root_mean_squared_error: 1.0067\n",
      "Epoch 117/125\n",
      "1/1 [==============================] - 0s 647us/step - loss: 1.0075 - root_mean_squared_error: 1.0037\n",
      "Epoch 118/125\n",
      "1/1 [==============================] - 0s 775us/step - loss: 1.0015 - root_mean_squared_error: 1.0008\n",
      "Epoch 119/125\n",
      "1/1 [==============================] - 0s 818us/step - loss: 0.9955 - root_mean_squared_error: 0.9978\n",
      "Epoch 120/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.9895 - root_mean_squared_error: 0.9947\n",
      "Epoch 121/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.9836 - root_mean_squared_error: 0.9918\n",
      "Epoch 122/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.9777 - root_mean_squared_error: 0.9888\n",
      "Epoch 123/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.9718 - root_mean_squared_error: 0.9858\n",
      "Epoch 124/125\n",
      "1/1 [==============================] - 0s 1ms/step - loss: 0.9661 - root_mean_squared_error: 0.9829\n",
      "Epoch 125/125\n",
      "1/1 [==============================] - 0s 730us/step - loss: 0.9604 - root_mean_squared_error: 0.9800\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/eric/.conda/envs/mlbase/lib/python3.8/site-packages/numpy/core/_asarray.py:136: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray\n",
      "  return array(a, dtype, copy=False, order=order, subok=True)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#@title Double-click to view a possible solution\n",
    "\n",
    "learning_rate=0.05\n",
    "epochs=125\n",
    "my_batch_size=1 # Wow, a batch size of 1 works!\n",
    "\n",
    "my_model = build_model(learning_rate)\n",
    "trained_weight, trained_bias, epochs, rmse = train_model(my_model, my_feature, \n",
    "                                                         my_label, epochs,\n",
    "                                                         my_batch_size)\n",
    "plot_the_model(trained_weight, trained_bias, my_feature, my_label)\n",
    "plot_the_loss_curve(epochs, rmse)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "aS3q7TIF9SFL"
   },
   "source": [
    "## Summary of hyperparameter tuning\n",
    "\n",
    "Most machine learning problems require a lot of hyperparameter tuning.  Unfortunately, we can't provide concrete tuning rules for every model. Lowering the learning rate can help one model converge efficiently but make another model converge much too slowly.  You must experiment to find the best set of hyperparameters for your dataset. That said, here are a few rules of thumb:\n",
    "\n",
    " * Training loss should steadily decrease, steeply at first, and then more slowly until the slope of the curve reaches or approaches zero. \n",
    " * If the training loss does not converge, train for more epochs.\n",
    " * If the training loss decreases too slowly, increase the learning rate. Note that setting the training loss too high may also prevent training loss from converging.\n",
    " * If the training loss varies wildly (that is, the training loss jumps around), decrease the learning rate.\n",
    " * Lowering the learning rate while increasing the number of epochs or the batch size is often a good combination.\n",
    " * Setting the batch size to a *very* small batch number can also cause instability. First, try large batch size values. Then, decrease the batch size until you see degradation.\n",
    " * For real-world datasets consisting of a very large number of examples, the entire dataset might not fit into memory. In such cases, you'll need to reduce the batch size to enable a batch to fit into memory. \n",
    "\n",
    "Remember: the ideal combination of hyperparameters is data dependent, so you must always experiment and verify."
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "Linear Regression with Synthetic Data.ipynb",
   "private_outputs": true,
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}