C1_W1_Lab_2_multi-output.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Ungraded Lab: Build a Multi-output Model\n",
    "\n",
    "In this lab, we'll show how you can build models with more than one output. The dataset we will be working on is available from the [UCI Machine Learning Repository](https://archive.ics.uci.edu/ml/datasets/Energy+efficiency). It is an Energy Efficiency dataset which uses the bulding features (e.g. wall area, roof area) as inputs and has two outputs: Cooling Load and Heating Load. Let's see how we can build a model to train on this data."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "0p84I7yFHRT2"
   },
   "source": [
    "## Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "try:\n",
    "  # %tensorflow_version only exists in Colab.\n",
    "  %tensorflow_version 2.x\n",
    "except Exception:\n",
    "  pass\n",
    "\n",
    "import tensorflow as tf\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "from tensorflow.keras.models import Model\n",
    "from tensorflow.keras.layers import Dense, Input\n",
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Utilities\n",
    "\n",
    "We define a few utilities for data conversion and visualization to make our code more neat."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "04Y-C9RYUTes"
   },
   "outputs": [],
   "source": [
    "def format_output(data):\n",
    "    y1 = data.pop('Y1')\n",
    "    y1 = np.array(y1)\n",
    "    y2 = data.pop('Y2')\n",
    "    y2 = np.array(y2)\n",
    "    return y1, y2\n",
    "\n",
    "\n",
    "def norm(x):\n",
    "    return (x - train_stats['mean']) / train_stats['std']\n",
    "\n",
    "\n",
    "def plot_diff(y_true, y_pred, title=''):\n",
    "    plt.scatter(y_true, y_pred)\n",
    "    plt.title(title)\n",
    "    plt.xlabel('True Values')\n",
    "    plt.ylabel('Predictions')\n",
    "    plt.axis('equal')\n",
    "    plt.axis('square')\n",
    "    plt.xlim(plt.xlim())\n",
    "    plt.ylim(plt.ylim())\n",
    "    plt.plot([-100, 100], [-100, 100])\n",
    "    plt.show()\n",
    "\n",
    "\n",
    "def plot_metrics(metric_name, title, ylim=5):\n",
    "    plt.title(title)\n",
    "    plt.ylim(0, ylim)\n",
    "    plt.plot(history.history[metric_name], color='blue', label=metric_name)\n",
    "    plt.plot(history.history['val_' + metric_name], color='green', label='val_' + metric_name)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Prepare the Data\n",
    "\n",
    "We download the dataset and format it for training."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Specify data URI\n",
    "URI = './data/ENB2012_data.xlsx'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Specify data URI\n",
    "URI = './data/ENB2012_data.xlsx'\n",
    "\n",
    "# Use pandas excel reader\n",
    "df = pd.read_excel(URI)\n",
    "df = df.sample(frac=1).reset_index(drop=True)\n",
    "\n",
    "# Split the data into train and test with 80 train / 20 test\n",
    "train, test = train_test_split(df, test_size=0.2)\n",
    "train_stats = train.describe()\n",
    "\n",
    "# Get Y1 and Y2 as the 2 outputs and format them as np arrays\n",
    "train_stats.pop('Y1')\n",
    "train_stats.pop('Y2')\n",
    "train_stats = train_stats.transpose()\n",
    "train_Y = format_output(train)\n",
    "test_Y = format_output(test)\n",
    "\n",
    "# Normalize the training and test data\n",
    "norm_train_X = norm(train)\n",
    "norm_test_X = norm(test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Build the Model\n",
    "\n",
    "Here is how we'll build the model using the functional syntax. Notice that we can specify a list of outputs (i.e. `[y1_output, y2_output]`) when we instantiate the `Model()` class."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model\"\n",
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "input_1 (InputLayer)            [(None, 8)]          0                                            \n",
      "__________________________________________________________________________________________________\n",
      "dense (Dense)                   (None, 128)          1152        input_1[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "dense_1 (Dense)                 (None, 128)          16512       dense[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "dense_2 (Dense)                 (None, 64)           8256        dense_1[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "y1_output (Dense)               (None, 1)            129         dense_1[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "y2_output (Dense)               (None, 1)            65          dense_2[0][0]                    \n",
      "==================================================================================================\n",
      "Total params: 26,114\n",
      "Trainable params: 26,114\n",
      "Non-trainable params: 0\n",
      "__________________________________________________________________________________________________\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "# Define model layers.\n",
    "input_layer = Input(shape=(len(train .columns),))\n",
    "first_dense = Dense(units='128', activation='relu')(input_layer)\n",
    "second_dense = Dense(units='128', activation='relu')(first_dense)\n",
    "\n",
    "# Y1 output will be fed directly from the second dense\n",
    "y1_output = Dense(units='1', name='y1_output')(second_dense)\n",
    "third_dense = Dense(units='64', activation='relu')(second_dense)\n",
    "\n",
    "# Y2 output will come via the third dense\n",
    "y2_output = Dense(units='1', name='y2_output')(third_dense)\n",
    "\n",
    "# Define the model with the input layer and a list of output layers\n",
    "model = Model(inputs=input_layer, outputs=[y1_output, y2_output])\n",
    "\n",
    "print(model.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Configure parameters\n",
    "\n",
    "We specify the optimizer as well as the loss and metrics for each output."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Specify the optimizer, and compile the model with loss functions for both outputs\n",
    "optimizer = tf.keras.optimizers.SGD(lr=0.001)\n",
    "model.compile(optimizer=optimizer,\n",
    "              loss={'y1_output': 'mse', 'y2_output': 'mse'},\n",
    "              metrics={'y1_output': tf.keras.metrics.RootMeanSquaredError(),\n",
    "                       'y2_output': tf.keras.metrics.RootMeanSquaredError()})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train the Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 614 samples, validate on 154 samples\n",
      "Epoch 1/500\n",
      "614/614 [==============================] - 1s 927us/sample - loss: 245.0307 - y1_output_loss: 114.7288 - y2_output_loss: 129.1645 - y1_output_root_mean_squared_error: 10.7331 - y2_output_root_mean_squared_error: 11.3943 - val_loss: 80.3185 - val_y1_output_loss: 26.8391 - val_y2_output_loss: 53.9628 - val_y1_output_root_mean_squared_error: 5.1865 - val_y2_output_root_mean_squared_error: 7.3088\n",
      "Epoch 2/500\n",
      "614/614 [==============================] - 0s 142us/sample - loss: 40.4920 - y1_output_loss: 14.0763 - y2_output_loss: 26.0914 - y1_output_root_mean_squared_error: 3.7664 - y2_output_root_mean_squared_error: 5.1290 - val_loss: 22.9740 - val_y1_output_loss: 9.9099 - val_y2_output_loss: 12.7589 - val_y1_output_root_mean_squared_error: 3.1688 - val_y2_output_root_mean_squared_error: 3.5962\n",
      "Epoch 3/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 27.5799 - y1_output_loss: 10.9163 - y2_output_loss: 16.8689 - y1_output_root_mean_squared_error: 3.2891 - y2_output_root_mean_squared_error: 4.0941 - val_loss: 25.7272 - val_y1_output_loss: 10.4786 - val_y2_output_loss: 15.1573 - val_y1_output_root_mean_squared_error: 3.2345 - val_y2_output_root_mean_squared_error: 3.9071\n",
      "Epoch 4/500\n",
      "614/614 [==============================] - 0s 131us/sample - loss: 23.3437 - y1_output_loss: 9.4451 - y2_output_loss: 13.9789 - y1_output_root_mean_squared_error: 3.0754 - y2_output_root_mean_squared_error: 3.7264 - val_loss: 29.3727 - val_y1_output_loss: 9.2583 - val_y2_output_loss: 20.0865 - val_y1_output_root_mean_squared_error: 3.0339 - val_y2_output_root_mean_squared_error: 4.4909\n",
      "Epoch 5/500\n",
      "614/614 [==============================] - 0s 132us/sample - loss: 23.5511 - y1_output_loss: 9.6549 - y2_output_loss: 14.2710 - y1_output_root_mean_squared_error: 3.0837 - y2_output_root_mean_squared_error: 3.7473 - val_loss: 24.4943 - val_y1_output_loss: 9.9599 - val_y2_output_loss: 15.5626 - val_y1_output_root_mean_squared_error: 3.1015 - val_y2_output_root_mean_squared_error: 3.8568\n",
      "Epoch 6/500\n",
      "614/614 [==============================] - 0s 131us/sample - loss: 22.6580 - y1_output_loss: 9.0400 - y2_output_loss: 13.7343 - y1_output_root_mean_squared_error: 2.9846 - y2_output_root_mean_squared_error: 3.7081 - val_loss: 20.8315 - val_y1_output_loss: 9.4298 - val_y2_output_loss: 12.1724 - val_y1_output_root_mean_squared_error: 3.0174 - val_y2_output_root_mean_squared_error: 3.4245\n",
      "Epoch 7/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 22.9794 - y1_output_loss: 8.8813 - y2_output_loss: 14.0844 - y1_output_root_mean_squared_error: 2.9914 - y2_output_root_mean_squared_error: 3.7458 - val_loss: 38.2626 - val_y1_output_loss: 10.9080 - val_y2_output_loss: 29.8957 - val_y1_output_root_mean_squared_error: 3.2043 - val_y2_output_root_mean_squared_error: 5.2911\n",
      "Epoch 8/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 23.0799 - y1_output_loss: 9.0092 - y2_output_loss: 14.3381 - y1_output_root_mean_squared_error: 2.9928 - y2_output_root_mean_squared_error: 3.7581 - val_loss: 23.0669 - val_y1_output_loss: 9.8428 - val_y2_output_loss: 14.0960 - val_y1_output_root_mean_squared_error: 3.0888 - val_y2_output_root_mean_squared_error: 3.6778\n",
      "Epoch 9/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 19.2934 - y1_output_loss: 7.9721 - y2_output_loss: 11.3639 - y1_output_root_mean_squared_error: 2.8053 - y2_output_root_mean_squared_error: 3.3799 - val_loss: 34.6399 - val_y1_output_loss: 15.3835 - val_y2_output_loss: 21.6598 - val_y1_output_root_mean_squared_error: 3.7973 - val_y2_output_root_mean_squared_error: 4.4967\n",
      "Epoch 10/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 20.6979 - y1_output_loss: 8.0859 - y2_output_loss: 12.7396 - y1_output_root_mean_squared_error: 2.8520 - y2_output_root_mean_squared_error: 3.5446 - val_loss: 35.1934 - val_y1_output_loss: 9.5766 - val_y2_output_loss: 25.6108 - val_y1_output_root_mean_squared_error: 3.0844 - val_y2_output_root_mean_squared_error: 5.0676\n",
      "Epoch 11/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 18.7483 - y1_output_loss: 7.4886 - y2_output_loss: 11.2805 - y1_output_root_mean_squared_error: 2.7343 - y2_output_root_mean_squared_error: 3.3574 - val_loss: 24.2515 - val_y1_output_loss: 9.3513 - val_y2_output_loss: 15.2433 - val_y1_output_root_mean_squared_error: 3.0325 - val_y2_output_root_mean_squared_error: 3.8801\n",
      "Epoch 12/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 20.8915 - y1_output_loss: 8.1305 - y2_output_loss: 12.5997 - y1_output_root_mean_squared_error: 2.8636 - y2_output_root_mean_squared_error: 3.5625 - val_loss: 19.2516 - val_y1_output_loss: 8.0731 - val_y2_output_loss: 11.5788 - val_y1_output_root_mean_squared_error: 2.8076 - val_y2_output_root_mean_squared_error: 3.3718\n",
      "Epoch 13/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 22.8436 - y1_output_loss: 8.2142 - y2_output_loss: 14.5629 - y1_output_root_mean_squared_error: 2.8757 - y2_output_root_mean_squared_error: 3.8176 - val_loss: 44.4728 - val_y1_output_loss: 11.0373 - val_y2_output_loss: 33.9795 - val_y1_output_root_mean_squared_error: 3.2883 - val_y2_output_root_mean_squared_error: 5.8017\n",
      "Epoch 14/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 17.5372 - y1_output_loss: 7.1608 - y2_output_loss: 10.2856 - y1_output_root_mean_squared_error: 2.6853 - y2_output_root_mean_squared_error: 3.2135 - val_loss: 18.3280 - val_y1_output_loss: 8.9250 - val_y2_output_loss: 10.4289 - val_y1_output_root_mean_squared_error: 2.8830 - val_y2_output_root_mean_squared_error: 3.1648\n",
      "Epoch 15/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 21.6316 - y1_output_loss: 7.9620 - y2_output_loss: 14.1365 - y1_output_root_mean_squared_error: 2.7958 - y2_output_root_mean_squared_error: 3.7169 - val_loss: 44.8525 - val_y1_output_loss: 13.6395 - val_y2_output_loss: 31.1068 - val_y1_output_root_mean_squared_error: 3.7077 - val_y2_output_root_mean_squared_error: 5.5772\n",
      "Epoch 16/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 20.0466 - y1_output_loss: 7.4083 - y2_output_loss: 12.5582 - y1_output_root_mean_squared_error: 2.7315 - y2_output_root_mean_squared_error: 3.5476 - val_loss: 16.6403 - val_y1_output_loss: 7.4238 - val_y2_output_loss: 9.9617 - val_y1_output_root_mean_squared_error: 2.6592 - val_y2_output_root_mean_squared_error: 3.0934\n",
      "Epoch 17/500\n",
      "614/614 [==============================] - 0s 132us/sample - loss: 23.2965 - y1_output_loss: 8.0057 - y2_output_loss: 15.2355 - y1_output_root_mean_squared_error: 2.8346 - y2_output_root_mean_squared_error: 3.9066 - val_loss: 19.1912 - val_y1_output_loss: 9.0187 - val_y2_output_loss: 11.2861 - val_y1_output_root_mean_squared_error: 2.9203 - val_y2_output_root_mean_squared_error: 3.2654\n",
      "Epoch 18/500\n",
      "614/614 [==============================] - 0s 133us/sample - loss: 17.8616 - y1_output_loss: 6.7233 - y2_output_loss: 11.1433 - y1_output_root_mean_squared_error: 2.5927 - y2_output_root_mean_squared_error: 3.3376 - val_loss: 17.2703 - val_y1_output_loss: 7.8584 - val_y2_output_loss: 10.6361 - val_y1_output_root_mean_squared_error: 2.7115 - val_y2_output_root_mean_squared_error: 3.1493\n",
      "Epoch 19/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 17.7483 - y1_output_loss: 6.7083 - y2_output_loss: 10.9353 - y1_output_root_mean_squared_error: 2.6020 - y2_output_root_mean_squared_error: 3.3133 - val_loss: 15.8051 - val_y1_output_loss: 7.1760 - val_y2_output_loss: 9.6501 - val_y1_output_root_mean_squared_error: 2.5796 - val_y2_output_root_mean_squared_error: 3.0250\n",
      "Epoch 20/500\n",
      "614/614 [==============================] - 0s 131us/sample - loss: 16.1098 - y1_output_loss: 6.2324 - y2_output_loss: 9.7425 - y1_output_root_mean_squared_error: 2.5074 - y2_output_root_mean_squared_error: 3.1341 - val_loss: 19.9291 - val_y1_output_loss: 8.7742 - val_y2_output_loss: 13.1469 - val_y1_output_root_mean_squared_error: 2.8293 - val_y2_output_root_mean_squared_error: 3.4532\n",
      "Epoch 21/500\n",
      "614/614 [==============================] - 0s 133us/sample - loss: 16.0469 - y1_output_loss: 6.0889 - y2_output_loss: 10.1142 - y1_output_root_mean_squared_error: 2.4602 - y2_output_root_mean_squared_error: 3.1614 - val_loss: 21.6538 - val_y1_output_loss: 7.5903 - val_y2_output_loss: 15.9892 - val_y1_output_root_mean_squared_error: 2.6550 - val_y2_output_root_mean_squared_error: 3.8216\n",
      "Epoch 22/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 18.9395 - y1_output_loss: 6.4566 - y2_output_loss: 12.3307 - y1_output_root_mean_squared_error: 2.5514 - y2_output_root_mean_squared_error: 3.5256 - val_loss: 18.1851 - val_y1_output_loss: 8.5062 - val_y2_output_loss: 11.3403 - val_y1_output_root_mean_squared_error: 2.7925 - val_y2_output_root_mean_squared_error: 3.2229\n",
      "Epoch 23/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 15.7926 - y1_output_loss: 6.0105 - y2_output_loss: 9.7016 - y1_output_root_mean_squared_error: 2.4591 - y2_output_root_mean_squared_error: 3.1218 - val_loss: 14.5336 - val_y1_output_loss: 6.1312 - val_y2_output_loss: 9.2222 - val_y1_output_root_mean_squared_error: 2.4025 - val_y2_output_root_mean_squared_error: 2.9600\n",
      "Epoch 24/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 13.1123 - y1_output_loss: 5.2682 - y2_output_loss: 8.0094 - y1_output_root_mean_squared_error: 2.2967 - y2_output_root_mean_squared_error: 2.7996 - val_loss: 24.0669 - val_y1_output_loss: 8.1641 - val_y2_output_loss: 18.0255 - val_y1_output_root_mean_squared_error: 2.7317 - val_y2_output_root_mean_squared_error: 4.0749\n",
      "Epoch 25/500\n",
      "614/614 [==============================] - 0s 133us/sample - loss: 16.7265 - y1_output_loss: 5.6678 - y2_output_loss: 11.1280 - y1_output_root_mean_squared_error: 2.3791 - y2_output_root_mean_squared_error: 3.3266 - val_loss: 38.6047 - val_y1_output_loss: 11.5480 - val_y2_output_loss: 30.2947 - val_y1_output_root_mean_squared_error: 3.2445 - val_y2_output_root_mean_squared_error: 5.2988\n",
      "Epoch 26/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 16.1266 - y1_output_loss: 5.6835 - y2_output_loss: 10.4118 - y1_output_root_mean_squared_error: 2.3888 - y2_output_root_mean_squared_error: 3.2280 - val_loss: 13.9239 - val_y1_output_loss: 6.3371 - val_y2_output_loss: 8.7422 - val_y1_output_root_mean_squared_error: 2.4243 - val_y2_output_root_mean_squared_error: 2.8366\n",
      "Epoch 27/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 13.1431 - y1_output_loss: 4.8885 - y2_output_loss: 8.2539 - y1_output_root_mean_squared_error: 2.2172 - y2_output_root_mean_squared_error: 2.8683 - val_loss: 25.2941 - val_y1_output_loss: 9.1775 - val_y2_output_loss: 18.4427 - val_y1_output_root_mean_squared_error: 2.8904 - val_y2_output_root_mean_squared_error: 4.1158\n",
      "Epoch 28/500\n",
      "614/614 [==============================] - 0s 132us/sample - loss: 13.3118 - y1_output_loss: 4.6470 - y2_output_loss: 8.6083 - y1_output_root_mean_squared_error: 2.1578 - y2_output_root_mean_squared_error: 2.9421 - val_loss: 14.1002 - val_y1_output_loss: 7.3048 - val_y2_output_loss: 8.0638 - val_y1_output_root_mean_squared_error: 2.5785 - val_y2_output_root_mean_squared_error: 2.7297\n",
      "Epoch 29/500\n",
      "614/614 [==============================] - 0s 132us/sample - loss: 11.8718 - y1_output_loss: 4.6123 - y2_output_loss: 7.1690 - y1_output_root_mean_squared_error: 2.1552 - y2_output_root_mean_squared_error: 2.6883 - val_loss: 12.1358 - val_y1_output_loss: 5.4909 - val_y2_output_loss: 7.8234 - val_y1_output_root_mean_squared_error: 2.2317 - val_y2_output_root_mean_squared_error: 2.6750\n",
      "Epoch 30/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 12.2862 - y1_output_loss: 4.4524 - y2_output_loss: 7.7913 - y1_output_root_mean_squared_error: 2.1059 - y2_output_root_mean_squared_error: 2.8020 - val_loss: 13.9022 - val_y1_output_loss: 7.2327 - val_y2_output_loss: 8.1468 - val_y1_output_root_mean_squared_error: 2.5566 - val_y2_output_root_mean_squared_error: 2.7140\n",
      "Epoch 31/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 11.1257 - y1_output_loss: 4.1434 - y2_output_loss: 6.9022 - y1_output_root_mean_squared_error: 2.0429 - y2_output_root_mean_squared_error: 2.6367 - val_loss: 12.2969 - val_y1_output_loss: 4.9409 - val_y2_output_loss: 8.2367 - val_y1_output_root_mean_squared_error: 2.1268 - val_y2_output_root_mean_squared_error: 2.7881\n",
      "Epoch 32/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 12.3813 - y1_output_loss: 4.3443 - y2_output_loss: 7.9375 - y1_output_root_mean_squared_error: 2.0934 - y2_output_root_mean_squared_error: 2.8283 - val_loss: 12.1366 - val_y1_output_loss: 5.3222 - val_y2_output_loss: 7.5534 - val_y1_output_root_mean_squared_error: 2.2469 - val_y2_output_root_mean_squared_error: 2.6623\n",
      "Epoch 33/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 12.4276 - y1_output_loss: 4.5807 - y2_output_loss: 8.1156 - y1_output_root_mean_squared_error: 2.1055 - y2_output_root_mean_squared_error: 2.8274 - val_loss: 40.0993 - val_y1_output_loss: 11.8040 - val_y2_output_loss: 29.0284 - val_y1_output_root_mean_squared_error: 3.4088 - val_y2_output_root_mean_squared_error: 5.3366\n",
      "Epoch 34/500\n",
      "614/614 [==============================] - 0s 134us/sample - loss: 17.6732 - y1_output_loss: 5.5646 - y2_output_loss: 12.0340 - y1_output_root_mean_squared_error: 2.3674 - y2_output_root_mean_squared_error: 3.4740 - val_loss: 15.3373 - val_y1_output_loss: 6.0899 - val_y2_output_loss: 10.6402 - val_y1_output_root_mean_squared_error: 2.3621 - val_y2_output_root_mean_squared_error: 3.1237\n",
      "Epoch 35/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 11.3483 - y1_output_loss: 3.8603 - y2_output_loss: 7.4536 - y1_output_root_mean_squared_error: 1.9565 - y2_output_root_mean_squared_error: 2.7423 - val_loss: 14.1409 - val_y1_output_loss: 5.7290 - val_y2_output_loss: 8.8973 - val_y1_output_root_mean_squared_error: 2.3566 - val_y2_output_root_mean_squared_error: 2.9304\n",
      "Epoch 36/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 9.6813 - y1_output_loss: 3.4299 - y2_output_loss: 6.3109 - y1_output_root_mean_squared_error: 1.8322 - y2_output_root_mean_squared_error: 2.5149 - val_loss: 14.5091 - val_y1_output_loss: 4.9523 - val_y2_output_loss: 10.2546 - val_y1_output_root_mean_squared_error: 2.1620 - val_y2_output_root_mean_squared_error: 3.1361\n",
      "Epoch 37/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 8.9033 - y1_output_loss: 2.9762 - y2_output_loss: 5.9475 - y1_output_root_mean_squared_error: 1.7309 - y2_output_root_mean_squared_error: 2.4305 - val_loss: 13.6118 - val_y1_output_loss: 5.0332 - val_y2_output_loss: 9.8803 - val_y1_output_root_mean_squared_error: 2.1314 - val_y2_output_root_mean_squared_error: 3.0114\n",
      "Epoch 38/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 9.9608 - y1_output_loss: 3.5385 - y2_output_loss: 6.3543 - y1_output_root_mean_squared_error: 1.8836 - y2_output_root_mean_squared_error: 2.5324 - val_loss: 9.4317 - val_y1_output_loss: 3.8022 - val_y2_output_loss: 6.3495 - val_y1_output_root_mean_squared_error: 1.8667 - val_y2_output_root_mean_squared_error: 2.4387\n",
      "Epoch 39/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 8.1798 - y1_output_loss: 2.8972 - y2_output_loss: 5.2255 - y1_output_root_mean_squared_error: 1.7075 - y2_output_root_mean_squared_error: 2.2944 - val_loss: 8.8240 - val_y1_output_loss: 3.8980 - val_y2_output_loss: 5.6850 - val_y1_output_root_mean_squared_error: 1.8997 - val_y2_output_root_mean_squared_error: 2.2837\n",
      "Epoch 40/500\n",
      "614/614 [==============================] - 0s 134us/sample - loss: 8.2385 - y1_output_loss: 2.8414 - y2_output_loss: 5.4738 - y1_output_root_mean_squared_error: 1.6894 - y2_output_root_mean_squared_error: 2.3204 - val_loss: 9.3977 - val_y1_output_loss: 3.8683 - val_y2_output_loss: 6.4277 - val_y1_output_root_mean_squared_error: 1.8743 - val_y2_output_root_mean_squared_error: 2.4259\n",
      "Epoch 41/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 8.7105 - y1_output_loss: 2.8026 - y2_output_loss: 5.8772 - y1_output_root_mean_squared_error: 1.6814 - y2_output_root_mean_squared_error: 2.4256 - val_loss: 10.1508 - val_y1_output_loss: 3.8851 - val_y2_output_loss: 7.3066 - val_y1_output_root_mean_squared_error: 1.8761 - val_y2_output_root_mean_squared_error: 2.5751\n",
      "Epoch 42/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 9.1558 - y1_output_loss: 2.8429 - y2_output_loss: 6.2364 - y1_output_root_mean_squared_error: 1.6934 - y2_output_root_mean_squared_error: 2.5076 - val_loss: 9.9421 - val_y1_output_loss: 4.2510 - val_y2_output_loss: 6.6994 - val_y1_output_root_mean_squared_error: 1.9675 - val_y2_output_root_mean_squared_error: 2.4640\n",
      "Epoch 43/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 8.0960 - y1_output_loss: 2.6192 - y2_output_loss: 5.4160 - y1_output_root_mean_squared_error: 1.6236 - y2_output_root_mean_squared_error: 2.3367 - val_loss: 8.2339 - val_y1_output_loss: 3.4371 - val_y2_output_loss: 5.3897 - val_y1_output_root_mean_squared_error: 1.7908 - val_y2_output_root_mean_squared_error: 2.2421\n",
      "Epoch 44/500\n",
      "614/614 [==============================] - 0s 132us/sample - loss: 8.2701 - y1_output_loss: 2.6849 - y2_output_loss: 5.6392 - y1_output_root_mean_squared_error: 1.6315 - y2_output_root_mean_squared_error: 2.3682 - val_loss: 7.8746 - val_y1_output_loss: 3.2975 - val_y2_output_loss: 5.2473 - val_y1_output_root_mean_squared_error: 1.7415 - val_y2_output_root_mean_squared_error: 2.2004\n",
      "Epoch 45/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 8.0676 - y1_output_loss: 2.5373 - y2_output_loss: 5.5781 - y1_output_root_mean_squared_error: 1.5854 - y2_output_root_mean_squared_error: 2.3567 - val_loss: 7.3855 - val_y1_output_loss: 3.0905 - val_y2_output_loss: 4.9448 - val_y1_output_root_mean_squared_error: 1.6871 - val_y2_output_root_mean_squared_error: 2.1305\n",
      "Epoch 46/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 7.3011 - y1_output_loss: 2.3925 - y2_output_loss: 4.8569 - y1_output_root_mean_squared_error: 1.5519 - y2_output_root_mean_squared_error: 2.2120 - val_loss: 7.9304 - val_y1_output_loss: 3.6168 - val_y2_output_loss: 5.0724 - val_y1_output_root_mean_squared_error: 1.8085 - val_y2_output_root_mean_squared_error: 2.1586\n",
      "Epoch 47/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 8.1582 - y1_output_loss: 2.5193 - y2_output_loss: 5.6900 - y1_output_root_mean_squared_error: 1.5864 - y2_output_root_mean_squared_error: 2.3752 - val_loss: 12.0807 - val_y1_output_loss: 4.2701 - val_y2_output_loss: 8.7544 - val_y1_output_root_mean_squared_error: 1.9718 - val_y2_output_root_mean_squared_error: 2.8623\n",
      "Epoch 48/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 7.0693 - y1_output_loss: 2.3158 - y2_output_loss: 4.6908 - y1_output_root_mean_squared_error: 1.5275 - y2_output_root_mean_squared_error: 2.1762 - val_loss: 6.5928 - val_y1_output_loss: 2.4444 - val_y2_output_loss: 4.5118 - val_y1_output_root_mean_squared_error: 1.5170 - val_y2_output_root_mean_squared_error: 2.0716\n",
      "Epoch 49/500\n",
      "614/614 [==============================] - 0s 133us/sample - loss: 6.1312 - y1_output_loss: 1.9254 - y2_output_loss: 4.2757 - y1_output_root_mean_squared_error: 1.3866 - y2_output_root_mean_squared_error: 2.0515 - val_loss: 17.0982 - val_y1_output_loss: 5.0198 - val_y2_output_loss: 12.5232 - val_y1_output_root_mean_squared_error: 2.1990 - val_y2_output_root_mean_squared_error: 3.5018\n",
      "Epoch 50/500\n",
      "614/614 [==============================] - 0s 137us/sample - loss: 8.6072 - y1_output_loss: 2.5850 - y2_output_loss: 6.0351 - y1_output_root_mean_squared_error: 1.6133 - y2_output_root_mean_squared_error: 2.4504 - val_loss: 7.8570 - val_y1_output_loss: 2.4888 - val_y2_output_loss: 5.9458 - val_y1_output_root_mean_squared_error: 1.5140 - val_y2_output_root_mean_squared_error: 2.3590\n",
      "Epoch 51/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 6.6914 - y1_output_loss: 1.9188 - y2_output_loss: 4.7318 - y1_output_root_mean_squared_error: 1.3896 - y2_output_root_mean_squared_error: 2.1818 - val_loss: 9.6348 - val_y1_output_loss: 2.8419 - val_y2_output_loss: 7.4923 - val_y1_output_root_mean_squared_error: 1.6131 - val_y2_output_root_mean_squared_error: 2.6519\n",
      "Epoch 52/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 7.0804 - y1_output_loss: 1.9841 - y2_output_loss: 5.0673 - y1_output_root_mean_squared_error: 1.4081 - y2_output_root_mean_squared_error: 2.2578 - val_loss: 15.6887 - val_y1_output_loss: 6.1117 - val_y2_output_loss: 10.6832 - val_y1_output_root_mean_squared_error: 2.3766 - val_y2_output_root_mean_squared_error: 3.1686\n",
      "Epoch 53/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 6.9436 - y1_output_loss: 2.1907 - y2_output_loss: 4.6970 - y1_output_root_mean_squared_error: 1.4851 - y2_output_root_mean_squared_error: 2.1767 - val_loss: 7.0061 - val_y1_output_loss: 2.6475 - val_y2_output_loss: 5.0172 - val_y1_output_root_mean_squared_error: 1.5456 - val_y2_output_root_mean_squared_error: 2.1488\n",
      "Epoch 54/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 6.6653 - y1_output_loss: 1.9129 - y2_output_loss: 4.7113 - y1_output_root_mean_squared_error: 1.3847 - y2_output_root_mean_squared_error: 2.1789 - val_loss: 7.4212 - val_y1_output_loss: 2.7516 - val_y2_output_loss: 4.9879 - val_y1_output_root_mean_squared_error: 1.6261 - val_y2_output_root_mean_squared_error: 2.1856\n",
      "Epoch 55/500\n",
      "614/614 [==============================] - 0s 123us/sample - loss: 6.7056 - y1_output_loss: 2.0075 - y2_output_loss: 4.6532 - y1_output_root_mean_squared_error: 1.4186 - y2_output_root_mean_squared_error: 2.1663 - val_loss: 7.0135 - val_y1_output_loss: 2.3657 - val_y2_output_loss: 5.2911 - val_y1_output_root_mean_squared_error: 1.4583 - val_y2_output_root_mean_squared_error: 2.2106\n",
      "Epoch 56/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 8.6051 - y1_output_loss: 2.4364 - y2_output_loss: 6.1146 - y1_output_root_mean_squared_error: 1.5661 - y2_output_root_mean_squared_error: 2.4804 - val_loss: 7.2113 - val_y1_output_loss: 2.6446 - val_y2_output_loss: 5.1603 - val_y1_output_root_mean_squared_error: 1.5497 - val_y2_output_root_mean_squared_error: 2.1931\n",
      "Epoch 57/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 5.1947 - y1_output_loss: 1.4457 - y2_output_loss: 3.7437 - y1_output_root_mean_squared_error: 1.2072 - y2_output_root_mean_squared_error: 1.9332 - val_loss: 9.7625 - val_y1_output_loss: 3.0435 - val_y2_output_loss: 7.3345 - val_y1_output_root_mean_squared_error: 1.6746 - val_y2_output_root_mean_squared_error: 2.6378\n",
      "Epoch 58/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 6.3363 - y1_output_loss: 1.7186 - y2_output_loss: 4.5982 - y1_output_root_mean_squared_error: 1.3168 - y2_output_root_mean_squared_error: 2.1453 - val_loss: 5.7078 - val_y1_output_loss: 1.9077 - val_y2_output_loss: 4.0287 - val_y1_output_root_mean_squared_error: 1.3513 - val_y2_output_root_mean_squared_error: 1.9703\n",
      "Epoch 59/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 5.2628 - y1_output_loss: 1.6372 - y2_output_loss: 3.6253 - y1_output_root_mean_squared_error: 1.2784 - y2_output_root_mean_squared_error: 1.9049 - val_loss: 14.5004 - val_y1_output_loss: 4.8812 - val_y2_output_loss: 10.0925 - val_y1_output_root_mean_squared_error: 2.1730 - val_y2_output_root_mean_squared_error: 3.1271\n",
      "Epoch 60/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 5.6782 - y1_output_loss: 1.5488 - y2_output_loss: 4.0907 - y1_output_root_mean_squared_error: 1.2447 - y2_output_root_mean_squared_error: 2.0320 - val_loss: 6.4851 - val_y1_output_loss: 1.6686 - val_y2_output_loss: 5.0981 - val_y1_output_root_mean_squared_error: 1.2383 - val_y2_output_root_mean_squared_error: 2.2252\n",
      "Epoch 61/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 4.6242 - y1_output_loss: 1.2226 - y2_output_loss: 3.3873 - y1_output_root_mean_squared_error: 1.1038 - y2_output_root_mean_squared_error: 1.8455 - val_loss: 4.8601 - val_y1_output_loss: 1.7798 - val_y2_output_loss: 3.4047 - val_y1_output_root_mean_squared_error: 1.2778 - val_y2_output_root_mean_squared_error: 1.7965\n",
      "Epoch 62/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 6.0048 - y1_output_loss: 1.7134 - y2_output_loss: 4.3237 - y1_output_root_mean_squared_error: 1.3024 - y2_output_root_mean_squared_error: 2.0757 - val_loss: 9.3363 - val_y1_output_loss: 2.2220 - val_y2_output_loss: 7.7489 - val_y1_output_root_mean_squared_error: 1.4179 - val_y2_output_root_mean_squared_error: 2.7067\n",
      "Epoch 63/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 4.3260 - y1_output_loss: 1.1509 - y2_output_loss: 3.1562 - y1_output_root_mean_squared_error: 1.0776 - y2_output_root_mean_squared_error: 1.7790 - val_loss: 5.3269 - val_y1_output_loss: 1.6423 - val_y2_output_loss: 4.0555 - val_y1_output_root_mean_squared_error: 1.2341 - val_y2_output_root_mean_squared_error: 1.9503\n",
      "Epoch 64/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 5.7459 - y1_output_loss: 1.5476 - y2_output_loss: 4.2691 - y1_output_root_mean_squared_error: 1.2399 - y2_output_root_mean_squared_error: 2.0515 - val_loss: 13.6694 - val_y1_output_loss: 4.2538 - val_y2_output_loss: 10.1835 - val_y1_output_root_mean_squared_error: 1.9871 - val_y2_output_root_mean_squared_error: 3.1178\n",
      "Epoch 65/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 5.0395 - y1_output_loss: 1.3398 - y2_output_loss: 3.6781 - y1_output_root_mean_squared_error: 1.1611 - y2_output_root_mean_squared_error: 1.9213 - val_loss: 4.6691 - val_y1_output_loss: 1.5247 - val_y2_output_loss: 3.4157 - val_y1_output_root_mean_squared_error: 1.1895 - val_y2_output_root_mean_squared_error: 1.8040\n",
      "Epoch 66/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 6.3980 - y1_output_loss: 1.6550 - y2_output_loss: 4.7516 - y1_output_root_mean_squared_error: 1.2797 - y2_output_root_mean_squared_error: 2.1818 - val_loss: 9.5476 - val_y1_output_loss: 3.1529 - val_y2_output_loss: 6.5981 - val_y1_output_root_mean_squared_error: 1.7536 - val_y2_output_root_mean_squared_error: 2.5441\n",
      "Epoch 67/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 5.4107 - y1_output_loss: 1.4378 - y2_output_loss: 3.9288 - y1_output_root_mean_squared_error: 1.2031 - y2_output_root_mean_squared_error: 1.9908 - val_loss: 5.7535 - val_y1_output_loss: 1.8457 - val_y2_output_loss: 4.2725 - val_y1_output_root_mean_squared_error: 1.3079 - val_y2_output_root_mean_squared_error: 2.0107\n",
      "Epoch 68/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 4.0315 - y1_output_loss: 1.0402 - y2_output_loss: 2.9549 - y1_output_root_mean_squared_error: 1.0244 - y2_output_root_mean_squared_error: 1.7269 - val_loss: 8.9480 - val_y1_output_loss: 3.3759 - val_y2_output_loss: 6.0354 - val_y1_output_root_mean_squared_error: 1.7752 - val_y2_output_root_mean_squared_error: 2.4076\n",
      "Epoch 69/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 4.6441 - y1_output_loss: 1.3070 - y2_output_loss: 3.3031 - y1_output_root_mean_squared_error: 1.1481 - y2_output_root_mean_squared_error: 1.8237 - val_loss: 5.9906 - val_y1_output_loss: 1.2788 - val_y2_output_loss: 5.0507 - val_y1_output_root_mean_squared_error: 1.0856 - val_y2_output_root_mean_squared_error: 2.1937\n",
      "Epoch 70/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 4.8822 - y1_output_loss: 1.3101 - y2_output_loss: 3.5317 - y1_output_root_mean_squared_error: 1.1485 - y2_output_root_mean_squared_error: 1.8876 - val_loss: 5.4140 - val_y1_output_loss: 1.5256 - val_y2_output_loss: 4.0424 - val_y1_output_root_mean_squared_error: 1.2152 - val_y2_output_root_mean_squared_error: 1.9842\n",
      "Epoch 71/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 4.7734 - y1_output_loss: 1.1392 - y2_output_loss: 3.5974 - y1_output_root_mean_squared_error: 1.0689 - y2_output_root_mean_squared_error: 1.9055 - val_loss: 4.4972 - val_y1_output_loss: 1.6793 - val_y2_output_loss: 3.0954 - val_y1_output_root_mean_squared_error: 1.2407 - val_y2_output_root_mean_squared_error: 1.7198\n",
      "Epoch 72/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 5.6358 - y1_output_loss: 1.4320 - y2_output_loss: 4.1977 - y1_output_root_mean_squared_error: 1.2019 - y2_output_root_mean_squared_error: 2.0472 - val_loss: 5.0731 - val_y1_output_loss: 1.8097 - val_y2_output_loss: 3.4851 - val_y1_output_root_mean_squared_error: 1.3199 - val_y2_output_root_mean_squared_error: 1.8251\n",
      "Epoch 73/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 4.7784 - y1_output_loss: 1.2418 - y2_output_loss: 3.5373 - y1_output_root_mean_squared_error: 1.1194 - y2_output_root_mean_squared_error: 1.8776 - val_loss: 5.4827 - val_y1_output_loss: 1.2833 - val_y2_output_loss: 4.3330 - val_y1_output_root_mean_squared_error: 1.1137 - val_y2_output_root_mean_squared_error: 2.0597\n",
      "Epoch 74/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 5.6256 - y1_output_loss: 1.4067 - y2_output_loss: 4.2723 - y1_output_root_mean_squared_error: 1.1690 - y2_output_root_mean_squared_error: 2.0638 - val_loss: 14.8038 - val_y1_output_loss: 4.2965 - val_y2_output_loss: 11.2409 - val_y1_output_root_mean_squared_error: 2.0061 - val_y2_output_root_mean_squared_error: 3.2832\n",
      "Epoch 75/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 6.4365 - y1_output_loss: 1.6654 - y2_output_loss: 4.7494 - y1_output_root_mean_squared_error: 1.2961 - y2_output_root_mean_squared_error: 2.1810 - val_loss: 5.7841 - val_y1_output_loss: 1.1306 - val_y2_output_loss: 4.9518 - val_y1_output_root_mean_squared_error: 1.0221 - val_y2_output_root_mean_squared_error: 2.1770\n",
      "Epoch 76/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 5.0318 - y1_output_loss: 1.2441 - y2_output_loss: 3.8023 - y1_output_root_mean_squared_error: 1.1200 - y2_output_root_mean_squared_error: 1.9435 - val_loss: 5.2908 - val_y1_output_loss: 1.5209 - val_y2_output_loss: 3.9854 - val_y1_output_root_mean_squared_error: 1.1859 - val_y2_output_root_mean_squared_error: 1.9709\n",
      "Epoch 77/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 4.2819 - y1_output_loss: 1.0323 - y2_output_loss: 3.2318 - y1_output_root_mean_squared_error: 1.0193 - y2_output_root_mean_squared_error: 1.8008 - val_loss: 6.4075 - val_y1_output_loss: 1.5618 - val_y2_output_loss: 5.2024 - val_y1_output_root_mean_squared_error: 1.1942 - val_y2_output_root_mean_squared_error: 2.2319\n",
      "Epoch 78/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 3.7354 - y1_output_loss: 0.9228 - y2_output_loss: 2.8469 - y1_output_root_mean_squared_error: 0.9640 - y2_output_root_mean_squared_error: 1.6751 - val_loss: 6.7605 - val_y1_output_loss: 1.5440 - val_y2_output_loss: 5.4273 - val_y1_output_root_mean_squared_error: 1.2115 - val_y2_output_root_mean_squared_error: 2.3006\n",
      "Epoch 79/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 4.3185 - y1_output_loss: 1.0503 - y2_output_loss: 3.2416 - y1_output_root_mean_squared_error: 1.0287 - y2_output_root_mean_squared_error: 1.8056 - val_loss: 4.1961 - val_y1_output_loss: 1.2708 - val_y2_output_loss: 3.0543 - val_y1_output_root_mean_squared_error: 1.1068 - val_y2_output_root_mean_squared_error: 1.7237\n",
      "Epoch 80/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 4.2281 - y1_output_loss: 1.1012 - y2_output_loss: 3.1474 - y1_output_root_mean_squared_error: 1.0532 - y2_output_root_mean_squared_error: 1.7660 - val_loss: 9.0928 - val_y1_output_loss: 2.6748 - val_y2_output_loss: 6.8021 - val_y1_output_root_mean_squared_error: 1.5852 - val_y2_output_root_mean_squared_error: 2.5651\n",
      "Epoch 81/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 4.3628 - y1_output_loss: 1.1627 - y2_output_loss: 3.1922 - y1_output_root_mean_squared_error: 1.0763 - y2_output_root_mean_squared_error: 1.7901 - val_loss: 5.0934 - val_y1_output_loss: 1.5738 - val_y2_output_loss: 3.6780 - val_y1_output_root_mean_squared_error: 1.2236 - val_y2_output_root_mean_squared_error: 1.8963\n",
      "Epoch 82/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 3.7708 - y1_output_loss: 0.8884 - y2_output_loss: 2.8549 - y1_output_root_mean_squared_error: 0.9450 - y2_output_root_mean_squared_error: 1.6964 - val_loss: 3.1904 - val_y1_output_loss: 0.9506 - val_y2_output_loss: 2.3772 - val_y1_output_root_mean_squared_error: 0.9339 - val_y2_output_root_mean_squared_error: 1.5226\n",
      "Epoch 83/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 4.0688 - y1_output_loss: 0.8959 - y2_output_loss: 3.1476 - y1_output_root_mean_squared_error: 0.9497 - y2_output_root_mean_squared_error: 1.7796 - val_loss: 3.5167 - val_y1_output_loss: 0.8945 - val_y2_output_loss: 2.7581 - val_y1_output_root_mean_squared_error: 0.9158 - val_y2_output_root_mean_squared_error: 1.6365\n",
      "Epoch 84/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 3.5432 - y1_output_loss: 0.8207 - y2_output_loss: 2.6960 - y1_output_root_mean_squared_error: 0.9092 - y2_output_root_mean_squared_error: 1.6482 - val_loss: 4.6894 - val_y1_output_loss: 1.2557 - val_y2_output_loss: 3.6412 - val_y1_output_root_mean_squared_error: 1.0810 - val_y2_output_root_mean_squared_error: 1.8764\n",
      "Epoch 85/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 5.2642 - y1_output_loss: 1.4656 - y2_output_loss: 3.7533 - y1_output_root_mean_squared_error: 1.2152 - y2_output_root_mean_squared_error: 1.9462 - val_loss: 3.3037 - val_y1_output_loss: 0.7668 - val_y2_output_loss: 2.5279 - val_y1_output_root_mean_squared_error: 0.8729 - val_y2_output_root_mean_squared_error: 1.5943\n",
      "Epoch 86/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 6.6735 - y1_output_loss: 1.7111 - y2_output_loss: 4.9271 - y1_output_root_mean_squared_error: 1.3076 - y2_output_root_mean_squared_error: 2.2279 - val_loss: 3.5356 - val_y1_output_loss: 0.7305 - val_y2_output_loss: 2.8626 - val_y1_output_root_mean_squared_error: 0.8395 - val_y2_output_root_mean_squared_error: 1.6825\n",
      "Epoch 87/500\n",
      "614/614 [==============================] - 0s 131us/sample - loss: 4.0472 - y1_output_loss: 1.0944 - y2_output_loss: 2.9814 - y1_output_root_mean_squared_error: 1.0493 - y2_output_root_mean_squared_error: 1.7164 - val_loss: 15.6243 - val_y1_output_loss: 2.3835 - val_y2_output_loss: 13.8125 - val_y1_output_root_mean_squared_error: 1.4868 - val_y2_output_root_mean_squared_error: 3.6625\n",
      "Epoch 88/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 3.9875 - y1_output_loss: 0.9171 - y2_output_loss: 3.0987 - y1_output_root_mean_squared_error: 0.9510 - y2_output_root_mean_squared_error: 1.7559 - val_loss: 21.8873 - val_y1_output_loss: 8.5038 - val_y2_output_loss: 14.5094 - val_y1_output_root_mean_squared_error: 2.8273 - val_y2_output_root_mean_squared_error: 3.7274\n",
      "Epoch 89/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 4.5099 - y1_output_loss: 1.0541 - y2_output_loss: 3.4432 - y1_output_root_mean_squared_error: 1.0287 - y2_output_root_mean_squared_error: 1.8579 - val_loss: 3.7169 - val_y1_output_loss: 0.7831 - val_y2_output_loss: 3.0436 - val_y1_output_root_mean_squared_error: 0.8559 - val_y2_output_root_mean_squared_error: 1.7275\n",
      "Epoch 90/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 4.2240 - y1_output_loss: 1.0666 - y2_output_loss: 3.1494 - y1_output_root_mean_squared_error: 1.0263 - y2_output_root_mean_squared_error: 1.7806 - val_loss: 4.6736 - val_y1_output_loss: 1.4806 - val_y2_output_loss: 3.1948 - val_y1_output_root_mean_squared_error: 1.2251 - val_y2_output_root_mean_squared_error: 1.7812\n",
      "Epoch 91/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 4.2126 - y1_output_loss: 1.1509 - y2_output_loss: 3.0382 - y1_output_root_mean_squared_error: 1.0760 - y2_output_root_mean_squared_error: 1.7478 - val_loss: 3.6531 - val_y1_output_loss: 1.1432 - val_y2_output_loss: 2.6164 - val_y1_output_root_mean_squared_error: 1.0429 - val_y2_output_root_mean_squared_error: 1.6017\n",
      "Epoch 92/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 3.1832 - y1_output_loss: 0.7445 - y2_output_loss: 2.4381 - y1_output_root_mean_squared_error: 0.8667 - y2_output_root_mean_squared_error: 1.5595 - val_loss: 3.2149 - val_y1_output_loss: 0.8741 - val_y2_output_loss: 2.4325 - val_y1_output_root_mean_squared_error: 0.9137 - val_y2_output_root_mean_squared_error: 1.5428\n",
      "Epoch 93/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 4.5279 - y1_output_loss: 1.1202 - y2_output_loss: 3.3852 - y1_output_root_mean_squared_error: 1.0629 - y2_output_root_mean_squared_error: 1.8434 - val_loss: 3.0911 - val_y1_output_loss: 0.9497 - val_y2_output_loss: 2.2335 - val_y1_output_root_mean_squared_error: 0.9489 - val_y2_output_root_mean_squared_error: 1.4801\n",
      "Epoch 94/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 3.5797 - y1_output_loss: 0.8488 - y2_output_loss: 2.7023 - y1_output_root_mean_squared_error: 0.9243 - y2_output_root_mean_squared_error: 1.6509 - val_loss: 3.8768 - val_y1_output_loss: 1.1422 - val_y2_output_loss: 2.8713 - val_y1_output_root_mean_squared_error: 1.0276 - val_y2_output_root_mean_squared_error: 1.6795\n",
      "Epoch 95/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 4.4637 - y1_output_loss: 1.0502 - y2_output_loss: 3.3811 - y1_output_root_mean_squared_error: 1.0286 - y2_output_root_mean_squared_error: 1.8454 - val_loss: 3.9835 - val_y1_output_loss: 1.2501 - val_y2_output_loss: 2.7306 - val_y1_output_root_mean_squared_error: 1.1196 - val_y2_output_root_mean_squared_error: 1.6523\n",
      "Epoch 96/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 3.1822 - y1_output_loss: 0.6959 - y2_output_loss: 2.4844 - y1_output_root_mean_squared_error: 0.8283 - y2_output_root_mean_squared_error: 1.5799 - val_loss: 3.9008 - val_y1_output_loss: 1.1248 - val_y2_output_loss: 2.7822 - val_y1_output_root_mean_squared_error: 1.0675 - val_y2_output_root_mean_squared_error: 1.6617\n",
      "Epoch 97/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 3.4224 - y1_output_loss: 0.7515 - y2_output_loss: 2.6584 - y1_output_root_mean_squared_error: 0.8685 - y2_output_root_mean_squared_error: 1.6335 - val_loss: 2.7251 - val_y1_output_loss: 0.6570 - val_y2_output_loss: 2.0855 - val_y1_output_root_mean_squared_error: 0.7958 - val_y2_output_root_mean_squared_error: 1.4463\n",
      "Epoch 98/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 3.1943 - y1_output_loss: 0.7334 - y2_output_loss: 2.4948 - y1_output_root_mean_squared_error: 0.8586 - y2_output_root_mean_squared_error: 1.5675 - val_loss: 7.5232 - val_y1_output_loss: 0.8388 - val_y2_output_loss: 6.7919 - val_y1_output_root_mean_squared_error: 0.9143 - val_y2_output_root_mean_squared_error: 2.5860\n",
      "Epoch 99/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 3.0424 - y1_output_loss: 0.6711 - y2_output_loss: 2.3583 - y1_output_root_mean_squared_error: 0.8195 - y2_output_root_mean_squared_error: 1.5397 - val_loss: 4.3693 - val_y1_output_loss: 0.7175 - val_y2_output_loss: 3.6902 - val_y1_output_root_mean_squared_error: 0.8326 - val_y2_output_root_mean_squared_error: 1.9173\n",
      "Epoch 100/500\n",
      "614/614 [==============================] - 0s 131us/sample - loss: 3.0969 - y1_output_loss: 0.6262 - y2_output_loss: 2.4837 - y1_output_root_mean_squared_error: 0.7917 - y2_output_root_mean_squared_error: 1.5717 - val_loss: 7.2186 - val_y1_output_loss: 0.9314 - val_y2_output_loss: 6.4470 - val_y1_output_root_mean_squared_error: 0.9557 - val_y2_output_root_mean_squared_error: 2.5110\n",
      "Epoch 101/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 4.0464 - y1_output_loss: 1.0167 - y2_output_loss: 3.0429 - y1_output_root_mean_squared_error: 1.0126 - y2_output_root_mean_squared_error: 1.7381 - val_loss: 3.3223 - val_y1_output_loss: 0.8550 - val_y2_output_loss: 2.4240 - val_y1_output_root_mean_squared_error: 0.9264 - val_y2_output_root_mean_squared_error: 1.5698\n",
      "Epoch 102/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 4.2509 - y1_output_loss: 0.7917 - y2_output_loss: 3.4396 - y1_output_root_mean_squared_error: 0.8824 - y2_output_root_mean_squared_error: 1.8634 - val_loss: 2.9709 - val_y1_output_loss: 0.7307 - val_y2_output_loss: 2.2869 - val_y1_output_root_mean_squared_error: 0.8346 - val_y2_output_root_mean_squared_error: 1.5081\n",
      "Epoch 103/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 4.0854 - y1_output_loss: 1.1420 - y2_output_loss: 2.9608 - y1_output_root_mean_squared_error: 1.0621 - y2_output_root_mean_squared_error: 1.7197 - val_loss: 6.1964 - val_y1_output_loss: 2.0713 - val_y2_output_loss: 4.3282 - val_y1_output_root_mean_squared_error: 1.4125 - val_y2_output_root_mean_squared_error: 2.0497\n",
      "Epoch 104/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 2.7778 - y1_output_loss: 0.5792 - y2_output_loss: 2.1766 - y1_output_root_mean_squared_error: 0.7646 - y2_output_root_mean_squared_error: 1.4809 - val_loss: 3.1741 - val_y1_output_loss: 0.7001 - val_y2_output_loss: 2.5026 - val_y1_output_root_mean_squared_error: 0.8366 - val_y2_output_root_mean_squared_error: 1.5730\n",
      "Epoch 105/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 3.2239 - y1_output_loss: 0.7146 - y2_output_loss: 2.4991 - y1_output_root_mean_squared_error: 0.8436 - y2_output_root_mean_squared_error: 1.5850 - val_loss: 4.3136 - val_y1_output_loss: 1.3758 - val_y2_output_loss: 2.9197 - val_y1_output_root_mean_squared_error: 1.1690 - val_y2_output_root_mean_squared_error: 1.7167\n",
      "Epoch 106/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 3.6780 - y1_output_loss: 0.7591 - y2_output_loss: 2.9174 - y1_output_root_mean_squared_error: 0.8753 - y2_output_root_mean_squared_error: 1.7064 - val_loss: 3.0894 - val_y1_output_loss: 0.7754 - val_y2_output_loss: 2.2814 - val_y1_output_root_mean_squared_error: 0.8851 - val_y2_output_root_mean_squared_error: 1.5185\n",
      "Epoch 107/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 2.5768 - y1_output_loss: 0.6529 - y2_output_loss: 1.9318 - y1_output_root_mean_squared_error: 0.8067 - y2_output_root_mean_squared_error: 1.3878 - val_loss: 2.4711 - val_y1_output_loss: 0.6340 - val_y2_output_loss: 1.8192 - val_y1_output_root_mean_squared_error: 0.7955 - val_y2_output_root_mean_squared_error: 1.3558\n",
      "Epoch 108/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 3.1630 - y1_output_loss: 0.6574 - y2_output_loss: 2.4829 - y1_output_root_mean_squared_error: 0.8137 - y2_output_root_mean_squared_error: 1.5814 - val_loss: 4.5437 - val_y1_output_loss: 1.6617 - val_y2_output_loss: 2.9119 - val_y1_output_root_mean_squared_error: 1.2642 - val_y2_output_root_mean_squared_error: 1.7162\n",
      "Epoch 109/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 4.1191 - y1_output_loss: 0.9273 - y2_output_loss: 3.1578 - y1_output_root_mean_squared_error: 0.9668 - y2_output_root_mean_squared_error: 1.7845 - val_loss: 2.4750 - val_y1_output_loss: 0.6653 - val_y2_output_loss: 1.7736 - val_y1_output_root_mean_squared_error: 0.8200 - val_y2_output_root_mean_squared_error: 1.3426\n",
      "Epoch 110/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 3.3665 - y1_output_loss: 0.7594 - y2_output_loss: 2.6116 - y1_output_root_mean_squared_error: 0.8741 - y2_output_root_mean_squared_error: 1.6132 - val_loss: 5.2248 - val_y1_output_loss: 1.8103 - val_y2_output_loss: 3.4472 - val_y1_output_root_mean_squared_error: 1.3286 - val_y2_output_root_mean_squared_error: 1.8600\n",
      "Epoch 111/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 3.3963 - y1_output_loss: 0.8165 - y2_output_loss: 2.5990 - y1_output_root_mean_squared_error: 0.9068 - y2_output_root_mean_squared_error: 1.6044 - val_loss: 4.4063 - val_y1_output_loss: 0.8267 - val_y2_output_loss: 3.6207 - val_y1_output_root_mean_squared_error: 0.9157 - val_y2_output_root_mean_squared_error: 1.8888\n",
      "Epoch 112/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 2.5852 - y1_output_loss: 0.6018 - y2_output_loss: 1.9705 - y1_output_root_mean_squared_error: 0.7778 - y2_output_root_mean_squared_error: 1.4072 - val_loss: 3.4906 - val_y1_output_loss: 0.9335 - val_y2_output_loss: 2.5973 - val_y1_output_root_mean_squared_error: 0.9347 - val_y2_output_root_mean_squared_error: 1.6177\n",
      "Epoch 113/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 3.8843 - y1_output_loss: 0.9566 - y2_output_loss: 2.9149 - y1_output_root_mean_squared_error: 0.9826 - y2_output_root_mean_squared_error: 1.7085 - val_loss: 4.1063 - val_y1_output_loss: 0.8150 - val_y2_output_loss: 3.2734 - val_y1_output_root_mean_squared_error: 0.9092 - val_y2_output_root_mean_squared_error: 1.8110\n",
      "Epoch 114/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 2.9343 - y1_output_loss: 0.7198 - y2_output_loss: 2.2181 - y1_output_root_mean_squared_error: 0.8393 - y2_output_root_mean_squared_error: 1.4932 - val_loss: 4.2780 - val_y1_output_loss: 1.3049 - val_y2_output_loss: 3.0959 - val_y1_output_root_mean_squared_error: 1.1206 - val_y2_output_root_mean_squared_error: 1.7385\n",
      "Epoch 115/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 6.3988 - y1_output_loss: 1.6911 - y2_output_loss: 4.6838 - y1_output_root_mean_squared_error: 1.3015 - y2_output_root_mean_squared_error: 2.1691 - val_loss: 5.3936 - val_y1_output_loss: 1.8969 - val_y2_output_loss: 3.6292 - val_y1_output_root_mean_squared_error: 1.3429 - val_y2_output_root_mean_squared_error: 1.8948\n",
      "Epoch 116/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 2.9353 - y1_output_loss: 0.6980 - y2_output_loss: 2.2341 - y1_output_root_mean_squared_error: 0.8263 - y2_output_root_mean_squared_error: 1.5008 - val_loss: 2.5264 - val_y1_output_loss: 0.7381 - val_y2_output_loss: 1.7516 - val_y1_output_root_mean_squared_error: 0.8692 - val_y2_output_root_mean_squared_error: 1.3308\n",
      "Epoch 117/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 4.1868 - y1_output_loss: 0.9847 - y2_output_loss: 3.1823 - y1_output_root_mean_squared_error: 0.9946 - y2_output_root_mean_squared_error: 1.7882 - val_loss: 2.7781 - val_y1_output_loss: 0.6700 - val_y2_output_loss: 2.1213 - val_y1_output_root_mean_squared_error: 0.8026 - val_y2_output_root_mean_squared_error: 1.4608\n",
      "Epoch 118/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 3.3641 - y1_output_loss: 0.7734 - y2_output_loss: 2.5779 - y1_output_root_mean_squared_error: 0.8804 - y2_output_root_mean_squared_error: 1.6090 - val_loss: 6.5466 - val_y1_output_loss: 1.1147 - val_y2_output_loss: 5.5402 - val_y1_output_root_mean_squared_error: 1.0243 - val_y2_output_root_mean_squared_error: 2.3447\n",
      "Epoch 119/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 3.0608 - y1_output_loss: 0.7278 - y2_output_loss: 2.3380 - y1_output_root_mean_squared_error: 0.8464 - y2_output_root_mean_squared_error: 1.5311 - val_loss: 4.8605 - val_y1_output_loss: 1.1399 - val_y2_output_loss: 3.7722 - val_y1_output_root_mean_squared_error: 1.0721 - val_y2_output_root_mean_squared_error: 1.9264\n",
      "Epoch 120/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 2.4146 - y1_output_loss: 0.6134 - y2_output_loss: 1.8292 - y1_output_root_mean_squared_error: 0.7827 - y2_output_root_mean_squared_error: 1.3424 - val_loss: 2.8570 - val_y1_output_loss: 0.9579 - val_y2_output_loss: 1.9051 - val_y1_output_root_mean_squared_error: 0.9659 - val_y2_output_root_mean_squared_error: 1.3871\n",
      "Epoch 121/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 2.9237 - y1_output_loss: 0.6795 - y2_output_loss: 2.2377 - y1_output_root_mean_squared_error: 0.8211 - y2_output_root_mean_squared_error: 1.4998 - val_loss: 6.3664 - val_y1_output_loss: 1.9910 - val_y2_output_loss: 4.3060 - val_y1_output_root_mean_squared_error: 1.4170 - val_y2_output_root_mean_squared_error: 2.0877\n",
      "Epoch 122/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 3.0915 - y1_output_loss: 0.8326 - y2_output_loss: 2.2377 - y1_output_root_mean_squared_error: 0.9163 - y2_output_root_mean_squared_error: 1.5006 - val_loss: 2.4145 - val_y1_output_loss: 0.5446 - val_y2_output_loss: 1.8548 - val_y1_output_root_mean_squared_error: 0.7366 - val_y2_output_root_mean_squared_error: 1.3682\n",
      "Epoch 123/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 2.5976 - y1_output_loss: 0.6113 - y2_output_loss: 1.9811 - y1_output_root_mean_squared_error: 0.7833 - y2_output_root_mean_squared_error: 1.4086 - val_loss: 2.8785 - val_y1_output_loss: 0.8498 - val_y2_output_loss: 2.0329 - val_y1_output_root_mean_squared_error: 0.9291 - val_y2_output_root_mean_squared_error: 1.4196\n",
      "Epoch 124/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 2.1707 - y1_output_loss: 0.5041 - y2_output_loss: 1.6718 - y1_output_root_mean_squared_error: 0.7130 - y2_output_root_mean_squared_error: 1.2893 - val_loss: 2.7908 - val_y1_output_loss: 0.4901 - val_y2_output_loss: 2.3213 - val_y1_output_root_mean_squared_error: 0.6973 - val_y2_output_root_mean_squared_error: 1.5181\n",
      "Epoch 125/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 2.4927 - y1_output_loss: 0.5161 - y2_output_loss: 1.9920 - y1_output_root_mean_squared_error: 0.7188 - y2_output_root_mean_squared_error: 1.4057 - val_loss: 7.4411 - val_y1_output_loss: 2.1348 - val_y2_output_loss: 5.3733 - val_y1_output_root_mean_squared_error: 1.4671 - val_y2_output_root_mean_squared_error: 2.2997\n",
      "Epoch 126/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 2.4119 - y1_output_loss: 0.5606 - y2_output_loss: 1.8455 - y1_output_root_mean_squared_error: 0.7518 - y2_output_root_mean_squared_error: 1.3589 - val_loss: 3.5438 - val_y1_output_loss: 1.1251 - val_y2_output_loss: 2.4101 - val_y1_output_root_mean_squared_error: 1.0638 - val_y2_output_root_mean_squared_error: 1.5531\n",
      "Epoch 127/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 2.6586 - y1_output_loss: 0.6856 - y2_output_loss: 1.9569 - y1_output_root_mean_squared_error: 0.8280 - y2_output_root_mean_squared_error: 1.4047 - val_loss: 3.3044 - val_y1_output_loss: 0.8259 - val_y2_output_loss: 2.5374 - val_y1_output_root_mean_squared_error: 0.9037 - val_y2_output_root_mean_squared_error: 1.5772\n",
      "Epoch 128/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 3.3272 - y1_output_loss: 0.8429 - y2_output_loss: 2.4570 - y1_output_root_mean_squared_error: 0.9216 - y2_output_root_mean_squared_error: 1.5741 - val_loss: 3.0218 - val_y1_output_loss: 0.7208 - val_y2_output_loss: 2.2964 - val_y1_output_root_mean_squared_error: 0.8556 - val_y2_output_root_mean_squared_error: 1.5132\n",
      "Epoch 129/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 2.4072 - y1_output_loss: 0.5623 - y2_output_loss: 1.8414 - y1_output_root_mean_squared_error: 0.7515 - y2_output_root_mean_squared_error: 1.3574 - val_loss: 7.5805 - val_y1_output_loss: 2.0422 - val_y2_output_loss: 5.5168 - val_y1_output_root_mean_squared_error: 1.4245 - val_y2_output_root_mean_squared_error: 2.3561\n",
      "Epoch 130/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 2.4658 - y1_output_loss: 0.6263 - y2_output_loss: 1.8213 - y1_output_root_mean_squared_error: 0.7942 - y2_output_root_mean_squared_error: 1.3546 - val_loss: 1.8976 - val_y1_output_loss: 0.4444 - val_y2_output_loss: 1.4430 - val_y1_output_root_mean_squared_error: 0.6554 - val_y2_output_root_mean_squared_error: 1.2116\n",
      "Epoch 131/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 3.9270 - y1_output_loss: 0.9308 - y2_output_loss: 2.9658 - y1_output_root_mean_squared_error: 0.9681 - y2_output_root_mean_squared_error: 1.7291 - val_loss: 2.5507 - val_y1_output_loss: 0.6510 - val_y2_output_loss: 1.8673 - val_y1_output_root_mean_squared_error: 0.8027 - val_y2_output_root_mean_squared_error: 1.3807\n",
      "Epoch 132/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 2.4755 - y1_output_loss: 0.6406 - y2_output_loss: 1.8217 - y1_output_root_mean_squared_error: 0.8002 - y2_output_root_mean_squared_error: 1.3547 - val_loss: 2.6840 - val_y1_output_loss: 0.4729 - val_y2_output_loss: 2.2426 - val_y1_output_root_mean_squared_error: 0.6870 - val_y2_output_root_mean_squared_error: 1.4873\n",
      "Epoch 133/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 2.1614 - y1_output_loss: 0.4811 - y2_output_loss: 1.6830 - y1_output_root_mean_squared_error: 0.6836 - y2_output_root_mean_squared_error: 1.3015 - val_loss: 2.5592 - val_y1_output_loss: 0.6486 - val_y2_output_loss: 1.8985 - val_y1_output_root_mean_squared_error: 0.8092 - val_y2_output_root_mean_squared_error: 1.3800\n",
      "Epoch 134/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 1.8007 - y1_output_loss: 0.4133 - y2_output_loss: 1.3984 - y1_output_root_mean_squared_error: 0.6452 - y2_output_root_mean_squared_error: 1.1766 - val_loss: 4.7319 - val_y1_output_loss: 0.5180 - val_y2_output_loss: 4.2431 - val_y1_output_root_mean_squared_error: 0.6984 - val_y2_output_root_mean_squared_error: 2.0601\n",
      "Epoch 135/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 1.9018 - y1_output_loss: 0.4841 - y2_output_loss: 1.4264 - y1_output_root_mean_squared_error: 0.6970 - y2_output_root_mean_squared_error: 1.1900 - val_loss: 3.8949 - val_y1_output_loss: 0.6526 - val_y2_output_loss: 3.2606 - val_y1_output_root_mean_squared_error: 0.8201 - val_y2_output_root_mean_squared_error: 1.7951\n",
      "Epoch 136/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 2.7969 - y1_output_loss: 0.6975 - y2_output_loss: 2.0915 - y1_output_root_mean_squared_error: 0.8361 - y2_output_root_mean_squared_error: 1.4484 - val_loss: 2.0701 - val_y1_output_loss: 0.4015 - val_y2_output_loss: 1.6710 - val_y1_output_root_mean_squared_error: 0.6206 - val_y2_output_root_mean_squared_error: 1.2981\n",
      "Epoch 137/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 2.1100 - y1_output_loss: 0.4312 - y2_output_loss: 1.6691 - y1_output_root_mean_squared_error: 0.6572 - y2_output_root_mean_squared_error: 1.2954 - val_loss: 2.7958 - val_y1_output_loss: 0.7870 - val_y2_output_loss: 1.9849 - val_y1_output_root_mean_squared_error: 0.8976 - val_y2_output_root_mean_squared_error: 1.4108\n",
      "Epoch 138/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 1.8879 - y1_output_loss: 0.4539 - y2_output_loss: 1.4702 - y1_output_root_mean_squared_error: 0.6711 - y2_output_root_mean_squared_error: 1.1990 - val_loss: 5.6266 - val_y1_output_loss: 0.8997 - val_y2_output_loss: 4.7421 - val_y1_output_root_mean_squared_error: 0.9297 - val_y2_output_root_mean_squared_error: 2.1823\n",
      "Epoch 139/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 2.1501 - y1_output_loss: 0.5147 - y2_output_loss: 1.6221 - y1_output_root_mean_squared_error: 0.7205 - y2_output_root_mean_squared_error: 1.2771 - val_loss: 1.7642 - val_y1_output_loss: 0.3780 - val_y2_output_loss: 1.3807 - val_y1_output_root_mean_squared_error: 0.6095 - val_y2_output_root_mean_squared_error: 1.1802\n",
      "Epoch 140/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 1.8219 - y1_output_loss: 0.4058 - y2_output_loss: 1.4045 - y1_output_root_mean_squared_error: 0.6398 - y2_output_root_mean_squared_error: 1.1885 - val_loss: 1.8347 - val_y1_output_loss: 0.4203 - val_y2_output_loss: 1.3914 - val_y1_output_root_mean_squared_error: 0.6431 - val_y2_output_root_mean_squared_error: 1.1921\n",
      "Epoch 141/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 1.8668 - y1_output_loss: 0.4254 - y2_output_loss: 1.4987 - y1_output_root_mean_squared_error: 0.6509 - y2_output_root_mean_squared_error: 1.2013 - val_loss: 10.9087 - val_y1_output_loss: 2.3690 - val_y2_output_loss: 8.7298 - val_y1_output_root_mean_squared_error: 1.5477 - val_y2_output_root_mean_squared_error: 2.9178\n",
      "Epoch 142/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 2.0264 - y1_output_loss: 0.5025 - y2_output_loss: 1.5106 - y1_output_root_mean_squared_error: 0.7118 - y2_output_root_mean_squared_error: 1.2328 - val_loss: 3.0597 - val_y1_output_loss: 0.8487 - val_y2_output_loss: 2.2372 - val_y1_output_root_mean_squared_error: 0.9199 - val_y2_output_root_mean_squared_error: 1.4878\n",
      "Epoch 143/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 2.0388 - y1_output_loss: 0.5065 - y2_output_loss: 1.5281 - y1_output_root_mean_squared_error: 0.7140 - y2_output_root_mean_squared_error: 1.2365 - val_loss: 3.5033 - val_y1_output_loss: 0.7613 - val_y2_output_loss: 2.8021 - val_y1_output_root_mean_squared_error: 0.8428 - val_y2_output_root_mean_squared_error: 1.6712\n",
      "Epoch 144/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 2.0584 - y1_output_loss: 0.4926 - y2_output_loss: 1.5516 - y1_output_root_mean_squared_error: 0.7047 - y2_output_root_mean_squared_error: 1.2497 - val_loss: 2.2978 - val_y1_output_loss: 0.4538 - val_y2_output_loss: 1.8435 - val_y1_output_root_mean_squared_error: 0.6601 - val_y2_output_root_mean_squared_error: 1.3646\n",
      "Epoch 145/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 2.3135 - y1_output_loss: 0.5091 - y2_output_loss: 1.8403 - y1_output_root_mean_squared_error: 0.7117 - y2_output_root_mean_squared_error: 1.3443 - val_loss: 5.7834 - val_y1_output_loss: 1.7656 - val_y2_output_loss: 3.9293 - val_y1_output_root_mean_squared_error: 1.3308 - val_y2_output_root_mean_squared_error: 2.0031\n",
      "Epoch 146/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 2.5642 - y1_output_loss: 0.6357 - y2_output_loss: 1.9107 - y1_output_root_mean_squared_error: 0.8006 - y2_output_root_mean_squared_error: 1.3868 - val_loss: 2.0220 - val_y1_output_loss: 0.5689 - val_y2_output_loss: 1.4313 - val_y1_output_root_mean_squared_error: 0.7371 - val_y2_output_root_mean_squared_error: 1.2160\n",
      "Epoch 147/500\n",
      "614/614 [==============================] - 0s 134us/sample - loss: 2.4173 - y1_output_loss: 0.5819 - y2_output_loss: 1.8313 - y1_output_root_mean_squared_error: 0.7640 - y2_output_root_mean_squared_error: 1.3541 - val_loss: 3.4368 - val_y1_output_loss: 1.3664 - val_y2_output_loss: 2.1766 - val_y1_output_root_mean_squared_error: 1.1339 - val_y2_output_root_mean_squared_error: 1.4667\n",
      "Epoch 148/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 2.1593 - y1_output_loss: 0.5107 - y2_output_loss: 1.6524 - y1_output_root_mean_squared_error: 0.7169 - y2_output_root_mean_squared_error: 1.2827 - val_loss: 3.0298 - val_y1_output_loss: 0.6863 - val_y2_output_loss: 2.4331 - val_y1_output_root_mean_squared_error: 0.8087 - val_y2_output_root_mean_squared_error: 1.5414\n",
      "Epoch 149/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 1.8470 - y1_output_loss: 0.4423 - y2_output_loss: 1.3952 - y1_output_root_mean_squared_error: 0.6658 - y2_output_root_mean_squared_error: 1.1848 - val_loss: 3.7992 - val_y1_output_loss: 0.8219 - val_y2_output_loss: 2.9624 - val_y1_output_root_mean_squared_error: 0.8870 - val_y2_output_root_mean_squared_error: 1.7357\n",
      "Epoch 150/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 2.1318 - y1_output_loss: 0.4329 - y2_output_loss: 1.6820 - y1_output_root_mean_squared_error: 0.6609 - y2_output_root_mean_squared_error: 1.3020 - val_loss: 1.8379 - val_y1_output_loss: 0.4378 - val_y2_output_loss: 1.3870 - val_y1_output_root_mean_squared_error: 0.6595 - val_y2_output_root_mean_squared_error: 1.1845\n",
      "Epoch 151/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 3.9205 - y1_output_loss: 1.0654 - y2_output_loss: 2.8685 - y1_output_root_mean_squared_error: 1.0344 - y2_output_root_mean_squared_error: 1.6883 - val_loss: 4.7963 - val_y1_output_loss: 0.4524 - val_y2_output_loss: 4.4784 - val_y1_output_root_mean_squared_error: 0.6786 - val_y2_output_root_mean_squared_error: 2.0823\n",
      "Epoch 152/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 1.7972 - y1_output_loss: 0.3759 - y2_output_loss: 1.4263 - y1_output_root_mean_squared_error: 0.6079 - y2_output_root_mean_squared_error: 1.1948 - val_loss: 3.5207 - val_y1_output_loss: 0.7813 - val_y2_output_loss: 2.8066 - val_y1_output_root_mean_squared_error: 0.8817 - val_y2_output_root_mean_squared_error: 1.6563\n",
      "Epoch 153/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 3.0344 - y1_output_loss: 0.7408 - y2_output_loss: 2.2682 - y1_output_root_mean_squared_error: 0.8635 - y2_output_root_mean_squared_error: 1.5128 - val_loss: 2.2878 - val_y1_output_loss: 0.4204 - val_y2_output_loss: 1.8990 - val_y1_output_root_mean_squared_error: 0.6440 - val_y2_output_root_mean_squared_error: 1.3686\n",
      "Epoch 154/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 1.7387 - y1_output_loss: 0.3387 - y2_output_loss: 1.3861 - y1_output_root_mean_squared_error: 0.5847 - y2_output_root_mean_squared_error: 1.1819 - val_loss: 1.6920 - val_y1_output_loss: 0.3766 - val_y2_output_loss: 1.3173 - val_y1_output_root_mean_squared_error: 0.5997 - val_y2_output_root_mean_squared_error: 1.1543\n",
      "Epoch 155/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 1.4032 - y1_output_loss: 0.3179 - y2_output_loss: 1.0735 - y1_output_root_mean_squared_error: 0.5660 - y2_output_root_mean_squared_error: 1.0406 - val_loss: 1.7997 - val_y1_output_loss: 0.3821 - val_y2_output_loss: 1.4000 - val_y1_output_root_mean_squared_error: 0.6026 - val_y2_output_root_mean_squared_error: 1.1986\n",
      "Epoch 156/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 1.5618 - y1_output_loss: 0.3734 - y2_output_loss: 1.1830 - y1_output_root_mean_squared_error: 0.6123 - y2_output_root_mean_squared_error: 1.0895 - val_loss: 1.5902 - val_y1_output_loss: 0.4173 - val_y2_output_loss: 1.1719 - val_y1_output_root_mean_squared_error: 0.6357 - val_y2_output_root_mean_squared_error: 1.0891\n",
      "Epoch 157/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 1.5589 - y1_output_loss: 0.3614 - y2_output_loss: 1.1894 - y1_output_root_mean_squared_error: 0.6034 - y2_output_root_mean_squared_error: 1.0931 - val_loss: 1.7482 - val_y1_output_loss: 0.3876 - val_y2_output_loss: 1.3358 - val_y1_output_root_mean_squared_error: 0.6241 - val_y2_output_root_mean_squared_error: 1.1656\n",
      "Epoch 158/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 1.7401 - y1_output_loss: 0.3790 - y2_output_loss: 1.3541 - y1_output_root_mean_squared_error: 0.6176 - y2_output_root_mean_squared_error: 1.1656 - val_loss: 1.8771 - val_y1_output_loss: 0.3575 - val_y2_output_loss: 1.5315 - val_y1_output_root_mean_squared_error: 0.5897 - val_y2_output_root_mean_squared_error: 1.2366\n",
      "Epoch 159/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 2.3399 - y1_output_loss: 0.5067 - y2_output_loss: 1.8421 - y1_output_root_mean_squared_error: 0.7117 - y2_output_root_mean_squared_error: 1.3540 - val_loss: 4.2810 - val_y1_output_loss: 0.6151 - val_y2_output_loss: 3.7889 - val_y1_output_root_mean_squared_error: 0.7709 - val_y2_output_root_mean_squared_error: 1.9201\n",
      "Epoch 160/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 2.2691 - y1_output_loss: 0.5106 - y2_output_loss: 1.7403 - y1_output_root_mean_squared_error: 0.7172 - y2_output_root_mean_squared_error: 1.3247 - val_loss: 1.7012 - val_y1_output_loss: 0.4030 - val_y2_output_loss: 1.2699 - val_y1_output_root_mean_squared_error: 0.6267 - val_y2_output_root_mean_squared_error: 1.1439\n",
      "Epoch 161/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 1.9284 - y1_output_loss: 0.4040 - y2_output_loss: 1.5243 - y1_output_root_mean_squared_error: 0.6327 - y2_output_root_mean_squared_error: 1.2362 - val_loss: 3.7307 - val_y1_output_loss: 1.1164 - val_y2_output_loss: 2.6300 - val_y1_output_root_mean_squared_error: 1.0677 - val_y2_output_root_mean_squared_error: 1.6096\n",
      "Epoch 162/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 2.1824 - y1_output_loss: 0.4843 - y2_output_loss: 1.6985 - y1_output_root_mean_squared_error: 0.6976 - y2_output_root_mean_squared_error: 1.3022 - val_loss: 1.5496 - val_y1_output_loss: 0.3186 - val_y2_output_loss: 1.2417 - val_y1_output_root_mean_squared_error: 0.5491 - val_y2_output_root_mean_squared_error: 1.1172\n",
      "Epoch 163/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 1.3592 - y1_output_loss: 0.3062 - y2_output_loss: 1.0420 - y1_output_root_mean_squared_error: 0.5553 - y2_output_root_mean_squared_error: 1.0251 - val_loss: 1.6087 - val_y1_output_loss: 0.3840 - val_y2_output_loss: 1.2148 - val_y1_output_root_mean_squared_error: 0.6111 - val_y2_output_root_mean_squared_error: 1.1114\n",
      "Epoch 164/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 1.8625 - y1_output_loss: 0.3556 - y2_output_loss: 1.4924 - y1_output_root_mean_squared_error: 0.5984 - y2_output_root_mean_squared_error: 1.2266 - val_loss: 1.5434 - val_y1_output_loss: 0.3452 - val_y2_output_loss: 1.1845 - val_y1_output_root_mean_squared_error: 0.5731 - val_y2_output_root_mean_squared_error: 1.1022\n",
      "Epoch 165/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 1.4845 - y1_output_loss: 0.3307 - y2_output_loss: 1.2134 - y1_output_root_mean_squared_error: 0.5759 - y2_output_root_mean_squared_error: 1.0737 - val_loss: 11.0526 - val_y1_output_loss: 0.7863 - val_y2_output_loss: 10.5961 - val_y1_output_root_mean_squared_error: 0.8812 - val_y2_output_root_mean_squared_error: 3.2056\n",
      "Epoch 166/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 1.7768 - y1_output_loss: 0.3953 - y2_output_loss: 1.3700 - y1_output_root_mean_squared_error: 0.6302 - y2_output_root_mean_squared_error: 1.1746 - val_loss: 1.5199 - val_y1_output_loss: 0.2934 - val_y2_output_loss: 1.2038 - val_y1_output_root_mean_squared_error: 0.5283 - val_y2_output_root_mean_squared_error: 1.1139\n",
      "Epoch 167/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 2.0721 - y1_output_loss: 0.4473 - y2_output_loss: 1.6173 - y1_output_root_mean_squared_error: 0.6700 - y2_output_root_mean_squared_error: 1.2741 - val_loss: 2.3627 - val_y1_output_loss: 0.4922 - val_y2_output_loss: 1.9153 - val_y1_output_root_mean_squared_error: 0.7035 - val_y2_output_root_mean_squared_error: 1.3666\n",
      "Epoch 168/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 1.3922 - y1_output_loss: 0.3213 - y2_output_loss: 1.0675 - y1_output_root_mean_squared_error: 0.5663 - y2_output_root_mean_squared_error: 1.0351 - val_loss: 1.6013 - val_y1_output_loss: 0.4140 - val_y2_output_loss: 1.1851 - val_y1_output_root_mean_squared_error: 0.6260 - val_y2_output_root_mean_squared_error: 1.0997\n",
      "Epoch 169/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 1.4180 - y1_output_loss: 0.3212 - y2_output_loss: 1.1144 - y1_output_root_mean_squared_error: 0.5662 - y2_output_root_mean_squared_error: 1.0476 - val_loss: 1.8267 - val_y1_output_loss: 0.4343 - val_y2_output_loss: 1.3935 - val_y1_output_root_mean_squared_error: 0.6605 - val_y2_output_root_mean_squared_error: 1.1791\n",
      "Epoch 170/500\n",
      "614/614 [==============================] - 0s 132us/sample - loss: 1.6686 - y1_output_loss: 0.4023 - y2_output_loss: 1.2854 - y1_output_root_mean_squared_error: 0.6142 - y2_output_root_mean_squared_error: 1.1363 - val_loss: 2.5734 - val_y1_output_loss: 0.8635 - val_y2_output_loss: 1.7288 - val_y1_output_root_mean_squared_error: 0.9225 - val_y2_output_root_mean_squared_error: 1.3124\n",
      "Epoch 171/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 2.0874 - y1_output_loss: 0.4685 - y2_output_loss: 1.6464 - y1_output_root_mean_squared_error: 0.6839 - y2_output_root_mean_squared_error: 1.2727 - val_loss: 10.0099 - val_y1_output_loss: 2.3607 - val_y2_output_loss: 7.8172 - val_y1_output_root_mean_squared_error: 1.4916 - val_y2_output_root_mean_squared_error: 2.7902\n",
      "Epoch 172/500\n",
      "614/614 [==============================] - 0s 131us/sample - loss: 2.2594 - y1_output_loss: 0.5436 - y2_output_loss: 1.6972 - y1_output_root_mean_squared_error: 0.7392 - y2_output_root_mean_squared_error: 1.3088 - val_loss: 1.4702 - val_y1_output_loss: 0.3354 - val_y2_output_loss: 1.1293 - val_y1_output_root_mean_squared_error: 0.5766 - val_y2_output_root_mean_squared_error: 1.0666\n",
      "Epoch 173/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 1.5137 - y1_output_loss: 0.2851 - y2_output_loss: 1.2175 - y1_output_root_mean_squared_error: 0.5358 - y2_output_root_mean_squared_error: 1.1075 - val_loss: 1.5197 - val_y1_output_loss: 0.3678 - val_y2_output_loss: 1.1776 - val_y1_output_root_mean_squared_error: 0.5910 - val_y2_output_root_mean_squared_error: 1.0819\n",
      "Epoch 174/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 1.7926 - y1_output_loss: 0.3748 - y2_output_loss: 1.4155 - y1_output_root_mean_squared_error: 0.6132 - y2_output_root_mean_squared_error: 1.1902 - val_loss: 1.4536 - val_y1_output_loss: 0.3064 - val_y2_output_loss: 1.1412 - val_y1_output_root_mean_squared_error: 0.5460 - val_y2_output_root_mean_squared_error: 1.0750\n",
      "Epoch 175/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 1.5096 - y1_output_loss: 0.3011 - y2_output_loss: 1.2114 - y1_output_root_mean_squared_error: 0.5499 - y2_output_root_mean_squared_error: 1.0987 - val_loss: 2.4684 - val_y1_output_loss: 0.4351 - val_y2_output_loss: 2.1080 - val_y1_output_root_mean_squared_error: 0.6640 - val_y2_output_root_mean_squared_error: 1.4239\n",
      "Epoch 176/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 1.9584 - y1_output_loss: 0.4197 - y2_output_loss: 1.5552 - y1_output_root_mean_squared_error: 0.6484 - y2_output_root_mean_squared_error: 1.2402 - val_loss: 2.7792 - val_y1_output_loss: 0.4716 - val_y2_output_loss: 2.2628 - val_y1_output_root_mean_squared_error: 0.6703 - val_y2_output_root_mean_squared_error: 1.5264\n",
      "Epoch 177/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 4.1718 - y1_output_loss: 1.2475 - y2_output_loss: 2.9086 - y1_output_root_mean_squared_error: 1.1187 - y2_output_root_mean_squared_error: 1.7089 - val_loss: 4.6496 - val_y1_output_loss: 1.0879 - val_y2_output_loss: 3.5242 - val_y1_output_root_mean_squared_error: 1.0214 - val_y2_output_root_mean_squared_error: 1.8990\n",
      "Epoch 178/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 1.4262 - y1_output_loss: 0.3146 - y2_output_loss: 1.1000 - y1_output_root_mean_squared_error: 0.5630 - y2_output_root_mean_squared_error: 1.0532 - val_loss: 1.4672 - val_y1_output_loss: 0.2856 - val_y2_output_loss: 1.1701 - val_y1_output_root_mean_squared_error: 0.5257 - val_y2_output_root_mean_squared_error: 1.0912\n",
      "Epoch 179/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 1.7261 - y1_output_loss: 0.3828 - y2_output_loss: 1.3299 - y1_output_root_mean_squared_error: 0.6201 - y2_output_root_mean_squared_error: 1.1582 - val_loss: 1.3784 - val_y1_output_loss: 0.3233 - val_y2_output_loss: 1.0529 - val_y1_output_root_mean_squared_error: 0.5545 - val_y2_output_root_mean_squared_error: 1.0349\n",
      "Epoch 180/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 1.2568 - y1_output_loss: 0.2797 - y2_output_loss: 0.9840 - y1_output_root_mean_squared_error: 0.5305 - y2_output_root_mean_squared_error: 0.9876 - val_loss: 4.0834 - val_y1_output_loss: 0.7695 - val_y2_output_loss: 3.4356 - val_y1_output_root_mean_squared_error: 0.8722 - val_y2_output_root_mean_squared_error: 1.8228\n",
      "Epoch 181/500\n",
      "614/614 [==============================] - 0s 132us/sample - loss: 1.6844 - y1_output_loss: 0.3377 - y2_output_loss: 1.3439 - y1_output_root_mean_squared_error: 0.5823 - y2_output_root_mean_squared_error: 1.1599 - val_loss: 1.6714 - val_y1_output_loss: 0.4547 - val_y2_output_loss: 1.2366 - val_y1_output_root_mean_squared_error: 0.6642 - val_y2_output_root_mean_squared_error: 1.1092\n",
      "Epoch 182/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 2.8073 - y1_output_loss: 0.6328 - y2_output_loss: 2.1497 - y1_output_root_mean_squared_error: 0.7989 - y2_output_root_mean_squared_error: 1.4728 - val_loss: 1.7459 - val_y1_output_loss: 0.3898 - val_y2_output_loss: 1.3828 - val_y1_output_root_mean_squared_error: 0.6151 - val_y2_output_root_mean_squared_error: 1.1694\n",
      "Epoch 183/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 1.8778 - y1_output_loss: 0.4644 - y2_output_loss: 1.3968 - y1_output_root_mean_squared_error: 0.6842 - y2_output_root_mean_squared_error: 1.1873 - val_loss: 1.9779 - val_y1_output_loss: 0.5946 - val_y2_output_loss: 1.3710 - val_y1_output_root_mean_squared_error: 0.7484 - val_y2_output_root_mean_squared_error: 1.1907\n",
      "Epoch 184/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 1.6648 - y1_output_loss: 0.3743 - y2_output_loss: 1.2772 - y1_output_root_mean_squared_error: 0.6144 - y2_output_root_mean_squared_error: 1.1346 - val_loss: 1.8769 - val_y1_output_loss: 0.3809 - val_y2_output_loss: 1.5660 - val_y1_output_root_mean_squared_error: 0.6046 - val_y2_output_root_mean_squared_error: 1.2294\n",
      "Epoch 185/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 1.5662 - y1_output_loss: 0.3465 - y2_output_loss: 1.2306 - y1_output_root_mean_squared_error: 0.5885 - y2_output_root_mean_squared_error: 1.1045 - val_loss: 5.7901 - val_y1_output_loss: 1.5347 - val_y2_output_loss: 4.3623 - val_y1_output_root_mean_squared_error: 1.1995 - val_y2_output_root_mean_squared_error: 2.0860\n",
      "Epoch 186/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 1.4243 - y1_output_loss: 0.3226 - y2_output_loss: 1.1394 - y1_output_root_mean_squared_error: 0.5680 - y2_output_root_mean_squared_error: 1.0496 - val_loss: 8.2518 - val_y1_output_loss: 1.2600 - val_y2_output_loss: 7.3200 - val_y1_output_root_mean_squared_error: 1.0978 - val_y2_output_root_mean_squared_error: 2.6545\n",
      "Epoch 187/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 1.3872 - y1_output_loss: 0.3066 - y2_output_loss: 1.0777 - y1_output_root_mean_squared_error: 0.5556 - y2_output_root_mean_squared_error: 1.0385 - val_loss: 5.1943 - val_y1_output_loss: 0.6098 - val_y2_output_loss: 4.7821 - val_y1_output_root_mean_squared_error: 0.7766 - val_y2_output_root_mean_squared_error: 2.1427\n",
      "Epoch 188/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 1.8601 - y1_output_loss: 0.3868 - y2_output_loss: 1.4588 - y1_output_root_mean_squared_error: 0.6230 - y2_output_root_mean_squared_error: 1.2132 - val_loss: 1.6512 - val_y1_output_loss: 0.3659 - val_y2_output_loss: 1.2813 - val_y1_output_root_mean_squared_error: 0.5932 - val_y2_output_root_mean_squared_error: 1.1399\n",
      "Epoch 189/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 1.3338 - y1_output_loss: 0.2847 - y2_output_loss: 1.0457 - y1_output_root_mean_squared_error: 0.5351 - y2_output_root_mean_squared_error: 1.0235 - val_loss: 1.9169 - val_y1_output_loss: 0.3853 - val_y2_output_loss: 1.5462 - val_y1_output_root_mean_squared_error: 0.6124 - val_y2_output_root_mean_squared_error: 1.2417\n",
      "Epoch 190/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 1.5777 - y1_output_loss: 0.3333 - y2_output_loss: 1.2376 - y1_output_root_mean_squared_error: 0.5748 - y2_output_root_mean_squared_error: 1.1168 - val_loss: 1.9127 - val_y1_output_loss: 0.3992 - val_y2_output_loss: 1.5437 - val_y1_output_root_mean_squared_error: 0.6323 - val_y2_output_root_mean_squared_error: 1.2300\n",
      "Epoch 191/500\n",
      "614/614 [==============================] - 0s 135us/sample - loss: 1.7352 - y1_output_loss: 0.3694 - y2_output_loss: 1.3558 - y1_output_root_mean_squared_error: 0.6085 - y2_output_root_mean_squared_error: 1.1683 - val_loss: 1.6961 - val_y1_output_loss: 0.3880 - val_y2_output_loss: 1.2839 - val_y1_output_root_mean_squared_error: 0.6042 - val_y2_output_root_mean_squared_error: 1.1537\n",
      "Epoch 192/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 1.3757 - y1_output_loss: 0.2974 - y2_output_loss: 1.0735 - y1_output_root_mean_squared_error: 0.5417 - y2_output_root_mean_squared_error: 1.0403 - val_loss: 1.6862 - val_y1_output_loss: 0.4168 - val_y2_output_loss: 1.2558 - val_y1_output_root_mean_squared_error: 0.6423 - val_y2_output_root_mean_squared_error: 1.1285\n",
      "Epoch 193/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 1.2497 - y1_output_loss: 0.2861 - y2_output_loss: 0.9698 - y1_output_root_mean_squared_error: 0.5349 - y2_output_root_mean_squared_error: 0.9816 - val_loss: 2.8252 - val_y1_output_loss: 0.6938 - val_y2_output_loss: 2.1141 - val_y1_output_root_mean_squared_error: 0.8030 - val_y2_output_root_mean_squared_error: 1.4766\n",
      "Epoch 194/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 1.2692 - y1_output_loss: 0.2468 - y2_output_loss: 1.0118 - y1_output_root_mean_squared_error: 0.4985 - y2_output_root_mean_squared_error: 1.0103 - val_loss: 1.1773 - val_y1_output_loss: 0.2783 - val_y2_output_loss: 0.9101 - val_y1_output_root_mean_squared_error: 0.5111 - val_y2_output_root_mean_squared_error: 0.9571\n",
      "Epoch 195/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 2.3147 - y1_output_loss: 0.5386 - y2_output_loss: 1.8057 - y1_output_root_mean_squared_error: 0.7261 - y2_output_root_mean_squared_error: 1.3370 - val_loss: 4.9764 - val_y1_output_loss: 2.0013 - val_y2_output_loss: 2.9558 - val_y1_output_root_mean_squared_error: 1.3967 - val_y2_output_root_mean_squared_error: 1.7394\n",
      "Epoch 196/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 1.8242 - y1_output_loss: 0.4251 - y2_output_loss: 1.5646 - y1_output_root_mean_squared_error: 0.6380 - y2_output_root_mean_squared_error: 1.1904 - val_loss: 64.8562 - val_y1_output_loss: 17.8806 - val_y2_output_loss: 49.4317 - val_y1_output_root_mean_squared_error: 4.1532 - val_y2_output_root_mean_squared_error: 6.8998\n",
      "Epoch 197/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 3.9468 - y1_output_loss: 0.9441 - y2_output_loss: 2.9776 - y1_output_root_mean_squared_error: 0.9757 - y2_output_root_mean_squared_error: 1.7305 - val_loss: 2.7421 - val_y1_output_loss: 0.4820 - val_y2_output_loss: 2.2278 - val_y1_output_root_mean_squared_error: 0.6775 - val_y2_output_root_mean_squared_error: 1.5110\n",
      "Epoch 198/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 1.1370 - y1_output_loss: 0.2866 - y2_output_loss: 0.8464 - y1_output_root_mean_squared_error: 0.5323 - y2_output_root_mean_squared_error: 0.9239 - val_loss: 1.3218 - val_y1_output_loss: 0.3567 - val_y2_output_loss: 0.9780 - val_y1_output_root_mean_squared_error: 0.5818 - val_y2_output_root_mean_squared_error: 0.9916\n",
      "Epoch 199/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 1.1029 - y1_output_loss: 0.2454 - y2_output_loss: 0.8558 - y1_output_root_mean_squared_error: 0.4976 - y2_output_root_mean_squared_error: 0.9248 - val_loss: 1.8040 - val_y1_output_loss: 0.3802 - val_y2_output_loss: 1.4010 - val_y1_output_root_mean_squared_error: 0.5913 - val_y2_output_root_mean_squared_error: 1.2060\n",
      "Epoch 200/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 1.8182 - y1_output_loss: 0.3851 - y2_output_loss: 1.4216 - y1_output_root_mean_squared_error: 0.6230 - y2_output_root_mean_squared_error: 1.1959 - val_loss: 1.6189 - val_y1_output_loss: 0.3438 - val_y2_output_loss: 1.2685 - val_y1_output_root_mean_squared_error: 0.5861 - val_y2_output_root_mean_squared_error: 1.1293\n",
      "Epoch 201/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 1.1810 - y1_output_loss: 0.2864 - y2_output_loss: 0.8851 - y1_output_root_mean_squared_error: 0.5370 - y2_output_root_mean_squared_error: 0.9448 - val_loss: 1.3760 - val_y1_output_loss: 0.3686 - val_y2_output_loss: 1.0108 - val_y1_output_root_mean_squared_error: 0.5906 - val_y2_output_root_mean_squared_error: 1.0135\n",
      "Epoch 202/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 1.3046 - y1_output_loss: 0.2828 - y2_output_loss: 1.0299 - y1_output_root_mean_squared_error: 0.5341 - y2_output_root_mean_squared_error: 1.0096 - val_loss: 3.0976 - val_y1_output_loss: 0.4348 - val_y2_output_loss: 2.6238 - val_y1_output_root_mean_squared_error: 0.6366 - val_y2_output_root_mean_squared_error: 1.6408\n",
      "Epoch 203/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 1.8658 - y1_output_loss: 0.3735 - y2_output_loss: 1.5135 - y1_output_root_mean_squared_error: 0.6081 - y2_output_root_mean_squared_error: 1.2231 - val_loss: 1.4568 - val_y1_output_loss: 0.2980 - val_y2_output_loss: 1.1244 - val_y1_output_root_mean_squared_error: 0.5423 - val_y2_output_root_mean_squared_error: 1.0783\n",
      "Epoch 204/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 1.2056 - y1_output_loss: 0.2623 - y2_output_loss: 0.9428 - y1_output_root_mean_squared_error: 0.5133 - y2_output_root_mean_squared_error: 0.9706 - val_loss: 1.6922 - val_y1_output_loss: 0.3430 - val_y2_output_loss: 1.3988 - val_y1_output_root_mean_squared_error: 0.5684 - val_y2_output_root_mean_squared_error: 1.1701\n",
      "Epoch 205/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 1.4157 - y1_output_loss: 0.3137 - y2_output_loss: 1.1469 - y1_output_root_mean_squared_error: 0.5603 - y2_output_root_mean_squared_error: 1.0496 - val_loss: 14.3190 - val_y1_output_loss: 3.3942 - val_y2_output_loss: 11.5290 - val_y1_output_root_mean_squared_error: 1.8094 - val_y2_output_root_mean_squared_error: 3.3234\n",
      "Epoch 206/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 2.3123 - y1_output_loss: 0.5430 - y2_output_loss: 1.7530 - y1_output_root_mean_squared_error: 0.7395 - y2_output_root_mean_squared_error: 1.3287 - val_loss: 1.5936 - val_y1_output_loss: 0.4924 - val_y2_output_loss: 1.1507 - val_y1_output_root_mean_squared_error: 0.6843 - val_y2_output_root_mean_squared_error: 1.0608\n",
      "Epoch 207/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 1.1585 - y1_output_loss: 0.2567 - y2_output_loss: 0.8957 - y1_output_root_mean_squared_error: 0.5086 - y2_output_root_mean_squared_error: 0.9486 - val_loss: 1.7757 - val_y1_output_loss: 0.4544 - val_y2_output_loss: 1.3188 - val_y1_output_root_mean_squared_error: 0.6480 - val_y2_output_root_mean_squared_error: 1.1644\n",
      "Epoch 208/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 1.6891 - y1_output_loss: 0.3073 - y2_output_loss: 1.3765 - y1_output_root_mean_squared_error: 0.5560 - y2_output_root_mean_squared_error: 1.1747 - val_loss: 1.9107 - val_y1_output_loss: 0.2981 - val_y2_output_loss: 1.6805 - val_y1_output_root_mean_squared_error: 0.5451 - val_y2_output_root_mean_squared_error: 1.2703\n",
      "Epoch 209/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 1.1968 - y1_output_loss: 0.2798 - y2_output_loss: 0.9240 - y1_output_root_mean_squared_error: 0.5264 - y2_output_root_mean_squared_error: 0.9590 - val_loss: 2.9170 - val_y1_output_loss: 0.9858 - val_y2_output_loss: 1.9446 - val_y1_output_root_mean_squared_error: 0.9591 - val_y2_output_root_mean_squared_error: 1.4132\n",
      "Epoch 210/500\n",
      "614/614 [==============================] - 0s 134us/sample - loss: 1.1901 - y1_output_loss: 0.2792 - y2_output_loss: 0.9087 - y1_output_root_mean_squared_error: 0.5286 - y2_output_root_mean_squared_error: 0.9543 - val_loss: 1.6019 - val_y1_output_loss: 0.3296 - val_y2_output_loss: 1.3015 - val_y1_output_root_mean_squared_error: 0.5585 - val_y2_output_root_mean_squared_error: 1.1358\n",
      "Epoch 211/500\n",
      "614/614 [==============================] - 0s 149us/sample - loss: 1.3552 - y1_output_loss: 0.2878 - y2_output_loss: 1.0587 - y1_output_root_mean_squared_error: 0.5373 - y2_output_root_mean_squared_error: 1.0327 - val_loss: 1.3794 - val_y1_output_loss: 0.3065 - val_y2_output_loss: 1.0571 - val_y1_output_root_mean_squared_error: 0.5389 - val_y2_output_root_mean_squared_error: 1.0435\n",
      "Epoch 212/500\n",
      "614/614 [==============================] - 0s 131us/sample - loss: 1.1712 - y1_output_loss: 0.2910 - y2_output_loss: 0.8833 - y1_output_root_mean_squared_error: 0.5406 - y2_output_root_mean_squared_error: 0.9375 - val_loss: 2.8002 - val_y1_output_loss: 0.6009 - val_y2_output_loss: 2.1689 - val_y1_output_root_mean_squared_error: 0.7523 - val_y2_output_root_mean_squared_error: 1.4948\n",
      "Epoch 213/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 1.1967 - y1_output_loss: 0.2921 - y2_output_loss: 0.8956 - y1_output_root_mean_squared_error: 0.5417 - y2_output_root_mean_squared_error: 0.9504 - val_loss: 1.4598 - val_y1_output_loss: 0.4217 - val_y2_output_loss: 1.0418 - val_y1_output_root_mean_squared_error: 0.6299 - val_y2_output_root_mean_squared_error: 1.0310\n",
      "Epoch 214/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 1.0077 - y1_output_loss: 0.2334 - y2_output_loss: 0.7689 - y1_output_root_mean_squared_error: 0.4844 - y2_output_root_mean_squared_error: 0.8792 - val_loss: 2.3322 - val_y1_output_loss: 0.4119 - val_y2_output_loss: 1.9944 - val_y1_output_root_mean_squared_error: 0.6416 - val_y2_output_root_mean_squared_error: 1.3858\n",
      "Epoch 215/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 1.1950 - y1_output_loss: 0.2836 - y2_output_loss: 0.9018 - y1_output_root_mean_squared_error: 0.5343 - y2_output_root_mean_squared_error: 0.9537 - val_loss: 1.7094 - val_y1_output_loss: 0.3086 - val_y2_output_loss: 1.4592 - val_y1_output_root_mean_squared_error: 0.5492 - val_y2_output_root_mean_squared_error: 1.1865\n",
      "Epoch 216/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 1.1488 - y1_output_loss: 0.2369 - y2_output_loss: 0.9055 - y1_output_root_mean_squared_error: 0.4875 - y2_output_root_mean_squared_error: 0.9545 - val_loss: 1.6029 - val_y1_output_loss: 0.3370 - val_y2_output_loss: 1.2895 - val_y1_output_root_mean_squared_error: 0.5731 - val_y2_output_root_mean_squared_error: 1.1289\n",
      "Epoch 217/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 1.0589 - y1_output_loss: 0.2349 - y2_output_loss: 0.8439 - y1_output_root_mean_squared_error: 0.4856 - y2_output_root_mean_squared_error: 0.9073 - val_loss: 3.5207 - val_y1_output_loss: 0.3795 - val_y2_output_loss: 3.2695 - val_y1_output_root_mean_squared_error: 0.6180 - val_y2_output_root_mean_squared_error: 1.7717\n",
      "Epoch 218/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 1.4251 - y1_output_loss: 0.3361 - y2_output_loss: 1.0848 - y1_output_root_mean_squared_error: 0.5805 - y2_output_root_mean_squared_error: 1.0431 - val_loss: 2.2111 - val_y1_output_loss: 0.8525 - val_y2_output_loss: 1.3677 - val_y1_output_root_mean_squared_error: 0.9003 - val_y2_output_root_mean_squared_error: 1.1835\n",
      "Epoch 219/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 1.3830 - y1_output_loss: 0.2784 - y2_output_loss: 1.1164 - y1_output_root_mean_squared_error: 0.5290 - y2_output_root_mean_squared_error: 1.0503 - val_loss: 2.3379 - val_y1_output_loss: 0.4833 - val_y2_output_loss: 1.9026 - val_y1_output_root_mean_squared_error: 0.6846 - val_y2_output_root_mean_squared_error: 1.3672\n",
      "Epoch 220/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 1.2230 - y1_output_loss: 0.2663 - y2_output_loss: 0.9652 - y1_output_root_mean_squared_error: 0.5057 - y2_output_root_mean_squared_error: 0.9835 - val_loss: 9.2763 - val_y1_output_loss: 3.0505 - val_y2_output_loss: 6.4387 - val_y1_output_root_mean_squared_error: 1.7007 - val_y2_output_root_mean_squared_error: 2.5266\n",
      "Epoch 221/500\n",
      "614/614 [==============================] - 0s 134us/sample - loss: 1.4621 - y1_output_loss: 0.3255 - y2_output_loss: 1.1260 - y1_output_root_mean_squared_error: 0.5709 - y2_output_root_mean_squared_error: 1.0659 - val_loss: 1.2148 - val_y1_output_loss: 0.2820 - val_y2_output_loss: 0.9360 - val_y1_output_root_mean_squared_error: 0.5265 - val_y2_output_root_mean_squared_error: 0.9683\n",
      "Epoch 222/500\n",
      "614/614 [==============================] - 0s 131us/sample - loss: 1.2543 - y1_output_loss: 0.2900 - y2_output_loss: 0.9549 - y1_output_root_mean_squared_error: 0.5398 - y2_output_root_mean_squared_error: 0.9813 - val_loss: 1.2950 - val_y1_output_loss: 0.2510 - val_y2_output_loss: 1.0203 - val_y1_output_root_mean_squared_error: 0.4937 - val_y2_output_root_mean_squared_error: 1.0253\n",
      "Epoch 223/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 1.0422 - y1_output_loss: 0.2290 - y2_output_loss: 0.8302 - y1_output_root_mean_squared_error: 0.4788 - y2_output_root_mean_squared_error: 0.9016 - val_loss: 3.4738 - val_y1_output_loss: 0.4961 - val_y2_output_loss: 2.9180 - val_y1_output_root_mean_squared_error: 0.6860 - val_y2_output_root_mean_squared_error: 1.7330\n",
      "Epoch 224/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 1.2374 - y1_output_loss: 0.2588 - y2_output_loss: 0.9846 - y1_output_root_mean_squared_error: 0.5084 - y2_output_root_mean_squared_error: 0.9894 - val_loss: 1.9509 - val_y1_output_loss: 0.2774 - val_y2_output_loss: 1.6323 - val_y1_output_root_mean_squared_error: 0.5090 - val_y2_output_root_mean_squared_error: 1.3007\n",
      "Epoch 225/500\n",
      "614/614 [==============================] - 0s 137us/sample - loss: 1.1482 - y1_output_loss: 0.2747 - y2_output_loss: 0.8890 - y1_output_root_mean_squared_error: 0.5235 - y2_output_root_mean_squared_error: 0.9349 - val_loss: 2.7476 - val_y1_output_loss: 0.2969 - val_y2_output_loss: 2.5824 - val_y1_output_root_mean_squared_error: 0.5369 - val_y2_output_root_mean_squared_error: 1.5682\n",
      "Epoch 226/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 1.1488 - y1_output_loss: 0.2349 - y2_output_loss: 0.9047 - y1_output_root_mean_squared_error: 0.4863 - y2_output_root_mean_squared_error: 0.9552 - val_loss: 1.2733 - val_y1_output_loss: 0.2761 - val_y2_output_loss: 0.9865 - val_y1_output_root_mean_squared_error: 0.5107 - val_y2_output_root_mean_squared_error: 1.0062\n",
      "Epoch 227/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 1.5307 - y1_output_loss: 0.3008 - y2_output_loss: 1.2245 - y1_output_root_mean_squared_error: 0.5507 - y2_output_root_mean_squared_error: 1.1079 - val_loss: 1.5938 - val_y1_output_loss: 0.3666 - val_y2_output_loss: 1.2707 - val_y1_output_root_mean_squared_error: 0.6048 - val_y2_output_root_mean_squared_error: 1.1081\n",
      "Epoch 228/500\n",
      "614/614 [==============================] - 0s 132us/sample - loss: 0.9324 - y1_output_loss: 0.2052 - y2_output_loss: 0.7413 - y1_output_root_mean_squared_error: 0.4544 - y2_output_root_mean_squared_error: 0.8520 - val_loss: 1.9867 - val_y1_output_loss: 0.4162 - val_y2_output_loss: 1.5527 - val_y1_output_root_mean_squared_error: 0.6281 - val_y2_output_root_mean_squared_error: 1.2618\n",
      "Epoch 229/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 1.1289 - y1_output_loss: 0.2419 - y2_output_loss: 0.8784 - y1_output_root_mean_squared_error: 0.4935 - y2_output_root_mean_squared_error: 0.9409 - val_loss: 1.3216 - val_y1_output_loss: 0.2593 - val_y2_output_loss: 1.0791 - val_y1_output_root_mean_squared_error: 0.5026 - val_y2_output_root_mean_squared_error: 1.0339\n",
      "Epoch 230/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 1.0034 - y1_output_loss: 0.2381 - y2_output_loss: 0.7606 - y1_output_root_mean_squared_error: 0.4874 - y2_output_root_mean_squared_error: 0.8751 - val_loss: 1.5296 - val_y1_output_loss: 0.4725 - val_y2_output_loss: 1.0669 - val_y1_output_root_mean_squared_error: 0.6840 - val_y2_output_root_mean_squared_error: 1.0304\n",
      "Epoch 231/500\n",
      "614/614 [==============================] - 0s 133us/sample - loss: 0.9802 - y1_output_loss: 0.2129 - y2_output_loss: 0.7605 - y1_output_root_mean_squared_error: 0.4617 - y2_output_root_mean_squared_error: 0.8758 - val_loss: 1.5510 - val_y1_output_loss: 0.3851 - val_y2_output_loss: 1.1514 - val_y1_output_root_mean_squared_error: 0.6055 - val_y2_output_root_mean_squared_error: 1.0883\n",
      "Epoch 232/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 1.1376 - y1_output_loss: 0.2475 - y2_output_loss: 0.8872 - y1_output_root_mean_squared_error: 0.4951 - y2_output_root_mean_squared_error: 0.9447 - val_loss: 1.5584 - val_y1_output_loss: 0.4912 - val_y2_output_loss: 1.0655 - val_y1_output_root_mean_squared_error: 0.6775 - val_y2_output_root_mean_squared_error: 1.0486\n",
      "Epoch 233/500\n",
      "614/614 [==============================] - 0s 131us/sample - loss: 0.9533 - y1_output_loss: 0.2277 - y2_output_loss: 0.7267 - y1_output_root_mean_squared_error: 0.4771 - y2_output_root_mean_squared_error: 0.8518 - val_loss: 2.1942 - val_y1_output_loss: 0.4347 - val_y2_output_loss: 1.7520 - val_y1_output_root_mean_squared_error: 0.6375 - val_y2_output_root_mean_squared_error: 1.3371\n",
      "Epoch 234/500\n",
      "614/614 [==============================] - 0s 132us/sample - loss: 1.2126 - y1_output_loss: 0.2890 - y2_output_loss: 0.9158 - y1_output_root_mean_squared_error: 0.5400 - y2_output_root_mean_squared_error: 0.9597 - val_loss: 1.5887 - val_y1_output_loss: 0.3695 - val_y2_output_loss: 1.2041 - val_y1_output_root_mean_squared_error: 0.5955 - val_y2_output_root_mean_squared_error: 1.1108\n",
      "Epoch 235/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 1.1707 - y1_output_loss: 0.2777 - y2_output_loss: 0.8856 - y1_output_root_mean_squared_error: 0.5275 - y2_output_root_mean_squared_error: 0.9447 - val_loss: 1.4474 - val_y1_output_loss: 0.3801 - val_y2_output_loss: 1.0953 - val_y1_output_root_mean_squared_error: 0.6151 - val_y2_output_root_mean_squared_error: 1.0340\n",
      "Epoch 236/500\n",
      "614/614 [==============================] - 0s 136us/sample - loss: 1.0411 - y1_output_loss: 0.2800 - y2_output_loss: 0.7549 - y1_output_root_mean_squared_error: 0.5317 - y2_output_root_mean_squared_error: 0.8708 - val_loss: 1.4424 - val_y1_output_loss: 0.2837 - val_y2_output_loss: 1.1985 - val_y1_output_root_mean_squared_error: 0.5270 - val_y2_output_root_mean_squared_error: 1.0792\n",
      "Epoch 237/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 0.9015 - y1_output_loss: 0.2198 - y2_output_loss: 0.6819 - y1_output_root_mean_squared_error: 0.4691 - y2_output_root_mean_squared_error: 0.8255 - val_loss: 1.6039 - val_y1_output_loss: 0.4455 - val_y2_output_loss: 1.1568 - val_y1_output_root_mean_squared_error: 0.6460 - val_y2_output_root_mean_squared_error: 1.0893\n",
      "Epoch 238/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 1.2401 - y1_output_loss: 0.3072 - y2_output_loss: 0.9243 - y1_output_root_mean_squared_error: 0.5567 - y2_output_root_mean_squared_error: 0.9645 - val_loss: 1.4472 - val_y1_output_loss: 0.3433 - val_y2_output_loss: 1.0864 - val_y1_output_root_mean_squared_error: 0.5718 - val_y2_output_root_mean_squared_error: 1.0584\n",
      "Epoch 239/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 1.1433 - y1_output_loss: 0.2747 - y2_output_loss: 0.8665 - y1_output_root_mean_squared_error: 0.5234 - y2_output_root_mean_squared_error: 0.9323 - val_loss: 1.3278 - val_y1_output_loss: 0.3737 - val_y2_output_loss: 0.9520 - val_y1_output_root_mean_squared_error: 0.5888 - val_y2_output_root_mean_squared_error: 0.9905\n",
      "Epoch 240/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 0.8894 - y1_output_loss: 0.2045 - y2_output_loss: 0.6886 - y1_output_root_mean_squared_error: 0.4521 - y2_output_root_mean_squared_error: 0.8276 - val_loss: 1.2809 - val_y1_output_loss: 0.2625 - val_y2_output_loss: 1.0417 - val_y1_output_root_mean_squared_error: 0.5053 - val_y2_output_root_mean_squared_error: 1.0127\n",
      "Epoch 241/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 1.5268 - y1_output_loss: 0.3107 - y2_output_loss: 1.2092 - y1_output_root_mean_squared_error: 0.5594 - y2_output_root_mean_squared_error: 1.1018 - val_loss: 1.4065 - val_y1_output_loss: 0.4868 - val_y2_output_loss: 0.9292 - val_y1_output_root_mean_squared_error: 0.6765 - val_y2_output_root_mean_squared_error: 0.9741\n",
      "Epoch 242/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 0.8503 - y1_output_loss: 0.2170 - y2_output_loss: 0.6351 - y1_output_root_mean_squared_error: 0.4658 - y2_output_root_mean_squared_error: 0.7958 - val_loss: 1.2422 - val_y1_output_loss: 0.3374 - val_y2_output_loss: 0.9162 - val_y1_output_root_mean_squared_error: 0.5591 - val_y2_output_root_mean_squared_error: 0.9642\n",
      "Epoch 243/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.9859 - y1_output_loss: 0.2248 - y2_output_loss: 0.7540 - y1_output_root_mean_squared_error: 0.4757 - y2_output_root_mean_squared_error: 0.8715 - val_loss: 1.0992 - val_y1_output_loss: 0.2428 - val_y2_output_loss: 0.8534 - val_y1_output_root_mean_squared_error: 0.4829 - val_y2_output_root_mean_squared_error: 0.9306\n",
      "Epoch 244/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 1.2008 - y1_output_loss: 0.2694 - y2_output_loss: 0.9225 - y1_output_root_mean_squared_error: 0.5201 - y2_output_root_mean_squared_error: 0.9645 - val_loss: 1.2922 - val_y1_output_loss: 0.4029 - val_y2_output_loss: 0.8949 - val_y1_output_root_mean_squared_error: 0.6133 - val_y2_output_root_mean_squared_error: 0.9571\n",
      "Epoch 245/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 1.0011 - y1_output_loss: 0.2177 - y2_output_loss: 0.7750 - y1_output_root_mean_squared_error: 0.4684 - y2_output_root_mean_squared_error: 0.8841 - val_loss: 1.1194 - val_y1_output_loss: 0.2413 - val_y2_output_loss: 0.8795 - val_y1_output_root_mean_squared_error: 0.4804 - val_y2_output_root_mean_squared_error: 0.9426\n",
      "Epoch 246/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.9457 - y1_output_loss: 0.2113 - y2_output_loss: 0.7291 - y1_output_root_mean_squared_error: 0.4613 - y2_output_root_mean_squared_error: 0.8561 - val_loss: 1.1775 - val_y1_output_loss: 0.3320 - val_y2_output_loss: 0.8547 - val_y1_output_root_mean_squared_error: 0.5592 - val_y2_output_root_mean_squared_error: 0.9299\n",
      "Epoch 247/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 1.0061 - y1_output_loss: 0.2121 - y2_output_loss: 0.8071 - y1_output_root_mean_squared_error: 0.4601 - y2_output_root_mean_squared_error: 0.8913 - val_loss: 1.4135 - val_y1_output_loss: 0.3121 - val_y2_output_loss: 1.0805 - val_y1_output_root_mean_squared_error: 0.5495 - val_y2_output_root_mean_squared_error: 1.0543\n",
      "Epoch 248/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 0.8217 - y1_output_loss: 0.1895 - y2_output_loss: 0.6251 - y1_output_root_mean_squared_error: 0.4373 - y2_output_root_mean_squared_error: 0.7940 - val_loss: 1.1846 - val_y1_output_loss: 0.2393 - val_y2_output_loss: 0.9274 - val_y1_output_root_mean_squared_error: 0.4779 - val_y2_output_root_mean_squared_error: 0.9779\n",
      "Epoch 249/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 1.0518 - y1_output_loss: 0.2486 - y2_output_loss: 0.7975 - y1_output_root_mean_squared_error: 0.5007 - y2_output_root_mean_squared_error: 0.8950 - val_loss: 1.5595 - val_y1_output_loss: 0.3380 - val_y2_output_loss: 1.2362 - val_y1_output_root_mean_squared_error: 0.5787 - val_y2_output_root_mean_squared_error: 1.1066\n",
      "Epoch 250/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.8259 - y1_output_loss: 0.2018 - y2_output_loss: 0.6342 - y1_output_root_mean_squared_error: 0.4417 - y2_output_root_mean_squared_error: 0.7942 - val_loss: 3.5645 - val_y1_output_loss: 0.9142 - val_y2_output_loss: 2.7745 - val_y1_output_root_mean_squared_error: 0.9491 - val_y2_output_root_mean_squared_error: 1.6321\n",
      "Epoch 251/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 1.0730 - y1_output_loss: 0.2364 - y2_output_loss: 0.8313 - y1_output_root_mean_squared_error: 0.4867 - y2_output_root_mean_squared_error: 0.9144 - val_loss: 1.8364 - val_y1_output_loss: 0.3575 - val_y2_output_loss: 1.5213 - val_y1_output_root_mean_squared_error: 0.5977 - val_y2_output_root_mean_squared_error: 1.2162\n",
      "Epoch 252/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 1.0777 - y1_output_loss: 0.2060 - y2_output_loss: 0.8747 - y1_output_root_mean_squared_error: 0.4527 - y2_output_root_mean_squared_error: 0.9342 - val_loss: 3.1768 - val_y1_output_loss: 0.7130 - val_y2_output_loss: 2.5783 - val_y1_output_root_mean_squared_error: 0.8491 - val_y2_output_root_mean_squared_error: 1.5671\n",
      "Epoch 253/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 1.2600 - y1_output_loss: 0.3026 - y2_output_loss: 0.9621 - y1_output_root_mean_squared_error: 0.5490 - y2_output_root_mean_squared_error: 0.9791 - val_loss: 2.4150 - val_y1_output_loss: 0.7554 - val_y2_output_loss: 1.6580 - val_y1_output_root_mean_squared_error: 0.8413 - val_y2_output_root_mean_squared_error: 1.3066\n",
      "Epoch 254/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 1.0518 - y1_output_loss: 0.2438 - y2_output_loss: 0.8038 - y1_output_root_mean_squared_error: 0.4957 - y2_output_root_mean_squared_error: 0.8978 - val_loss: 1.3428 - val_y1_output_loss: 0.3792 - val_y2_output_loss: 0.9654 - val_y1_output_root_mean_squared_error: 0.6021 - val_y2_output_root_mean_squared_error: 0.9901\n",
      "Epoch 255/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.9707 - y1_output_loss: 0.2215 - y2_output_loss: 0.7418 - y1_output_root_mean_squared_error: 0.4723 - y2_output_root_mean_squared_error: 0.8647 - val_loss: 1.0924 - val_y1_output_loss: 0.2571 - val_y2_output_loss: 0.8327 - val_y1_output_root_mean_squared_error: 0.5002 - val_y2_output_root_mean_squared_error: 0.9177\n",
      "Epoch 256/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.9662 - y1_output_loss: 0.2140 - y2_output_loss: 0.7557 - y1_output_root_mean_squared_error: 0.4600 - y2_output_root_mean_squared_error: 0.8687 - val_loss: 2.5152 - val_y1_output_loss: 0.5942 - val_y2_output_loss: 1.9148 - val_y1_output_root_mean_squared_error: 0.7452 - val_y2_output_root_mean_squared_error: 1.3999\n",
      "Epoch 257/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.8648 - y1_output_loss: 0.2492 - y2_output_loss: 0.6235 - y1_output_root_mean_squared_error: 0.5003 - y2_output_root_mean_squared_error: 0.7839 - val_loss: 1.5175 - val_y1_output_loss: 0.2410 - val_y2_output_loss: 1.2569 - val_y1_output_root_mean_squared_error: 0.4775 - val_y2_output_root_mean_squared_error: 1.1356\n",
      "Epoch 258/500\n",
      "614/614 [==============================] - 0s 123us/sample - loss: 0.6969 - y1_output_loss: 0.1779 - y2_output_loss: 0.5136 - y1_output_root_mean_squared_error: 0.4235 - y2_output_root_mean_squared_error: 0.7194 - val_loss: 1.6073 - val_y1_output_loss: 0.3223 - val_y2_output_loss: 1.2735 - val_y1_output_root_mean_squared_error: 0.5458 - val_y2_output_root_mean_squared_error: 1.1443\n",
      "Epoch 259/500\n",
      "614/614 [==============================] - 0s 123us/sample - loss: 0.9065 - y1_output_loss: 0.2059 - y2_output_loss: 0.6939 - y1_output_root_mean_squared_error: 0.4551 - y2_output_root_mean_squared_error: 0.8363 - val_loss: 1.9532 - val_y1_output_loss: 0.4777 - val_y2_output_loss: 1.4696 - val_y1_output_root_mean_squared_error: 0.6677 - val_y2_output_root_mean_squared_error: 1.2277\n",
      "Epoch 260/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.8571 - y1_output_loss: 0.2089 - y2_output_loss: 0.6414 - y1_output_root_mean_squared_error: 0.4588 - y2_output_root_mean_squared_error: 0.8041 - val_loss: 1.2386 - val_y1_output_loss: 0.2739 - val_y2_output_loss: 0.9432 - val_y1_output_root_mean_squared_error: 0.5171 - val_y2_output_root_mean_squared_error: 0.9855\n",
      "Epoch 261/500\n",
      "614/614 [==============================] - 0s 123us/sample - loss: 0.9353 - y1_output_loss: 0.2104 - y2_output_loss: 0.7219 - y1_output_root_mean_squared_error: 0.4601 - y2_output_root_mean_squared_error: 0.8506 - val_loss: 1.5753 - val_y1_output_loss: 0.4050 - val_y2_output_loss: 1.1939 - val_y1_output_root_mean_squared_error: 0.6327 - val_y2_output_root_mean_squared_error: 1.0840\n",
      "Epoch 262/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 1.0113 - y1_output_loss: 0.2145 - y2_output_loss: 0.7884 - y1_output_root_mean_squared_error: 0.4647 - y2_output_root_mean_squared_error: 0.8918 - val_loss: 1.5093 - val_y1_output_loss: 0.2614 - val_y2_output_loss: 1.3075 - val_y1_output_root_mean_squared_error: 0.5008 - val_y2_output_root_mean_squared_error: 1.1218\n",
      "Epoch 263/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 1.0869 - y1_output_loss: 0.2246 - y2_output_loss: 0.8579 - y1_output_root_mean_squared_error: 0.4736 - y2_output_root_mean_squared_error: 0.9288 - val_loss: 1.1386 - val_y1_output_loss: 0.2699 - val_y2_output_loss: 0.8560 - val_y1_output_root_mean_squared_error: 0.5100 - val_y2_output_root_mean_squared_error: 0.9373\n",
      "Epoch 264/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 1.4923 - y1_output_loss: 0.3344 - y2_output_loss: 1.1567 - y1_output_root_mean_squared_error: 0.5766 - y2_output_root_mean_squared_error: 1.0770 - val_loss: 2.8048 - val_y1_output_loss: 0.8128 - val_y2_output_loss: 2.1082 - val_y1_output_root_mean_squared_error: 0.8823 - val_y2_output_root_mean_squared_error: 1.4235\n",
      "Epoch 265/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.8306 - y1_output_loss: 0.2296 - y2_output_loss: 0.6221 - y1_output_root_mean_squared_error: 0.4792 - y2_output_root_mean_squared_error: 0.7753 - val_loss: 4.3426 - val_y1_output_loss: 0.4775 - val_y2_output_loss: 3.8323 - val_y1_output_root_mean_squared_error: 0.6728 - val_y2_output_root_mean_squared_error: 1.9723\n",
      "Epoch 266/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 1.1760 - y1_output_loss: 0.2681 - y2_output_loss: 0.9011 - y1_output_root_mean_squared_error: 0.5192 - y2_output_root_mean_squared_error: 0.9521 - val_loss: 1.5474 - val_y1_output_loss: 0.3949 - val_y2_output_loss: 1.1427 - val_y1_output_root_mean_squared_error: 0.6070 - val_y2_output_root_mean_squared_error: 1.0858\n",
      "Epoch 267/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 1.1356 - y1_output_loss: 0.3092 - y2_output_loss: 0.8182 - y1_output_root_mean_squared_error: 0.5580 - y2_output_root_mean_squared_error: 0.9079 - val_loss: 1.3971 - val_y1_output_loss: 0.3316 - val_y2_output_loss: 1.0531 - val_y1_output_root_mean_squared_error: 0.5556 - val_y2_output_root_mean_squared_error: 1.0432\n",
      "Epoch 268/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 0.8184 - y1_output_loss: 0.2132 - y2_output_loss: 0.6260 - y1_output_root_mean_squared_error: 0.4547 - y2_output_root_mean_squared_error: 0.7821 - val_loss: 4.6897 - val_y1_output_loss: 1.5041 - val_y2_output_loss: 3.3590 - val_y1_output_root_mean_squared_error: 1.2090 - val_y2_output_root_mean_squared_error: 1.7967\n",
      "Epoch 269/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 1.3086 - y1_output_loss: 0.3207 - y2_output_loss: 0.9806 - y1_output_root_mean_squared_error: 0.5683 - y2_output_root_mean_squared_error: 0.9928 - val_loss: 1.7628 - val_y1_output_loss: 0.5136 - val_y2_output_loss: 1.2470 - val_y1_output_root_mean_squared_error: 0.6914 - val_y2_output_root_mean_squared_error: 1.1335\n",
      "Epoch 270/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 0.9046 - y1_output_loss: 0.2198 - y2_output_loss: 0.6933 - y1_output_root_mean_squared_error: 0.4670 - y2_output_root_mean_squared_error: 0.8286 - val_loss: 1.7200 - val_y1_output_loss: 0.4287 - val_y2_output_loss: 1.2731 - val_y1_output_root_mean_squared_error: 0.6436 - val_y2_output_root_mean_squared_error: 1.1427\n",
      "Epoch 271/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.7972 - y1_output_loss: 0.1875 - y2_output_loss: 0.6090 - y1_output_root_mean_squared_error: 0.4327 - y2_output_root_mean_squared_error: 0.7810 - val_loss: 1.5099 - val_y1_output_loss: 0.4395 - val_y2_output_loss: 1.0589 - val_y1_output_root_mean_squared_error: 0.6480 - val_y2_output_root_mean_squared_error: 1.0440\n",
      "Epoch 272/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 1.2263 - y1_output_loss: 0.3132 - y2_output_loss: 0.9056 - y1_output_root_mean_squared_error: 0.5615 - y2_output_root_mean_squared_error: 0.9545 - val_loss: 1.2669 - val_y1_output_loss: 0.4448 - val_y2_output_loss: 0.8277 - val_y1_output_root_mean_squared_error: 0.6454 - val_y2_output_root_mean_squared_error: 0.9221\n",
      "Epoch 273/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 2.1181 - y1_output_loss: 0.5776 - y2_output_loss: 1.5325 - y1_output_root_mean_squared_error: 0.7624 - y2_output_root_mean_squared_error: 1.2397 - val_loss: 1.9755 - val_y1_output_loss: 0.4973 - val_y2_output_loss: 1.4262 - val_y1_output_root_mean_squared_error: 0.7069 - val_y2_output_root_mean_squared_error: 1.2148\n",
      "Epoch 274/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 0.9118 - y1_output_loss: 0.2318 - y2_output_loss: 0.6738 - y1_output_root_mean_squared_error: 0.4833 - y2_output_root_mean_squared_error: 0.8235 - val_loss: 1.3348 - val_y1_output_loss: 0.2735 - val_y2_output_loss: 1.1012 - val_y1_output_root_mean_squared_error: 0.5119 - val_y2_output_root_mean_squared_error: 1.0357\n",
      "Epoch 275/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 0.7289 - y1_output_loss: 0.1825 - y2_output_loss: 0.5508 - y1_output_root_mean_squared_error: 0.4267 - y2_output_root_mean_squared_error: 0.7395 - val_loss: 2.5119 - val_y1_output_loss: 0.2524 - val_y2_output_loss: 2.3175 - val_y1_output_root_mean_squared_error: 0.4932 - val_y2_output_root_mean_squared_error: 1.5062\n",
      "Epoch 276/500\n",
      "614/614 [==============================] - 0s 133us/sample - loss: 1.2688 - y1_output_loss: 0.3250 - y2_output_loss: 0.9497 - y1_output_root_mean_squared_error: 0.5694 - y2_output_root_mean_squared_error: 0.9719 - val_loss: 3.4164 - val_y1_output_loss: 0.5927 - val_y2_output_loss: 2.9341 - val_y1_output_root_mean_squared_error: 0.7711 - val_y2_output_root_mean_squared_error: 1.6799\n",
      "Epoch 277/500\n",
      "614/614 [==============================] - 0s 131us/sample - loss: 0.8508 - y1_output_loss: 0.2117 - y2_output_loss: 0.6344 - y1_output_root_mean_squared_error: 0.4605 - y2_output_root_mean_squared_error: 0.7992 - val_loss: 1.1529 - val_y1_output_loss: 0.3522 - val_y2_output_loss: 0.8041 - val_y1_output_root_mean_squared_error: 0.5736 - val_y2_output_root_mean_squared_error: 0.9077\n",
      "Epoch 278/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.8849 - y1_output_loss: 0.2141 - y2_output_loss: 0.7372 - y1_output_root_mean_squared_error: 0.4486 - y2_output_root_mean_squared_error: 0.8269 - val_loss: 13.3203 - val_y1_output_loss: 2.5191 - val_y2_output_loss: 10.9072 - val_y1_output_root_mean_squared_error: 1.5555 - val_y2_output_root_mean_squared_error: 3.3016\n",
      "Epoch 279/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 1.3366 - y1_output_loss: 0.2631 - y2_output_loss: 1.0973 - y1_output_root_mean_squared_error: 0.5130 - y2_output_root_mean_squared_error: 1.0361 - val_loss: 5.4283 - val_y1_output_loss: 0.2821 - val_y2_output_loss: 5.3944 - val_y1_output_root_mean_squared_error: 0.5237 - val_y2_output_root_mean_squared_error: 2.2703\n",
      "Epoch 280/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 0.9634 - y1_output_loss: 0.2200 - y2_output_loss: 0.7402 - y1_output_root_mean_squared_error: 0.4701 - y2_output_root_mean_squared_error: 0.8617 - val_loss: 1.6835 - val_y1_output_loss: 0.4477 - val_y2_output_loss: 1.2222 - val_y1_output_root_mean_squared_error: 0.6522 - val_y2_output_root_mean_squared_error: 1.1217\n",
      "Epoch 281/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.8010 - y1_output_loss: 0.1957 - y2_output_loss: 0.5988 - y1_output_root_mean_squared_error: 0.4437 - y2_output_root_mean_squared_error: 0.7773 - val_loss: 1.1462 - val_y1_output_loss: 0.3317 - val_y2_output_loss: 0.8294 - val_y1_output_root_mean_squared_error: 0.5617 - val_y2_output_root_mean_squared_error: 0.9114\n",
      "Epoch 282/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.7867 - y1_output_loss: 0.1939 - y2_output_loss: 0.5937 - y1_output_root_mean_squared_error: 0.4422 - y2_output_root_mean_squared_error: 0.7689 - val_loss: 2.0906 - val_y1_output_loss: 0.2873 - val_y2_output_loss: 1.8878 - val_y1_output_root_mean_squared_error: 0.5290 - val_y2_output_root_mean_squared_error: 1.3456\n",
      "Epoch 283/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.8831 - y1_output_loss: 0.2003 - y2_output_loss: 0.6805 - y1_output_root_mean_squared_error: 0.4485 - y2_output_root_mean_squared_error: 0.8258 - val_loss: 1.5690 - val_y1_output_loss: 0.3608 - val_y2_output_loss: 1.2511 - val_y1_output_root_mean_squared_error: 0.5980 - val_y2_output_root_mean_squared_error: 1.1006\n",
      "Epoch 284/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.8973 - y1_output_loss: 0.2692 - y2_output_loss: 0.6304 - y1_output_root_mean_squared_error: 0.5185 - y2_output_root_mean_squared_error: 0.7927 - val_loss: 1.1077 - val_y1_output_loss: 0.3095 - val_y2_output_loss: 0.8077 - val_y1_output_root_mean_squared_error: 0.5452 - val_y2_output_root_mean_squared_error: 0.9003\n",
      "Epoch 285/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.7735 - y1_output_loss: 0.1812 - y2_output_loss: 0.5892 - y1_output_root_mean_squared_error: 0.4263 - y2_output_root_mean_squared_error: 0.7693 - val_loss: 1.5279 - val_y1_output_loss: 0.2709 - val_y2_output_loss: 1.2321 - val_y1_output_root_mean_squared_error: 0.5058 - val_y2_output_root_mean_squared_error: 1.1278\n",
      "Epoch 286/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.6430 - y1_output_loss: 0.1773 - y2_output_loss: 0.4664 - y1_output_root_mean_squared_error: 0.4202 - y2_output_root_mean_squared_error: 0.6830 - val_loss: 2.0072 - val_y1_output_loss: 0.5511 - val_y2_output_loss: 1.5132 - val_y1_output_root_mean_squared_error: 0.7305 - val_y2_output_root_mean_squared_error: 1.2139\n",
      "Epoch 287/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.5880 - y1_output_loss: 0.1643 - y2_output_loss: 0.4323 - y1_output_root_mean_squared_error: 0.4061 - y2_output_root_mean_squared_error: 0.6505 - val_loss: 2.5376 - val_y1_output_loss: 0.3868 - val_y2_output_loss: 2.1329 - val_y1_output_root_mean_squared_error: 0.6074 - val_y2_output_root_mean_squared_error: 1.4726\n",
      "Epoch 288/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.8540 - y1_output_loss: 0.2144 - y2_output_loss: 0.6492 - y1_output_root_mean_squared_error: 0.4633 - y2_output_root_mean_squared_error: 0.7996 - val_loss: 1.5301 - val_y1_output_loss: 0.3299 - val_y2_output_loss: 1.1697 - val_y1_output_root_mean_squared_error: 0.5705 - val_y2_output_root_mean_squared_error: 1.0975\n",
      "Epoch 289/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 1.1196 - y1_output_loss: 0.2424 - y2_output_loss: 0.8715 - y1_output_root_mean_squared_error: 0.4918 - y2_output_root_mean_squared_error: 0.9369 - val_loss: 1.0750 - val_y1_output_loss: 0.2942 - val_y2_output_loss: 0.7845 - val_y1_output_root_mean_squared_error: 0.5220 - val_y2_output_root_mean_squared_error: 0.8958\n",
      "Epoch 290/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.7274 - y1_output_loss: 0.1962 - y2_output_loss: 0.5271 - y1_output_root_mean_squared_error: 0.4444 - y2_output_root_mean_squared_error: 0.7279 - val_loss: 1.2705 - val_y1_output_loss: 0.3053 - val_y2_output_loss: 0.9915 - val_y1_output_root_mean_squared_error: 0.5392 - val_y2_output_root_mean_squared_error: 0.9898\n",
      "Epoch 291/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 0.7300 - y1_output_loss: 0.1701 - y2_output_loss: 0.5703 - y1_output_root_mean_squared_error: 0.4135 - y2_output_root_mean_squared_error: 0.7477 - val_loss: 2.7736 - val_y1_output_loss: 0.3381 - val_y2_output_loss: 2.5295 - val_y1_output_root_mean_squared_error: 0.5703 - val_y2_output_root_mean_squared_error: 1.5647\n",
      "Epoch 292/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.8276 - y1_output_loss: 0.1897 - y2_output_loss: 0.6489 - y1_output_root_mean_squared_error: 0.4361 - y2_output_root_mean_squared_error: 0.7984 - val_loss: 2.8982 - val_y1_output_loss: 0.3567 - val_y2_output_loss: 2.6586 - val_y1_output_root_mean_squared_error: 0.5982 - val_y2_output_root_mean_squared_error: 1.5939\n",
      "Epoch 293/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.8197 - y1_output_loss: 0.2091 - y2_output_loss: 0.6060 - y1_output_root_mean_squared_error: 0.4568 - y2_output_root_mean_squared_error: 0.7816 - val_loss: 2.0325 - val_y1_output_loss: 0.5022 - val_y2_output_loss: 1.5007 - val_y1_output_root_mean_squared_error: 0.7002 - val_y2_output_root_mean_squared_error: 1.2419\n",
      "Epoch 294/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 1.2142 - y1_output_loss: 0.3402 - y2_output_loss: 0.8719 - y1_output_root_mean_squared_error: 0.5839 - y2_output_root_mean_squared_error: 0.9345 - val_loss: 1.8540 - val_y1_output_loss: 0.2761 - val_y2_output_loss: 1.6466 - val_y1_output_root_mean_squared_error: 0.5254 - val_y2_output_root_mean_squared_error: 1.2562\n",
      "Epoch 295/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.9487 - y1_output_loss: 0.2639 - y2_output_loss: 0.6828 - y1_output_root_mean_squared_error: 0.5158 - y2_output_root_mean_squared_error: 0.8262 - val_loss: 1.4532 - val_y1_output_loss: 0.3177 - val_y2_output_loss: 1.1266 - val_y1_output_root_mean_squared_error: 0.5520 - val_y2_output_root_mean_squared_error: 1.0717\n",
      "Epoch 296/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 0.7370 - y1_output_loss: 0.1835 - y2_output_loss: 0.5524 - y1_output_root_mean_squared_error: 0.4297 - y2_output_root_mean_squared_error: 0.7432 - val_loss: 1.3122 - val_y1_output_loss: 0.2349 - val_y2_output_loss: 1.0536 - val_y1_output_root_mean_squared_error: 0.4698 - val_y2_output_root_mean_squared_error: 1.0448\n",
      "Epoch 297/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.6867 - y1_output_loss: 0.1686 - y2_output_loss: 0.5146 - y1_output_root_mean_squared_error: 0.4111 - y2_output_root_mean_squared_error: 0.7195 - val_loss: 1.1442 - val_y1_output_loss: 0.2822 - val_y2_output_loss: 0.8658 - val_y1_output_root_mean_squared_error: 0.5204 - val_y2_output_root_mean_squared_error: 0.9345\n",
      "Epoch 298/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 0.7019 - y1_output_loss: 0.1866 - y2_output_loss: 0.5151 - y1_output_root_mean_squared_error: 0.4315 - y2_output_root_mean_squared_error: 0.7181 - val_loss: 1.8708 - val_y1_output_loss: 0.4517 - val_y2_output_loss: 1.4067 - val_y1_output_root_mean_squared_error: 0.6510 - val_y2_output_root_mean_squared_error: 1.2029\n",
      "Epoch 299/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.8095 - y1_output_loss: 0.1861 - y2_output_loss: 0.6186 - y1_output_root_mean_squared_error: 0.4321 - y2_output_root_mean_squared_error: 0.7892 - val_loss: 1.2290 - val_y1_output_loss: 0.2425 - val_y2_output_loss: 1.0415 - val_y1_output_root_mean_squared_error: 0.4843 - val_y2_output_root_mean_squared_error: 0.9972\n",
      "Epoch 300/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.7616 - y1_output_loss: 0.1732 - y2_output_loss: 0.5980 - y1_output_root_mean_squared_error: 0.4155 - y2_output_root_mean_squared_error: 0.7674 - val_loss: 3.0047 - val_y1_output_loss: 0.4105 - val_y2_output_loss: 2.7376 - val_y1_output_root_mean_squared_error: 0.6361 - val_y2_output_root_mean_squared_error: 1.6125\n",
      "Epoch 301/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.9866 - y1_output_loss: 0.2275 - y2_output_loss: 0.7538 - y1_output_root_mean_squared_error: 0.4764 - y2_output_root_mean_squared_error: 0.8716 - val_loss: 1.0167 - val_y1_output_loss: 0.2814 - val_y2_output_loss: 0.7609 - val_y1_output_root_mean_squared_error: 0.5206 - val_y2_output_root_mean_squared_error: 0.8635\n",
      "Epoch 302/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 0.6727 - y1_output_loss: 0.1891 - y2_output_loss: 0.4840 - y1_output_root_mean_squared_error: 0.4337 - y2_output_root_mean_squared_error: 0.6961 - val_loss: 1.1779 - val_y1_output_loss: 0.3110 - val_y2_output_loss: 0.8654 - val_y1_output_root_mean_squared_error: 0.5530 - val_y2_output_root_mean_squared_error: 0.9338\n",
      "Epoch 303/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.8014 - y1_output_loss: 0.2162 - y2_output_loss: 0.5847 - y1_output_root_mean_squared_error: 0.4669 - y2_output_root_mean_squared_error: 0.7638 - val_loss: 1.1745 - val_y1_output_loss: 0.2774 - val_y2_output_loss: 0.9210 - val_y1_output_root_mean_squared_error: 0.5196 - val_y2_output_root_mean_squared_error: 0.9510\n",
      "Epoch 304/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.7393 - y1_output_loss: 0.2227 - y2_output_loss: 0.5386 - y1_output_root_mean_squared_error: 0.4660 - y2_output_root_mean_squared_error: 0.7226 - val_loss: 3.3278 - val_y1_output_loss: 0.7279 - val_y2_output_loss: 2.5853 - val_y1_output_root_mean_squared_error: 0.8329 - val_y2_output_root_mean_squared_error: 1.6230\n",
      "Epoch 305/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.6685 - y1_output_loss: 0.1669 - y2_output_loss: 0.4976 - y1_output_root_mean_squared_error: 0.4090 - y2_output_root_mean_squared_error: 0.7080 - val_loss: 0.9137 - val_y1_output_loss: 0.2572 - val_y2_output_loss: 0.6752 - val_y1_output_root_mean_squared_error: 0.4929 - val_y2_output_root_mean_squared_error: 0.8190\n",
      "Epoch 306/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.7637 - y1_output_loss: 0.1969 - y2_output_loss: 0.5651 - y1_output_root_mean_squared_error: 0.4452 - y2_output_root_mean_squared_error: 0.7520 - val_loss: 1.0778 - val_y1_output_loss: 0.2585 - val_y2_output_loss: 0.8090 - val_y1_output_root_mean_squared_error: 0.4986 - val_y2_output_root_mean_squared_error: 0.9106\n",
      "Epoch 307/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.6093 - y1_output_loss: 0.1769 - y2_output_loss: 0.4359 - y1_output_root_mean_squared_error: 0.4215 - y2_output_root_mean_squared_error: 0.6570 - val_loss: 0.9811 - val_y1_output_loss: 0.3129 - val_y2_output_loss: 0.6847 - val_y1_output_root_mean_squared_error: 0.5391 - val_y2_output_root_mean_squared_error: 0.8309\n",
      "Epoch 308/500\n",
      "614/614 [==============================] - 0s 123us/sample - loss: 0.6255 - y1_output_loss: 0.1734 - y2_output_loss: 0.4557 - y1_output_root_mean_squared_error: 0.4178 - y2_output_root_mean_squared_error: 0.6715 - val_loss: 1.8078 - val_y1_output_loss: 0.3605 - val_y2_output_loss: 1.4172 - val_y1_output_root_mean_squared_error: 0.5872 - val_y2_output_root_mean_squared_error: 1.2095\n",
      "Epoch 309/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.6933 - y1_output_loss: 0.1934 - y2_output_loss: 0.4942 - y1_output_root_mean_squared_error: 0.4416 - y2_output_root_mean_squared_error: 0.7059 - val_loss: 1.2053 - val_y1_output_loss: 0.3131 - val_y2_output_loss: 0.9075 - val_y1_output_root_mean_squared_error: 0.5500 - val_y2_output_root_mean_squared_error: 0.9501\n",
      "Epoch 310/500\n",
      "614/614 [==============================] - 0s 123us/sample - loss: 0.5759 - y1_output_loss: 0.1507 - y2_output_loss: 0.4211 - y1_output_root_mean_squared_error: 0.3896 - y2_output_root_mean_squared_error: 0.6512 - val_loss: 0.9686 - val_y1_output_loss: 0.2176 - val_y2_output_loss: 0.7641 - val_y1_output_root_mean_squared_error: 0.4537 - val_y2_output_root_mean_squared_error: 0.8734\n",
      "Epoch 311/500\n",
      "614/614 [==============================] - 0s 123us/sample - loss: 0.6564 - y1_output_loss: 0.1781 - y2_output_loss: 0.4746 - y1_output_root_mean_squared_error: 0.4232 - y2_output_root_mean_squared_error: 0.6909 - val_loss: 1.0452 - val_y1_output_loss: 0.2798 - val_y2_output_loss: 0.7849 - val_y1_output_root_mean_squared_error: 0.5176 - val_y2_output_root_mean_squared_error: 0.8816\n",
      "Epoch 312/500\n",
      "614/614 [==============================] - 0s 123us/sample - loss: 0.7699 - y1_output_loss: 0.1764 - y2_output_loss: 0.5943 - y1_output_root_mean_squared_error: 0.4211 - y2_output_root_mean_squared_error: 0.7698 - val_loss: 1.7542 - val_y1_output_loss: 0.4353 - val_y2_output_loss: 1.3006 - val_y1_output_root_mean_squared_error: 0.6424 - val_y2_output_root_mean_squared_error: 1.1583\n",
      "Epoch 313/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 0.9540 - y1_output_loss: 0.2594 - y2_output_loss: 0.6910 - y1_output_root_mean_squared_error: 0.5100 - y2_output_root_mean_squared_error: 0.8330 - val_loss: 0.8859 - val_y1_output_loss: 0.2271 - val_y2_output_loss: 0.6676 - val_y1_output_root_mean_squared_error: 0.4688 - val_y2_output_root_mean_squared_error: 0.8162\n",
      "Epoch 314/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.6759 - y1_output_loss: 0.1754 - y2_output_loss: 0.5083 - y1_output_root_mean_squared_error: 0.4201 - y2_output_root_mean_squared_error: 0.7067 - val_loss: 2.6517 - val_y1_output_loss: 0.3404 - val_y2_output_loss: 2.3914 - val_y1_output_root_mean_squared_error: 0.5878 - val_y2_output_root_mean_squared_error: 1.5186\n",
      "Epoch 315/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 0.5910 - y1_output_loss: 0.1538 - y2_output_loss: 0.4350 - y1_output_root_mean_squared_error: 0.3934 - y2_output_root_mean_squared_error: 0.6604 - val_loss: 1.0956 - val_y1_output_loss: 0.2680 - val_y2_output_loss: 0.8497 - val_y1_output_root_mean_squared_error: 0.5034 - val_y2_output_root_mean_squared_error: 0.9177\n",
      "Epoch 316/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.7839 - y1_output_loss: 0.1681 - y2_output_loss: 0.6112 - y1_output_root_mean_squared_error: 0.4107 - y2_output_root_mean_squared_error: 0.7844 - val_loss: 1.0052 - val_y1_output_loss: 0.2550 - val_y2_output_loss: 0.7480 - val_y1_output_root_mean_squared_error: 0.4898 - val_y2_output_root_mean_squared_error: 0.8748\n",
      "Epoch 317/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.6865 - y1_output_loss: 0.1848 - y2_output_loss: 0.4976 - y1_output_root_mean_squared_error: 0.4310 - y2_output_root_mean_squared_error: 0.7076 - val_loss: 1.1130 - val_y1_output_loss: 0.2616 - val_y2_output_loss: 0.8416 - val_y1_output_root_mean_squared_error: 0.4935 - val_y2_output_root_mean_squared_error: 0.9325\n",
      "Epoch 318/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 0.6671 - y1_output_loss: 0.1638 - y2_output_loss: 0.4982 - y1_output_root_mean_squared_error: 0.4058 - y2_output_root_mean_squared_error: 0.7089 - val_loss: 0.9169 - val_y1_output_loss: 0.2745 - val_y2_output_loss: 0.6548 - val_y1_output_root_mean_squared_error: 0.5073 - val_y2_output_root_mean_squared_error: 0.8121\n",
      "Epoch 319/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.7026 - y1_output_loss: 0.1732 - y2_output_loss: 0.5247 - y1_output_root_mean_squared_error: 0.4180 - y2_output_root_mean_squared_error: 0.7265 - val_loss: 1.0329 - val_y1_output_loss: 0.2202 - val_y2_output_loss: 0.8033 - val_y1_output_root_mean_squared_error: 0.4521 - val_y2_output_root_mean_squared_error: 0.9102\n",
      "Epoch 320/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.6688 - y1_output_loss: 0.1792 - y2_output_loss: 0.4974 - y1_output_root_mean_squared_error: 0.4241 - y2_output_root_mean_squared_error: 0.6993 - val_loss: 1.7969 - val_y1_output_loss: 0.2454 - val_y2_output_loss: 1.6053 - val_y1_output_root_mean_squared_error: 0.4861 - val_y2_output_root_mean_squared_error: 1.2493\n",
      "Epoch 321/500\n",
      "614/614 [==============================] - 0s 123us/sample - loss: 0.7835 - y1_output_loss: 0.2178 - y2_output_loss: 0.5621 - y1_output_root_mean_squared_error: 0.4661 - y2_output_root_mean_squared_error: 0.7525 - val_loss: 1.2482 - val_y1_output_loss: 0.4719 - val_y2_output_loss: 0.7874 - val_y1_output_root_mean_squared_error: 0.6696 - val_y2_output_root_mean_squared_error: 0.8943\n",
      "Epoch 322/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.6566 - y1_output_loss: 0.1871 - y2_output_loss: 0.4852 - y1_output_root_mean_squared_error: 0.4292 - y2_output_root_mean_squared_error: 0.6873 - val_loss: 1.9934 - val_y1_output_loss: 0.5072 - val_y2_output_loss: 1.5650 - val_y1_output_root_mean_squared_error: 0.7050 - val_y2_output_root_mean_squared_error: 1.2233\n",
      "Epoch 323/500\n",
      "614/614 [==============================] - 0s 123us/sample - loss: 0.6277 - y1_output_loss: 0.1710 - y2_output_loss: 0.4539 - y1_output_root_mean_squared_error: 0.4138 - y2_output_root_mean_squared_error: 0.6756 - val_loss: 0.9397 - val_y1_output_loss: 0.2472 - val_y2_output_loss: 0.7015 - val_y1_output_root_mean_squared_error: 0.4907 - val_y2_output_root_mean_squared_error: 0.8360\n",
      "Epoch 324/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.7026 - y1_output_loss: 0.2024 - y2_output_loss: 0.5013 - y1_output_root_mean_squared_error: 0.4460 - y2_output_root_mean_squared_error: 0.7097 - val_loss: 1.0318 - val_y1_output_loss: 0.3170 - val_y2_output_loss: 0.7412 - val_y1_output_root_mean_squared_error: 0.5397 - val_y2_output_root_mean_squared_error: 0.8605\n",
      "Epoch 325/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.6756 - y1_output_loss: 0.1718 - y2_output_loss: 0.5087 - y1_output_root_mean_squared_error: 0.4113 - y2_output_root_mean_squared_error: 0.7116 - val_loss: 1.3891 - val_y1_output_loss: 0.4070 - val_y2_output_loss: 0.9851 - val_y1_output_root_mean_squared_error: 0.6192 - val_y2_output_root_mean_squared_error: 1.0028\n",
      "Epoch 326/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.6769 - y1_output_loss: 0.1552 - y2_output_loss: 0.5177 - y1_output_root_mean_squared_error: 0.3955 - y2_output_root_mean_squared_error: 0.7215 - val_loss: 1.1197 - val_y1_output_loss: 0.2653 - val_y2_output_loss: 0.8770 - val_y1_output_root_mean_squared_error: 0.5047 - val_y2_output_root_mean_squared_error: 0.9300\n",
      "Epoch 327/500\n",
      "614/614 [==============================] - 0s 123us/sample - loss: 0.6599 - y1_output_loss: 0.1841 - y2_output_loss: 0.4710 - y1_output_root_mean_squared_error: 0.4301 - y2_output_root_mean_squared_error: 0.6891 - val_loss: 1.3495 - val_y1_output_loss: 0.5337 - val_y2_output_loss: 0.8433 - val_y1_output_root_mean_squared_error: 0.7109 - val_y2_output_root_mean_squared_error: 0.9187\n",
      "Epoch 328/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.6724 - y1_output_loss: 0.1851 - y2_output_loss: 0.4852 - y1_output_root_mean_squared_error: 0.4294 - y2_output_root_mean_squared_error: 0.6985 - val_loss: 1.0570 - val_y1_output_loss: 0.2858 - val_y2_output_loss: 0.7627 - val_y1_output_root_mean_squared_error: 0.5243 - val_y2_output_root_mean_squared_error: 0.8844\n",
      "Epoch 329/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 1.4465 - y1_output_loss: 0.3617 - y2_output_loss: 1.0743 - y1_output_root_mean_squared_error: 0.6041 - y2_output_root_mean_squared_error: 1.0399 - val_loss: 1.8429 - val_y1_output_loss: 0.3776 - val_y2_output_loss: 1.4675 - val_y1_output_root_mean_squared_error: 0.5925 - val_y2_output_root_mean_squared_error: 1.2214\n",
      "Epoch 330/500\n",
      "614/614 [==============================] - 0s 123us/sample - loss: 0.5086 - y1_output_loss: 0.1472 - y2_output_loss: 0.3594 - y1_output_root_mean_squared_error: 0.3827 - y2_output_root_mean_squared_error: 0.6017 - val_loss: 1.1701 - val_y1_output_loss: 0.5010 - val_y2_output_loss: 0.6879 - val_y1_output_root_mean_squared_error: 0.6830 - val_y2_output_root_mean_squared_error: 0.8388\n",
      "Epoch 331/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.4718 - y1_output_loss: 0.1417 - y2_output_loss: 0.3274 - y1_output_root_mean_squared_error: 0.3773 - y2_output_root_mean_squared_error: 0.5740 - val_loss: 1.0364 - val_y1_output_loss: 0.2780 - val_y2_output_loss: 0.7647 - val_y1_output_root_mean_squared_error: 0.5111 - val_y2_output_root_mean_squared_error: 0.8805\n",
      "Epoch 332/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.5127 - y1_output_loss: 0.1466 - y2_output_loss: 0.3631 - y1_output_root_mean_squared_error: 0.3823 - y2_output_root_mean_squared_error: 0.6054 - val_loss: 0.9867 - val_y1_output_loss: 0.2417 - val_y2_output_loss: 0.7431 - val_y1_output_root_mean_squared_error: 0.4822 - val_y2_output_root_mean_squared_error: 0.8684\n",
      "Epoch 333/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.8136 - y1_output_loss: 0.2204 - y2_output_loss: 0.5864 - y1_output_root_mean_squared_error: 0.4712 - y2_output_root_mean_squared_error: 0.7691 - val_loss: 1.0580 - val_y1_output_loss: 0.3043 - val_y2_output_loss: 0.7730 - val_y1_output_root_mean_squared_error: 0.5312 - val_y2_output_root_mean_squared_error: 0.8808\n",
      "Epoch 334/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.6344 - y1_output_loss: 0.1700 - y2_output_loss: 0.4664 - y1_output_root_mean_squared_error: 0.4139 - y2_output_root_mean_squared_error: 0.6805 - val_loss: 1.2806 - val_y1_output_loss: 0.2750 - val_y2_output_loss: 0.9940 - val_y1_output_root_mean_squared_error: 0.5088 - val_y2_output_root_mean_squared_error: 1.0108\n",
      "Epoch 335/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.6020 - y1_output_loss: 0.1774 - y2_output_loss: 0.4417 - y1_output_root_mean_squared_error: 0.4135 - y2_output_root_mean_squared_error: 0.6565 - val_loss: 2.4449 - val_y1_output_loss: 0.7043 - val_y2_output_loss: 1.8292 - val_y1_output_root_mean_squared_error: 0.8427 - val_y2_output_root_mean_squared_error: 1.3171\n",
      "Epoch 336/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.7388 - y1_output_loss: 0.2019 - y2_output_loss: 0.5323 - y1_output_root_mean_squared_error: 0.4502 - y2_output_root_mean_squared_error: 0.7322 - val_loss: 1.3380 - val_y1_output_loss: 0.2657 - val_y2_output_loss: 1.1108 - val_y1_output_root_mean_squared_error: 0.5088 - val_y2_output_root_mean_squared_error: 1.0388\n",
      "Epoch 337/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 0.5821 - y1_output_loss: 0.1556 - y2_output_loss: 0.4282 - y1_output_root_mean_squared_error: 0.3917 - y2_output_root_mean_squared_error: 0.6547 - val_loss: 0.9755 - val_y1_output_loss: 0.2431 - val_y2_output_loss: 0.7391 - val_y1_output_root_mean_squared_error: 0.4848 - val_y2_output_root_mean_squared_error: 0.8605\n",
      "Epoch 338/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.6023 - y1_output_loss: 0.1521 - y2_output_loss: 0.4489 - y1_output_root_mean_squared_error: 0.3912 - y2_output_root_mean_squared_error: 0.6703 - val_loss: 1.1590 - val_y1_output_loss: 0.2665 - val_y2_output_loss: 0.9321 - val_y1_output_root_mean_squared_error: 0.5108 - val_y2_output_root_mean_squared_error: 0.9477\n",
      "Epoch 339/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.6437 - y1_output_loss: 0.1767 - y2_output_loss: 0.4622 - y1_output_root_mean_squared_error: 0.4221 - y2_output_root_mean_squared_error: 0.6824 - val_loss: 1.0161 - val_y1_output_loss: 0.2799 - val_y2_output_loss: 0.7439 - val_y1_output_root_mean_squared_error: 0.5217 - val_y2_output_root_mean_squared_error: 0.8625\n",
      "Epoch 340/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 0.7394 - y1_output_loss: 0.2007 - y2_output_loss: 0.5361 - y1_output_root_mean_squared_error: 0.4465 - y2_output_root_mean_squared_error: 0.7349 - val_loss: 0.9416 - val_y1_output_loss: 0.2873 - val_y2_output_loss: 0.6607 - val_y1_output_root_mean_squared_error: 0.5218 - val_y2_output_root_mean_squared_error: 0.8181\n",
      "Epoch 341/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.5686 - y1_output_loss: 0.1534 - y2_output_loss: 0.4102 - y1_output_root_mean_squared_error: 0.3935 - y2_output_root_mean_squared_error: 0.6432 - val_loss: 1.0168 - val_y1_output_loss: 0.3328 - val_y2_output_loss: 0.6923 - val_y1_output_root_mean_squared_error: 0.5556 - val_y2_output_root_mean_squared_error: 0.8415\n",
      "Epoch 342/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 0.5356 - y1_output_loss: 0.1657 - y2_output_loss: 0.3680 - y1_output_root_mean_squared_error: 0.4062 - y2_output_root_mean_squared_error: 0.6087 - val_loss: 0.9657 - val_y1_output_loss: 0.2494 - val_y2_output_loss: 0.7351 - val_y1_output_root_mean_squared_error: 0.4917 - val_y2_output_root_mean_squared_error: 0.8508\n",
      "Epoch 343/500\n",
      "614/614 [==============================] - 0s 132us/sample - loss: 0.4896 - y1_output_loss: 0.1408 - y2_output_loss: 0.3474 - y1_output_root_mean_squared_error: 0.3750 - y2_output_root_mean_squared_error: 0.5907 - val_loss: 1.0512 - val_y1_output_loss: 0.2898 - val_y2_output_loss: 0.7905 - val_y1_output_root_mean_squared_error: 0.5303 - val_y2_output_root_mean_squared_error: 0.8775\n",
      "Epoch 344/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 0.6054 - y1_output_loss: 0.1671 - y2_output_loss: 0.4347 - y1_output_root_mean_squared_error: 0.4088 - y2_output_root_mean_squared_error: 0.6620 - val_loss: 0.9144 - val_y1_output_loss: 0.2445 - val_y2_output_loss: 0.6800 - val_y1_output_root_mean_squared_error: 0.4827 - val_y2_output_root_mean_squared_error: 0.8255\n",
      "Epoch 345/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.5883 - y1_output_loss: 0.1517 - y2_output_loss: 0.4326 - y1_output_root_mean_squared_error: 0.3910 - y2_output_root_mean_squared_error: 0.6598 - val_loss: 0.9608 - val_y1_output_loss: 0.2536 - val_y2_output_loss: 0.7061 - val_y1_output_root_mean_squared_error: 0.4905 - val_y2_output_root_mean_squared_error: 0.8487\n",
      "Epoch 346/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.6239 - y1_output_loss: 0.1782 - y2_output_loss: 0.4421 - y1_output_root_mean_squared_error: 0.4229 - y2_output_root_mean_squared_error: 0.6671 - val_loss: 0.8821 - val_y1_output_loss: 0.2536 - val_y2_output_loss: 0.6493 - val_y1_output_root_mean_squared_error: 0.4893 - val_y2_output_root_mean_squared_error: 0.8017\n",
      "Epoch 347/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.5556 - y1_output_loss: 0.1578 - y2_output_loss: 0.4048 - y1_output_root_mean_squared_error: 0.3969 - y2_output_root_mean_squared_error: 0.6309 - val_loss: 0.9083 - val_y1_output_loss: 0.2650 - val_y2_output_loss: 0.6796 - val_y1_output_root_mean_squared_error: 0.5007 - val_y2_output_root_mean_squared_error: 0.8109\n",
      "Epoch 348/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.6394 - y1_output_loss: 0.1646 - y2_output_loss: 0.4716 - y1_output_root_mean_squared_error: 0.4066 - y2_output_root_mean_squared_error: 0.6885 - val_loss: 0.9688 - val_y1_output_loss: 0.2948 - val_y2_output_loss: 0.6830 - val_y1_output_root_mean_squared_error: 0.5262 - val_y2_output_root_mean_squared_error: 0.8318\n",
      "Epoch 349/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 0.5164 - y1_output_loss: 0.1532 - y2_output_loss: 0.3721 - y1_output_root_mean_squared_error: 0.3876 - y2_output_root_mean_squared_error: 0.6051 - val_loss: 1.0869 - val_y1_output_loss: 0.4198 - val_y2_output_loss: 0.6969 - val_y1_output_root_mean_squared_error: 0.6292 - val_y2_output_root_mean_squared_error: 0.8313\n",
      "Epoch 350/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.5650 - y1_output_loss: 0.1596 - y2_output_loss: 0.4032 - y1_output_root_mean_squared_error: 0.3992 - y2_output_root_mean_squared_error: 0.6369 - val_loss: 0.9678 - val_y1_output_loss: 0.2258 - val_y2_output_loss: 0.7673 - val_y1_output_root_mean_squared_error: 0.4634 - val_y2_output_root_mean_squared_error: 0.8677\n",
      "Epoch 351/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 0.6287 - y1_output_loss: 0.1771 - y2_output_loss: 0.4541 - y1_output_root_mean_squared_error: 0.4180 - y2_output_root_mean_squared_error: 0.6738 - val_loss: 3.1153 - val_y1_output_loss: 0.4347 - val_y2_output_loss: 2.7733 - val_y1_output_root_mean_squared_error: 0.6651 - val_y2_output_root_mean_squared_error: 1.6349\n",
      "Epoch 352/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.6593 - y1_output_loss: 0.1784 - y2_output_loss: 0.4791 - y1_output_root_mean_squared_error: 0.4217 - y2_output_root_mean_squared_error: 0.6939 - val_loss: 1.1593 - val_y1_output_loss: 0.2703 - val_y2_output_loss: 0.8932 - val_y1_output_root_mean_squared_error: 0.5146 - val_y2_output_root_mean_squared_error: 0.9458\n",
      "Epoch 353/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.6550 - y1_output_loss: 0.1655 - y2_output_loss: 0.4886 - y1_output_root_mean_squared_error: 0.4087 - y2_output_root_mean_squared_error: 0.6985 - val_loss: 1.1026 - val_y1_output_loss: 0.2518 - val_y2_output_loss: 0.8449 - val_y1_output_root_mean_squared_error: 0.4825 - val_y2_output_root_mean_squared_error: 0.9326\n",
      "Epoch 354/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.4929 - y1_output_loss: 0.1436 - y2_output_loss: 0.3661 - y1_output_root_mean_squared_error: 0.3763 - y2_output_root_mean_squared_error: 0.5928 - val_loss: 2.6409 - val_y1_output_loss: 0.5778 - val_y2_output_loss: 2.0582 - val_y1_output_root_mean_squared_error: 0.7403 - val_y2_output_root_mean_squared_error: 1.4467\n",
      "Epoch 355/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 0.5913 - y1_output_loss: 0.1667 - y2_output_loss: 0.4363 - y1_output_root_mean_squared_error: 0.4083 - y2_output_root_mean_squared_error: 0.6516 - val_loss: 3.2388 - val_y1_output_loss: 0.2967 - val_y2_output_loss: 3.0416 - val_y1_output_root_mean_squared_error: 0.5424 - val_y2_output_root_mean_squared_error: 1.7160\n",
      "Epoch 356/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 0.5933 - y1_output_loss: 0.1497 - y2_output_loss: 0.4477 - y1_output_root_mean_squared_error: 0.3878 - y2_output_root_mean_squared_error: 0.6655 - val_loss: 0.9699 - val_y1_output_loss: 0.2357 - val_y2_output_loss: 0.7317 - val_y1_output_root_mean_squared_error: 0.4743 - val_y2_output_root_mean_squared_error: 0.8631\n",
      "Epoch 357/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 0.5954 - y1_output_loss: 0.1646 - y2_output_loss: 0.4278 - y1_output_root_mean_squared_error: 0.4060 - y2_output_root_mean_squared_error: 0.6561 - val_loss: 1.1192 - val_y1_output_loss: 0.3374 - val_y2_output_loss: 0.7793 - val_y1_output_root_mean_squared_error: 0.5731 - val_y2_output_root_mean_squared_error: 0.8893\n",
      "Epoch 358/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 0.7443 - y1_output_loss: 0.2186 - y2_output_loss: 0.5380 - y1_output_root_mean_squared_error: 0.4616 - y2_output_root_mean_squared_error: 0.7288 - val_loss: 2.3356 - val_y1_output_loss: 0.8328 - val_y2_output_loss: 1.4903 - val_y1_output_root_mean_squared_error: 0.8954 - val_y2_output_root_mean_squared_error: 1.2385\n",
      "Epoch 359/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 0.7352 - y1_output_loss: 0.1990 - y2_output_loss: 0.5312 - y1_output_root_mean_squared_error: 0.4471 - y2_output_root_mean_squared_error: 0.7317 - val_loss: 0.9822 - val_y1_output_loss: 0.3037 - val_y2_output_loss: 0.6823 - val_y1_output_root_mean_squared_error: 0.5348 - val_y2_output_root_mean_squared_error: 0.8344\n",
      "Epoch 360/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.7257 - y1_output_loss: 0.1993 - y2_output_loss: 0.5224 - y1_output_root_mean_squared_error: 0.4475 - y2_output_root_mean_squared_error: 0.7249 - val_loss: 1.1222 - val_y1_output_loss: 0.2394 - val_y2_output_loss: 0.9057 - val_y1_output_root_mean_squared_error: 0.4773 - val_y2_output_root_mean_squared_error: 0.9457\n",
      "Epoch 361/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 0.5675 - y1_output_loss: 0.1460 - y2_output_loss: 0.4227 - y1_output_root_mean_squared_error: 0.3806 - y2_output_root_mean_squared_error: 0.6501 - val_loss: 1.3842 - val_y1_output_loss: 0.3970 - val_y2_output_loss: 0.9773 - val_y1_output_root_mean_squared_error: 0.6217 - val_y2_output_root_mean_squared_error: 0.9988\n",
      "Epoch 362/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 0.4638 - y1_output_loss: 0.1433 - y2_output_loss: 0.3274 - y1_output_root_mean_squared_error: 0.3757 - y2_output_root_mean_squared_error: 0.5680 - val_loss: 1.4669 - val_y1_output_loss: 0.4560 - val_y2_output_loss: 1.0104 - val_y1_output_root_mean_squared_error: 0.6554 - val_y2_output_root_mean_squared_error: 1.0185\n",
      "Epoch 363/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.6083 - y1_output_loss: 0.1705 - y2_output_loss: 0.4354 - y1_output_root_mean_squared_error: 0.4134 - y2_output_root_mean_squared_error: 0.6613 - val_loss: 1.0725 - val_y1_output_loss: 0.2507 - val_y2_output_loss: 0.8207 - val_y1_output_root_mean_squared_error: 0.4834 - val_y2_output_root_mean_squared_error: 0.9159\n",
      "Epoch 364/500\n",
      "614/614 [==============================] - 0s 131us/sample - loss: 0.5905 - y1_output_loss: 0.1699 - y2_output_loss: 0.4159 - y1_output_root_mean_squared_error: 0.4133 - y2_output_root_mean_squared_error: 0.6478 - val_loss: 0.9524 - val_y1_output_loss: 0.2854 - val_y2_output_loss: 0.6751 - val_y1_output_root_mean_squared_error: 0.5181 - val_y2_output_root_mean_squared_error: 0.8270\n",
      "Epoch 365/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.4288 - y1_output_loss: 0.1211 - y2_output_loss: 0.3043 - y1_output_root_mean_squared_error: 0.3488 - y2_output_root_mean_squared_error: 0.5542 - val_loss: 0.9706 - val_y1_output_loss: 0.2200 - val_y2_output_loss: 0.7472 - val_y1_output_root_mean_squared_error: 0.4558 - val_y2_output_root_mean_squared_error: 0.8734\n",
      "Epoch 366/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.4478 - y1_output_loss: 0.1488 - y2_output_loss: 0.2963 - y1_output_root_mean_squared_error: 0.3870 - y2_output_root_mean_squared_error: 0.5459 - val_loss: 1.0450 - val_y1_output_loss: 0.2533 - val_y2_output_loss: 0.7836 - val_y1_output_root_mean_squared_error: 0.4875 - val_y2_output_root_mean_squared_error: 0.8985\n",
      "Epoch 367/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.4791 - y1_output_loss: 0.1400 - y2_output_loss: 0.3384 - y1_output_root_mean_squared_error: 0.3736 - y2_output_root_mean_squared_error: 0.5827 - val_loss: 1.1426 - val_y1_output_loss: 0.3474 - val_y2_output_loss: 0.7980 - val_y1_output_root_mean_squared_error: 0.5671 - val_y2_output_root_mean_squared_error: 0.9061\n",
      "Epoch 368/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.4697 - y1_output_loss: 0.1290 - y2_output_loss: 0.3408 - y1_output_root_mean_squared_error: 0.3606 - y2_output_root_mean_squared_error: 0.5828 - val_loss: 1.3519 - val_y1_output_loss: 0.2554 - val_y2_output_loss: 1.0819 - val_y1_output_root_mean_squared_error: 0.4921 - val_y2_output_root_mean_squared_error: 1.0535\n",
      "Epoch 369/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.6113 - y1_output_loss: 0.1748 - y2_output_loss: 0.4342 - y1_output_root_mean_squared_error: 0.4176 - y2_output_root_mean_squared_error: 0.6610 - val_loss: 1.1680 - val_y1_output_loss: 0.4460 - val_y2_output_loss: 0.7469 - val_y1_output_root_mean_squared_error: 0.6428 - val_y2_output_root_mean_squared_error: 0.8688\n",
      "Epoch 370/500\n",
      "614/614 [==============================] - 0s 123us/sample - loss: 0.5597 - y1_output_loss: 0.1635 - y2_output_loss: 0.3952 - y1_output_root_mean_squared_error: 0.4044 - y2_output_root_mean_squared_error: 0.6294 - val_loss: 1.0846 - val_y1_output_loss: 0.2705 - val_y2_output_loss: 0.8114 - val_y1_output_root_mean_squared_error: 0.5016 - val_y2_output_root_mean_squared_error: 0.9127\n",
      "Epoch 371/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.4916 - y1_output_loss: 0.1587 - y2_output_loss: 0.3337 - y1_output_root_mean_squared_error: 0.3976 - y2_output_root_mean_squared_error: 0.5775 - val_loss: 1.0708 - val_y1_output_loss: 0.3153 - val_y2_output_loss: 0.7453 - val_y1_output_root_mean_squared_error: 0.5553 - val_y2_output_root_mean_squared_error: 0.8732\n",
      "Epoch 372/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.9483 - y1_output_loss: 0.2454 - y2_output_loss: 0.7157 - y1_output_root_mean_squared_error: 0.4910 - y2_output_root_mean_squared_error: 0.8410 - val_loss: 3.4703 - val_y1_output_loss: 1.2438 - val_y2_output_loss: 2.2121 - val_y1_output_root_mean_squared_error: 1.1061 - val_y2_output_root_mean_squared_error: 1.4990\n",
      "Epoch 373/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 0.4933 - y1_output_loss: 0.1587 - y2_output_loss: 0.3459 - y1_output_root_mean_squared_error: 0.3991 - y2_output_root_mean_squared_error: 0.5779 - val_loss: 1.3772 - val_y1_output_loss: 0.2629 - val_y2_output_loss: 1.1488 - val_y1_output_root_mean_squared_error: 0.5074 - val_y2_output_root_mean_squared_error: 1.0582\n",
      "Epoch 374/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.5670 - y1_output_loss: 0.1590 - y2_output_loss: 0.4033 - y1_output_root_mean_squared_error: 0.4000 - y2_output_root_mean_squared_error: 0.6380 - val_loss: 1.1172 - val_y1_output_loss: 0.2833 - val_y2_output_loss: 0.8314 - val_y1_output_root_mean_squared_error: 0.5149 - val_y2_output_root_mean_squared_error: 0.9231\n",
      "Epoch 375/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 0.7003 - y1_output_loss: 0.1784 - y2_output_loss: 0.5189 - y1_output_root_mean_squared_error: 0.4234 - y2_output_root_mean_squared_error: 0.7218 - val_loss: 0.8719 - val_y1_output_loss: 0.2261 - val_y2_output_loss: 0.6452 - val_y1_output_root_mean_squared_error: 0.4590 - val_y2_output_root_mean_squared_error: 0.8131\n",
      "Epoch 376/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 0.6344 - y1_output_loss: 0.1736 - y2_output_loss: 0.4571 - y1_output_root_mean_squared_error: 0.4179 - y2_output_root_mean_squared_error: 0.6780 - val_loss: 1.1326 - val_y1_output_loss: 0.3151 - val_y2_output_loss: 0.8237 - val_y1_output_root_mean_squared_error: 0.5395 - val_y2_output_root_mean_squared_error: 0.9174\n",
      "Epoch 377/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.3935 - y1_output_loss: 0.1190 - y2_output_loss: 0.2766 - y1_output_root_mean_squared_error: 0.3433 - y2_output_root_mean_squared_error: 0.5250 - val_loss: 1.4058 - val_y1_output_loss: 0.3968 - val_y2_output_loss: 1.0110 - val_y1_output_root_mean_squared_error: 0.6069 - val_y2_output_root_mean_squared_error: 1.0186\n",
      "Epoch 378/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.4946 - y1_output_loss: 0.1503 - y2_output_loss: 0.3443 - y1_output_root_mean_squared_error: 0.3882 - y2_output_root_mean_squared_error: 0.5865 - val_loss: 0.8504 - val_y1_output_loss: 0.2142 - val_y2_output_loss: 0.6297 - val_y1_output_root_mean_squared_error: 0.4550 - val_y2_output_root_mean_squared_error: 0.8021\n",
      "Epoch 379/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.5838 - y1_output_loss: 0.1754 - y2_output_loss: 0.4041 - y1_output_root_mean_squared_error: 0.4206 - y2_output_root_mean_squared_error: 0.6379 - val_loss: 0.8032 - val_y1_output_loss: 0.2163 - val_y2_output_loss: 0.5882 - val_y1_output_root_mean_squared_error: 0.4503 - val_y2_output_root_mean_squared_error: 0.7749\n",
      "Epoch 380/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.6740 - y1_output_loss: 0.1867 - y2_output_loss: 0.4822 - y1_output_root_mean_squared_error: 0.4329 - y2_output_root_mean_squared_error: 0.6976 - val_loss: 0.8127 - val_y1_output_loss: 0.2101 - val_y2_output_loss: 0.6166 - val_y1_output_root_mean_squared_error: 0.4431 - val_y2_output_root_mean_squared_error: 0.7851\n",
      "Epoch 381/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.5105 - y1_output_loss: 0.1390 - y2_output_loss: 0.3698 - y1_output_root_mean_squared_error: 0.3738 - y2_output_root_mean_squared_error: 0.6089 - val_loss: 0.8412 - val_y1_output_loss: 0.2915 - val_y2_output_loss: 0.5724 - val_y1_output_root_mean_squared_error: 0.5207 - val_y2_output_root_mean_squared_error: 0.7551\n",
      "Epoch 382/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 0.4543 - y1_output_loss: 0.1370 - y2_output_loss: 0.3151 - y1_output_root_mean_squared_error: 0.3694 - y2_output_root_mean_squared_error: 0.5638 - val_loss: 0.8682 - val_y1_output_loss: 0.2542 - val_y2_output_loss: 0.6212 - val_y1_output_root_mean_squared_error: 0.4899 - val_y2_output_root_mean_squared_error: 0.7926\n",
      "Epoch 383/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 0.4320 - y1_output_loss: 0.1322 - y2_output_loss: 0.2975 - y1_output_root_mean_squared_error: 0.3641 - y2_output_root_mean_squared_error: 0.5472 - val_loss: 0.9629 - val_y1_output_loss: 0.3017 - val_y2_output_loss: 0.6811 - val_y1_output_root_mean_squared_error: 0.5329 - val_y2_output_root_mean_squared_error: 0.8240\n",
      "Epoch 384/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.4280 - y1_output_loss: 0.1373 - y2_output_loss: 0.2909 - y1_output_root_mean_squared_error: 0.3716 - y2_output_root_mean_squared_error: 0.5384 - val_loss: 1.2326 - val_y1_output_loss: 0.2503 - val_y2_output_loss: 1.0070 - val_y1_output_root_mean_squared_error: 0.4969 - val_y2_output_root_mean_squared_error: 0.9928\n",
      "Epoch 385/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.5146 - y1_output_loss: 0.1603 - y2_output_loss: 0.3643 - y1_output_root_mean_squared_error: 0.3918 - y2_output_root_mean_squared_error: 0.6009 - val_loss: 3.6024 - val_y1_output_loss: 1.3114 - val_y2_output_loss: 2.4496 - val_y1_output_root_mean_squared_error: 1.1246 - val_y2_output_root_mean_squared_error: 1.5290\n",
      "Epoch 386/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.7150 - y1_output_loss: 0.1897 - y2_output_loss: 0.5205 - y1_output_root_mean_squared_error: 0.4371 - y2_output_root_mean_squared_error: 0.7238 - val_loss: 0.8736 - val_y1_output_loss: 0.2752 - val_y2_output_loss: 0.6293 - val_y1_output_root_mean_squared_error: 0.5146 - val_y2_output_root_mean_squared_error: 0.7803\n",
      "Epoch 387/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.7367 - y1_output_loss: 0.1862 - y2_output_loss: 0.5468 - y1_output_root_mean_squared_error: 0.4308 - y2_output_root_mean_squared_error: 0.7424 - val_loss: 1.0523 - val_y1_output_loss: 0.3078 - val_y2_output_loss: 0.7392 - val_y1_output_root_mean_squared_error: 0.5403 - val_y2_output_root_mean_squared_error: 0.8720\n",
      "Epoch 388/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 0.6098 - y1_output_loss: 0.1607 - y2_output_loss: 0.4467 - y1_output_root_mean_squared_error: 0.4002 - y2_output_root_mean_squared_error: 0.6706 - val_loss: 1.2726 - val_y1_output_loss: 0.4061 - val_y2_output_loss: 0.8659 - val_y1_output_root_mean_squared_error: 0.6320 - val_y2_output_root_mean_squared_error: 0.9345\n",
      "Epoch 389/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.4055 - y1_output_loss: 0.1352 - y2_output_loss: 0.2701 - y1_output_root_mean_squared_error: 0.3690 - y2_output_root_mean_squared_error: 0.5190 - val_loss: 0.9297 - val_y1_output_loss: 0.2435 - val_y2_output_loss: 0.6911 - val_y1_output_root_mean_squared_error: 0.4781 - val_y2_output_root_mean_squared_error: 0.8373\n",
      "Epoch 390/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.4278 - y1_output_loss: 0.1341 - y2_output_loss: 0.2945 - y1_output_root_mean_squared_error: 0.3665 - y2_output_root_mean_squared_error: 0.5418 - val_loss: 1.4829 - val_y1_output_loss: 0.3998 - val_y2_output_loss: 1.1324 - val_y1_output_root_mean_squared_error: 0.6183 - val_y2_output_root_mean_squared_error: 1.0491\n",
      "Epoch 391/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.6988 - y1_output_loss: 0.1974 - y2_output_loss: 0.4978 - y1_output_root_mean_squared_error: 0.4459 - y2_output_root_mean_squared_error: 0.7071 - val_loss: 1.1131 - val_y1_output_loss: 0.3226 - val_y2_output_loss: 0.7925 - val_y1_output_root_mean_squared_error: 0.5474 - val_y2_output_root_mean_squared_error: 0.9019\n",
      "Epoch 392/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.4271 - y1_output_loss: 0.1236 - y2_output_loss: 0.3029 - y1_output_root_mean_squared_error: 0.3525 - y2_output_root_mean_squared_error: 0.5503 - val_loss: 0.8505 - val_y1_output_loss: 0.2396 - val_y2_output_loss: 0.6351 - val_y1_output_root_mean_squared_error: 0.4816 - val_y2_output_root_mean_squared_error: 0.7865\n",
      "Epoch 393/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.6395 - y1_output_loss: 0.1777 - y2_output_loss: 0.4572 - y1_output_root_mean_squared_error: 0.4232 - y2_output_root_mean_squared_error: 0.6786 - val_loss: 1.1359 - val_y1_output_loss: 0.3509 - val_y2_output_loss: 0.7873 - val_y1_output_root_mean_squared_error: 0.5892 - val_y2_output_root_mean_squared_error: 0.8881\n",
      "Epoch 394/500\n",
      "614/614 [==============================] - 0s 123us/sample - loss: 0.6180 - y1_output_loss: 0.1787 - y2_output_loss: 0.4352 - y1_output_root_mean_squared_error: 0.4231 - y2_output_root_mean_squared_error: 0.6626 - val_loss: 1.0484 - val_y1_output_loss: 0.3521 - val_y2_output_loss: 0.7248 - val_y1_output_root_mean_squared_error: 0.5839 - val_y2_output_root_mean_squared_error: 0.8411\n",
      "Epoch 395/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.5221 - y1_output_loss: 0.1557 - y2_output_loss: 0.3630 - y1_output_root_mean_squared_error: 0.3962 - y2_output_root_mean_squared_error: 0.6043 - val_loss: 0.9474 - val_y1_output_loss: 0.2572 - val_y2_output_loss: 0.6899 - val_y1_output_root_mean_squared_error: 0.4898 - val_y2_output_root_mean_squared_error: 0.8411\n",
      "Epoch 396/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.4670 - y1_output_loss: 0.1258 - y2_output_loss: 0.3429 - y1_output_root_mean_squared_error: 0.3539 - y2_output_root_mean_squared_error: 0.5846 - val_loss: 2.4909 - val_y1_output_loss: 0.4149 - val_y2_output_loss: 2.0783 - val_y1_output_root_mean_squared_error: 0.6226 - val_y2_output_root_mean_squared_error: 1.4502\n",
      "Epoch 397/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.4586 - y1_output_loss: 0.1289 - y2_output_loss: 0.3390 - y1_output_root_mean_squared_error: 0.3593 - y2_output_root_mean_squared_error: 0.5741 - val_loss: 2.3888 - val_y1_output_loss: 0.3080 - val_y2_output_loss: 2.1745 - val_y1_output_root_mean_squared_error: 0.5430 - val_y2_output_root_mean_squared_error: 1.4471\n",
      "Epoch 398/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.4078 - y1_output_loss: 0.1304 - y2_output_loss: 0.2779 - y1_output_root_mean_squared_error: 0.3610 - y2_output_root_mean_squared_error: 0.5267 - val_loss: 1.1022 - val_y1_output_loss: 0.2662 - val_y2_output_loss: 0.8753 - val_y1_output_root_mean_squared_error: 0.5023 - val_y2_output_root_mean_squared_error: 0.9219\n",
      "Epoch 399/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.5843 - y1_output_loss: 0.1640 - y2_output_loss: 0.4161 - y1_output_root_mean_squared_error: 0.4063 - y2_output_root_mean_squared_error: 0.6475 - val_loss: 0.8477 - val_y1_output_loss: 0.2255 - val_y2_output_loss: 0.6400 - val_y1_output_root_mean_squared_error: 0.4614 - val_y2_output_root_mean_squared_error: 0.7967\n",
      "Epoch 400/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 0.5123 - y1_output_loss: 0.1466 - y2_output_loss: 0.3643 - y1_output_root_mean_squared_error: 0.3839 - y2_output_root_mean_squared_error: 0.6041 - val_loss: 1.1556 - val_y1_output_loss: 0.2626 - val_y2_output_loss: 0.8839 - val_y1_output_root_mean_squared_error: 0.4997 - val_y2_output_root_mean_squared_error: 0.9518\n",
      "Epoch 401/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.6228 - y1_output_loss: 0.1809 - y2_output_loss: 0.4447 - y1_output_root_mean_squared_error: 0.4215 - y2_output_root_mean_squared_error: 0.6672 - val_loss: 2.6809 - val_y1_output_loss: 0.8400 - val_y2_output_loss: 1.9410 - val_y1_output_root_mean_squared_error: 0.8982 - val_y2_output_root_mean_squared_error: 1.3690\n",
      "Epoch 402/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.6408 - y1_output_loss: 0.1880 - y2_output_loss: 0.4487 - y1_output_root_mean_squared_error: 0.4352 - y2_output_root_mean_squared_error: 0.6719 - val_loss: 0.8139 - val_y1_output_loss: 0.2279 - val_y2_output_loss: 0.6021 - val_y1_output_root_mean_squared_error: 0.4673 - val_y2_output_root_mean_squared_error: 0.7717\n",
      "Epoch 403/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.4483 - y1_output_loss: 0.1495 - y2_output_loss: 0.2955 - y1_output_root_mean_squared_error: 0.3879 - y2_output_root_mean_squared_error: 0.5457 - val_loss: 0.8233 - val_y1_output_loss: 0.2692 - val_y2_output_loss: 0.5652 - val_y1_output_root_mean_squared_error: 0.4986 - val_y2_output_root_mean_squared_error: 0.7581\n",
      "Epoch 404/500\n",
      "614/614 [==============================] - 0s 133us/sample - loss: 0.3721 - y1_output_loss: 0.1179 - y2_output_loss: 0.2523 - y1_output_root_mean_squared_error: 0.3441 - y2_output_root_mean_squared_error: 0.5037 - val_loss: 0.9187 - val_y1_output_loss: 0.2542 - val_y2_output_loss: 0.6647 - val_y1_output_root_mean_squared_error: 0.4872 - val_y2_output_root_mean_squared_error: 0.8254\n",
      "Epoch 405/500\n",
      "614/614 [==============================] - 0s 131us/sample - loss: 0.5949 - y1_output_loss: 0.1622 - y2_output_loss: 0.4283 - y1_output_root_mean_squared_error: 0.4044 - y2_output_root_mean_squared_error: 0.6568 - val_loss: 0.8490 - val_y1_output_loss: 0.2334 - val_y2_output_loss: 0.6213 - val_y1_output_root_mean_squared_error: 0.4660 - val_y2_output_root_mean_squared_error: 0.7948\n",
      "Epoch 406/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 0.5805 - y1_output_loss: 0.1662 - y2_output_loss: 0.4104 - y1_output_root_mean_squared_error: 0.4085 - y2_output_root_mean_squared_error: 0.6431 - val_loss: 1.0862 - val_y1_output_loss: 0.2262 - val_y2_output_loss: 0.8989 - val_y1_output_root_mean_squared_error: 0.4645 - val_y2_output_root_mean_squared_error: 0.9330\n",
      "Epoch 407/500\n",
      "614/614 [==============================] - 0s 134us/sample - loss: 0.5336 - y1_output_loss: 0.1363 - y2_output_loss: 0.3953 - y1_output_root_mean_squared_error: 0.3700 - y2_output_root_mean_squared_error: 0.6298 - val_loss: 0.9763 - val_y1_output_loss: 0.2304 - val_y2_output_loss: 0.7766 - val_y1_output_root_mean_squared_error: 0.4685 - val_y2_output_root_mean_squared_error: 0.8700\n",
      "Epoch 408/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 0.5598 - y1_output_loss: 0.1516 - y2_output_loss: 0.4073 - y1_output_root_mean_squared_error: 0.3886 - y2_output_root_mean_squared_error: 0.6394 - val_loss: 1.4547 - val_y1_output_loss: 0.5281 - val_y2_output_loss: 0.9562 - val_y1_output_root_mean_squared_error: 0.7019 - val_y2_output_root_mean_squared_error: 0.9808\n",
      "Epoch 409/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.5728 - y1_output_loss: 0.1592 - y2_output_loss: 0.4163 - y1_output_root_mean_squared_error: 0.3997 - y2_output_root_mean_squared_error: 0.6427 - val_loss: 1.2984 - val_y1_output_loss: 0.2811 - val_y2_output_loss: 1.0476 - val_y1_output_root_mean_squared_error: 0.5287 - val_y2_output_root_mean_squared_error: 1.0094\n",
      "Epoch 410/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 0.4962 - y1_output_loss: 0.1444 - y2_output_loss: 0.3592 - y1_output_root_mean_squared_error: 0.3764 - y2_output_root_mean_squared_error: 0.5954 - val_loss: 2.8388 - val_y1_output_loss: 0.7937 - val_y2_output_loss: 2.0683 - val_y1_output_root_mean_squared_error: 0.8718 - val_y2_output_root_mean_squared_error: 1.4418\n",
      "Epoch 411/500\n",
      "614/614 [==============================] - 0s 135us/sample - loss: 0.7152 - y1_output_loss: 0.2155 - y2_output_loss: 0.4933 - y1_output_root_mean_squared_error: 0.4663 - y2_output_root_mean_squared_error: 0.7055 - val_loss: 0.8145 - val_y1_output_loss: 0.2423 - val_y2_output_loss: 0.5757 - val_y1_output_root_mean_squared_error: 0.4812 - val_y2_output_root_mean_squared_error: 0.7635\n",
      "Epoch 412/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 0.6324 - y1_output_loss: 0.1791 - y2_output_loss: 0.4489 - y1_output_root_mean_squared_error: 0.4238 - y2_output_root_mean_squared_error: 0.6728 - val_loss: 0.9096 - val_y1_output_loss: 0.2567 - val_y2_output_loss: 0.6491 - val_y1_output_root_mean_squared_error: 0.4970 - val_y2_output_root_mean_squared_error: 0.8140\n",
      "Epoch 413/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 0.4278 - y1_output_loss: 0.1205 - y2_output_loss: 0.3052 - y1_output_root_mean_squared_error: 0.3481 - y2_output_root_mean_squared_error: 0.5537 - val_loss: 0.8486 - val_y1_output_loss: 0.2738 - val_y2_output_loss: 0.5876 - val_y1_output_root_mean_squared_error: 0.5066 - val_y2_output_root_mean_squared_error: 0.7693\n",
      "Epoch 414/500\n",
      "614/614 [==============================] - 0s 131us/sample - loss: 0.5527 - y1_output_loss: 0.1727 - y2_output_loss: 0.3772 - y1_output_root_mean_squared_error: 0.4152 - y2_output_root_mean_squared_error: 0.6167 - val_loss: 0.9294 - val_y1_output_loss: 0.3100 - val_y2_output_loss: 0.6352 - val_y1_output_root_mean_squared_error: 0.5439 - val_y2_output_root_mean_squared_error: 0.7960\n",
      "Epoch 415/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.4613 - y1_output_loss: 0.1507 - y2_output_loss: 0.3104 - y1_output_root_mean_squared_error: 0.3870 - y2_output_root_mean_squared_error: 0.5582 - val_loss: 1.0336 - val_y1_output_loss: 0.3217 - val_y2_output_loss: 0.7122 - val_y1_output_root_mean_squared_error: 0.5589 - val_y2_output_root_mean_squared_error: 0.8492\n",
      "Epoch 416/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 0.4299 - y1_output_loss: 0.1339 - y2_output_loss: 0.2944 - y1_output_root_mean_squared_error: 0.3674 - y2_output_root_mean_squared_error: 0.5430 - val_loss: 1.4751 - val_y1_output_loss: 0.3851 - val_y2_output_loss: 1.0879 - val_y1_output_root_mean_squared_error: 0.5973 - val_y2_output_root_mean_squared_error: 1.0575\n",
      "Epoch 417/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.5331 - y1_output_loss: 0.1552 - y2_output_loss: 0.3761 - y1_output_root_mean_squared_error: 0.3928 - y2_output_root_mean_squared_error: 0.6155 - val_loss: 0.8022 - val_y1_output_loss: 0.2250 - val_y2_output_loss: 0.5860 - val_y1_output_root_mean_squared_error: 0.4558 - val_y2_output_root_mean_squared_error: 0.7710\n",
      "Epoch 418/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 0.3964 - y1_output_loss: 0.1306 - y2_output_loss: 0.2694 - y1_output_root_mean_squared_error: 0.3594 - y2_output_root_mean_squared_error: 0.5170 - val_loss: 1.0800 - val_y1_output_loss: 0.2882 - val_y2_output_loss: 0.8291 - val_y1_output_root_mean_squared_error: 0.5327 - val_y2_output_root_mean_squared_error: 0.8923\n",
      "Epoch 419/500\n",
      "614/614 [==============================] - 0s 131us/sample - loss: 0.5177 - y1_output_loss: 0.1595 - y2_output_loss: 0.3617 - y1_output_root_mean_squared_error: 0.3991 - y2_output_root_mean_squared_error: 0.5987 - val_loss: 0.8937 - val_y1_output_loss: 0.2711 - val_y2_output_loss: 0.6371 - val_y1_output_root_mean_squared_error: 0.4997 - val_y2_output_root_mean_squared_error: 0.8025\n",
      "Epoch 420/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 0.5416 - y1_output_loss: 0.1532 - y2_output_loss: 0.3840 - y1_output_root_mean_squared_error: 0.3932 - y2_output_root_mean_squared_error: 0.6221 - val_loss: 0.7867 - val_y1_output_loss: 0.2369 - val_y2_output_loss: 0.5638 - val_y1_output_root_mean_squared_error: 0.4718 - val_y2_output_root_mean_squared_error: 0.7511\n",
      "Epoch 421/500\n",
      "614/614 [==============================] - 0s 131us/sample - loss: 0.4428 - y1_output_loss: 0.1367 - y2_output_loss: 0.3053 - y1_output_root_mean_squared_error: 0.3709 - y2_output_root_mean_squared_error: 0.5524 - val_loss: 0.7969 - val_y1_output_loss: 0.2270 - val_y2_output_loss: 0.5813 - val_y1_output_root_mean_squared_error: 0.4642 - val_y2_output_root_mean_squared_error: 0.7625\n",
      "Epoch 422/500\n",
      "614/614 [==============================] - 0s 131us/sample - loss: 0.5613 - y1_output_loss: 0.1696 - y2_output_loss: 0.3894 - y1_output_root_mean_squared_error: 0.4112 - y2_output_root_mean_squared_error: 0.6262 - val_loss: 1.0590 - val_y1_output_loss: 0.2783 - val_y2_output_loss: 0.7838 - val_y1_output_root_mean_squared_error: 0.5226 - val_y2_output_root_mean_squared_error: 0.8865\n",
      "Epoch 423/500\n",
      "614/614 [==============================] - 0s 122us/sample - loss: 0.7107 - y1_output_loss: 0.1793 - y2_output_loss: 0.5255 - y1_output_root_mean_squared_error: 0.4252 - y2_output_root_mean_squared_error: 0.7280 - val_loss: 0.7443 - val_y1_output_loss: 0.2031 - val_y2_output_loss: 0.5490 - val_y1_output_root_mean_squared_error: 0.4361 - val_y2_output_root_mean_squared_error: 0.7444\n",
      "Epoch 424/500\n",
      "614/614 [==============================] - 0s 123us/sample - loss: 0.6619 - y1_output_loss: 0.1564 - y2_output_loss: 0.5020 - y1_output_root_mean_squared_error: 0.3965 - y2_output_root_mean_squared_error: 0.7104 - val_loss: 0.9474 - val_y1_output_loss: 0.2902 - val_y2_output_loss: 0.6732 - val_y1_output_root_mean_squared_error: 0.5227 - val_y2_output_root_mean_squared_error: 0.8211\n",
      "Epoch 425/500\n",
      "614/614 [==============================] - 0s 123us/sample - loss: 0.4407 - y1_output_loss: 0.1233 - y2_output_loss: 0.3162 - y1_output_root_mean_squared_error: 0.3518 - y2_output_root_mean_squared_error: 0.5630 - val_loss: 0.9156 - val_y1_output_loss: 0.2285 - val_y2_output_loss: 0.7096 - val_y1_output_root_mean_squared_error: 0.4673 - val_y2_output_root_mean_squared_error: 0.8350\n",
      "Epoch 426/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.4553 - y1_output_loss: 0.1332 - y2_output_loss: 0.3270 - y1_output_root_mean_squared_error: 0.3622 - y2_output_root_mean_squared_error: 0.5694 - val_loss: 1.6960 - val_y1_output_loss: 0.3888 - val_y2_output_loss: 1.2959 - val_y1_output_root_mean_squared_error: 0.6039 - val_y2_output_root_mean_squared_error: 1.1538\n",
      "Epoch 427/500\n",
      "614/614 [==============================] - 0s 132us/sample - loss: 0.4142 - y1_output_loss: 0.1356 - y2_output_loss: 0.2767 - y1_output_root_mean_squared_error: 0.3681 - y2_output_root_mean_squared_error: 0.5279 - val_loss: 0.9121 - val_y1_output_loss: 0.2166 - val_y2_output_loss: 0.6899 - val_y1_output_root_mean_squared_error: 0.4529 - val_y2_output_root_mean_squared_error: 0.8408\n",
      "Epoch 428/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.4744 - y1_output_loss: 0.1468 - y2_output_loss: 0.3245 - y1_output_root_mean_squared_error: 0.3841 - y2_output_root_mean_squared_error: 0.5717 - val_loss: 0.9147 - val_y1_output_loss: 0.2365 - val_y2_output_loss: 0.6807 - val_y1_output_root_mean_squared_error: 0.4731 - val_y2_output_root_mean_squared_error: 0.8312\n",
      "Epoch 429/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 0.4453 - y1_output_loss: 0.1422 - y2_output_loss: 0.2999 - y1_output_root_mean_squared_error: 0.3786 - y2_output_root_mean_squared_error: 0.5495 - val_loss: 0.7835 - val_y1_output_loss: 0.2270 - val_y2_output_loss: 0.5762 - val_y1_output_root_mean_squared_error: 0.4573 - val_y2_output_root_mean_squared_error: 0.7579\n",
      "Epoch 430/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.4629 - y1_output_loss: 0.1460 - y2_output_loss: 0.3174 - y1_output_root_mean_squared_error: 0.3817 - y2_output_root_mean_squared_error: 0.5632 - val_loss: 1.0578 - val_y1_output_loss: 0.3266 - val_y2_output_loss: 0.7375 - val_y1_output_root_mean_squared_error: 0.5508 - val_y2_output_root_mean_squared_error: 0.8685\n",
      "Epoch 431/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 0.4184 - y1_output_loss: 0.1394 - y2_output_loss: 0.2835 - y1_output_root_mean_squared_error: 0.3727 - y2_output_root_mean_squared_error: 0.5286 - val_loss: 1.1623 - val_y1_output_loss: 0.2150 - val_y2_output_loss: 0.9919 - val_y1_output_root_mean_squared_error: 0.4467 - val_y2_output_root_mean_squared_error: 0.9812\n",
      "Epoch 432/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.4470 - y1_output_loss: 0.1384 - y2_output_loss: 0.3097 - y1_output_root_mean_squared_error: 0.3701 - y2_output_root_mean_squared_error: 0.5568 - val_loss: 0.8303 - val_y1_output_loss: 0.2299 - val_y2_output_loss: 0.6121 - val_y1_output_root_mean_squared_error: 0.4744 - val_y2_output_root_mean_squared_error: 0.7780\n",
      "Epoch 433/500\n",
      "614/614 [==============================] - 0s 123us/sample - loss: 0.3619 - y1_output_loss: 0.1248 - y2_output_loss: 0.2374 - y1_output_root_mean_squared_error: 0.3519 - y2_output_root_mean_squared_error: 0.4879 - val_loss: 1.0668 - val_y1_output_loss: 0.3034 - val_y2_output_loss: 0.7848 - val_y1_output_root_mean_squared_error: 0.5458 - val_y2_output_root_mean_squared_error: 0.8769\n",
      "Epoch 434/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.4054 - y1_output_loss: 0.1378 - y2_output_loss: 0.2674 - y1_output_root_mean_squared_error: 0.3704 - y2_output_root_mean_squared_error: 0.5179 - val_loss: 0.8809 - val_y1_output_loss: 0.2808 - val_y2_output_loss: 0.6038 - val_y1_output_root_mean_squared_error: 0.5192 - val_y2_output_root_mean_squared_error: 0.7819\n",
      "Epoch 435/500\n",
      "614/614 [==============================] - 0s 123us/sample - loss: 0.4592 - y1_output_loss: 0.1499 - y2_output_loss: 0.3176 - y1_output_root_mean_squared_error: 0.3867 - y2_output_root_mean_squared_error: 0.5565 - val_loss: 2.1331 - val_y1_output_loss: 0.4805 - val_y2_output_loss: 1.7215 - val_y1_output_root_mean_squared_error: 0.6923 - val_y2_output_root_mean_squared_error: 1.2860\n",
      "Epoch 436/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.4697 - y1_output_loss: 0.1558 - y2_output_loss: 0.3171 - y1_output_root_mean_squared_error: 0.3928 - y2_output_root_mean_squared_error: 0.5616 - val_loss: 0.9993 - val_y1_output_loss: 0.2227 - val_y2_output_loss: 0.8221 - val_y1_output_root_mean_squared_error: 0.4578 - val_y2_output_root_mean_squared_error: 0.8887\n",
      "Epoch 437/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.3773 - y1_output_loss: 0.1132 - y2_output_loss: 0.2618 - y1_output_root_mean_squared_error: 0.3371 - y2_output_root_mean_squared_error: 0.5135 - val_loss: 0.8247 - val_y1_output_loss: 0.2498 - val_y2_output_loss: 0.5743 - val_y1_output_root_mean_squared_error: 0.4912 - val_y2_output_root_mean_squared_error: 0.7638\n",
      "Epoch 438/500\n",
      "614/614 [==============================] - 0s 123us/sample - loss: 0.4237 - y1_output_loss: 0.1276 - y2_output_loss: 0.2929 - y1_output_root_mean_squared_error: 0.3584 - y2_output_root_mean_squared_error: 0.5434 - val_loss: 0.7659 - val_y1_output_loss: 0.2121 - val_y2_output_loss: 0.5593 - val_y1_output_root_mean_squared_error: 0.4529 - val_y2_output_root_mean_squared_error: 0.7489\n",
      "Epoch 439/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.3954 - y1_output_loss: 0.1255 - y2_output_loss: 0.2672 - y1_output_root_mean_squared_error: 0.3545 - y2_output_root_mean_squared_error: 0.5194 - val_loss: 0.7721 - val_y1_output_loss: 0.2059 - val_y2_output_loss: 0.5753 - val_y1_output_root_mean_squared_error: 0.4374 - val_y2_output_root_mean_squared_error: 0.7621\n",
      "Epoch 440/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.3593 - y1_output_loss: 0.1188 - y2_output_loss: 0.2374 - y1_output_root_mean_squared_error: 0.3462 - y2_output_root_mean_squared_error: 0.4893 - val_loss: 0.7440 - val_y1_output_loss: 0.2107 - val_y2_output_loss: 0.5402 - val_y1_output_root_mean_squared_error: 0.4493 - val_y2_output_root_mean_squared_error: 0.7363\n",
      "Epoch 441/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.3542 - y1_output_loss: 0.1234 - y2_output_loss: 0.2308 - y1_output_root_mean_squared_error: 0.3506 - y2_output_root_mean_squared_error: 0.4809 - val_loss: 1.2672 - val_y1_output_loss: 0.3315 - val_y2_output_loss: 0.9250 - val_y1_output_root_mean_squared_error: 0.5611 - val_y2_output_root_mean_squared_error: 0.9759\n",
      "Epoch 442/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.4444 - y1_output_loss: 0.1317 - y2_output_loss: 0.3089 - y1_output_root_mean_squared_error: 0.3644 - y2_output_root_mean_squared_error: 0.5582 - val_loss: 0.7784 - val_y1_output_loss: 0.1953 - val_y2_output_loss: 0.5896 - val_y1_output_root_mean_squared_error: 0.4291 - val_y2_output_root_mean_squared_error: 0.7709\n",
      "Epoch 443/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 0.3470 - y1_output_loss: 0.1100 - y2_output_loss: 0.2356 - y1_output_root_mean_squared_error: 0.3328 - y2_output_root_mean_squared_error: 0.4861 - val_loss: 0.8867 - val_y1_output_loss: 0.1977 - val_y2_output_loss: 0.6839 - val_y1_output_root_mean_squared_error: 0.4318 - val_y2_output_root_mean_squared_error: 0.8368\n",
      "Epoch 444/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 0.3969 - y1_output_loss: 0.1245 - y2_output_loss: 0.2695 - y1_output_root_mean_squared_error: 0.3536 - y2_output_root_mean_squared_error: 0.5214 - val_loss: 0.7411 - val_y1_output_loss: 0.2020 - val_y2_output_loss: 0.5645 - val_y1_output_root_mean_squared_error: 0.4330 - val_y2_output_root_mean_squared_error: 0.7441\n",
      "Epoch 445/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.3815 - y1_output_loss: 0.1310 - y2_output_loss: 0.2496 - y1_output_root_mean_squared_error: 0.3629 - y2_output_root_mean_squared_error: 0.4998 - val_loss: 0.8702 - val_y1_output_loss: 0.2595 - val_y2_output_loss: 0.6388 - val_y1_output_root_mean_squared_error: 0.4937 - val_y2_output_root_mean_squared_error: 0.7915\n",
      "Epoch 446/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.5443 - y1_output_loss: 0.1645 - y2_output_loss: 0.3758 - y1_output_root_mean_squared_error: 0.4067 - y2_output_root_mean_squared_error: 0.6155 - val_loss: 1.0002 - val_y1_output_loss: 0.2667 - val_y2_output_loss: 0.7302 - val_y1_output_root_mean_squared_error: 0.5026 - val_y2_output_root_mean_squared_error: 0.8646\n",
      "Epoch 447/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.3662 - y1_output_loss: 0.1114 - y2_output_loss: 0.2523 - y1_output_root_mean_squared_error: 0.3352 - y2_output_root_mean_squared_error: 0.5038 - val_loss: 0.7757 - val_y1_output_loss: 0.2047 - val_y2_output_loss: 0.5953 - val_y1_output_root_mean_squared_error: 0.4357 - val_y2_output_root_mean_squared_error: 0.7654\n",
      "Epoch 448/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.5341 - y1_output_loss: 0.1567 - y2_output_loss: 0.3795 - y1_output_root_mean_squared_error: 0.3964 - y2_output_root_mean_squared_error: 0.6140 - val_loss: 1.1591 - val_y1_output_loss: 0.3159 - val_y2_output_loss: 0.8387 - val_y1_output_root_mean_squared_error: 0.5579 - val_y2_output_root_mean_squared_error: 0.9208\n",
      "Epoch 449/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.6218 - y1_output_loss: 0.1541 - y2_output_loss: 0.4642 - y1_output_root_mean_squared_error: 0.3932 - y2_output_root_mean_squared_error: 0.6835 - val_loss: 1.4573 - val_y1_output_loss: 0.4846 - val_y2_output_loss: 0.9876 - val_y1_output_root_mean_squared_error: 0.6759 - val_y2_output_root_mean_squared_error: 1.0003\n",
      "Epoch 450/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.4301 - y1_output_loss: 0.1363 - y2_output_loss: 0.2994 - y1_output_root_mean_squared_error: 0.3674 - y2_output_root_mean_squared_error: 0.5432 - val_loss: 1.2298 - val_y1_output_loss: 0.2654 - val_y2_output_loss: 0.9629 - val_y1_output_root_mean_squared_error: 0.4971 - val_y2_output_root_mean_squared_error: 0.9913\n",
      "Epoch 451/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.3269 - y1_output_loss: 0.1121 - y2_output_loss: 0.2128 - y1_output_root_mean_squared_error: 0.3359 - y2_output_root_mean_squared_error: 0.4627 - val_loss: 0.7756 - val_y1_output_loss: 0.2098 - val_y2_output_loss: 0.5878 - val_y1_output_root_mean_squared_error: 0.4417 - val_y2_output_root_mean_squared_error: 0.7619\n",
      "Epoch 452/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.6107 - y1_output_loss: 0.1621 - y2_output_loss: 0.4450 - y1_output_root_mean_squared_error: 0.4038 - y2_output_root_mean_squared_error: 0.6691 - val_loss: 0.9470 - val_y1_output_loss: 0.2599 - val_y2_output_loss: 0.6920 - val_y1_output_root_mean_squared_error: 0.4884 - val_y2_output_root_mean_squared_error: 0.8417\n",
      "Epoch 453/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 0.3338 - y1_output_loss: 0.1122 - y2_output_loss: 0.2224 - y1_output_root_mean_squared_error: 0.3357 - y2_output_root_mean_squared_error: 0.4702 - val_loss: 1.3959 - val_y1_output_loss: 0.2206 - val_y2_output_loss: 1.2410 - val_y1_output_root_mean_squared_error: 0.4583 - val_y2_output_root_mean_squared_error: 1.0889\n",
      "Epoch 454/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.3777 - y1_output_loss: 0.1200 - y2_output_loss: 0.2568 - y1_output_root_mean_squared_error: 0.3475 - y2_output_root_mean_squared_error: 0.5069 - val_loss: 0.9278 - val_y1_output_loss: 0.2721 - val_y2_output_loss: 0.6622 - val_y1_output_root_mean_squared_error: 0.5014 - val_y2_output_root_mean_squared_error: 0.8224\n",
      "Epoch 455/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.3325 - y1_output_loss: 0.1054 - y2_output_loss: 0.2247 - y1_output_root_mean_squared_error: 0.3259 - y2_output_root_mean_squared_error: 0.4757 - val_loss: 0.7390 - val_y1_output_loss: 0.2034 - val_y2_output_loss: 0.5424 - val_y1_output_root_mean_squared_error: 0.4373 - val_y2_output_root_mean_squared_error: 0.7401\n",
      "Epoch 456/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 0.5885 - y1_output_loss: 0.1491 - y2_output_loss: 0.4400 - y1_output_root_mean_squared_error: 0.3871 - y2_output_root_mean_squared_error: 0.6623 - val_loss: 1.3613 - val_y1_output_loss: 0.2608 - val_y2_output_loss: 1.1602 - val_y1_output_root_mean_squared_error: 0.5081 - val_y2_output_root_mean_squared_error: 1.0503\n",
      "Epoch 457/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.3913 - y1_output_loss: 0.1253 - y2_output_loss: 0.2721 - y1_output_root_mean_squared_error: 0.3538 - y2_output_root_mean_squared_error: 0.5159 - val_loss: 2.2090 - val_y1_output_loss: 0.2363 - val_y2_output_loss: 2.0510 - val_y1_output_root_mean_squared_error: 0.4740 - val_y2_output_root_mean_squared_error: 1.4087\n",
      "Epoch 458/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 0.6267 - y1_output_loss: 0.1662 - y2_output_loss: 0.4615 - y1_output_root_mean_squared_error: 0.4082 - y2_output_root_mean_squared_error: 0.6783 - val_loss: 1.9858 - val_y1_output_loss: 0.2650 - val_y2_output_loss: 1.8097 - val_y1_output_root_mean_squared_error: 0.5078 - val_y2_output_root_mean_squared_error: 1.3145\n",
      "Epoch 459/500\n",
      "614/614 [==============================] - 0s 131us/sample - loss: 0.4675 - y1_output_loss: 0.1394 - y2_output_loss: 0.3263 - y1_output_root_mean_squared_error: 0.3748 - y2_output_root_mean_squared_error: 0.5719 - val_loss: 0.7331 - val_y1_output_loss: 0.2060 - val_y2_output_loss: 0.5415 - val_y1_output_root_mean_squared_error: 0.4397 - val_y2_output_root_mean_squared_error: 0.7347\n",
      "Epoch 460/500\n",
      "614/614 [==============================] - 0s 131us/sample - loss: 0.6217 - y1_output_loss: 0.1883 - y2_output_loss: 0.4335 - y1_output_root_mean_squared_error: 0.4347 - y2_output_root_mean_squared_error: 0.6578 - val_loss: 1.5930 - val_y1_output_loss: 0.3089 - val_y2_output_loss: 1.2834 - val_y1_output_root_mean_squared_error: 0.5335 - val_y2_output_root_mean_squared_error: 1.1439\n",
      "Epoch 461/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 0.5600 - y1_output_loss: 0.1607 - y2_output_loss: 0.3970 - y1_output_root_mean_squared_error: 0.4023 - y2_output_root_mean_squared_error: 0.6310 - val_loss: 2.1966 - val_y1_output_loss: 0.5302 - val_y2_output_loss: 1.7462 - val_y1_output_root_mean_squared_error: 0.7240 - val_y2_output_root_mean_squared_error: 1.2932\n",
      "Epoch 462/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 0.4044 - y1_output_loss: 0.1326 - y2_output_loss: 0.2718 - y1_output_root_mean_squared_error: 0.3647 - y2_output_root_mean_squared_error: 0.5209 - val_loss: 0.9406 - val_y1_output_loss: 0.2616 - val_y2_output_loss: 0.6897 - val_y1_output_root_mean_squared_error: 0.5051 - val_y2_output_root_mean_squared_error: 0.8279\n",
      "Epoch 463/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.3592 - y1_output_loss: 0.1171 - y2_output_loss: 0.2471 - y1_output_root_mean_squared_error: 0.3438 - y2_output_root_mean_squared_error: 0.4909 - val_loss: 2.0871 - val_y1_output_loss: 0.2857 - val_y2_output_loss: 1.7801 - val_y1_output_root_mean_squared_error: 0.5177 - val_y2_output_root_mean_squared_error: 1.3487\n",
      "Epoch 464/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.5956 - y1_output_loss: 0.1624 - y2_output_loss: 0.4308 - y1_output_root_mean_squared_error: 0.4036 - y2_output_root_mean_squared_error: 0.6578 - val_loss: 0.7388 - val_y1_output_loss: 0.2486 - val_y2_output_loss: 0.5022 - val_y1_output_root_mean_squared_error: 0.4823 - val_y2_output_root_mean_squared_error: 0.7114\n",
      "Epoch 465/500\n",
      "614/614 [==============================] - 0s 124us/sample - loss: 0.3516 - y1_output_loss: 0.1104 - y2_output_loss: 0.2399 - y1_output_root_mean_squared_error: 0.3334 - y2_output_root_mean_squared_error: 0.4903 - val_loss: 0.8880 - val_y1_output_loss: 0.3406 - val_y2_output_loss: 0.5643 - val_y1_output_root_mean_squared_error: 0.5629 - val_y2_output_root_mean_squared_error: 0.7558\n",
      "Epoch 466/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.3683 - y1_output_loss: 0.1280 - y2_output_loss: 0.2463 - y1_output_root_mean_squared_error: 0.3514 - y2_output_root_mean_squared_error: 0.4947 - val_loss: 1.3606 - val_y1_output_loss: 0.4613 - val_y2_output_loss: 0.9055 - val_y1_output_root_mean_squared_error: 0.6587 - val_y2_output_root_mean_squared_error: 0.9626\n",
      "Epoch 467/500\n",
      "614/614 [==============================] - 0s 127us/sample - loss: 0.3812 - y1_output_loss: 0.1317 - y2_output_loss: 0.2490 - y1_output_root_mean_squared_error: 0.3620 - y2_output_root_mean_squared_error: 0.5001 - val_loss: 0.9290 - val_y1_output_loss: 0.2607 - val_y2_output_loss: 0.6702 - val_y1_output_root_mean_squared_error: 0.4914 - val_y2_output_root_mean_squared_error: 0.8292\n",
      "Epoch 468/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.3889 - y1_output_loss: 0.1259 - y2_output_loss: 0.2604 - y1_output_root_mean_squared_error: 0.3560 - y2_output_root_mean_squared_error: 0.5120 - val_loss: 0.7744 - val_y1_output_loss: 0.2516 - val_y2_output_loss: 0.5468 - val_y1_output_root_mean_squared_error: 0.4793 - val_y2_output_root_mean_squared_error: 0.7380\n",
      "Epoch 469/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.4653 - y1_output_loss: 0.1466 - y2_output_loss: 0.3158 - y1_output_root_mean_squared_error: 0.3842 - y2_output_root_mean_squared_error: 0.5636 - val_loss: 0.7311 - val_y1_output_loss: 0.2113 - val_y2_output_loss: 0.5234 - val_y1_output_root_mean_squared_error: 0.4472 - val_y2_output_root_mean_squared_error: 0.7288\n",
      "Epoch 470/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 0.3784 - y1_output_loss: 0.1133 - y2_output_loss: 0.2645 - y1_output_root_mean_squared_error: 0.3367 - y2_output_root_mean_squared_error: 0.5148 - val_loss: 1.0655 - val_y1_output_loss: 0.2693 - val_y2_output_loss: 0.7966 - val_y1_output_root_mean_squared_error: 0.5015 - val_y2_output_root_mean_squared_error: 0.9022\n",
      "Epoch 471/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 0.3670 - y1_output_loss: 0.1280 - y2_output_loss: 0.2436 - y1_output_root_mean_squared_error: 0.3576 - y2_output_root_mean_squared_error: 0.4890 - val_loss: 1.6937 - val_y1_output_loss: 0.5917 - val_y2_output_loss: 1.1377 - val_y1_output_root_mean_squared_error: 0.7428 - val_y2_output_root_mean_squared_error: 1.0686\n",
      "Epoch 472/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 0.4397 - y1_output_loss: 0.1442 - y2_output_loss: 0.2935 - y1_output_root_mean_squared_error: 0.3796 - y2_output_root_mean_squared_error: 0.5437 - val_loss: 0.9017 - val_y1_output_loss: 0.2927 - val_y2_output_loss: 0.6167 - val_y1_output_root_mean_squared_error: 0.5368 - val_y2_output_root_mean_squared_error: 0.7833\n",
      "Epoch 473/500\n",
      "614/614 [==============================] - 0s 132us/sample - loss: 0.4034 - y1_output_loss: 0.1373 - y2_output_loss: 0.2631 - y1_output_root_mean_squared_error: 0.3723 - y2_output_root_mean_squared_error: 0.5146 - val_loss: 0.7079 - val_y1_output_loss: 0.1913 - val_y2_output_loss: 0.5285 - val_y1_output_root_mean_squared_error: 0.4258 - val_y2_output_root_mean_squared_error: 0.7257\n",
      "Epoch 474/500\n",
      "614/614 [==============================] - 0s 136us/sample - loss: 0.3130 - y1_output_loss: 0.1138 - y2_output_loss: 0.1999 - y1_output_root_mean_squared_error: 0.3379 - y2_output_root_mean_squared_error: 0.4459 - val_loss: 0.8933 - val_y1_output_loss: 0.2844 - val_y2_output_loss: 0.6221 - val_y1_output_root_mean_squared_error: 0.5143 - val_y2_output_root_mean_squared_error: 0.7930\n",
      "Epoch 475/500\n",
      "614/614 [==============================] - 0s 133us/sample - loss: 0.5566 - y1_output_loss: 0.1525 - y2_output_loss: 0.3990 - y1_output_root_mean_squared_error: 0.3923 - y2_output_root_mean_squared_error: 0.6346 - val_loss: 0.7743 - val_y1_output_loss: 0.2506 - val_y2_output_loss: 0.5329 - val_y1_output_root_mean_squared_error: 0.4825 - val_y2_output_root_mean_squared_error: 0.7358\n",
      "Epoch 476/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.3701 - y1_output_loss: 0.1184 - y2_output_loss: 0.2529 - y1_output_root_mean_squared_error: 0.3429 - y2_output_root_mean_squared_error: 0.5026 - val_loss: 1.0048 - val_y1_output_loss: 0.3430 - val_y2_output_loss: 0.6707 - val_y1_output_root_mean_squared_error: 0.5622 - val_y2_output_root_mean_squared_error: 0.8299\n",
      "Epoch 477/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.4159 - y1_output_loss: 0.1383 - y2_output_loss: 0.2810 - y1_output_root_mean_squared_error: 0.3694 - y2_output_root_mean_squared_error: 0.5286 - val_loss: 1.7405 - val_y1_output_loss: 0.5325 - val_y2_output_loss: 1.2011 - val_y1_output_root_mean_squared_error: 0.7239 - val_y2_output_root_mean_squared_error: 1.1030\n",
      "Epoch 478/500\n",
      "614/614 [==============================] - 0s 135us/sample - loss: 0.4507 - y1_output_loss: 0.1368 - y2_output_loss: 0.3119 - y1_output_root_mean_squared_error: 0.3698 - y2_output_root_mean_squared_error: 0.5603 - val_loss: 0.8418 - val_y1_output_loss: 0.2648 - val_y2_output_loss: 0.5876 - val_y1_output_root_mean_squared_error: 0.5018 - val_y2_output_root_mean_squared_error: 0.7681\n",
      "Epoch 479/500\n",
      "614/614 [==============================] - 0s 131us/sample - loss: 0.3945 - y1_output_loss: 0.1329 - y2_output_loss: 0.2625 - y1_output_root_mean_squared_error: 0.3656 - y2_output_root_mean_squared_error: 0.5107 - val_loss: 1.0778 - val_y1_output_loss: 0.2109 - val_y2_output_loss: 0.9077 - val_y1_output_root_mean_squared_error: 0.4489 - val_y2_output_root_mean_squared_error: 0.9361\n",
      "Epoch 480/500\n",
      "614/614 [==============================] - 0s 132us/sample - loss: 0.5659 - y1_output_loss: 0.1573 - y2_output_loss: 0.4053 - y1_output_root_mean_squared_error: 0.3984 - y2_output_root_mean_squared_error: 0.6381 - val_loss: 0.8415 - val_y1_output_loss: 0.3041 - val_y2_output_loss: 0.5608 - val_y1_output_root_mean_squared_error: 0.5300 - val_y2_output_root_mean_squared_error: 0.7487\n",
      "Epoch 481/500\n",
      "614/614 [==============================] - 0s 132us/sample - loss: 0.3494 - y1_output_loss: 0.1025 - y2_output_loss: 0.2451 - y1_output_root_mean_squared_error: 0.3215 - y2_output_root_mean_squared_error: 0.4961 - val_loss: 0.6824 - val_y1_output_loss: 0.1886 - val_y2_output_loss: 0.5001 - val_y1_output_root_mean_squared_error: 0.4231 - val_y2_output_root_mean_squared_error: 0.7095\n",
      "Epoch 482/500\n",
      "614/614 [==============================] - 0s 131us/sample - loss: 0.5270 - y1_output_loss: 0.1338 - y2_output_loss: 0.3932 - y1_output_root_mean_squared_error: 0.3670 - y2_output_root_mean_squared_error: 0.6264 - val_loss: 1.2778 - val_y1_output_loss: 0.2257 - val_y2_output_loss: 1.1059 - val_y1_output_root_mean_squared_error: 0.4698 - val_y2_output_root_mean_squared_error: 1.0281\n",
      "Epoch 483/500\n",
      "614/614 [==============================] - 0s 136us/sample - loss: 0.3585 - y1_output_loss: 0.1179 - y2_output_loss: 0.2407 - y1_output_root_mean_squared_error: 0.3423 - y2_output_root_mean_squared_error: 0.4913 - val_loss: 0.9190 - val_y1_output_loss: 0.2927 - val_y2_output_loss: 0.6423 - val_y1_output_root_mean_squared_error: 0.5323 - val_y2_output_root_mean_squared_error: 0.7973\n",
      "Epoch 484/500\n",
      "614/614 [==============================] - 0s 131us/sample - loss: 0.2984 - y1_output_loss: 0.1151 - y2_output_loss: 0.1818 - y1_output_root_mean_squared_error: 0.3407 - y2_output_root_mean_squared_error: 0.4270 - val_loss: 0.7828 - val_y1_output_loss: 0.2012 - val_y2_output_loss: 0.5789 - val_y1_output_root_mean_squared_error: 0.4377 - val_y2_output_root_mean_squared_error: 0.7689\n",
      "Epoch 485/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 0.3393 - y1_output_loss: 0.1083 - y2_output_loss: 0.2291 - y1_output_root_mean_squared_error: 0.3290 - y2_output_root_mean_squared_error: 0.4807 - val_loss: 0.9002 - val_y1_output_loss: 0.2782 - val_y2_output_loss: 0.6245 - val_y1_output_root_mean_squared_error: 0.5156 - val_y2_output_root_mean_squared_error: 0.7964\n",
      "Epoch 486/500\n",
      "614/614 [==============================] - 0s 132us/sample - loss: 0.3549 - y1_output_loss: 0.1143 - y2_output_loss: 0.2387 - y1_output_root_mean_squared_error: 0.3383 - y2_output_root_mean_squared_error: 0.4903 - val_loss: 0.6916 - val_y1_output_loss: 0.2233 - val_y2_output_loss: 0.4789 - val_y1_output_root_mean_squared_error: 0.4587 - val_y2_output_root_mean_squared_error: 0.6937\n",
      "Epoch 487/500\n",
      "614/614 [==============================] - 0s 133us/sample - loss: 0.3335 - y1_output_loss: 0.1170 - y2_output_loss: 0.2160 - y1_output_root_mean_squared_error: 0.3430 - y2_output_root_mean_squared_error: 0.4646 - val_loss: 0.8689 - val_y1_output_loss: 0.2091 - val_y2_output_loss: 0.6580 - val_y1_output_root_mean_squared_error: 0.4394 - val_y2_output_root_mean_squared_error: 0.8221\n",
      "Epoch 488/500\n",
      "614/614 [==============================] - 0s 130us/sample - loss: 0.4429 - y1_output_loss: 0.1505 - y2_output_loss: 0.2904 - y1_output_root_mean_squared_error: 0.3897 - y2_output_root_mean_squared_error: 0.5395 - val_loss: 0.7915 - val_y1_output_loss: 0.2738 - val_y2_output_loss: 0.5293 - val_y1_output_root_mean_squared_error: 0.5066 - val_y2_output_root_mean_squared_error: 0.7313\n",
      "Epoch 489/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 0.3871 - y1_output_loss: 0.1454 - y2_output_loss: 0.2386 - y1_output_root_mean_squared_error: 0.3830 - y2_output_root_mean_squared_error: 0.4904 - val_loss: 0.7720 - val_y1_output_loss: 0.2036 - val_y2_output_loss: 0.5719 - val_y1_output_root_mean_squared_error: 0.4382 - val_y2_output_root_mean_squared_error: 0.7616\n",
      "Epoch 490/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.3305 - y1_output_loss: 0.1209 - y2_output_loss: 0.2135 - y1_output_root_mean_squared_error: 0.3451 - y2_output_root_mean_squared_error: 0.4598 - val_loss: 1.3914 - val_y1_output_loss: 0.4392 - val_y2_output_loss: 0.9945 - val_y1_output_root_mean_squared_error: 0.6552 - val_y2_output_root_mean_squared_error: 0.9809\n",
      "Epoch 491/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.3372 - y1_output_loss: 0.1256 - y2_output_loss: 0.2229 - y1_output_root_mean_squared_error: 0.3502 - y2_output_root_mean_squared_error: 0.4632 - val_loss: 1.5928 - val_y1_output_loss: 0.2907 - val_y2_output_loss: 1.3630 - val_y1_output_root_mean_squared_error: 0.5403 - val_y2_output_root_mean_squared_error: 1.1405\n",
      "Epoch 492/500\n",
      "614/614 [==============================] - 0s 125us/sample - loss: 0.4118 - y1_output_loss: 0.1279 - y2_output_loss: 0.2804 - y1_output_root_mean_squared_error: 0.3591 - y2_output_root_mean_squared_error: 0.5319 - val_loss: 0.7342 - val_y1_output_loss: 0.2181 - val_y2_output_loss: 0.5216 - val_y1_output_root_mean_squared_error: 0.4555 - val_y2_output_root_mean_squared_error: 0.7258\n",
      "Epoch 493/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.3392 - y1_output_loss: 0.1194 - y2_output_loss: 0.2184 - y1_output_root_mean_squared_error: 0.3460 - y2_output_root_mean_squared_error: 0.4685 - val_loss: 0.6965 - val_y1_output_loss: 0.1970 - val_y2_output_loss: 0.5095 - val_y1_output_root_mean_squared_error: 0.4267 - val_y2_output_root_mean_squared_error: 0.7172\n",
      "Epoch 494/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.3778 - y1_output_loss: 0.1275 - y2_output_loss: 0.2477 - y1_output_root_mean_squared_error: 0.3583 - y2_output_root_mean_squared_error: 0.4994 - val_loss: 0.7498 - val_y1_output_loss: 0.1961 - val_y2_output_loss: 0.5527 - val_y1_output_root_mean_squared_error: 0.4319 - val_y2_output_root_mean_squared_error: 0.7505\n",
      "Epoch 495/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.3839 - y1_output_loss: 0.1202 - y2_output_loss: 0.2618 - y1_output_root_mean_squared_error: 0.3475 - y2_output_root_mean_squared_error: 0.5129 - val_loss: 0.8585 - val_y1_output_loss: 0.2432 - val_y2_output_loss: 0.6221 - val_y1_output_root_mean_squared_error: 0.4831 - val_y2_output_root_mean_squared_error: 0.7906\n",
      "Epoch 496/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.3825 - y1_output_loss: 0.1268 - y2_output_loss: 0.2548 - y1_output_root_mean_squared_error: 0.3570 - y2_output_root_mean_squared_error: 0.5051 - val_loss: 0.8139 - val_y1_output_loss: 0.2668 - val_y2_output_loss: 0.5478 - val_y1_output_root_mean_squared_error: 0.5041 - val_y2_output_root_mean_squared_error: 0.7482\n",
      "Epoch 497/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.3163 - y1_output_loss: 0.1035 - y2_output_loss: 0.2146 - y1_output_root_mean_squared_error: 0.3217 - y2_output_root_mean_squared_error: 0.4614 - val_loss: 0.8657 - val_y1_output_loss: 0.2336 - val_y2_output_loss: 0.6322 - val_y1_output_root_mean_squared_error: 0.4691 - val_y2_output_root_mean_squared_error: 0.8035\n",
      "Epoch 498/500\n",
      "614/614 [==============================] - 0s 126us/sample - loss: 0.4041 - y1_output_loss: 0.1368 - y2_output_loss: 0.2796 - y1_output_root_mean_squared_error: 0.3676 - y2_output_root_mean_squared_error: 0.5186 - val_loss: 1.8678 - val_y1_output_loss: 0.2850 - val_y2_output_loss: 1.6518 - val_y1_output_root_mean_squared_error: 0.5299 - val_y2_output_root_mean_squared_error: 1.2598\n",
      "Epoch 499/500\n",
      "614/614 [==============================] - 0s 129us/sample - loss: 0.4089 - y1_output_loss: 0.1332 - y2_output_loss: 0.3007 - y1_output_root_mean_squared_error: 0.3643 - y2_output_root_mean_squared_error: 0.5256 - val_loss: 1.4734 - val_y1_output_loss: 0.4473 - val_y2_output_loss: 1.0144 - val_y1_output_root_mean_squared_error: 0.6630 - val_y2_output_root_mean_squared_error: 1.0168\n",
      "Epoch 500/500\n",
      "614/614 [==============================] - 0s 128us/sample - loss: 0.4112 - y1_output_loss: 0.1311 - y2_output_loss: 0.2783 - y1_output_root_mean_squared_error: 0.3634 - y2_output_root_mean_squared_error: 0.5283 - val_loss: 0.7536 - val_y1_output_loss: 0.2140 - val_y2_output_loss: 0.5607 - val_y1_output_root_mean_squared_error: 0.4535 - val_y2_output_root_mean_squared_error: 0.7403\n"
     ]
    }
   ],
   "source": [
    "# Train the model for 500 epochs\n",
    "history = model.fit(norm_train_X, train_Y,\n",
    "                    epochs=500, batch_size=10, validation_data=(norm_test_X, test_Y))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Evaluate the Model and Plot Metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "154/154 [==============================] - 0s 30us/sample - loss: 0.7536 - y1_output_loss: 0.2078 - y2_output_loss: 0.5479 - y1_output_root_mean_squared_error: 0.4535 - y2_output_root_mean_squared_error: 0.7403\n",
      "Loss = 0.7536452048784726, Y1_loss = 0.20781919360160828, Y1_mse = 0.45345839858055115, Y2_loss = 0.5478577613830566, Y2_mse = 0.7402842044830322\n"
     ]
    }
   ],
   "source": [
    "# Test the model and print loss and mse for both outputs\n",
    "loss, Y1_loss, Y2_loss, Y1_rmse, Y2_rmse = model.evaluate(x=norm_test_X, y=test_Y)\n",
    "print(\"Loss = {}, Y1_loss = {}, Y1_mse = {}, Y2_loss = {}, Y2_mse = {}\".format(loss, Y1_loss, Y1_rmse, Y2_loss, Y2_rmse))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "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"
    },
    {
     "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": [
    "# Plot the loss and mse\n",
    "Y_pred = model.predict(norm_test_X)\n",
    "plot_diff(test_Y[0], Y_pred[0], title='Y1')\n",
    "plot_diff(test_Y[1], Y_pred[1], title='Y2')\n",
    "plot_metrics(metric_name='y1_output_root_mean_squared_error', title='Y1 RMSE', ylim=6)\n",
    "plot_metrics(metric_name='y2_output_root_mean_squared_error', title='Y2 RMSE', ylim=7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "include_colab_link": true,
   "name": "Multi-Output.ipynb",
   "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}