Multioutput_model_Functional_API.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "7b4210fd",
   "metadata": {},
   "source": [
    "# Multiple outputs NN with functional API"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "6b71412e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing the libraries\n",
    "import os\n",
    "from os import listdir\n",
    "from os.path import isfile, join\n",
    "import pandas as pd\n",
    "import tensorflow as tf\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from tensorflow.keras.models import Model, Sequential\n",
    "from tensorflow.keras.layers import Dense, Input\n",
    "from tensorflow.keras.optimizers import Adam"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "af80f1e9",
   "metadata": {},
   "source": [
    "Helper functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "fcd8fb0d",
   "metadata": {},
   "outputs": [],
   "source": [
    "#function that retrieves the input variable columns from the CSV and returns a pandas dataframe containing the columns\n",
    "def retrieve_X_df(df, verbose=True):\n",
    "  \"\"\"\n",
    "  :param df: Required, the dataframe containing the input and output variables. The first 900 values for each observation are assumed the input variables.\n",
    "  :param verbose: Optional, default==True, prints the shape of the dataframe after retrieving the input columns.\n",
    "\n",
    "  :return: dataframe containing the input data. \n",
    "  \"\"\"\n",
    "  df = df.iloc[:,0:900]\n",
    "  if verbose:\n",
    "    print(\"df shape: \", df.shape)\n",
    "  return df\n",
    "\n",
    "#function that retrieves the target columns from the CSV and returns a pandas dataframe containing the columns\n",
    "def retrieve_y_df(df, n_last_cols = 3, verbose=True):\n",
    "  \"\"\"\n",
    "  :param df: Required, the dataframe containing the input and target variables. \n",
    "  :param n_last_cols: Optional, default==3, The number of column values for each observation that are the target variables.\n",
    "  :param verbose: Optional, default==True, prints the shape of the dataframe after retrieving the target columns.\n",
    "\n",
    "  :return: dataframe containing the target data. \n",
    "  \"\"\"\n",
    "  df = df.iloc[:,-n_last_cols:]\n",
    "  df.columns = ['amplitude','X','Y']\n",
    "  if verbose:\n",
    "    print(\"df shape: \", df.shape)\n",
    "  return df\n",
    "\n",
    "#function that gets the 3 arrays from the dataframe\n",
    "def format_output(df):\n",
    "    \"\"\"\n",
    "    :param df: Required, the dataframe containing the target variables. \n",
    "    \n",
    "    :return: 3 numpy arrays with the target data. \n",
    "    \"\"\"\n",
    "    y1 = df[:, 0]\n",
    "    y1 = np.array(y1)\n",
    "    y2 = df[:, 1]\n",
    "    y2 = np.array(y2)\n",
    "    y3 = df[:, 2]\n",
    "    y3 = np.array(y3)\n",
    "    return y1, y2, y3"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3c4aad9a",
   "metadata": {},
   "source": [
    "Read csv-files and retrieve the train, test and judge datasets. \n",
    "Split train dataset into train and validation datasets with ratio 90%"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "ac544558",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "df shape:  (9000, 900)\n",
      "df shape:  (9000, 3)\n",
      "df shape:  (1000, 900)\n",
      "df shape:  (1000, 3)\n",
      "df shape:  (2000, 900)\n",
      "df shape:  (2000, 3)\n",
      "df shape:  (2000, 900)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>890</th>\n",
       "      <th>891</th>\n",
       "      <th>892</th>\n",
       "      <th>893</th>\n",
       "      <th>894</th>\n",
       "      <th>895</th>\n",
       "      <th>896</th>\n",
       "      <th>897</th>\n",
       "      <th>898</th>\n",
       "      <th>899</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1995</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1996</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1997</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1999</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2000 rows × 900 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      0    1    2    3    4    5    6    7    8    9    ...  890  891  892  \\\n",
       "0     0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  ...  0.0  0.0  0.0   \n",
       "1     0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  ...  0.0  0.0  0.0   \n",
       "2     0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  ...  0.0  0.0  0.0   \n",
       "3     0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  ...  0.0  0.0  0.0   \n",
       "4     0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  ...  0.0  0.0  0.0   \n",
       "...   ...  ...  ...  ...  ...  ...  ...  ...  ...  ...  ...  ...  ...  ...   \n",
       "1995  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  ...  0.0  0.0  0.0   \n",
       "1996  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  ...  0.0  0.0  0.0   \n",
       "1997  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  ...  0.0  0.0  0.0   \n",
       "1998  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  ...  0.0  0.0  0.0   \n",
       "1999  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  ...  0.0  0.0  0.0   \n",
       "\n",
       "      893  894  895  896  897  898  899  \n",
       "0     0.0  0.0  0.0  0.0  0.0  0.0  0.0  \n",
       "1     0.0  0.0  0.0  0.0  0.0  0.0  0.0  \n",
       "2     0.0  0.0  0.0  0.0  0.0  0.0  0.0  \n",
       "3     0.0  0.0  0.0  0.0  0.0  0.0  0.0  \n",
       "4     0.0  0.0  0.0  0.0  0.0  0.0  0.0  \n",
       "...   ...  ...  ...  ...  ...  ...  ...  \n",
       "1995  0.0  0.0  0.0  0.0  0.0  0.0  0.0  \n",
       "1996  0.0  0.0  0.0  0.0  0.0  0.0  0.0  \n",
       "1997  0.0  0.0  0.0  0.0  0.0  0.0  0.0  \n",
       "1998  0.0  0.0  0.0  0.0  0.0  0.0  0.0  \n",
       "1999  0.0  0.0  0.0  0.0  0.0  0.0  0.0  \n",
       "\n",
       "[2000 rows x 900 columns]"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "source = \"https://userweb.jlab.org/~tbritton/Hackathon2021_DataSets/Problem1/\"\n",
    "\n",
    "#Read csv file with train data\n",
    "df_train = pd.read_csv(os.path.join(source,\"train.csv\"), na_values=['NA', '?'], header=None)\n",
    "\n",
    "#Split train data into train (90%) and validation datasets (10%)\n",
    "train_pct = 0.9\n",
    "train_cut = int(len(df_train) * train_pct)\n",
    "\n",
    "#Get train dataframes for X and Y and transform them to numpy arrays\n",
    "df_tr = df_train[0:train_cut]\n",
    "df_train_X = retrieve_X_df(df_tr)\n",
    "df_tr_y = retrieve_y_df(df_tr).to_numpy()\n",
    "df_train_y = format_output(df_tr_y)#3 separate arrays for each output\n",
    "\n",
    "#Get validation dataframes for X and Y and transform them to numpy arrays\n",
    "df_val = df_train[train_cut:]\n",
    "df_val_X = retrieve_X_df(df_val)\n",
    "df_v_y = retrieve_y_df(df_val).to_numpy()\n",
    "df_val_y = format_output(df_v_y)#3 separate arrays for each output\n",
    "\n",
    "#Read csv file with test data\n",
    "df_test = pd.read_csv(os.path.join(source,\"test.csv\"), na_values=['NA', '?'], header=None)\n",
    "\n",
    "#Get train dataframes for X and Y and transform them to numpy arrays\n",
    "df_test_X = retrieve_X_df(df_test)\n",
    "df_t_y = retrieve_y_df(df_test).to_numpy()\n",
    "df_test_y = format_output(df_t_y)#3 separate arrays for each output\n",
    "\n",
    "#Read csv file with judge data\n",
    "df_judge = pd.read_csv(os.path.join(source,\"judge.csv\"), na_values=['NA', '?'], header=None)\n",
    "df_judge_X = retrieve_X_df(df_judge)\n",
    "\n",
    "df_judge_X"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f6aeb7f3",
   "metadata": {},
   "source": [
    "Transform the train and validation inputs into the numpy arrays"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "ca9e8e70",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(9000, 900)\n",
      "(1000, 900)\n"
     ]
    }
   ],
   "source": [
    "train_X = np.array(df_train_X)\n",
    "print(train_X.shape)\n",
    "val_X = np.array(df_val_X)\n",
    "print(val_X.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3fbe6da1",
   "metadata": {},
   "source": [
    "Build the model with functional API"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "2a33091d",
   "metadata": {},
   "outputs": [],
   "source": [
    "def build_model():\n",
    "    # Define model layers.\n",
    "    input_layer = Input(shape=900,)\n",
    "    first_dense = Dense(units='128', activation='relu')(input_layer)\n",
    "    \n",
    "    # Y1 output will be fed from the first dense\n",
    "    y1_output = Dense(units='1', name='amplitude')(first_dense)\n",
    "\n",
    "    second_dense = Dense(units='128', activation='relu')(first_dense)\n",
    "    \n",
    "    # Y2 output will be fed from the second dense\n",
    "    y2_output = Dense(units='1', name='coord_X')(second_dense)\n",
    "\n",
    "    third_dense = Dense(units='128', activation='relu')(second_dense)\n",
    "    \n",
    "    # Y3 output will be fed from the third dense\n",
    "    y3_output = Dense(units='1', name='coord_Y')(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, y3_output])\n",
    "\n",
    "    return model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "154059f7",
   "metadata": {},
   "source": [
    "Create and compile the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "25638886",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = build_model()\n",
    "\n",
    "# Specify the optimizer, and compile the model with loss functions for each output\n",
    "optimizer = Adam(learning_rate=1e-3)\n",
    "model.compile(optimizer=optimizer,\n",
    "              loss={'amplitude': 'mse', \n",
    "                    'coord_X': 'mse', \n",
    "                    'coord_Y': 'mse'},\n",
    "              metrics={'amplitude': tf.keras.metrics.RootMeanSquaredError(),\n",
    "                       'coord_X': tf.keras.metrics.RootMeanSquaredError(),\n",
    "                       'coord_Y': tf.keras.metrics.RootMeanSquaredError()})"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c8a511de",
   "metadata": {},
   "source": [
    "Train the model for 100 epochs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "78acd35d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 9000 samples, validate on 1000 samples\n",
      "Epoch 1/100\n",
      "9000/9000 [==============================] - 1s 161us/sample - loss: 119109.2785 - amplitude_loss: 2.1646 - coord_X_loss: 79818.9062 - coord_Y_loss: 39288.3516 - amplitude_root_mean_squared_error: 1.4713 - coord_X_root_mean_squared_error: 282.5223 - coord_Y_root_mean_squared_error: 198.2130 - val_loss: 32174.1465 - val_amplitude_loss: 1.3236 - val_coord_X_loss: 24928.3320 - val_coord_Y_loss: 7244.4907 - val_amplitude_root_mean_squared_error: 1.1505 - val_coord_X_root_mean_squared_error: 157.8871 - val_coord_Y_root_mean_squared_error: 85.1146\n",
      "Epoch 2/100\n",
      "9000/9000 [==============================] - 1s 131us/sample - loss: 27876.6408 - amplitude_loss: 1.0246 - coord_X_loss: 22271.7852 - coord_Y_loss: 5603.8252 - amplitude_root_mean_squared_error: 1.0122 - coord_X_root_mean_squared_error: 149.2373 - coord_Y_root_mean_squared_error: 74.8587 - val_loss: 19332.5374 - val_amplitude_loss: 0.7747 - val_coord_X_loss: 16913.5605 - val_coord_Y_loss: 2418.2036 - val_amplitude_root_mean_squared_error: 0.8802 - val_coord_X_root_mean_squared_error: 130.0522 - val_coord_Y_root_mean_squared_error: 49.1752\n",
      "Epoch 3/100\n",
      "9000/9000 [==============================] - 1s 133us/sample - loss: 16755.0412 - amplitude_loss: 0.8378 - coord_X_loss: 14654.1680 - coord_Y_loss: 2100.0322 - amplitude_root_mean_squared_error: 0.9153 - coord_X_root_mean_squared_error: 121.0544 - coord_Y_root_mean_squared_error: 45.8261 - val_loss: 12413.0911 - val_amplitude_loss: 0.7473 - val_coord_X_loss: 11445.8975 - val_coord_Y_loss: 966.4455 - val_amplitude_root_mean_squared_error: 0.8644 - val_coord_X_root_mean_squared_error: 106.9855 - val_coord_Y_root_mean_squared_error: 31.0877\n",
      "Epoch 4/100\n",
      "9000/9000 [==============================] - 1s 133us/sample - loss: 10188.2917 - amplitude_loss: 0.8283 - coord_X_loss: 9043.3906 - coord_Y_loss: 1144.0658 - amplitude_root_mean_squared_error: 0.9101 - coord_X_root_mean_squared_error: 95.0968 - coord_Y_root_mean_squared_error: 33.8240 - val_loss: 8323.5213 - val_amplitude_loss: 0.8117 - val_coord_X_loss: 7106.8730 - val_coord_Y_loss: 1215.8359 - val_amplitude_root_mean_squared_error: 0.9010 - val_coord_X_root_mean_squared_error: 84.3023 - val_coord_Y_root_mean_squared_error: 34.8688\n",
      "Epoch 5/100\n",
      "9000/9000 [==============================] - 1s 134us/sample - loss: 6029.3944 - amplitude_loss: 0.9031 - coord_X_loss: 5036.0410 - coord_Y_loss: 992.4496 - amplitude_root_mean_squared_error: 0.9503 - coord_X_root_mean_squared_error: 70.9651 - coord_Y_root_mean_squared_error: 31.5032 - val_loss: 4575.9138 - val_amplitude_loss: 0.8999 - val_coord_X_loss: 3819.4829 - val_coord_Y_loss: 755.5319 - val_amplitude_root_mean_squared_error: 0.9486 - val_coord_X_root_mean_squared_error: 61.8020 - val_coord_Y_root_mean_squared_error: 27.4869\n",
      "Epoch 6/100\n",
      "9000/9000 [==============================] - 1s 132us/sample - loss: 3536.6705 - amplitude_loss: 0.8901 - coord_X_loss: 2763.3755 - coord_Y_loss: 772.4053 - amplitude_root_mean_squared_error: 0.9434 - coord_X_root_mean_squared_error: 52.5678 - coord_Y_root_mean_squared_error: 27.7922 - val_loss: 3553.5744 - val_amplitude_loss: 0.7891 - val_coord_X_loss: 2204.9512 - val_coord_Y_loss: 1347.8342 - val_amplitude_root_mean_squared_error: 0.8883 - val_coord_X_root_mean_squared_error: 46.9569 - val_coord_Y_root_mean_squared_error: 36.7129\n",
      "Epoch 7/100\n",
      "9000/9000 [==============================] - 1s 132us/sample - loss: 2498.7585 - amplitude_loss: 0.7903 - coord_X_loss: 1709.3998 - coord_Y_loss: 788.5676 - amplitude_root_mean_squared_error: 0.8890 - coord_X_root_mean_squared_error: 41.3449 - coord_Y_root_mean_squared_error: 28.0814 - val_loss: 2522.1395 - val_amplitude_loss: 0.7311 - val_coord_X_loss: 1480.2271 - val_coord_Y_loss: 1041.1813 - val_amplitude_root_mean_squared_error: 0.8550 - val_coord_X_root_mean_squared_error: 38.4737 - val_coord_Y_root_mean_squared_error: 32.2673\n",
      "Epoch 8/100\n",
      "9000/9000 [==============================] - 1s 134us/sample - loss: 1705.4869 - amplitude_loss: 0.6944 - coord_X_loss: 1094.9199 - coord_Y_loss: 609.8726 - amplitude_root_mean_squared_error: 0.8333 - coord_X_root_mean_squared_error: 33.0896 - coord_Y_root_mean_squared_error: 24.6956 - val_loss: 1733.5382 - val_amplitude_loss: 0.6234 - val_coord_X_loss: 1119.0875 - val_coord_Y_loss: 613.8275 - val_amplitude_root_mean_squared_error: 0.7895 - val_coord_X_root_mean_squared_error: 33.4528 - val_coord_Y_root_mean_squared_error: 24.7755\n",
      "Epoch 9/100\n",
      "9000/9000 [==============================] - 1s 134us/sample - loss: 1391.7941 - amplitude_loss: 0.6092 - coord_X_loss: 794.0154 - coord_Y_loss: 597.1691 - amplitude_root_mean_squared_error: 0.7805 - coord_X_root_mean_squared_error: 28.1783 - coord_Y_root_mean_squared_error: 24.4370 - val_loss: 1321.4842 - val_amplitude_loss: 0.5730 - val_coord_X_loss: 911.4744 - val_coord_Y_loss: 409.4368 - val_amplitude_root_mean_squared_error: 0.7570 - val_coord_X_root_mean_squared_error: 30.1906 - val_coord_Y_root_mean_squared_error: 20.2345\n",
      "Epoch 10/100\n",
      "9000/9000 [==============================] - 1s 134us/sample - loss: 1201.4086 - amplitude_loss: 0.5476 - coord_X_loss: 666.1162 - coord_Y_loss: 534.7446 - amplitude_root_mean_squared_error: 0.7400 - coord_X_root_mean_squared_error: 25.8092 - coord_Y_root_mean_squared_error: 23.1245 - val_loss: 1184.9499 - val_amplitude_loss: 0.5421 - val_coord_X_loss: 794.6932 - val_coord_Y_loss: 389.7147 - val_amplitude_root_mean_squared_error: 0.7363 - val_coord_X_root_mean_squared_error: 28.1903 - val_coord_Y_root_mean_squared_error: 19.7412\n",
      "Epoch 11/100\n",
      "9000/9000 [==============================] - 1s 133us/sample - loss: 962.8253 - amplitude_loss: 0.5017 - coord_X_loss: 518.3530 - coord_Y_loss: 443.9703 - amplitude_root_mean_squared_error: 0.7083 - coord_X_root_mean_squared_error: 22.7674 - coord_Y_root_mean_squared_error: 21.0706 - val_loss: 1057.0849 - val_amplitude_loss: 0.5034 - val_coord_X_loss: 706.9485 - val_coord_Y_loss: 349.6328 - val_amplitude_root_mean_squared_error: 0.7095 - val_coord_X_root_mean_squared_error: 26.5885 - val_coord_Y_root_mean_squared_error: 18.6985\n",
      "Epoch 12/100\n",
      "9000/9000 [==============================] - 1s 133us/sample - loss: 980.5845 - amplitude_loss: 0.4715 - coord_X_loss: 526.3713 - coord_Y_loss: 453.7420 - amplitude_root_mean_squared_error: 0.6867 - coord_X_root_mean_squared_error: 22.9428 - coord_Y_root_mean_squared_error: 21.3012 - val_loss: 1143.3384 - val_amplitude_loss: 0.4984 - val_coord_X_loss: 633.7591 - val_coord_Y_loss: 509.0811 - val_amplitude_root_mean_squared_error: 0.7060 - val_coord_X_root_mean_squared_error: 25.1746 - val_coord_Y_root_mean_squared_error: 22.5628\n",
      "Epoch 13/100\n",
      "9000/9000 [==============================] - 1s 135us/sample - loss: 893.8678 - amplitude_loss: 0.4512 - coord_X_loss: 446.1597 - coord_Y_loss: 447.2569 - amplitude_root_mean_squared_error: 0.6717 - coord_X_root_mean_squared_error: 21.1225 - coord_Y_root_mean_squared_error: 21.1485 - val_loss: 966.0747 - val_amplitude_loss: 0.4484 - val_coord_X_loss: 628.2561 - val_coord_Y_loss: 337.3701 - val_amplitude_root_mean_squared_error: 0.6696 - val_coord_X_root_mean_squared_error: 25.0650 - val_coord_Y_root_mean_squared_error: 18.3676\n",
      "Epoch 14/100\n",
      "9000/9000 [==============================] - 1s 134us/sample - loss: 714.9266 - amplitude_loss: 0.4233 - coord_X_loss: 366.6633 - coord_Y_loss: 347.8401 - amplitude_root_mean_squared_error: 0.6506 - coord_X_root_mean_squared_error: 19.1485 - coord_Y_root_mean_squared_error: 18.6505 - val_loss: 861.2797 - val_amplitude_loss: 0.4170 - val_coord_X_loss: 505.8425 - val_coord_Y_loss: 355.0201 - val_amplitude_root_mean_squared_error: 0.6458 - val_coord_X_root_mean_squared_error: 22.4909 - val_coord_Y_root_mean_squared_error: 18.8420\n",
      "Epoch 15/100\n",
      "9000/9000 [==============================] - 1s 134us/sample - loss: 733.8011 - amplitude_loss: 0.4100 - coord_X_loss: 379.5532 - coord_Y_loss: 353.8379 - amplitude_root_mean_squared_error: 0.6403 - coord_X_root_mean_squared_error: 19.4821 - coord_Y_root_mean_squared_error: 18.8106 - val_loss: 1167.3375 - val_amplitude_loss: 0.4035 - val_coord_X_loss: 594.5936 - val_coord_Y_loss: 572.3406 - val_amplitude_root_mean_squared_error: 0.6352 - val_coord_X_root_mean_squared_error: 24.3843 - val_coord_Y_root_mean_squared_error: 23.9236\n",
      "Epoch 16/100\n",
      "9000/9000 [==============================] - 1s 134us/sample - loss: 631.6849 - amplitude_loss: 0.3812 - coord_X_loss: 335.6285 - coord_Y_loss: 295.6751 - amplitude_root_mean_squared_error: 0.6174 - coord_X_root_mean_squared_error: 18.3202 - coord_Y_root_mean_squared_error: 17.1952 - val_loss: 1110.8127 - val_amplitude_loss: 0.3658 - val_coord_X_loss: 625.5554 - val_coord_Y_loss: 484.8915 - val_amplitude_root_mean_squared_error: 0.6048 - val_coord_X_root_mean_squared_error: 25.0111 - val_coord_Y_root_mean_squared_error: 22.0203\n",
      "Epoch 17/100\n",
      "9000/9000 [==============================] - 1s 135us/sample - loss: 679.6300 - amplitude_loss: 0.3641 - coord_X_loss: 325.1904 - coord_Y_loss: 354.0757 - amplitude_root_mean_squared_error: 0.6034 - coord_X_root_mean_squared_error: 18.0330 - coord_Y_root_mean_squared_error: 18.8169 - val_loss: 820.2581 - val_amplitude_loss: 0.3778 - val_coord_X_loss: 484.4124 - val_coord_Y_loss: 335.4680 - val_amplitude_root_mean_squared_error: 0.6147 - val_coord_X_root_mean_squared_error: 22.0094 - val_coord_Y_root_mean_squared_error: 18.3158\n",
      "Epoch 18/100\n",
      "9000/9000 [==============================] - 1s 137us/sample - loss: 574.8420 - amplitude_loss: 0.3512 - coord_X_loss: 302.3887 - coord_Y_loss: 272.1020 - amplitude_root_mean_squared_error: 0.5927 - coord_X_root_mean_squared_error: 17.3893 - coord_Y_root_mean_squared_error: 16.4955 - val_loss: 690.9946 - val_amplitude_loss: 0.3281 - val_coord_X_loss: 392.4118 - val_coord_Y_loss: 298.2547 - val_amplitude_root_mean_squared_error: 0.5728 - val_coord_X_root_mean_squared_error: 19.8094 - val_coord_Y_root_mean_squared_error: 17.2701\n",
      "Epoch 19/100\n",
      "9000/9000 [==============================] - 1s 138us/sample - loss: 549.9315 - amplitude_loss: 0.3202 - coord_X_loss: 251.2112 - coord_Y_loss: 298.4004 - amplitude_root_mean_squared_error: 0.5658 - coord_X_root_mean_squared_error: 15.8496 - coord_Y_root_mean_squared_error: 17.2743 - val_loss: 782.7737 - val_amplitude_loss: 0.3050 - val_coord_X_loss: 552.5057 - val_coord_Y_loss: 229.9629 - val_amplitude_root_mean_squared_error: 0.5523 - val_coord_X_root_mean_squared_error: 23.5054 - val_coord_Y_root_mean_squared_error: 15.1645\n",
      "Epoch 20/100\n",
      "9000/9000 [==============================] - 1s 135us/sample - loss: 531.5989 - amplitude_loss: 0.3039 - coord_X_loss: 270.2501 - coord_Y_loss: 261.0448 - amplitude_root_mean_squared_error: 0.5513 - coord_X_root_mean_squared_error: 16.4393 - coord_Y_root_mean_squared_error: 16.1569 - val_loss: 672.0425 - val_amplitude_loss: 0.2926 - val_coord_X_loss: 395.9567 - val_coord_Y_loss: 275.7932 - val_amplitude_root_mean_squared_error: 0.5409 - val_coord_X_root_mean_squared_error: 19.8987 - val_coord_Y_root_mean_squared_error: 16.6070\n",
      "Epoch 21/100\n",
      "9000/9000 [==============================] - 1s 135us/sample - loss: 493.9068 - amplitude_loss: 0.2980 - coord_X_loss: 253.6114 - coord_Y_loss: 239.9973 - amplitude_root_mean_squared_error: 0.5459 - coord_X_root_mean_squared_error: 15.9252 - coord_Y_root_mean_squared_error: 15.4918 - val_loss: 798.2323 - val_amplitude_loss: 0.3006 - val_coord_X_loss: 394.5241 - val_coord_Y_loss: 403.4077 - val_amplitude_root_mean_squared_error: 0.5483 - val_coord_X_root_mean_squared_error: 19.8626 - val_coord_Y_root_mean_squared_error: 20.0850\n",
      "Epoch 22/100\n",
      "9000/9000 [==============================] - 1s 138us/sample - loss: 508.4577 - amplitude_loss: 0.2855 - coord_X_loss: 230.2751 - coord_Y_loss: 277.8971 - amplitude_root_mean_squared_error: 0.5343 - coord_X_root_mean_squared_error: 15.1748 - coord_Y_root_mean_squared_error: 16.6703 - val_loss: 739.8755 - val_amplitude_loss: 0.2637 - val_coord_X_loss: 452.6098 - val_coord_Y_loss: 287.0019 - val_amplitude_root_mean_squared_error: 0.5135 - val_coord_X_root_mean_squared_error: 21.2746 - val_coord_Y_root_mean_squared_error: 16.9411\n",
      "Epoch 23/100\n",
      "9000/9000 [==============================] - 1s 136us/sample - loss: 451.5623 - amplitude_loss: 0.2636 - coord_X_loss: 233.5223 - coord_Y_loss: 217.7760 - amplitude_root_mean_squared_error: 0.5134 - coord_X_root_mean_squared_error: 15.2814 - coord_Y_root_mean_squared_error: 14.7572 - val_loss: 659.2916 - val_amplitude_loss: 0.2456 - val_coord_X_loss: 365.4682 - val_coord_Y_loss: 293.5779 - val_amplitude_root_mean_squared_error: 0.4956 - val_coord_X_root_mean_squared_error: 19.1172 - val_coord_Y_root_mean_squared_error: 17.1341\n",
      "Epoch 24/100\n",
      "9000/9000 [==============================] - 1s 132us/sample - loss: 497.1807 - amplitude_loss: 0.2485 - coord_X_loss: 243.7668 - coord_Y_loss: 253.1652 - amplitude_root_mean_squared_error: 0.4985 - coord_X_root_mean_squared_error: 15.6130 - coord_Y_root_mean_squared_error: 15.9112 - val_loss: 846.8294 - val_amplitude_loss: 0.2460 - val_coord_X_loss: 450.9455 - val_coord_Y_loss: 395.6377 - val_amplitude_root_mean_squared_error: 0.4960 - val_coord_X_root_mean_squared_error: 21.2355 - val_coord_Y_root_mean_squared_error: 19.8906\n",
      "Epoch 25/100\n",
      "9000/9000 [==============================] - 1s 136us/sample - loss: 406.7611 - amplitude_loss: 0.2410 - coord_X_loss: 186.0952 - coord_Y_loss: 220.4249 - amplitude_root_mean_squared_error: 0.4909 - coord_X_root_mean_squared_error: 13.6417 - coord_Y_root_mean_squared_error: 14.8467 - val_loss: 668.9085 - val_amplitude_loss: 0.2123 - val_coord_X_loss: 377.4687 - val_coord_Y_loss: 291.2275 - val_amplitude_root_mean_squared_error: 0.4608 - val_coord_X_root_mean_squared_error: 19.4286 - val_coord_Y_root_mean_squared_error: 17.0654\n",
      "Epoch 26/100\n",
      "9000/9000 [==============================] - 1s 137us/sample - loss: 382.6474 - amplitude_loss: 0.2198 - coord_X_loss: 201.8730 - coord_Y_loss: 180.5544 - amplitude_root_mean_squared_error: 0.4688 - coord_X_root_mean_squared_error: 14.2082 - coord_Y_root_mean_squared_error: 13.4370 - val_loss: 554.3631 - val_amplitude_loss: 0.1982 - val_coord_X_loss: 363.1350 - val_coord_Y_loss: 191.0300 - val_amplitude_root_mean_squared_error: 0.4452 - val_coord_X_root_mean_squared_error: 19.0561 - val_coord_Y_root_mean_squared_error: 13.8214\n",
      "Epoch 27/100\n",
      "9000/9000 [==============================] - 1s 137us/sample - loss: 433.3886 - amplitude_loss: 0.2125 - coord_X_loss: 195.1063 - coord_Y_loss: 238.0697 - amplitude_root_mean_squared_error: 0.4610 - coord_X_root_mean_squared_error: 13.9681 - coord_Y_root_mean_squared_error: 15.4295 - val_loss: 723.2825 - val_amplitude_loss: 0.1959 - val_coord_X_loss: 451.0382 - val_coord_Y_loss: 272.0486 - val_amplitude_root_mean_squared_error: 0.4426 - val_coord_X_root_mean_squared_error: 21.2377 - val_coord_Y_root_mean_squared_error: 16.4939\n",
      "Epoch 28/100\n",
      "9000/9000 [==============================] - 1s 136us/sample - loss: 375.2241 - amplitude_loss: 0.2158 - coord_X_loss: 186.1405 - coord_Y_loss: 188.8680 - amplitude_root_mean_squared_error: 0.4645 - coord_X_root_mean_squared_error: 13.6433 - coord_Y_root_mean_squared_error: 13.7429 - val_loss: 614.6855 - val_amplitude_loss: 0.2052 - val_coord_X_loss: 350.1714 - val_coord_Y_loss: 264.3088 - val_amplitude_root_mean_squared_error: 0.4530 - val_coord_X_root_mean_squared_error: 18.7129 - val_coord_Y_root_mean_squared_error: 16.2576\n",
      "Epoch 29/100\n",
      "9000/9000 [==============================] - 1s 135us/sample - loss: 380.7668 - amplitude_loss: 0.2060 - coord_X_loss: 188.3855 - coord_Y_loss: 192.1752 - amplitude_root_mean_squared_error: 0.4539 - coord_X_root_mean_squared_error: 13.7254 - coord_Y_root_mean_squared_error: 13.8627 - val_loss: 641.6114 - val_amplitude_loss: 0.1830 - val_coord_X_loss: 434.0961 - val_coord_Y_loss: 207.3324 - val_amplitude_root_mean_squared_error: 0.4278 - val_coord_X_root_mean_squared_error: 20.8350 - val_coord_Y_root_mean_squared_error: 14.3990\n",
      "Epoch 30/100\n",
      "9000/9000 [==============================] - 1s 137us/sample - loss: 361.5547 - amplitude_loss: 0.1885 - coord_X_loss: 170.9501 - coord_Y_loss: 190.4160 - amplitude_root_mean_squared_error: 0.4342 - coord_X_root_mean_squared_error: 13.0748 - coord_Y_root_mean_squared_error: 13.7991 - val_loss: 589.0282 - val_amplitude_loss: 0.1801 - val_coord_X_loss: 337.8088 - val_coord_Y_loss: 251.0394 - val_amplitude_root_mean_squared_error: 0.4243 - val_coord_X_root_mean_squared_error: 18.3796 - val_coord_Y_root_mean_squared_error: 15.8442\n",
      "Epoch 31/100\n",
      "9000/9000 [==============================] - 1s 131us/sample - loss: 399.6967 - amplitude_loss: 0.1819 - coord_X_loss: 167.9437 - coord_Y_loss: 231.5714 - amplitude_root_mean_squared_error: 0.4265 - coord_X_root_mean_squared_error: 12.9593 - coord_Y_root_mean_squared_error: 15.2175 - val_loss: 558.0778 - val_amplitude_loss: 0.1782 - val_coord_X_loss: 331.6638 - val_coord_Y_loss: 226.2357 - val_amplitude_root_mean_squared_error: 0.4222 - val_coord_X_root_mean_squared_error: 18.2116 - val_coord_Y_root_mean_squared_error: 15.0411\n",
      "Epoch 32/100\n",
      "9000/9000 [==============================] - 1s 132us/sample - loss: 314.2313 - amplitude_loss: 0.1732 - coord_X_loss: 144.1372 - coord_Y_loss: 169.9208 - amplitude_root_mean_squared_error: 0.4162 - coord_X_root_mean_squared_error: 12.0057 - coord_Y_root_mean_squared_error: 13.0354 - val_loss: 503.1596 - val_amplitude_loss: 0.1719 - val_coord_X_loss: 292.4965 - val_coord_Y_loss: 210.4913 - val_amplitude_root_mean_squared_error: 0.4146 - val_coord_X_root_mean_squared_error: 17.1025 - val_coord_Y_root_mean_squared_error: 14.5083\n",
      "Epoch 33/100\n",
      "9000/9000 [==============================] - 1s 132us/sample - loss: 366.4505 - amplitude_loss: 0.1648 - coord_X_loss: 160.5616 - coord_Y_loss: 205.7242 - amplitude_root_mean_squared_error: 0.4059 - coord_X_root_mean_squared_error: 12.6713 - coord_Y_root_mean_squared_error: 14.3431 - val_loss: 506.1475 - val_amplitude_loss: 0.1563 - val_coord_X_loss: 282.3988 - val_coord_Y_loss: 223.5924 - val_amplitude_root_mean_squared_error: 0.3953 - val_coord_X_root_mean_squared_error: 16.8047 - val_coord_Y_root_mean_squared_error: 14.9530\n",
      "Epoch 34/100\n",
      "9000/9000 [==============================] - 1s 132us/sample - loss: 310.8026 - amplitude_loss: 0.1550 - coord_X_loss: 163.1879 - coord_Y_loss: 147.4596 - amplitude_root_mean_squared_error: 0.3937 - coord_X_root_mean_squared_error: 12.7745 - coord_Y_root_mean_squared_error: 12.1433 - val_loss: 641.9965 - val_amplitude_loss: 0.1532 - val_coord_X_loss: 360.2451 - val_coord_Y_loss: 281.5981 - val_amplitude_root_mean_squared_error: 0.3914 - val_coord_X_root_mean_squared_error: 18.9801 - val_coord_Y_root_mean_squared_error: 16.7809\n",
      "Epoch 35/100\n",
      "9000/9000 [==============================] - 1s 133us/sample - loss: 315.6371 - amplitude_loss: 0.1565 - coord_X_loss: 131.5401 - coord_Y_loss: 183.9405 - amplitude_root_mean_squared_error: 0.3956 - coord_X_root_mean_squared_error: 11.4691 - coord_Y_root_mean_squared_error: 13.5625 - val_loss: 504.2887 - val_amplitude_loss: 0.1582 - val_coord_X_loss: 280.8485 - val_coord_Y_loss: 223.2820 - val_amplitude_root_mean_squared_error: 0.3977 - val_coord_X_root_mean_squared_error: 16.7585 - val_coord_Y_root_mean_squared_error: 14.9426\n",
      "Epoch 36/100\n",
      "9000/9000 [==============================] - 1s 133us/sample - loss: 300.8924 - amplitude_loss: 0.1524 - coord_X_loss: 158.3911 - coord_Y_loss: 142.3490 - amplitude_root_mean_squared_error: 0.3904 - coord_X_root_mean_squared_error: 12.5853 - coord_Y_root_mean_squared_error: 11.9310 - val_loss: 512.4092 - val_amplitude_loss: 0.1537 - val_coord_X_loss: 319.3538 - val_coord_Y_loss: 192.9016 - val_amplitude_root_mean_squared_error: 0.3921 - val_coord_X_root_mean_squared_error: 17.8705 - val_coord_Y_root_mean_squared_error: 13.8889\n",
      "Epoch 37/100\n",
      "9000/9000 [==============================] - 1s 135us/sample - loss: 357.7434 - amplitude_loss: 0.1487 - coord_X_loss: 153.5077 - coord_Y_loss: 204.0870 - amplitude_root_mean_squared_error: 0.3856 - coord_X_root_mean_squared_error: 12.3898 - coord_Y_root_mean_squared_error: 14.2859 - val_loss: 556.3321 - val_amplitude_loss: 0.1629 - val_coord_X_loss: 306.7298 - val_coord_Y_loss: 249.4394 - val_amplitude_root_mean_squared_error: 0.4036 - val_coord_X_root_mean_squared_error: 17.5137 - val_coord_Y_root_mean_squared_error: 15.7936\n",
      "Epoch 38/100\n",
      "9000/9000 [==============================] - 1s 134us/sample - loss: 276.4223 - amplitude_loss: 0.1421 - coord_X_loss: 122.1882 - coord_Y_loss: 154.0920 - amplitude_root_mean_squared_error: 0.3770 - coord_X_root_mean_squared_error: 11.0539 - coord_Y_root_mean_squared_error: 12.4134 - val_loss: 496.0741 - val_amplitude_loss: 0.1476 - val_coord_X_loss: 302.1059 - val_coord_Y_loss: 193.8206 - val_amplitude_root_mean_squared_error: 0.3842 - val_coord_X_root_mean_squared_error: 17.3812 - val_coord_Y_root_mean_squared_error: 13.9219\n",
      "Epoch 39/100\n",
      "9000/9000 [==============================] - 1s 134us/sample - loss: 268.5320 - amplitude_loss: 0.1409 - coord_X_loss: 115.3051 - coord_Y_loss: 153.0860 - amplitude_root_mean_squared_error: 0.3754 - coord_X_root_mean_squared_error: 10.7380 - coord_Y_root_mean_squared_error: 12.3728 - val_loss: 555.8273 - val_amplitude_loss: 0.1478 - val_coord_X_loss: 327.4521 - val_coord_Y_loss: 228.2274 - val_amplitude_root_mean_squared_error: 0.3844 - val_coord_X_root_mean_squared_error: 18.0956 - val_coord_Y_root_mean_squared_error: 15.1072\n",
      "Epoch 40/100\n",
      "9000/9000 [==============================] - 1s 140us/sample - loss: 265.9432 - amplitude_loss: 0.1361 - coord_X_loss: 127.7238 - coord_Y_loss: 138.0831 - amplitude_root_mean_squared_error: 0.3690 - coord_X_root_mean_squared_error: 11.3015 - coord_Y_root_mean_squared_error: 11.7509 - val_loss: 472.9638 - val_amplitude_loss: 0.1497 - val_coord_X_loss: 309.2619 - val_coord_Y_loss: 163.5522 - val_amplitude_root_mean_squared_error: 0.3869 - val_coord_X_root_mean_squared_error: 17.5858 - val_coord_Y_root_mean_squared_error: 12.7888\n",
      "Epoch 41/100\n",
      "9000/9000 [==============================] - 1s 139us/sample - loss: 303.4340 - amplitude_loss: 0.1368 - coord_X_loss: 141.2320 - coord_Y_loss: 162.0653 - amplitude_root_mean_squared_error: 0.3698 - coord_X_root_mean_squared_error: 11.8841 - coord_Y_root_mean_squared_error: 12.7305 - val_loss: 481.1674 - val_amplitude_loss: 0.1418 - val_coord_X_loss: 305.3811 - val_coord_Y_loss: 175.6443 - val_amplitude_root_mean_squared_error: 0.3766 - val_coord_X_root_mean_squared_error: 17.4752 - val_coord_Y_root_mean_squared_error: 13.2531\n",
      "Epoch 42/100\n",
      "9000/9000 [==============================] - 1s 138us/sample - loss: 250.8602 - amplitude_loss: 0.1292 - coord_X_loss: 101.9963 - coord_Y_loss: 148.7345 - amplitude_root_mean_squared_error: 0.3595 - coord_X_root_mean_squared_error: 10.0993 - coord_Y_root_mean_squared_error: 12.1957 - val_loss: 602.0374 - val_amplitude_loss: 0.1493 - val_coord_X_loss: 308.4281 - val_coord_Y_loss: 293.4600 - val_amplitude_root_mean_squared_error: 0.3864 - val_coord_X_root_mean_squared_error: 17.5621 - val_coord_Y_root_mean_squared_error: 17.1307\n",
      "Epoch 43/100\n",
      "9000/9000 [==============================] - 1s 137us/sample - loss: 280.1313 - amplitude_loss: 0.1297 - coord_X_loss: 119.9922 - coord_Y_loss: 160.0095 - amplitude_root_mean_squared_error: 0.3601 - coord_X_root_mean_squared_error: 10.9541 - coord_Y_root_mean_squared_error: 12.6495 - val_loss: 502.6423 - val_amplitude_loss: 0.1720 - val_coord_X_loss: 290.7528 - val_coord_Y_loss: 211.7175 - val_amplitude_root_mean_squared_error: 0.4147 - val_coord_X_root_mean_squared_error: 17.0515 - val_coord_Y_root_mean_squared_error: 14.5505\n",
      "Epoch 44/100\n",
      "9000/9000 [==============================] - 1s 138us/sample - loss: 261.7867 - amplitude_loss: 0.1323 - coord_X_loss: 118.5762 - coord_Y_loss: 143.0782 - amplitude_root_mean_squared_error: 0.3638 - coord_X_root_mean_squared_error: 10.8893 - coord_Y_root_mean_squared_error: 11.9615 - val_loss: 466.6725 - val_amplitude_loss: 0.1444 - val_coord_X_loss: 281.2714 - val_coord_Y_loss: 185.2566 - val_amplitude_root_mean_squared_error: 0.3800 - val_coord_X_root_mean_squared_error: 16.7711 - val_coord_Y_root_mean_squared_error: 13.6109\n",
      "Epoch 45/100\n",
      "9000/9000 [==============================] - 1s 138us/sample - loss: 268.1978 - amplitude_loss: 0.1276 - coord_X_loss: 110.6727 - coord_Y_loss: 157.3976 - amplitude_root_mean_squared_error: 0.3572 - coord_X_root_mean_squared_error: 10.5201 - coord_Y_root_mean_squared_error: 12.5458 - val_loss: 399.2328 - val_amplitude_loss: 0.1457 - val_coord_X_loss: 242.4860 - val_coord_Y_loss: 156.6011 - val_amplitude_root_mean_squared_error: 0.3818 - val_coord_X_root_mean_squared_error: 15.5720 - val_coord_Y_root_mean_squared_error: 12.5140\n",
      "Epoch 46/100\n",
      "9000/9000 [==============================] - 1s 137us/sample - loss: 255.6246 - amplitude_loss: 0.1261 - coord_X_loss: 114.1024 - coord_Y_loss: 141.3961 - amplitude_root_mean_squared_error: 0.3550 - coord_X_root_mean_squared_error: 10.6819 - coord_Y_root_mean_squared_error: 11.8910 - val_loss: 489.2918 - val_amplitude_loss: 0.1413 - val_coord_X_loss: 284.9655 - val_coord_Y_loss: 204.1850 - val_amplitude_root_mean_squared_error: 0.3759 - val_coord_X_root_mean_squared_error: 16.8809 - val_coord_Y_root_mean_squared_error: 14.2893\n",
      "Epoch 47/100\n",
      "9000/9000 [==============================] - 1s 141us/sample - loss: 227.6963 - amplitude_loss: 0.1239 - coord_X_loss: 101.0886 - coord_Y_loss: 126.4836 - amplitude_root_mean_squared_error: 0.3521 - coord_X_root_mean_squared_error: 10.0543 - coord_Y_root_mean_squared_error: 11.2465 - val_loss: 483.7446 - val_amplitude_loss: 0.1352 - val_coord_X_loss: 273.8440 - val_coord_Y_loss: 209.7654 - val_amplitude_root_mean_squared_error: 0.3677 - val_coord_X_root_mean_squared_error: 16.5482 - val_coord_Y_root_mean_squared_error: 14.4833\n",
      "Epoch 48/100\n",
      "9000/9000 [==============================] - 1s 134us/sample - loss: 273.2357 - amplitude_loss: 0.1252 - coord_X_loss: 122.5629 - coord_Y_loss: 150.5476 - amplitude_root_mean_squared_error: 0.3538 - coord_X_root_mean_squared_error: 11.0708 - coord_Y_root_mean_squared_error: 12.2698 - val_loss: 539.2095 - val_amplitude_loss: 0.1506 - val_coord_X_loss: 256.9402 - val_coord_Y_loss: 282.1186 - val_amplitude_root_mean_squared_error: 0.3881 - val_coord_X_root_mean_squared_error: 16.0294 - val_coord_Y_root_mean_squared_error: 16.7964\n",
      "Epoch 49/100\n",
      "9000/9000 [==============================] - 1s 135us/sample - loss: 222.9197 - amplitude_loss: 0.1259 - coord_X_loss: 94.1743 - coord_Y_loss: 128.6195 - amplitude_root_mean_squared_error: 0.3548 - coord_X_root_mean_squared_error: 9.7043 - coord_Y_root_mean_squared_error: 11.3411 - val_loss: 460.9273 - val_amplitude_loss: 0.1366 - val_coord_X_loss: 268.7186 - val_coord_Y_loss: 192.0722 - val_amplitude_root_mean_squared_error: 0.3696 - val_coord_X_root_mean_squared_error: 16.3926 - val_coord_Y_root_mean_squared_error: 13.8590\n",
      "Epoch 50/100\n",
      "9000/9000 [==============================] - 1s 135us/sample - loss: 237.8958 - amplitude_loss: 0.1199 - coord_X_loss: 111.6006 - coord_Y_loss: 126.1755 - amplitude_root_mean_squared_error: 0.3463 - coord_X_root_mean_squared_error: 10.5641 - coord_Y_root_mean_squared_error: 11.2328 - val_loss: 391.4551 - val_amplitude_loss: 0.1296 - val_coord_X_loss: 250.9433 - val_coord_Y_loss: 140.3822 - val_amplitude_root_mean_squared_error: 0.3600 - val_coord_X_root_mean_squared_error: 15.8412 - val_coord_Y_root_mean_squared_error: 11.8483\n",
      "Epoch 51/100\n",
      "9000/9000 [==============================] - 1s 133us/sample - loss: 214.2578 - amplitude_loss: 0.1184 - coord_X_loss: 92.2027 - coord_Y_loss: 121.9366 - amplitude_root_mean_squared_error: 0.3441 - coord_X_root_mean_squared_error: 9.6022 - coord_Y_root_mean_squared_error: 11.0425 - val_loss: 365.3825 - val_amplitude_loss: 0.1267 - val_coord_X_loss: 224.5592 - val_coord_Y_loss: 140.6966 - val_amplitude_root_mean_squared_error: 0.3560 - val_coord_X_root_mean_squared_error: 14.9853 - val_coord_Y_root_mean_squared_error: 11.8616\n",
      "Epoch 52/100\n",
      "9000/9000 [==============================] - 1s 133us/sample - loss: 256.7147 - amplitude_loss: 0.1182 - coord_X_loss: 108.2901 - coord_Y_loss: 148.3065 - amplitude_root_mean_squared_error: 0.3438 - coord_X_root_mean_squared_error: 10.4062 - coord_Y_root_mean_squared_error: 12.1781 - val_loss: 452.9583 - val_amplitude_loss: 0.1328 - val_coord_X_loss: 264.6792 - val_coord_Y_loss: 188.1463 - val_amplitude_root_mean_squared_error: 0.3644 - val_coord_X_root_mean_squared_error: 16.2690 - val_coord_Y_root_mean_squared_error: 13.7166\n",
      "Epoch 53/100\n",
      "9000/9000 [==============================] - 1s 131us/sample - loss: 238.5503 - amplitude_loss: 0.1170 - coord_X_loss: 107.0240 - coord_Y_loss: 131.4094 - amplitude_root_mean_squared_error: 0.3421 - coord_X_root_mean_squared_error: 10.3452 - coord_Y_root_mean_squared_error: 11.4634 - val_loss: 488.7907 - val_amplitude_loss: 0.1144 - val_coord_X_loss: 213.4030 - val_coord_Y_loss: 275.2732 - val_amplitude_root_mean_squared_error: 0.3382 - val_coord_X_root_mean_squared_error: 14.6083 - val_coord_Y_root_mean_squared_error: 16.5914\n",
      "Epoch 54/100\n",
      "9000/9000 [==============================] - 1s 133us/sample - loss: 193.6409 - amplitude_loss: 0.1134 - coord_X_loss: 86.3948 - coord_Y_loss: 107.1326 - amplitude_root_mean_squared_error: 0.3368 - coord_X_root_mean_squared_error: 9.2949 - coord_Y_root_mean_squared_error: 10.3505 - val_loss: 461.8913 - val_amplitude_loss: 0.1233 - val_coord_X_loss: 236.9224 - val_coord_Y_loss: 224.8455 - val_amplitude_root_mean_squared_error: 0.3512 - val_coord_X_root_mean_squared_error: 15.3923 - val_coord_Y_root_mean_squared_error: 14.9949\n",
      "Epoch 55/100\n",
      "9000/9000 [==============================] - 1s 132us/sample - loss: 244.1440 - amplitude_loss: 0.1141 - coord_X_loss: 107.6866 - coord_Y_loss: 136.3432 - amplitude_root_mean_squared_error: 0.3378 - coord_X_root_mean_squared_error: 10.3772 - coord_Y_root_mean_squared_error: 11.6766 - val_loss: 415.4876 - val_amplitude_loss: 0.1146 - val_coord_X_loss: 229.6774 - val_coord_Y_loss: 185.6955 - val_amplitude_root_mean_squared_error: 0.3386 - val_coord_X_root_mean_squared_error: 15.1551 - val_coord_Y_root_mean_squared_error: 13.6270\n",
      "Epoch 56/100\n",
      "9000/9000 [==============================] - 1s 134us/sample - loss: 206.6342 - amplitude_loss: 0.1107 - coord_X_loss: 89.6200 - coord_Y_loss: 116.9035 - amplitude_root_mean_squared_error: 0.3327 - coord_X_root_mean_squared_error: 9.4668 - coord_Y_root_mean_squared_error: 10.8122 - val_loss: 376.5029 - val_amplitude_loss: 0.1138 - val_coord_X_loss: 232.8493 - val_coord_Y_loss: 143.5397 - val_amplitude_root_mean_squared_error: 0.3374 - val_coord_X_root_mean_squared_error: 15.2594 - val_coord_Y_root_mean_squared_error: 11.9808\n",
      "Epoch 57/100\n",
      "9000/9000 [==============================] - 1s 130us/sample - loss: 200.8453 - amplitude_loss: 0.1111 - coord_X_loss: 86.0069 - coord_Y_loss: 114.7272 - amplitude_root_mean_squared_error: 0.3333 - coord_X_root_mean_squared_error: 9.2740 - coord_Y_root_mean_squared_error: 10.7111 - val_loss: 449.6998 - val_amplitude_loss: 0.1147 - val_coord_X_loss: 236.9932 - val_coord_Y_loss: 212.5919 - val_amplitude_root_mean_squared_error: 0.3387 - val_coord_X_root_mean_squared_error: 15.3946 - val_coord_Y_root_mean_squared_error: 14.5805\n",
      "Epoch 58/100\n",
      "9000/9000 [==============================] - 1s 131us/sample - loss: 208.9743 - amplitude_loss: 0.1128 - coord_X_loss: 93.7207 - coord_Y_loss: 115.1409 - amplitude_root_mean_squared_error: 0.3359 - coord_X_root_mean_squared_error: 9.6809 - coord_Y_root_mean_squared_error: 10.7304 - val_loss: 365.6812 - val_amplitude_loss: 0.1179 - val_coord_X_loss: 229.2589 - val_coord_Y_loss: 136.3044 - val_amplitude_root_mean_squared_error: 0.3433 - val_coord_X_root_mean_squared_error: 15.1413 - val_coord_Y_root_mean_squared_error: 11.6749\n",
      "Epoch 59/100\n",
      "9000/9000 [==============================] - 1s 131us/sample - loss: 240.9672 - amplitude_loss: 0.1089 - coord_X_loss: 96.7389 - coord_Y_loss: 144.1195 - amplitude_root_mean_squared_error: 0.3300 - coord_X_root_mean_squared_error: 9.8356 - coord_Y_root_mean_squared_error: 12.0050 - val_loss: 384.7695 - val_amplitude_loss: 0.1106 - val_coord_X_loss: 203.7795 - val_coord_Y_loss: 180.8795 - val_amplitude_root_mean_squared_error: 0.3325 - val_coord_X_root_mean_squared_error: 14.2751 - val_coord_Y_root_mean_squared_error: 13.4491\n",
      "Epoch 60/100\n",
      "9000/9000 [==============================] - 1s 131us/sample - loss: 178.0755 - amplitude_loss: 0.1041 - coord_X_loss: 78.0490 - coord_Y_loss: 99.9224 - amplitude_root_mean_squared_error: 0.3227 - coord_X_root_mean_squared_error: 8.8345 - coord_Y_root_mean_squared_error: 9.9961 - val_loss: 366.3579 - val_amplitude_loss: 0.1076 - val_coord_X_loss: 237.7791 - val_coord_Y_loss: 128.4713 - val_amplitude_root_mean_squared_error: 0.3280 - val_coord_X_root_mean_squared_error: 15.4201 - val_coord_Y_root_mean_squared_error: 11.3345\n",
      "Epoch 61/100\n",
      "9000/9000 [==============================] - 1s 131us/sample - loss: 182.4766 - amplitude_loss: 0.1019 - coord_X_loss: 80.2150 - coord_Y_loss: 102.1596 - amplitude_root_mean_squared_error: 0.3192 - coord_X_root_mean_squared_error: 8.9563 - coord_Y_root_mean_squared_error: 10.1074 - val_loss: 353.6036 - val_amplitude_loss: 0.1141 - val_coord_X_loss: 202.4955 - val_coord_Y_loss: 150.9940 - val_amplitude_root_mean_squared_error: 0.3378 - val_coord_X_root_mean_squared_error: 14.2301 - val_coord_Y_root_mean_squared_error: 12.2880\n",
      "Epoch 62/100\n",
      "9000/9000 [==============================] - 1s 132us/sample - loss: 193.1802 - amplitude_loss: 0.1046 - coord_X_loss: 87.4597 - coord_Y_loss: 105.6160 - amplitude_root_mean_squared_error: 0.3233 - coord_X_root_mean_squared_error: 9.3520 - coord_Y_root_mean_squared_error: 10.2770 - val_loss: 431.0921 - val_amplitude_loss: 0.1335 - val_coord_X_loss: 233.5581 - val_coord_Y_loss: 197.4004 - val_amplitude_root_mean_squared_error: 0.3654 - val_coord_X_root_mean_squared_error: 15.2826 - val_coord_Y_root_mean_squared_error: 14.0499\n",
      "Epoch 63/100\n",
      "9000/9000 [==============================] - 1s 132us/sample - loss: 204.3940 - amplitude_loss: 0.1080 - coord_X_loss: 83.1657 - coord_Y_loss: 121.1205 - amplitude_root_mean_squared_error: 0.3286 - coord_X_root_mean_squared_error: 9.1195 - coord_Y_root_mean_squared_error: 11.0055 - val_loss: 393.7276 - val_amplitude_loss: 0.1114 - val_coord_X_loss: 228.5440 - val_coord_Y_loss: 165.0721 - val_amplitude_root_mean_squared_error: 0.3338 - val_coord_X_root_mean_squared_error: 15.1177 - val_coord_Y_root_mean_squared_error: 12.8480\n",
      "Epoch 64/100\n",
      "9000/9000 [==============================] - 1s 132us/sample - loss: 182.9676 - amplitude_loss: 0.1066 - coord_X_loss: 75.1015 - coord_Y_loss: 107.7596 - amplitude_root_mean_squared_error: 0.3265 - coord_X_root_mean_squared_error: 8.6661 - coord_Y_root_mean_squared_error: 10.3807 - val_loss: 392.3937 - val_amplitude_loss: 0.1132 - val_coord_X_loss: 234.2579 - val_coord_Y_loss: 158.0225 - val_amplitude_root_mean_squared_error: 0.3365 - val_coord_X_root_mean_squared_error: 15.3055 - val_coord_Y_root_mean_squared_error: 12.5707\n",
      "Epoch 65/100\n",
      "9000/9000 [==============================] - 1s 132us/sample - loss: 181.1824 - amplitude_loss: 0.1050 - coord_X_loss: 79.5391 - coord_Y_loss: 101.5384 - amplitude_root_mean_squared_error: 0.3240 - coord_X_root_mean_squared_error: 8.9185 - coord_Y_root_mean_squared_error: 10.0766 - val_loss: 347.6956 - val_amplitude_loss: 0.1125 - val_coord_X_loss: 203.3611 - val_coord_Y_loss: 144.2219 - val_amplitude_root_mean_squared_error: 0.3354 - val_coord_X_root_mean_squared_error: 14.2605 - val_coord_Y_root_mean_squared_error: 12.0092\n",
      "Epoch 66/100\n",
      "9000/9000 [==============================] - 1s 133us/sample - loss: 190.9019 - amplitude_loss: 0.1051 - coord_X_loss: 87.1717 - coord_Y_loss: 103.6251 - amplitude_root_mean_squared_error: 0.3242 - coord_X_root_mean_squared_error: 9.3366 - coord_Y_root_mean_squared_error: 10.1796 - val_loss: 349.3414 - val_amplitude_loss: 0.1165 - val_coord_X_loss: 224.5479 - val_coord_Y_loss: 124.6770 - val_amplitude_root_mean_squared_error: 0.3414 - val_coord_X_root_mean_squared_error: 14.9849 - val_coord_Y_root_mean_squared_error: 11.1659\n",
      "Epoch 67/100\n",
      "9000/9000 [==============================] - 1s 131us/sample - loss: 191.3578 - amplitude_loss: 0.1048 - coord_X_loss: 83.5492 - coord_Y_loss: 107.7037 - amplitude_root_mean_squared_error: 0.3238 - coord_X_root_mean_squared_error: 9.1405 - coord_Y_root_mean_squared_error: 10.3780 - val_loss: 332.3545 - val_amplitude_loss: 0.1022 - val_coord_X_loss: 204.4736 - val_coord_Y_loss: 127.7787 - val_amplitude_root_mean_squared_error: 0.3197 - val_coord_X_root_mean_squared_error: 14.2994 - val_coord_Y_root_mean_squared_error: 11.3039\n",
      "Epoch 68/100\n",
      "9000/9000 [==============================] - 1s 134us/sample - loss: 188.1426 - amplitude_loss: 0.1043 - coord_X_loss: 87.9662 - coord_Y_loss: 100.0722 - amplitude_root_mean_squared_error: 0.3229 - coord_X_root_mean_squared_error: 9.3790 - coord_Y_root_mean_squared_error: 10.0036 - val_loss: 334.2707 - val_amplitude_loss: 0.1024 - val_coord_X_loss: 220.2157 - val_coord_Y_loss: 113.9526 - val_amplitude_root_mean_squared_error: 0.3200 - val_coord_X_root_mean_squared_error: 14.8397 - val_coord_Y_root_mean_squared_error: 10.6749\n",
      "Epoch 69/100\n",
      "9000/9000 [==============================] - 1s 133us/sample - loss: 167.9382 - amplitude_loss: 0.1031 - coord_X_loss: 74.7445 - coord_Y_loss: 93.0905 - amplitude_root_mean_squared_error: 0.3211 - coord_X_root_mean_squared_error: 8.6455 - coord_Y_root_mean_squared_error: 9.6483 - val_loss: 396.1913 - val_amplitude_loss: 0.1118 - val_coord_X_loss: 212.7612 - val_coord_Y_loss: 183.3182 - val_amplitude_root_mean_squared_error: 0.3343 - val_coord_X_root_mean_squared_error: 14.5863 - val_coord_Y_root_mean_squared_error: 13.5395\n",
      "Epoch 70/100\n",
      "9000/9000 [==============================] - 1s 133us/sample - loss: 177.2279 - amplitude_loss: 0.1037 - coord_X_loss: 74.3728 - coord_Y_loss: 102.7513 - amplitude_root_mean_squared_error: 0.3220 - coord_X_root_mean_squared_error: 8.6240 - coord_Y_root_mean_squared_error: 10.1366 - val_loss: 410.6945 - val_amplitude_loss: 0.1123 - val_coord_X_loss: 243.3120 - val_coord_Y_loss: 167.2702 - val_amplitude_root_mean_squared_error: 0.3351 - val_coord_X_root_mean_squared_error: 15.5985 - val_coord_Y_root_mean_squared_error: 12.9333\n",
      "Epoch 71/100\n",
      "9000/9000 [==============================] - 1s 132us/sample - loss: 229.8503 - amplitude_loss: 0.1050 - coord_X_loss: 94.7979 - coord_Y_loss: 134.9474 - amplitude_root_mean_squared_error: 0.3240 - coord_X_root_mean_squared_error: 9.7364 - coord_Y_root_mean_squared_error: 11.6167 - val_loss: 349.6014 - val_amplitude_loss: 0.1034 - val_coord_X_loss: 198.7179 - val_coord_Y_loss: 150.7802 - val_amplitude_root_mean_squared_error: 0.3215 - val_coord_X_root_mean_squared_error: 14.0967 - val_coord_Y_root_mean_squared_error: 12.2793\n",
      "Epoch 72/100\n",
      "9000/9000 [==============================] - 1s 134us/sample - loss: 156.0160 - amplitude_loss: 0.1050 - coord_X_loss: 70.9879 - coord_Y_loss: 84.9231 - amplitude_root_mean_squared_error: 0.3240 - coord_X_root_mean_squared_error: 8.4254 - coord_Y_root_mean_squared_error: 9.2154 - val_loss: 401.7451 - val_amplitude_loss: 0.1122 - val_coord_X_loss: 220.0295 - val_coord_Y_loss: 181.6033 - val_amplitude_root_mean_squared_error: 0.3350 - val_coord_X_root_mean_squared_error: 14.8334 - val_coord_Y_root_mean_squared_error: 13.4760\n",
      "Epoch 73/100\n",
      "9000/9000 [==============================] - 1s 135us/sample - loss: 169.7204 - amplitude_loss: 0.1021 - coord_X_loss: 66.5652 - coord_Y_loss: 103.0531 - amplitude_root_mean_squared_error: 0.3196 - coord_X_root_mean_squared_error: 8.1587 - coord_Y_root_mean_squared_error: 10.1515 - val_loss: 345.7080 - val_amplitude_loss: 0.1076 - val_coord_X_loss: 207.4614 - val_coord_Y_loss: 138.1390 - val_amplitude_root_mean_squared_error: 0.3280 - val_coord_X_root_mean_squared_error: 14.4035 - val_coord_Y_root_mean_squared_error: 11.7533\n",
      "Epoch 74/100\n",
      "9000/9000 [==============================] - 1s 134us/sample - loss: 170.0551 - amplitude_loss: 0.1003 - coord_X_loss: 74.1041 - coord_Y_loss: 95.8508 - amplitude_root_mean_squared_error: 0.3167 - coord_X_root_mean_squared_error: 8.6084 - coord_Y_root_mean_squared_error: 9.7903 - val_loss: 358.8894 - val_amplitude_loss: 0.0951 - val_coord_X_loss: 167.7980 - val_coord_Y_loss: 190.9962 - val_amplitude_root_mean_squared_error: 0.3084 - val_coord_X_root_mean_squared_error: 12.9537 - val_coord_Y_root_mean_squared_error: 13.8201\n",
      "Epoch 75/100\n",
      "9000/9000 [==============================] - 1s 133us/sample - loss: 157.2235 - amplitude_loss: 0.1007 - coord_X_loss: 65.6537 - coord_Y_loss: 91.4691 - amplitude_root_mean_squared_error: 0.3174 - coord_X_root_mean_squared_error: 8.1027 - coord_Y_root_mean_squared_error: 9.5639 - val_loss: 444.1042 - val_amplitude_loss: 0.1020 - val_coord_X_loss: 190.6458 - val_coord_Y_loss: 253.3564 - val_amplitude_root_mean_squared_error: 0.3193 - val_coord_X_root_mean_squared_error: 13.8075 - val_coord_Y_root_mean_squared_error: 15.9172\n",
      "Epoch 76/100\n",
      "9000/9000 [==============================] - 1s 133us/sample - loss: 161.2392 - amplitude_loss: 0.1008 - coord_X_loss: 77.1051 - coord_Y_loss: 84.0333 - amplitude_root_mean_squared_error: 0.3175 - coord_X_root_mean_squared_error: 8.7809 - coord_Y_root_mean_squared_error: 9.1670 - val_loss: 371.6372 - val_amplitude_loss: 0.1026 - val_coord_X_loss: 211.3698 - val_coord_Y_loss: 160.1649 - val_amplitude_root_mean_squared_error: 0.3202 - val_coord_X_root_mean_squared_error: 14.5386 - val_coord_Y_root_mean_squared_error: 12.6556\n",
      "Epoch 77/100\n",
      "9000/9000 [==============================] - 1s 133us/sample - loss: 188.9562 - amplitude_loss: 0.1019 - coord_X_loss: 74.6641 - coord_Y_loss: 114.1902 - amplitude_root_mean_squared_error: 0.3192 - coord_X_root_mean_squared_error: 8.6408 - coord_Y_root_mean_squared_error: 10.6860 - val_loss: 329.0831 - val_amplitude_loss: 0.1068 - val_coord_X_loss: 188.6696 - val_coord_Y_loss: 140.3068 - val_amplitude_root_mean_squared_error: 0.3268 - val_coord_X_root_mean_squared_error: 13.7357 - val_coord_Y_root_mean_squared_error: 11.8451\n",
      "Epoch 78/100\n",
      "9000/9000 [==============================] - 1s 135us/sample - loss: 133.4208 - amplitude_loss: 0.0986 - coord_X_loss: 62.0992 - coord_Y_loss: 71.2230 - amplitude_root_mean_squared_error: 0.3140 - coord_X_root_mean_squared_error: 7.8803 - coord_Y_root_mean_squared_error: 8.4394 - val_loss: 300.7882 - val_amplitude_loss: 0.1039 - val_coord_X_loss: 173.6838 - val_coord_Y_loss: 127.0004 - val_amplitude_root_mean_squared_error: 0.3223 - val_coord_X_root_mean_squared_error: 13.1789 - val_coord_Y_root_mean_squared_error: 11.2694\n",
      "Epoch 79/100\n",
      "9000/9000 [==============================] - 1s 135us/sample - loss: 159.3615 - amplitude_loss: 0.0976 - coord_X_loss: 66.6409 - coord_Y_loss: 92.6230 - amplitude_root_mean_squared_error: 0.3124 - coord_X_root_mean_squared_error: 8.1634 - coord_Y_root_mean_squared_error: 9.6241 - val_loss: 361.4718 - val_amplitude_loss: 0.1012 - val_coord_X_loss: 196.1841 - val_coord_Y_loss: 165.1865 - val_amplitude_root_mean_squared_error: 0.3182 - val_coord_X_root_mean_squared_error: 14.0066 - val_coord_Y_root_mean_squared_error: 12.8525\n",
      "Epoch 80/100\n",
      "9000/9000 [==============================] - 1s 136us/sample - loss: 176.1914 - amplitude_loss: 0.0985 - coord_X_loss: 69.5066 - coord_Y_loss: 106.5863 - amplitude_root_mean_squared_error: 0.3138 - coord_X_root_mean_squared_error: 8.3371 - coord_Y_root_mean_squared_error: 10.3241 - val_loss: 305.7348 - val_amplitude_loss: 0.1085 - val_coord_X_loss: 180.6155 - val_coord_Y_loss: 125.0108 - val_amplitude_root_mean_squared_error: 0.3293 - val_coord_X_root_mean_squared_error: 13.4393 - val_coord_Y_root_mean_squared_error: 11.1808\n",
      "Epoch 81/100\n",
      "9000/9000 [==============================] - 1s 134us/sample - loss: 138.4593 - amplitude_loss: 0.0976 - coord_X_loss: 65.7136 - coord_Y_loss: 72.6480 - amplitude_root_mean_squared_error: 0.3124 - coord_X_root_mean_squared_error: 8.1064 - coord_Y_root_mean_squared_error: 8.5234 - val_loss: 400.6540 - val_amplitude_loss: 0.1032 - val_coord_X_loss: 225.5594 - val_coord_Y_loss: 174.9915 - val_amplitude_root_mean_squared_error: 0.3213 - val_coord_X_root_mean_squared_error: 15.0186 - val_coord_Y_root_mean_squared_error: 13.2284\n",
      "Epoch 82/100\n",
      "9000/9000 [==============================] - 1s 136us/sample - loss: 177.7093 - amplitude_loss: 0.0970 - coord_X_loss: 71.1535 - coord_Y_loss: 106.4587 - amplitude_root_mean_squared_error: 0.3115 - coord_X_root_mean_squared_error: 8.4353 - coord_Y_root_mean_squared_error: 10.3179 - val_loss: 336.8915 - val_amplitude_loss: 0.1049 - val_coord_X_loss: 203.0164 - val_coord_Y_loss: 133.7703 - val_amplitude_root_mean_squared_error: 0.3239 - val_coord_X_root_mean_squared_error: 14.2484 - val_coord_Y_root_mean_squared_error: 11.5659\n",
      "Epoch 83/100\n",
      "9000/9000 [==============================] - 1s 133us/sample - loss: 141.4716 - amplitude_loss: 0.0968 - coord_X_loss: 59.1710 - coord_Y_loss: 82.2039 - amplitude_root_mean_squared_error: 0.3112 - coord_X_root_mean_squared_error: 7.6923 - coord_Y_root_mean_squared_error: 9.0666 - val_loss: 371.3939 - val_amplitude_loss: 0.0977 - val_coord_X_loss: 225.4537 - val_coord_Y_loss: 145.8425 - val_amplitude_root_mean_squared_error: 0.3126 - val_coord_X_root_mean_squared_error: 15.0151 - val_coord_Y_root_mean_squared_error: 12.0765\n",
      "Epoch 84/100\n",
      "9000/9000 [==============================] - 1s 134us/sample - loss: 174.2728 - amplitude_loss: 0.0952 - coord_X_loss: 80.0362 - coord_Y_loss: 94.1414 - amplitude_root_mean_squared_error: 0.3085 - coord_X_root_mean_squared_error: 8.9463 - coord_Y_root_mean_squared_error: 9.7026 - val_loss: 348.5854 - val_amplitude_loss: 0.0978 - val_coord_X_loss: 184.3508 - val_coord_Y_loss: 164.1369 - val_amplitude_root_mean_squared_error: 0.3127 - val_coord_X_root_mean_squared_error: 13.5776 - val_coord_Y_root_mean_squared_error: 12.8116\n",
      "Epoch 85/100\n",
      "9000/9000 [==============================] - 1s 132us/sample - loss: 157.2393 - amplitude_loss: 0.0946 - coord_X_loss: 57.3689 - coord_Y_loss: 99.7758 - amplitude_root_mean_squared_error: 0.3075 - coord_X_root_mean_squared_error: 7.5742 - coord_Y_root_mean_squared_error: 9.9888 - val_loss: 411.4754 - val_amplitude_loss: 0.1137 - val_coord_X_loss: 235.4813 - val_coord_Y_loss: 175.8804 - val_amplitude_root_mean_squared_error: 0.3372 - val_coord_X_root_mean_squared_error: 15.3454 - val_coord_Y_root_mean_squared_error: 13.2620\n",
      "Epoch 86/100\n",
      "9000/9000 [==============================] - 1s 131us/sample - loss: 146.4266 - amplitude_loss: 0.0942 - coord_X_loss: 66.9025 - coord_Y_loss: 79.4299 - amplitude_root_mean_squared_error: 0.3068 - coord_X_root_mean_squared_error: 8.1794 - coord_Y_root_mean_squared_error: 8.9123 - val_loss: 319.4566 - val_amplitude_loss: 0.1129 - val_coord_X_loss: 197.1813 - val_coord_Y_loss: 122.1624 - val_amplitude_root_mean_squared_error: 0.3360 - val_coord_X_root_mean_squared_error: 14.0421 - val_coord_Y_root_mean_squared_error: 11.0527\n",
      "Epoch 87/100\n",
      "9000/9000 [==============================] - 1s 130us/sample - loss: 158.3742 - amplitude_loss: 0.0920 - coord_X_loss: 64.3820 - coord_Y_loss: 93.9002 - amplitude_root_mean_squared_error: 0.3034 - coord_X_root_mean_squared_error: 8.0238 - coord_Y_root_mean_squared_error: 9.6902 - val_loss: 333.3628 - val_amplitude_loss: 0.1144 - val_coord_X_loss: 195.7207 - val_coord_Y_loss: 137.5277 - val_amplitude_root_mean_squared_error: 0.3382 - val_coord_X_root_mean_squared_error: 13.9900 - val_coord_Y_root_mean_squared_error: 11.7272\n",
      "Epoch 88/100\n",
      "9000/9000 [==============================] - 1s 131us/sample - loss: 133.7087 - amplitude_loss: 0.0915 - coord_X_loss: 59.4341 - coord_Y_loss: 74.1831 - amplitude_root_mean_squared_error: 0.3024 - coord_X_root_mean_squared_error: 7.7094 - coord_Y_root_mean_squared_error: 8.6130 - val_loss: 348.3241 - val_amplitude_loss: 0.1002 - val_coord_X_loss: 210.5803 - val_coord_Y_loss: 137.6436 - val_amplitude_root_mean_squared_error: 0.3166 - val_coord_X_root_mean_squared_error: 14.5114 - val_coord_Y_root_mean_squared_error: 11.7322\n",
      "Epoch 89/100\n",
      "9000/9000 [==============================] - 1s 133us/sample - loss: 143.9334 - amplitude_loss: 0.0930 - coord_X_loss: 61.1334 - coord_Y_loss: 82.7070 - amplitude_root_mean_squared_error: 0.3049 - coord_X_root_mean_squared_error: 7.8188 - coord_Y_root_mean_squared_error: 9.0943 - val_loss: 324.5833 - val_amplitude_loss: 0.1014 - val_coord_X_loss: 173.7924 - val_coord_Y_loss: 150.6895 - val_amplitude_root_mean_squared_error: 0.3184 - val_coord_X_root_mean_squared_error: 13.1830 - val_coord_Y_root_mean_squared_error: 12.2756\n",
      "Epoch 90/100\n",
      "9000/9000 [==============================] - 1s 132us/sample - loss: 152.4463 - amplitude_loss: 0.0906 - coord_X_loss: 60.9217 - coord_Y_loss: 91.4339 - amplitude_root_mean_squared_error: 0.3011 - coord_X_root_mean_squared_error: 7.8052 - coord_Y_root_mean_squared_error: 9.5621 - val_loss: 327.5359 - val_amplitude_loss: 0.0975 - val_coord_X_loss: 186.5214 - val_coord_Y_loss: 140.9170 - val_amplitude_root_mean_squared_error: 0.3122 - val_coord_X_root_mean_squared_error: 13.6573 - val_coord_Y_root_mean_squared_error: 11.8708\n",
      "Epoch 91/100\n",
      "9000/9000 [==============================] - 1s 132us/sample - loss: 154.7898 - amplitude_loss: 0.0939 - coord_X_loss: 70.9116 - coord_Y_loss: 83.7843 - amplitude_root_mean_squared_error: 0.3064 - coord_X_root_mean_squared_error: 8.4209 - coord_Y_root_mean_squared_error: 9.1534 - val_loss: 282.7590 - val_amplitude_loss: 0.0923 - val_coord_X_loss: 158.4292 - val_coord_Y_loss: 124.2375 - val_amplitude_root_mean_squared_error: 0.3038 - val_coord_X_root_mean_squared_error: 12.5869 - val_coord_Y_root_mean_squared_error: 11.1462\n",
      "Epoch 92/100\n",
      "9000/9000 [==============================] - ETA: 0s - loss: 122.9603 - amplitude_loss: 0.0907 - coord_X_loss: 52.1410 - coord_Y_loss: 70.7286 - amplitude_root_mean_squared_error: 0.3012 - coord_X_root_mean_squared_error: 7.2209 - coord_Y_root_mean_squared_error: 8.41 - 1s 132us/sample - loss: 121.7400 - amplitude_loss: 0.0912 - coord_X_loss: 51.4348 - coord_Y_loss: 70.2140 - amplitude_root_mean_squared_error: 0.3021 - coord_X_root_mean_squared_error: 7.1718 - coord_Y_root_mean_squared_error: 8.3794 - val_loss: 315.0667 - val_amplitude_loss: 0.0899 - val_coord_X_loss: 183.6074 - val_coord_Y_loss: 131.3694 - val_amplitude_root_mean_squared_error: 0.2999 - val_coord_X_root_mean_squared_error: 13.5502 - val_coord_Y_root_mean_squared_error: 11.4617\n",
      "Epoch 93/100\n",
      "9000/9000 [==============================] - 1s 131us/sample - loss: 131.6905 - amplitude_loss: 0.0890 - coord_X_loss: 61.6344 - coord_Y_loss: 69.9671 - amplitude_root_mean_squared_error: 0.2983 - coord_X_root_mean_squared_error: 7.8508 - coord_Y_root_mean_squared_error: 8.3646 - val_loss: 344.7803 - val_amplitude_loss: 0.0855 - val_coord_X_loss: 193.7006 - val_coord_Y_loss: 150.9942 - val_amplitude_root_mean_squared_error: 0.2924 - val_coord_X_root_mean_squared_error: 13.9176 - val_coord_Y_root_mean_squared_error: 12.2880\n",
      "Epoch 94/100\n",
      "9000/9000 [==============================] - 1s 138us/sample - loss: 156.3723 - amplitude_loss: 0.0902 - coord_X_loss: 68.6894 - coord_Y_loss: 87.5927 - amplitude_root_mean_squared_error: 0.3004 - coord_X_root_mean_squared_error: 8.2879 - coord_Y_root_mean_squared_error: 9.3591 - val_loss: 278.9158 - val_amplitude_loss: 0.0920 - val_coord_X_loss: 164.5418 - val_coord_Y_loss: 114.2820 - val_amplitude_root_mean_squared_error: 0.3033 - val_coord_X_root_mean_squared_error: 12.8274 - val_coord_Y_root_mean_squared_error: 10.6903\n",
      "Epoch 95/100\n",
      "9000/9000 [==============================] - 1s 133us/sample - loss: 131.6743 - amplitude_loss: 0.0889 - coord_X_loss: 55.4808 - coord_Y_loss: 76.1046 - amplitude_root_mean_squared_error: 0.2982 - coord_X_root_mean_squared_error: 7.4485 - coord_Y_root_mean_squared_error: 8.7238 - val_loss: 309.7527 - val_amplitude_loss: 0.0925 - val_coord_X_loss: 206.8118 - val_coord_Y_loss: 102.8483 - val_amplitude_root_mean_squared_error: 0.3041 - val_coord_X_root_mean_squared_error: 14.3810 - val_coord_Y_root_mean_squared_error: 10.1414\n",
      "Epoch 96/100\n",
      "9000/9000 [==============================] - 1s 132us/sample - loss: 115.9853 - amplitude_loss: 0.0877 - coord_X_loss: 58.7524 - coord_Y_loss: 57.1451 - amplitude_root_mean_squared_error: 0.2962 - coord_X_root_mean_squared_error: 7.6650 - coord_Y_root_mean_squared_error: 7.5594 - val_loss: 293.8417 - val_amplitude_loss: 0.0905 - val_coord_X_loss: 173.7878 - val_coord_Y_loss: 119.9634 - val_amplitude_root_mean_squared_error: 0.3009 - val_coord_X_root_mean_squared_error: 13.1829 - val_coord_Y_root_mean_squared_error: 10.9528\n",
      "Epoch 97/100\n",
      "9000/9000 [==============================] - 1s 132us/sample - loss: 129.7114 - amplitude_loss: 0.0859 - coord_X_loss: 58.4238 - coord_Y_loss: 71.2017 - amplitude_root_mean_squared_error: 0.2931 - coord_X_root_mean_squared_error: 7.6435 - coord_Y_root_mean_squared_error: 8.4381 - val_loss: 374.0188 - val_amplitude_loss: 0.0911 - val_coord_X_loss: 190.2723 - val_coord_Y_loss: 183.6554 - val_amplitude_root_mean_squared_error: 0.3019 - val_coord_X_root_mean_squared_error: 13.7939 - val_coord_Y_root_mean_squared_error: 13.5520\n",
      "Epoch 98/100\n",
      "9000/9000 [==============================] - 1s 131us/sample - loss: 157.3140 - amplitude_loss: 0.0851 - coord_X_loss: 70.7551 - coord_Y_loss: 86.4737 - amplitude_root_mean_squared_error: 0.2917 - coord_X_root_mean_squared_error: 8.4116 - coord_Y_root_mean_squared_error: 9.2991 - val_loss: 353.1592 - val_amplitude_loss: 0.0944 - val_coord_X_loss: 212.4596 - val_coord_Y_loss: 140.6052 - val_amplitude_root_mean_squared_error: 0.3073 - val_coord_X_root_mean_squared_error: 14.5760 - val_coord_Y_root_mean_squared_error: 11.8577\n",
      "Epoch 99/100\n",
      "9000/9000 [==============================] - 1s 132us/sample - loss: 145.8636 - amplitude_loss: 0.0902 - coord_X_loss: 61.1946 - coord_Y_loss: 84.5788 - amplitude_root_mean_squared_error: 0.3004 - coord_X_root_mean_squared_error: 7.8227 - coord_Y_root_mean_squared_error: 9.1967 - val_loss: 340.1913 - val_amplitude_loss: 0.0919 - val_coord_X_loss: 187.1844 - val_coord_Y_loss: 152.9149 - val_amplitude_root_mean_squared_error: 0.3031 - val_coord_X_root_mean_squared_error: 13.6815 - val_coord_Y_root_mean_squared_error: 12.3659\n",
      "Epoch 100/100\n",
      "9000/9000 [==============================] - 1s 132us/sample - loss: 131.6348 - amplitude_loss: 0.0904 - coord_X_loss: 55.1268 - coord_Y_loss: 76.4175 - amplitude_root_mean_squared_error: 0.3007 - coord_X_root_mean_squared_error: 7.4247 - coord_Y_root_mean_squared_error: 8.7417 - val_loss: 301.7947 - val_amplitude_loss: 0.1015 - val_coord_X_loss: 189.9845 - val_coord_Y_loss: 111.7087 - val_amplitude_root_mean_squared_error: 0.3186 - val_coord_X_root_mean_squared_error: 13.7835 - val_coord_Y_root_mean_squared_error: 10.5692\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(train_X, df_train_y,\n",
    "                    epochs=100, batch_size=10, validation_data=(df_val_X, df_val_y))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b54f0b5b",
   "metadata": {},
   "source": [
    "Test the model and print loss and rmse for each output"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "f98c2e4a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1000/1000 [==============================] - 0s 24us/sample - loss: 301.7947 - amplitude_loss: 0.1015 - coord_X_loss: 193.5195 - coord_Y_loss: 156.3393 - amplitude_root_mean_squared_error: 0.3186 - coord_X_root_mean_squared_error: 13.7835 - coord_Y_root_mean_squared_error: 10.5692\n",
      "\n",
      "loss: 301.7947414550781\n",
      "amp_loss: 0.10145990550518036\n",
      "X_loss: 193.51953125\n",
      "Y_loss: 156.3392791748047\n",
      "amp_rmse: 0.3186219334602356\n",
      "X_rmse: 13.783487319946289\n",
      "Y_rmse: 10.569231986999512\n"
     ]
    }
   ],
   "source": [
    "loss, Y1_loss, Y2_loss, Y3_loss, Y1_rmse, Y2_rmse, Y3_rmse  = model.evaluate(x = df_val_X, y = df_val_y)\n",
    "\n",
    "print()\n",
    "print(f'loss: {loss}')\n",
    "\n",
    "print(f'amp_loss: {Y1_loss}')\n",
    "print(f'X_loss: {Y2_loss}')\n",
    "print(f'Y_loss: {Y3_loss}')\n",
    "\n",
    "print(f'amp_rmse: {Y1_rmse}')\n",
    "print(f'X_rmse: {Y2_rmse}')\n",
    "print(f'Y_rmse: {Y3_rmse}')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e1dbcc13",
   "metadata": {},
   "source": [
    "Helper functions for building plots"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "75acdf1d",
   "metadata": {},
   "outputs": [],
   "source": [
    "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",
    "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",
   "id": "e04f899d",
   "metadata": {},
   "source": [
    "Predict the outputs fot the test dataset and biuld the plots to visualize results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "f38bd2cd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "Y_pred = model.predict(df_test_X)\n",
    "\n",
    "amp_pred = Y_pred[0]\n",
    "coord_X_pred = Y_pred[1]\n",
    "coord_Y_pred = Y_pred[2]\n",
    "\n",
    "plot_diff(df_test_y[0], amp_pred, title='AMPLITUDE')\n",
    "plot_diff(df_test_y[1], coord_X_pred, title='COORDINATE_X')\n",
    "plot_diff(df_test_y[2], coord_Y_pred, title='COORDINATE_Y')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "f971ddc8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot RMSE\n",
    "plot_metrics(metric_name='amplitude_root_mean_squared_error', title='Amplitude RMSE', ylim=1.75)\n",
    "plot_metrics(metric_name='coord_X_root_mean_squared_error', title='Coordinate X RMSE', ylim=60)\n",
    "plot_metrics(metric_name='coord_Y_root_mean_squared_error', title='Coordinate Y RMSE', ylim=60)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8ed8aac9",
   "metadata": {},
   "source": [
    "Predict outputs for judge data and create a dataframe with results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "6f64e9ed",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>amplitude</th>\n",
       "      <th>X</th>\n",
       "      <th>Y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3.776049</td>\n",
       "      <td>286.604523</td>\n",
       "      <td>-79.535728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.255513</td>\n",
       "      <td>203.767838</td>\n",
       "      <td>-824.392517</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2.450529</td>\n",
       "      <td>-35.189953</td>\n",
       "      <td>25.657576</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.748884</td>\n",
       "      <td>554.403137</td>\n",
       "      <td>-192.849564</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3.834386</td>\n",
       "      <td>-282.510345</td>\n",
       "      <td>-364.280304</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1995</th>\n",
       "      <td>0.383825</td>\n",
       "      <td>-992.262146</td>\n",
       "      <td>386.812378</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1996</th>\n",
       "      <td>5.058572</td>\n",
       "      <td>311.958069</td>\n",
       "      <td>-242.305527</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1997</th>\n",
       "      <td>0.291929</td>\n",
       "      <td>276.632355</td>\n",
       "      <td>-32.820583</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998</th>\n",
       "      <td>4.944913</td>\n",
       "      <td>-438.278931</td>\n",
       "      <td>132.765823</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1999</th>\n",
       "      <td>4.745356</td>\n",
       "      <td>627.073914</td>\n",
       "      <td>-102.175629</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2000 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      amplitude           X           Y\n",
       "0      3.776049  286.604523  -79.535728\n",
       "1      0.255513  203.767838 -824.392517\n",
       "2      2.450529  -35.189953   25.657576\n",
       "3      1.748884  554.403137 -192.849564\n",
       "4      3.834386 -282.510345 -364.280304\n",
       "...         ...         ...         ...\n",
       "1995   0.383825 -992.262146  386.812378\n",
       "1996   5.058572  311.958069 -242.305527\n",
       "1997   0.291929  276.632355  -32.820583\n",
       "1998   4.944913 -438.278931  132.765823\n",
       "1999   4.745356  627.073914 -102.175629\n",
       "\n",
       "[2000 rows x 3 columns]"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "judge_predictions = model.predict(df_judge_X)\n",
    "\n",
    "amp_pred = judge_predictions[0]\n",
    "coord_X_pred = judge_predictions[1]\n",
    "coord_Y_pred = judge_predictions[2]\n",
    "\n",
    "output_judge = pd.DataFrame({'amplitude':amp_pred.flatten(), 'X':coord_X_pred.flatten(),'Y':coord_Y_pred.flatten()})\n",
    "output_judge"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "498157af",
   "metadata": {},
   "source": [
    "Create the csv-file with outputs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "e4c48c8e",
   "metadata": {},
   "outputs": [],
   "source": [
    "path = \"C:/Users/matsiuk/Documents/Python Scripts/DATA/Problem1\"\n",
    "output_judge.to_csv(os.path.join(path,\"output_judge_Pr1_multioutput_model.csv\"),index=False,header=False)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "gametrick",
   "language": "python",
   "name": "gametrick"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}