kerasSyntaxNotes.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 154,
   "id": "2b20a18c",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "id": "4d9770e2",
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 156,
   "id": "a0e90370",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('../DATA/fake_reg.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "id": "e678941c",
   "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>price</th>\n",
       "      <th>feature1</th>\n",
       "      <th>feature2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>461.527929</td>\n",
       "      <td>999.787558</td>\n",
       "      <td>999.766096</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>548.130011</td>\n",
       "      <td>998.861615</td>\n",
       "      <td>1001.042403</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>410.297162</td>\n",
       "      <td>1000.070267</td>\n",
       "      <td>998.844015</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>540.382220</td>\n",
       "      <td>999.952251</td>\n",
       "      <td>1000.440940</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>546.024553</td>\n",
       "      <td>1000.446011</td>\n",
       "      <td>1000.338531</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        price     feature1     feature2\n",
       "0  461.527929   999.787558   999.766096\n",
       "1  548.130011   998.861615  1001.042403\n",
       "2  410.297162  1000.070267   998.844015\n",
       "3  540.382220   999.952251  1000.440940\n",
       "4  546.024553  1000.446011  1000.338531"
      ]
     },
     "execution_count": 157,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "id": "883069be",
   "metadata": {},
   "outputs": [],
   "source": [
    "#will treat this as a regression problem. Predict price based on \n",
    "#feature 1 and feature 2\n",
    "\n",
    "#This is a supervised learning problem because we have historical info."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 158,
   "id": "f50b70ac",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.PairGrid at 0x7f9a115b15e0>"
      ]
     },
     "execution_count": 158,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 750x750 with 12 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.pairplot(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "2283bf1e",
   "metadata": {},
   "outputs": [],
   "source": [
    "#feature 2 has a high correlation with the price (this is an indication\n",
    "#that the dataset is fake)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "e985954d",
   "metadata": {},
   "outputs": [],
   "source": [
    "#General steps:\n",
    "\n",
    "#1 - read in data\n",
    "#2 - feature engineering and exploration\n",
    "#3 - train test split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "e9476e23",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 159,
   "id": "5536575c",
   "metadata": {},
   "outputs": [],
   "source": [
    "X = df[['feature1','feature2']].values\n",
    "y = df['price'].values\n",
    "\n",
    "#for Tensorflow, numpy arrays must be used instead of pandas dataframes\n",
    "#or series.\n",
    "#This is accomplished by the .values method\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, \n",
    "    test_size=0.3, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "id": "e52e2af4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(700, 2)"
      ]
     },
     "execution_count": 160,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 161,
   "id": "8b9441b5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(300, 2)"
      ]
     },
     "execution_count": 161,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "id": "112825c7",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "4db7149e",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Help on class MinMaxScaler in module sklearn.preprocessing._data:\n",
      "\n",
      "class MinMaxScaler(sklearn.base._OneToOneFeatureMixin, sklearn.base.TransformerMixin, sklearn.base.BaseEstimator)\n",
      " |  MinMaxScaler(feature_range=(0, 1), *, copy=True, clip=False)\n",
      " |  \n",
      " |  Transform features by scaling each feature to a given range.\n",
      " |  \n",
      " |  This estimator scales and translates each feature individually such\n",
      " |  that it is in the given range on the training set, e.g. between\n",
      " |  zero and one.\n",
      " |  \n",
      " |  The transformation is given by::\n",
      " |  \n",
      " |      X_std = (X - X.min(axis=0)) / (X.max(axis=0) - X.min(axis=0))\n",
      " |      X_scaled = X_std * (max - min) + min\n",
      " |  \n",
      " |  where min, max = feature_range.\n",
      " |  \n",
      " |  This transformation is often used as an alternative to zero mean,\n",
      " |  unit variance scaling.\n",
      " |  \n",
      " |  Read more in the :ref:`User Guide <preprocessing_scaler>`.\n",
      " |  \n",
      " |  Parameters\n",
      " |  ----------\n",
      " |  feature_range : tuple (min, max), default=(0, 1)\n",
      " |      Desired range of transformed data.\n",
      " |  \n",
      " |  copy : bool, default=True\n",
      " |      Set to False to perform inplace row normalization and avoid a\n",
      " |      copy (if the input is already a numpy array).\n",
      " |  \n",
      " |  clip : bool, default=False\n",
      " |      Set to True to clip transformed values of held-out data to\n",
      " |      provided `feature range`.\n",
      " |  \n",
      " |      .. versionadded:: 0.24\n",
      " |  \n",
      " |  Attributes\n",
      " |  ----------\n",
      " |  min_ : ndarray of shape (n_features,)\n",
      " |      Per feature adjustment for minimum. Equivalent to\n",
      " |      ``min - X.min(axis=0) * self.scale_``\n",
      " |  \n",
      " |  scale_ : ndarray of shape (n_features,)\n",
      " |      Per feature relative scaling of the data. Equivalent to\n",
      " |      ``(max - min) / (X.max(axis=0) - X.min(axis=0))``\n",
      " |  \n",
      " |      .. versionadded:: 0.17\n",
      " |         *scale_* attribute.\n",
      " |  \n",
      " |  data_min_ : ndarray of shape (n_features,)\n",
      " |      Per feature minimum seen in the data\n",
      " |  \n",
      " |      .. versionadded:: 0.17\n",
      " |         *data_min_*\n",
      " |  \n",
      " |  data_max_ : ndarray of shape (n_features,)\n",
      " |      Per feature maximum seen in the data\n",
      " |  \n",
      " |      .. versionadded:: 0.17\n",
      " |         *data_max_*\n",
      " |  \n",
      " |  data_range_ : ndarray of shape (n_features,)\n",
      " |      Per feature range ``(data_max_ - data_min_)`` seen in the data\n",
      " |  \n",
      " |      .. versionadded:: 0.17\n",
      " |         *data_range_*\n",
      " |  \n",
      " |  n_features_in_ : int\n",
      " |      Number of features seen during :term:`fit`.\n",
      " |  \n",
      " |      .. versionadded:: 0.24\n",
      " |  \n",
      " |  n_samples_seen_ : int\n",
      " |      The number of samples processed by the estimator.\n",
      " |      It will be reset on new calls to fit, but increments across\n",
      " |      ``partial_fit`` calls.\n",
      " |  \n",
      " |  feature_names_in_ : ndarray of shape (`n_features_in_`,)\n",
      " |      Names of features seen during :term:`fit`. Defined only when `X`\n",
      " |      has feature names that are all strings.\n",
      " |  \n",
      " |      .. versionadded:: 1.0\n",
      " |  \n",
      " |  See Also\n",
      " |  --------\n",
      " |  minmax_scale : Equivalent function without the estimator API.\n",
      " |  \n",
      " |  Notes\n",
      " |  -----\n",
      " |  NaNs are treated as missing values: disregarded in fit, and maintained in\n",
      " |  transform.\n",
      " |  \n",
      " |  For a comparison of the different scalers, transformers, and normalizers,\n",
      " |  see :ref:`examples/preprocessing/plot_all_scaling.py\n",
      " |  <sphx_glr_auto_examples_preprocessing_plot_all_scaling.py>`.\n",
      " |  \n",
      " |  Examples\n",
      " |  --------\n",
      " |  >>> from sklearn.preprocessing import MinMaxScaler\n",
      " |  >>> data = [[-1, 2], [-0.5, 6], [0, 10], [1, 18]]\n",
      " |  >>> scaler = MinMaxScaler()\n",
      " |  >>> print(scaler.fit(data))\n",
      " |  MinMaxScaler()\n",
      " |  >>> print(scaler.data_max_)\n",
      " |  [ 1. 18.]\n",
      " |  >>> print(scaler.transform(data))\n",
      " |  [[0.   0.  ]\n",
      " |   [0.25 0.25]\n",
      " |   [0.5  0.5 ]\n",
      " |   [1.   1.  ]]\n",
      " |  >>> print(scaler.transform([[2, 2]]))\n",
      " |  [[1.5 0. ]]\n",
      " |  \n",
      " |  Method resolution order:\n",
      " |      MinMaxScaler\n",
      " |      sklearn.base._OneToOneFeatureMixin\n",
      " |      sklearn.base.TransformerMixin\n",
      " |      sklearn.base.BaseEstimator\n",
      " |      builtins.object\n",
      " |  \n",
      " |  Methods defined here:\n",
      " |  \n",
      " |  __init__(self, feature_range=(0, 1), *, copy=True, clip=False)\n",
      " |      Initialize self.  See help(type(self)) for accurate signature.\n",
      " |  \n",
      " |  fit(self, X, y=None)\n",
      " |      Compute the minimum and maximum to be used for later scaling.\n",
      " |      \n",
      " |      Parameters\n",
      " |      ----------\n",
      " |      X : array-like of shape (n_samples, n_features)\n",
      " |          The data used to compute the per-feature minimum and maximum\n",
      " |          used for later scaling along the features axis.\n",
      " |      \n",
      " |      y : None\n",
      " |          Ignored.\n",
      " |      \n",
      " |      Returns\n",
      " |      -------\n",
      " |      self : object\n",
      " |          Fitted scaler.\n",
      " |  \n",
      " |  inverse_transform(self, X)\n",
      " |      Undo the scaling of X according to feature_range.\n",
      " |      \n",
      " |      Parameters\n",
      " |      ----------\n",
      " |      X : array-like of shape (n_samples, n_features)\n",
      " |          Input data that will be transformed. It cannot be sparse.\n",
      " |      \n",
      " |      Returns\n",
      " |      -------\n",
      " |      Xt : ndarray of shape (n_samples, n_features)\n",
      " |          Transformed data.\n",
      " |  \n",
      " |  partial_fit(self, X, y=None)\n",
      " |      Online computation of min and max on X for later scaling.\n",
      " |      \n",
      " |      All of X is processed as a single batch. This is intended for cases\n",
      " |      when :meth:`fit` is not feasible due to very large number of\n",
      " |      `n_samples` or because X is read from a continuous stream.\n",
      " |      \n",
      " |      Parameters\n",
      " |      ----------\n",
      " |      X : array-like of shape (n_samples, n_features)\n",
      " |          The data used to compute the mean and standard deviation\n",
      " |          used for later scaling along the features axis.\n",
      " |      \n",
      " |      y : None\n",
      " |          Ignored.\n",
      " |      \n",
      " |      Returns\n",
      " |      -------\n",
      " |      self : object\n",
      " |          Fitted scaler.\n",
      " |  \n",
      " |  transform(self, X)\n",
      " |      Scale features of X according to feature_range.\n",
      " |      \n",
      " |      Parameters\n",
      " |      ----------\n",
      " |      X : array-like of shape (n_samples, n_features)\n",
      " |          Input data that will be transformed.\n",
      " |      \n",
      " |      Returns\n",
      " |      -------\n",
      " |      Xt : ndarray of shape (n_samples, n_features)\n",
      " |          Transformed data.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Methods inherited from sklearn.base._OneToOneFeatureMixin:\n",
      " |  \n",
      " |  get_feature_names_out(self, input_features=None)\n",
      " |      Get output feature names for transformation.\n",
      " |      \n",
      " |      Parameters\n",
      " |      ----------\n",
      " |      input_features : array-like of str or None, default=None\n",
      " |          Input features.\n",
      " |      \n",
      " |          - If `input_features` is `None`, then `feature_names_in_` is\n",
      " |            used as feature names in. If `feature_names_in_` is not defined,\n",
      " |            then names are generated: `[x0, x1, ..., x(n_features_in_)]`.\n",
      " |          - If `input_features` is an array-like, then `input_features` must\n",
      " |            match `feature_names_in_` if `feature_names_in_` is defined.\n",
      " |      \n",
      " |      Returns\n",
      " |      -------\n",
      " |      feature_names_out : ndarray of str objects\n",
      " |          Same as input features.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Data descriptors inherited from sklearn.base._OneToOneFeatureMixin:\n",
      " |  \n",
      " |  __dict__\n",
      " |      dictionary for instance variables (if defined)\n",
      " |  \n",
      " |  __weakref__\n",
      " |      list of weak references to the object (if defined)\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Methods inherited from sklearn.base.TransformerMixin:\n",
      " |  \n",
      " |  fit_transform(self, X, y=None, **fit_params)\n",
      " |      Fit to data, then transform it.\n",
      " |      \n",
      " |      Fits transformer to `X` and `y` with optional parameters `fit_params`\n",
      " |      and returns a transformed version of `X`.\n",
      " |      \n",
      " |      Parameters\n",
      " |      ----------\n",
      " |      X : array-like of shape (n_samples, n_features)\n",
      " |          Input samples.\n",
      " |      \n",
      " |      y :  array-like of shape (n_samples,) or (n_samples, n_outputs),                 default=None\n",
      " |          Target values (None for unsupervised transformations).\n",
      " |      \n",
      " |      **fit_params : dict\n",
      " |          Additional fit parameters.\n",
      " |      \n",
      " |      Returns\n",
      " |      -------\n",
      " |      X_new : ndarray array of shape (n_samples, n_features_new)\n",
      " |          Transformed array.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Methods inherited from sklearn.base.BaseEstimator:\n",
      " |  \n",
      " |  __getstate__(self)\n",
      " |  \n",
      " |  __repr__(self, N_CHAR_MAX=700)\n",
      " |      Return repr(self).\n",
      " |  \n",
      " |  __setstate__(self, state)\n",
      " |  \n",
      " |  get_params(self, deep=True)\n",
      " |      Get parameters for this estimator.\n",
      " |      \n",
      " |      Parameters\n",
      " |      ----------\n",
      " |      deep : bool, default=True\n",
      " |          If True, will return the parameters for this estimator and\n",
      " |          contained subobjects that are estimators.\n",
      " |      \n",
      " |      Returns\n",
      " |      -------\n",
      " |      params : dict\n",
      " |          Parameter names mapped to their values.\n",
      " |  \n",
      " |  set_params(self, **params)\n",
      " |      Set the parameters of this estimator.\n",
      " |      \n",
      " |      The method works on simple estimators as well as on nested objects\n",
      " |      (such as :class:`~sklearn.pipeline.Pipeline`). The latter have\n",
      " |      parameters of the form ``<component>__<parameter>`` so that it's\n",
      " |      possible to update each component of a nested object.\n",
      " |      \n",
      " |      Parameters\n",
      " |      ----------\n",
      " |      **params : dict\n",
      " |          Estimator parameters.\n",
      " |      \n",
      " |      Returns\n",
      " |      -------\n",
      " |      self : estimator instance\n",
      " |          Estimator instance.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "help(MinMaxScaler)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "2fb9d606",
   "metadata": {},
   "outputs": [],
   "source": [
    "#essentially transforms data based on standard deviation as well as \n",
    "#min and max values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "18292ea1",
   "metadata": {},
   "outputs": [],
   "source": [
    "#we will only scale the features, not the label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "id": "c3f9b33f",
   "metadata": {},
   "outputs": [],
   "source": [
    "scaler = MinMaxScaler()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "id": "0f3f4126",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "MinMaxScaler()"
      ]
     },
     "execution_count": 164,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scaler.fit(X_train)\n",
    "#this only calculates the parameters it needs to perform the actual \n",
    "#scaling later\n",
    "#fit essentially calculates the standard deviation, min, and max"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "id": "df88091e",
   "metadata": {},
   "outputs": [],
   "source": [
    "#this is not done on the test set to prevent data leakage; we do not\n",
    "#want to assume that we have prior info on the thest set, so the scaler\n",
    "#is set only to the training set. Otherwise this could be considered\n",
    "#'cheating' by allowing the scaler to look at the test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "id": "307346b7",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = scaler.transform(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "id": "084ef0b0",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_test = scaler.transform(X_test)\n",
    "\n",
    "#we still scale this, but we do not allow the fit to account for test \n",
    "#data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "id": "e7c865d4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.74046017, 0.32583248],\n",
       "       [0.43166001, 0.2555088 ],\n",
       "       [0.18468554, 0.70500664],\n",
       "       ...,\n",
       "       [0.54913363, 0.79933822],\n",
       "       [0.2834197 , 0.38818708],\n",
       "       [0.56282703, 0.42371827]])"
      ]
     },
     "execution_count": 168,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "id": "56e40325",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0"
      ]
     },
     "execution_count": 148,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.max()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 149,
   "id": "02e6ba07",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0"
      ]
     },
     "execution_count": 149,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.min()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "1c03d821",
   "metadata": {},
   "outputs": [],
   "source": [
    "#everything is now scaled between 0 and 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "fa267e07",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Next step is to create the model with keras syntax"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 169,
   "id": "2ff4b4a0",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "450f9512",
   "metadata": {},
   "outputs": [],
   "source": [
    "#we will set up a base sequential model and then keep adding layers to\n",
    "#it, which in this case will be a simple dense layer."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "1f84b40e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Help on class Sequential in module keras.engine.sequential:\n",
      "\n",
      "class Sequential(keras.engine.functional.Functional)\n",
      " |  Sequential(layers=None, name=None)\n",
      " |  \n",
      " |  `Sequential` groups a linear stack of layers into a `tf.keras.Model`.\n",
      " |  \n",
      " |  `Sequential` provides training and inference features on this model.\n",
      " |  \n",
      " |  Examples:\n",
      " |  \n",
      " |  ```python\n",
      " |  # Optionally, the first layer can receive an `input_shape` argument:\n",
      " |  model = tf.keras.Sequential()\n",
      " |  model.add(tf.keras.layers.Dense(8, input_shape=(16,)))\n",
      " |  # Afterwards, we do automatic shape inference:\n",
      " |  model.add(tf.keras.layers.Dense(4))\n",
      " |  \n",
      " |  # This is identical to the following:\n",
      " |  model = tf.keras.Sequential()\n",
      " |  model.add(tf.keras.Input(shape=(16,)))\n",
      " |  model.add(tf.keras.layers.Dense(8))\n",
      " |  \n",
      " |  # Note that you can also omit the `input_shape` argument.\n",
      " |  # In that case the model doesn't have any weights until the first call\n",
      " |  # to a training/evaluation method (since it isn't yet built):\n",
      " |  model = tf.keras.Sequential()\n",
      " |  model.add(tf.keras.layers.Dense(8))\n",
      " |  model.add(tf.keras.layers.Dense(4))\n",
      " |  # model.weights not created yet\n",
      " |  \n",
      " |  # Whereas if you specify the input shape, the model gets built\n",
      " |  # continuously as you are adding layers:\n",
      " |  model = tf.keras.Sequential()\n",
      " |  model.add(tf.keras.layers.Dense(8, input_shape=(16,)))\n",
      " |  model.add(tf.keras.layers.Dense(4))\n",
      " |  len(model.weights)\n",
      " |  # Returns \"4\"\n",
      " |  \n",
      " |  # When using the delayed-build pattern (no input shape specified), you can\n",
      " |  # choose to manually build your model by calling\n",
      " |  # `build(batch_input_shape)`:\n",
      " |  model = tf.keras.Sequential()\n",
      " |  model.add(tf.keras.layers.Dense(8))\n",
      " |  model.add(tf.keras.layers.Dense(4))\n",
      " |  model.build((None, 16))\n",
      " |  len(model.weights)\n",
      " |  # Returns \"4\"\n",
      " |  \n",
      " |  # Note that when using the delayed-build pattern (no input shape specified),\n",
      " |  # the model gets built the first time you call `fit`, `eval`, or `predict`,\n",
      " |  # or the first time you call the model on some input data.\n",
      " |  model = tf.keras.Sequential()\n",
      " |  model.add(tf.keras.layers.Dense(8))\n",
      " |  model.add(tf.keras.layers.Dense(1))\n",
      " |  model.compile(optimizer='sgd', loss='mse')\n",
      " |  # This builds the model for the first time:\n",
      " |  model.fit(x, y, batch_size=32, epochs=10)\n",
      " |  ```\n",
      " |  \n",
      " |  Method resolution order:\n",
      " |      Sequential\n",
      " |      keras.engine.functional.Functional\n",
      " |      keras.engine.training.Model\n",
      " |      keras.engine.base_layer.Layer\n",
      " |      tensorflow.python.module.module.Module\n",
      " |      tensorflow.python.trackable.autotrackable.AutoTrackable\n",
      " |      tensorflow.python.trackable.base.Trackable\n",
      " |      keras.utils.version_utils.LayerVersionSelector\n",
      " |      keras.utils.version_utils.ModelVersionSelector\n",
      " |      builtins.object\n",
      " |  \n",
      " |  Methods defined here:\n",
      " |  \n",
      " |  __init__(self, layers=None, name=None)\n",
      " |      Creates a `Sequential` model instance.\n",
      " |      \n",
      " |      Args:\n",
      " |        layers: Optional list of layers to add to the model.\n",
      " |        name: Optional name for the model.\n",
      " |  \n",
      " |  add(self, layer)\n",
      " |      Adds a layer instance on top of the layer stack.\n",
      " |      \n",
      " |      Args:\n",
      " |          layer: layer instance.\n",
      " |      \n",
      " |      Raises:\n",
      " |          TypeError: If `layer` is not a layer instance.\n",
      " |          ValueError: In case the `layer` argument does not\n",
      " |              know its input shape.\n",
      " |          ValueError: In case the `layer` argument has\n",
      " |              multiple output tensors, or is already connected\n",
      " |              somewhere else (forbidden in `Sequential` models).\n",
      " |  \n",
      " |  build(self, input_shape=None)\n",
      " |      Builds the model based on input shapes received.\n",
      " |      \n",
      " |      This is to be used for subclassed models, which do not know at\n",
      " |      instantiation time what their inputs look like.\n",
      " |      \n",
      " |      This method only exists for users who want to call `model.build()` in a\n",
      " |      standalone way (as a substitute for calling the model on real data to\n",
      " |      build it). It will never be called by the framework (and thus it will\n",
      " |      never throw unexpected errors in an unrelated workflow).\n",
      " |      \n",
      " |      Args:\n",
      " |       input_shape: Single tuple, `TensorShape` instance, or list/dict of\n",
      " |         shapes, where shapes are tuples, integers, or `TensorShape`\n",
      " |         instances.\n",
      " |      \n",
      " |      Raises:\n",
      " |        ValueError:\n",
      " |          1. In case of invalid user-provided data (not of type tuple,\n",
      " |             list, `TensorShape`, or dict).\n",
      " |          2. If the model requires call arguments that are agnostic\n",
      " |             to the input shapes (positional or keyword arg in call\n",
      " |             signature).\n",
      " |          3. If not all layers were properly built.\n",
      " |          4. If float type inputs are not supported within the layers.\n",
      " |      \n",
      " |        In each of these cases, the user should build their model by calling\n",
      " |        it on real tensor data.\n",
      " |  \n",
      " |  call(self, inputs, training=None, mask=None)\n",
      " |      Calls the model on new inputs.\n",
      " |      \n",
      " |      In this case `call` just reapplies\n",
      " |      all ops in the graph to the new inputs\n",
      " |      (e.g. build a new computational graph from the provided inputs).\n",
      " |      \n",
      " |      Args:\n",
      " |          inputs: A tensor or list of tensors.\n",
      " |          training: Boolean or boolean scalar tensor, indicating whether to\n",
      " |              run the `Network` in training mode or inference mode.\n",
      " |          mask: A mask or list of masks. A mask can be\n",
      " |              either a tensor or None (no mask).\n",
      " |      \n",
      " |      Returns:\n",
      " |          A tensor if there is a single output, or\n",
      " |          a list of tensors if there are more than one outputs.\n",
      " |  \n",
      " |  compute_mask(self, inputs, mask)\n",
      " |      Computes an output mask tensor.\n",
      " |      \n",
      " |      Args:\n",
      " |          inputs: Tensor or list of tensors.\n",
      " |          mask: Tensor or list of tensors.\n",
      " |      \n",
      " |      Returns:\n",
      " |          None or a tensor (or list of tensors,\n",
      " |              one per output tensor of the layer).\n",
      " |  \n",
      " |  compute_output_shape(self, input_shape)\n",
      " |      Computes the output shape of the layer.\n",
      " |      \n",
      " |      This method will cause the layer's state to be built, if that has not\n",
      " |      happened before. This requires that the layer will later be used with\n",
      " |      inputs that match the input shape provided here.\n",
      " |      \n",
      " |      Args:\n",
      " |          input_shape: Shape tuple (tuple of integers) or `tf.TensorShape`,\n",
      " |              or structure of shape tuples / `tf.TensorShape` instances\n",
      " |              (one per output tensor of the layer).\n",
      " |              Shape tuples can include None for free dimensions,\n",
      " |              instead of an integer.\n",
      " |      \n",
      " |      Returns:\n",
      " |          A `tf.TensorShape` instance\n",
      " |          or structure of `tf.TensorShape` instances.\n",
      " |  \n",
      " |  get_config(self)\n",
      " |      Returns the config of the `Model`.\n",
      " |      \n",
      " |      Config is a Python dictionary (serializable) containing the\n",
      " |      configuration of an object, which in this case is a `Model`. This allows\n",
      " |      the `Model` to be be reinstantiated later (without its trained weights)\n",
      " |      from this configuration.\n",
      " |      \n",
      " |      Note that `get_config()` does not guarantee to return a fresh copy of\n",
      " |      dict every time it is called. The callers should make a copy of the\n",
      " |      returned dict if they want to modify it.\n",
      " |      \n",
      " |      Developers of subclassed `Model` are advised to override this method,\n",
      " |      and continue to update the dict from `super(MyModel, self).get_config()`\n",
      " |      to provide the proper configuration of this `Model`. The default config\n",
      " |      is an empty dict. Optionally, raise `NotImplementedError` to allow Keras\n",
      " |      to attempt a default serialization.\n",
      " |      \n",
      " |      Returns:\n",
      " |          Python dictionary containing the configuration of this `Model`.\n",
      " |  \n",
      " |  pop(self)\n",
      " |      Removes the last layer in the model.\n",
      " |      \n",
      " |      Raises:\n",
      " |          TypeError: if there are no layers in the model.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Class methods defined here:\n",
      " |  \n",
      " |  from_config(config, custom_objects=None) from builtins.type\n",
      " |      Creates a layer from its config.\n",
      " |      \n",
      " |      This method is the reverse of `get_config`,\n",
      " |      capable of instantiating the same layer from the config\n",
      " |      dictionary. It does not handle layer connectivity\n",
      " |      (handled by Network), nor weights (handled by `set_weights`).\n",
      " |      \n",
      " |      Args:\n",
      " |          config: A Python dictionary, typically the\n",
      " |              output of get_config.\n",
      " |      \n",
      " |      Returns:\n",
      " |          A layer instance.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Readonly properties defined here:\n",
      " |  \n",
      " |  layers\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Data descriptors defined here:\n",
      " |  \n",
      " |  input_spec\n",
      " |      `InputSpec` instance(s) describing the input format for this layer.\n",
      " |      \n",
      " |      When you create a layer subclass, you can set `self.input_spec` to\n",
      " |      enable the layer to run input compatibility checks when it is called.\n",
      " |      Consider a `Conv2D` layer: it can only be called on a single input\n",
      " |      tensor of rank 4. As such, you can set, in `__init__()`:\n",
      " |      \n",
      " |      ```python\n",
      " |      self.input_spec = tf.keras.layers.InputSpec(ndim=4)\n",
      " |      ```\n",
      " |      \n",
      " |      Now, if you try to call the layer on an input that isn't rank 4\n",
      " |      (for instance, an input of shape `(2,)`, it will raise a\n",
      " |      nicely-formatted error:\n",
      " |      \n",
      " |      ```\n",
      " |      ValueError: Input 0 of layer conv2d is incompatible with the layer:\n",
      " |      expected ndim=4, found ndim=1. Full shape received: [2]\n",
      " |      ```\n",
      " |      \n",
      " |      Input checks that can be specified via `input_spec` include:\n",
      " |      - Structure (e.g. a single input, a list of 2 inputs, etc)\n",
      " |      - Shape\n",
      " |      - Rank (ndim)\n",
      " |      - Dtype\n",
      " |      \n",
      " |      For more information, see `tf.keras.layers.InputSpec`.\n",
      " |      \n",
      " |      Returns:\n",
      " |        A `tf.keras.layers.InputSpec` instance, or nested structure thereof.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Methods inherited from keras.engine.functional.Functional:\n",
      " |  \n",
      " |  get_weight_paths(self)\n",
      " |      Retrieve all the variables and their paths for the model.\n",
      " |      \n",
      " |      The variable path (string) is a stable key to indentify a `tf.Variable`\n",
      " |      instance owned by the model. It can be used to specify variable-specific\n",
      " |      configurations (e.g. DTensor, quantization) from a global view.\n",
      " |      \n",
      " |      This method returns a dict with weight object paths as keys\n",
      " |      and the corresponding `tf.Variable` instances as values.\n",
      " |      \n",
      " |      Note that if the model is a subclassed model and the weights haven't\n",
      " |      been initialized, an empty dict will be returned.\n",
      " |      \n",
      " |      Returns:\n",
      " |          A dict where keys are variable paths and values are `tf.Variable`\n",
      " |           instances.\n",
      " |      \n",
      " |      Example:\n",
      " |      \n",
      " |      ```python\n",
      " |      class SubclassModel(tf.keras.Model):\n",
      " |      \n",
      " |        def __init__(self, name=None):\n",
      " |          super().__init__(name=name)\n",
      " |          self.d1 = tf.keras.layers.Dense(10)\n",
      " |          self.d2 = tf.keras.layers.Dense(20)\n",
      " |      \n",
      " |        def call(self, inputs):\n",
      " |          x = self.d1(inputs)\n",
      " |          return self.d2(x)\n",
      " |      \n",
      " |      model = SubclassModel()\n",
      " |      model(tf.zeros((10, 10)))\n",
      " |      weight_paths = model.get_weight_paths()\n",
      " |      # weight_paths:\n",
      " |      # {\n",
      " |      #    'd1.kernel': model.d1.kernel,\n",
      " |      #    'd1.bias': model.d1.bias,\n",
      " |      #    'd2.kernel': model.d2.kernel,\n",
      " |      #    'd2.bias': model.d2.bias,\n",
      " |      # }\n",
      " |      \n",
      " |      # Functional model\n",
      " |      inputs = tf.keras.Input((10,), batch_size=10)\n",
      " |      x = tf.keras.layers.Dense(20, name='d1')(inputs)\n",
      " |      output = tf.keras.layers.Dense(30, name='d2')(x)\n",
      " |      model = tf.keras.Model(inputs, output)\n",
      " |      d1 = model.layers[1]\n",
      " |      d2 = model.layers[2]\n",
      " |      weight_paths = model.get_weight_paths()\n",
      " |      # weight_paths:\n",
      " |      # {\n",
      " |      #    'd1.kernel': d1.kernel,\n",
      " |      #    'd1.bias': d1.bias,\n",
      " |      #    'd2.kernel': d2.kernel,\n",
      " |      #    'd2.bias': d2.bias,\n",
      " |      # }\n",
      " |      ```\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Readonly properties inherited from keras.engine.functional.Functional:\n",
      " |  \n",
      " |  input\n",
      " |      Retrieves the input tensor(s) of a layer.\n",
      " |      \n",
      " |      Only applicable if the layer has exactly one input,\n",
      " |      i.e. if it is connected to one incoming layer.\n",
      " |      \n",
      " |      Returns:\n",
      " |          Input tensor or list of input tensors.\n",
      " |      \n",
      " |      Raises:\n",
      " |        RuntimeError: If called in Eager mode.\n",
      " |        AttributeError: If no inbound nodes are found.\n",
      " |  \n",
      " |  input_shape\n",
      " |      Retrieves the input shape(s) of a layer.\n",
      " |      \n",
      " |      Only applicable if the layer has exactly one input,\n",
      " |      i.e. if it is connected to one incoming layer, or if all inputs\n",
      " |      have the same shape.\n",
      " |      \n",
      " |      Returns:\n",
      " |          Input shape, as an integer shape tuple\n",
      " |          (or list of shape tuples, one tuple per input tensor).\n",
      " |      \n",
      " |      Raises:\n",
      " |          AttributeError: if the layer has no defined input_shape.\n",
      " |          RuntimeError: if called in Eager mode.\n",
      " |  \n",
      " |  output\n",
      " |      Retrieves the output tensor(s) of a layer.\n",
      " |      \n",
      " |      Only applicable if the layer has exactly one output,\n",
      " |      i.e. if it is connected to one incoming layer.\n",
      " |      \n",
      " |      Returns:\n",
      " |        Output tensor or list of output tensors.\n",
      " |      \n",
      " |      Raises:\n",
      " |        AttributeError: if the layer is connected to more than one incoming\n",
      " |          layers.\n",
      " |        RuntimeError: if called in Eager mode.\n",
      " |  \n",
      " |  output_shape\n",
      " |      Retrieves the output shape(s) of a layer.\n",
      " |      \n",
      " |      Only applicable if the layer has one output,\n",
      " |      or if all outputs have the same shape.\n",
      " |      \n",
      " |      Returns:\n",
      " |          Output shape, as an integer shape tuple\n",
      " |          (or list of shape tuples, one tuple per output tensor).\n",
      " |      \n",
      " |      Raises:\n",
      " |          AttributeError: if the layer has no defined output shape.\n",
      " |          RuntimeError: if called in Eager mode.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Methods inherited from keras.engine.training.Model:\n",
      " |  \n",
      " |  __call__(self, *args, **kwargs)\n",
      " |  \n",
      " |  __copy__(self)\n",
      " |  \n",
      " |  __deepcopy__(self, memo)\n",
      " |  \n",
      " |  __reduce__(self)\n",
      " |      Helper for pickle.\n",
      " |  \n",
      " |  __setattr__(self, name, value)\n",
      " |      Support self.foo = trackable syntax.\n",
      " |  \n",
      " |  compile(self, optimizer='rmsprop', loss=None, metrics=None, loss_weights=None, weighted_metrics=None, run_eagerly=None, steps_per_execution=None, jit_compile=None, **kwargs)\n",
      " |      Configures the model for training.\n",
      " |      \n",
      " |      Example:\n",
      " |      \n",
      " |      ```python\n",
      " |      model.compile(optimizer=tf.keras.optimizers.Adam(learning_rate=1e-3),\n",
      " |                    loss=tf.keras.losses.BinaryCrossentropy(),\n",
      " |                    metrics=[tf.keras.metrics.BinaryAccuracy(),\n",
      " |                             tf.keras.metrics.FalseNegatives()])\n",
      " |      ```\n",
      " |      \n",
      " |      Args:\n",
      " |          optimizer: String (name of optimizer) or optimizer instance. See\n",
      " |            `tf.keras.optimizers`.\n",
      " |          loss: Loss function. May be a string (name of loss function), or\n",
      " |            a `tf.keras.losses.Loss` instance. See `tf.keras.losses`. A loss\n",
      " |            function is any callable with the signature `loss = fn(y_true,\n",
      " |            y_pred)`, where `y_true` are the ground truth values, and\n",
      " |            `y_pred` are the model's predictions.\n",
      " |            `y_true` should have shape\n",
      " |            `(batch_size, d0, .. dN)` (except in the case of\n",
      " |            sparse loss functions such as\n",
      " |            sparse categorical crossentropy which expects integer arrays of\n",
      " |            shape `(batch_size, d0, .. dN-1)`).\n",
      " |            `y_pred` should have shape `(batch_size, d0, .. dN)`.\n",
      " |            The loss function should return a float tensor.\n",
      " |            If a custom `Loss` instance is\n",
      " |            used and reduction is set to `None`, return value has shape\n",
      " |            `(batch_size, d0, .. dN-1)` i.e. per-sample or per-timestep loss\n",
      " |            values; otherwise, it is a scalar. If the model has multiple\n",
      " |            outputs, you can use a different loss on each output by passing a\n",
      " |            dictionary or a list of losses. The loss value that will be\n",
      " |            minimized by the model will then be the sum of all individual\n",
      " |            losses, unless `loss_weights` is specified.\n",
      " |          metrics: List of metrics to be evaluated by the model during\n",
      " |            training and testing. Each of this can be a string (name of a\n",
      " |            built-in function), function or a `tf.keras.metrics.Metric`\n",
      " |            instance. See `tf.keras.metrics`. Typically you will use\n",
      " |            `metrics=['accuracy']`.\n",
      " |            A function is any callable with the signature `result = fn(y_true,\n",
      " |            y_pred)`. To specify different metrics for different outputs of a\n",
      " |            multi-output model, you could also pass a dictionary, such as\n",
      " |            `metrics={'output_a':'accuracy', 'output_b':['accuracy', 'mse']}`.\n",
      " |            You can also pass a list to specify a metric or a list of metrics\n",
      " |            for each output, such as\n",
      " |            `metrics=[['accuracy'], ['accuracy', 'mse']]`\n",
      " |            or `metrics=['accuracy', ['accuracy', 'mse']]`. When you pass the\n",
      " |            strings 'accuracy' or 'acc', we convert this to one of\n",
      " |            `tf.keras.metrics.BinaryAccuracy`,\n",
      " |            `tf.keras.metrics.CategoricalAccuracy`,\n",
      " |            `tf.keras.metrics.SparseCategoricalAccuracy` based on the shapes\n",
      " |            of the targets and of the model output. We do a similar\n",
      " |            conversion for the strings 'crossentropy' and 'ce' as well.\n",
      " |            The metrics passed here are evaluated without sample weighting; if\n",
      " |            you would like sample weighting to apply, you can specify your\n",
      " |            metrics via the `weighted_metrics` argument instead.\n",
      " |          loss_weights: Optional list or dictionary specifying scalar\n",
      " |            coefficients (Python floats) to weight the loss contributions of\n",
      " |            different model outputs. The loss value that will be minimized by\n",
      " |            the model will then be the *weighted sum* of all individual\n",
      " |            losses, weighted by the `loss_weights` coefficients.  If a list,\n",
      " |            it is expected to have a 1:1 mapping to the model's outputs. If a\n",
      " |            dict, it is expected to map output names (strings) to scalar\n",
      " |            coefficients.\n",
      " |          weighted_metrics: List of metrics to be evaluated and weighted by\n",
      " |            `sample_weight` or `class_weight` during training and testing.\n",
      " |          run_eagerly: Bool. Defaults to `False`. If `True`, this `Model`'s\n",
      " |            logic will not be wrapped in a `tf.function`. Recommended to leave\n",
      " |            this as `None` unless your `Model` cannot be run inside a\n",
      " |            `tf.function`. `run_eagerly=True` is not supported when using\n",
      " |            `tf.distribute.experimental.ParameterServerStrategy`.\n",
      " |          steps_per_execution: Int. Defaults to 1. The number of batches to\n",
      " |            run during each `tf.function` call. Running multiple batches\n",
      " |            inside a single `tf.function` call can greatly improve performance\n",
      " |            on TPUs or small models with a large Python overhead. At most, one\n",
      " |            full epoch will be run each execution. If a number larger than the\n",
      " |            size of the epoch is passed, the execution will be truncated to\n",
      " |            the size of the epoch. Note that if `steps_per_execution` is set\n",
      " |            to `N`, `Callback.on_batch_begin` and `Callback.on_batch_end`\n",
      " |            methods will only be called every `N` batches (i.e. before/after\n",
      " |            each `tf.function` execution).\n",
      " |          jit_compile: If `True`, compile the model training step with XLA.\n",
      " |            [XLA](https://www.tensorflow.org/xla) is an optimizing compiler\n",
      " |            for machine learning.\n",
      " |            `jit_compile` is not enabled for by default.\n",
      " |            This option cannot be enabled with `run_eagerly=True`.\n",
      " |            Note that `jit_compile=True`\n",
      " |            may not necessarily work for all models.\n",
      " |            For more information on supported operations please refer to the\n",
      " |            [XLA documentation](https://www.tensorflow.org/xla).\n",
      " |            Also refer to\n",
      " |            [known XLA issues](https://www.tensorflow.org/xla/known_issues)\n",
      " |            for more details.\n",
      " |          **kwargs: Arguments supported for backwards compatibility only.\n",
      " |  \n",
      " |  compute_loss(self, x=None, y=None, y_pred=None, sample_weight=None)\n",
      " |      Compute the total loss, validate it, and return it.\n",
      " |      \n",
      " |      Subclasses can optionally override this method to provide custom loss\n",
      " |      computation logic.\n",
      " |      \n",
      " |      Example:\n",
      " |      ```python\n",
      " |      class MyModel(tf.keras.Model):\n",
      " |      \n",
      " |        def __init__(self, *args, **kwargs):\n",
      " |          super(MyModel, self).__init__(*args, **kwargs)\n",
      " |          self.loss_tracker = tf.keras.metrics.Mean(name='loss')\n",
      " |      \n",
      " |        def compute_loss(self, x, y, y_pred, sample_weight):\n",
      " |          loss = tf.reduce_mean(tf.math.squared_difference(y_pred, y))\n",
      " |          loss += tf.add_n(self.losses)\n",
      " |          self.loss_tracker.update_state(loss)\n",
      " |          return loss\n",
      " |      \n",
      " |        def reset_metrics(self):\n",
      " |          self.loss_tracker.reset_states()\n",
      " |      \n",
      " |        @property\n",
      " |        def metrics(self):\n",
      " |          return [self.loss_tracker]\n",
      " |      \n",
      " |      tensors = tf.random.uniform((10, 10)), tf.random.uniform((10,))\n",
      " |      dataset = tf.data.Dataset.from_tensor_slices(tensors).repeat().batch(1)\n",
      " |      \n",
      " |      inputs = tf.keras.layers.Input(shape=(10,), name='my_input')\n",
      " |      outputs = tf.keras.layers.Dense(10)(inputs)\n",
      " |      model = MyModel(inputs, outputs)\n",
      " |      model.add_loss(tf.reduce_sum(outputs))\n",
      " |      \n",
      " |      optimizer = tf.keras.optimizers.SGD()\n",
      " |      model.compile(optimizer, loss='mse', steps_per_execution=10)\n",
      " |      model.fit(dataset, epochs=2, steps_per_epoch=10)\n",
      " |      print('My custom loss: ', model.loss_tracker.result().numpy())\n",
      " |      ```\n",
      " |      \n",
      " |      Args:\n",
      " |        x: Input data.\n",
      " |        y: Target data.\n",
      " |        y_pred: Predictions returned by the model (output of `model(x)`)\n",
      " |        sample_weight: Sample weights for weighting the loss function.\n",
      " |      \n",
      " |      Returns:\n",
      " |        The total loss as a `tf.Tensor`, or `None` if no loss results (which\n",
      " |        is the case when called by `Model.test_step`).\n",
      " |  \n",
      " |  compute_metrics(self, x, y, y_pred, sample_weight)\n",
      " |      Update metric states and collect all metrics to be returned.\n",
      " |      \n",
      " |      Subclasses can optionally override this method to provide custom metric\n",
      " |      updating and collection logic.\n",
      " |      \n",
      " |      Example:\n",
      " |      ```python\n",
      " |      class MyModel(tf.keras.Sequential):\n",
      " |      \n",
      " |        def compute_metrics(self, x, y, y_pred, sample_weight):\n",
      " |      \n",
      " |          # This super call updates `self.compiled_metrics` and returns\n",
      " |          # results for all metrics listed in `self.metrics`.\n",
      " |          metric_results = super(MyModel, self).compute_metrics(\n",
      " |              x, y, y_pred, sample_weight)\n",
      " |      \n",
      " |          # Note that `self.custom_metric` is not listed in `self.metrics`.\n",
      " |          self.custom_metric.update_state(x, y, y_pred, sample_weight)\n",
      " |          metric_results['custom_metric_name'] = self.custom_metric.result()\n",
      " |          return metric_results\n",
      " |      ```\n",
      " |      \n",
      " |      Args:\n",
      " |        x: Input data.\n",
      " |        y: Target data.\n",
      " |        y_pred: Predictions returned by the model (output of `model.call(x)`)\n",
      " |        sample_weight: Sample weights for weighting the loss function.\n",
      " |      \n",
      " |      Returns:\n",
      " |        A `dict` containing values that will be passed to\n",
      " |        `tf.keras.callbacks.CallbackList.on_train_batch_end()`. Typically, the\n",
      " |        values of the metrics listed in `self.metrics` are returned. Example:\n",
      " |        `{'loss': 0.2, 'accuracy': 0.7}`.\n",
      " |  \n",
      " |  evaluate(self, x=None, y=None, batch_size=None, verbose='auto', sample_weight=None, steps=None, callbacks=None, max_queue_size=10, workers=1, use_multiprocessing=False, return_dict=False, **kwargs)\n",
      " |      Returns the loss value & metrics values for the model in test mode.\n",
      " |      \n",
      " |      Computation is done in batches (see the `batch_size` arg.)\n",
      " |      \n",
      " |      Args:\n",
      " |          x: Input data. It could be:\n",
      " |            - A Numpy array (or array-like), or a list of arrays\n",
      " |              (in case the model has multiple inputs).\n",
      " |            - A TensorFlow tensor, or a list of tensors\n",
      " |              (in case the model has multiple inputs).\n",
      " |            - A dict mapping input names to the corresponding array/tensors,\n",
      " |              if the model has named inputs.\n",
      " |            - A `tf.data` dataset. Should return a tuple\n",
      " |              of either `(inputs, targets)` or\n",
      " |              `(inputs, targets, sample_weights)`.\n",
      " |            - A generator or `keras.utils.Sequence` returning `(inputs,\n",
      " |              targets)` or `(inputs, targets, sample_weights)`.\n",
      " |            A more detailed description of unpacking behavior for iterator\n",
      " |            types (Dataset, generator, Sequence) is given in the `Unpacking\n",
      " |            behavior for iterator-like inputs` section of `Model.fit`.\n",
      " |          y: Target data. Like the input data `x`, it could be either Numpy\n",
      " |            array(s) or TensorFlow tensor(s). It should be consistent with `x`\n",
      " |            (you cannot have Numpy inputs and tensor targets, or inversely).\n",
      " |            If `x` is a dataset, generator or `keras.utils.Sequence` instance,\n",
      " |            `y` should not be specified (since targets will be obtained from\n",
      " |            the iterator/dataset).\n",
      " |          batch_size: Integer or `None`. Number of samples per batch of\n",
      " |            computation. If unspecified, `batch_size` will default to 32. Do\n",
      " |            not specify the `batch_size` if your data is in the form of a\n",
      " |            dataset, generators, or `keras.utils.Sequence` instances (since\n",
      " |            they generate batches).\n",
      " |          verbose: `\"auto\"`, 0, 1, or 2. Verbosity mode.\n",
      " |              0 = silent, 1 = progress bar, 2 = single line.\n",
      " |              `\"auto\"` defaults to 1 for most cases, and to 2 when used with\n",
      " |              `ParameterServerStrategy`. Note that the progress bar is not\n",
      " |              particularly useful when logged to a file, so `verbose=2` is\n",
      " |              recommended when not running interactively (e.g. in a production\n",
      " |              environment).\n",
      " |          sample_weight: Optional Numpy array of weights for the test samples,\n",
      " |            used for weighting the loss function. You can either pass a flat\n",
      " |            (1D) Numpy array with the same length as the input samples\n",
      " |              (1:1 mapping between weights and samples), or in the case of\n",
      " |                temporal data, you can pass a 2D array with shape `(samples,\n",
      " |                sequence_length)`, to apply a different weight to every\n",
      " |                timestep of every sample. This argument is not supported when\n",
      " |                `x` is a dataset, instead pass sample weights as the third\n",
      " |                element of `x`.\n",
      " |          steps: Integer or `None`. Total number of steps (batches of samples)\n",
      " |            before declaring the evaluation round finished. Ignored with the\n",
      " |            default value of `None`. If x is a `tf.data` dataset and `steps`\n",
      " |            is None, 'evaluate' will run until the dataset is exhausted. This\n",
      " |            argument is not supported with array inputs.\n",
      " |          callbacks: List of `keras.callbacks.Callback` instances. List of\n",
      " |            callbacks to apply during evaluation. See\n",
      " |            [callbacks](https://www.tensorflow.org/api_docs/python/tf/keras/callbacks).\n",
      " |          max_queue_size: Integer. Used for generator or\n",
      " |            `keras.utils.Sequence` input only. Maximum size for the generator\n",
      " |            queue. If unspecified, `max_queue_size` will default to 10.\n",
      " |          workers: Integer. Used for generator or `keras.utils.Sequence` input\n",
      " |            only. Maximum number of processes to spin up when using\n",
      " |            process-based threading. If unspecified, `workers` will default to\n",
      " |            1.\n",
      " |          use_multiprocessing: Boolean. Used for generator or\n",
      " |            `keras.utils.Sequence` input only. If `True`, use process-based\n",
      " |            threading. If unspecified, `use_multiprocessing` will default to\n",
      " |            `False`. Note that because this implementation relies on\n",
      " |            multiprocessing, you should not pass non-picklable arguments to\n",
      " |            the generator as they can't be passed easily to children\n",
      " |            processes.\n",
      " |          return_dict: If `True`, loss and metric results are returned as a\n",
      " |            dict, with each key being the name of the metric. If `False`, they\n",
      " |            are returned as a list.\n",
      " |          **kwargs: Unused at this time.\n",
      " |      \n",
      " |      See the discussion of `Unpacking behavior for iterator-like inputs` for\n",
      " |      `Model.fit`.\n",
      " |      \n",
      " |      Returns:\n",
      " |          Scalar test loss (if the model has a single output and no metrics)\n",
      " |          or list of scalars (if the model has multiple outputs\n",
      " |          and/or metrics). The attribute `model.metrics_names` will give you\n",
      " |          the display labels for the scalar outputs.\n",
      " |      \n",
      " |      Raises:\n",
      " |          RuntimeError: If `model.evaluate` is wrapped in a `tf.function`.\n",
      " |  \n",
      " |  evaluate_generator(self, generator, steps=None, callbacks=None, max_queue_size=10, workers=1, use_multiprocessing=False, verbose=0)\n",
      " |      Evaluates the model on a data generator.\n",
      " |      \n",
      " |      DEPRECATED:\n",
      " |        `Model.evaluate` now supports generators, so there is no longer any\n",
      " |        need to use this endpoint.\n",
      " |  \n",
      " |  fit(self, x=None, y=None, batch_size=None, epochs=1, verbose='auto', callbacks=None, validation_split=0.0, validation_data=None, shuffle=True, class_weight=None, sample_weight=None, initial_epoch=0, steps_per_epoch=None, validation_steps=None, validation_batch_size=None, validation_freq=1, max_queue_size=10, workers=1, use_multiprocessing=False)\n",
      " |      Trains the model for a fixed number of epochs (iterations on a dataset).\n",
      " |      \n",
      " |      Args:\n",
      " |          x: Input data. It could be:\n",
      " |            - A Numpy array (or array-like), or a list of arrays\n",
      " |              (in case the model has multiple inputs).\n",
      " |            - A TensorFlow tensor, or a list of tensors\n",
      " |              (in case the model has multiple inputs).\n",
      " |            - A dict mapping input names to the corresponding array/tensors,\n",
      " |              if the model has named inputs.\n",
      " |            - A `tf.data` dataset. Should return a tuple\n",
      " |              of either `(inputs, targets)` or\n",
      " |              `(inputs, targets, sample_weights)`.\n",
      " |            - A generator or `keras.utils.Sequence` returning `(inputs,\n",
      " |              targets)` or `(inputs, targets, sample_weights)`.\n",
      " |            - A `tf.keras.utils.experimental.DatasetCreator`, which wraps a\n",
      " |              callable that takes a single argument of type\n",
      " |              `tf.distribute.InputContext`, and returns a `tf.data.Dataset`.\n",
      " |              `DatasetCreator` should be used when users prefer to specify the\n",
      " |              per-replica batching and sharding logic for the `Dataset`.\n",
      " |              See `tf.keras.utils.experimental.DatasetCreator` doc for more\n",
      " |              information.\n",
      " |            A more detailed description of unpacking behavior for iterator\n",
      " |            types (Dataset, generator, Sequence) is given below. If these\n",
      " |            include `sample_weights` as a third component, note that sample\n",
      " |            weighting applies to the `weighted_metrics` argument but not the\n",
      " |            `metrics` argument in `compile()`. If using\n",
      " |            `tf.distribute.experimental.ParameterServerStrategy`, only\n",
      " |            `DatasetCreator` type is supported for `x`.\n",
      " |          y: Target data. Like the input data `x`,\n",
      " |            it could be either Numpy array(s) or TensorFlow tensor(s).\n",
      " |            It should be consistent with `x` (you cannot have Numpy inputs and\n",
      " |            tensor targets, or inversely). If `x` is a dataset, generator,\n",
      " |            or `keras.utils.Sequence` instance, `y` should\n",
      " |            not be specified (since targets will be obtained from `x`).\n",
      " |          batch_size: Integer or `None`.\n",
      " |              Number of samples per gradient update.\n",
      " |              If unspecified, `batch_size` will default to 32.\n",
      " |              Do not specify the `batch_size` if your data is in the\n",
      " |              form of datasets, generators, or `keras.utils.Sequence`\n",
      " |              instances (since they generate batches).\n",
      " |          epochs: Integer. Number of epochs to train the model.\n",
      " |              An epoch is an iteration over the entire `x` and `y`\n",
      " |              data provided\n",
      " |              (unless the `steps_per_epoch` flag is set to\n",
      " |              something other than None).\n",
      " |              Note that in conjunction with `initial_epoch`,\n",
      " |              `epochs` is to be understood as \"final epoch\".\n",
      " |              The model is not trained for a number of iterations\n",
      " |              given by `epochs`, but merely until the epoch\n",
      " |              of index `epochs` is reached.\n",
      " |          verbose: 'auto', 0, 1, or 2. Verbosity mode.\n",
      " |              0 = silent, 1 = progress bar, 2 = one line per epoch.\n",
      " |              'auto' defaults to 1 for most cases, but 2 when used with\n",
      " |              `ParameterServerStrategy`. Note that the progress bar is not\n",
      " |              particularly useful when logged to a file, so verbose=2 is\n",
      " |              recommended when not running interactively (eg, in a production\n",
      " |              environment).\n",
      " |          callbacks: List of `keras.callbacks.Callback` instances.\n",
      " |              List of callbacks to apply during training.\n",
      " |              See `tf.keras.callbacks`. Note\n",
      " |              `tf.keras.callbacks.ProgbarLogger` and\n",
      " |              `tf.keras.callbacks.History` callbacks are created automatically\n",
      " |              and need not be passed into `model.fit`.\n",
      " |              `tf.keras.callbacks.ProgbarLogger` is created or not based on\n",
      " |              `verbose` argument to `model.fit`.\n",
      " |              Callbacks with batch-level calls are currently unsupported with\n",
      " |              `tf.distribute.experimental.ParameterServerStrategy`, and users\n",
      " |              are advised to implement epoch-level calls instead with an\n",
      " |              appropriate `steps_per_epoch` value.\n",
      " |          validation_split: Float between 0 and 1.\n",
      " |              Fraction of the training data to be used as validation data.\n",
      " |              The model will set apart this fraction of the training data,\n",
      " |              will not train on it, and will evaluate\n",
      " |              the loss and any model metrics\n",
      " |              on this data at the end of each epoch.\n",
      " |              The validation data is selected from the last samples\n",
      " |              in the `x` and `y` data provided, before shuffling. This\n",
      " |              argument is not supported when `x` is a dataset, generator or\n",
      " |              `keras.utils.Sequence` instance.\n",
      " |              If both `validation_data` and `validation_split` are provided,\n",
      " |              `validation_data` will override `validation_split`.\n",
      " |              `validation_split` is not yet supported with\n",
      " |              `tf.distribute.experimental.ParameterServerStrategy`.\n",
      " |          validation_data: Data on which to evaluate\n",
      " |              the loss and any model metrics at the end of each epoch.\n",
      " |              The model will not be trained on this data. Thus, note the fact\n",
      " |              that the validation loss of data provided using\n",
      " |              `validation_split` or `validation_data` is not affected by\n",
      " |              regularization layers like noise and dropout.\n",
      " |              `validation_data` will override `validation_split`.\n",
      " |              `validation_data` could be:\n",
      " |                - A tuple `(x_val, y_val)` of Numpy arrays or tensors.\n",
      " |                - A tuple `(x_val, y_val, val_sample_weights)` of NumPy\n",
      " |                  arrays.\n",
      " |                - A `tf.data.Dataset`.\n",
      " |                - A Python generator or `keras.utils.Sequence` returning\n",
      " |                `(inputs, targets)` or `(inputs, targets, sample_weights)`.\n",
      " |              `validation_data` is not yet supported with\n",
      " |              `tf.distribute.experimental.ParameterServerStrategy`.\n",
      " |          shuffle: Boolean (whether to shuffle the training data\n",
      " |              before each epoch) or str (for 'batch'). This argument is\n",
      " |              ignored when `x` is a generator or an object of tf.data.Dataset.\n",
      " |              'batch' is a special option for dealing\n",
      " |              with the limitations of HDF5 data; it shuffles in batch-sized\n",
      " |              chunks. Has no effect when `steps_per_epoch` is not `None`.\n",
      " |          class_weight: Optional dictionary mapping class indices (integers)\n",
      " |              to a weight (float) value, used for weighting the loss function\n",
      " |              (during training only).\n",
      " |              This can be useful to tell the model to\n",
      " |              \"pay more attention\" to samples from\n",
      " |              an under-represented class.\n",
      " |          sample_weight: Optional Numpy array of weights for\n",
      " |              the training samples, used for weighting the loss function\n",
      " |              (during training only). You can either pass a flat (1D)\n",
      " |              Numpy array with the same length as the input samples\n",
      " |              (1:1 mapping between weights and samples),\n",
      " |              or in the case of temporal data,\n",
      " |              you can pass a 2D array with shape\n",
      " |              `(samples, sequence_length)`,\n",
      " |              to apply a different weight to every timestep of every sample.\n",
      " |              This argument is not supported when `x` is a dataset, generator,\n",
      " |              or `keras.utils.Sequence` instance, instead provide the\n",
      " |              sample_weights as the third element of `x`.\n",
      " |              Note that sample weighting does not apply to metrics specified\n",
      " |              via the `metrics` argument in `compile()`. To apply sample\n",
      " |              weighting to your metrics, you can specify them via the\n",
      " |              `weighted_metrics` in `compile()` instead.\n",
      " |          initial_epoch: Integer.\n",
      " |              Epoch at which to start training\n",
      " |              (useful for resuming a previous training run).\n",
      " |          steps_per_epoch: Integer or `None`.\n",
      " |              Total number of steps (batches of samples)\n",
      " |              before declaring one epoch finished and starting the\n",
      " |              next epoch. When training with input tensors such as\n",
      " |              TensorFlow data tensors, the default `None` is equal to\n",
      " |              the number of samples in your dataset divided by\n",
      " |              the batch size, or 1 if that cannot be determined. If x is a\n",
      " |              `tf.data` dataset, and 'steps_per_epoch'\n",
      " |              is None, the epoch will run until the input dataset is\n",
      " |              exhausted.  When passing an infinitely repeating dataset, you\n",
      " |              must specify the `steps_per_epoch` argument. If\n",
      " |              `steps_per_epoch=-1` the training will run indefinitely with an\n",
      " |              infinitely repeating dataset.  This argument is not supported\n",
      " |              with array inputs.\n",
      " |              When using `tf.distribute.experimental.ParameterServerStrategy`:\n",
      " |                * `steps_per_epoch=None` is not supported.\n",
      " |          validation_steps: Only relevant if `validation_data` is provided and\n",
      " |              is a `tf.data` dataset. Total number of steps (batches of\n",
      " |              samples) to draw before stopping when performing validation\n",
      " |              at the end of every epoch. If 'validation_steps' is None,\n",
      " |              validation will run until the `validation_data` dataset is\n",
      " |              exhausted. In the case of an infinitely repeated dataset, it\n",
      " |              will run into an infinite loop. If 'validation_steps' is\n",
      " |              specified and only part of the dataset will be consumed, the\n",
      " |              evaluation will start from the beginning of the dataset at each\n",
      " |              epoch. This ensures that the same validation samples are used\n",
      " |              every time.\n",
      " |          validation_batch_size: Integer or `None`.\n",
      " |              Number of samples per validation batch.\n",
      " |              If unspecified, will default to `batch_size`.\n",
      " |              Do not specify the `validation_batch_size` if your data is in\n",
      " |              the form of datasets, generators, or `keras.utils.Sequence`\n",
      " |              instances (since they generate batches).\n",
      " |          validation_freq: Only relevant if validation data is provided.\n",
      " |            Integer or `collections.abc.Container` instance (e.g. list, tuple,\n",
      " |            etc.).  If an integer, specifies how many training epochs to run\n",
      " |            before a new validation run is performed, e.g. `validation_freq=2`\n",
      " |            runs validation every 2 epochs. If a Container, specifies the\n",
      " |            epochs on which to run validation, e.g.\n",
      " |            `validation_freq=[1, 2, 10]` runs validation at the end of the\n",
      " |            1st, 2nd, and 10th epochs.\n",
      " |          max_queue_size: Integer. Used for generator or\n",
      " |            `keras.utils.Sequence` input only. Maximum size for the generator\n",
      " |            queue.  If unspecified, `max_queue_size` will default to 10.\n",
      " |          workers: Integer. Used for generator or `keras.utils.Sequence` input\n",
      " |              only. Maximum number of processes to spin up\n",
      " |              when using process-based threading. If unspecified, `workers`\n",
      " |              will default to 1.\n",
      " |          use_multiprocessing: Boolean. Used for generator or\n",
      " |              `keras.utils.Sequence` input only. If `True`, use process-based\n",
      " |              threading. If unspecified, `use_multiprocessing` will default to\n",
      " |              `False`. Note that because this implementation relies on\n",
      " |              multiprocessing, you should not pass non-picklable arguments to\n",
      " |              the generator as they can't be passed easily to children\n",
      " |              processes.\n",
      " |      \n",
      " |      Unpacking behavior for iterator-like inputs:\n",
      " |          A common pattern is to pass a tf.data.Dataset, generator, or\n",
      " |        tf.keras.utils.Sequence to the `x` argument of fit, which will in fact\n",
      " |        yield not only features (x) but optionally targets (y) and sample\n",
      " |        weights.  Keras requires that the output of such iterator-likes be\n",
      " |        unambiguous. The iterator should return a tuple of length 1, 2, or 3,\n",
      " |        where the optional second and third elements will be used for y and\n",
      " |        sample_weight respectively. Any other type provided will be wrapped in\n",
      " |        a length one tuple, effectively treating everything as 'x'. When\n",
      " |        yielding dicts, they should still adhere to the top-level tuple\n",
      " |        structure.\n",
      " |        e.g. `({\"x0\": x0, \"x1\": x1}, y)`. Keras will not attempt to separate\n",
      " |        features, targets, and weights from the keys of a single dict.\n",
      " |          A notable unsupported data type is the namedtuple. The reason is\n",
      " |        that it behaves like both an ordered datatype (tuple) and a mapping\n",
      " |        datatype (dict). So given a namedtuple of the form:\n",
      " |            `namedtuple(\"example_tuple\", [\"y\", \"x\"])`\n",
      " |        it is ambiguous whether to reverse the order of the elements when\n",
      " |        interpreting the value. Even worse is a tuple of the form:\n",
      " |            `namedtuple(\"other_tuple\", [\"x\", \"y\", \"z\"])`\n",
      " |        where it is unclear if the tuple was intended to be unpacked into x,\n",
      " |        y, and sample_weight or passed through as a single element to `x`. As\n",
      " |        a result the data processing code will simply raise a ValueError if it\n",
      " |        encounters a namedtuple. (Along with instructions to remedy the\n",
      " |        issue.)\n",
      " |      \n",
      " |      Returns:\n",
      " |          A `History` object. Its `History.history` attribute is\n",
      " |          a record of training loss values and metrics values\n",
      " |          at successive epochs, as well as validation loss values\n",
      " |          and validation metrics values (if applicable).\n",
      " |      \n",
      " |      Raises:\n",
      " |          RuntimeError: 1. If the model was never compiled or,\n",
      " |          2. If `model.fit` is  wrapped in `tf.function`.\n",
      " |      \n",
      " |          ValueError: In case of mismatch between the provided input data\n",
      " |              and what the model expects or when the input data is empty.\n",
      " |  \n",
      " |  fit_generator(self, generator, steps_per_epoch=None, epochs=1, verbose=1, callbacks=None, validation_data=None, validation_steps=None, validation_freq=1, class_weight=None, max_queue_size=10, workers=1, use_multiprocessing=False, shuffle=True, initial_epoch=0)\n",
      " |      Fits the model on data yielded batch-by-batch by a Python generator.\n",
      " |      \n",
      " |      DEPRECATED:\n",
      " |        `Model.fit` now supports generators, so there is no longer any need to\n",
      " |        use this endpoint.\n",
      " |  \n",
      " |  get_layer(self, name=None, index=None)\n",
      " |      Retrieves a layer based on either its name (unique) or index.\n",
      " |      \n",
      " |      If `name` and `index` are both provided, `index` will take precedence.\n",
      " |      Indices are based on order of horizontal graph traversal (bottom-up).\n",
      " |      \n",
      " |      Args:\n",
      " |          name: String, name of layer.\n",
      " |          index: Integer, index of layer.\n",
      " |      \n",
      " |      Returns:\n",
      " |          A layer instance.\n",
      " |  \n",
      " |  get_metrics_result(self)\n",
      " |      Returns the model's metrics values as a dict.\n",
      " |      \n",
      " |      If any of the metric result is a dict (containing multiple metrics),\n",
      " |      each of them gets added to the top level returned dict of this method.\n",
      " |      \n",
      " |      Returns:\n",
      " |        A `dict` containing values of the metrics listed in `self.metrics`.\n",
      " |        Example:\n",
      " |        `{'loss': 0.2, 'accuracy': 0.7}`.\n",
      " |  \n",
      " |  get_weights(self)\n",
      " |      Retrieves the weights of the model.\n",
      " |      \n",
      " |      Returns:\n",
      " |          A flat list of Numpy arrays.\n",
      " |  \n",
      " |  load_weights(self, filepath, by_name=False, skip_mismatch=False, options=None)\n",
      " |      Loads all layer weights, either from a TensorFlow or an HDF5 weight file.\n",
      " |      \n",
      " |      If `by_name` is False weights are loaded based on the network's\n",
      " |      topology. This means the architecture should be the same as when the\n",
      " |      weights were saved.  Note that layers that don't have weights are not\n",
      " |      taken into account in the topological ordering, so adding or removing\n",
      " |      layers is fine as long as they don't have weights.\n",
      " |      \n",
      " |      If `by_name` is True, weights are loaded into layers only if they share\n",
      " |      the same name. This is useful for fine-tuning or transfer-learning\n",
      " |      models where some of the layers have changed.\n",
      " |      \n",
      " |      Only topological loading (`by_name=False`) is supported when loading\n",
      " |      weights from the TensorFlow format. Note that topological loading\n",
      " |      differs slightly between TensorFlow and HDF5 formats for user-defined\n",
      " |      classes inheriting from `tf.keras.Model`: HDF5 loads based on a\n",
      " |      flattened list of weights, while the TensorFlow format loads based on\n",
      " |      the object-local names of attributes to which layers are assigned in the\n",
      " |      `Model`'s constructor.\n",
      " |      \n",
      " |      Args:\n",
      " |          filepath: String, path to the weights file to load. For weight files\n",
      " |              in TensorFlow format, this is the file prefix (the same as was\n",
      " |              passed to `save_weights`). This can also be a path to a\n",
      " |              SavedModel saved from `model.save`.\n",
      " |          by_name: Boolean, whether to load weights by name or by topological\n",
      " |              order. Only topological loading is supported for weight files in\n",
      " |              TensorFlow format.\n",
      " |          skip_mismatch: Boolean, whether to skip loading of layers where\n",
      " |              there is a mismatch in the number of weights, or a mismatch in\n",
      " |              the shape of the weight (only valid when `by_name=True`).\n",
      " |          options: Optional `tf.train.CheckpointOptions` object that specifies\n",
      " |              options for loading weights.\n",
      " |      \n",
      " |      Returns:\n",
      " |          When loading a weight file in TensorFlow format, returns the same\n",
      " |          status object as `tf.train.Checkpoint.restore`. When graph building,\n",
      " |          restore ops are run automatically as soon as the network is built\n",
      " |          (on first call for user-defined classes inheriting from `Model`,\n",
      " |          immediately if it is already built).\n",
      " |      \n",
      " |          When loading weights in HDF5 format, returns `None`.\n",
      " |      \n",
      " |      Raises:\n",
      " |          ImportError: If `h5py` is not available and the weight file is in\n",
      " |            HDF5 format.\n",
      " |          ValueError: If `skip_mismatch` is set to `True` when `by_name` is\n",
      " |            `False`.\n",
      " |  \n",
      " |  make_predict_function(self, force=False)\n",
      " |      Creates a function that executes one step of inference.\n",
      " |      \n",
      " |      This method can be overridden to support custom inference logic.\n",
      " |      This method is called by `Model.predict` and `Model.predict_on_batch`.\n",
      " |      \n",
      " |      Typically, this method directly controls `tf.function` and\n",
      " |      `tf.distribute.Strategy` settings, and delegates the actual evaluation\n",
      " |      logic to `Model.predict_step`.\n",
      " |      \n",
      " |      This function is cached the first time `Model.predict` or\n",
      " |      `Model.predict_on_batch` is called. The cache is cleared whenever\n",
      " |      `Model.compile` is called. You can skip the cache and generate again the\n",
      " |      function with `force=True`.\n",
      " |      \n",
      " |      Args:\n",
      " |        force: Whether to regenerate the predict function and skip the cached\n",
      " |          function if available.\n",
      " |      \n",
      " |      Returns:\n",
      " |        Function. The function created by this method should accept a\n",
      " |        `tf.data.Iterator`, and return the outputs of the `Model`.\n",
      " |  \n",
      " |  make_test_function(self, force=False)\n",
      " |      Creates a function that executes one step of evaluation.\n",
      " |      \n",
      " |      This method can be overridden to support custom evaluation logic.\n",
      " |      This method is called by `Model.evaluate` and `Model.test_on_batch`.\n",
      " |      \n",
      " |      Typically, this method directly controls `tf.function` and\n",
      " |      `tf.distribute.Strategy` settings, and delegates the actual evaluation\n",
      " |      logic to `Model.test_step`.\n",
      " |      \n",
      " |      This function is cached the first time `Model.evaluate` or\n",
      " |      `Model.test_on_batch` is called. The cache is cleared whenever\n",
      " |      `Model.compile` is called. You can skip the cache and generate again the\n",
      " |      function with `force=True`.\n",
      " |      \n",
      " |      Args:\n",
      " |        force: Whether to regenerate the test function and skip the cached\n",
      " |          function if available.\n",
      " |      \n",
      " |      Returns:\n",
      " |        Function. The function created by this method should accept a\n",
      " |        `tf.data.Iterator`, and return a `dict` containing values that will\n",
      " |        be passed to `tf.keras.Callbacks.on_test_batch_end`.\n",
      " |  \n",
      " |  make_train_function(self, force=False)\n",
      " |      Creates a function that executes one step of training.\n",
      " |      \n",
      " |      This method can be overridden to support custom training logic.\n",
      " |      This method is called by `Model.fit` and `Model.train_on_batch`.\n",
      " |      \n",
      " |      Typically, this method directly controls `tf.function` and\n",
      " |      `tf.distribute.Strategy` settings, and delegates the actual training\n",
      " |      logic to `Model.train_step`.\n",
      " |      \n",
      " |      This function is cached the first time `Model.fit` or\n",
      " |      `Model.train_on_batch` is called. The cache is cleared whenever\n",
      " |      `Model.compile` is called. You can skip the cache and generate again the\n",
      " |      function with `force=True`.\n",
      " |      \n",
      " |      Args:\n",
      " |        force: Whether to regenerate the train function and skip the cached\n",
      " |          function if available.\n",
      " |      \n",
      " |      Returns:\n",
      " |        Function. The function created by this method should accept a\n",
      " |        `tf.data.Iterator`, and return a `dict` containing values that will\n",
      " |        be passed to `tf.keras.Callbacks.on_train_batch_end`, such as\n",
      " |        `{'loss': 0.2, 'accuracy': 0.7}`.\n",
      " |  \n",
      " |  predict(self, x, batch_size=None, verbose='auto', steps=None, callbacks=None, max_queue_size=10, workers=1, use_multiprocessing=False)\n",
      " |      Generates output predictions for the input samples.\n",
      " |      \n",
      " |      Computation is done in batches. This method is designed for batch\n",
      " |      processing of large numbers of inputs. It is not intended for use inside\n",
      " |      of loops that iterate over your data and process small numbers of inputs\n",
      " |      at a time.\n",
      " |      \n",
      " |      For small numbers of inputs that fit in one batch,\n",
      " |      directly use `__call__()` for faster execution, e.g.,\n",
      " |      `model(x)`, or `model(x, training=False)` if you have layers such as\n",
      " |      `tf.keras.layers.BatchNormalization` that behave differently during\n",
      " |      inference. You may pair the individual model call with a `tf.function`\n",
      " |      for additional performance inside your inner loop.\n",
      " |      If you need access to numpy array values instead of tensors after your\n",
      " |      model call, you can use `tensor.numpy()` to get the numpy array value of\n",
      " |      an eager tensor.\n",
      " |      \n",
      " |      Also, note the fact that test loss is not affected by\n",
      " |      regularization layers like noise and dropout.\n",
      " |      \n",
      " |      Note: See [this FAQ entry](\n",
      " |      https://keras.io/getting_started/faq/#whats-the-difference-between-model-methods-predict-and-call)\n",
      " |      for more details about the difference between `Model` methods\n",
      " |      `predict()` and `__call__()`.\n",
      " |      \n",
      " |      Args:\n",
      " |          x: Input samples. It could be:\n",
      " |            - A Numpy array (or array-like), or a list of arrays\n",
      " |              (in case the model has multiple inputs).\n",
      " |            - A TensorFlow tensor, or a list of tensors\n",
      " |              (in case the model has multiple inputs).\n",
      " |            - A `tf.data` dataset.\n",
      " |            - A generator or `keras.utils.Sequence` instance.\n",
      " |            A more detailed description of unpacking behavior for iterator\n",
      " |            types (Dataset, generator, Sequence) is given in the `Unpacking\n",
      " |            behavior for iterator-like inputs` section of `Model.fit`.\n",
      " |          batch_size: Integer or `None`.\n",
      " |              Number of samples per batch.\n",
      " |              If unspecified, `batch_size` will default to 32.\n",
      " |              Do not specify the `batch_size` if your data is in the\n",
      " |              form of dataset, generators, or `keras.utils.Sequence` instances\n",
      " |              (since they generate batches).\n",
      " |          verbose: `\"auto\"`, 0, 1, or 2. Verbosity mode.\n",
      " |              0 = silent, 1 = progress bar, 2 = single line.\n",
      " |              `\"auto\"` defaults to 1 for most cases, and to 2 when used with\n",
      " |              `ParameterServerStrategy`. Note that the progress bar is not\n",
      " |              particularly useful when logged to a file, so `verbose=2` is\n",
      " |              recommended when not running interactively (e.g. in a production\n",
      " |              environment).\n",
      " |          steps: Total number of steps (batches of samples)\n",
      " |              before declaring the prediction round finished.\n",
      " |              Ignored with the default value of `None`. If x is a `tf.data`\n",
      " |              dataset and `steps` is None, `predict()` will\n",
      " |              run until the input dataset is exhausted.\n",
      " |          callbacks: List of `keras.callbacks.Callback` instances.\n",
      " |              List of callbacks to apply during prediction.\n",
      " |              See [callbacks](\n",
      " |              https://www.tensorflow.org/api_docs/python/tf/keras/callbacks).\n",
      " |          max_queue_size: Integer. Used for generator or\n",
      " |              `keras.utils.Sequence` input only. Maximum size for the\n",
      " |              generator queue. If unspecified, `max_queue_size` will default\n",
      " |              to 10.\n",
      " |          workers: Integer. Used for generator or `keras.utils.Sequence` input\n",
      " |              only. Maximum number of processes to spin up when using\n",
      " |              process-based threading. If unspecified, `workers` will default\n",
      " |              to 1.\n",
      " |          use_multiprocessing: Boolean. Used for generator or\n",
      " |              `keras.utils.Sequence` input only. If `True`, use process-based\n",
      " |              threading. If unspecified, `use_multiprocessing` will default to\n",
      " |              `False`. Note that because this implementation relies on\n",
      " |              multiprocessing, you should not pass non-picklable arguments to\n",
      " |              the generator as they can't be passed easily to children\n",
      " |              processes.\n",
      " |      \n",
      " |      See the discussion of `Unpacking behavior for iterator-like inputs` for\n",
      " |      `Model.fit`. Note that Model.predict uses the same interpretation rules\n",
      " |      as `Model.fit` and `Model.evaluate`, so inputs must be unambiguous for\n",
      " |      all three methods.\n",
      " |      \n",
      " |      Returns:\n",
      " |          Numpy array(s) of predictions.\n",
      " |      \n",
      " |      Raises:\n",
      " |          RuntimeError: If `model.predict` is wrapped in a `tf.function`.\n",
      " |          ValueError: In case of mismatch between the provided\n",
      " |              input data and the model's expectations,\n",
      " |              or in case a stateful model receives a number of samples\n",
      " |              that is not a multiple of the batch size.\n",
      " |  \n",
      " |  predict_generator(self, generator, steps=None, callbacks=None, max_queue_size=10, workers=1, use_multiprocessing=False, verbose=0)\n",
      " |      Generates predictions for the input samples from a data generator.\n",
      " |      \n",
      " |      DEPRECATED:\n",
      " |        `Model.predict` now supports generators, so there is no longer any\n",
      " |        need to use this endpoint.\n",
      " |  \n",
      " |  predict_on_batch(self, x)\n",
      " |      Returns predictions for a single batch of samples.\n",
      " |      \n",
      " |      Args:\n",
      " |          x: Input data. It could be:\n",
      " |            - A Numpy array (or array-like), or a list of arrays (in case the\n",
      " |                model has multiple inputs).\n",
      " |            - A TensorFlow tensor, or a list of tensors (in case the model has\n",
      " |                multiple inputs).\n",
      " |      \n",
      " |      Returns:\n",
      " |          Numpy array(s) of predictions.\n",
      " |      \n",
      " |      Raises:\n",
      " |          RuntimeError: If `model.predict_on_batch` is wrapped in a\n",
      " |            `tf.function`.\n",
      " |  \n",
      " |  predict_step(self, data)\n",
      " |      The logic for one inference step.\n",
      " |      \n",
      " |      This method can be overridden to support custom inference logic.\n",
      " |      This method is called by `Model.make_predict_function`.\n",
      " |      \n",
      " |      This method should contain the mathematical logic for one step of\n",
      " |      inference.  This typically includes the forward pass.\n",
      " |      \n",
      " |      Configuration details for *how* this logic is run (e.g. `tf.function`\n",
      " |      and `tf.distribute.Strategy` settings), should be left to\n",
      " |      `Model.make_predict_function`, which can also be overridden.\n",
      " |      \n",
      " |      Args:\n",
      " |        data: A nested structure of `Tensor`s.\n",
      " |      \n",
      " |      Returns:\n",
      " |        The result of one inference step, typically the output of calling the\n",
      " |        `Model` on data.\n",
      " |  \n",
      " |  reset_metrics(self)\n",
      " |      Resets the state of all the metrics in the model.\n",
      " |      \n",
      " |      Examples:\n",
      " |      \n",
      " |      >>> inputs = tf.keras.layers.Input(shape=(3,))\n",
      " |      >>> outputs = tf.keras.layers.Dense(2)(inputs)\n",
      " |      >>> model = tf.keras.models.Model(inputs=inputs, outputs=outputs)\n",
      " |      >>> model.compile(optimizer=\"Adam\", loss=\"mse\", metrics=[\"mae\"])\n",
      " |      \n",
      " |      >>> x = np.random.random((2, 3))\n",
      " |      >>> y = np.random.randint(0, 2, (2, 2))\n",
      " |      >>> _ = model.fit(x, y, verbose=0)\n",
      " |      >>> assert all(float(m.result()) for m in model.metrics)\n",
      " |      \n",
      " |      >>> model.reset_metrics()\n",
      " |      >>> assert all(float(m.result()) == 0 for m in model.metrics)\n",
      " |  \n",
      " |  reset_states(self)\n",
      " |  \n",
      " |  save(self, filepath, overwrite=True, include_optimizer=True, save_format=None, signatures=None, options=None, save_traces=True)\n",
      " |      Saves the model to Tensorflow SavedModel or a single HDF5 file.\n",
      " |      \n",
      " |      Please see `tf.keras.models.save_model` or the\n",
      " |      [Serialization and Saving guide](\n",
      " |      https://keras.io/guides/serialization_and_saving/)\n",
      " |      for details.\n",
      " |      \n",
      " |      Args:\n",
      " |          filepath: String, PathLike, path to SavedModel or H5 file to save\n",
      " |              the model.\n",
      " |          overwrite: Whether to silently overwrite any existing file at the\n",
      " |              target location, or provide the user with a manual prompt.\n",
      " |          include_optimizer: If True, save optimizer's state together.\n",
      " |          save_format: Either `'tf'` or `'h5'`, indicating whether to save the\n",
      " |              model to Tensorflow SavedModel or HDF5. Defaults to 'tf' in TF\n",
      " |              2.X, and 'h5' in TF 1.X.\n",
      " |          signatures: Signatures to save with the SavedModel. Applicable to\n",
      " |              the 'tf' format only. Please see the `signatures` argument in\n",
      " |              `tf.saved_model.save` for details.\n",
      " |          options: (only applies to SavedModel format)\n",
      " |              `tf.saved_model.SaveOptions` object that specifies options for\n",
      " |              saving to SavedModel.\n",
      " |          save_traces: (only applies to SavedModel format) When enabled, the\n",
      " |              SavedModel will store the function traces for each layer. This\n",
      " |              can be disabled, so that only the configs of each layer are\n",
      " |              stored.  Defaults to `True`. Disabling this will decrease\n",
      " |              serialization time and reduce file size, but it requires that\n",
      " |              all custom layers/models implement a `get_config()` method.\n",
      " |      \n",
      " |      Example:\n",
      " |      \n",
      " |      ```python\n",
      " |      from keras.models import load_model\n",
      " |      \n",
      " |      model.save('my_model.h5')  # creates a HDF5 file 'my_model.h5'\n",
      " |      del model  # deletes the existing model\n",
      " |      \n",
      " |      # returns a compiled model\n",
      " |      # identical to the previous one\n",
      " |      model = load_model('my_model.h5')\n",
      " |      ```\n",
      " |  \n",
      " |  save_spec(self, dynamic_batch=True)\n",
      " |      Returns the `tf.TensorSpec` of call inputs as a tuple `(args, kwargs)`.\n",
      " |      \n",
      " |      This value is automatically defined after calling the model for the\n",
      " |      first time. Afterwards, you can use it when exporting the model for\n",
      " |      serving:\n",
      " |      \n",
      " |      ```python\n",
      " |      model = tf.keras.Model(...)\n",
      " |      \n",
      " |      @tf.function\n",
      " |      def serve(*args, **kwargs):\n",
      " |        outputs = model(*args, **kwargs)\n",
      " |        # Apply postprocessing steps, or add additional outputs.\n",
      " |        ...\n",
      " |        return outputs\n",
      " |      \n",
      " |      # arg_specs is `[tf.TensorSpec(...), ...]`. kwarg_specs, in this\n",
      " |      # example, is an empty dict since functional models do not use keyword\n",
      " |      # arguments.\n",
      " |      arg_specs, kwarg_specs = model.save_spec()\n",
      " |      \n",
      " |      model.save(path, signatures={\n",
      " |        'serving_default': serve.get_concrete_function(*arg_specs,\n",
      " |                                                       **kwarg_specs)\n",
      " |      })\n",
      " |      ```\n",
      " |      \n",
      " |      Args:\n",
      " |        dynamic_batch: Whether to set the batch sizes of all the returned\n",
      " |          `tf.TensorSpec` to `None`. (Note that when defining functional or\n",
      " |          Sequential models with `tf.keras.Input([...], batch_size=X)`, the\n",
      " |          batch size will always be preserved). Defaults to `True`.\n",
      " |      Returns:\n",
      " |        If the model inputs are defined, returns a tuple `(args, kwargs)`. All\n",
      " |        elements in `args` and `kwargs` are `tf.TensorSpec`.\n",
      " |        If the model inputs are not defined, returns `None`.\n",
      " |        The model inputs are automatically set when calling the model,\n",
      " |        `model.fit`, `model.evaluate` or `model.predict`.\n",
      " |  \n",
      " |  save_weights(self, filepath, overwrite=True, save_format=None, options=None)\n",
      " |      Saves all layer weights.\n",
      " |      \n",
      " |      Either saves in HDF5 or in TensorFlow format based on the `save_format`\n",
      " |      argument.\n",
      " |      \n",
      " |      When saving in HDF5 format, the weight file has:\n",
      " |        - `layer_names` (attribute), a list of strings\n",
      " |            (ordered names of model layers).\n",
      " |        - For every layer, a `group` named `layer.name`\n",
      " |            - For every such layer group, a group attribute `weight_names`,\n",
      " |                a list of strings\n",
      " |                (ordered names of weights tensor of the layer).\n",
      " |            - For every weight in the layer, a dataset\n",
      " |                storing the weight value, named after the weight tensor.\n",
      " |      \n",
      " |      When saving in TensorFlow format, all objects referenced by the network\n",
      " |      are saved in the same format as `tf.train.Checkpoint`, including any\n",
      " |      `Layer` instances or `Optimizer` instances assigned to object\n",
      " |      attributes. For networks constructed from inputs and outputs using\n",
      " |      `tf.keras.Model(inputs, outputs)`, `Layer` instances used by the network\n",
      " |      are tracked/saved automatically. For user-defined classes which inherit\n",
      " |      from `tf.keras.Model`, `Layer` instances must be assigned to object\n",
      " |      attributes, typically in the constructor. See the documentation of\n",
      " |      `tf.train.Checkpoint` and `tf.keras.Model` for details.\n",
      " |      \n",
      " |      While the formats are the same, do not mix `save_weights` and\n",
      " |      `tf.train.Checkpoint`. Checkpoints saved by `Model.save_weights` should\n",
      " |      be loaded using `Model.load_weights`. Checkpoints saved using\n",
      " |      `tf.train.Checkpoint.save` should be restored using the corresponding\n",
      " |      `tf.train.Checkpoint.restore`. Prefer `tf.train.Checkpoint` over\n",
      " |      `save_weights` for training checkpoints.\n",
      " |      \n",
      " |      The TensorFlow format matches objects and variables by starting at a\n",
      " |      root object, `self` for `save_weights`, and greedily matching attribute\n",
      " |      names. For `Model.save` this is the `Model`, and for `Checkpoint.save`\n",
      " |      this is the `Checkpoint` even if the `Checkpoint` has a model attached.\n",
      " |      This means saving a `tf.keras.Model` using `save_weights` and loading\n",
      " |      into a `tf.train.Checkpoint` with a `Model` attached (or vice versa)\n",
      " |      will not match the `Model`'s variables. See the\n",
      " |      [guide to training checkpoints](\n",
      " |      https://www.tensorflow.org/guide/checkpoint) for details on\n",
      " |      the TensorFlow format.\n",
      " |      \n",
      " |      Args:\n",
      " |          filepath: String or PathLike, path to the file to save the weights\n",
      " |              to. When saving in TensorFlow format, this is the prefix used\n",
      " |              for checkpoint files (multiple files are generated). Note that\n",
      " |              the '.h5' suffix causes weights to be saved in HDF5 format.\n",
      " |          overwrite: Whether to silently overwrite any existing file at the\n",
      " |              target location, or provide the user with a manual prompt.\n",
      " |          save_format: Either 'tf' or 'h5'. A `filepath` ending in '.h5' or\n",
      " |              '.keras' will default to HDF5 if `save_format` is `None`.\n",
      " |              Otherwise `None` defaults to 'tf'.\n",
      " |          options: Optional `tf.train.CheckpointOptions` object that specifies\n",
      " |              options for saving weights.\n",
      " |      \n",
      " |      Raises:\n",
      " |          ImportError: If `h5py` is not available when attempting to save in\n",
      " |              HDF5 format.\n",
      " |  \n",
      " |  summary(self, line_length=None, positions=None, print_fn=None, expand_nested=False, show_trainable=False, layer_range=None)\n",
      " |      Prints a string summary of the network.\n",
      " |      \n",
      " |      Args:\n",
      " |          line_length: Total length of printed lines\n",
      " |              (e.g. set this to adapt the display to different\n",
      " |              terminal window sizes).\n",
      " |          positions: Relative or absolute positions of log elements\n",
      " |              in each line. If not provided,\n",
      " |              defaults to `[.33, .55, .67, 1.]`.\n",
      " |          print_fn: Print function to use. Defaults to `print`.\n",
      " |              It will be called on each line of the summary.\n",
      " |              You can set it to a custom function\n",
      " |              in order to capture the string summary.\n",
      " |          expand_nested: Whether to expand the nested models.\n",
      " |              If not provided, defaults to `False`.\n",
      " |          show_trainable: Whether to show if a layer is trainable.\n",
      " |              If not provided, defaults to `False`.\n",
      " |          layer_range: a list or tuple of 2 strings,\n",
      " |              which is the starting layer name and ending layer name\n",
      " |              (both inclusive) indicating the range of layers to be printed\n",
      " |              in summary. It also accepts regex patterns instead of exact\n",
      " |              name. In such case, start predicate will be the first element\n",
      " |              it matches to `layer_range[0]` and the end predicate will be\n",
      " |              the last element it matches to `layer_range[1]`.\n",
      " |              By default `None` which considers all layers of model.\n",
      " |      \n",
      " |      Raises:\n",
      " |          ValueError: if `summary()` is called before the model is built.\n",
      " |  \n",
      " |  test_on_batch(self, x, y=None, sample_weight=None, reset_metrics=True, return_dict=False)\n",
      " |      Test the model on a single batch of samples.\n",
      " |      \n",
      " |      Args:\n",
      " |          x: Input data. It could be:\n",
      " |            - A Numpy array (or array-like), or a list of arrays (in case the\n",
      " |                model has multiple inputs).\n",
      " |            - A TensorFlow tensor, or a list of tensors (in case the model has\n",
      " |                multiple inputs).\n",
      " |            - A dict mapping input names to the corresponding array/tensors,\n",
      " |                if the model has named inputs.\n",
      " |          y: Target data. Like the input data `x`, it could be either Numpy\n",
      " |            array(s) or TensorFlow tensor(s). It should be consistent with `x`\n",
      " |            (you cannot have Numpy inputs and tensor targets, or inversely).\n",
      " |          sample_weight: Optional array of the same length as x, containing\n",
      " |            weights to apply to the model's loss for each sample. In the case\n",
      " |            of temporal data, you can pass a 2D array with shape (samples,\n",
      " |            sequence_length), to apply a different weight to every timestep of\n",
      " |            every sample.\n",
      " |          reset_metrics: If `True`, the metrics returned will be only for this\n",
      " |            batch. If `False`, the metrics will be statefully accumulated\n",
      " |            across batches.\n",
      " |          return_dict: If `True`, loss and metric results are returned as a\n",
      " |            dict, with each key being the name of the metric. If `False`, they\n",
      " |            are returned as a list.\n",
      " |      \n",
      " |      Returns:\n",
      " |          Scalar test loss (if the model has a single output and no metrics)\n",
      " |          or list of scalars (if the model has multiple outputs\n",
      " |          and/or metrics). The attribute `model.metrics_names` will give you\n",
      " |          the display labels for the scalar outputs.\n",
      " |      \n",
      " |      Raises:\n",
      " |          RuntimeError: If `model.test_on_batch` is wrapped in a\n",
      " |            `tf.function`.\n",
      " |  \n",
      " |  test_step(self, data)\n",
      " |      The logic for one evaluation step.\n",
      " |      \n",
      " |      This method can be overridden to support custom evaluation logic.\n",
      " |      This method is called by `Model.make_test_function`.\n",
      " |      \n",
      " |      This function should contain the mathematical logic for one step of\n",
      " |      evaluation.\n",
      " |      This typically includes the forward pass, loss calculation, and metrics\n",
      " |      updates.\n",
      " |      \n",
      " |      Configuration details for *how* this logic is run (e.g. `tf.function`\n",
      " |      and `tf.distribute.Strategy` settings), should be left to\n",
      " |      `Model.make_test_function`, which can also be overridden.\n",
      " |      \n",
      " |      Args:\n",
      " |        data: A nested structure of `Tensor`s.\n",
      " |      \n",
      " |      Returns:\n",
      " |        A `dict` containing values that will be passed to\n",
      " |        `tf.keras.callbacks.CallbackList.on_train_batch_end`. Typically, the\n",
      " |        values of the `Model`'s metrics are returned.\n",
      " |  \n",
      " |  to_json(self, **kwargs)\n",
      " |      Returns a JSON string containing the network configuration.\n",
      " |      \n",
      " |      To load a network from a JSON save file, use\n",
      " |      `keras.models.model_from_json(json_string, custom_objects={})`.\n",
      " |      \n",
      " |      Args:\n",
      " |          **kwargs: Additional keyword arguments to be passed to\n",
      " |              *`json.dumps()`.\n",
      " |      \n",
      " |      Returns:\n",
      " |          A JSON string.\n",
      " |  \n",
      " |  to_yaml(self, **kwargs)\n",
      " |      Returns a yaml string containing the network configuration.\n",
      " |      \n",
      " |      Note: Since TF 2.6, this method is no longer supported and will raise a\n",
      " |      RuntimeError.\n",
      " |      \n",
      " |      To load a network from a yaml save file, use\n",
      " |      `keras.models.model_from_yaml(yaml_string, custom_objects={})`.\n",
      " |      \n",
      " |      `custom_objects` should be a dictionary mapping\n",
      " |      the names of custom losses / layers / etc to the corresponding\n",
      " |      functions / classes.\n",
      " |      \n",
      " |      Args:\n",
      " |          **kwargs: Additional keyword arguments\n",
      " |              to be passed to `yaml.dump()`.\n",
      " |      \n",
      " |      Returns:\n",
      " |          A YAML string.\n",
      " |      \n",
      " |      Raises:\n",
      " |          RuntimeError: announces that the method poses a security risk\n",
      " |  \n",
      " |  train_on_batch(self, x, y=None, sample_weight=None, class_weight=None, reset_metrics=True, return_dict=False)\n",
      " |      Runs a single gradient update on a single batch of data.\n",
      " |      \n",
      " |      Args:\n",
      " |          x: Input data. It could be:\n",
      " |            - A Numpy array (or array-like), or a list of arrays\n",
      " |                (in case the model has multiple inputs).\n",
      " |            - A TensorFlow tensor, or a list of tensors\n",
      " |                (in case the model has multiple inputs).\n",
      " |            - A dict mapping input names to the corresponding array/tensors,\n",
      " |                if the model has named inputs.\n",
      " |          y: Target data. Like the input data `x`, it could be either Numpy\n",
      " |            array(s) or TensorFlow tensor(s).\n",
      " |          sample_weight: Optional array of the same length as x, containing\n",
      " |            weights to apply to the model's loss for each sample. In the case\n",
      " |            of temporal data, you can pass a 2D array with shape (samples,\n",
      " |            sequence_length), to apply a different weight to every timestep of\n",
      " |            every sample.\n",
      " |          class_weight: Optional dictionary mapping class indices (integers)\n",
      " |            to a weight (float) to apply to the model's loss for the samples\n",
      " |            from this class during training. This can be useful to tell the\n",
      " |            model to \"pay more attention\" to samples from an under-represented\n",
      " |            class.\n",
      " |          reset_metrics: If `True`, the metrics returned will be only for this\n",
      " |            batch. If `False`, the metrics will be statefully accumulated\n",
      " |            across batches.\n",
      " |          return_dict: If `True`, loss and metric results are returned as a\n",
      " |            dict, with each key being the name of the metric. If `False`, they\n",
      " |            are returned as a list.\n",
      " |      \n",
      " |      Returns:\n",
      " |          Scalar training loss\n",
      " |          (if the model has a single output and no metrics)\n",
      " |          or list of scalars (if the model has multiple outputs\n",
      " |          and/or metrics). The attribute `model.metrics_names` will give you\n",
      " |          the display labels for the scalar outputs.\n",
      " |      \n",
      " |      Raises:\n",
      " |        RuntimeError: If `model.train_on_batch` is wrapped in a `tf.function`.\n",
      " |  \n",
      " |  train_step(self, data)\n",
      " |      The logic for one training step.\n",
      " |      \n",
      " |      This method can be overridden to support custom training logic.\n",
      " |      For concrete examples of how to override this method see\n",
      " |      [Customizing what happens in fit](\n",
      " |      https://www.tensorflow.org/guide/keras/customizing_what_happens_in_fit).\n",
      " |      This method is called by `Model.make_train_function`.\n",
      " |      \n",
      " |      This method should contain the mathematical logic for one step of\n",
      " |      training.  This typically includes the forward pass, loss calculation,\n",
      " |      backpropagation, and metric updates.\n",
      " |      \n",
      " |      Configuration details for *how* this logic is run (e.g. `tf.function`\n",
      " |      and `tf.distribute.Strategy` settings), should be left to\n",
      " |      `Model.make_train_function`, which can also be overridden.\n",
      " |      \n",
      " |      Args:\n",
      " |        data: A nested structure of `Tensor`s.\n",
      " |      \n",
      " |      Returns:\n",
      " |        A `dict` containing values that will be passed to\n",
      " |        `tf.keras.callbacks.CallbackList.on_train_batch_end`. Typically, the\n",
      " |        values of the `Model`'s metrics are returned. Example:\n",
      " |        `{'loss': 0.2, 'accuracy': 0.7}`.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Static methods inherited from keras.engine.training.Model:\n",
      " |  \n",
      " |  __new__(cls, *args, **kwargs)\n",
      " |      Create and return a new object.  See help(type) for accurate signature.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Readonly properties inherited from keras.engine.training.Model:\n",
      " |  \n",
      " |  distribute_strategy\n",
      " |      The `tf.distribute.Strategy` this model was created under.\n",
      " |  \n",
      " |  metrics\n",
      " |      Returns the model's metrics added using `compile()`, `add_metric()` APIs.\n",
      " |      \n",
      " |      Note: Metrics passed to `compile()` are available only after a\n",
      " |      `keras.Model` has been trained/evaluated on actual data.\n",
      " |      \n",
      " |      Examples:\n",
      " |      \n",
      " |      >>> inputs = tf.keras.layers.Input(shape=(3,))\n",
      " |      >>> outputs = tf.keras.layers.Dense(2)(inputs)\n",
      " |      >>> model = tf.keras.models.Model(inputs=inputs, outputs=outputs)\n",
      " |      >>> model.compile(optimizer=\"Adam\", loss=\"mse\", metrics=[\"mae\"])\n",
      " |      >>> [m.name for m in model.metrics]\n",
      " |      []\n",
      " |      \n",
      " |      >>> x = np.random.random((2, 3))\n",
      " |      >>> y = np.random.randint(0, 2, (2, 2))\n",
      " |      >>> model.fit(x, y)\n",
      " |      >>> [m.name for m in model.metrics]\n",
      " |      ['loss', 'mae']\n",
      " |      \n",
      " |      >>> inputs = tf.keras.layers.Input(shape=(3,))\n",
      " |      >>> d = tf.keras.layers.Dense(2, name='out')\n",
      " |      >>> output_1 = d(inputs)\n",
      " |      >>> output_2 = d(inputs)\n",
      " |      >>> model = tf.keras.models.Model(\n",
      " |      ...    inputs=inputs, outputs=[output_1, output_2])\n",
      " |      >>> model.add_metric(\n",
      " |      ...    tf.reduce_sum(output_2), name='mean', aggregation='mean')\n",
      " |      >>> model.compile(optimizer=\"Adam\", loss=\"mse\", metrics=[\"mae\", \"acc\"])\n",
      " |      >>> model.fit(x, (y, y))\n",
      " |      >>> [m.name for m in model.metrics]\n",
      " |      ['loss', 'out_loss', 'out_1_loss', 'out_mae', 'out_acc', 'out_1_mae',\n",
      " |      'out_1_acc', 'mean']\n",
      " |  \n",
      " |  metrics_names\n",
      " |      Returns the model's display labels for all outputs.\n",
      " |      \n",
      " |      Note: `metrics_names` are available only after a `keras.Model` has been\n",
      " |      trained/evaluated on actual data.\n",
      " |      \n",
      " |      Examples:\n",
      " |      \n",
      " |      >>> inputs = tf.keras.layers.Input(shape=(3,))\n",
      " |      >>> outputs = tf.keras.layers.Dense(2)(inputs)\n",
      " |      >>> model = tf.keras.models.Model(inputs=inputs, outputs=outputs)\n",
      " |      >>> model.compile(optimizer=\"Adam\", loss=\"mse\", metrics=[\"mae\"])\n",
      " |      >>> model.metrics_names\n",
      " |      []\n",
      " |      \n",
      " |      >>> x = np.random.random((2, 3))\n",
      " |      >>> y = np.random.randint(0, 2, (2, 2))\n",
      " |      >>> model.fit(x, y)\n",
      " |      >>> model.metrics_names\n",
      " |      ['loss', 'mae']\n",
      " |      \n",
      " |      >>> inputs = tf.keras.layers.Input(shape=(3,))\n",
      " |      >>> d = tf.keras.layers.Dense(2, name='out')\n",
      " |      >>> output_1 = d(inputs)\n",
      " |      >>> output_2 = d(inputs)\n",
      " |      >>> model = tf.keras.models.Model(\n",
      " |      ...    inputs=inputs, outputs=[output_1, output_2])\n",
      " |      >>> model.compile(optimizer=\"Adam\", loss=\"mse\", metrics=[\"mae\", \"acc\"])\n",
      " |      >>> model.fit(x, (y, y))\n",
      " |      >>> model.metrics_names\n",
      " |      ['loss', 'out_loss', 'out_1_loss', 'out_mae', 'out_acc', 'out_1_mae',\n",
      " |      'out_1_acc']\n",
      " |  \n",
      " |  non_trainable_weights\n",
      " |      List of all non-trainable weights tracked by this layer.\n",
      " |      \n",
      " |      Non-trainable weights are *not* updated during training. They are\n",
      " |      expected to be updated manually in `call()`.\n",
      " |      \n",
      " |      Returns:\n",
      " |        A list of non-trainable variables.\n",
      " |  \n",
      " |  state_updates\n",
      " |      Deprecated, do NOT use!\n",
      " |      \n",
      " |      Returns the `updates` from all layers that are stateful.\n",
      " |      \n",
      " |      This is useful for separating training updates and\n",
      " |      state updates, e.g. when we need to update a layer's internal state\n",
      " |      during prediction.\n",
      " |      \n",
      " |      Returns:\n",
      " |          A list of update ops.\n",
      " |  \n",
      " |  trainable_weights\n",
      " |      List of all trainable weights tracked by this layer.\n",
      " |      \n",
      " |      Trainable weights are updated via gradient descent during training.\n",
      " |      \n",
      " |      Returns:\n",
      " |        A list of trainable variables.\n",
      " |  \n",
      " |  weights\n",
      " |      Returns the list of all layer variables/weights.\n",
      " |      \n",
      " |      Note: This will not track the weights of nested `tf.Modules` that are\n",
      " |      not themselves Keras layers.\n",
      " |      \n",
      " |      Returns:\n",
      " |        A list of variables.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Data descriptors inherited from keras.engine.training.Model:\n",
      " |  \n",
      " |  distribute_reduction_method\n",
      " |      The method employed to reduce per-replica values during training.\n",
      " |      \n",
      " |      Unless specified, the value \"auto\" will be assumed, indicating that\n",
      " |      the reduction strategy should be chosen based on the current\n",
      " |      running environment.\n",
      " |      See `reduce_per_replica` function for more details.\n",
      " |  \n",
      " |  run_eagerly\n",
      " |      Settable attribute indicating whether the model should run eagerly.\n",
      " |      \n",
      " |      Running eagerly means that your model will be run step by step,\n",
      " |      like Python code. Your model might run slower, but it should become\n",
      " |      easier for you to debug it by stepping into individual layer calls.\n",
      " |      \n",
      " |      By default, we will attempt to compile your model to a static graph to\n",
      " |      deliver the best execution performance.\n",
      " |      \n",
      " |      Returns:\n",
      " |        Boolean, whether the model should run eagerly.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Methods inherited from keras.engine.base_layer.Layer:\n",
      " |  \n",
      " |  __delattr__(self, name)\n",
      " |      Implement delattr(self, name).\n",
      " |  \n",
      " |  __getstate__(self)\n",
      " |  \n",
      " |  __setstate__(self, state)\n",
      " |  \n",
      " |  add_loss(self, losses, **kwargs)\n",
      " |      Add loss tensor(s), potentially dependent on layer inputs.\n",
      " |      \n",
      " |      Some losses (for instance, activity regularization losses) may be\n",
      " |      dependent on the inputs passed when calling a layer. Hence, when reusing\n",
      " |      the same layer on different inputs `a` and `b`, some entries in\n",
      " |      `layer.losses` may be dependent on `a` and some on `b`. This method\n",
      " |      automatically keeps track of dependencies.\n",
      " |      \n",
      " |      This method can be used inside a subclassed layer or model's `call`\n",
      " |      function, in which case `losses` should be a Tensor or list of Tensors.\n",
      " |      \n",
      " |      Example:\n",
      " |      \n",
      " |      ```python\n",
      " |      class MyLayer(tf.keras.layers.Layer):\n",
      " |        def call(self, inputs):\n",
      " |          self.add_loss(tf.abs(tf.reduce_mean(inputs)))\n",
      " |          return inputs\n",
      " |      ```\n",
      " |      \n",
      " |      This method can also be called directly on a Functional Model during\n",
      " |      construction. In this case, any loss Tensors passed to this Model must\n",
      " |      be symbolic and be able to be traced back to the model's `Input`s. These\n",
      " |      losses become part of the model's topology and are tracked in\n",
      " |      `get_config`.\n",
      " |      \n",
      " |      Example:\n",
      " |      \n",
      " |      ```python\n",
      " |      inputs = tf.keras.Input(shape=(10,))\n",
      " |      x = tf.keras.layers.Dense(10)(inputs)\n",
      " |      outputs = tf.keras.layers.Dense(1)(x)\n",
      " |      model = tf.keras.Model(inputs, outputs)\n",
      " |      # Activity regularization.\n",
      " |      model.add_loss(tf.abs(tf.reduce_mean(x)))\n",
      " |      ```\n",
      " |      \n",
      " |      If this is not the case for your loss (if, for example, your loss\n",
      " |      references a `Variable` of one of the model's layers), you can wrap your\n",
      " |      loss in a zero-argument lambda. These losses are not tracked as part of\n",
      " |      the model's topology since they can't be serialized.\n",
      " |      \n",
      " |      Example:\n",
      " |      \n",
      " |      ```python\n",
      " |      inputs = tf.keras.Input(shape=(10,))\n",
      " |      d = tf.keras.layers.Dense(10)\n",
      " |      x = d(inputs)\n",
      " |      outputs = tf.keras.layers.Dense(1)(x)\n",
      " |      model = tf.keras.Model(inputs, outputs)\n",
      " |      # Weight regularization.\n",
      " |      model.add_loss(lambda: tf.reduce_mean(d.kernel))\n",
      " |      ```\n",
      " |      \n",
      " |      Args:\n",
      " |        losses: Loss tensor, or list/tuple of tensors. Rather than tensors,\n",
      " |          losses may also be zero-argument callables which create a loss\n",
      " |          tensor.\n",
      " |        **kwargs: Used for backwards compatibility only.\n",
      " |  \n",
      " |  add_metric(self, value, name=None, **kwargs)\n",
      " |      Adds metric tensor to the layer.\n",
      " |      \n",
      " |      This method can be used inside the `call()` method of a subclassed layer\n",
      " |      or model.\n",
      " |      \n",
      " |      ```python\n",
      " |      class MyMetricLayer(tf.keras.layers.Layer):\n",
      " |        def __init__(self):\n",
      " |          super(MyMetricLayer, self).__init__(name='my_metric_layer')\n",
      " |          self.mean = tf.keras.metrics.Mean(name='metric_1')\n",
      " |      \n",
      " |        def call(self, inputs):\n",
      " |          self.add_metric(self.mean(inputs))\n",
      " |          self.add_metric(tf.reduce_sum(inputs), name='metric_2')\n",
      " |          return inputs\n",
      " |      ```\n",
      " |      \n",
      " |      This method can also be called directly on a Functional Model during\n",
      " |      construction. In this case, any tensor passed to this Model must\n",
      " |      be symbolic and be able to be traced back to the model's `Input`s. These\n",
      " |      metrics become part of the model's topology and are tracked when you\n",
      " |      save the model via `save()`.\n",
      " |      \n",
      " |      ```python\n",
      " |      inputs = tf.keras.Input(shape=(10,))\n",
      " |      x = tf.keras.layers.Dense(10)(inputs)\n",
      " |      outputs = tf.keras.layers.Dense(1)(x)\n",
      " |      model = tf.keras.Model(inputs, outputs)\n",
      " |      model.add_metric(math_ops.reduce_sum(x), name='metric_1')\n",
      " |      ```\n",
      " |      \n",
      " |      Note: Calling `add_metric()` with the result of a metric object on a\n",
      " |      Functional Model, as shown in the example below, is not supported. This\n",
      " |      is because we cannot trace the metric result tensor back to the model's\n",
      " |      inputs.\n",
      " |      \n",
      " |      ```python\n",
      " |      inputs = tf.keras.Input(shape=(10,))\n",
      " |      x = tf.keras.layers.Dense(10)(inputs)\n",
      " |      outputs = tf.keras.layers.Dense(1)(x)\n",
      " |      model = tf.keras.Model(inputs, outputs)\n",
      " |      model.add_metric(tf.keras.metrics.Mean()(x), name='metric_1')\n",
      " |      ```\n",
      " |      \n",
      " |      Args:\n",
      " |        value: Metric tensor.\n",
      " |        name: String metric name.\n",
      " |        **kwargs: Additional keyword arguments for backward compatibility.\n",
      " |          Accepted values:\n",
      " |          `aggregation` - When the `value` tensor provided is not the result\n",
      " |          of calling a `keras.Metric` instance, it will be aggregated by\n",
      " |          default using a `keras.Metric.Mean`.\n",
      " |  \n",
      " |  add_update(self, updates)\n",
      " |      Add update op(s), potentially dependent on layer inputs.\n",
      " |      \n",
      " |      Weight updates (for instance, the updates of the moving mean and\n",
      " |      variance in a BatchNormalization layer) may be dependent on the inputs\n",
      " |      passed when calling a layer. Hence, when reusing the same layer on\n",
      " |      different inputs `a` and `b`, some entries in `layer.updates` may be\n",
      " |      dependent on `a` and some on `b`. This method automatically keeps track\n",
      " |      of dependencies.\n",
      " |      \n",
      " |      This call is ignored when eager execution is enabled (in that case,\n",
      " |      variable updates are run on the fly and thus do not need to be tracked\n",
      " |      for later execution).\n",
      " |      \n",
      " |      Args:\n",
      " |        updates: Update op, or list/tuple of update ops, or zero-arg callable\n",
      " |          that returns an update op. A zero-arg callable should be passed in\n",
      " |          order to disable running the updates by setting `trainable=False`\n",
      " |          on this Layer, when executing in Eager mode.\n",
      " |  \n",
      " |  add_variable(self, *args, **kwargs)\n",
      " |      Deprecated, do NOT use! Alias for `add_weight`.\n",
      " |  \n",
      " |  add_weight(self, name=None, shape=None, dtype=None, initializer=None, regularizer=None, trainable=None, constraint=None, use_resource=None, synchronization=<VariableSynchronization.AUTO: 0>, aggregation=<VariableAggregationV2.NONE: 0>, **kwargs)\n",
      " |      Adds a new variable to the layer.\n",
      " |      \n",
      " |      Args:\n",
      " |        name: Variable name.\n",
      " |        shape: Variable shape. Defaults to scalar if unspecified.\n",
      " |        dtype: The type of the variable. Defaults to `self.dtype`.\n",
      " |        initializer: Initializer instance (callable).\n",
      " |        regularizer: Regularizer instance (callable).\n",
      " |        trainable: Boolean, whether the variable should be part of the layer's\n",
      " |          \"trainable_variables\" (e.g. variables, biases)\n",
      " |          or \"non_trainable_variables\" (e.g. BatchNorm mean and variance).\n",
      " |          Note that `trainable` cannot be `True` if `synchronization`\n",
      " |          is set to `ON_READ`.\n",
      " |        constraint: Constraint instance (callable).\n",
      " |        use_resource: Whether to use a `ResourceVariable` or not.\n",
      " |          See [this guide](\n",
      " |          https://www.tensorflow.org/guide/migrate/tf1_vs_tf2#resourcevariables_instead_of_referencevariables)\n",
      " |           for more information.\n",
      " |        synchronization: Indicates when a distributed a variable will be\n",
      " |          aggregated. Accepted values are constants defined in the class\n",
      " |          `tf.VariableSynchronization`. By default the synchronization is set\n",
      " |          to `AUTO` and the current `DistributionStrategy` chooses when to\n",
      " |          synchronize. If `synchronization` is set to `ON_READ`, `trainable`\n",
      " |          must not be set to `True`.\n",
      " |        aggregation: Indicates how a distributed variable will be aggregated.\n",
      " |          Accepted values are constants defined in the class\n",
      " |          `tf.VariableAggregation`.\n",
      " |        **kwargs: Additional keyword arguments. Accepted values are `getter`,\n",
      " |          `collections`, `experimental_autocast` and `caching_device`.\n",
      " |      \n",
      " |      Returns:\n",
      " |        The variable created.\n",
      " |      \n",
      " |      Raises:\n",
      " |        ValueError: When giving unsupported dtype and no initializer or when\n",
      " |          trainable has been set to True with synchronization set as\n",
      " |          `ON_READ`.\n",
      " |  \n",
      " |  compute_output_signature(self, input_signature)\n",
      " |      Compute the output tensor signature of the layer based on the inputs.\n",
      " |      \n",
      " |      Unlike a TensorShape object, a TensorSpec object contains both shape\n",
      " |      and dtype information for a tensor. This method allows layers to provide\n",
      " |      output dtype information if it is different from the input dtype.\n",
      " |      For any layer that doesn't implement this function,\n",
      " |      the framework will fall back to use `compute_output_shape`, and will\n",
      " |      assume that the output dtype matches the input dtype.\n",
      " |      \n",
      " |      Args:\n",
      " |        input_signature: Single TensorSpec or nested structure of TensorSpec\n",
      " |          objects, describing a candidate input for the layer.\n",
      " |      \n",
      " |      Returns:\n",
      " |        Single TensorSpec or nested structure of TensorSpec objects,\n",
      " |          describing how the layer would transform the provided input.\n",
      " |      \n",
      " |      Raises:\n",
      " |        TypeError: If input_signature contains a non-TensorSpec object.\n",
      " |  \n",
      " |  count_params(self)\n",
      " |      Count the total number of scalars composing the weights.\n",
      " |      \n",
      " |      Returns:\n",
      " |          An integer count.\n",
      " |      \n",
      " |      Raises:\n",
      " |          ValueError: if the layer isn't yet built\n",
      " |            (in which case its weights aren't yet defined).\n",
      " |  \n",
      " |  finalize_state(self)\n",
      " |      Finalizes the layers state after updating layer weights.\n",
      " |      \n",
      " |      This function can be subclassed in a layer and will be called after\n",
      " |      updating a layer weights. It can be overridden to finalize any\n",
      " |      additional layer state after a weight update.\n",
      " |      \n",
      " |      This function will be called after weights of a layer have been restored\n",
      " |      from a loaded model.\n",
      " |  \n",
      " |  get_input_at(self, node_index)\n",
      " |      Retrieves the input tensor(s) of a layer at a given node.\n",
      " |      \n",
      " |      Args:\n",
      " |          node_index: Integer, index of the node\n",
      " |              from which to retrieve the attribute.\n",
      " |              E.g. `node_index=0` will correspond to the\n",
      " |              first input node of the layer.\n",
      " |      \n",
      " |      Returns:\n",
      " |          A tensor (or list of tensors if the layer has multiple inputs).\n",
      " |      \n",
      " |      Raises:\n",
      " |        RuntimeError: If called in Eager mode.\n",
      " |  \n",
      " |  get_input_mask_at(self, node_index)\n",
      " |      Retrieves the input mask tensor(s) of a layer at a given node.\n",
      " |      \n",
      " |      Args:\n",
      " |          node_index: Integer, index of the node\n",
      " |              from which to retrieve the attribute.\n",
      " |              E.g. `node_index=0` will correspond to the\n",
      " |              first time the layer was called.\n",
      " |      \n",
      " |      Returns:\n",
      " |          A mask tensor\n",
      " |          (or list of tensors if the layer has multiple inputs).\n",
      " |  \n",
      " |  get_input_shape_at(self, node_index)\n",
      " |      Retrieves the input shape(s) of a layer at a given node.\n",
      " |      \n",
      " |      Args:\n",
      " |          node_index: Integer, index of the node\n",
      " |              from which to retrieve the attribute.\n",
      " |              E.g. `node_index=0` will correspond to the\n",
      " |              first time the layer was called.\n",
      " |      \n",
      " |      Returns:\n",
      " |          A shape tuple\n",
      " |          (or list of shape tuples if the layer has multiple inputs).\n",
      " |      \n",
      " |      Raises:\n",
      " |        RuntimeError: If called in Eager mode.\n",
      " |  \n",
      " |  get_output_at(self, node_index)\n",
      " |      Retrieves the output tensor(s) of a layer at a given node.\n",
      " |      \n",
      " |      Args:\n",
      " |          node_index: Integer, index of the node\n",
      " |              from which to retrieve the attribute.\n",
      " |              E.g. `node_index=0` will correspond to the\n",
      " |              first output node of the layer.\n",
      " |      \n",
      " |      Returns:\n",
      " |          A tensor (or list of tensors if the layer has multiple outputs).\n",
      " |      \n",
      " |      Raises:\n",
      " |        RuntimeError: If called in Eager mode.\n",
      " |  \n",
      " |  get_output_mask_at(self, node_index)\n",
      " |      Retrieves the output mask tensor(s) of a layer at a given node.\n",
      " |      \n",
      " |      Args:\n",
      " |          node_index: Integer, index of the node\n",
      " |              from which to retrieve the attribute.\n",
      " |              E.g. `node_index=0` will correspond to the\n",
      " |              first time the layer was called.\n",
      " |      \n",
      " |      Returns:\n",
      " |          A mask tensor\n",
      " |          (or list of tensors if the layer has multiple outputs).\n",
      " |  \n",
      " |  get_output_shape_at(self, node_index)\n",
      " |      Retrieves the output shape(s) of a layer at a given node.\n",
      " |      \n",
      " |      Args:\n",
      " |          node_index: Integer, index of the node\n",
      " |              from which to retrieve the attribute.\n",
      " |              E.g. `node_index=0` will correspond to the\n",
      " |              first time the layer was called.\n",
      " |      \n",
      " |      Returns:\n",
      " |          A shape tuple\n",
      " |          (or list of shape tuples if the layer has multiple outputs).\n",
      " |      \n",
      " |      Raises:\n",
      " |        RuntimeError: If called in Eager mode.\n",
      " |  \n",
      " |  set_weights(self, weights)\n",
      " |      Sets the weights of the layer, from NumPy arrays.\n",
      " |      \n",
      " |      The weights of a layer represent the state of the layer. This function\n",
      " |      sets the weight values from numpy arrays. The weight values should be\n",
      " |      passed in the order they are created by the layer. Note that the layer's\n",
      " |      weights must be instantiated before calling this function, by calling\n",
      " |      the layer.\n",
      " |      \n",
      " |      For example, a `Dense` layer returns a list of two values: the kernel\n",
      " |      matrix and the bias vector. These can be used to set the weights of\n",
      " |      another `Dense` layer:\n",
      " |      \n",
      " |      >>> layer_a = tf.keras.layers.Dense(1,\n",
      " |      ...   kernel_initializer=tf.constant_initializer(1.))\n",
      " |      >>> a_out = layer_a(tf.convert_to_tensor([[1., 2., 3.]]))\n",
      " |      >>> layer_a.get_weights()\n",
      " |      [array([[1.],\n",
      " |             [1.],\n",
      " |             [1.]], dtype=float32), array([0.], dtype=float32)]\n",
      " |      >>> layer_b = tf.keras.layers.Dense(1,\n",
      " |      ...   kernel_initializer=tf.constant_initializer(2.))\n",
      " |      >>> b_out = layer_b(tf.convert_to_tensor([[10., 20., 30.]]))\n",
      " |      >>> layer_b.get_weights()\n",
      " |      [array([[2.],\n",
      " |             [2.],\n",
      " |             [2.]], dtype=float32), array([0.], dtype=float32)]\n",
      " |      >>> layer_b.set_weights(layer_a.get_weights())\n",
      " |      >>> layer_b.get_weights()\n",
      " |      [array([[1.],\n",
      " |             [1.],\n",
      " |             [1.]], dtype=float32), array([0.], dtype=float32)]\n",
      " |      \n",
      " |      Args:\n",
      " |        weights: a list of NumPy arrays. The number\n",
      " |          of arrays and their shape must match\n",
      " |          number of the dimensions of the weights\n",
      " |          of the layer (i.e. it should match the\n",
      " |          output of `get_weights`).\n",
      " |      \n",
      " |      Raises:\n",
      " |        ValueError: If the provided weights list does not match the\n",
      " |          layer's specifications.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Readonly properties inherited from keras.engine.base_layer.Layer:\n",
      " |  \n",
      " |  compute_dtype\n",
      " |      The dtype of the layer's computations.\n",
      " |      \n",
      " |      This is equivalent to `Layer.dtype_policy.compute_dtype`. Unless\n",
      " |      mixed precision is used, this is the same as `Layer.dtype`, the dtype of\n",
      " |      the weights.\n",
      " |      \n",
      " |      Layers automatically cast their inputs to the compute dtype, which\n",
      " |      causes computations and the output to be in the compute dtype as well.\n",
      " |      This is done by the base Layer class in `Layer.__call__`, so you do not\n",
      " |      have to insert these casts if implementing your own layer.\n",
      " |      \n",
      " |      Layers often perform certain internal computations in higher precision\n",
      " |      when `compute_dtype` is float16 or bfloat16 for numeric stability. The\n",
      " |      output will still typically be float16 or bfloat16 in such cases.\n",
      " |      \n",
      " |      Returns:\n",
      " |        The layer's compute dtype.\n",
      " |  \n",
      " |  dtype\n",
      " |      The dtype of the layer weights.\n",
      " |      \n",
      " |      This is equivalent to `Layer.dtype_policy.variable_dtype`. Unless\n",
      " |      mixed precision is used, this is the same as `Layer.compute_dtype`, the\n",
      " |      dtype of the layer's computations.\n",
      " |  \n",
      " |  dtype_policy\n",
      " |      The dtype policy associated with this layer.\n",
      " |      \n",
      " |      This is an instance of a `tf.keras.mixed_precision.Policy`.\n",
      " |  \n",
      " |  dynamic\n",
      " |      Whether the layer is dynamic (eager-only); set in the constructor.\n",
      " |  \n",
      " |  inbound_nodes\n",
      " |      Return Functional API nodes upstream of this layer.\n",
      " |  \n",
      " |  input_mask\n",
      " |      Retrieves the input mask tensor(s) of a layer.\n",
      " |      \n",
      " |      Only applicable if the layer has exactly one inbound node,\n",
      " |      i.e. if it is connected to one incoming layer.\n",
      " |      \n",
      " |      Returns:\n",
      " |          Input mask tensor (potentially None) or list of input\n",
      " |          mask tensors.\n",
      " |      \n",
      " |      Raises:\n",
      " |          AttributeError: if the layer is connected to\n",
      " |          more than one incoming layers.\n",
      " |  \n",
      " |  losses\n",
      " |      List of losses added using the `add_loss()` API.\n",
      " |      \n",
      " |      Variable regularization tensors are created when this property is\n",
      " |      accessed, so it is eager safe: accessing `losses` under a\n",
      " |      `tf.GradientTape` will propagate gradients back to the corresponding\n",
      " |      variables.\n",
      " |      \n",
      " |      Examples:\n",
      " |      \n",
      " |      >>> class MyLayer(tf.keras.layers.Layer):\n",
      " |      ...   def call(self, inputs):\n",
      " |      ...     self.add_loss(tf.abs(tf.reduce_mean(inputs)))\n",
      " |      ...     return inputs\n",
      " |      >>> l = MyLayer()\n",
      " |      >>> l(np.ones((10, 1)))\n",
      " |      >>> l.losses\n",
      " |      [1.0]\n",
      " |      \n",
      " |      >>> inputs = tf.keras.Input(shape=(10,))\n",
      " |      >>> x = tf.keras.layers.Dense(10)(inputs)\n",
      " |      >>> outputs = tf.keras.layers.Dense(1)(x)\n",
      " |      >>> model = tf.keras.Model(inputs, outputs)\n",
      " |      >>> # Activity regularization.\n",
      " |      >>> len(model.losses)\n",
      " |      0\n",
      " |      >>> model.add_loss(tf.abs(tf.reduce_mean(x)))\n",
      " |      >>> len(model.losses)\n",
      " |      1\n",
      " |      \n",
      " |      >>> inputs = tf.keras.Input(shape=(10,))\n",
      " |      >>> d = tf.keras.layers.Dense(10, kernel_initializer='ones')\n",
      " |      >>> x = d(inputs)\n",
      " |      >>> outputs = tf.keras.layers.Dense(1)(x)\n",
      " |      >>> model = tf.keras.Model(inputs, outputs)\n",
      " |      >>> # Weight regularization.\n",
      " |      >>> model.add_loss(lambda: tf.reduce_mean(d.kernel))\n",
      " |      >>> model.losses\n",
      " |      [<tf.Tensor: shape=(), dtype=float32, numpy=1.0>]\n",
      " |      \n",
      " |      Returns:\n",
      " |        A list of tensors.\n",
      " |  \n",
      " |  name\n",
      " |      Name of the layer (string), set in the constructor.\n",
      " |  \n",
      " |  non_trainable_variables\n",
      " |      Sequence of non-trainable variables owned by this module and its submodules.\n",
      " |      \n",
      " |      Note: this method uses reflection to find variables on the current instance\n",
      " |      and submodules. For performance reasons you may wish to cache the result\n",
      " |      of calling this method if you don't expect the return value to change.\n",
      " |      \n",
      " |      Returns:\n",
      " |        A sequence of variables for the current module (sorted by attribute\n",
      " |        name) followed by variables from all submodules recursively (breadth\n",
      " |        first).\n",
      " |  \n",
      " |  outbound_nodes\n",
      " |      Return Functional API nodes downstream of this layer.\n",
      " |  \n",
      " |  output_mask\n",
      " |      Retrieves the output mask tensor(s) of a layer.\n",
      " |      \n",
      " |      Only applicable if the layer has exactly one inbound node,\n",
      " |      i.e. if it is connected to one incoming layer.\n",
      " |      \n",
      " |      Returns:\n",
      " |          Output mask tensor (potentially None) or list of output\n",
      " |          mask tensors.\n",
      " |      \n",
      " |      Raises:\n",
      " |          AttributeError: if the layer is connected to\n",
      " |          more than one incoming layers.\n",
      " |  \n",
      " |  trainable_variables\n",
      " |      Sequence of trainable variables owned by this module and its submodules.\n",
      " |      \n",
      " |      Note: this method uses reflection to find variables on the current instance\n",
      " |      and submodules. For performance reasons you may wish to cache the result\n",
      " |      of calling this method if you don't expect the return value to change.\n",
      " |      \n",
      " |      Returns:\n",
      " |        A sequence of variables for the current module (sorted by attribute\n",
      " |        name) followed by variables from all submodules recursively (breadth\n",
      " |        first).\n",
      " |  \n",
      " |  updates\n",
      " |  \n",
      " |  variable_dtype\n",
      " |      Alias of `Layer.dtype`, the dtype of the weights.\n",
      " |  \n",
      " |  variables\n",
      " |      Returns the list of all layer variables/weights.\n",
      " |      \n",
      " |      Alias of `self.weights`.\n",
      " |      \n",
      " |      Note: This will not track the weights of nested `tf.Modules` that are\n",
      " |      not themselves Keras layers.\n",
      " |      \n",
      " |      Returns:\n",
      " |        A list of variables.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Data descriptors inherited from keras.engine.base_layer.Layer:\n",
      " |  \n",
      " |  activity_regularizer\n",
      " |      Optional regularizer function for the output of this layer.\n",
      " |  \n",
      " |  stateful\n",
      " |  \n",
      " |  supports_masking\n",
      " |      Whether this layer supports computing a mask using `compute_mask`.\n",
      " |  \n",
      " |  trainable\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Class methods inherited from tensorflow.python.module.module.Module:\n",
      " |  \n",
      " |  with_name_scope(method) from builtins.type\n",
      " |      Decorator to automatically enter the module name scope.\n",
      " |      \n",
      " |      >>> class MyModule(tf.Module):\n",
      " |      ...   @tf.Module.with_name_scope\n",
      " |      ...   def __call__(self, x):\n",
      " |      ...     if not hasattr(self, 'w'):\n",
      " |      ...       self.w = tf.Variable(tf.random.normal([x.shape[1], 3]))\n",
      " |      ...     return tf.matmul(x, self.w)\n",
      " |      \n",
      " |      Using the above module would produce `tf.Variable`s and `tf.Tensor`s whose\n",
      " |      names included the module name:\n",
      " |      \n",
      " |      >>> mod = MyModule()\n",
      " |      >>> mod(tf.ones([1, 2]))\n",
      " |      <tf.Tensor: shape=(1, 3), dtype=float32, numpy=..., dtype=float32)>\n",
      " |      >>> mod.w\n",
      " |      <tf.Variable 'my_module/Variable:0' shape=(2, 3) dtype=float32,\n",
      " |      numpy=..., dtype=float32)>\n",
      " |      \n",
      " |      Args:\n",
      " |        method: The method to wrap.\n",
      " |      \n",
      " |      Returns:\n",
      " |        The original method wrapped such that it enters the module's name scope.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Readonly properties inherited from tensorflow.python.module.module.Module:\n",
      " |  \n",
      " |  name_scope\n",
      " |      Returns a `tf.name_scope` instance for this class.\n",
      " |  \n",
      " |  submodules\n",
      " |      Sequence of all sub-modules.\n",
      " |      \n",
      " |      Submodules are modules which are properties of this module, or found as\n",
      " |      properties of modules which are properties of this module (and so on).\n",
      " |      \n",
      " |      >>> a = tf.Module()\n",
      " |      >>> b = tf.Module()\n",
      " |      >>> c = tf.Module()\n",
      " |      >>> a.b = b\n",
      " |      >>> b.c = c\n",
      " |      >>> list(a.submodules) == [b, c]\n",
      " |      True\n",
      " |      >>> list(b.submodules) == [c]\n",
      " |      True\n",
      " |      >>> list(c.submodules) == []\n",
      " |      True\n",
      " |      \n",
      " |      Returns:\n",
      " |        A sequence of all submodules.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Data descriptors inherited from tensorflow.python.trackable.base.Trackable:\n",
      " |  \n",
      " |  __dict__\n",
      " |      dictionary for instance variables (if defined)\n",
      " |  \n",
      " |  __weakref__\n",
      " |      list of weak references to the object (if defined)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "help(Sequential)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "31ab7428",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Help on class Dense in module keras.layers.core.dense:\n",
      "\n",
      "class Dense(keras.engine.base_layer.Layer)\n",
      " |  Dense(units, activation=None, use_bias=True, kernel_initializer='glorot_uniform', bias_initializer='zeros', kernel_regularizer=None, bias_regularizer=None, activity_regularizer=None, kernel_constraint=None, bias_constraint=None, **kwargs)\n",
      " |  \n",
      " |  Just your regular densely-connected NN layer.\n",
      " |  \n",
      " |  `Dense` implements the operation:\n",
      " |  `output = activation(dot(input, kernel) + bias)`\n",
      " |  where `activation` is the element-wise activation function\n",
      " |  passed as the `activation` argument, `kernel` is a weights matrix\n",
      " |  created by the layer, and `bias` is a bias vector created by the layer\n",
      " |  (only applicable if `use_bias` is `True`). These are all attributes of\n",
      " |  `Dense`.\n",
      " |  \n",
      " |  Note: If the input to the layer has a rank greater than 2, then `Dense`\n",
      " |  computes the dot product between the `inputs` and the `kernel` along the\n",
      " |  last axis of the `inputs` and axis 0 of the `kernel` (using `tf.tensordot`).\n",
      " |  For example, if input has dimensions `(batch_size, d0, d1)`, then we create\n",
      " |  a `kernel` with shape `(d1, units)`, and the `kernel` operates along axis 2\n",
      " |  of the `input`, on every sub-tensor of shape `(1, 1, d1)` (there are\n",
      " |  `batch_size * d0` such sub-tensors).  The output in this case will have\n",
      " |  shape `(batch_size, d0, units)`.\n",
      " |  \n",
      " |  Besides, layer attributes cannot be modified after the layer has been called\n",
      " |  once (except the `trainable` attribute).\n",
      " |  When a popular kwarg `input_shape` is passed, then keras will create\n",
      " |  an input layer to insert before the current layer. This can be treated\n",
      " |  equivalent to explicitly defining an `InputLayer`.\n",
      " |  \n",
      " |  Example:\n",
      " |  \n",
      " |  >>> # Create a `Sequential` model and add a Dense layer as the first layer.\n",
      " |  >>> model = tf.keras.models.Sequential()\n",
      " |  >>> model.add(tf.keras.Input(shape=(16,)))\n",
      " |  >>> model.add(tf.keras.layers.Dense(32, activation='relu'))\n",
      " |  >>> # Now the model will take as input arrays of shape (None, 16)\n",
      " |  >>> # and output arrays of shape (None, 32).\n",
      " |  >>> # Note that after the first layer, you don't need to specify\n",
      " |  >>> # the size of the input anymore:\n",
      " |  >>> model.add(tf.keras.layers.Dense(32))\n",
      " |  >>> model.output_shape\n",
      " |  (None, 32)\n",
      " |  \n",
      " |  Args:\n",
      " |    units: Positive integer, dimensionality of the output space.\n",
      " |    activation: Activation function to use.\n",
      " |      If you don't specify anything, no activation is applied\n",
      " |      (ie. \"linear\" activation: `a(x) = x`).\n",
      " |    use_bias: Boolean, whether the layer uses a bias vector.\n",
      " |    kernel_initializer: Initializer for the `kernel` weights matrix.\n",
      " |    bias_initializer: Initializer for the bias vector.\n",
      " |    kernel_regularizer: Regularizer function applied to\n",
      " |      the `kernel` weights matrix.\n",
      " |    bias_regularizer: Regularizer function applied to the bias vector.\n",
      " |    activity_regularizer: Regularizer function applied to\n",
      " |      the output of the layer (its \"activation\").\n",
      " |    kernel_constraint: Constraint function applied to\n",
      " |      the `kernel` weights matrix.\n",
      " |    bias_constraint: Constraint function applied to the bias vector.\n",
      " |  \n",
      " |  Input shape:\n",
      " |    N-D tensor with shape: `(batch_size, ..., input_dim)`.\n",
      " |    The most common situation would be\n",
      " |    a 2D input with shape `(batch_size, input_dim)`.\n",
      " |  \n",
      " |  Output shape:\n",
      " |    N-D tensor with shape: `(batch_size, ..., units)`.\n",
      " |    For instance, for a 2D input with shape `(batch_size, input_dim)`,\n",
      " |    the output would have shape `(batch_size, units)`.\n",
      " |  \n",
      " |  Method resolution order:\n",
      " |      Dense\n",
      " |      keras.engine.base_layer.Layer\n",
      " |      tensorflow.python.module.module.Module\n",
      " |      tensorflow.python.trackable.autotrackable.AutoTrackable\n",
      " |      tensorflow.python.trackable.base.Trackable\n",
      " |      keras.utils.version_utils.LayerVersionSelector\n",
      " |      builtins.object\n",
      " |  \n",
      " |  Methods defined here:\n",
      " |  \n",
      " |  __init__(self, units, activation=None, use_bias=True, kernel_initializer='glorot_uniform', bias_initializer='zeros', kernel_regularizer=None, bias_regularizer=None, activity_regularizer=None, kernel_constraint=None, bias_constraint=None, **kwargs)\n",
      " |  \n",
      " |  build(self, input_shape)\n",
      " |      Creates the variables of the layer (optional, for subclass implementers).\n",
      " |      \n",
      " |      This is a method that implementers of subclasses of `Layer` or `Model`\n",
      " |      can override if they need a state-creation step in-between\n",
      " |      layer instantiation and layer call. It is invoked automatically before\n",
      " |      the first execution of `call()`.\n",
      " |      \n",
      " |      This is typically used to create the weights of `Layer` subclasses\n",
      " |      (at the discretion of the subclass implementer).\n",
      " |      \n",
      " |      Args:\n",
      " |        input_shape: Instance of `TensorShape`, or list of instances of\n",
      " |          `TensorShape` if the layer expects a list of inputs\n",
      " |          (one instance per input).\n",
      " |  \n",
      " |  call(self, inputs)\n",
      " |      This is where the layer's logic lives.\n",
      " |      \n",
      " |      The `call()` method may not create state (except in its first\n",
      " |      invocation, wrapping the creation of variables or other resources in\n",
      " |      `tf.init_scope()`).  It is recommended to create state, including\n",
      " |      `tf.Variable` instances and nested `Layer` instances,\n",
      " |       in `__init__()`, or in the `build()` method that is\n",
      " |      called automatically before `call()` executes for the first time.\n",
      " |      \n",
      " |      Args:\n",
      " |        inputs: Input tensor, or dict/list/tuple of input tensors.\n",
      " |          The first positional `inputs` argument is subject to special rules:\n",
      " |          - `inputs` must be explicitly passed. A layer cannot have zero\n",
      " |            arguments, and `inputs` cannot be provided via the default value\n",
      " |            of a keyword argument.\n",
      " |          - NumPy array or Python scalar values in `inputs` get cast as\n",
      " |            tensors.\n",
      " |          - Keras mask metadata is only collected from `inputs`.\n",
      " |          - Layers are built (`build(input_shape)` method)\n",
      " |            using shape info from `inputs` only.\n",
      " |          - `input_spec` compatibility is only checked against `inputs`.\n",
      " |          - Mixed precision input casting is only applied to `inputs`.\n",
      " |            If a layer has tensor arguments in `*args` or `**kwargs`, their\n",
      " |            casting behavior in mixed precision should be handled manually.\n",
      " |          - The SavedModel input specification is generated using `inputs`\n",
      " |            only.\n",
      " |          - Integration with various ecosystem packages like TFMOT, TFLite,\n",
      " |            TF.js, etc is only supported for `inputs` and not for tensors in\n",
      " |            positional and keyword arguments.\n",
      " |        *args: Additional positional arguments. May contain tensors, although\n",
      " |          this is not recommended, for the reasons above.\n",
      " |        **kwargs: Additional keyword arguments. May contain tensors, although\n",
      " |          this is not recommended, for the reasons above.\n",
      " |          The following optional keyword arguments are reserved:\n",
      " |          - `training`: Boolean scalar tensor of Python boolean indicating\n",
      " |            whether the `call` is meant for training or inference.\n",
      " |          - `mask`: Boolean input mask. If the layer's `call()` method takes a\n",
      " |            `mask` argument, its default value will be set to the mask\n",
      " |            generated for `inputs` by the previous layer (if `input` did come\n",
      " |            from a layer that generated a corresponding mask, i.e. if it came\n",
      " |            from a Keras layer with masking support).\n",
      " |      \n",
      " |      Returns:\n",
      " |        A tensor or list/tuple of tensors.\n",
      " |  \n",
      " |  compute_output_shape(self, input_shape)\n",
      " |      Computes the output shape of the layer.\n",
      " |      \n",
      " |      This method will cause the layer's state to be built, if that has not\n",
      " |      happened before. This requires that the layer will later be used with\n",
      " |      inputs that match the input shape provided here.\n",
      " |      \n",
      " |      Args:\n",
      " |          input_shape: Shape tuple (tuple of integers) or `tf.TensorShape`,\n",
      " |              or structure of shape tuples / `tf.TensorShape` instances\n",
      " |              (one per output tensor of the layer).\n",
      " |              Shape tuples can include None for free dimensions,\n",
      " |              instead of an integer.\n",
      " |      \n",
      " |      Returns:\n",
      " |          A `tf.TensorShape` instance\n",
      " |          or structure of `tf.TensorShape` instances.\n",
      " |  \n",
      " |  get_config(self)\n",
      " |      Returns the config of the layer.\n",
      " |      \n",
      " |      A layer config is a Python dictionary (serializable)\n",
      " |      containing the configuration of a layer.\n",
      " |      The same layer can be reinstantiated later\n",
      " |      (without its trained weights) from this configuration.\n",
      " |      \n",
      " |      The config of a layer does not include connectivity\n",
      " |      information, nor the layer class name. These are handled\n",
      " |      by `Network` (one layer of abstraction above).\n",
      " |      \n",
      " |      Note that `get_config()` does not guarantee to return a fresh copy of\n",
      " |      dict every time it is called. The callers should make a copy of the\n",
      " |      returned dict if they want to modify it.\n",
      " |      \n",
      " |      Returns:\n",
      " |          Python dictionary.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Methods inherited from keras.engine.base_layer.Layer:\n",
      " |  \n",
      " |  __call__(self, *args, **kwargs)\n",
      " |      Wraps `call`, applying pre- and post-processing steps.\n",
      " |      \n",
      " |      Args:\n",
      " |        *args: Positional arguments to be passed to `self.call`.\n",
      " |        **kwargs: Keyword arguments to be passed to `self.call`.\n",
      " |      \n",
      " |      Returns:\n",
      " |        Output tensor(s).\n",
      " |      \n",
      " |      Note:\n",
      " |        - The following optional keyword arguments are reserved for specific\n",
      " |          uses:\n",
      " |          * `training`: Boolean scalar tensor of Python boolean indicating\n",
      " |            whether the `call` is meant for training or inference.\n",
      " |          * `mask`: Boolean input mask.\n",
      " |        - If the layer's `call` method takes a `mask` argument (as some Keras\n",
      " |          layers do), its default value will be set to the mask generated\n",
      " |          for `inputs` by the previous layer (if `input` did come from\n",
      " |          a layer that generated a corresponding mask, i.e. if it came from\n",
      " |          a Keras layer with masking support.\n",
      " |        - If the layer is not built, the method will call `build`.\n",
      " |      \n",
      " |      Raises:\n",
      " |        ValueError: if the layer's `call` method returns None (an invalid\n",
      " |          value).\n",
      " |        RuntimeError: if `super().__init__()` was not called in the\n",
      " |          constructor.\n",
      " |  \n",
      " |  __delattr__(self, name)\n",
      " |      Implement delattr(self, name).\n",
      " |  \n",
      " |  __getstate__(self)\n",
      " |  \n",
      " |  __setattr__(self, name, value)\n",
      " |      Support self.foo = trackable syntax.\n",
      " |  \n",
      " |  __setstate__(self, state)\n",
      " |  \n",
      " |  add_loss(self, losses, **kwargs)\n",
      " |      Add loss tensor(s), potentially dependent on layer inputs.\n",
      " |      \n",
      " |      Some losses (for instance, activity regularization losses) may be\n",
      " |      dependent on the inputs passed when calling a layer. Hence, when reusing\n",
      " |      the same layer on different inputs `a` and `b`, some entries in\n",
      " |      `layer.losses` may be dependent on `a` and some on `b`. This method\n",
      " |      automatically keeps track of dependencies.\n",
      " |      \n",
      " |      This method can be used inside a subclassed layer or model's `call`\n",
      " |      function, in which case `losses` should be a Tensor or list of Tensors.\n",
      " |      \n",
      " |      Example:\n",
      " |      \n",
      " |      ```python\n",
      " |      class MyLayer(tf.keras.layers.Layer):\n",
      " |        def call(self, inputs):\n",
      " |          self.add_loss(tf.abs(tf.reduce_mean(inputs)))\n",
      " |          return inputs\n",
      " |      ```\n",
      " |      \n",
      " |      This method can also be called directly on a Functional Model during\n",
      " |      construction. In this case, any loss Tensors passed to this Model must\n",
      " |      be symbolic and be able to be traced back to the model's `Input`s. These\n",
      " |      losses become part of the model's topology and are tracked in\n",
      " |      `get_config`.\n",
      " |      \n",
      " |      Example:\n",
      " |      \n",
      " |      ```python\n",
      " |      inputs = tf.keras.Input(shape=(10,))\n",
      " |      x = tf.keras.layers.Dense(10)(inputs)\n",
      " |      outputs = tf.keras.layers.Dense(1)(x)\n",
      " |      model = tf.keras.Model(inputs, outputs)\n",
      " |      # Activity regularization.\n",
      " |      model.add_loss(tf.abs(tf.reduce_mean(x)))\n",
      " |      ```\n",
      " |      \n",
      " |      If this is not the case for your loss (if, for example, your loss\n",
      " |      references a `Variable` of one of the model's layers), you can wrap your\n",
      " |      loss in a zero-argument lambda. These losses are not tracked as part of\n",
      " |      the model's topology since they can't be serialized.\n",
      " |      \n",
      " |      Example:\n",
      " |      \n",
      " |      ```python\n",
      " |      inputs = tf.keras.Input(shape=(10,))\n",
      " |      d = tf.keras.layers.Dense(10)\n",
      " |      x = d(inputs)\n",
      " |      outputs = tf.keras.layers.Dense(1)(x)\n",
      " |      model = tf.keras.Model(inputs, outputs)\n",
      " |      # Weight regularization.\n",
      " |      model.add_loss(lambda: tf.reduce_mean(d.kernel))\n",
      " |      ```\n",
      " |      \n",
      " |      Args:\n",
      " |        losses: Loss tensor, or list/tuple of tensors. Rather than tensors,\n",
      " |          losses may also be zero-argument callables which create a loss\n",
      " |          tensor.\n",
      " |        **kwargs: Used for backwards compatibility only.\n",
      " |  \n",
      " |  add_metric(self, value, name=None, **kwargs)\n",
      " |      Adds metric tensor to the layer.\n",
      " |      \n",
      " |      This method can be used inside the `call()` method of a subclassed layer\n",
      " |      or model.\n",
      " |      \n",
      " |      ```python\n",
      " |      class MyMetricLayer(tf.keras.layers.Layer):\n",
      " |        def __init__(self):\n",
      " |          super(MyMetricLayer, self).__init__(name='my_metric_layer')\n",
      " |          self.mean = tf.keras.metrics.Mean(name='metric_1')\n",
      " |      \n",
      " |        def call(self, inputs):\n",
      " |          self.add_metric(self.mean(inputs))\n",
      " |          self.add_metric(tf.reduce_sum(inputs), name='metric_2')\n",
      " |          return inputs\n",
      " |      ```\n",
      " |      \n",
      " |      This method can also be called directly on a Functional Model during\n",
      " |      construction. In this case, any tensor passed to this Model must\n",
      " |      be symbolic and be able to be traced back to the model's `Input`s. These\n",
      " |      metrics become part of the model's topology and are tracked when you\n",
      " |      save the model via `save()`.\n",
      " |      \n",
      " |      ```python\n",
      " |      inputs = tf.keras.Input(shape=(10,))\n",
      " |      x = tf.keras.layers.Dense(10)(inputs)\n",
      " |      outputs = tf.keras.layers.Dense(1)(x)\n",
      " |      model = tf.keras.Model(inputs, outputs)\n",
      " |      model.add_metric(math_ops.reduce_sum(x), name='metric_1')\n",
      " |      ```\n",
      " |      \n",
      " |      Note: Calling `add_metric()` with the result of a metric object on a\n",
      " |      Functional Model, as shown in the example below, is not supported. This\n",
      " |      is because we cannot trace the metric result tensor back to the model's\n",
      " |      inputs.\n",
      " |      \n",
      " |      ```python\n",
      " |      inputs = tf.keras.Input(shape=(10,))\n",
      " |      x = tf.keras.layers.Dense(10)(inputs)\n",
      " |      outputs = tf.keras.layers.Dense(1)(x)\n",
      " |      model = tf.keras.Model(inputs, outputs)\n",
      " |      model.add_metric(tf.keras.metrics.Mean()(x), name='metric_1')\n",
      " |      ```\n",
      " |      \n",
      " |      Args:\n",
      " |        value: Metric tensor.\n",
      " |        name: String metric name.\n",
      " |        **kwargs: Additional keyword arguments for backward compatibility.\n",
      " |          Accepted values:\n",
      " |          `aggregation` - When the `value` tensor provided is not the result\n",
      " |          of calling a `keras.Metric` instance, it will be aggregated by\n",
      " |          default using a `keras.Metric.Mean`.\n",
      " |  \n",
      " |  add_update(self, updates)\n",
      " |      Add update op(s), potentially dependent on layer inputs.\n",
      " |      \n",
      " |      Weight updates (for instance, the updates of the moving mean and\n",
      " |      variance in a BatchNormalization layer) may be dependent on the inputs\n",
      " |      passed when calling a layer. Hence, when reusing the same layer on\n",
      " |      different inputs `a` and `b`, some entries in `layer.updates` may be\n",
      " |      dependent on `a` and some on `b`. This method automatically keeps track\n",
      " |      of dependencies.\n",
      " |      \n",
      " |      This call is ignored when eager execution is enabled (in that case,\n",
      " |      variable updates are run on the fly and thus do not need to be tracked\n",
      " |      for later execution).\n",
      " |      \n",
      " |      Args:\n",
      " |        updates: Update op, or list/tuple of update ops, or zero-arg callable\n",
      " |          that returns an update op. A zero-arg callable should be passed in\n",
      " |          order to disable running the updates by setting `trainable=False`\n",
      " |          on this Layer, when executing in Eager mode.\n",
      " |  \n",
      " |  add_variable(self, *args, **kwargs)\n",
      " |      Deprecated, do NOT use! Alias for `add_weight`.\n",
      " |  \n",
      " |  add_weight(self, name=None, shape=None, dtype=None, initializer=None, regularizer=None, trainable=None, constraint=None, use_resource=None, synchronization=<VariableSynchronization.AUTO: 0>, aggregation=<VariableAggregationV2.NONE: 0>, **kwargs)\n",
      " |      Adds a new variable to the layer.\n",
      " |      \n",
      " |      Args:\n",
      " |        name: Variable name.\n",
      " |        shape: Variable shape. Defaults to scalar if unspecified.\n",
      " |        dtype: The type of the variable. Defaults to `self.dtype`.\n",
      " |        initializer: Initializer instance (callable).\n",
      " |        regularizer: Regularizer instance (callable).\n",
      " |        trainable: Boolean, whether the variable should be part of the layer's\n",
      " |          \"trainable_variables\" (e.g. variables, biases)\n",
      " |          or \"non_trainable_variables\" (e.g. BatchNorm mean and variance).\n",
      " |          Note that `trainable` cannot be `True` if `synchronization`\n",
      " |          is set to `ON_READ`.\n",
      " |        constraint: Constraint instance (callable).\n",
      " |        use_resource: Whether to use a `ResourceVariable` or not.\n",
      " |          See [this guide](\n",
      " |          https://www.tensorflow.org/guide/migrate/tf1_vs_tf2#resourcevariables_instead_of_referencevariables)\n",
      " |           for more information.\n",
      " |        synchronization: Indicates when a distributed a variable will be\n",
      " |          aggregated. Accepted values are constants defined in the class\n",
      " |          `tf.VariableSynchronization`. By default the synchronization is set\n",
      " |          to `AUTO` and the current `DistributionStrategy` chooses when to\n",
      " |          synchronize. If `synchronization` is set to `ON_READ`, `trainable`\n",
      " |          must not be set to `True`.\n",
      " |        aggregation: Indicates how a distributed variable will be aggregated.\n",
      " |          Accepted values are constants defined in the class\n",
      " |          `tf.VariableAggregation`.\n",
      " |        **kwargs: Additional keyword arguments. Accepted values are `getter`,\n",
      " |          `collections`, `experimental_autocast` and `caching_device`.\n",
      " |      \n",
      " |      Returns:\n",
      " |        The variable created.\n",
      " |      \n",
      " |      Raises:\n",
      " |        ValueError: When giving unsupported dtype and no initializer or when\n",
      " |          trainable has been set to True with synchronization set as\n",
      " |          `ON_READ`.\n",
      " |  \n",
      " |  compute_mask(self, inputs, mask=None)\n",
      " |      Computes an output mask tensor.\n",
      " |      \n",
      " |      Args:\n",
      " |          inputs: Tensor or list of tensors.\n",
      " |          mask: Tensor or list of tensors.\n",
      " |      \n",
      " |      Returns:\n",
      " |          None or a tensor (or list of tensors,\n",
      " |              one per output tensor of the layer).\n",
      " |  \n",
      " |  compute_output_signature(self, input_signature)\n",
      " |      Compute the output tensor signature of the layer based on the inputs.\n",
      " |      \n",
      " |      Unlike a TensorShape object, a TensorSpec object contains both shape\n",
      " |      and dtype information for a tensor. This method allows layers to provide\n",
      " |      output dtype information if it is different from the input dtype.\n",
      " |      For any layer that doesn't implement this function,\n",
      " |      the framework will fall back to use `compute_output_shape`, and will\n",
      " |      assume that the output dtype matches the input dtype.\n",
      " |      \n",
      " |      Args:\n",
      " |        input_signature: Single TensorSpec or nested structure of TensorSpec\n",
      " |          objects, describing a candidate input for the layer.\n",
      " |      \n",
      " |      Returns:\n",
      " |        Single TensorSpec or nested structure of TensorSpec objects,\n",
      " |          describing how the layer would transform the provided input.\n",
      " |      \n",
      " |      Raises:\n",
      " |        TypeError: If input_signature contains a non-TensorSpec object.\n",
      " |  \n",
      " |  count_params(self)\n",
      " |      Count the total number of scalars composing the weights.\n",
      " |      \n",
      " |      Returns:\n",
      " |          An integer count.\n",
      " |      \n",
      " |      Raises:\n",
      " |          ValueError: if the layer isn't yet built\n",
      " |            (in which case its weights aren't yet defined).\n",
      " |  \n",
      " |  finalize_state(self)\n",
      " |      Finalizes the layers state after updating layer weights.\n",
      " |      \n",
      " |      This function can be subclassed in a layer and will be called after\n",
      " |      updating a layer weights. It can be overridden to finalize any\n",
      " |      additional layer state after a weight update.\n",
      " |      \n",
      " |      This function will be called after weights of a layer have been restored\n",
      " |      from a loaded model.\n",
      " |  \n",
      " |  get_input_at(self, node_index)\n",
      " |      Retrieves the input tensor(s) of a layer at a given node.\n",
      " |      \n",
      " |      Args:\n",
      " |          node_index: Integer, index of the node\n",
      " |              from which to retrieve the attribute.\n",
      " |              E.g. `node_index=0` will correspond to the\n",
      " |              first input node of the layer.\n",
      " |      \n",
      " |      Returns:\n",
      " |          A tensor (or list of tensors if the layer has multiple inputs).\n",
      " |      \n",
      " |      Raises:\n",
      " |        RuntimeError: If called in Eager mode.\n",
      " |  \n",
      " |  get_input_mask_at(self, node_index)\n",
      " |      Retrieves the input mask tensor(s) of a layer at a given node.\n",
      " |      \n",
      " |      Args:\n",
      " |          node_index: Integer, index of the node\n",
      " |              from which to retrieve the attribute.\n",
      " |              E.g. `node_index=0` will correspond to the\n",
      " |              first time the layer was called.\n",
      " |      \n",
      " |      Returns:\n",
      " |          A mask tensor\n",
      " |          (or list of tensors if the layer has multiple inputs).\n",
      " |  \n",
      " |  get_input_shape_at(self, node_index)\n",
      " |      Retrieves the input shape(s) of a layer at a given node.\n",
      " |      \n",
      " |      Args:\n",
      " |          node_index: Integer, index of the node\n",
      " |              from which to retrieve the attribute.\n",
      " |              E.g. `node_index=0` will correspond to the\n",
      " |              first time the layer was called.\n",
      " |      \n",
      " |      Returns:\n",
      " |          A shape tuple\n",
      " |          (or list of shape tuples if the layer has multiple inputs).\n",
      " |      \n",
      " |      Raises:\n",
      " |        RuntimeError: If called in Eager mode.\n",
      " |  \n",
      " |  get_output_at(self, node_index)\n",
      " |      Retrieves the output tensor(s) of a layer at a given node.\n",
      " |      \n",
      " |      Args:\n",
      " |          node_index: Integer, index of the node\n",
      " |              from which to retrieve the attribute.\n",
      " |              E.g. `node_index=0` will correspond to the\n",
      " |              first output node of the layer.\n",
      " |      \n",
      " |      Returns:\n",
      " |          A tensor (or list of tensors if the layer has multiple outputs).\n",
      " |      \n",
      " |      Raises:\n",
      " |        RuntimeError: If called in Eager mode.\n",
      " |  \n",
      " |  get_output_mask_at(self, node_index)\n",
      " |      Retrieves the output mask tensor(s) of a layer at a given node.\n",
      " |      \n",
      " |      Args:\n",
      " |          node_index: Integer, index of the node\n",
      " |              from which to retrieve the attribute.\n",
      " |              E.g. `node_index=0` will correspond to the\n",
      " |              first time the layer was called.\n",
      " |      \n",
      " |      Returns:\n",
      " |          A mask tensor\n",
      " |          (or list of tensors if the layer has multiple outputs).\n",
      " |  \n",
      " |  get_output_shape_at(self, node_index)\n",
      " |      Retrieves the output shape(s) of a layer at a given node.\n",
      " |      \n",
      " |      Args:\n",
      " |          node_index: Integer, index of the node\n",
      " |              from which to retrieve the attribute.\n",
      " |              E.g. `node_index=0` will correspond to the\n",
      " |              first time the layer was called.\n",
      " |      \n",
      " |      Returns:\n",
      " |          A shape tuple\n",
      " |          (or list of shape tuples if the layer has multiple outputs).\n",
      " |      \n",
      " |      Raises:\n",
      " |        RuntimeError: If called in Eager mode.\n",
      " |  \n",
      " |  get_weights(self)\n",
      " |      Returns the current weights of the layer, as NumPy arrays.\n",
      " |      \n",
      " |      The weights of a layer represent the state of the layer. This function\n",
      " |      returns both trainable and non-trainable weight values associated with\n",
      " |      this layer as a list of NumPy arrays, which can in turn be used to load\n",
      " |      state into similarly parameterized layers.\n",
      " |      \n",
      " |      For example, a `Dense` layer returns a list of two values: the kernel\n",
      " |      matrix and the bias vector. These can be used to set the weights of\n",
      " |      another `Dense` layer:\n",
      " |      \n",
      " |      >>> layer_a = tf.keras.layers.Dense(1,\n",
      " |      ...   kernel_initializer=tf.constant_initializer(1.))\n",
      " |      >>> a_out = layer_a(tf.convert_to_tensor([[1., 2., 3.]]))\n",
      " |      >>> layer_a.get_weights()\n",
      " |      [array([[1.],\n",
      " |             [1.],\n",
      " |             [1.]], dtype=float32), array([0.], dtype=float32)]\n",
      " |      >>> layer_b = tf.keras.layers.Dense(1,\n",
      " |      ...   kernel_initializer=tf.constant_initializer(2.))\n",
      " |      >>> b_out = layer_b(tf.convert_to_tensor([[10., 20., 30.]]))\n",
      " |      >>> layer_b.get_weights()\n",
      " |      [array([[2.],\n",
      " |             [2.],\n",
      " |             [2.]], dtype=float32), array([0.], dtype=float32)]\n",
      " |      >>> layer_b.set_weights(layer_a.get_weights())\n",
      " |      >>> layer_b.get_weights()\n",
      " |      [array([[1.],\n",
      " |             [1.],\n",
      " |             [1.]], dtype=float32), array([0.], dtype=float32)]\n",
      " |      \n",
      " |      Returns:\n",
      " |          Weights values as a list of NumPy arrays.\n",
      " |  \n",
      " |  set_weights(self, weights)\n",
      " |      Sets the weights of the layer, from NumPy arrays.\n",
      " |      \n",
      " |      The weights of a layer represent the state of the layer. This function\n",
      " |      sets the weight values from numpy arrays. The weight values should be\n",
      " |      passed in the order they are created by the layer. Note that the layer's\n",
      " |      weights must be instantiated before calling this function, by calling\n",
      " |      the layer.\n",
      " |      \n",
      " |      For example, a `Dense` layer returns a list of two values: the kernel\n",
      " |      matrix and the bias vector. These can be used to set the weights of\n",
      " |      another `Dense` layer:\n",
      " |      \n",
      " |      >>> layer_a = tf.keras.layers.Dense(1,\n",
      " |      ...   kernel_initializer=tf.constant_initializer(1.))\n",
      " |      >>> a_out = layer_a(tf.convert_to_tensor([[1., 2., 3.]]))\n",
      " |      >>> layer_a.get_weights()\n",
      " |      [array([[1.],\n",
      " |             [1.],\n",
      " |             [1.]], dtype=float32), array([0.], dtype=float32)]\n",
      " |      >>> layer_b = tf.keras.layers.Dense(1,\n",
      " |      ...   kernel_initializer=tf.constant_initializer(2.))\n",
      " |      >>> b_out = layer_b(tf.convert_to_tensor([[10., 20., 30.]]))\n",
      " |      >>> layer_b.get_weights()\n",
      " |      [array([[2.],\n",
      " |             [2.],\n",
      " |             [2.]], dtype=float32), array([0.], dtype=float32)]\n",
      " |      >>> layer_b.set_weights(layer_a.get_weights())\n",
      " |      >>> layer_b.get_weights()\n",
      " |      [array([[1.],\n",
      " |             [1.],\n",
      " |             [1.]], dtype=float32), array([0.], dtype=float32)]\n",
      " |      \n",
      " |      Args:\n",
      " |        weights: a list of NumPy arrays. The number\n",
      " |          of arrays and their shape must match\n",
      " |          number of the dimensions of the weights\n",
      " |          of the layer (i.e. it should match the\n",
      " |          output of `get_weights`).\n",
      " |      \n",
      " |      Raises:\n",
      " |        ValueError: If the provided weights list does not match the\n",
      " |          layer's specifications.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Class methods inherited from keras.engine.base_layer.Layer:\n",
      " |  \n",
      " |  from_config(config) from builtins.type\n",
      " |      Creates a layer from its config.\n",
      " |      \n",
      " |      This method is the reverse of `get_config`,\n",
      " |      capable of instantiating the same layer from the config\n",
      " |      dictionary. It does not handle layer connectivity\n",
      " |      (handled by Network), nor weights (handled by `set_weights`).\n",
      " |      \n",
      " |      Args:\n",
      " |          config: A Python dictionary, typically the\n",
      " |              output of get_config.\n",
      " |      \n",
      " |      Returns:\n",
      " |          A layer instance.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Static methods inherited from keras.engine.base_layer.Layer:\n",
      " |  \n",
      " |  __new__(cls, *args, **kwargs)\n",
      " |      Create and return a new object.  See help(type) for accurate signature.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Readonly properties inherited from keras.engine.base_layer.Layer:\n",
      " |  \n",
      " |  compute_dtype\n",
      " |      The dtype of the layer's computations.\n",
      " |      \n",
      " |      This is equivalent to `Layer.dtype_policy.compute_dtype`. Unless\n",
      " |      mixed precision is used, this is the same as `Layer.dtype`, the dtype of\n",
      " |      the weights.\n",
      " |      \n",
      " |      Layers automatically cast their inputs to the compute dtype, which\n",
      " |      causes computations and the output to be in the compute dtype as well.\n",
      " |      This is done by the base Layer class in `Layer.__call__`, so you do not\n",
      " |      have to insert these casts if implementing your own layer.\n",
      " |      \n",
      " |      Layers often perform certain internal computations in higher precision\n",
      " |      when `compute_dtype` is float16 or bfloat16 for numeric stability. The\n",
      " |      output will still typically be float16 or bfloat16 in such cases.\n",
      " |      \n",
      " |      Returns:\n",
      " |        The layer's compute dtype.\n",
      " |  \n",
      " |  dtype\n",
      " |      The dtype of the layer weights.\n",
      " |      \n",
      " |      This is equivalent to `Layer.dtype_policy.variable_dtype`. Unless\n",
      " |      mixed precision is used, this is the same as `Layer.compute_dtype`, the\n",
      " |      dtype of the layer's computations.\n",
      " |  \n",
      " |  dtype_policy\n",
      " |      The dtype policy associated with this layer.\n",
      " |      \n",
      " |      This is an instance of a `tf.keras.mixed_precision.Policy`.\n",
      " |  \n",
      " |  dynamic\n",
      " |      Whether the layer is dynamic (eager-only); set in the constructor.\n",
      " |  \n",
      " |  inbound_nodes\n",
      " |      Return Functional API nodes upstream of this layer.\n",
      " |  \n",
      " |  input\n",
      " |      Retrieves the input tensor(s) of a layer.\n",
      " |      \n",
      " |      Only applicable if the layer has exactly one input,\n",
      " |      i.e. if it is connected to one incoming layer.\n",
      " |      \n",
      " |      Returns:\n",
      " |          Input tensor or list of input tensors.\n",
      " |      \n",
      " |      Raises:\n",
      " |        RuntimeError: If called in Eager mode.\n",
      " |        AttributeError: If no inbound nodes are found.\n",
      " |  \n",
      " |  input_mask\n",
      " |      Retrieves the input mask tensor(s) of a layer.\n",
      " |      \n",
      " |      Only applicable if the layer has exactly one inbound node,\n",
      " |      i.e. if it is connected to one incoming layer.\n",
      " |      \n",
      " |      Returns:\n",
      " |          Input mask tensor (potentially None) or list of input\n",
      " |          mask tensors.\n",
      " |      \n",
      " |      Raises:\n",
      " |          AttributeError: if the layer is connected to\n",
      " |          more than one incoming layers.\n",
      " |  \n",
      " |  input_shape\n",
      " |      Retrieves the input shape(s) of a layer.\n",
      " |      \n",
      " |      Only applicable if the layer has exactly one input,\n",
      " |      i.e. if it is connected to one incoming layer, or if all inputs\n",
      " |      have the same shape.\n",
      " |      \n",
      " |      Returns:\n",
      " |          Input shape, as an integer shape tuple\n",
      " |          (or list of shape tuples, one tuple per input tensor).\n",
      " |      \n",
      " |      Raises:\n",
      " |          AttributeError: if the layer has no defined input_shape.\n",
      " |          RuntimeError: if called in Eager mode.\n",
      " |  \n",
      " |  losses\n",
      " |      List of losses added using the `add_loss()` API.\n",
      " |      \n",
      " |      Variable regularization tensors are created when this property is\n",
      " |      accessed, so it is eager safe: accessing `losses` under a\n",
      " |      `tf.GradientTape` will propagate gradients back to the corresponding\n",
      " |      variables.\n",
      " |      \n",
      " |      Examples:\n",
      " |      \n",
      " |      >>> class MyLayer(tf.keras.layers.Layer):\n",
      " |      ...   def call(self, inputs):\n",
      " |      ...     self.add_loss(tf.abs(tf.reduce_mean(inputs)))\n",
      " |      ...     return inputs\n",
      " |      >>> l = MyLayer()\n",
      " |      >>> l(np.ones((10, 1)))\n",
      " |      >>> l.losses\n",
      " |      [1.0]\n",
      " |      \n",
      " |      >>> inputs = tf.keras.Input(shape=(10,))\n",
      " |      >>> x = tf.keras.layers.Dense(10)(inputs)\n",
      " |      >>> outputs = tf.keras.layers.Dense(1)(x)\n",
      " |      >>> model = tf.keras.Model(inputs, outputs)\n",
      " |      >>> # Activity regularization.\n",
      " |      >>> len(model.losses)\n",
      " |      0\n",
      " |      >>> model.add_loss(tf.abs(tf.reduce_mean(x)))\n",
      " |      >>> len(model.losses)\n",
      " |      1\n",
      " |      \n",
      " |      >>> inputs = tf.keras.Input(shape=(10,))\n",
      " |      >>> d = tf.keras.layers.Dense(10, kernel_initializer='ones')\n",
      " |      >>> x = d(inputs)\n",
      " |      >>> outputs = tf.keras.layers.Dense(1)(x)\n",
      " |      >>> model = tf.keras.Model(inputs, outputs)\n",
      " |      >>> # Weight regularization.\n",
      " |      >>> model.add_loss(lambda: tf.reduce_mean(d.kernel))\n",
      " |      >>> model.losses\n",
      " |      [<tf.Tensor: shape=(), dtype=float32, numpy=1.0>]\n",
      " |      \n",
      " |      Returns:\n",
      " |        A list of tensors.\n",
      " |  \n",
      " |  metrics\n",
      " |      List of metrics added using the `add_metric()` API.\n",
      " |      \n",
      " |      Example:\n",
      " |      \n",
      " |      >>> input = tf.keras.layers.Input(shape=(3,))\n",
      " |      >>> d = tf.keras.layers.Dense(2)\n",
      " |      >>> output = d(input)\n",
      " |      >>> d.add_metric(tf.reduce_max(output), name='max')\n",
      " |      >>> d.add_metric(tf.reduce_min(output), name='min')\n",
      " |      >>> [m.name for m in d.metrics]\n",
      " |      ['max', 'min']\n",
      " |      \n",
      " |      Returns:\n",
      " |        A list of `Metric` objects.\n",
      " |  \n",
      " |  name\n",
      " |      Name of the layer (string), set in the constructor.\n",
      " |  \n",
      " |  non_trainable_variables\n",
      " |      Sequence of non-trainable variables owned by this module and its submodules.\n",
      " |      \n",
      " |      Note: this method uses reflection to find variables on the current instance\n",
      " |      and submodules. For performance reasons you may wish to cache the result\n",
      " |      of calling this method if you don't expect the return value to change.\n",
      " |      \n",
      " |      Returns:\n",
      " |        A sequence of variables for the current module (sorted by attribute\n",
      " |        name) followed by variables from all submodules recursively (breadth\n",
      " |        first).\n",
      " |  \n",
      " |  non_trainable_weights\n",
      " |      List of all non-trainable weights tracked by this layer.\n",
      " |      \n",
      " |      Non-trainable weights are *not* updated during training. They are\n",
      " |      expected to be updated manually in `call()`.\n",
      " |      \n",
      " |      Returns:\n",
      " |        A list of non-trainable variables.\n",
      " |  \n",
      " |  outbound_nodes\n",
      " |      Return Functional API nodes downstream of this layer.\n",
      " |  \n",
      " |  output\n",
      " |      Retrieves the output tensor(s) of a layer.\n",
      " |      \n",
      " |      Only applicable if the layer has exactly one output,\n",
      " |      i.e. if it is connected to one incoming layer.\n",
      " |      \n",
      " |      Returns:\n",
      " |        Output tensor or list of output tensors.\n",
      " |      \n",
      " |      Raises:\n",
      " |        AttributeError: if the layer is connected to more than one incoming\n",
      " |          layers.\n",
      " |        RuntimeError: if called in Eager mode.\n",
      " |  \n",
      " |  output_mask\n",
      " |      Retrieves the output mask tensor(s) of a layer.\n",
      " |      \n",
      " |      Only applicable if the layer has exactly one inbound node,\n",
      " |      i.e. if it is connected to one incoming layer.\n",
      " |      \n",
      " |      Returns:\n",
      " |          Output mask tensor (potentially None) or list of output\n",
      " |          mask tensors.\n",
      " |      \n",
      " |      Raises:\n",
      " |          AttributeError: if the layer is connected to\n",
      " |          more than one incoming layers.\n",
      " |  \n",
      " |  output_shape\n",
      " |      Retrieves the output shape(s) of a layer.\n",
      " |      \n",
      " |      Only applicable if the layer has one output,\n",
      " |      or if all outputs have the same shape.\n",
      " |      \n",
      " |      Returns:\n",
      " |          Output shape, as an integer shape tuple\n",
      " |          (or list of shape tuples, one tuple per output tensor).\n",
      " |      \n",
      " |      Raises:\n",
      " |          AttributeError: if the layer has no defined output shape.\n",
      " |          RuntimeError: if called in Eager mode.\n",
      " |  \n",
      " |  trainable_variables\n",
      " |      Sequence of trainable variables owned by this module and its submodules.\n",
      " |      \n",
      " |      Note: this method uses reflection to find variables on the current instance\n",
      " |      and submodules. For performance reasons you may wish to cache the result\n",
      " |      of calling this method if you don't expect the return value to change.\n",
      " |      \n",
      " |      Returns:\n",
      " |        A sequence of variables for the current module (sorted by attribute\n",
      " |        name) followed by variables from all submodules recursively (breadth\n",
      " |        first).\n",
      " |  \n",
      " |  trainable_weights\n",
      " |      List of all trainable weights tracked by this layer.\n",
      " |      \n",
      " |      Trainable weights are updated via gradient descent during training.\n",
      " |      \n",
      " |      Returns:\n",
      " |        A list of trainable variables.\n",
      " |  \n",
      " |  updates\n",
      " |  \n",
      " |  variable_dtype\n",
      " |      Alias of `Layer.dtype`, the dtype of the weights.\n",
      " |  \n",
      " |  variables\n",
      " |      Returns the list of all layer variables/weights.\n",
      " |      \n",
      " |      Alias of `self.weights`.\n",
      " |      \n",
      " |      Note: This will not track the weights of nested `tf.Modules` that are\n",
      " |      not themselves Keras layers.\n",
      " |      \n",
      " |      Returns:\n",
      " |        A list of variables.\n",
      " |  \n",
      " |  weights\n",
      " |      Returns the list of all layer variables/weights.\n",
      " |      \n",
      " |      Returns:\n",
      " |        A list of variables.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Data descriptors inherited from keras.engine.base_layer.Layer:\n",
      " |  \n",
      " |  activity_regularizer\n",
      " |      Optional regularizer function for the output of this layer.\n",
      " |  \n",
      " |  input_spec\n",
      " |      `InputSpec` instance(s) describing the input format for this layer.\n",
      " |      \n",
      " |      When you create a layer subclass, you can set `self.input_spec` to\n",
      " |      enable the layer to run input compatibility checks when it is called.\n",
      " |      Consider a `Conv2D` layer: it can only be called on a single input\n",
      " |      tensor of rank 4. As such, you can set, in `__init__()`:\n",
      " |      \n",
      " |      ```python\n",
      " |      self.input_spec = tf.keras.layers.InputSpec(ndim=4)\n",
      " |      ```\n",
      " |      \n",
      " |      Now, if you try to call the layer on an input that isn't rank 4\n",
      " |      (for instance, an input of shape `(2,)`, it will raise a\n",
      " |      nicely-formatted error:\n",
      " |      \n",
      " |      ```\n",
      " |      ValueError: Input 0 of layer conv2d is incompatible with the layer:\n",
      " |      expected ndim=4, found ndim=1. Full shape received: [2]\n",
      " |      ```\n",
      " |      \n",
      " |      Input checks that can be specified via `input_spec` include:\n",
      " |      - Structure (e.g. a single input, a list of 2 inputs, etc)\n",
      " |      - Shape\n",
      " |      - Rank (ndim)\n",
      " |      - Dtype\n",
      " |      \n",
      " |      For more information, see `tf.keras.layers.InputSpec`.\n",
      " |      \n",
      " |      Returns:\n",
      " |        A `tf.keras.layers.InputSpec` instance, or nested structure thereof.\n",
      " |  \n",
      " |  stateful\n",
      " |  \n",
      " |  supports_masking\n",
      " |      Whether this layer supports computing a mask using `compute_mask`.\n",
      " |  \n",
      " |  trainable\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Class methods inherited from tensorflow.python.module.module.Module:\n",
      " |  \n",
      " |  with_name_scope(method) from builtins.type\n",
      " |      Decorator to automatically enter the module name scope.\n",
      " |      \n",
      " |      >>> class MyModule(tf.Module):\n",
      " |      ...   @tf.Module.with_name_scope\n",
      " |      ...   def __call__(self, x):\n",
      " |      ...     if not hasattr(self, 'w'):\n",
      " |      ...       self.w = tf.Variable(tf.random.normal([x.shape[1], 3]))\n",
      " |      ...     return tf.matmul(x, self.w)\n",
      " |      \n",
      " |      Using the above module would produce `tf.Variable`s and `tf.Tensor`s whose\n",
      " |      names included the module name:\n",
      " |      \n",
      " |      >>> mod = MyModule()\n",
      " |      >>> mod(tf.ones([1, 2]))\n",
      " |      <tf.Tensor: shape=(1, 3), dtype=float32, numpy=..., dtype=float32)>\n",
      " |      >>> mod.w\n",
      " |      <tf.Variable 'my_module/Variable:0' shape=(2, 3) dtype=float32,\n",
      " |      numpy=..., dtype=float32)>\n",
      " |      \n",
      " |      Args:\n",
      " |        method: The method to wrap.\n",
      " |      \n",
      " |      Returns:\n",
      " |        The original method wrapped such that it enters the module's name scope.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Readonly properties inherited from tensorflow.python.module.module.Module:\n",
      " |  \n",
      " |  name_scope\n",
      " |      Returns a `tf.name_scope` instance for this class.\n",
      " |  \n",
      " |  submodules\n",
      " |      Sequence of all sub-modules.\n",
      " |      \n",
      " |      Submodules are modules which are properties of this module, or found as\n",
      " |      properties of modules which are properties of this module (and so on).\n",
      " |      \n",
      " |      >>> a = tf.Module()\n",
      " |      >>> b = tf.Module()\n",
      " |      >>> c = tf.Module()\n",
      " |      >>> a.b = b\n",
      " |      >>> b.c = c\n",
      " |      >>> list(a.submodules) == [b, c]\n",
      " |      True\n",
      " |      >>> list(b.submodules) == [c]\n",
      " |      True\n",
      " |      >>> list(c.submodules) == []\n",
      " |      True\n",
      " |      \n",
      " |      Returns:\n",
      " |        A sequence of all submodules.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Data descriptors inherited from tensorflow.python.trackable.base.Trackable:\n",
      " |  \n",
      " |  __dict__\n",
      " |      dictionary for instance variables (if defined)\n",
      " |  \n",
      " |  __weakref__\n",
      " |      list of weak references to the object (if defined)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "help(Dense)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "6d12312f",
   "metadata": {},
   "outputs": [],
   "source": [
    "#help() includes several helpful examples on how to construct models."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "7eaaefd0",
   "metadata": {},
   "outputs": [],
   "source": [
    "#two ways of creating a keras based model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "id": "a51acdfb",
   "metadata": {},
   "outputs": [],
   "source": [
    "#first way:\n",
    "\n",
    "model = Sequential([Dense(4,activation='relu'),\n",
    "                   Dense(2,activation='relu'),\n",
    "                   Dense(1)])\n",
    "\n",
    "#Dense means \"Just your regular densely-connected NN layer.\" If \n",
    "#something is densely connected, it's a normal feedforward network\n",
    "#where every neuron is connected to every other neuron in the next \n",
    "#layer\n",
    "\n",
    "#The above is 4 neurons densely connected and the activation is\n",
    "#rectified linear unit.\n",
    "#A second layer with 2 neurons is then added also with RELU,\n",
    "#and then finally an output layer with 1 unit is added"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 196,
   "id": "f1448571",
   "metadata": {},
   "outputs": [],
   "source": [
    "#second way (preferred way):\n",
    "\n",
    "model = Sequential()\n",
    "\n",
    "model.add(Dense(4,activation='relu'))\n",
    "model.add(Dense(4,activation='relu'))\n",
    "model.add(Dense(4,activation='relu'))\n",
    "model.add(Dense(1))\n",
    "\n",
    "model.compile(optimizer='rmsprop',loss='mse')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "d7f57b5c",
   "metadata": {},
   "outputs": [],
   "source": [
    "#doing this in separate lines is more convenient, as the model can\n",
    "#be more easily edited. You can comment out rows and then run the model\n",
    "#again quickly."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "6ec0b7f4",
   "metadata": {},
   "outputs": [],
   "source": [
    "#final output layer is very important and will be determined by your\n",
    "#actual data and what you are trying to predict. In this case\n",
    "#we are trying to predict a single price so we want the final layer\n",
    "#to be a single neuron.\n",
    "\n",
    "#final output is then measured against the true price using a loss \n",
    "#function, which is where the final line (compiling) comes into play.\n",
    "\n",
    "\n",
    "#Lecture notebook contains guide on choosing an optimizer and loss\n",
    "#optimizer asks how you want to perform gradient descent (we want to\n",
    "#use Adam typically). \n",
    "#loss parameter - for a regression proglem, you use mean squared error.\n",
    "#this makes sense because we are taking the mean squared error of our \n",
    "#predictions, comparing against the true values, and then trying to \n",
    "#minimize that through the optimizer call."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 197,
   "id": "e78c22af",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 257172.7500\n",
      "Epoch 2/250\n",
      "22/22 [==============================] - 0s 928us/step - loss: 256957.8281\n",
      "Epoch 3/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 256805.5469\n",
      "Epoch 4/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 256697.7344\n",
      "Epoch 5/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 256625.5781\n",
      "Epoch 6/250\n",
      "22/22 [==============================] - 0s 998us/step - loss: 256579.9531\n",
      "Epoch 7/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 256542.0781\n",
      "Epoch 8/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 256500.8750\n",
      "Epoch 9/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 256451.7969\n",
      "Epoch 10/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 256393.5938\n",
      "Epoch 11/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 256325.1406\n",
      "Epoch 12/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 256245.3438\n",
      "Epoch 13/250\n",
      "22/22 [==============================] - 0s 983us/step - loss: 256152.8750\n",
      "Epoch 14/250\n",
      "22/22 [==============================] - 0s 977us/step - loss: 256045.4688\n",
      "Epoch 15/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 255922.5625\n",
      "Epoch 16/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 255783.5000\n",
      "Epoch 17/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 255626.2188\n",
      "Epoch 18/250\n",
      "22/22 [==============================] - 0s 981us/step - loss: 255449.0469\n",
      "Epoch 19/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 255250.7031\n",
      "Epoch 20/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 255028.3125\n",
      "Epoch 21/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 254781.4688\n",
      "Epoch 22/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 254505.4688\n",
      "Epoch 23/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 254200.9531\n",
      "Epoch 24/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 253864.5781\n",
      "Epoch 25/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 253495.1719\n",
      "Epoch 26/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 253087.5625\n",
      "Epoch 27/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 252644.8438\n",
      "Epoch 28/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 252161.0938\n",
      "Epoch 29/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 251634.9062\n",
      "Epoch 30/250\n",
      "22/22 [==============================] - 0s 961us/step - loss: 251063.6406\n",
      "Epoch 31/250\n",
      "22/22 [==============================] - 0s 940us/step - loss: 250442.0156\n",
      "Epoch 32/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 249774.7656\n",
      "Epoch 33/250\n",
      "22/22 [==============================] - 0s 958us/step - loss: 249049.2969\n",
      "Epoch 34/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 248272.7500\n",
      "Epoch 35/250\n",
      "22/22 [==============================] - 0s 914us/step - loss: 247437.6406\n",
      "Epoch 36/250\n",
      "22/22 [==============================] - 0s 919us/step - loss: 246540.4531\n",
      "Epoch 37/250\n",
      "22/22 [==============================] - 0s 980us/step - loss: 245580.0625\n",
      "Epoch 38/250\n",
      "22/22 [==============================] - 0s 903us/step - loss: 244554.5781\n",
      "Epoch 39/250\n",
      "22/22 [==============================] - 0s 907us/step - loss: 243465.0781\n",
      "Epoch 40/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 242301.2969\n",
      "Epoch 41/250\n",
      "22/22 [==============================] - 0s 968us/step - loss: 241060.1406\n",
      "Epoch 42/250\n",
      "22/22 [==============================] - 0s 966us/step - loss: 239750.3594\n",
      "Epoch 43/250\n",
      "22/22 [==============================] - 0s 985us/step - loss: 238358.7812\n",
      "Epoch 44/250\n",
      "22/22 [==============================] - 0s 996us/step - loss: 236883.7500\n",
      "Epoch 45/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 235320.1875\n",
      "Epoch 46/250\n",
      "22/22 [==============================] - 0s 987us/step - loss: 233669.1406\n",
      "Epoch 47/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 231933.0781\n",
      "Epoch 48/250\n",
      "22/22 [==============================] - 0s 978us/step - loss: 230109.6250\n",
      "Epoch 49/250\n",
      "22/22 [==============================] - 0s 965us/step - loss: 228200.5781\n",
      "Epoch 50/250\n",
      "22/22 [==============================] - 0s 886us/step - loss: 226175.8438\n",
      "Epoch 51/250\n",
      "22/22 [==============================] - 0s 968us/step - loss: 224055.5156\n",
      "Epoch 52/250\n",
      "22/22 [==============================] - 0s 950us/step - loss: 221830.1250\n",
      "Epoch 53/250\n",
      "22/22 [==============================] - 0s 977us/step - loss: 219510.6562\n",
      "Epoch 54/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 217076.6875\n",
      "Epoch 55/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 214533.0469\n",
      "Epoch 56/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 211898.2188\n",
      "Epoch 57/250\n",
      "22/22 [==============================] - 0s 905us/step - loss: 209145.0469\n",
      "Epoch 58/250\n",
      "22/22 [==============================] - 0s 947us/step - loss: 206276.6562\n",
      "Epoch 59/250\n",
      "22/22 [==============================] - 0s 961us/step - loss: 203301.3438\n",
      "Epoch 60/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 200201.0781\n",
      "Epoch 61/250\n",
      "22/22 [==============================] - 0s 958us/step - loss: 196975.9375\n",
      "Epoch 62/250\n",
      "22/22 [==============================] - 0s 982us/step - loss: 193642.0156\n",
      "Epoch 63/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 190200.8906\n",
      "Epoch 64/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 186643.0469\n",
      "Epoch 65/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 182981.5938\n",
      "Epoch 66/250\n",
      "22/22 [==============================] - 0s 985us/step - loss: 179175.0000\n",
      "Epoch 67/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 175279.6250\n",
      "Epoch 68/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 171249.2500\n",
      "Epoch 69/250\n",
      "22/22 [==============================] - 0s 979us/step - loss: 167112.2031\n",
      "Epoch 70/250\n",
      "22/22 [==============================] - 0s 989us/step - loss: 162875.6875\n",
      "Epoch 71/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 158526.1094\n",
      "Epoch 72/250\n",
      "22/22 [==============================] - 0s 932us/step - loss: 154070.1875\n",
      "Epoch 73/250\n",
      "22/22 [==============================] - 0s 915us/step - loss: 149510.7031\n",
      "Epoch 74/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 144849.2031\n",
      "Epoch 75/250\n",
      "22/22 [==============================] - 0s 964us/step - loss: 140083.8125\n",
      "Epoch 76/250\n",
      "22/22 [==============================] - 0s 924us/step - loss: 135241.5938\n",
      "Epoch 77/250\n",
      "22/22 [==============================] - 0s 983us/step - loss: 130326.4766\n",
      "Epoch 78/250\n",
      "22/22 [==============================] - 0s 925us/step - loss: 125329.2422\n",
      "Epoch 79/250\n",
      "22/22 [==============================] - 0s 918us/step - loss: 120236.6406\n",
      "Epoch 80/250\n",
      "22/22 [==============================] - 0s 870us/step - loss: 115127.2031\n",
      "Epoch 81/250\n",
      "22/22 [==============================] - 0s 938us/step - loss: 109932.1953\n",
      "Epoch 82/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 104659.3359\n",
      "Epoch 83/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 99422.9141\n",
      "Epoch 84/250\n",
      "22/22 [==============================] - 0s 920us/step - loss: 94098.0625\n",
      "Epoch 85/250\n",
      "22/22 [==============================] - 0s 987us/step - loss: 88753.5547\n",
      "Epoch 86/250\n",
      "22/22 [==============================] - 0s 979us/step - loss: 83435.7031\n",
      "Epoch 87/250\n",
      "22/22 [==============================] - 0s 924us/step - loss: 78118.0703\n",
      "Epoch 88/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 72826.3750\n",
      "Epoch 89/250\n",
      "22/22 [==============================] - 0s 994us/step - loss: 67554.7031\n",
      "Epoch 90/250\n",
      "22/22 [==============================] - 0s 967us/step - loss: 62347.6016\n",
      "Epoch 91/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 57207.5703\n",
      "Epoch 92/250\n",
      "22/22 [==============================] - 0s 974us/step - loss: 52137.1445\n",
      "Epoch 93/250\n",
      "22/22 [==============================] - 0s 904us/step - loss: 47198.3906\n",
      "Epoch 94/250\n",
      "22/22 [==============================] - 0s 983us/step - loss: 42363.5938\n",
      "Epoch 95/250\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "22/22 [==============================] - 0s 949us/step - loss: 37679.5703\n",
      "Epoch 96/250\n",
      "22/22 [==============================] - 0s 904us/step - loss: 33181.8633\n",
      "Epoch 97/250\n",
      "22/22 [==============================] - 0s 964us/step - loss: 28912.0078\n",
      "Epoch 98/250\n",
      "22/22 [==============================] - 0s 899us/step - loss: 24797.3809\n",
      "Epoch 99/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 20914.8105\n",
      "Epoch 100/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 17333.9199\n",
      "Epoch 101/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 14059.2871\n",
      "Epoch 102/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 11109.1309\n",
      "Epoch 103/250\n",
      "22/22 [==============================] - 0s 992us/step - loss: 8535.5947\n",
      "Epoch 104/250\n",
      "22/22 [==============================] - 0s 999us/step - loss: 6312.2798\n",
      "Epoch 105/250\n",
      "22/22 [==============================] - 0s 980us/step - loss: 4466.5386\n",
      "Epoch 106/250\n",
      "22/22 [==============================] - 0s 1000us/step - loss: 3070.7649\n",
      "Epoch 107/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 2123.0322\n",
      "Epoch 108/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 1586.2778\n",
      "Epoch 109/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 1374.7714\n",
      "Epoch 110/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 1326.2599\n",
      "Epoch 111/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 1307.0638\n",
      "Epoch 112/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 1282.0702\n",
      "Epoch 113/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 1261.2817\n",
      "Epoch 114/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 1237.9906\n",
      "Epoch 115/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 1216.8477\n",
      "Epoch 116/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 1193.5648\n",
      "Epoch 117/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 1167.2208\n",
      "Epoch 118/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 1148.0934\n",
      "Epoch 119/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 1122.9673\n",
      "Epoch 120/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 1104.1298\n",
      "Epoch 121/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 1084.6727\n",
      "Epoch 122/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 1066.1912\n",
      "Epoch 123/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 1044.0641\n",
      "Epoch 124/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 1021.3212\n",
      "Epoch 125/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 999.8380\n",
      "Epoch 126/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 978.3920\n",
      "Epoch 127/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 957.3561\n",
      "Epoch 128/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 940.1335\n",
      "Epoch 129/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 920.5853\n",
      "Epoch 130/250\n",
      "22/22 [==============================] - 0s 1000us/step - loss: 900.5737\n",
      "Epoch 131/250\n",
      "22/22 [==============================] - 0s 966us/step - loss: 878.5322\n",
      "Epoch 132/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 862.3939\n",
      "Epoch 133/250\n",
      "22/22 [==============================] - 0s 961us/step - loss: 844.4258\n",
      "Epoch 134/250\n",
      "22/22 [==============================] - 0s 977us/step - loss: 825.6877\n",
      "Epoch 135/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 809.2900\n",
      "Epoch 136/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 789.0845\n",
      "Epoch 137/250\n",
      "22/22 [==============================] - 0s 985us/step - loss: 769.8428\n",
      "Epoch 138/250\n",
      "22/22 [==============================] - 0s 969us/step - loss: 750.2626\n",
      "Epoch 139/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 731.3394\n",
      "Epoch 140/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 714.7363\n",
      "Epoch 141/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 697.5771\n",
      "Epoch 142/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 680.2286\n",
      "Epoch 143/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 662.5814\n",
      "Epoch 144/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 646.1339\n",
      "Epoch 145/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 627.4380\n",
      "Epoch 146/250\n",
      "22/22 [==============================] - 0s 998us/step - loss: 610.7582\n",
      "Epoch 147/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 594.2882\n",
      "Epoch 148/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 579.0416\n",
      "Epoch 149/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 563.3976\n",
      "Epoch 150/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 546.7811\n",
      "Epoch 151/250\n",
      "22/22 [==============================] - 0s 993us/step - loss: 530.4022\n",
      "Epoch 152/250\n",
      "22/22 [==============================] - 0s 996us/step - loss: 514.3455\n",
      "Epoch 153/250\n",
      "22/22 [==============================] - 0s 983us/step - loss: 501.4859\n",
      "Epoch 154/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 486.5758\n",
      "Epoch 155/250\n",
      "22/22 [==============================] - 0s 941us/step - loss: 471.6917\n",
      "Epoch 156/250\n",
      "22/22 [==============================] - 0s 912us/step - loss: 457.5762\n",
      "Epoch 157/250\n",
      "22/22 [==============================] - 0s 869us/step - loss: 444.0784\n",
      "Epoch 158/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 431.6013\n",
      "Epoch 159/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 418.1478\n",
      "Epoch 160/250\n",
      "22/22 [==============================] - 0s 975us/step - loss: 405.9102\n",
      "Epoch 161/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 392.3878\n",
      "Epoch 162/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 380.0342\n",
      "Epoch 163/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 366.4429\n",
      "Epoch 164/250\n",
      "22/22 [==============================] - 0s 996us/step - loss: 353.4259\n",
      "Epoch 165/250\n",
      "22/22 [==============================] - 0s 936us/step - loss: 339.7578\n",
      "Epoch 166/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 328.3154\n",
      "Epoch 167/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 314.9561\n",
      "Epoch 168/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 303.3364\n",
      "Epoch 169/250\n",
      "22/22 [==============================] - 0s 934us/step - loss: 291.5527\n",
      "Epoch 170/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 279.7840\n",
      "Epoch 171/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 269.8734\n",
      "Epoch 172/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 258.1868\n",
      "Epoch 173/250\n",
      "22/22 [==============================] - 0s 938us/step - loss: 246.7487\n",
      "Epoch 174/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 236.3708\n",
      "Epoch 175/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 226.3752\n",
      "Epoch 176/250\n",
      "22/22 [==============================] - 0s 912us/step - loss: 215.7621\n",
      "Epoch 177/250\n",
      "22/22 [==============================] - 0s 988us/step - loss: 205.5177\n",
      "Epoch 178/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 197.1566\n",
      "Epoch 179/250\n",
      "22/22 [==============================] - 0s 912us/step - loss: 188.4902\n",
      "Epoch 180/250\n",
      "22/22 [==============================] - 0s 917us/step - loss: 179.7245\n",
      "Epoch 181/250\n",
      "22/22 [==============================] - 0s 978us/step - loss: 171.7323\n",
      "Epoch 182/250\n",
      "22/22 [==============================] - 0s 962us/step - loss: 162.9580\n",
      "Epoch 183/250\n",
      "22/22 [==============================] - 0s 947us/step - loss: 153.7692\n",
      "Epoch 184/250\n",
      "22/22 [==============================] - 0s 989us/step - loss: 145.9915\n",
      "Epoch 185/250\n",
      "22/22 [==============================] - 0s 955us/step - loss: 138.4485\n",
      "Epoch 186/250\n",
      "22/22 [==============================] - 0s 930us/step - loss: 131.6926\n",
      "Epoch 187/250\n",
      "22/22 [==============================] - 0s 956us/step - loss: 124.2791\n",
      "Epoch 188/250\n",
      "22/22 [==============================] - 0s 871us/step - loss: 117.7777\n",
      "Epoch 189/250\n",
      "22/22 [==============================] - 0s 910us/step - loss: 111.2060\n",
      "Epoch 190/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 104.8933\n",
      "Epoch 191/250\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "22/22 [==============================] - 0s 911us/step - loss: 99.5571\n",
      "Epoch 192/250\n",
      "22/22 [==============================] - 0s 898us/step - loss: 93.0576\n",
      "Epoch 193/250\n",
      "22/22 [==============================] - 0s 911us/step - loss: 87.3499\n",
      "Epoch 194/250\n",
      "22/22 [==============================] - 0s 845us/step - loss: 82.8014\n",
      "Epoch 195/250\n",
      "22/22 [==============================] - 0s 989us/step - loss: 77.9241\n",
      "Epoch 196/250\n",
      "22/22 [==============================] - 0s 989us/step - loss: 73.5760\n",
      "Epoch 197/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 69.2674\n",
      "Epoch 198/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 65.6718\n",
      "Epoch 199/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 62.3063\n",
      "Epoch 200/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 59.0242\n",
      "Epoch 201/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 55.2160\n",
      "Epoch 202/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 51.7484\n",
      "Epoch 203/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 48.8784\n",
      "Epoch 204/250\n",
      "22/22 [==============================] - 0s 990us/step - loss: 46.5039\n",
      "Epoch 205/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 43.9487\n",
      "Epoch 206/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 42.1710\n",
      "Epoch 207/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 39.6639\n",
      "Epoch 208/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 37.2819\n",
      "Epoch 209/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 35.6165\n",
      "Epoch 210/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 34.0394\n",
      "Epoch 211/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 32.5975\n",
      "Epoch 212/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 31.5115\n",
      "Epoch 213/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 30.7565\n",
      "Epoch 214/250\n",
      "22/22 [==============================] - 0s 995us/step - loss: 30.3762\n",
      "Epoch 215/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 29.2542\n",
      "Epoch 216/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 28.6641\n",
      "Epoch 217/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 28.1663\n",
      "Epoch 218/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 27.5853\n",
      "Epoch 219/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 27.3538\n",
      "Epoch 220/250\n",
      "22/22 [==============================] - 0s 956us/step - loss: 26.6515\n",
      "Epoch 221/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 26.4669\n",
      "Epoch 222/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 25.9861\n",
      "Epoch 223/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 25.6929\n",
      "Epoch 224/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 25.6600\n",
      "Epoch 225/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 24.8798\n",
      "Epoch 226/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 25.3776\n",
      "Epoch 227/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 25.0932\n",
      "Epoch 228/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 24.9343\n",
      "Epoch 229/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 24.8837\n",
      "Epoch 230/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 24.5024\n",
      "Epoch 231/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 24.5821\n",
      "Epoch 232/250\n",
      "22/22 [==============================] - 0s 996us/step - loss: 24.7274\n",
      "Epoch 233/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 24.4878\n",
      "Epoch 234/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 24.0122\n",
      "Epoch 235/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 24.1219\n",
      "Epoch 236/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 24.2179\n",
      "Epoch 237/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 24.4692\n",
      "Epoch 238/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 24.5729\n",
      "Epoch 239/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 24.3216\n",
      "Epoch 240/250\n",
      "22/22 [==============================] - 0s 992us/step - loss: 24.3890\n",
      "Epoch 241/250\n",
      "22/22 [==============================] - 0s 896us/step - loss: 24.1508\n",
      "Epoch 242/250\n",
      "22/22 [==============================] - 0s 970us/step - loss: 24.2693\n",
      "Epoch 243/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 24.3759\n",
      "Epoch 244/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 24.3579\n",
      "Epoch 245/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 24.2680\n",
      "Epoch 246/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 24.3722\n",
      "Epoch 247/250\n",
      "22/22 [==============================] - 0s 962us/step - loss: 24.0457\n",
      "Epoch 248/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 24.2521\n",
      "Epoch 249/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 24.1656\n",
      "Epoch 250/250\n",
      "22/22 [==============================] - 0s 1ms/step - loss: 24.3821\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x7f99f182fc40>"
      ]
     },
     "execution_count": 197,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(x=X_train,y=y_train,epochs=250,verbose=1)\n",
    "\n",
    "#x=features we are training on\n",
    "#y=training labels that correspond to trainig feature points\n",
    "#epochs= number of times you go through the dataset\n",
    "#verbose= amount of info that is outputted while the model runs\n",
    "\n",
    "#later on we will cover making the model smart enough to cut off at a\n",
    "#certain time where it is optimized based on validation loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 198,
   "id": "c0daa189",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'loss': [257172.75,\n",
       "  256957.828125,\n",
       "  256805.546875,\n",
       "  256697.734375,\n",
       "  256625.578125,\n",
       "  256579.953125,\n",
       "  256542.078125,\n",
       "  256500.875,\n",
       "  256451.796875,\n",
       "  256393.59375,\n",
       "  256325.140625,\n",
       "  256245.34375,\n",
       "  256152.875,\n",
       "  256045.46875,\n",
       "  255922.5625,\n",
       "  255783.5,\n",
       "  255626.21875,\n",
       "  255449.046875,\n",
       "  255250.703125,\n",
       "  255028.3125,\n",
       "  254781.46875,\n",
       "  254505.46875,\n",
       "  254200.953125,\n",
       "  253864.578125,\n",
       "  253495.171875,\n",
       "  253087.5625,\n",
       "  252644.84375,\n",
       "  252161.09375,\n",
       "  251634.90625,\n",
       "  251063.640625,\n",
       "  250442.015625,\n",
       "  249774.765625,\n",
       "  249049.296875,\n",
       "  248272.75,\n",
       "  247437.640625,\n",
       "  246540.453125,\n",
       "  245580.0625,\n",
       "  244554.578125,\n",
       "  243465.078125,\n",
       "  242301.296875,\n",
       "  241060.140625,\n",
       "  239750.359375,\n",
       "  238358.78125,\n",
       "  236883.75,\n",
       "  235320.1875,\n",
       "  233669.140625,\n",
       "  231933.078125,\n",
       "  230109.625,\n",
       "  228200.578125,\n",
       "  226175.84375,\n",
       "  224055.515625,\n",
       "  221830.125,\n",
       "  219510.65625,\n",
       "  217076.6875,\n",
       "  214533.046875,\n",
       "  211898.21875,\n",
       "  209145.046875,\n",
       "  206276.65625,\n",
       "  203301.34375,\n",
       "  200201.078125,\n",
       "  196975.9375,\n",
       "  193642.015625,\n",
       "  190200.890625,\n",
       "  186643.046875,\n",
       "  182981.59375,\n",
       "  179175.0,\n",
       "  175279.625,\n",
       "  171249.25,\n",
       "  167112.203125,\n",
       "  162875.6875,\n",
       "  158526.109375,\n",
       "  154070.1875,\n",
       "  149510.703125,\n",
       "  144849.203125,\n",
       "  140083.8125,\n",
       "  135241.59375,\n",
       "  130326.4765625,\n",
       "  125329.2421875,\n",
       "  120236.640625,\n",
       "  115127.203125,\n",
       "  109932.1953125,\n",
       "  104659.3359375,\n",
       "  99422.9140625,\n",
       "  94098.0625,\n",
       "  88753.5546875,\n",
       "  83435.703125,\n",
       "  78118.0703125,\n",
       "  72826.375,\n",
       "  67554.703125,\n",
       "  62347.6015625,\n",
       "  57207.5703125,\n",
       "  52137.14453125,\n",
       "  47198.390625,\n",
       "  42363.59375,\n",
       "  37679.5703125,\n",
       "  33181.86328125,\n",
       "  28912.0078125,\n",
       "  24797.380859375,\n",
       "  20914.810546875,\n",
       "  17333.919921875,\n",
       "  14059.287109375,\n",
       "  11109.130859375,\n",
       "  8535.5947265625,\n",
       "  6312.27978515625,\n",
       "  4466.53857421875,\n",
       "  3070.764892578125,\n",
       "  2123.0322265625,\n",
       "  1586.27783203125,\n",
       "  1374.7713623046875,\n",
       "  1326.2598876953125,\n",
       "  1307.0638427734375,\n",
       "  1282.0701904296875,\n",
       "  1261.28173828125,\n",
       "  1237.9906005859375,\n",
       "  1216.84765625,\n",
       "  1193.5648193359375,\n",
       "  1167.2208251953125,\n",
       "  1148.0933837890625,\n",
       "  1122.96728515625,\n",
       "  1104.1297607421875,\n",
       "  1084.6727294921875,\n",
       "  1066.191162109375,\n",
       "  1044.0640869140625,\n",
       "  1021.3211669921875,\n",
       "  999.8380126953125,\n",
       "  978.3920288085938,\n",
       "  957.3561401367188,\n",
       "  940.1334838867188,\n",
       "  920.5852661132812,\n",
       "  900.57373046875,\n",
       "  878.5322265625,\n",
       "  862.3938598632812,\n",
       "  844.42578125,\n",
       "  825.687744140625,\n",
       "  809.2899780273438,\n",
       "  789.0845336914062,\n",
       "  769.8427734375,\n",
       "  750.2625732421875,\n",
       "  731.33935546875,\n",
       "  714.7362670898438,\n",
       "  697.5770874023438,\n",
       "  680.2286376953125,\n",
       "  662.5813598632812,\n",
       "  646.1339111328125,\n",
       "  627.4380493164062,\n",
       "  610.7582397460938,\n",
       "  594.2882080078125,\n",
       "  579.0415649414062,\n",
       "  563.3976440429688,\n",
       "  546.7811279296875,\n",
       "  530.4021606445312,\n",
       "  514.345458984375,\n",
       "  501.4859313964844,\n",
       "  486.5758056640625,\n",
       "  471.6917419433594,\n",
       "  457.576171875,\n",
       "  444.0783996582031,\n",
       "  431.6012878417969,\n",
       "  418.1478271484375,\n",
       "  405.9101867675781,\n",
       "  392.38775634765625,\n",
       "  380.0342102050781,\n",
       "  366.44287109375,\n",
       "  353.4258728027344,\n",
       "  339.75775146484375,\n",
       "  328.3153991699219,\n",
       "  314.9560852050781,\n",
       "  303.33642578125,\n",
       "  291.55267333984375,\n",
       "  279.7840270996094,\n",
       "  269.8734436035156,\n",
       "  258.18682861328125,\n",
       "  246.74874877929688,\n",
       "  236.3708038330078,\n",
       "  226.37515258789062,\n",
       "  215.76211547851562,\n",
       "  205.5177459716797,\n",
       "  197.15655517578125,\n",
       "  188.49024963378906,\n",
       "  179.72454833984375,\n",
       "  171.7322540283203,\n",
       "  162.9580078125,\n",
       "  153.76918029785156,\n",
       "  145.991455078125,\n",
       "  138.44851684570312,\n",
       "  131.69259643554688,\n",
       "  124.27906036376953,\n",
       "  117.7776870727539,\n",
       "  111.20602416992188,\n",
       "  104.89325714111328,\n",
       "  99.55706787109375,\n",
       "  93.0575942993164,\n",
       "  87.34992980957031,\n",
       "  82.8013916015625,\n",
       "  77.92413330078125,\n",
       "  73.57597351074219,\n",
       "  69.2673568725586,\n",
       "  65.67180633544922,\n",
       "  62.30631637573242,\n",
       "  59.02421951293945,\n",
       "  55.21598815917969,\n",
       "  51.748374938964844,\n",
       "  48.87843322753906,\n",
       "  46.50392150878906,\n",
       "  43.94866943359375,\n",
       "  42.170982360839844,\n",
       "  39.66389846801758,\n",
       "  37.281883239746094,\n",
       "  35.616539001464844,\n",
       "  34.03940963745117,\n",
       "  32.597496032714844,\n",
       "  31.511531829833984,\n",
       "  30.756481170654297,\n",
       "  30.376169204711914,\n",
       "  29.254207611083984,\n",
       "  28.664127349853516,\n",
       "  28.166261672973633,\n",
       "  27.585290908813477,\n",
       "  27.353809356689453,\n",
       "  26.6514835357666,\n",
       "  26.46685028076172,\n",
       "  25.98610496520996,\n",
       "  25.692949295043945,\n",
       "  25.659957885742188,\n",
       "  24.879798889160156,\n",
       "  25.3775691986084,\n",
       "  25.093236923217773,\n",
       "  24.934288024902344,\n",
       "  24.883739471435547,\n",
       "  24.502410888671875,\n",
       "  24.58214569091797,\n",
       "  24.72736930847168,\n",
       "  24.487764358520508,\n",
       "  24.01217269897461,\n",
       "  24.12186622619629,\n",
       "  24.217941284179688,\n",
       "  24.469173431396484,\n",
       "  24.572893142700195,\n",
       "  24.321609497070312,\n",
       "  24.38901710510254,\n",
       "  24.15082359313965,\n",
       "  24.26932716369629,\n",
       "  24.375879287719727,\n",
       "  24.357934951782227,\n",
       "  24.26799201965332,\n",
       "  24.372249603271484,\n",
       "  24.045711517333984,\n",
       "  24.252126693725586,\n",
       "  24.165557861328125,\n",
       "  24.382081985473633]}"
      ]
     },
     "execution_count": 198,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.history.history"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 199,
   "id": "1ff9193d",
   "metadata": {},
   "outputs": [],
   "source": [
    "loss_df = pd.DataFrame(model.history.history)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 200,
   "id": "a0e3cc99",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 200,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "loss_df.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 201,
   "id": "31d26c72",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "25.037519454956055"
      ]
     },
     "execution_count": 201,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.evaluate(X_test,y_test,verbose=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 202,
   "id": "0f540ad2",
   "metadata": {},
   "outputs": [],
   "source": [
    "#on the test dataset, the model is getting back a MSE of 25.81"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 203,
   "id": "f475e5f0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "22/22 [==============================] - 0s 938us/step - loss: 23.7901\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "23.790124893188477"
      ]
     },
     "execution_count": 203,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.evaluate(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "95e4a796",
   "metadata": {},
   "outputs": [],
   "source": [
    "#this shows the loss on our training set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 204,
   "id": "68c76a42",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10/10 [==============================] - 0s 1ms/step\n"
     ]
    }
   ],
   "source": [
    "test_predictions = model.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 205,
   "id": "ec4e88d6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[405.48212],\n",
       "       [623.60223],\n",
       "       [592.17505],\n",
       "       [572.40607],\n",
       "       [366.99783],\n",
       "       [579.2931 ],\n",
       "       [515.34106],\n",
       "       [459.20297],\n",
       "       [549.4357 ],\n",
       "       [447.60895],\n",
       "       [611.8514 ],\n",
       "       [548.8884 ],\n",
       "       [419.2967 ],\n",
       "       [409.09885],\n",
       "       [651.11523],\n",
       "       [437.42603],\n",
       "       [508.4463 ],\n",
       "       [660.11456],\n",
       "       [662.6839 ],\n",
       "       [565.5687 ],\n",
       "       [334.54758],\n",
       "       [445.01312],\n",
       "       [382.79532],\n",
       "       [378.9559 ],\n",
       "       [566.7035 ],\n",
       "       [610.74963],\n",
       "       [532.5325 ],\n",
       "       [428.02374],\n",
       "       [655.58636],\n",
       "       [414.20386],\n",
       "       [442.74277],\n",
       "       [485.27252],\n",
       "       [438.5749 ],\n",
       "       [681.95184],\n",
       "       [424.8307 ],\n",
       "       [417.9416 ],\n",
       "       [501.8855 ],\n",
       "       [550.5681 ],\n",
       "       [509.99188],\n",
       "       [395.71326],\n",
       "       [618.62067],\n",
       "       [416.7659 ],\n",
       "       [604.4934 ],\n",
       "       [446.01108],\n",
       "       [502.1275 ],\n",
       "       [581.6233 ],\n",
       "       [668.90393],\n",
       "       [490.4804 ],\n",
       "       [318.74802],\n",
       "       [485.58826],\n",
       "       [517.36414],\n",
       "       [381.92065],\n",
       "       [541.89716],\n",
       "       [408.78006],\n",
       "       [641.5295 ],\n",
       "       [491.24902],\n",
       "       [628.13477],\n",
       "       [626.9222 ],\n",
       "       [447.22916],\n",
       "       [484.93057],\n",
       "       [491.34058],\n",
       "       [474.64847],\n",
       "       [682.78894],\n",
       "       [403.539  ],\n",
       "       [701.14386],\n",
       "       [586.5311 ],\n",
       "       [583.19275],\n",
       "       [537.9029 ],\n",
       "       [484.73093],\n",
       "       [516.76825],\n",
       "       [361.42532],\n",
       "       [540.8705 ],\n",
       "       [570.7664 ],\n",
       "       [528.4137 ],\n",
       "       [453.85788],\n",
       "       [531.2157 ],\n",
       "       [507.24496],\n",
       "       [443.5986 ],\n",
       "       [543.5119 ],\n",
       "       [640.69904],\n",
       "       [466.32684],\n",
       "       [567.3176 ],\n",
       "       [690.6799 ],\n",
       "       [458.95242],\n",
       "       [709.02185],\n",
       "       [472.8674 ],\n",
       "       [403.35446],\n",
       "       [585.20605],\n",
       "       [436.8035 ],\n",
       "       [489.04745],\n",
       "       [617.0753 ],\n",
       "       [439.57318],\n",
       "       [455.585  ],\n",
       "       [435.46725],\n",
       "       [507.1477 ],\n",
       "       [608.56525],\n",
       "       [321.88663],\n",
       "       [436.40756],\n",
       "       [536.4211 ],\n",
       "       [518.8387 ],\n",
       "       [605.2369 ],\n",
       "       [525.6033 ],\n",
       "       [334.28162],\n",
       "       [576.1417 ],\n",
       "       [432.14127],\n",
       "       [562.48334],\n",
       "       [513.51953],\n",
       "       [391.34235],\n",
       "       [566.513  ],\n",
       "       [454.81207],\n",
       "       [448.70828],\n",
       "       [641.0043 ],\n",
       "       [524.41266],\n",
       "       [550.7181 ],\n",
       "       [417.89072],\n",
       "       [478.92557],\n",
       "       [586.61115],\n",
       "       [667.49036],\n",
       "       [700.55115],\n",
       "       [659.6058 ],\n",
       "       [560.6726 ],\n",
       "       [503.2368 ],\n",
       "       [390.67123],\n",
       "       [281.65994],\n",
       "       [479.72885],\n",
       "       [616.33325],\n",
       "       [373.4144 ],\n",
       "       [512.3885 ],\n",
       "       [511.2913 ],\n",
       "       [493.69614],\n",
       "       [480.60526],\n",
       "       [424.09134],\n",
       "       [493.63858],\n",
       "       [471.615  ],\n",
       "       [600.40845],\n",
       "       [573.5177 ],\n",
       "       [415.30383],\n",
       "       [630.68555],\n",
       "       [466.4757 ],\n",
       "       [564.3725 ],\n",
       "       [405.89563],\n",
       "       [532.31433],\n",
       "       [572.7504 ],\n",
       "       [357.36313],\n",
       "       [549.9954 ],\n",
       "       [603.44684],\n",
       "       [384.54953],\n",
       "       [542.5908 ],\n",
       "       [562.6927 ],\n",
       "       [452.9715 ],\n",
       "       [632.2578 ],\n",
       "       [372.59644],\n",
       "       [474.49005],\n",
       "       [529.10223],\n",
       "       [372.4675 ],\n",
       "       [461.48877],\n",
       "       [436.579  ],\n",
       "       [498.65454],\n",
       "       [346.24564],\n",
       "       [395.42825],\n",
       "       [604.61975],\n",
       "       [506.7554 ],\n",
       "       [468.80014],\n",
       "       [490.44827],\n",
       "       [535.2787 ],\n",
       "       [344.76602],\n",
       "       [512.2571 ],\n",
       "       [251.04416],\n",
       "       [504.22455],\n",
       "       [541.36444],\n",
       "       [489.54572],\n",
       "       [471.078  ],\n",
       "       [392.73917],\n",
       "       [416.29843],\n",
       "       [549.73413],\n",
       "       [476.04828],\n",
       "       [580.13165],\n",
       "       [490.02927],\n",
       "       [601.2151 ],\n",
       "       [547.0403 ],\n",
       "       [542.0293 ],\n",
       "       [500.71338],\n",
       "       [645.9331 ],\n",
       "       [560.5548 ],\n",
       "       [577.8757 ],\n",
       "       [444.17514],\n",
       "       [415.613  ],\n",
       "       [420.3312 ],\n",
       "       [569.1446 ],\n",
       "       [609.1433 ],\n",
       "       [437.8633 ],\n",
       "       [488.08276],\n",
       "       [587.9523 ],\n",
       "       [525.58997],\n",
       "       [357.62518],\n",
       "       [645.5342 ],\n",
       "       [528.2324 ],\n",
       "       [337.4501 ],\n",
       "       [492.853  ],\n",
       "       [410.24823],\n",
       "       [606.65686],\n",
       "       [346.6626 ],\n",
       "       [522.45   ],\n",
       "       [404.837  ],\n",
       "       [258.8724 ],\n",
       "       [519.9645 ],\n",
       "       [340.96484],\n",
       "       [362.39133],\n",
       "       [576.4053 ],\n",
       "       [416.76004],\n",
       "       [551.16016],\n",
       "       [521.26874],\n",
       "       [511.13037],\n",
       "       [324.9603 ],\n",
       "       [404.4883 ],\n",
       "       [602.5748 ],\n",
       "       [618.01013],\n",
       "       [603.5395 ],\n",
       "       [566.2736 ],\n",
       "       [473.67966],\n",
       "       [460.5217 ],\n",
       "       [508.5938 ],\n",
       "       [446.00333],\n",
       "       [511.32962],\n",
       "       [502.66895],\n",
       "       [400.28558],\n",
       "       [605.62244],\n",
       "       [258.4471 ],\n",
       "       [628.2783 ],\n",
       "       [589.03577],\n",
       "       [327.309  ],\n",
       "       [479.7981 ],\n",
       "       [595.40314],\n",
       "       [378.58484],\n",
       "       [460.3356 ],\n",
       "       [325.15936],\n",
       "       [518.73145],\n",
       "       [409.68512],\n",
       "       [556.31647],\n",
       "       [641.9079 ],\n",
       "       [536.9701 ],\n",
       "       [503.26038],\n",
       "       [635.0128 ],\n",
       "       [515.32117],\n",
       "       [531.8362 ],\n",
       "       [519.04285],\n",
       "       [457.57874],\n",
       "       [505.87296],\n",
       "       [461.23117],\n",
       "       [591.6441 ],\n",
       "       [466.11383],\n",
       "       [427.44714],\n",
       "       [541.8063 ],\n",
       "       [493.97495],\n",
       "       [679.9311 ],\n",
       "       [372.9967 ],\n",
       "       [551.71814],\n",
       "       [578.055  ],\n",
       "       [434.23206],\n",
       "       [543.1951 ],\n",
       "       [585.9733 ],\n",
       "       [579.2413 ],\n",
       "       [720.9754 ],\n",
       "       [433.0666 ],\n",
       "       [398.74094],\n",
       "       [314.20392],\n",
       "       [448.51123],\n",
       "       [388.15048],\n",
       "       [543.3936 ],\n",
       "       [522.89624],\n",
       "       [564.44037],\n",
       "       [448.18002],\n",
       "       [534.4894 ],\n",
       "       [381.97318],\n",
       "       [501.63467],\n",
       "       [637.24634],\n",
       "       [496.55896],\n",
       "       [568.5564 ],\n",
       "       [470.3348 ],\n",
       "       [273.40872],\n",
       "       [517.56683],\n",
       "       [621.50775],\n",
       "       [350.74585],\n",
       "       [450.7416 ],\n",
       "       [499.49872],\n",
       "       [543.10144],\n",
       "       [611.83636],\n",
       "       [388.5228 ],\n",
       "       [449.7132 ],\n",
       "       [482.52664],\n",
       "       [598.4462 ],\n",
       "       [499.65033],\n",
       "       [321.69934],\n",
       "       [555.2654 ],\n",
       "       [444.7954 ],\n",
       "       [529.2378 ],\n",
       "       [515.6968 ],\n",
       "       [609.7037 ],\n",
       "       [417.06247],\n",
       "       [411.12842]], dtype=float32)"
      ]
     },
     "execution_count": 205,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 206,
   "id": "b0a7cb96",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0      405.482117\n",
       "1      623.602234\n",
       "2      592.175049\n",
       "3      572.406067\n",
       "4      366.997833\n",
       "          ...    \n",
       "295    529.237793\n",
       "296    515.696777\n",
       "297    609.703674\n",
       "298    417.062469\n",
       "299    411.128418\n",
       "Length: 300, dtype: float32"
      ]
     },
     "execution_count": 206,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_predictions = pd.Series(test_predictions.reshape(300,))\n",
    "test_predictions\n",
    "\n",
    "#reshape is required to make the dimensions match with what the series\n",
    "#call expects"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 207,
   "id": "2c4f4d2e",
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_df = pd.DataFrame(y_test,columns = ['Test True Y'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 208,
   "id": "c2b0fb2d",
   "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>Test True Y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>402.296319</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>624.156198</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>582.455066</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>578.588606</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>371.224104</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>295</th>\n",
       "      <td>525.704657</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>296</th>\n",
       "      <td>502.909473</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>297</th>\n",
       "      <td>612.727910</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>298</th>\n",
       "      <td>417.569725</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>299</th>\n",
       "      <td>410.538250</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>300 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     Test True Y\n",
       "0     402.296319\n",
       "1     624.156198\n",
       "2     582.455066\n",
       "3     578.588606\n",
       "4     371.224104\n",
       "..           ...\n",
       "295   525.704657\n",
       "296   502.909473\n",
       "297   612.727910\n",
       "298   417.569725\n",
       "299   410.538250\n",
       "\n",
       "[300 rows x 1 columns]"
      ]
     },
     "execution_count": 208,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 209,
   "id": "14784970",
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_df = pd.concat([pred_df,test_predictions],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 210,
   "id": "6d53fbb4",
   "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>Test True Y</th>\n",
       "      <th>0</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>402.296319</td>\n",
       "      <td>405.482117</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>624.156198</td>\n",
       "      <td>623.602234</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>582.455066</td>\n",
       "      <td>592.175049</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>578.588606</td>\n",
       "      <td>572.406067</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>371.224104</td>\n",
       "      <td>366.997833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>295</th>\n",
       "      <td>525.704657</td>\n",
       "      <td>529.237793</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>296</th>\n",
       "      <td>502.909473</td>\n",
       "      <td>515.696777</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>297</th>\n",
       "      <td>612.727910</td>\n",
       "      <td>609.703674</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>298</th>\n",
       "      <td>417.569725</td>\n",
       "      <td>417.062469</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>299</th>\n",
       "      <td>410.538250</td>\n",
       "      <td>411.128418</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>300 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     Test True Y           0\n",
       "0     402.296319  405.482117\n",
       "1     624.156198  623.602234\n",
       "2     582.455066  592.175049\n",
       "3     578.588606  572.406067\n",
       "4     371.224104  366.997833\n",
       "..           ...         ...\n",
       "295   525.704657  529.237793\n",
       "296   502.909473  515.696777\n",
       "297   612.727910  609.703674\n",
       "298   417.569725  417.062469\n",
       "299   410.538250  411.128418\n",
       "\n",
       "[300 rows x 2 columns]"
      ]
     },
     "execution_count": 210,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 211,
   "id": "0c461656",
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_df.columns = ['Test True Y','Model Predictions']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 212,
   "id": "79b316af",
   "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>Test True Y</th>\n",
       "      <th>Model Predictions</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>402.296319</td>\n",
       "      <td>405.482117</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>624.156198</td>\n",
       "      <td>623.602234</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>582.455066</td>\n",
       "      <td>592.175049</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>578.588606</td>\n",
       "      <td>572.406067</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>371.224104</td>\n",
       "      <td>366.997833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>295</th>\n",
       "      <td>525.704657</td>\n",
       "      <td>529.237793</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>296</th>\n",
       "      <td>502.909473</td>\n",
       "      <td>515.696777</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>297</th>\n",
       "      <td>612.727910</td>\n",
       "      <td>609.703674</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>298</th>\n",
       "      <td>417.569725</td>\n",
       "      <td>417.062469</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>299</th>\n",
       "      <td>410.538250</td>\n",
       "      <td>411.128418</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>300 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     Test True Y  Model Predictions\n",
       "0     402.296319         405.482117\n",
       "1     624.156198         623.602234\n",
       "2     582.455066         592.175049\n",
       "3     578.588606         572.406067\n",
       "4     371.224104         366.997833\n",
       "..           ...                ...\n",
       "295   525.704657         529.237793\n",
       "296   502.909473         515.696777\n",
       "297   612.727910         609.703674\n",
       "298   417.569725         417.062469\n",
       "299   410.538250         411.128418\n",
       "\n",
       "[300 rows x 2 columns]"
      ]
     },
     "execution_count": 212,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 213,
   "id": "b238edcb",
   "metadata": {},
   "outputs": [],
   "source": [
    "#we can now see the true price value vs the model's predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 214,
   "id": "02ead846",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='Test True Y', ylabel='Model Predictions'>"
      ]
     },
     "execution_count": 214,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.scatterplot(data=pred_df,x='Test True Y',y='Model Predictions')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "id": "ab0ee440",
   "metadata": {},
   "outputs": [],
   "source": [
    "#perfect match would be a perfect straight line. This is very close so\n",
    "#the model is performing very well."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 186,
   "id": "d36f8ba3",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import mean_absolute_error,mean_squared_error"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 215,
   "id": "32659abe",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4.013332291073785"
      ]
     },
     "execution_count": 215,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mean_absolute_error(pred_df['Test True Y'],pred_df['Model Predictions'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 216,
   "id": "f0bff770",
   "metadata": {},
   "outputs": [],
   "source": [
    "#mean absolute error is 4.07; on average, we are about $4 off from\n",
    "#our price point."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 217,
   "id": "a1d406f0",
   "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>price</th>\n",
       "      <th>feature1</th>\n",
       "      <th>feature2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1000.000000</td>\n",
       "      <td>1000.000000</td>\n",
       "      <td>1000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>498.673029</td>\n",
       "      <td>1000.014171</td>\n",
       "      <td>999.979847</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>93.785431</td>\n",
       "      <td>0.974018</td>\n",
       "      <td>0.948330</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>223.346793</td>\n",
       "      <td>997.058347</td>\n",
       "      <td>996.995651</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>433.025732</td>\n",
       "      <td>999.332068</td>\n",
       "      <td>999.316106</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>502.382117</td>\n",
       "      <td>1000.009915</td>\n",
       "      <td>1000.002243</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>564.921588</td>\n",
       "      <td>1000.637580</td>\n",
       "      <td>1000.645380</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>774.407854</td>\n",
       "      <td>1003.207934</td>\n",
       "      <td>1002.666308</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             price     feature1     feature2\n",
       "count  1000.000000  1000.000000  1000.000000\n",
       "mean    498.673029  1000.014171   999.979847\n",
       "std      93.785431     0.974018     0.948330\n",
       "min     223.346793   997.058347   996.995651\n",
       "25%     433.025732   999.332068   999.316106\n",
       "50%     502.382117  1000.009915  1000.002243\n",
       "75%     564.921588  1000.637580  1000.645380\n",
       "max     774.407854  1003.207934  1002.666308"
      ]
     },
     "execution_count": 217,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "id": "40847dbb",
   "metadata": {},
   "outputs": [],
   "source": [
    "#avg price is about $500 with min of $223 and max of $774. We are \n",
    "#on average off by about $4, which is less than 1% off the mean.\n",
    "\n",
    "#mean values and actual distribution of your label must always be \n",
    "#accounted for."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 218,
   "id": "76203e5a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "25.037524266636876"
      ]
     },
     "execution_count": 218,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mean_squared_error(pred_df['Test True Y'],pred_df['Model Predictions'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 219,
   "id": "09570298",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "25.037519454956055"
      ]
     },
     "execution_count": 219,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.evaluate(X_test,y_test,verbose=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "id": "41cb5d60",
   "metadata": {},
   "outputs": [],
   "source": [
    "#note that the MSE of the dataframe is essentially the same as the\n",
    "#MSE as calculated by the original loss model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 220,
   "id": "4d3b4310",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "5.003751019648847"
      ]
     },
     "execution_count": 220,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mean_squared_error(pred_df['Test True Y'],\n",
    "        pred_df['Model Predictions'])**0.5\n",
    "\n",
    "#root mean squared error"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 221,
   "id": "2c5182c4",
   "metadata": {},
   "outputs": [],
   "source": [
    "#now to predict brand new data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 222,
   "id": "8027716b",
   "metadata": {},
   "outputs": [],
   "source": [
    "new_gem = [[998,1000]]\n",
    "\n",
    "#remember that the model is trained on scaled data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 223,
   "id": "71a08538",
   "metadata": {},
   "outputs": [],
   "source": [
    "new_gem = scaler.transform(new_gem)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 224,
   "id": "6a1f5c7a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.14117652, 0.53968792]])"
      ]
     },
     "execution_count": 224,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_gem"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 225,
   "id": "b920a3a1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 19ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[420.02463]], dtype=float32)"
      ]
     },
     "execution_count": 225,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.predict(new_gem)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 226,
   "id": "3a3002b9",
   "metadata": {},
   "outputs": [],
   "source": [
    "#note that this is the same process as predicting the test set because\n",
    "#the test set is essentially new data that the model hasn't seen \n",
    "#before."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 227,
   "id": "37310f18",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.models import load_model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ce651095",
   "metadata": {},
   "outputs": [],
   "source": [
    "#how to save and load a model in another file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 228,
   "id": "c6ed5eb1",
   "metadata": {},
   "outputs": [],
   "source": [
    "model.save('my_gem_model.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 229,
   "id": "91609621",
   "metadata": {},
   "outputs": [],
   "source": [
    "later_model = load_model('my_gem_model.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 230,
   "id": "32500f95",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 51ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[420.02463]], dtype=float32)"
      ]
     },
     "execution_count": 230,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "later_model.predict(new_gem)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "870012fc",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}