Advanced-Time-Series-Analysis-for-Order-Predictions.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "64b0e515",
   "metadata": {},
   "source": [
    "# Load and Analyze Time Series Data\n",
    "\n",
    "<div data-purpose=\"safely-set-inner-html:rich-text-viewer:html\" class=\"lab-rich-text-viewer--lab-html-content--22XsM rt-scaffolding\"><p>You're required to perform initial data exploration for the 'orders.csv' file. Start by importing the necessary libraries. After which, load the data into a DataFrame. Ensure that the 'date' column serves as the index and is parsed as a date. Subsequently, view the last five entries to validate your data import. Lastly, retrieve a statistical summary of the loaded data to gauge its distribution and characteristics.</p></div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "dfcc6f97",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import the required Pandas library.\n",
    "\n",
    "# Import the Pandas library\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "2cda3c0b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load the data from 'orders.csv' into a DataFrame, setting 'date' as the index and parsing it appropriately.\n",
    "\n",
    "# Load data from 'orders.csv' with 'date' as index and parsed as date\n",
    "df = pd.read_csv(\"orders.csv\", parse_dates=True, index_col = ['date'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "4eb03cd7",
   "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>orders</th>\n",
       "      <th>marketing_investment</th>\n",
       "      <th>discount_rate</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2022-12-27</th>\n",
       "      <td>27450.6</td>\n",
       "      <td>302.904844</td>\n",
       "      <td>0.070321</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-12-28</th>\n",
       "      <td>29131.6</td>\n",
       "      <td>303.511201</td>\n",
       "      <td>0.073613</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-12-29</th>\n",
       "      <td>19133.3</td>\n",
       "      <td>300.464154</td>\n",
       "      <td>0.006089</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-12-30</th>\n",
       "      <td>18347.7</td>\n",
       "      <td>297.994770</td>\n",
       "      <td>0.005196</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2022-12-31</th>\n",
       "      <td>12451.0</td>\n",
       "      <td>300.681191</td>\n",
       "      <td>0.009529</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             orders  marketing_investment  discount_rate\n",
       "date                                                    \n",
       "2022-12-27  27450.6            302.904844       0.070321\n",
       "2022-12-28  29131.6            303.511201       0.073613\n",
       "2022-12-29  19133.3            300.464154       0.006089\n",
       "2022-12-30  18347.7            297.994770       0.005196\n",
       "2022-12-31  12451.0            300.681191       0.009529"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Display the last five records to ensure correct data loading.\n",
    "\n",
    "# Preview the last five entries in the DataFrame\n",
    "df.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "2b28120f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Summary Statistics:\n",
      "             orders  marketing_investment  discount_rate\n",
      "count   1826.000000           1826.000000    1826.000000\n",
      "mean   16102.564239            190.488288       0.011012\n",
      "std     7048.729220             52.782529       0.015189\n",
      "min     4925.700000             76.739679       0.002250\n",
      "25%    10769.175000            149.182568       0.004157\n",
      "50%    14673.650000            186.819914       0.006233\n",
      "75%    20026.800000            229.652842       0.010576\n",
      "max    93604.900000            352.655988       0.126732\n"
     ]
    }
   ],
   "source": [
    "# Obtain and display a statistical summary of the DataFrame to inspect its data distribution.\n",
    "\n",
    "# Retrieve and print a statistical summary of the DataFrame\n",
    "print(\"\\nSummary Statistics:\")\n",
    "print(df.describe())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "51d84072",
   "metadata": {},
   "source": [
    "# Decompose and Visualize the 'orders' Time Series Using STL Method\n",
    "\n",
    "<p>Your objective is to break down the 'orders' time series data into its underlying components using the STL method. After the decomposition, visualize the original data, trend, seasonal, and residual components for a comprehensive view of the 'orders' time series data.</p>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "b94a1d09",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Start by importing the necessary Matplotlib library.\n",
    "\n",
    "# Import the required Matplotlib library for plotting\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "d98d5872",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Orders')"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1500x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Initialize a figure to host the three subplots with a suitable size for clear visualization. \n",
    "# Create the first subplot to visualize 'orders' over time.\n",
    "\n",
    "# Initialize a figure for the subplots\n",
    "plt.figure(figsize=(15, 6))\n",
    " \n",
    "# Create a subplot to visualize 'orders' over time\n",
    "plt.subplot(1, 3, 1)\n",
    "plt.plot(df.index, df['orders'])\n",
    "plt.title('Orders Over Time')\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Orders')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "89c2900f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1500x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the 'marketing_investment' against time in the second subplot. Adjust the layout to ensure no overlaps and display the plotted charts.\n",
    "\n",
    "# Initialize a figure for the subplots\n",
    "plt.figure(figsize=(15, 6))\n",
    " \n",
    "# Create a subplot to visualize 'orders' over time\n",
    "plt.subplot(1, 3, 1)\n",
    "plt.plot(df.index, df['orders'])\n",
    "plt.title('Orders Over Time')\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Orders')\n",
    " \n",
    "# Add a subplot to visualize 'marketing_investment' over time\n",
    "plt.subplot(1, 3, 2)\n",
    "plt.plot(df.index, df['marketing_investment'])\n",
    "plt.title('Marketing Investment Over Time')\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Marketing Investment')\n",
    " \n",
    "# Adjust the layout and display the figure\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "c45df763",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1500x600 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Lastly, visualize the 'discount_rate' over time in the third subplot.\n",
    "\n",
    "# Initialize a figure for the subplots\n",
    "plt.figure(figsize=(15, 6))\n",
    " \n",
    "# Create a subplot to visualize 'orders' over time\n",
    "plt.subplot(1, 3, 1)\n",
    "plt.plot(df.index, df['orders'])\n",
    "plt.title('Orders Over Time')\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Orders')\n",
    " \n",
    "# Add a subplot to visualize 'marketing_investment' over time\n",
    "plt.subplot(1, 3, 2)\n",
    "plt.plot(df.index, df['marketing_investment'])\n",
    "plt.title('Marketing Investment Over Time')\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Marketing Investment')\n",
    " \n",
    "# Add a subplot to visualize 'discount_rate' over time\n",
    "plt.subplot(1, 3, 3)\n",
    "plt.plot(df.index, df['discount_rate'])\n",
    "plt.title('Discount Rate Over Time')\n",
    "plt.xlabel('Date')\n",
    "plt.ylabel('Discount Rate')\n",
    " \n",
    "# Adjust the layout and display the figure\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1213d16b",
   "metadata": {},
   "source": [
    "# Incorporate Trend and Seasonal Components into the DataFrame\n",
    "\n",
    "<p>Integrate the previously obtained trend and seasonal components from the 'orders' time series decomposition into the primary DataFrame. After this integration, ensure that the DataFrame is clean by removing any rows that contain NaN values.</p>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "ace5678d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import the required library for STL decomposition.\n",
    "\n",
    "# Required Libraries\n",
    "from statsmodels.tsa.seasonal import STL"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "3f6dcc8d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Decompose the 'orders' time series into its trend, seasonal, and residual components. Assuming daily data, set the seasonality period to 365 days.\n",
    "\n",
    "# Decompose the 'orders' time series using STL\n",
    "stl = STL(df['orders'], period=365)  # Assuming daily data and 365 days for seasonality\n",
    "result = stl.fit()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "1520ef95",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Extract the respective components from the STL result.\n",
    "\n",
    "# Extract the trend, seasonal, and residual components\n",
    "trend = result.trend\n",
    "seasonal = result.seasonal\n",
    "residual = result.resid"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "5e3b77df",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1500x800 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot and visualize the original time series data alongside the decomposed components.\n",
    "\n",
    "# Plotting\n",
    "plt.figure(figsize=(15, 8))\n",
    " \n",
    "# Original Data\n",
    "plt.subplot(4, 1, 1)\n",
    "plt.plot(df.index, df['orders'], label='Original')\n",
    "plt.legend(loc='upper left')\n",
    "plt.title('Original Time Series')\n",
    " \n",
    "# Trend Component\n",
    "plt.subplot(4, 1, 2)\n",
    "plt.plot(df.index, trend, label='Trend')\n",
    "plt.legend(loc='upper left')\n",
    "plt.title('Trend Component')\n",
    " \n",
    "# Seasonal Component\n",
    "plt.subplot(4, 1, 3)\n",
    "plt.plot(df.index, seasonal, label='Seasonal')\n",
    "plt.legend(loc='upper left')\n",
    "plt.title('Seasonal Component')\n",
    " \n",
    "# Residual Component\n",
    "plt.subplot(4, 1, 4)\n",
    "plt.plot(df.index, residual, label='Residual')\n",
    "plt.legend(loc='upper left')\n",
    "plt.title('Residual Component')\n",
    " \n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5ca7a3f6",
   "metadata": {},
   "source": [
    "# Preprocess Data for LSTM and Train a Sequential Model\n",
    "\n",
    "<div data-purpose=\"safely-set-inner-html:rich-text-viewer:html\" class=\"lab-rich-text-viewer--lab-html-content--22XsM rt-scaffolding\"><p>Prepare the provided dataset for LSTM processing. You need to:</p><ul><li><p>Scale the dataset using MinMax scaling.</p></li><li><p>Segment the data into feature (X) and target (y) variables.</p></li><li><p>Ensure the data is shaped to fit into LSTM layers.</p></li><li><p>Split the dataset into training and test sets. The test set should consist of the last 31 records.</p></li><li><p>Construct an LSTM model:</p><ul><li><p>Use 50 neurons in the LSTM layer with the activation function set to 'relu'.</p></li><li><p>The dense layer should contain one neuron.</p></li><li><p>Compile the model using the 'adam' optimizer and the mean absolute error ('mae') as the loss metric.</p></li></ul></li><li><p>Train the model for 50 epochs.</p></li></ul></div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "15e31503",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Add the trend and seasonal components to the DataFrame.\n",
    "\n",
    "# Add trend and seasonal components to the dataframe\n",
    "df['trend'] = result.trend\n",
    "df['seasonal'] = result.seasonal\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "7b8e2bd4",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Clean the DataFrame by eliminating any rows that might have NaN values, ensuring a seamless dataset.\n",
    "\n",
    "# We'll need to remove any remaining NaN values after decomposition\n",
    "df.dropna(inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "bea088f2",
   "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>orders</th>\n",
       "      <th>marketing_investment</th>\n",
       "      <th>discount_rate</th>\n",
       "      <th>trend</th>\n",
       "      <th>seasonal</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2018-01-01</th>\n",
       "      <td>7766.7</td>\n",
       "      <td>342.634293</td>\n",
       "      <td>0.015126</td>\n",
       "      <td>8755.194669</td>\n",
       "      <td>-1522.105256</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-01-02</th>\n",
       "      <td>10686.5</td>\n",
       "      <td>304.345270</td>\n",
       "      <td>0.015907</td>\n",
       "      <td>8760.481296</td>\n",
       "      <td>2420.759849</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-01-03</th>\n",
       "      <td>11032.3</td>\n",
       "      <td>284.074519</td>\n",
       "      <td>0.014387</td>\n",
       "      <td>8765.768729</td>\n",
       "      <td>2521.026909</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-01-04</th>\n",
       "      <td>10034.9</td>\n",
       "      <td>275.267608</td>\n",
       "      <td>0.013852</td>\n",
       "      <td>8771.056972</td>\n",
       "      <td>1003.363837</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-01-05</th>\n",
       "      <td>8452.1</td>\n",
       "      <td>264.312049</td>\n",
       "      <td>0.012928</td>\n",
       "      <td>8776.346029</td>\n",
       "      <td>-435.974023</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             orders  marketing_investment  discount_rate        trend  \\\n",
       "date                                                                    \n",
       "2018-01-01   7766.7            342.634293       0.015126  8755.194669   \n",
       "2018-01-02  10686.5            304.345270       0.015907  8760.481296   \n",
       "2018-01-03  11032.3            284.074519       0.014387  8765.768729   \n",
       "2018-01-04  10034.9            275.267608       0.013852  8771.056972   \n",
       "2018-01-05   8452.1            264.312049       0.012928  8776.346029   \n",
       "\n",
       "               seasonal  \n",
       "date                     \n",
       "2018-01-01 -1522.105256  \n",
       "2018-01-02  2420.759849  \n",
       "2018-01-03  2521.026909  \n",
       "2018-01-04  1003.363837  \n",
       "2018-01-05  -435.974023  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Inspect the first 5 rows to check the outcome.\n",
    "\n",
    "# Preview the DataFrame\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5793d75f",
   "metadata": {},
   "source": [
    "# Compute Performance Metrics for LSTM Model Predictions on Time Series Data\n",
    "\n",
    "<div data-purpose=\"safely-set-inner-html:rich-text-viewer:html\" class=\"lab-rich-text-viewer--lab-html-content--22XsM rt-scaffolding\"><p>Utilizing the trained LSTM model, carry out predictions on the test dataset. Subsequently, you should:</p><ul><li><p>Predict the last 31 days of orders using the LSTM model.</p></li><li><p>Reverse the scaling process on the predicted values to acquire the original orders scale.</p></li><li><p>Collate the predicted orders to the actual last 31 days of orders.</p></li><li><p>Measure the performance of the LSTM model by computing both the Mean Absolute Error (MAE) and the Mean Absolute Percentage Error (MAPE).</p></li></ul></div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "b6cb5c4f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import all the necessary libraries and tools for data processing and LSTM modeling.\n",
    "\n",
    "# Required Libraries\n",
    "import numpy as np\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import LSTM, Dense"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "d61a6f45",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Utilize the MinMaxScaler to scale the dataset for neural network modeling.\n",
    "\n",
    "# Preprocess data for LSTM\n",
    "scaler = MinMaxScaler()\n",
    "scaled_data = scaler.fit_transform(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "5b5838e8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Segment the data into feature set (X) and target variable (y).\n",
    "\n",
    "# Isolate X and Y\n",
    "X = scaled_data[:, 1:]\n",
    "y = scaled_data[:, 0]   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "003f72aa",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Alter the shape of the feature set to be compatible with LSTM layers.\n",
    "\n",
    "# Reshape X for LSTM\n",
    "X = X.reshape((X.shape[0], 1, X.shape[1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "fe54cbc3",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split the dataset into training and testing sets, ensuring you separate the last 31 records for testing.\n",
    "\n",
    "# Split data into training and test sets\n",
    "X_train, X_test = X[:-31], X[-31:]\n",
    "y_train, y_test = y[:-31], y[-31:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "eed61171",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Design and compile an LSTM model architecture, specifying the layers, activation functions, and loss optimization strategy.\n",
    "\n",
    "# LSTM Model architecture\n",
    "model = Sequential()\n",
    "model.add(LSTM(50, \n",
    "               activation='relu', \n",
    "               input_shape=(X_train.shape[1], \n",
    "                            X_train.shape[2])))\n",
    "model.add(Dense(1))\n",
    "model.compile(optimizer='adam', loss='mae')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "b2cd2c35",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-02-05 12:28:15.265986: W tensorflow/tsl/platform/profile_utils/cpu_utils.cc:128] Failed to get CPU frequency: 0 Hz\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "57/57 [==============================] - 0s 694us/step - loss: 0.0745\n",
      "Epoch 2/50\n",
      "57/57 [==============================] - 0s 588us/step - loss: 0.0308\n",
      "Epoch 3/50\n",
      "57/57 [==============================] - 0s 578us/step - loss: 0.0246\n",
      "Epoch 4/50\n",
      "57/57 [==============================] - 0s 591us/step - loss: 0.0229\n",
      "Epoch 5/50\n",
      "57/57 [==============================] - 0s 605us/step - loss: 0.0214\n",
      "Epoch 6/50\n",
      "57/57 [==============================] - 0s 579us/step - loss: 0.0202\n",
      "Epoch 7/50\n",
      "57/57 [==============================] - 0s 592us/step - loss: 0.0189\n",
      "Epoch 8/50\n",
      "57/57 [==============================] - 0s 604us/step - loss: 0.0177\n",
      "Epoch 9/50\n",
      "57/57 [==============================] - 0s 583us/step - loss: 0.0165\n",
      "Epoch 10/50\n",
      "57/57 [==============================] - 0s 589us/step - loss: 0.0158\n",
      "Epoch 11/50\n",
      "57/57 [==============================] - 0s 594us/step - loss: 0.0155\n",
      "Epoch 12/50\n",
      "57/57 [==============================] - 0s 589us/step - loss: 0.0152\n",
      "Epoch 13/50\n",
      "57/57 [==============================] - 0s 609us/step - loss: 0.0147\n",
      "Epoch 14/50\n",
      "57/57 [==============================] - 0s 586us/step - loss: 0.0147\n",
      "Epoch 15/50\n",
      "57/57 [==============================] - 0s 667us/step - loss: 0.0147\n",
      "Epoch 16/50\n",
      "57/57 [==============================] - 0s 590us/step - loss: 0.0145\n",
      "Epoch 17/50\n",
      "57/57 [==============================] - 0s 576us/step - loss: 0.0145\n",
      "Epoch 18/50\n",
      "57/57 [==============================] - 0s 586us/step - loss: 0.0145\n",
      "Epoch 19/50\n",
      "57/57 [==============================] - 0s 608us/step - loss: 0.0145\n",
      "Epoch 20/50\n",
      "57/57 [==============================] - 0s 574us/step - loss: 0.0143\n",
      "Epoch 21/50\n",
      "57/57 [==============================] - 0s 594us/step - loss: 0.0143\n",
      "Epoch 22/50\n",
      "57/57 [==============================] - 0s 571us/step - loss: 0.0146\n",
      "Epoch 23/50\n",
      "57/57 [==============================] - 0s 609us/step - loss: 0.0147\n",
      "Epoch 24/50\n",
      "57/57 [==============================] - 0s 757us/step - loss: 0.0144\n",
      "Epoch 25/50\n",
      "57/57 [==============================] - 0s 590us/step - loss: 0.0150\n",
      "Epoch 26/50\n",
      "57/57 [==============================] - 0s 598us/step - loss: 0.0144\n",
      "Epoch 27/50\n",
      "57/57 [==============================] - 0s 576us/step - loss: 0.0142\n",
      "Epoch 28/50\n",
      "57/57 [==============================] - 0s 595us/step - loss: 0.0144\n",
      "Epoch 29/50\n",
      "57/57 [==============================] - 0s 582us/step - loss: 0.0142\n",
      "Epoch 30/50\n",
      "57/57 [==============================] - 0s 598us/step - loss: 0.0143\n",
      "Epoch 31/50\n",
      "57/57 [==============================] - 0s 586us/step - loss: 0.0143\n",
      "Epoch 32/50\n",
      "57/57 [==============================] - 0s 598us/step - loss: 0.0146\n",
      "Epoch 33/50\n",
      "57/57 [==============================] - 0s 589us/step - loss: 0.0143\n",
      "Epoch 34/50\n",
      "57/57 [==============================] - 0s 576us/step - loss: 0.0142\n",
      "Epoch 35/50\n",
      "57/57 [==============================] - 0s 593us/step - loss: 0.0143\n",
      "Epoch 36/50\n",
      "57/57 [==============================] - 0s 580us/step - loss: 0.0142\n",
      "Epoch 37/50\n",
      "57/57 [==============================] - 0s 624us/step - loss: 0.0147\n",
      "Epoch 38/50\n",
      "57/57 [==============================] - 0s 601us/step - loss: 0.0142\n",
      "Epoch 39/50\n",
      "57/57 [==============================] - 0s 585us/step - loss: 0.0143\n",
      "Epoch 40/50\n",
      "57/57 [==============================] - 0s 610us/step - loss: 0.0142\n",
      "Epoch 41/50\n",
      "57/57 [==============================] - 0s 572us/step - loss: 0.0143\n",
      "Epoch 42/50\n",
      "57/57 [==============================] - 0s 581us/step - loss: 0.0141\n",
      "Epoch 43/50\n",
      "57/57 [==============================] - 0s 569us/step - loss: 0.0142\n",
      "Epoch 44/50\n",
      "57/57 [==============================] - 0s 579us/step - loss: 0.0144\n",
      "Epoch 45/50\n",
      "57/57 [==============================] - 0s 594us/step - loss: 0.0143\n",
      "Epoch 46/50\n",
      "57/57 [==============================] - 0s 603us/step - loss: 0.0141\n",
      "Epoch 47/50\n",
      "57/57 [==============================] - 0s 581us/step - loss: 0.0142\n",
      "Epoch 48/50\n",
      "57/57 [==============================] - 0s 584us/step - loss: 0.0142\n",
      "Epoch 49/50\n",
      "57/57 [==============================] - 0s 580us/step - loss: 0.0141\n",
      "Epoch 50/50\n",
      "57/57 [==============================] - 0s 578us/step - loss: 0.0144\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x159517d00>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Train the prepared LSTM model using the segmented training data.\n",
    "\n",
    "# Train the LSTM model\n",
    "model.fit(X_train, y_train, epochs=50, verbose=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8d422bbd",
   "metadata": {},
   "source": [
    "# Design, Train, and Evaluate an Advanced LSTM Model with Dropout Layers on Time Series Data\n",
    "\n",
    "<div data-purpose=\"safely-set-inner-html:rich-text-viewer:html\" class=\"lab-rich-text-viewer--lab-html-content--22XsM rt-scaffolding\"><p>Your objective is to develop an advanced LSTM model with precise parameters and layer configurations to enhance the prediction accuracy of a time series dataset. This model should incorporate two LSTM layers with dropout regularization. Ensure you follow the detailed steps below:</p><ul><li><p><strong>Define an Advanced LSTM Model</strong>:</p><ul><li><p>First LSTM layer:</p><ul><li><p>Units: 100</p></li><li><p>Activation function: ReLU</p></li><li><p>Return sequences: True</p></li><li><p>Dropout: 0.2 (20% dropout rate)</p></li></ul></li><li><p>Second LSTM layer:</p><ul><li><p>Units: 50</p></li><li><p>Activation function: ReLU</p></li><li><p>Dropout: 0.2 (20% dropout rate)</p></li></ul></li><li><p>Final Dense layer with one output unit.</p></li></ul></li><li><p><strong>Compile and Train the Advanced LSTM Model</strong>:</p><ul><li><p>Optimizer: Adam</p></li><li><p>Loss function: Mean Absolute Error (MAE)</p></li><li><p>Epochs: 50</p></li></ul></li><li><p><strong>Evaluate and Report the Advanced Model's Performance</strong>:</p><ul><li><p>Compute and compare the Mean Absolute Error (MAE) and the Mean Absolute Percentage Error (MAPE) of the advanced model predictions against the actual values of the last 31 days.</p></li></ul></li></ul></div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "efacfdc0",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import the required metric utilities:\n",
    "\n",
    "# Importing the metric utilities\n",
    "from sklearn.metrics import mean_absolute_error, mean_absolute_percentage_error"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "b34ebff7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 67ms/step\n"
     ]
    }
   ],
   "source": [
    "# Predict on the test data using LSTM model:\n",
    "\n",
    "# Predicting the orders for the last 31 days\n",
    "y_pred = model.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "db75f1fe",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Revert the scaling process for predicted data:\n",
    "\n",
    "# Setting up a zero matrix to facilitate the reverse scaling process\n",
    "num_features = df.shape[1]\n",
    "zero_matrix = np.zeros((y_pred.shape[0], num_features - 1))\n",
    " \n",
    "# Merging the predictions with the zero matrix\n",
    "inverse_input = np.hstack((y_pred, zero_matrix))\n",
    " \n",
    "# Reverting the scaling to obtain predicted values in the original order magnitude\n",
    "predicted_orders = scaler.inverse_transform(inverse_input)[:, 0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "26f23b59",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Retrieve the actual last 31 days of orders for comparison:\n",
    "\n",
    "# Retrieving the actual orders for the last 31 days\n",
    "actual_orders = df['orders'].values[-31:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "ea3d6e6f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculate the performance metrics:\n",
    "\n",
    "# Calculating Mean Absolute Error\n",
    "MAE = mean_absolute_error(actual_orders, predicted_orders)\n",
    " \n",
    "# Calculating Mean Absolute Percentage Error\n",
    "MAPE = mean_absolute_percentage_error(actual_orders, predicted_orders)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "5755a1fb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean Absolute Error (MAE): 1237.1537547269172\n",
      "Mean Absolute Percentage Error (MAPE): 0.05507738709709705%\n"
     ]
    }
   ],
   "source": [
    "# Output the derived esults:\n",
    "\n",
    "# Printing the derived performance metrics\n",
    "print(f\"Mean Absolute Error (MAE): {MAE}\")\n",
    "print(f\"Mean Absolute Percentage Error (MAPE): {MAPE}%\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f45a27d8",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "8942f908",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import the necessary layer:\n",
    "\n",
    "# Importing the Dropout layer from Keras\n",
    "from tensorflow.keras.layers import Dropout"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "64a4855a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define the advanced LSTM model:\n",
    "\n",
    "# Setting up an advanced LSTM architecture\n",
    "advanced_model = Sequential()\n",
    " \n",
    "# First LSTM layer along with Dropout\n",
    "advanced_model.add(LSTM(100, \n",
    "                        activation='relu', \n",
    "                        return_sequences=True,\n",
    "                        input_shape=(X_train.shape[1], X_train.shape[2])))\n",
    "advanced_model.add(Dropout(0.2))\n",
    " \n",
    "# Second LSTM layer followed by Dropout\n",
    "advanced_model.add(LSTM(50, activation='relu'))\n",
    "advanced_model.add(Dropout(0.2))\n",
    " \n",
    "# Final Dense layer\n",
    "advanced_model.add(Dense(1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "ce5baf52",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "57/57 [==============================] - 1s 1ms/step - loss: 0.0644\n",
      "Epoch 2/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0320\n",
      "Epoch 3/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0271\n",
      "Epoch 4/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0252\n",
      "Epoch 5/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0234\n",
      "Epoch 6/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0233\n",
      "Epoch 7/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0212\n",
      "Epoch 8/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0204\n",
      "Epoch 9/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0191\n",
      "Epoch 10/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0188\n",
      "Epoch 11/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0189\n",
      "Epoch 12/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0183\n",
      "Epoch 13/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0184\n",
      "Epoch 14/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0183\n",
      "Epoch 15/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0188\n",
      "Epoch 16/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0184\n",
      "Epoch 17/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0179\n",
      "Epoch 18/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0187\n",
      "Epoch 19/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0174\n",
      "Epoch 20/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0183\n",
      "Epoch 21/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0178\n",
      "Epoch 22/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0178\n",
      "Epoch 23/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0182\n",
      "Epoch 24/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0179\n",
      "Epoch 25/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0179\n",
      "Epoch 26/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0180\n",
      "Epoch 27/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0178\n",
      "Epoch 28/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0180\n",
      "Epoch 29/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0179\n",
      "Epoch 30/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0174\n",
      "Epoch 31/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0174\n",
      "Epoch 32/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0179\n",
      "Epoch 33/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0173\n",
      "Epoch 34/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0175\n",
      "Epoch 35/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0177\n",
      "Epoch 36/50\n",
      "57/57 [==============================] - 0s 2ms/step - loss: 0.0177\n",
      "Epoch 37/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0181\n",
      "Epoch 38/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0177\n",
      "Epoch 39/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0175\n",
      "Epoch 40/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0169\n",
      "Epoch 41/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0177\n",
      "Epoch 42/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0171\n",
      "Epoch 43/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0172\n",
      "Epoch 44/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0172\n",
      "Epoch 45/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0173\n",
      "Epoch 46/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0174\n",
      "Epoch 47/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0173\n",
      "Epoch 48/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0171\n",
      "Epoch 49/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0175\n",
      "Epoch 50/50\n",
      "57/57 [==============================] - 0s 1ms/step - loss: 0.0171\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x15a618580>"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Compile and train the model:\n",
    "\n",
    "# Compiling the model\n",
    "advanced_model.compile(optimizer='adam', loss='mae')\n",
    " \n",
    "# Model training\n",
    "advanced_model.fit(X_train, y_train, epochs=50, verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "efecc8df",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 93ms/step\n"
     ]
    }
   ],
   "source": [
    "# Predict with the advanced model:\n",
    "\n",
    "# Making predictions with the advanced model\n",
    "y_pred_advanced = advanced_model.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "80dbab74",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Revert scaling for predicted data:\n",
    "\n",
    "# Constructing an inverse input matrix with predictions and zero columns\n",
    "inverse_input_advanced = np.hstack((y_pred_advanced, zero_matrix))\n",
    " \n",
    "# Undoing the scaling to get predicted values in the original order magnitude\n",
    "predicted_orders_advanced = scaler.inverse_transform(inverse_input_advanced)[:, 0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "3e1b380a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Advanced Model - Mean Absolute Error (MAE): 1242.1598839732917\n",
      "Advanced Model - Mean Absolute Percentage Error (MAPE): 0.05082466016233073%\n"
     ]
    }
   ],
   "source": [
    "# Calculate the performance metrics for the advanced model:\n",
    "\n",
    "# Calculating the Mean Absolute Error for the advanced model\n",
    "MAE_advanced = mean_absolute_error(actual_orders, predicted_orders_advanced)\n",
    " \n",
    "# Calculating the Mean Absolute Percentage Error for the advanced model\n",
    "MAPE_advanced = mean_absolute_percentage_error(actual_orders, predicted_orders_advanced)\n",
    " \n",
    "# Displaying the computed metrics\n",
    "print(f\"Advanced Model - Mean Absolute Error (MAE): {MAE_advanced}\")\n",
    "print(f\"Advanced Model - Mean Absolute Percentage Error (MAPE): {MAPE_advanced}%\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e7199648",
   "metadata": {},
   "source": [
    "# Implement and Evaluate an Advanced LSTM Model using Rolling Forecast Evaluation\n",
    "\n",
    "<div data-purpose=\"safely-set-inner-html:rich-text-viewer:html\" class=\"lab-rich-text-viewer--lab-html-content--22XsM rt-scaffolding\"><p>You are required to implement and assess an advanced LSTM model utilizing a rolling forecast approach for time series data. This method involves making successive forecasts and advancing the training window over the dataset. The forecasts should be performed for the next 30 days at every 15-day interval during 2022. This is to observe the model's capability to generalize over varying temporal subsets of data. Here are the specifics:</p><ul><li><p><strong>Create an Advanced LSTM Model</strong>:</p><ul><li><p>Define a function, <code><strong>create_advanced_lstm_model</strong></code>, which returns an LSTM model.</p></li><li><p>First LSTM layer:</p><ul><li><p>Units: 50</p></li><li><p>Activation function: ReLU</p></li><li><p>Return sequences: True</p></li></ul></li><li><p>Second LSTM layer:</p><ul><li><p>Units: 50</p></li><li><p>Activation function: ReLU</p></li></ul></li><li><p>First Dense layer:</p><ul><li><p>Units: 25</p></li><li><p>Activation function: ReLU</p></li></ul></li><li><p>Second Dense layer with one output unit.</p></li><li><p>Compilation:</p><ul><li><p>Optimizer: Adam</p></li><li><p>Loss function: Mean Absolute Error (MAE)</p></li></ul></li></ul></li><li><p><strong>Implement the Rolling Forecast Evaluation</strong>:</p><ul><li><p>Define a function, <code><strong>rolling_forecast_evaluation_advanced</strong></code>, with parameters:</p><ul><li><p>df: input dataframe.</p></li><li><p>start_date: beginning date for the rolling forecast.</p></li><li><p>forecast_length (default value of 30): duration of the forecast.</p></li><li><p>step (default value of 15): interval to slide the window.</p></li></ul></li><li><p>Within this function:</p><ul><li><p>Iterate over the dataset by advancing the training window.</p></li><li><p>For each window, scale the data using MinMaxScaler.</p></li><li><p>Split the data into training and test subsets.</p></li><li><p>Train the LSTM model on the training subset.</p></li><li><p>Generate predictions for the test subset.</p></li><li><p>Compute the MAE and MAPE for each forecast.</p></li><li><p>Return the MAEs and MAPEs for each forecast window.</p></li></ul></li></ul></li><li><p><strong>Evaluate the Model</strong>:</p><ul><li><p>Report the MAEs and MAPEs for each window and their averages.</p><p><br></p></li></ul></li></ul></div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "e3be1f74",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing necessary libraries:\n",
    "\n",
    "from datetime import timedelta"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "118f5476",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define the advanced LSTM model:\n",
    "\n",
    "def create_advanced_lstm_model(input_shape):\n",
    "    model = Sequential()\n",
    "    # Adding the first LSTM layer with 50 units and ReLU activation.\n",
    "    model.add(LSTM(50, activation='relu', return_sequences=True, input_shape=input_shape))\n",
    "    \n",
    "    # Adding the second LSTM layer with 50 units and ReLU activation.\n",
    "    model.add(LSTM(50, activation='relu'))\n",
    "    \n",
    "    # Adding a dense layer with 25 units and ReLU activation.\n",
    "    model.add(Dense(25, activation='relu'))\n",
    "    \n",
    "    # Adding the output dense layer with a single unit.\n",
    "    model.add(Dense(1))\n",
    "    \n",
    "    # Compiling the LSTM model.\n",
    "    model.compile(optimizer='adam', loss='mae')\n",
    "    \n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "f61c9e07",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Implement the rolling forecast evaluation:\n",
    "\n",
    "def rolling_forecast_evaluation_advanced(df, start_date, forecast_length=30, step=15):\n",
    "    MAEs = []  # List to store Mean Absolute Errors.\n",
    "    MAPEs = []  # List to store Mean Absolute Percentage Errors.\n",
    " \n",
    "    current_date = pd.to_datetime(start_date)\n",
    "    \n",
    "    # Continue as long as the end date doesn't exceed the final date in the dataset.\n",
    "    while current_date < df.index[-1] - timedelta(days=forecast_length):\n",
    "        \n",
    "        # Define training and test periods based on current_date.\n",
    "        train_end = current_date\n",
    "        test_start = train_end + timedelta(days=1)\n",
    "        test_end = test_start + timedelta(days=forecast_length - 1)\n",
    " \n",
    "        # Split data into training and test sets.\n",
    "        train = df[:train_end]\n",
    "        test = df[test_start:test_end]\n",
    " \n",
    "        # Normalize train and test data.\n",
    "        scaler = MinMaxScaler()\n",
    "        train_scaled = scaler.fit_transform(train)\n",
    "        test_scaled = scaler.transform(test)\n",
    " \n",
    "        # Create train and test sets for X (features) and y (target).\n",
    "        X_train, y_train = train_scaled[:, 1:], train_scaled[:, 0]\n",
    "        X_test, y_test = test_scaled[:, 1:], test_scaled[:, 0]\n",
    " \n",
    "        # Adjust data shape for LSTM input.\n",
    "        X_train = X_train.reshape((X_train.shape[0], 1, X_train.shape[1]))\n",
    "        X_test = X_test.reshape((X_test.shape[0], 1, X_test.shape[1]))\n",
    " \n",
    "        # Create and train the model.\n",
    "        model = create_advanced_lstm_model((X_train.shape[1], X_train.shape[2]))\n",
    "        model.fit(X_train, y_train, epochs=50, verbose=0)\n",
    " \n",
    "        # Forecast with the trained model.\n",
    "        y_pred = model.predict(X_test)\n",
    "        \n",
    "        # Process the predictions for inverse scaling.\n",
    "        inverse_input = np.hstack((y_pred, np.zeros((y_pred.shape[0], df.shape[1]-1))))\n",
    "        predicted = scaler.inverse_transform(inverse_input)[:, 0]\n",
    "        actual = test['orders'].values\n",
    " \n",
    "        # Calculate error metrics.\n",
    "        MAE = mean_absolute_error(actual, predicted)\n",
    "        MAPE = mean_absolute_percentage_error(actual, predicted)\n",
    "        \n",
    "        # Append the errors to their respective lists.\n",
    "        MAEs.append(MAE)\n",
    "        MAPEs.append(MAPE)\n",
    " \n",
    "        # Advance the current_date by the specified step.\n",
    "        current_date += timedelta(days=step)\n",
    " \n",
    "    return MAEs, MAPEs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "5a8b113c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "WARNING:tensorflow:5 out of the last 5 calls to <function Model.make_predict_function.<locals>.predict_function at 0x10c1365f0> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has reduce_retracing=True option that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "WARNING:tensorflow:6 out of the last 6 calls to <function Model.make_predict_function.<locals>.predict_function at 0x15ba37e20> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has reduce_retracing=True option that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 94ms/step\n",
      "1/1 [==============================] - 0s 97ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 93ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 91ms/step\n",
      "1/1 [==============================] - 0s 90ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n",
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    }
   ],
   "source": [
    "# Use the function and evaluate the model:\n",
    "\n",
    "# Set the start date for the rolling forecast evaluation.\n",
    "start_date = '2022-01-01'\n",
    "MAEs, MAPEs = rolling_forecast_evaluation_advanced(df, start_date)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "f9acdb21",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average MAE: 1548.597003965261\n",
      "Average MAPE: 0.06928386893364587%\n"
     ]
    }
   ],
   "source": [
    "# View the outcome of the cross-validation:\n",
    "\n",
    "# Display the MAE and MAPE for each evaluation window and their averages.\n",
    "print(f\"Average MAE: {np.mean(MAEs)}\")\n",
    "print(f\"Average MAPE: {np.mean(MAPEs)}%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6dd5ce4a",
   "metadata": {},
   "source": [
    "# Hyperparameter Tuning for LSTM Model\n",
    "\n",
    "<div data-purpose=\"safely-set-inner-html:rich-text-viewer:html\" class=\"lab-rich-text-viewer--lab-html-content--22XsM rt-scaffolding\"><p>Conduct a hyperparameter tuning process to optimize the structure of an LSTM neural network designed to predict a target value based on 2022 data. Develop a function, <code><strong>create_tuned_model</strong></code>, that defines an LSTM model based on specific hyperparameters. This model consists of two LSTM layers followed by two Dense layers, with the ability to adjust hyperparameters like the number of neurons, the optimizer type, and the dropout rate. After defining the model structure, the model should be compiled using the specified optimizer and a loss function set to Mean Absolute Error (MAE). Return the compiled model.</p><p><strong>Function: create_tuned_model</strong></p><ul><li><p><strong>Layers &amp; Structure</strong>:</p><ol><li><p><strong>LSTM Layer</strong>:</p><ul><li><p>Number of neurons: <code><strong>neurons</strong></code>.</p></li><li><p>Activation function: <code><strong>ReLU</strong></code>.</p></li><li><p>Configuration: This layer is set to return sequences, making it suitable to feed into another LSTM layer.</p></li><li><p>Input Shape: Depends on the shape of the training data (specifically, <code><strong>X_train.shape[1]</strong></code> is the number of time steps and <code><strong>X_train.shape[2]</strong></code> is the number of features per time step).</p></li></ul></li><li><p><strong>LSTM Layer</strong>:</p><ul><li><p>Number of neurons: <code><strong>neurons</strong></code>.</p></li><li><p>Activation function: <code><strong>ReLU</strong></code>.</p></li></ul></li><li><p><strong>Dense Layer</strong>:</p><ul><li><p>Number of neurons: <code><strong>neurons//2</strong></code>.</p></li><li><p>Activation function: <code><strong>ReLU</strong></code>.</p></li></ul></li><li><p><strong>Dense Layer</strong> (Output Layer):</p><ul><li><p>Number of neurons: 1. This implies that it's a regression task, predicting a single continuous value.</p></li><li><p>Activation function: None specified, meaning it will be a linear activation.</p></li></ul></li></ol></li><li><p><strong>Compilation</strong>:</p><ul><li><p><strong>Optimizer</strong>: Specified by the <code><strong>optimizer</strong></code> parameter.</p></li><li><p><strong>Loss Function</strong>: Mean Absolute Error (MAE).</p></li></ul></li><li><p><strong>Parameters</strong>:</p><ul><li><p><strong>neurons</strong> (int, default: 50): Number of neurons for the LSTM layers. Represents the dimensionality of the output space.</p></li><li><p><strong>optimizer</strong> (class, default: Adam): The optimizer class to be used during model compilation.</p></li><li><p><strong>dropout_rate</strong> (float, default: 0.0): Although the parameter exists, it's not applied in the layers in the provided code. Ideally, it should define the fraction of the inputs to drop, typically added to prevent overfitting.</p></li></ul></li><li><p><strong>Returns</strong>:</p><ul><li><p>A compiled LSTM model constructed based on the provided hyperparameters.</p></li></ul></li></ul><p><strong>Parameters to be tuned</strong>:</p><ul><li><p><strong>neurons</strong>: List of integers representing the number of neurons to be tested ( [30, 50]).</p></li><li><p><strong>optimizer</strong>: List of optimizer classes to be tested ([SGD, Adam]).</p></li><li><p><strong>dropout_rate</strong>: List of dropout rates to be tested ([0.0, 0.2]).</p><p><br></p></li></ul></div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "b65c3f40",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import the optimizers to be tested\n",
    "\n",
    "from tensorflow.keras.optimizers import SGD, Adam"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "bf032939",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create the LSTM Model Function based on provided hyperparameters:\n",
    "\n",
    "def create_tuned_model(neurons=50, optimizer=Adam, dropout_rate=0.0):\n",
    "    # Initialize a sequential model\n",
    "    model = Sequential()\n",
    "    \n",
    "    # Add the first LSTM layer with the 'relu' activation function\n",
    "    model.add(LSTM(neurons, activation='relu', return_sequences=True, input_shape=(X_train.shape[1], X_train.shape[2])))\n",
    "    \n",
    "    # Add the second LSTM layer with the 'relu' activation function\n",
    "    model.add(LSTM(neurons, activation='relu'))\n",
    "    \n",
    "    # Add a Dense layer which has half the number of neurons compared to the LSTM layers and uses the 'relu' activation function\n",
    "    model.add(Dense(neurons//2, activation='relu'))\n",
    "    \n",
    "    # Add an output Dense layer to produce a single output value\n",
    "    model.add(Dense(1))\n",
    "    \n",
    "    # Compile the model with the provided optimizer and Mean Absolute Error as the loss function\n",
    "    model.compile(optimizer=optimizer(), loss='mae')\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "8070c2f5",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define Hyperparameters for Tuning and combine the hyperparameters to test all possible combinations:\n",
    "\n",
    "# Define hyperparameters to be tested\n",
    "neurons = [30, 50]\n",
    "optimizer_classes = [SGD, Adam]\n",
    "dropout_rate = [0.0, 0.2]\n",
    " \n",
    "# Generate all combinations of hyperparameters\n",
    "all_params = [(n, o, d) for n in neurons for o in optimizer_classes for d in dropout_rate]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "5d25f90e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Set up variables to track the best hyperparameters and their respective scores:\n",
    "\n",
    "# Initialize variables to track best hyperparameters and scores\n",
    "best_mae_params = None\n",
    "best_mae_score = float('inf')\n",
    "best_mape_params = None\n",
    "best_mape_score = float('inf')\n",
    "results = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "2e9df941",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define the initial date for the time series and calculate the number of required splits:\n",
    "\n",
    "# Define starting date and calculate number of time series splits\n",
    "start_date = pd.Timestamp('2022-01-01')\n",
    "num_splits = (365 - 30) // 15"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "3f67c5ac",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "30 0 0.0\n",
      "30 0 0.2\n",
      "30 0 0.0\n",
      "30 0 0.2\n",
      "50 0 0.0\n",
      "50 0 0.2\n",
      "50 0 0.0\n",
      "50 0 0.2\n"
     ]
    }
   ],
   "source": [
    "# Start by structuring the loop to go through all hyperparameter combinations:\n",
    "\n",
    "# Loop over all hyperparameter combinations\n",
    "for params in all_params:\n",
    "    total_mae = 0.0\n",
    "    total_mape = 0.0\n",
    "    n, o, d = params\n",
    "    print(n, 0, d)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "44333862",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 93ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 97ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 93ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 290ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 96ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 96ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 97ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 95ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 96ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 94ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 93ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 93ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 150ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 94ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 98ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 93ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 97ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 100ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 332ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 93ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 97ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 315ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 93ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 101ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 99ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 90ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 94ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 93ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 95ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 316ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 93ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 91ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 92ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 93ms/step\n"
     ]
    }
   ],
   "source": [
    "# Perform cross-validation, data preprocessing, model training, prediction, and performance evaluation.\n",
    "\n",
    "# Loop over all hyperparameter combinations\n",
    "for params in all_params:\n",
    "    total_mae = 0.0\n",
    "    total_mape = 0.0\n",
    "    n, o, d = params\n",
    "    \n",
    "    # Sliding window cross-validation over 2022 data\n",
    "    for i in range(0, 365 - 30, 15): \n",
    "        split_date = start_date + pd.Timedelta(days=i)\n",
    "        train = df[df.index < split_date]\n",
    "        test = df[(df.index >= split_date) & (df.index < split_date + pd.Timedelta(days=30))]\n",
    "        \n",
    "        # Preprocessing, scaling\n",
    "        scaler = MinMaxScaler()\n",
    "        train_scaled = scaler.fit_transform(train)\n",
    "        test_scaled = scaler.transform(test)\n",
    "        \n",
    "        X_train, y_train = train_scaled[:, 1:], train_scaled[:, 0]\n",
    "        X_test, y_test = test_scaled[:, 1:], test_scaled[:, 0]\n",
    "        \n",
    "        X_train = X_train.reshape((X_train.shape[0], 1, X_train.shape[1]))\n",
    "        X_test = X_test.reshape((X_test.shape[0], 1, X_test.shape[1]))\n",
    "        \n",
    "        # Create and fit the model\n",
    "        model = create_tuned_model(neurons=n, optimizer=o, dropout_rate=d)\n",
    "        model.fit(X_train, y_train, epochs=50, verbose=0)\n",
    "        \n",
    "        # Predict and evaluate\n",
    "        y_pred = model.predict(X_test)\n",
    "        inverse_input = np.hstack((y_pred, np.zeros((y_pred.shape[0], df.shape[1] - 1))))\n",
    "        y_pred_rescaled = scaler.inverse_transform(inverse_input)[:, 0]\n",
    "        \n",
    "        # Convert predictions and test targets back to the original scale\n",
    "        inverse_input_test = np.hstack((y_test.reshape(-1, 1), np.zeros((y_test.shape[0], df.shape[1] - 1))))\n",
    "        y_test_rescaled = scaler.inverse_transform(inverse_input_test)[:, 0]\n",
    "        \n",
    "        # Compute the evaluation metrics.\n",
    "        mae = mean_absolute_error(y_test_rescaled, y_pred_rescaled)\n",
    "        mape = mean_absolute_percentage_error(y_test_rescaled, y_pred_rescaled)\n",
    " \n",
    "        # adding the MAPE and MAE for each fold\n",
    "        total_mae += mae\n",
    "        total_mape += mape\n",
    " \n",
    "    # Computing the Average MAE and MAPE\n",
    "    avg_mae = total_mae / num_splits\n",
    "    avg_mape = total_mape / num_splits\n",
    " \n",
    "    # Storing the results\n",
    "    results.append({'params': params,\n",
    "                    'avg_mae': avg_mae,\n",
    "                    'avg_mape': avg_mape\n",
    "                   })\n",
    " \n",
    "    if avg_mae < best_mae_score:\n",
    "        best_mae_score = avg_mae\n",
    "        best_mae_params = params\n",
    "    \n",
    "    if avg_mape < best_mape_score:\n",
    "        best_mape_score = avg_mape\n",
    "        best_mape_params = params"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "40d5c7bc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best MAE: 1500.068371973579\n",
      "Best MAPE: 0.06622386084863913\n",
      "Best Hyperparameters for MAE - Neurons: 50, Optimizer: Adam, Dropout Rate: 0.2\n"
     ]
    }
   ],
   "source": [
    "# Display the best hyperparameters based on evaluation metrics:\n",
    "\n",
    "# Print the best MAE and MAPE scores with their respective hyperparameters\n",
    "print(f\"Best MAE: {best_mae_score}\")\n",
    "print(f\"Best MAPE: {best_mape_score}\")\n",
    "print(f\"Best Hyperparameters for MAE - Neurons: {best_mae_params[0]}, Optimizer: {best_mae_params[1].__name__}, Dropout Rate: {best_mae_params[2]}\")"
   ]
  }
 ],
 "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": "3.10.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}