{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Crude Oil Price Forecasting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 194,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import scipy\n",
    "import torch\n",
    "import os\n",
    "import scipy\n",
    "import datetime\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Data Preparation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 206,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Dated brent data\n",
    "dated_brent_data = pd.read_csv(os.path.join('data', \n",
    "                                            'dated_brent_allbate.csv')\n",
    "                              ).rename({'assessDate':'Date'}, axis=1\n",
    "                                      ).drop(columns=['Unnamed: 0', \n",
    "                                                      'isCorrected', \n",
    "                                                      'modDate', \n",
    "                                                      'symbol'])\n",
    "\n",
    "dated_c = dated_brent_data[dated_brent_data['bate'] == 'c'].drop(columns=['bate']).rename({'value':'Closing'}, axis=1)\n",
    "dated_h = dated_brent_data[dated_brent_data['bate'] == 'h'].drop(columns=['bate']).rename({'value':'High'}, axis=1)\n",
    "dated_l = dated_brent_data[dated_brent_data['bate'] == 'l'].drop(columns=['bate']).rename({'value':'Low'}, axis=1)\n",
    "\n",
    "dated_brent_data = dated_c.merge(dated_h, on=\"Date\").merge(dated_l, on=\"Date\")\n",
    "dated_brent_data['Date'] = pd.to_datetime(dated_brent_data['Date'])\n",
    "\n",
    "# Date conversion\n",
    "\n",
    "def convert_date(date):\n",
    "    format = '%m/%d/%Y'\n",
    "    date_converted = datetime.datetime.strptime(date, format).date()\n",
    "\n",
    "    return date_converted\n",
    "\n",
    "# # Crude oil volatility data\n",
    "# crude_volatility_data = pd.read_csv(os.path.join('data', 'cboe_ovx_futures.csv')\n",
    "#                                    ).drop(columns=['Unnamed: 0', 'Volume', 'Open'])\n",
    "# crude_volatility_data['Date'] = pd.to_datetime(crude_volatility_data['Date'].apply(convert_date))\n",
    "\n",
    "# News data\n",
    "\n",
    "# --- #\n",
    "\n",
    "# Brent futures data\n",
    "brent_futures_data = pd.read_csv(os.path.join('data', 'brent_nmx.csv')\n",
    "                                ).drop(columns=['Open', 'Volume']).rename({'Close/Last':'Close'},axis=1)\n",
    "brent_futures_data['Date'] = pd.to_datetime(brent_futures_data['Date'].apply(convert_date))\n",
    "\n",
    "# Gasoline futures data\n",
    "gasoline_data = pd.read_csv(os.path.join('data', 'gasoline_nmx.csv')\n",
    "                           ).drop(columns=['Open', 'Volume']).rename({'Close/Last':'Close'},axis=1)\n",
    "gasoline_data['Date'] = pd.to_datetime(gasoline_data['Date'].apply(convert_date))\n",
    "\n",
    "# Dollar Index data\n",
    "dollar_data = pd.read_csv(os.path.join('data', 'dollar_index.csv')\n",
    "                         ).drop(columns=['Adj Close', 'Volume', 'Open']).dropna().drop_duplicates()\n",
    "dollar_data['Date'] = pd.to_datetime(dollar_data['Date'])\n",
    "\n",
    "# U.S. Crude Oil Production data\n",
    "production_data = pd.read_csv(os.path.join('data', 'U.S._Crude_Production_ThousandPerDay.csv'), delimiter = ',')\n",
    "\n",
    "# GPR \n",
    "gpr_data = pd.read_excel(os.path.join('data', 'data_gpr_daily_recent.xlsx')).drop(columns=['event', 'var_name', 'var_label', 'N10D', 'DAY']).rename({'date':'Date'}, axis=1)\n",
    "\n",
    "# Gold Price\n",
    "gold_price_data = pd.read_csv(os.path.join('data', 'GoldPrice.csv')\n",
    "                                ).drop(columns=['Open', 'Volume']).rename({'Close/Last':'Close'},axis=1)\n",
    "gold_price_data['Date'] = pd.to_datetime(gold_price_data['Date'].apply(convert_date))\n",
    "\n",
    "# Silver Price\n",
    "silver_price_data = pd.read_csv(os.path.join('data', 'Silver(CMX).csv')\n",
    "                                ).drop(columns=['Open', 'Volume']).rename({'Close/Last':'Close'},axis=1)\n",
    "silver_price_data['Date'] = pd.to_datetime(silver_price_data['Date'].apply(convert_date))\n",
    "\n",
    "# Platinum Price\n",
    "platinum_price_data = pd.read_csv(os.path.join('data', 'Platinum(NMX).csv')\n",
    "                                ).drop(columns=['Open', 'Volume']).rename({'Close/Last':'Close'},axis=1)\n",
    "platinum_price_data['Date'] = pd.to_datetime(platinum_price_data['Date'].apply(convert_date))\n",
    "\n",
    "# Platinum Price\n",
    "palladium_price_data = pd.read_csv(os.path.join('data', 'Palladium(NMX).csv')\n",
    "                                ).drop(columns=['Open', 'Volume']).rename({'Close/Last':'Close'},axis=1)\n",
    "palladium_price_data['Date'] = pd.to_datetime(palladium_price_data['Date'].apply(convert_date))\n",
    "\n",
    "\n",
    "# US Bond Rate\n",
    "us_bond_data = pd.read_csv(os.path.join('data', 'USBondRate.csv')\n",
    "                                ).drop(columns=['Open', 'Volume']).rename({'Close/Last':'Close'},axis=1)\n",
    "us_bond_data['Date'] = pd.to_datetime(us_bond_data['Date'].apply(convert_date))\n",
    "\n",
    "# S&P500\n",
    "sp500_data = pd.read_csv(os.path.join('data', 'S&P500 (SPX).csv')\n",
    "                                ).drop(columns=['Open']).rename({'Close/Last':'Close'},axis=1)\n",
    "sp500_data['Date'] = pd.to_datetime(sp500_data['Date'].apply(convert_date))\n",
    "\n",
    "# EUR to USD\n",
    "eur_usd_data = pd.read_csv(os.path.join('data', 'EURUSD.csv')\n",
    "                                ).drop(columns=['Open', 'Volume']).rename({'Close/Last':'Close'},axis=1)\n",
    "eur_usd_data['Date'] = pd.to_datetime(eur_usd_data['Date'].apply(convert_date))\n",
    "\n",
    "# Gold miners\n",
    "gold_miners_data = pd.read_csv(os.path.join('data', 'GoldMiners(GDX).csv')\n",
    "                                ).drop(columns=['Open', 'Volume']).rename({'Close/Last':'Close'},axis=1)\n",
    "gold_miners_data['Date'] = pd.to_datetime(gold_miners_data['Date'].apply(convert_date))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Predictor Variables:\n",
    "1. Silver Price\n",
    "2. Platinum Price\n",
    "3. Palladium Price\n",
    "4. Gold Price\n",
    "5. Gold Miners Price\n",
    "6. S&P500 Price\n",
    "7. Dollar Index (Closing), Percent Change -> NASDAQ\n",
    "8. Gasoline (Closing), Percent Change -> NASDAQ\n",
    "9. ? U.S. Crude Oil Production, MioBarrels/Day -> U.S. Energy Information Administration (EIA)\n",
    "10. GPR Data, Percent Change -> Economic Policy Uncertainty (EPU)\n",
    "11. US Bond Rate\n",
    "12. ? Euro to USD\n",
    "\n",
    "Target Variable:\n",
    "Brent Futures (Closing), $ (MA30) -> NASDAQ\n",
    "Dated Brent Price -> S&P Global Commodity Insights"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 207,
   "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>date</th>\n",
       "      <th>Gas</th>\n",
       "      <th>DXY</th>\n",
       "      <th>Gold(CMX)</th>\n",
       "      <th>Silver(CMX)</th>\n",
       "      <th>Platinum(NMX)</th>\n",
       "      <th>Palladium(NMX)</th>\n",
       "      <th>USBond</th>\n",
       "      <th>S&amp;P500</th>\n",
       "      <th>GoldMiners</th>\n",
       "      <th>GPRD</th>\n",
       "      <th>BrFu</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2024-08-05</td>\n",
       "      <td>2.3336</td>\n",
       "      <td>102.690002</td>\n",
       "      <td>2444.4</td>\n",
       "      <td>27.207</td>\n",
       "      <td>915.5</td>\n",
       "      <td>826.10</td>\n",
       "      <td>125.4062</td>\n",
       "      <td>5186.33</td>\n",
       "      <td>35.34</td>\n",
       "      <td>210.206711</td>\n",
       "      <td>72.94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2024-08-02</td>\n",
       "      <td>2.3176</td>\n",
       "      <td>103.209999</td>\n",
       "      <td>2469.8</td>\n",
       "      <td>28.392</td>\n",
       "      <td>967.6</td>\n",
       "      <td>882.50</td>\n",
       "      <td>125.0625</td>\n",
       "      <td>5346.56</td>\n",
       "      <td>36.48</td>\n",
       "      <td>207.799438</td>\n",
       "      <td>73.52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2024-08-01</td>\n",
       "      <td>2.3980</td>\n",
       "      <td>104.419998</td>\n",
       "      <td>2480.8</td>\n",
       "      <td>28.477</td>\n",
       "      <td>970.5</td>\n",
       "      <td>895.10</td>\n",
       "      <td>122.5312</td>\n",
       "      <td>5446.68</td>\n",
       "      <td>37.27</td>\n",
       "      <td>139.878098</td>\n",
       "      <td>76.93</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2024-07-31</td>\n",
       "      <td>2.4425</td>\n",
       "      <td>104.099998</td>\n",
       "      <td>2473.0</td>\n",
       "      <td>28.938</td>\n",
       "      <td>986.4</td>\n",
       "      <td>925.20</td>\n",
       "      <td>120.7812</td>\n",
       "      <td>5522.30</td>\n",
       "      <td>37.93</td>\n",
       "      <td>135.206848</td>\n",
       "      <td>77.91</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2024-07-30</td>\n",
       "      <td>2.3443</td>\n",
       "      <td>104.550003</td>\n",
       "      <td>2451.9</td>\n",
       "      <td>28.525</td>\n",
       "      <td>971.0</td>\n",
       "      <td>881.70</td>\n",
       "      <td>120.1875</td>\n",
       "      <td>5436.44</td>\n",
       "      <td>36.95</td>\n",
       "      <td>95.696396</td>\n",
       "      <td>74.73</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2496</th>\n",
       "      <td>2014-08-29</td>\n",
       "      <td>2.6229</td>\n",
       "      <td>82.750000</td>\n",
       "      <td>1287.4</td>\n",
       "      <td>19.492</td>\n",
       "      <td>1424.7</td>\n",
       "      <td>909.55</td>\n",
       "      <td>140.0938</td>\n",
       "      <td>2003.37</td>\n",
       "      <td>26.69</td>\n",
       "      <td>157.336334</td>\n",
       "      <td>95.96</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2497</th>\n",
       "      <td>2014-08-28</td>\n",
       "      <td>2.5908</td>\n",
       "      <td>82.480003</td>\n",
       "      <td>1290.4</td>\n",
       "      <td>19.609</td>\n",
       "      <td>1425.2</td>\n",
       "      <td>898.10</td>\n",
       "      <td>141.7188</td>\n",
       "      <td>1996.74</td>\n",
       "      <td>26.46</td>\n",
       "      <td>97.534706</td>\n",
       "      <td>94.55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2498</th>\n",
       "      <td>2014-08-27</td>\n",
       "      <td>2.5905</td>\n",
       "      <td>82.430000</td>\n",
       "      <td>1283.4</td>\n",
       "      <td>19.475</td>\n",
       "      <td>1419.9</td>\n",
       "      <td>894.70</td>\n",
       "      <td>141.1562</td>\n",
       "      <td>2000.12</td>\n",
       "      <td>26.11</td>\n",
       "      <td>143.082108</td>\n",
       "      <td>93.88</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2499</th>\n",
       "      <td>2014-08-26</td>\n",
       "      <td>2.6001</td>\n",
       "      <td>82.650002</td>\n",
       "      <td>1285.2</td>\n",
       "      <td>19.459</td>\n",
       "      <td>1419.6</td>\n",
       "      <td>890.15</td>\n",
       "      <td>140.5312</td>\n",
       "      <td>2000.02</td>\n",
       "      <td>26.19</td>\n",
       "      <td>118.143707</td>\n",
       "      <td>93.86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2500</th>\n",
       "      <td>2014-08-25</td>\n",
       "      <td>2.5954</td>\n",
       "      <td>82.550003</td>\n",
       "      <td>1278.9</td>\n",
       "      <td>19.431</td>\n",
       "      <td>1418.4</td>\n",
       "      <td>891.50</td>\n",
       "      <td>140.7188</td>\n",
       "      <td>1997.92</td>\n",
       "      <td>25.62</td>\n",
       "      <td>167.600159</td>\n",
       "      <td>93.35</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2501 rows × 12 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           date     Gas         DXY  Gold(CMX)  Silver(CMX)  Platinum(NMX)  \\\n",
       "0    2024-08-05  2.3336  102.690002     2444.4       27.207          915.5   \n",
       "1    2024-08-02  2.3176  103.209999     2469.8       28.392          967.6   \n",
       "2    2024-08-01  2.3980  104.419998     2480.8       28.477          970.5   \n",
       "3    2024-07-31  2.4425  104.099998     2473.0       28.938          986.4   \n",
       "4    2024-07-30  2.3443  104.550003     2451.9       28.525          971.0   \n",
       "...         ...     ...         ...        ...          ...            ...   \n",
       "2496 2014-08-29  2.6229   82.750000     1287.4       19.492         1424.7   \n",
       "2497 2014-08-28  2.5908   82.480003     1290.4       19.609         1425.2   \n",
       "2498 2014-08-27  2.5905   82.430000     1283.4       19.475         1419.9   \n",
       "2499 2014-08-26  2.6001   82.650002     1285.2       19.459         1419.6   \n",
       "2500 2014-08-25  2.5954   82.550003     1278.9       19.431         1418.4   \n",
       "\n",
       "      Palladium(NMX)    USBond   S&P500  GoldMiners        GPRD   BrFu  \n",
       "0             826.10  125.4062  5186.33       35.34  210.206711  72.94  \n",
       "1             882.50  125.0625  5346.56       36.48  207.799438  73.52  \n",
       "2             895.10  122.5312  5446.68       37.27  139.878098  76.93  \n",
       "3             925.20  120.7812  5522.30       37.93  135.206848  77.91  \n",
       "4             881.70  120.1875  5436.44       36.95   95.696396  74.73  \n",
       "...              ...       ...      ...         ...         ...    ...  \n",
       "2496          909.55  140.0938  2003.37       26.69  157.336334  95.96  \n",
       "2497          898.10  141.7188  1996.74       26.46   97.534706  94.55  \n",
       "2498          894.70  141.1562  2000.12       26.11  143.082108  93.88  \n",
       "2499          890.15  140.5312  2000.02       26.19  118.143707  93.86  \n",
       "2500          891.50  140.7188  1997.92       25.62  167.600159  93.35  \n",
       "\n",
       "[2501 rows x 12 columns]"
      ]
     },
     "execution_count": 207,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictor = gasoline_data.drop(columns=['High', 'Low']).rename({'Close' : 'Gas'}, axis=1).merge(\n",
    "    dollar_data.drop(columns=['High', 'Low']).rename({'Close' : 'DXY'}, axis=1), on='Date').merge(\n",
    "    gold_price_data.drop(columns=['High', 'Low']).rename({'Close' : 'Gold(CMX)'}, axis=1), on='Date').merge(\n",
    "    silver_price_data.drop(columns=['High', 'Low']).rename({'Close' : 'Silver(CMX)'}, axis=1), on='Date').merge(\n",
    "    platinum_price_data.drop(columns=['High', 'Low']).rename({'Close' : 'Platinum(NMX)'}, axis=1), on='Date').merge(\n",
    "    palladium_price_data.drop(columns=['High', 'Low']).rename({'Close' : 'Palladium(NMX)'}, axis=1), on='Date').merge(\n",
    "    us_bond_data.drop(columns=['High', 'Low']).rename({'Close' : 'USBond'}, axis=1), on='Date').merge(\n",
    "    sp500_data.drop(columns=['High', 'Low']).rename({'Close' : 'S&P500'}, axis=1), on='Date').merge(\n",
    "    gold_miners_data.drop(columns=['High', 'Low']).rename({'Close' : 'GoldMiners'}, axis=1), on='Date').merge(\n",
    "    gpr_data[['Date', 'GPRD']])\n",
    "\n",
    "target = brent_futures_data[['Date','Close']].rename({'Close' : 'BrFu'}, axis=1)\n",
    "\n",
    "df = predictor.merge(target, on ='Date').rename({'Date' : 'date'}, axis=1)\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Modelling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 208,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from neuralforecast.core import NeuralForecast\n",
    "from neuralforecast.models import TSMixer, TSMixerx, NHITS, MLPMultivariate, NBEATSx\n",
    "from neuralforecast.losses.pytorch import MSE, MAE, MAPE\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "import numpy as np\n",
    "import os"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 209,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Functions\n",
    "\n",
    "def createLag(data, amt=10):\n",
    "    \"\"\"\n",
    "    Create a lag inside dataframe, in business days\n",
    "\n",
    "    Input:\n",
    "        data -> Pandas dataframe \n",
    "        amt -> int\n",
    "\n",
    "    Output:\n",
    "        Copy of pandas Dataframe\n",
    "    \"\"\"\n",
    "    if 'ds' in data:\n",
    "        # Ensure 'ds' is a datetime column\n",
    "        data['ds'] = pd.to_datetime(data['ds'], errors='coerce')\n",
    "        \n",
    "        # Check for any null values after conversion\n",
    "        if data['ds'].isnull().any():\n",
    "            print(\"Warning: Some dates couldn't be converted to datetime.\")\n",
    "\n",
    "        copy = data.copy()\n",
    "        # Apply the business day offset\n",
    "        copy['ds'] = copy['ds'] + pd.tseries.offsets.BusinessDay(amt)\n",
    "        return copy\n",
    "    else:\n",
    "        print(\"No 'ds' column found inside dataframe\")\n",
    "        return data\n",
    "\n",
    "def trainTestValSplit(data, test_size, val_size):\n",
    "    \"\"\"\n",
    "    Splits data into train-test-validation sets\n",
    "\n",
    "    Input:\n",
    "        data -> Pandas dataframe\n",
    "        test_size -> Proportion of data for test set\n",
    "        val_size -> Proportiion of data fro validation set\n",
    "\n",
    "    Output:\n",
    "        This is not needed yet, actually\n",
    "    \"\"\"\n",
    "    pass\n",
    "\n",
    "def scaleStandard(df_col):\n",
    "    \"\"\"\n",
    "    Fits and returns a standard scaled version of a dataframe column\n",
    "    \"\"\"\n",
    "    scaler = StandardScaler()\n",
    "    df_col = scaler.fit_transform(df_col)\n",
    "    df_col = pd.DataFrame(df_col)\n",
    "    return df_col, scaler\n",
    "\n",
    "def logReturn(data, df_col):\n",
    "    \"\"\"\n",
    "    Perform log return for a dataframe column\n",
    "    \"\"\"\n",
    "    new_col = np.log1p(data[df_col].pct_change())\n",
    "    return new_col\n",
    "\n",
    "def transformData(data, log_return=[], standard_scale=[]):\n",
    "    \"\"\"\n",
    "    Perform essential transformations towards the dataframe.\n",
    "    \n",
    "    Args:\n",
    "        data (pd.DataFrame): DataFrame with data to be transformed.\n",
    "        log_return (list): List of columns for which log returns should be computed.\n",
    "        standard_scale (list): List of columns to be standard scaled.\n",
    "        \n",
    "    Returns:\n",
    "        data (pd.DataFrame): Transformed DataFrame.\n",
    "        Optional: yScaler if 'y' is in standard_scale.\n",
    "    \"\"\"\n",
    "    y_log_ret = False\n",
    "    y_std_scale = False\n",
    "\n",
    "    data.sort_values(by='ds', inplace=True)\n",
    "\n",
    "    # Apply log return transformation\n",
    "    if len(log_return) > 0:\n",
    "        for col1 in log_return:\n",
    "            try:\n",
    "                data[col1] = logReturn(data, col1)\n",
    "            except Exception as e:\n",
    "                print(f\"Error processing log return for column '{col1}': {e}\")\n",
    "                pass\n",
    "            \n",
    "        if 'y' in log_return:\n",
    "            y_log_ret = True\n",
    "\n",
    "    # Apply standard scaling\n",
    "    yScaler = None  # Initialize to None\n",
    "    if len(standard_scale) > 0:\n",
    "        for col2 in standard_scale:\n",
    "            try:\n",
    "                data[col2], _ = scaleStandard(data[[col2]])  # Assuming scaleStandard handles 1D arrays\n",
    "            except Exception as e:\n",
    "                print(f\"Error processing standard scaling for column '{col2}': {e}\")\n",
    "                pass\n",
    "            \n",
    "        if 'y' in standard_scale:\n",
    "            try:\n",
    "                data['y'], yScaler = scaleStandard(data['y'])  # Scale 'y' and get scaler\n",
    "                y_std_scale = True\n",
    "            except Exception as e:\n",
    "                print(f\"Error processing standard scaling for 'y': {e}\")\n",
    "                pass\n",
    "\n",
    "    # If 'yScaler' exists, return it along with the transformed data\n",
    "    if yScaler:\n",
    "        return data, yScaler\n",
    "\n",
    "    return data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 210,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total length is 2501, with validation and test size of 250 for each\n"
     ]
    }
   ],
   "source": [
    "# Exogenous \n",
    "\n",
    "Y_df = df.rename({'date' : 'ds', 'BrFu' : 'y'}, axis=1\n",
    "                          )\n",
    "Y_df['unique_id'] = 'Dated'\n",
    "Y_df['ds'] = pd.to_datetime(Y_df['ds'])\n",
    "\n",
    "# We make validation and test splits\n",
    "n_time = len(Y_df.ds.unique())\n",
    "val_size = int(.1 * n_time)\n",
    "test_size = int(.1 * n_time)\n",
    "\n",
    "print(f'Total length is {n_time}, with validation and test size of {val_size} for each')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 211,
   "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>ds</th>\n",
       "      <th>y</th>\n",
       "      <th>Gas</th>\n",
       "      <th>DXY</th>\n",
       "      <th>Gold(CMX)</th>\n",
       "      <th>Silver(CMX)</th>\n",
       "      <th>Platinum(NMX)</th>\n",
       "      <th>Palladium(NMX)</th>\n",
       "      <th>USBond</th>\n",
       "      <th>S&amp;P500</th>\n",
       "      <th>GoldMiners</th>\n",
       "      <th>GPRD</th>\n",
       "      <th>BrFu</th>\n",
       "      <th>unique_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2024-08-05</td>\n",
       "      <td>72.94</td>\n",
       "      <td>2.3336</td>\n",
       "      <td>102.690002</td>\n",
       "      <td>2444.4</td>\n",
       "      <td>27.207</td>\n",
       "      <td>915.5</td>\n",
       "      <td>826.10</td>\n",
       "      <td>125.4062</td>\n",
       "      <td>5186.33</td>\n",
       "      <td>35.34</td>\n",
       "      <td>210.206711</td>\n",
       "      <td>72.94</td>\n",
       "      <td>Dated</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2024-08-02</td>\n",
       "      <td>73.52</td>\n",
       "      <td>2.3176</td>\n",
       "      <td>103.209999</td>\n",
       "      <td>2469.8</td>\n",
       "      <td>28.392</td>\n",
       "      <td>967.6</td>\n",
       "      <td>882.50</td>\n",
       "      <td>125.0625</td>\n",
       "      <td>5346.56</td>\n",
       "      <td>36.48</td>\n",
       "      <td>207.799438</td>\n",
       "      <td>73.52</td>\n",
       "      <td>Dated</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2024-08-01</td>\n",
       "      <td>76.93</td>\n",
       "      <td>2.3980</td>\n",
       "      <td>104.419998</td>\n",
       "      <td>2480.8</td>\n",
       "      <td>28.477</td>\n",
       "      <td>970.5</td>\n",
       "      <td>895.10</td>\n",
       "      <td>122.5312</td>\n",
       "      <td>5446.68</td>\n",
       "      <td>37.27</td>\n",
       "      <td>139.878098</td>\n",
       "      <td>76.93</td>\n",
       "      <td>Dated</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2024-07-31</td>\n",
       "      <td>77.91</td>\n",
       "      <td>2.4425</td>\n",
       "      <td>104.099998</td>\n",
       "      <td>2473.0</td>\n",
       "      <td>28.938</td>\n",
       "      <td>986.4</td>\n",
       "      <td>925.20</td>\n",
       "      <td>120.7812</td>\n",
       "      <td>5522.30</td>\n",
       "      <td>37.93</td>\n",
       "      <td>135.206848</td>\n",
       "      <td>77.91</td>\n",
       "      <td>Dated</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2024-07-30</td>\n",
       "      <td>74.73</td>\n",
       "      <td>2.3443</td>\n",
       "      <td>104.550003</td>\n",
       "      <td>2451.9</td>\n",
       "      <td>28.525</td>\n",
       "      <td>971.0</td>\n",
       "      <td>881.70</td>\n",
       "      <td>120.1875</td>\n",
       "      <td>5436.44</td>\n",
       "      <td>36.95</td>\n",
       "      <td>95.696396</td>\n",
       "      <td>74.73</td>\n",
       "      <td>Dated</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2496</th>\n",
       "      <td>2014-08-29</td>\n",
       "      <td>95.96</td>\n",
       "      <td>2.6229</td>\n",
       "      <td>82.750000</td>\n",
       "      <td>1287.4</td>\n",
       "      <td>19.492</td>\n",
       "      <td>1424.7</td>\n",
       "      <td>909.55</td>\n",
       "      <td>140.0938</td>\n",
       "      <td>2003.37</td>\n",
       "      <td>26.69</td>\n",
       "      <td>157.336334</td>\n",
       "      <td>95.96</td>\n",
       "      <td>Dated</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2497</th>\n",
       "      <td>2014-08-28</td>\n",
       "      <td>94.55</td>\n",
       "      <td>2.5908</td>\n",
       "      <td>82.480003</td>\n",
       "      <td>1290.4</td>\n",
       "      <td>19.609</td>\n",
       "      <td>1425.2</td>\n",
       "      <td>898.10</td>\n",
       "      <td>141.7188</td>\n",
       "      <td>1996.74</td>\n",
       "      <td>26.46</td>\n",
       "      <td>97.534706</td>\n",
       "      <td>94.55</td>\n",
       "      <td>Dated</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2498</th>\n",
       "      <td>2014-08-27</td>\n",
       "      <td>93.88</td>\n",
       "      <td>2.5905</td>\n",
       "      <td>82.430000</td>\n",
       "      <td>1283.4</td>\n",
       "      <td>19.475</td>\n",
       "      <td>1419.9</td>\n",
       "      <td>894.70</td>\n",
       "      <td>141.1562</td>\n",
       "      <td>2000.12</td>\n",
       "      <td>26.11</td>\n",
       "      <td>143.082108</td>\n",
       "      <td>93.88</td>\n",
       "      <td>Dated</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2499</th>\n",
       "      <td>2014-08-26</td>\n",
       "      <td>93.86</td>\n",
       "      <td>2.6001</td>\n",
       "      <td>82.650002</td>\n",
       "      <td>1285.2</td>\n",
       "      <td>19.459</td>\n",
       "      <td>1419.6</td>\n",
       "      <td>890.15</td>\n",
       "      <td>140.5312</td>\n",
       "      <td>2000.02</td>\n",
       "      <td>26.19</td>\n",
       "      <td>118.143707</td>\n",
       "      <td>93.86</td>\n",
       "      <td>Dated</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2500</th>\n",
       "      <td>2014-08-25</td>\n",
       "      <td>93.35</td>\n",
       "      <td>2.5954</td>\n",
       "      <td>82.550003</td>\n",
       "      <td>1278.9</td>\n",
       "      <td>19.431</td>\n",
       "      <td>1418.4</td>\n",
       "      <td>891.50</td>\n",
       "      <td>140.7188</td>\n",
       "      <td>1997.92</td>\n",
       "      <td>25.62</td>\n",
       "      <td>167.600159</td>\n",
       "      <td>93.35</td>\n",
       "      <td>Dated</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2501 rows × 14 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             ds      y     Gas         DXY  Gold(CMX)  Silver(CMX)  \\\n",
       "0    2024-08-05  72.94  2.3336  102.690002     2444.4       27.207   \n",
       "1    2024-08-02  73.52  2.3176  103.209999     2469.8       28.392   \n",
       "2    2024-08-01  76.93  2.3980  104.419998     2480.8       28.477   \n",
       "3    2024-07-31  77.91  2.4425  104.099998     2473.0       28.938   \n",
       "4    2024-07-30  74.73  2.3443  104.550003     2451.9       28.525   \n",
       "...         ...    ...     ...         ...        ...          ...   \n",
       "2496 2014-08-29  95.96  2.6229   82.750000     1287.4       19.492   \n",
       "2497 2014-08-28  94.55  2.5908   82.480003     1290.4       19.609   \n",
       "2498 2014-08-27  93.88  2.5905   82.430000     1283.4       19.475   \n",
       "2499 2014-08-26  93.86  2.6001   82.650002     1285.2       19.459   \n",
       "2500 2014-08-25  93.35  2.5954   82.550003     1278.9       19.431   \n",
       "\n",
       "      Platinum(NMX)  Palladium(NMX)    USBond   S&P500  GoldMiners  \\\n",
       "0             915.5          826.10  125.4062  5186.33       35.34   \n",
       "1             967.6          882.50  125.0625  5346.56       36.48   \n",
       "2             970.5          895.10  122.5312  5446.68       37.27   \n",
       "3             986.4          925.20  120.7812  5522.30       37.93   \n",
       "4             971.0          881.70  120.1875  5436.44       36.95   \n",
       "...             ...             ...       ...      ...         ...   \n",
       "2496         1424.7          909.55  140.0938  2003.37       26.69   \n",
       "2497         1425.2          898.10  141.7188  1996.74       26.46   \n",
       "2498         1419.9          894.70  141.1562  2000.12       26.11   \n",
       "2499         1419.6          890.15  140.5312  2000.02       26.19   \n",
       "2500         1418.4          891.50  140.7188  1997.92       25.62   \n",
       "\n",
       "            GPRD   BrFu unique_id  \n",
       "0     210.206711  72.94     Dated  \n",
       "1     207.799438  73.52     Dated  \n",
       "2     139.878098  76.93     Dated  \n",
       "3     135.206848  77.91     Dated  \n",
       "4      95.696396  74.73     Dated  \n",
       "...          ...    ...       ...  \n",
       "2496  157.336334  95.96     Dated  \n",
       "2497   97.534706  94.55     Dated  \n",
       "2498  143.082108  93.88     Dated  \n",
       "2499  118.143707  93.86     Dated  \n",
       "2500  167.600159  93.35     Dated  \n",
       "\n",
       "[2501 rows x 14 columns]"
      ]
     },
     "execution_count": 211,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y_df_test = Y_df.copy().rename({'y' : 'BrFu'}, axis=1)\n",
    "\n",
    "# Y_df_test = createLag(Y_df_test, amt=30)\n",
    "last_df = Y_df[['ds', 'y']].merge(Y_df_test, on = 'ds')\n",
    "last_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 212,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Error processing standard scaling for 'y': Expected 2D array, got 1D array instead:\n",
      "array=[0.57819032 0.6095309  0.79379192 ... 1.70969345 1.70861274 1.68105464].\n",
      "Reshape your data either using array.reshape(-1, 1) if your data has a single feature or array.reshape(1, -1) if it contains a single sample.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ds</th>\n",
       "      <th>y</th>\n",
       "      <th>Gas</th>\n",
       "      <th>DXY</th>\n",
       "      <th>Gold(CMX)</th>\n",
       "      <th>Silver(CMX)</th>\n",
       "      <th>Platinum(NMX)</th>\n",
       "      <th>Palladium(NMX)</th>\n",
       "      <th>USBond</th>\n",
       "      <th>S&amp;P500</th>\n",
       "      <th>GoldMiners</th>\n",
       "      <th>GPRD</th>\n",
       "      <th>BrFu</th>\n",
       "      <th>unique_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2500</th>\n",
       "      <td>2014-08-25</td>\n",
       "      <td>0.578190</td>\n",
       "      <td>0.696193</td>\n",
       "      <td>1.014414</td>\n",
       "      <td>2.601005</td>\n",
       "      <td>1.767282</td>\n",
       "      <td>-0.488608</td>\n",
       "      <td>-0.829241</td>\n",
       "      <td>-1.441077</td>\n",
       "      <td>1.996384</td>\n",
       "      <td>1.317266</td>\n",
       "      <td>1.747204</td>\n",
       "      <td>93.35</td>\n",
       "      <td>Dated</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2499</th>\n",
       "      <td>2014-08-26</td>\n",
       "      <td>0.609531</td>\n",
       "      <td>0.668536</td>\n",
       "      <td>1.114362</td>\n",
       "      <td>2.675076</td>\n",
       "      <td>2.043063</td>\n",
       "      <td>-0.048414</td>\n",
       "      <td>-0.742670</td>\n",
       "      <td>-1.461573</td>\n",
       "      <td>2.158139</td>\n",
       "      <td>1.488741</td>\n",
       "      <td>1.702180</td>\n",
       "      <td>93.86</td>\n",
       "      <td>Dated</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2498</th>\n",
       "      <td>2014-08-27</td>\n",
       "      <td>0.793792</td>\n",
       "      <td>0.807512</td>\n",
       "      <td>1.346933</td>\n",
       "      <td>2.707154</td>\n",
       "      <td>2.062844</td>\n",
       "      <td>-0.023911</td>\n",
       "      <td>-0.723329</td>\n",
       "      <td>-1.612523</td>\n",
       "      <td>2.259212</td>\n",
       "      <td>1.607570</td>\n",
       "      <td>0.431829</td>\n",
       "      <td>93.88</td>\n",
       "      <td>Dated</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2497</th>\n",
       "      <td>2014-08-28</td>\n",
       "      <td>0.846747</td>\n",
       "      <td>0.884432</td>\n",
       "      <td>1.285427</td>\n",
       "      <td>2.684408</td>\n",
       "      <td>2.170131</td>\n",
       "      <td>0.110428</td>\n",
       "      <td>-0.677127</td>\n",
       "      <td>-1.716881</td>\n",
       "      <td>2.335551</td>\n",
       "      <td>1.706845</td>\n",
       "      <td>0.344462</td>\n",
       "      <td>94.55</td>\n",
       "      <td>Dated</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2496</th>\n",
       "      <td>2014-08-29</td>\n",
       "      <td>0.674914</td>\n",
       "      <td>0.714688</td>\n",
       "      <td>1.371921</td>\n",
       "      <td>2.622877</td>\n",
       "      <td>2.074015</td>\n",
       "      <td>-0.019687</td>\n",
       "      <td>-0.743898</td>\n",
       "      <td>-1.752285</td>\n",
       "      <td>2.248874</td>\n",
       "      <td>1.559436</td>\n",
       "      <td>-0.394513</td>\n",
       "      <td>95.96</td>\n",
       "      <td>Dated</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2024-07-30</td>\n",
       "      <td>1.822087</td>\n",
       "      <td>1.196264</td>\n",
       "      <td>-2.818213</td>\n",
       "      <td>-0.773021</td>\n",
       "      <td>-0.028204</td>\n",
       "      <td>3.813639</td>\n",
       "      <td>-0.701149</td>\n",
       "      <td>-0.565209</td>\n",
       "      <td>-1.216868</td>\n",
       "      <td>0.016163</td>\n",
       "      <td>0.758355</td>\n",
       "      <td>74.73</td>\n",
       "      <td>Dated</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2024-07-31</td>\n",
       "      <td>1.745897</td>\n",
       "      <td>1.140777</td>\n",
       "      <td>-2.870109</td>\n",
       "      <td>-0.764272</td>\n",
       "      <td>-0.000975</td>\n",
       "      <td>3.817864</td>\n",
       "      <td>-0.718724</td>\n",
       "      <td>-0.468305</td>\n",
       "      <td>-1.223561</td>\n",
       "      <td>-0.018432</td>\n",
       "      <td>-0.360130</td>\n",
       "      <td>77.91</td>\n",
       "      <td>Dated</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2024-08-01</td>\n",
       "      <td>1.709693</td>\n",
       "      <td>1.140259</td>\n",
       "      <td>-2.879720</td>\n",
       "      <td>-0.784686</td>\n",
       "      <td>-0.032161</td>\n",
       "      <td>3.773084</td>\n",
       "      <td>-0.723943</td>\n",
       "      <td>-0.501855</td>\n",
       "      <td>-1.220149</td>\n",
       "      <td>-0.071078</td>\n",
       "      <td>0.491755</td>\n",
       "      <td>76.93</td>\n",
       "      <td>Dated</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2024-08-02</td>\n",
       "      <td>1.708613</td>\n",
       "      <td>1.156853</td>\n",
       "      <td>-2.837434</td>\n",
       "      <td>-0.779436</td>\n",
       "      <td>-0.035884</td>\n",
       "      <td>3.770549</td>\n",
       "      <td>-0.730927</td>\n",
       "      <td>-0.539126</td>\n",
       "      <td>-1.220250</td>\n",
       "      <td>-0.059045</td>\n",
       "      <td>0.025325</td>\n",
       "      <td>73.52</td>\n",
       "      <td>Dated</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2024-08-05</td>\n",
       "      <td>1.681055</td>\n",
       "      <td>1.148729</td>\n",
       "      <td>-2.856654</td>\n",
       "      <td>-0.797808</td>\n",
       "      <td>-0.042401</td>\n",
       "      <td>3.760410</td>\n",
       "      <td>-0.728855</td>\n",
       "      <td>-0.527938</td>\n",
       "      <td>-1.222370</td>\n",
       "      <td>-0.144782</td>\n",
       "      <td>0.950322</td>\n",
       "      <td>72.94</td>\n",
       "      <td>Dated</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2501 rows × 14 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             ds         y       Gas       DXY  Gold(CMX)  Silver(CMX)  \\\n",
       "2500 2014-08-25  0.578190  0.696193  1.014414   2.601005     1.767282   \n",
       "2499 2014-08-26  0.609531  0.668536  1.114362   2.675076     2.043063   \n",
       "2498 2014-08-27  0.793792  0.807512  1.346933   2.707154     2.062844   \n",
       "2497 2014-08-28  0.846747  0.884432  1.285427   2.684408     2.170131   \n",
       "2496 2014-08-29  0.674914  0.714688  1.371921   2.622877     2.074015   \n",
       "...         ...       ...       ...       ...        ...          ...   \n",
       "4    2024-07-30  1.822087  1.196264 -2.818213  -0.773021    -0.028204   \n",
       "3    2024-07-31  1.745897  1.140777 -2.870109  -0.764272    -0.000975   \n",
       "2    2024-08-01  1.709693  1.140259 -2.879720  -0.784686    -0.032161   \n",
       "1    2024-08-02  1.708613  1.156853 -2.837434  -0.779436    -0.035884   \n",
       "0    2024-08-05  1.681055  1.148729 -2.856654  -0.797808    -0.042401   \n",
       "\n",
       "      Platinum(NMX)  Palladium(NMX)    USBond    S&P500  GoldMiners      GPRD  \\\n",
       "2500      -0.488608       -0.829241 -1.441077  1.996384    1.317266  1.747204   \n",
       "2499      -0.048414       -0.742670 -1.461573  2.158139    1.488741  1.702180   \n",
       "2498      -0.023911       -0.723329 -1.612523  2.259212    1.607570  0.431829   \n",
       "2497       0.110428       -0.677127 -1.716881  2.335551    1.706845  0.344462   \n",
       "2496      -0.019687       -0.743898 -1.752285  2.248874    1.559436 -0.394513   \n",
       "...             ...             ...       ...       ...         ...       ...   \n",
       "4          3.813639       -0.701149 -0.565209 -1.216868    0.016163  0.758355   \n",
       "3          3.817864       -0.718724 -0.468305 -1.223561   -0.018432 -0.360130   \n",
       "2          3.773084       -0.723943 -0.501855 -1.220149   -0.071078  0.491755   \n",
       "1          3.770549       -0.730927 -0.539126 -1.220250   -0.059045  0.025325   \n",
       "0          3.760410       -0.728855 -0.527938 -1.222370   -0.144782  0.950322   \n",
       "\n",
       "       BrFu unique_id  \n",
       "2500  93.35     Dated  \n",
       "2499  93.86     Dated  \n",
       "2498  93.88     Dated  \n",
       "2497  94.55     Dated  \n",
       "2496  95.96     Dated  \n",
       "...     ...       ...  \n",
       "4     74.73     Dated  \n",
       "3     77.91     Dated  \n",
       "2     76.93     Dated  \n",
       "1     73.52     Dated  \n",
       "0     72.94     Dated  \n",
       "\n",
       "[2501 rows x 14 columns]"
      ]
     },
     "execution_count": 212,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_data = transformData(last_df,\n",
    "                         #log_return = ['y', 'BrFu', 'Gas', 'DXY', 'Gold(CMX)', 'Silver(CMX)', 'Platinum(NMX)', 'Palladium(NMX)', 'USBond', 'S&P500', 'GoldMiners', 'GPRD'],\n",
    "                         standard_scale = ['y', 'Gas', 'DXY', 'Gold(CMX)', 'Silver(CMX)', 'Platinum(NMX)', 'Palladium(NMX)', 'USBond', 'S&P500', 'GoldMiners', 'GPRD']\n",
    "                         )\n",
    "test_data.dropna(inplace=True)\n",
    "test_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "import os\n",
    "\n",
    "def plot_correlation_heatmap(df):\n",
    "    \"\"\"\n",
    "    Plot a correlation heatmap for all columns in a pandas DataFrame,\n",
    "    showing only the lower triangle of the matrix.\n",
    "    \n",
    "    Input:\n",
    "        df -> pandas DataFrame\n",
    "        \n",
    "    Output:\n",
    "        Correlation heatmap (lower triangle)\n",
    "    \"\"\"\n",
    "    corr_matrix = df.corr()\n",
    "\n",
    "    mask = np.triu(np.ones_like(corr_matrix, dtype=bool))\n",
    "\n",
    "    plt.figure(figsize=(10, 8))\n",
    "\n",
    "    # Lower triangle correlation heatmap\n",
    "    sns.heatmap(corr_matrix, mask=mask, annot=True, cmap='coolwarm', vmin=-1, vmax=1, \n",
    "                square=True, linewidths=.5, cbar_kws={\"shrink\": .8})\n",
    "\n",
    "    plt.title(\"Lower Triangle Correlation Heatmap\", fontsize=16)\n",
    "\n",
    "    plt.show()\n",
    "\n",
    "test = pd.read_csv(os.path.join('artifacts', '4b950ff8-8caf-4f33-9a68-345ea4464fc8', 'transformed_dataset.csv'))\n",
    "# plot_correlation_heatmap(test_data.drop(columns=['ds', 'unique_id']))\n",
    "plot_correlation_heatmap(test.drop(columns=['Date', 'Unnamed: 0']))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Statsforecast"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 341,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "e:\\VM\\miniconda3\\envs\\multi-its\\lib\\site-packages\\statsforecast\\core.py:510: FutureWarning: The `df` argument of the StatsForecast constructor as well as reusing stored dfs from other methods is deprecated and will raise an error in a future version. Please provide the `df` argument to the corresponding method instead, e.g. fit/forecast.\n",
      "  warnings.warn(\n",
      "e:\\VM\\miniconda3\\envs\\multi-its\\lib\\site-packages\\statsforecast\\core.py:510: FutureWarning: The `df` argument of the StatsForecast constructor as well as reusing stored dfs from other methods is deprecated and will raise an error in a future version. Please provide the `df` argument to the corresponding method instead, e.g. fit/forecast.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dated  -  ARIMA(5,0,2)(0,0,1)[30] with zero mean    \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "e:\\VM\\miniconda3\\envs\\multi-its\\lib\\site-packages\\statsforecast\\core.py:528: FutureWarning: In a future version the predictions will have the id as a column. You can set the `NIXTLA_ID_AS_COL` environment variable to adopt the new behavior and to suppress this warning.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ds</th>\n",
       "      <th>AutoARIMA</th>\n",
       "      <th>AutoARIMA-lo-95</th>\n",
       "      <th>AutoARIMA-hi-95</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-08-06</td>\n",
       "      <td>0.000161</td>\n",
       "      <td>-0.057762</td>\n",
       "      <td>0.058084</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-08-07</td>\n",
       "      <td>0.005220</td>\n",
       "      <td>-0.052705</td>\n",
       "      <td>0.063145</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-08-08</td>\n",
       "      <td>-0.002106</td>\n",
       "      <td>-0.060039</td>\n",
       "      <td>0.055827</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-08-09</td>\n",
       "      <td>-0.002936</td>\n",
       "      <td>-0.061125</td>\n",
       "      <td>0.055254</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-08-10</td>\n",
       "      <td>-0.002981</td>\n",
       "      <td>-0.061185</td>\n",
       "      <td>0.055223</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-08-11</td>\n",
       "      <td>-0.002235</td>\n",
       "      <td>-0.060556</td>\n",
       "      <td>0.056086</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-08-12</td>\n",
       "      <td>-0.001244</td>\n",
       "      <td>-0.059578</td>\n",
       "      <td>0.057090</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-08-13</td>\n",
       "      <td>-0.001598</td>\n",
       "      <td>-0.059981</td>\n",
       "      <td>0.056785</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-08-14</td>\n",
       "      <td>-0.001097</td>\n",
       "      <td>-0.059480</td>\n",
       "      <td>0.057286</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-08-15</td>\n",
       "      <td>-0.000821</td>\n",
       "      <td>-0.059239</td>\n",
       "      <td>0.057597</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-08-16</td>\n",
       "      <td>-0.000784</td>\n",
       "      <td>-0.059202</td>\n",
       "      <td>0.057635</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-08-17</td>\n",
       "      <td>-0.001267</td>\n",
       "      <td>-0.059713</td>\n",
       "      <td>0.057179</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-08-18</td>\n",
       "      <td>-0.001201</td>\n",
       "      <td>-0.059648</td>\n",
       "      <td>0.057245</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-08-19</td>\n",
       "      <td>-0.000658</td>\n",
       "      <td>-0.059121</td>\n",
       "      <td>0.057806</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-08-20</td>\n",
       "      <td>-0.000674</td>\n",
       "      <td>-0.059138</td>\n",
       "      <td>0.057790</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-08-21</td>\n",
       "      <td>-0.000226</td>\n",
       "      <td>-0.058701</td>\n",
       "      <td>0.058248</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-08-22</td>\n",
       "      <td>-0.001502</td>\n",
       "      <td>-0.059978</td>\n",
       "      <td>0.056973</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-08-23</td>\n",
       "      <td>-0.000578</td>\n",
       "      <td>-0.059061</td>\n",
       "      <td>0.057904</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-08-24</td>\n",
       "      <td>0.000512</td>\n",
       "      <td>-0.057971</td>\n",
       "      <td>0.058994</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-08-25</td>\n",
       "      <td>0.000194</td>\n",
       "      <td>-0.058294</td>\n",
       "      <td>0.058681</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-08-26</td>\n",
       "      <td>0.000129</td>\n",
       "      <td>-0.058359</td>\n",
       "      <td>0.058617</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-08-27</td>\n",
       "      <td>-0.000743</td>\n",
       "      <td>-0.059234</td>\n",
       "      <td>0.057748</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-08-28</td>\n",
       "      <td>-0.000714</td>\n",
       "      <td>-0.059205</td>\n",
       "      <td>0.057778</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-08-29</td>\n",
       "      <td>0.000102</td>\n",
       "      <td>-0.058391</td>\n",
       "      <td>0.058595</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-08-30</td>\n",
       "      <td>0.000239</td>\n",
       "      <td>-0.058255</td>\n",
       "      <td>0.058732</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-08-31</td>\n",
       "      <td>0.000167</td>\n",
       "      <td>-0.058328</td>\n",
       "      <td>0.058662</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-09-01</td>\n",
       "      <td>-0.001718</td>\n",
       "      <td>-0.060214</td>\n",
       "      <td>0.056777</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-09-02</td>\n",
       "      <td>0.000138</td>\n",
       "      <td>-0.058358</td>\n",
       "      <td>0.058634</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-09-03</td>\n",
       "      <td>0.001258</td>\n",
       "      <td>-0.057239</td>\n",
       "      <td>0.059754</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>2024-09-04</td>\n",
       "      <td>0.000057</td>\n",
       "      <td>-0.058440</td>\n",
       "      <td>0.058554</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  ds  AutoARIMA  AutoARIMA-lo-95  AutoARIMA-hi-95\n",
       "unique_id                                                        \n",
       "Dated     2024-08-06   0.000161        -0.057762         0.058084\n",
       "Dated     2024-08-07   0.005220        -0.052705         0.063145\n",
       "Dated     2024-08-08  -0.002106        -0.060039         0.055827\n",
       "Dated     2024-08-09  -0.002936        -0.061125         0.055254\n",
       "Dated     2024-08-10  -0.002981        -0.061185         0.055223\n",
       "Dated     2024-08-11  -0.002235        -0.060556         0.056086\n",
       "Dated     2024-08-12  -0.001244        -0.059578         0.057090\n",
       "Dated     2024-08-13  -0.001598        -0.059981         0.056785\n",
       "Dated     2024-08-14  -0.001097        -0.059480         0.057286\n",
       "Dated     2024-08-15  -0.000821        -0.059239         0.057597\n",
       "Dated     2024-08-16  -0.000784        -0.059202         0.057635\n",
       "Dated     2024-08-17  -0.001267        -0.059713         0.057179\n",
       "Dated     2024-08-18  -0.001201        -0.059648         0.057245\n",
       "Dated     2024-08-19  -0.000658        -0.059121         0.057806\n",
       "Dated     2024-08-20  -0.000674        -0.059138         0.057790\n",
       "Dated     2024-08-21  -0.000226        -0.058701         0.058248\n",
       "Dated     2024-08-22  -0.001502        -0.059978         0.056973\n",
       "Dated     2024-08-23  -0.000578        -0.059061         0.057904\n",
       "Dated     2024-08-24   0.000512        -0.057971         0.058994\n",
       "Dated     2024-08-25   0.000194        -0.058294         0.058681\n",
       "Dated     2024-08-26   0.000129        -0.058359         0.058617\n",
       "Dated     2024-08-27  -0.000743        -0.059234         0.057748\n",
       "Dated     2024-08-28  -0.000714        -0.059205         0.057778\n",
       "Dated     2024-08-29   0.000102        -0.058391         0.058595\n",
       "Dated     2024-08-30   0.000239        -0.058255         0.058732\n",
       "Dated     2024-08-31   0.000167        -0.058328         0.058662\n",
       "Dated     2024-09-01  -0.001718        -0.060214         0.056777\n",
       "Dated     2024-09-02   0.000138        -0.058358         0.058634\n",
       "Dated     2024-09-03   0.001258        -0.057239         0.059754\n",
       "Dated     2024-09-04   0.000057        -0.058440         0.058554"
      ]
     },
     "execution_count": 341,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "from statsforecast.models import AutoARIMA\n",
    "from statsforecast import StatsForecast\n",
    "from statsforecast.arima import arima_string\n",
    "\n",
    "# Create a list of models and instantiation parameters\n",
    "models = [\n",
    "    AutoARIMA(season_length=30, max_p=5, max_d=2, max_q=5, seasonal=True),\n",
    "]\n",
    "\n",
    "# Instantiate StatsForecast class as sf\n",
    "sf = StatsForecast(\n",
    "    df=test_data[['ds','y','unique_id']], \n",
    "    models=models,\n",
    "    freq='D', \n",
    "    n_jobs=2\n",
    ")\n",
    "\n",
    "forecasts_df = sf.fit().predict(h=30, level=[95])\n",
    "\n",
    "for model, uid in zip(sf.fitted_, forecasts_df.index.unique()):\n",
    "    print(uid, ' - ', arima_string(model[0].model_))\n",
    "\n",
    "forecasts_df\n",
    "\n",
    "# Check parameters\n",
    "# https://stackoverflow.com/questions/77436740/statforecast-autoarima-how-to-run-different-models-for-each-unique-ids"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 342,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "\n",
    "def plot_forecast(test_data, forecasts_df, horizon=30):\n",
    "    \"\"\"\n",
    "    Plots historical data and forecasted values, emphasizing the forecast.\n",
    "    \n",
    "    Args:\n",
    "        test_data (pd.DataFrame): Historical data with columns 'ds' (date) and 'y' (values).\n",
    "        forecasts_df (pd.DataFrame): Forecasted values with index 'ds' and columns for predictions.\n",
    "        horizon (int): Number of forecasted steps.\n",
    "    \"\"\"\n",
    "    # Prepare the forecasted data\n",
    "    forecasted_dates = pd.date_range(test_data['ds'].max(), periods=horizon+1, freq='D')[1:]\n",
    "    forecasted_values = forecasts_df['AutoARIMA'].values[:horizon]\n",
    "\n",
    "    # Combine forecasted values with their corresponding dates\n",
    "    forecast_df = pd.DataFrame({'ds': forecasted_dates, 'forecast': forecasted_values})\n",
    "\n",
    "    # Plot the historical and forecasted data\n",
    "    plt.figure(figsize=(10, 6))\n",
    "    \n",
    "    # Plot historical data with lighter color and transparency\n",
    "    plt.plot(test_data['ds'], test_data['y'], label='Historical', color='blue', alpha=0.6, linewidth=2)\n",
    "    \n",
    "    # Plot forecasted data with a thicker, dashed line to stand out\n",
    "    plt.plot(forecast_df['ds'], forecast_df['forecast'], label='Forecast', color='orange' , linewidth=3)\n",
    "    \n",
    "    # Add labels and title\n",
    "    plt.xlabel('Date')\n",
    "    plt.ylabel('Values')\n",
    "    plt.title('Crude Oil Price Forecast vs Historical Data')\n",
    "    plt.legend()\n",
    "    \n",
    "    # Show the plot\n",
    "    plt.show()\n",
    "\n",
    "# Example usage:\n",
    "plot_forecast(last_df, forecasts_df, horizon=30)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Neuralforecast"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Seed set to 12345678\n",
      "Seed set to 12345678\n",
      "Seed set to 12345678\n"
     ]
    }
   ],
   "source": [
    "horizon = 15\n",
    "input_size = 30\n",
    "models = [\n",
    "    TSMixer(h=horizon,\n",
    "                input_size=input_size,\n",
    "                n_series=1,\n",
    "                max_steps=1000,\n",
    "                val_check_steps=50,\n",
    "                early_stop_patience_steps=5,\n",
    "                scaler_type='identity',\n",
    "                loss=MAE(),\n",
    "                valid_loss=MAE(),\n",
    "                random_seed=12345678,\n",
    "                ),\n",
    "    TSMixerx(h=horizon,\n",
    "                input_size=input_size,\n",
    "                n_series=1,\n",
    "                max_steps=1000,\n",
    "                val_check_steps=50,\n",
    "                early_stop_patience_steps=5,\n",
    "                scaler_type='identity',\n",
    "                dropout=0.7,\n",
    "                loss=MAE(),\n",
    "                valid_loss=MAE(),\n",
    "                random_seed=12345678,\n",
    "                hist_exog_list=['Gas', 'DXY', 'Gold(CMX)', 'Silver(CMX)', 'Platinum(NMX)', 'Palladium(NMX)', 'USBond', 'S&P500', 'GoldMiners', 'GPRD'],\n",
    "                ),\n",
    "    NBEATSx(h=horizon,\n",
    "                input_size=horizon,\n",
    "                max_steps=1000,\n",
    "                val_check_steps=50,\n",
    "                early_stop_patience_steps=5,\n",
    "                scaler_type='identity',\n",
    "                loss=MAE(),\n",
    "                valid_loss=MAE(),\n",
    "                random_seed=12345678,\n",
    "                hist_exog_list=['Gas', 'DXY', 'Gold(CMX)', 'Silver(CMX)', 'Platinum(NMX)', 'Palladium(NMX)', 'USBond', 'S&P500', 'GoldMiners', 'GPRD']\n",
    "                ),\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "nf = NeuralForecast(\n",
    "    models=models,\n",
    "    freq='D')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "GPU available: True (cuda), used: True\n",
      "TPU available: False, using: 0 TPU cores\n",
      "HPU available: False, using: 0 HPUs\n",
      "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n",
      "\n",
      "  | Name          | Type                     | Params | Mode \n",
      "-------------------------------------------------------------------\n",
      "0 | loss          | MAE                      | 0      | train\n",
      "1 | valid_loss    | MAE                      | 0      | train\n",
      "2 | padder        | ConstantPad1d            | 0      | train\n",
      "3 | scaler        | TemporalNorm             | 0      | train\n",
      "4 | norm          | ReversibleInstanceNorm1d | 2      | train\n",
      "5 | mixing_layers | Sequential               | 2.5 K  | train\n",
      "6 | out           | Linear                   | 465    | train\n",
      "-------------------------------------------------------------------\n",
      "3.0 K     Trainable params\n",
      "0         Non-trainable params\n",
      "3.0 K     Total params\n",
      "0.012     Total estimated model params size (MB)\n",
      "29        Modules in train mode\n",
      "0         Modules in eval mode\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 549: 100%|██████████| 1/1 [00:00<00:00, 15.68it/s, v_num=12, train_loss_step=3.290, train_loss_epoch=3.290, valid_loss=4.060]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Trainer already configured with model summary callbacks: [<class 'pytorch_lightning.callbacks.model_summary.ModelSummary'>]. Skipping setting a default `ModelSummary` callback.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "GPU available: True (cuda), used: True\n",
      "TPU available: False, using: 0 TPU cores\n",
      "HPU available: False, using: 0 HPUs\n",
      "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predicting DataLoader 0: 100%|██████████| 1/1 [00:00<00:00, 66.03it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "GPU available: True (cuda), used: True\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "TPU available: False, using: 0 TPU cores\n",
      "HPU available: False, using: 0 HPUs\n",
      "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n",
      "\n",
      "  | Name                | Type                     | Params | Mode \n",
      "-------------------------------------------------------------------------\n",
      "0 | loss                | MAE                      | 0      | train\n",
      "1 | valid_loss          | MAE                      | 0      | train\n",
      "2 | padder              | ConstantPad1d            | 0      | train\n",
      "3 | scaler              | TemporalNorm             | 0      | train\n",
      "4 | norm                | ReversibleInstanceNorm1d | 2      | train\n",
      "5 | temporal_projection | Linear                   | 465    | train\n",
      "6 | feature_mixer_hist  | FeatureMixing            | 7.6 K  | train\n",
      "7 | first_mixing        | MixingLayer              | 12.4 K | train\n",
      "8 | mixing_block        | Sequential               | 24.8 K | train\n",
      "9 | out                 | Linear                   | 65     | train\n",
      "-------------------------------------------------------------------------\n",
      "45.3 K    Trainable params\n",
      "0         Non-trainable params\n",
      "45.3 K    Total params\n",
      "0.181     Total estimated model params size (MB)\n",
      "48        Modules in train mode\n",
      "0         Modules in eval mode\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 299: 100%|██████████| 1/1 [00:00<00:00, 12.99it/s, v_num=14, train_loss_step=3.880, train_loss_epoch=3.880, valid_loss=6.120]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Trainer already configured with model summary callbacks: [<class 'pytorch_lightning.callbacks.model_summary.ModelSummary'>]. Skipping setting a default `ModelSummary` callback.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "GPU available: True (cuda), used: True\n",
      "TPU available: False, using: 0 TPU cores\n",
      "HPU available: False, using: 0 HPUs\n",
      "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predicting DataLoader 0: 100%|██████████| 1/1 [00:00<00:00, 68.88it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "GPU available: True (cuda), used: True\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "TPU available: False, using: 0 TPU cores\n",
      "HPU available: False, using: 0 HPUs\n",
      "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n",
      "\n",
      "  | Name         | Type          | Params | Mode \n",
      "-------------------------------------------------------\n",
      "0 | loss         | MAE           | 0      | train\n",
      "1 | valid_loss   | MAE           | 0      | train\n",
      "2 | padder_train | ConstantPad1d | 0      | train\n",
      "3 | scaler       | TemporalNorm  | 0      | train\n",
      "4 | blocks       | ModuleList    | 2.7 M  | train\n",
      "-------------------------------------------------------\n",
      "2.7 M     Trainable params\n",
      "930       Non-trainable params\n",
      "2.7 M     Total params\n",
      "10.668    Total estimated model params size (MB)\n",
      "32        Modules in train mode\n",
      "0         Modules in eval mode\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 549: 100%|██████████| 1/1 [00:00<00:00, 14.88it/s, v_num=16, train_loss_step=2.540, train_loss_epoch=2.540, valid_loss=4.240]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Trainer already configured with model summary callbacks: [<class 'pytorch_lightning.callbacks.model_summary.ModelSummary'>]. Skipping setting a default `ModelSummary` callback.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "GPU available: True (cuda), used: True\n",
      "TPU available: False, using: 0 TPU cores\n",
      "HPU available: False, using: 0 HPUs\n",
      "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predicting DataLoader 0: 100%|██████████| 1/1 [00:00<00:00, 95.13it/s] \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "e:\\VM\\miniconda3\\envs\\multi-its\\lib\\site-packages\\neuralforecast\\core.py:209: FutureWarning: In a future version the predictions will have the id as a column. You can set the `NIXTLA_ID_AS_COL` environment variable to adopt the new behavior and to suppress this warning.\n",
      "  warnings.warn(\n"
     ]
    }
   ],
   "source": [
    "Y_hat_df = nf.cross_validation(df=last_df,\n",
    "                               val_size=val_size,\n",
    "                               test_size=test_size,\n",
    "                               n_windows=None\n",
    "                               )\n",
    "Y_hat_df = Y_hat_df.reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Y_hat_df['y'] = yScaler.inverse_transform(Y_hat_df[['y']])\n",
    "# Y_hat_df['TSMixer'] = yScaler.inverse_transform(Y_hat_df[['TSMixer']])\n",
    "# Y_hat_df['TSMixerx'] = yScaler.inverse_transform(Y_hat_df[['TSMixerx']])\n",
    "# Y_hat_df['NBEATSx'] = yScaler.inverse_transform(Y_hat_df[['NBEATSx']])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "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>unique_id</th>\n",
       "      <th>ds</th>\n",
       "      <th>cutoff</th>\n",
       "      <th>TSMixer</th>\n",
       "      <th>TSMixerx</th>\n",
       "      <th>NBEATSx</th>\n",
       "      <th>y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2023-08-08</td>\n",
       "      <td>2023-08-07</td>\n",
       "      <td>82.474472</td>\n",
       "      <td>85.802956</td>\n",
       "      <td>82.134666</td>\n",
       "      <td>82.92</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2023-08-09</td>\n",
       "      <td>2023-08-07</td>\n",
       "      <td>82.759224</td>\n",
       "      <td>86.319862</td>\n",
       "      <td>82.286011</td>\n",
       "      <td>84.40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2023-08-10</td>\n",
       "      <td>2023-08-07</td>\n",
       "      <td>83.013039</td>\n",
       "      <td>85.747772</td>\n",
       "      <td>82.536789</td>\n",
       "      <td>82.82</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2023-08-11</td>\n",
       "      <td>2023-08-07</td>\n",
       "      <td>83.141472</td>\n",
       "      <td>86.653404</td>\n",
       "      <td>82.569160</td>\n",
       "      <td>83.19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2023-08-14</td>\n",
       "      <td>2023-08-07</td>\n",
       "      <td>82.938721</td>\n",
       "      <td>86.651329</td>\n",
       "      <td>82.851303</td>\n",
       "      <td>82.51</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3535</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-07-30</td>\n",
       "      <td>2024-07-15</td>\n",
       "      <td>83.346069</td>\n",
       "      <td>84.194443</td>\n",
       "      <td>83.034058</td>\n",
       "      <td>74.73</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3536</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-07-31</td>\n",
       "      <td>2024-07-15</td>\n",
       "      <td>83.281624</td>\n",
       "      <td>85.323982</td>\n",
       "      <td>83.138260</td>\n",
       "      <td>77.91</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3537</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-01</td>\n",
       "      <td>2024-07-15</td>\n",
       "      <td>82.876152</td>\n",
       "      <td>84.959213</td>\n",
       "      <td>82.852654</td>\n",
       "      <td>76.93</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3538</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-02</td>\n",
       "      <td>2024-07-15</td>\n",
       "      <td>83.144974</td>\n",
       "      <td>84.943405</td>\n",
       "      <td>82.720123</td>\n",
       "      <td>73.52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3539</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-05</td>\n",
       "      <td>2024-07-15</td>\n",
       "      <td>83.181046</td>\n",
       "      <td>84.934311</td>\n",
       "      <td>82.854950</td>\n",
       "      <td>72.94</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3540 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     unique_id         ds     cutoff    TSMixer   TSMixerx    NBEATSx      y\n",
       "0        Dated 2023-08-08 2023-08-07  82.474472  85.802956  82.134666  82.92\n",
       "1        Dated 2023-08-09 2023-08-07  82.759224  86.319862  82.286011  84.40\n",
       "2        Dated 2023-08-10 2023-08-07  83.013039  85.747772  82.536789  82.82\n",
       "3        Dated 2023-08-11 2023-08-07  83.141472  86.653404  82.569160  83.19\n",
       "4        Dated 2023-08-14 2023-08-07  82.938721  86.651329  82.851303  82.51\n",
       "...        ...        ...        ...        ...        ...        ...    ...\n",
       "3535     Dated 2024-07-30 2024-07-15  83.346069  84.194443  83.034058  74.73\n",
       "3536     Dated 2024-07-31 2024-07-15  83.281624  85.323982  83.138260  77.91\n",
       "3537     Dated 2024-08-01 2024-07-15  82.876152  84.959213  82.852654  76.93\n",
       "3538     Dated 2024-08-02 2024-07-15  83.144974  84.943405  82.720123  73.52\n",
       "3539     Dated 2024-08-05 2024-07-15  83.181046  84.934311  82.854950  72.94\n",
       "\n",
       "[3540 rows x 7 columns]"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y_hat_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x18d29307fd0>"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "Y_plot = Y_hat_df[Y_hat_df['unique_id']=='Dated']\n",
    "cutoffs = Y_hat_df['cutoff'].unique()[::horizon]\n",
    "Y_plot = Y_plot[Y_hat_df['cutoff'].isin(cutoffs)]\n",
    "\n",
    "plt.figure(figsize=(20,5))\n",
    "plt.plot(Y_plot['ds'], Y_plot['y'], label='True')\n",
    "for model in models:\n",
    "    plt.plot(Y_plot['ds'], Y_plot[f'{model}'], label=f'{model}')\n",
    "plt.xlabel('Datestamp')\n",
    "plt.ylabel('OT')\n",
    "plt.grid()\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TSMixer horizon 2 - MAE: 1.445\n",
      "TSMixer horizon 2 - MSE: 3.415\n",
      "TSMixer horizon 2 - MAPE: 0.018\n",
      "TSMixerx horizon 2 - MAE: 3.171\n",
      "TSMixerx horizon 2 - MSE: 14.609\n",
      "TSMixerx horizon 2 - MAPE: 0.040\n",
      "NBEATSx horizon 2 - MAE: 1.364\n",
      "NBEATSx horizon 2 - MSE: 3.087\n",
      "NBEATSx horizon 2 - MAPE: 0.017\n"
     ]
    }
   ],
   "source": [
    "from neuralforecast.losses.numpy import mse, mae, mape\n",
    "\n",
    "for model in models:\n",
    "    mae_model = mae(Y_hat_df['y'], Y_hat_df[f'{model}']) \n",
    "    mse_model = mse(Y_hat_df['y'], Y_hat_df[f'{model}'])\n",
    "    mape_model = mape(Y_hat_df['y'], Y_hat_df[f'{model}'])\n",
    "    print(f'{model} horizon {horizon} - MAE: {mae_model:.3f}')\n",
    "    print(f'{model} horizon {horizon} - MSE: {mse_model:.3f}')\n",
    "    print(f'{model} horizon {horizon} - MAPE: {mape_model:.3f}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Hyperparameter Tuning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [],
   "source": [
    "from neuralforecast.losses.pytorch import MAE\n",
    "from neuralforecast.auto import AutoNHITS, AutoTSMixer\n",
    "from neuralforecast import NeuralForecast"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [],
   "source": [
    "import optuna\n",
    "optuna.logging.set_verbosity(optuna.logging.WARNING) # Use this to disable training prints from optuna"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [],
   "source": [
    "horizon_len = 12\n",
    "\n",
    "def config_nhits(trial):\n",
    "    return {\n",
    "        \"max_steps\": 1000,                                                                                               # Number of SGD steps\n",
    "        \"input_size\" : trial.suggest_categorical(\"input_size\", [horizon_len, horizon_len*2, horizon_len*3]),                                                                                               # Size of input window\n",
    "        \"learning_rate\": trial.suggest_loguniform(\"learning_rate\", 1e-5, 1e-1),                                         # Initial Learning rate\n",
    "        \"n_pool_kernel_size\": trial.suggest_categorical(\"n_pool_kernel_size\", [[2, 2, 2], [16, 8, 1]]),                 # MaxPool's Kernel size\n",
    "        \"n_freq_downsample\": trial.suggest_categorical(\"n_freq_downsample\", [[168, 24, 1], [24, 12, 1], [1, 1, 1]]),    # Interpolation expressivity ratios\n",
    "        \"val_check_steps\": 50,                                                                                          # Compute validation every 50 steps\n",
    "        \"early_stop_patience_steps\": 5,                                                                                 # Stops at 5 steps max if loss doesn't get beter\n",
    "        \"random_seed\": trial.suggest_int(\"random_seed\", 1, 10),                                                         # Random seed\n",
    "    }\n",
    "\n",
    "def config_tsmixer(trial):\n",
    "    return {\n",
    "        \"max_steps\": 1000,\n",
    "        \"n_series\" : 1,\n",
    "        \"input_size\" : trial.suggest_categorical(\"input_size\", [horizon_len, horizon_len*2, horizon_len*3]),\n",
    "        \"learning_rate\": trial.suggest_loguniform(\"learning_rate\", 1e-5, 1e-1),\n",
    "        \"ff_dim\": trial.suggest_categorical(\"ff_dim\", [32,64,128]),\n",
    "        \"n_block\": trial.suggest_categorical(\"n_block\", [2,4,8]),\n",
    "        \"val_check_steps\": 50,\n",
    "        \"early_stop_patience_steps\": 5,\n",
    "        \"scaler_type\": 'identity',\n",
    "    }"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = [AutoNHITS(h=horizon_len,\n",
    "                  loss=MAE(),\n",
    "                  valid_loss=MAE(),\n",
    "                  config=config_nhits,\n",
    "                  search_alg=optuna.samplers.TPESampler(),\n",
    "                  backend='optuna',\n",
    "                  num_samples=10),\n",
    "        AutoTSMixer(h=horizon_len,\n",
    "                    n_series=1,\n",
    "                  loss=MAE(),\n",
    "                  valid_loss=MAE(),\n",
    "                  config=config_tsmixer,\n",
    "                  search_alg=optuna.samplers.TPESampler(),\n",
    "                  backend='optuna',\n",
    "                  num_samples=10)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nf = NeuralForecast(models=model, freq='D')\n",
    "nf.fit(df=last_df, val_size=24)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "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>number</th>\n",
       "      <th>value</th>\n",
       "      <th>datetime_start</th>\n",
       "      <th>datetime_complete</th>\n",
       "      <th>duration</th>\n",
       "      <th>params_ff_dim</th>\n",
       "      <th>params_input_size</th>\n",
       "      <th>params_learning_rate</th>\n",
       "      <th>params_n_block</th>\n",
       "      <th>user_attrs_METRICS</th>\n",
       "      <th>state</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>2.930035</td>\n",
       "      <td>2024-09-22 11:47:26.290977</td>\n",
       "      <td>2024-09-22 11:47:55.230830</td>\n",
       "      <td>0 days 00:00:28.939853</td>\n",
       "      <td>128</td>\n",
       "      <td>12</td>\n",
       "      <td>0.000020</td>\n",
       "      <td>2</td>\n",
       "      <td>{'loss': tensor(2.9300), 'train_loss': tensor(...</td>\n",
       "      <td>COMPLETE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>3.161288</td>\n",
       "      <td>2024-09-22 11:47:55.230830</td>\n",
       "      <td>2024-09-22 11:48:05.877961</td>\n",
       "      <td>0 days 00:00:10.647131</td>\n",
       "      <td>64</td>\n",
       "      <td>24</td>\n",
       "      <td>0.001222</td>\n",
       "      <td>2</td>\n",
       "      <td>{'loss': tensor(3.1613), 'train_loss': tensor(...</td>\n",
       "      <td>COMPLETE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>2.700425</td>\n",
       "      <td>2024-09-22 11:48:05.878972</td>\n",
       "      <td>2024-09-22 11:48:21.498122</td>\n",
       "      <td>0 days 00:00:15.619150</td>\n",
       "      <td>32</td>\n",
       "      <td>36</td>\n",
       "      <td>0.000025</td>\n",
       "      <td>8</td>\n",
       "      <td>{'loss': tensor(2.7004), 'train_loss': tensor(...</td>\n",
       "      <td>COMPLETE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>2.985046</td>\n",
       "      <td>2024-09-22 11:48:21.499665</td>\n",
       "      <td>2024-09-22 11:48:37.470886</td>\n",
       "      <td>0 days 00:00:15.971221</td>\n",
       "      <td>128</td>\n",
       "      <td>36</td>\n",
       "      <td>0.070560</td>\n",
       "      <td>4</td>\n",
       "      <td>{'loss': tensor(2.9850), 'train_loss': tensor(...</td>\n",
       "      <td>COMPLETE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>2.906783</td>\n",
       "      <td>2024-09-22 11:48:37.471895</td>\n",
       "      <td>2024-09-22 11:49:12.608907</td>\n",
       "      <td>0 days 00:00:35.137012</td>\n",
       "      <td>32</td>\n",
       "      <td>36</td>\n",
       "      <td>0.000031</td>\n",
       "      <td>4</td>\n",
       "      <td>{'loss': tensor(2.9068), 'train_loss': tensor(...</td>\n",
       "      <td>COMPLETE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>5</td>\n",
       "      <td>2.779932</td>\n",
       "      <td>2024-09-22 11:49:12.610470</td>\n",
       "      <td>2024-09-22 11:49:29.336763</td>\n",
       "      <td>0 days 00:00:16.726293</td>\n",
       "      <td>32</td>\n",
       "      <td>12</td>\n",
       "      <td>0.001010</td>\n",
       "      <td>2</td>\n",
       "      <td>{'loss': tensor(2.7799), 'train_loss': tensor(...</td>\n",
       "      <td>COMPLETE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>6</td>\n",
       "      <td>2.776175</td>\n",
       "      <td>2024-09-22 11:49:29.336763</td>\n",
       "      <td>2024-09-22 11:50:10.076748</td>\n",
       "      <td>0 days 00:00:40.739985</td>\n",
       "      <td>128</td>\n",
       "      <td>12</td>\n",
       "      <td>0.000924</td>\n",
       "      <td>8</td>\n",
       "      <td>{'loss': tensor(2.7762), 'train_loss': tensor(...</td>\n",
       "      <td>COMPLETE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>7</td>\n",
       "      <td>3.037341</td>\n",
       "      <td>2024-09-22 11:50:10.076748</td>\n",
       "      <td>2024-09-22 11:50:19.131097</td>\n",
       "      <td>0 days 00:00:09.054349</td>\n",
       "      <td>128</td>\n",
       "      <td>24</td>\n",
       "      <td>0.000014</td>\n",
       "      <td>2</td>\n",
       "      <td>{'loss': tensor(3.0373), 'train_loss': tensor(...</td>\n",
       "      <td>COMPLETE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>8</td>\n",
       "      <td>3.202243</td>\n",
       "      <td>2024-09-22 11:50:19.131097</td>\n",
       "      <td>2024-09-22 11:50:32.949219</td>\n",
       "      <td>0 days 00:00:13.818122</td>\n",
       "      <td>32</td>\n",
       "      <td>24</td>\n",
       "      <td>0.000631</td>\n",
       "      <td>2</td>\n",
       "      <td>{'loss': tensor(3.2022), 'train_loss': tensor(...</td>\n",
       "      <td>COMPLETE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>9</td>\n",
       "      <td>3.245659</td>\n",
       "      <td>2024-09-22 11:50:32.950220</td>\n",
       "      <td>2024-09-22 11:50:53.980985</td>\n",
       "      <td>0 days 00:00:21.030765</td>\n",
       "      <td>64</td>\n",
       "      <td>24</td>\n",
       "      <td>0.008169</td>\n",
       "      <td>8</td>\n",
       "      <td>{'loss': tensor(3.2457), 'train_loss': tensor(...</td>\n",
       "      <td>COMPLETE</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   number     value             datetime_start          datetime_complete  \\\n",
       "0       0  2.930035 2024-09-22 11:47:26.290977 2024-09-22 11:47:55.230830   \n",
       "1       1  3.161288 2024-09-22 11:47:55.230830 2024-09-22 11:48:05.877961   \n",
       "2       2  2.700425 2024-09-22 11:48:05.878972 2024-09-22 11:48:21.498122   \n",
       "3       3  2.985046 2024-09-22 11:48:21.499665 2024-09-22 11:48:37.470886   \n",
       "4       4  2.906783 2024-09-22 11:48:37.471895 2024-09-22 11:49:12.608907   \n",
       "5       5  2.779932 2024-09-22 11:49:12.610470 2024-09-22 11:49:29.336763   \n",
       "6       6  2.776175 2024-09-22 11:49:29.336763 2024-09-22 11:50:10.076748   \n",
       "7       7  3.037341 2024-09-22 11:50:10.076748 2024-09-22 11:50:19.131097   \n",
       "8       8  3.202243 2024-09-22 11:50:19.131097 2024-09-22 11:50:32.949219   \n",
       "9       9  3.245659 2024-09-22 11:50:32.950220 2024-09-22 11:50:53.980985   \n",
       "\n",
       "                duration  params_ff_dim  params_input_size  \\\n",
       "0 0 days 00:00:28.939853            128                 12   \n",
       "1 0 days 00:00:10.647131             64                 24   \n",
       "2 0 days 00:00:15.619150             32                 36   \n",
       "3 0 days 00:00:15.971221            128                 36   \n",
       "4 0 days 00:00:35.137012             32                 36   \n",
       "5 0 days 00:00:16.726293             32                 12   \n",
       "6 0 days 00:00:40.739985            128                 12   \n",
       "7 0 days 00:00:09.054349            128                 24   \n",
       "8 0 days 00:00:13.818122             32                 24   \n",
       "9 0 days 00:00:21.030765             64                 24   \n",
       "\n",
       "   params_learning_rate  params_n_block  \\\n",
       "0              0.000020               2   \n",
       "1              0.001222               2   \n",
       "2              0.000025               8   \n",
       "3              0.070560               4   \n",
       "4              0.000031               4   \n",
       "5              0.001010               2   \n",
       "6              0.000924               8   \n",
       "7              0.000014               2   \n",
       "8              0.000631               2   \n",
       "9              0.008169               8   \n",
       "\n",
       "                                  user_attrs_METRICS     state  \n",
       "0  {'loss': tensor(2.9300), 'train_loss': tensor(...  COMPLETE  \n",
       "1  {'loss': tensor(3.1613), 'train_loss': tensor(...  COMPLETE  \n",
       "2  {'loss': tensor(2.7004), 'train_loss': tensor(...  COMPLETE  \n",
       "3  {'loss': tensor(2.9850), 'train_loss': tensor(...  COMPLETE  \n",
       "4  {'loss': tensor(2.9068), 'train_loss': tensor(...  COMPLETE  \n",
       "5  {'loss': tensor(2.7799), 'train_loss': tensor(...  COMPLETE  \n",
       "6  {'loss': tensor(2.7762), 'train_loss': tensor(...  COMPLETE  \n",
       "7  {'loss': tensor(3.0373), 'train_loss': tensor(...  COMPLETE  \n",
       "8  {'loss': tensor(3.2022), 'train_loss': tensor(...  COMPLETE  \n",
       "9  {'loss': tensor(3.2457), 'train_loss': tensor(...  COMPLETE  "
      ]
     },
     "execution_count": 133,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results = nf.models[1].results.trials_dataframe()\n",
    "results.drop(columns='user_attrs_ALL_PARAMS')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Trainer already configured with model summary callbacks: [<class 'pytorch_lightning.callbacks.model_summary.ModelSummary'>]. Skipping setting a default `ModelSummary` callback.\n",
      "GPU available: True (cuda), used: True\n",
      "TPU available: False, using: 0 TPU cores\n",
      "HPU available: False, using: 0 HPUs\n",
      "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predicting DataLoader 0: 100%|██████████| 1/1 [00:00<00:00, 73.97it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Trainer already configured with model summary callbacks: [<class 'pytorch_lightning.callbacks.model_summary.ModelSummary'>]. Skipping setting a default `ModelSummary` callback.\n",
      "GPU available: True (cuda), used: True\n",
      "TPU available: False, using: 0 TPU cores\n",
      "HPU available: False, using: 0 HPUs\n",
      "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Predicting DataLoader 0: 100%|██████████| 1/1 [00:00<00:00, 53.50it/s]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "e:\\VM\\miniconda3\\envs\\multi-its\\lib\\site-packages\\neuralforecast\\core.py:209: FutureWarning: In a future version the predictions will have the id as a column. You can set the `NIXTLA_ID_AS_COL` environment variable to adopt the new behavior and to suppress this warning.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>unique_id</th>\n",
       "      <th>ds</th>\n",
       "      <th>AutoNHITS</th>\n",
       "      <th>AutoTSMixer</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-06</td>\n",
       "      <td>71.889229</td>\n",
       "      <td>80.282181</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-07</td>\n",
       "      <td>71.929710</td>\n",
       "      <td>77.726891</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-08</td>\n",
       "      <td>71.966599</td>\n",
       "      <td>78.719620</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-09</td>\n",
       "      <td>71.985695</td>\n",
       "      <td>79.689644</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-10</td>\n",
       "      <td>71.991013</td>\n",
       "      <td>81.262947</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-11</td>\n",
       "      <td>72.061859</td>\n",
       "      <td>81.586983</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-12</td>\n",
       "      <td>72.137817</td>\n",
       "      <td>80.715485</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-13</td>\n",
       "      <td>72.140602</td>\n",
       "      <td>80.055084</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-14</td>\n",
       "      <td>72.211769</td>\n",
       "      <td>80.489311</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-15</td>\n",
       "      <td>72.257744</td>\n",
       "      <td>81.713593</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-16</td>\n",
       "      <td>72.288536</td>\n",
       "      <td>79.390053</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-17</td>\n",
       "      <td>72.298134</td>\n",
       "      <td>81.773094</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   unique_id         ds  AutoNHITS  AutoTSMixer\n",
       "0      Dated 2024-08-06  71.889229    80.282181\n",
       "1      Dated 2024-08-07  71.929710    77.726891\n",
       "2      Dated 2024-08-08  71.966599    78.719620\n",
       "3      Dated 2024-08-09  71.985695    79.689644\n",
       "4      Dated 2024-08-10  71.991013    81.262947\n",
       "5      Dated 2024-08-11  72.061859    81.586983\n",
       "6      Dated 2024-08-12  72.137817    80.715485\n",
       "7      Dated 2024-08-13  72.140602    80.055084\n",
       "8      Dated 2024-08-14  72.211769    80.489311\n",
       "9      Dated 2024-08-15  72.257744    81.713593\n",
       "10     Dated 2024-08-16  72.288536    79.390053\n",
       "11     Dated 2024-08-17  72.298134    81.773094"
      ]
     },
     "execution_count": 128,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y_hat_df_optuna = nf.predict()\n",
    "Y_hat_df_optuna = Y_hat_df_optuna.reset_index()\n",
    "Y_hat_df_optuna"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x700 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "fig, ax = plt.subplots(1, 1, figsize = (20, 7))\n",
    "plot_df = pd.concat([Y_df, Y_hat_df]).reset_index()\n",
    "\n",
    "#plt.plot(plot_df['ds'], plot_df['y'], label='y')\n",
    "plt.plot(Y_hat_df_optuna['ds'], Y_hat_df_optuna['AutoNHITS'], label='AutoNHITS')\n",
    "plt.plot(Y_hat_df_optuna['ds'], Y_hat_df_optuna['AutoTSMixer'], label='AutoTSMixer')\n",
    "\n",
    "ax.set_title('AirPassengers Forecast', fontsize=22)\n",
    "ax.set_ylabel('Monthly Passengers', fontsize=20)\n",
    "ax.set_xlabel('Timestamp [t]', fontsize=20)\n",
    "ax.legend(prop={'size': 15})\n",
    "ax.grid()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature Engineering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 184,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_df = gasoline_data.drop(columns=['High', 'Low']).rename({'Close' : 'Gas'}, axis=1).merge(\n",
    "    dollar_data.drop(columns=['High', 'Low']).rename({'Close' : 'DXY'}, axis=1), on='Date').merge(\n",
    "    gold_price_data.drop(columns=['High', 'Low']).rename({'Close' : 'Gold(CMX)'}, axis=1), on='Date').merge(\n",
    "    silver_price_data.drop(columns=['High', 'Low']).rename({'Close' : 'Silver(CMX)'}, axis=1), on='Date').merge(\n",
    "    platinum_price_data.drop(columns=['High', 'Low']).rename({'Close' : 'Platinum(NMX)'}, axis=1), on='Date').merge(\n",
    "    palladium_price_data.drop(columns=['High', 'Low']).rename({'Close' : 'Palladium(NMX)'}, axis=1), on='Date').merge(\n",
    "    us_bond_data.drop(columns=['High', 'Low']).rename({'Close' : 'USBond'}, axis=1), on='Date').merge(\n",
    "    sp500_data.drop(columns=['High', 'Low']).rename({'Close' : 'S&P500'}, axis=1), on='Date').merge(\n",
    "    gold_miners_data.drop(columns=['High', 'Low']).rename({'Close' : 'GoldMiners'}, axis=1), on='Date').merge(\n",
    "    gpr_data[['Date', 'GPRD']])\n",
    "\n",
    "y_df = brent_futures_data[['Date','Close']].rename({'Close' : 'BrFu'}, axis=1)\n",
    "\n",
    "df = X_df.merge(y_df, on ='Date').rename({'Date' : 'ds'}, axis=1)\n",
    "df['ds'] = pd.to_datetime(df['ds'])\n",
    "\n",
    "y_df = df['BrFu']\n",
    "\n",
    "df.drop(columns=['BrFu'], inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 187,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['unique_id'] = 1\n",
    "\n",
    "y_df = y_df.rename({'BrFu' : 'y'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 189,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Feature Extraction: 100%|██████████| 10/10 [00:22<00:00,  2.20s/it]\n"
     ]
    }
   ],
   "source": [
    "from tsfresh import extract_features\n",
    "extracted_features = extract_features(df, column_id=\"unique_id\", column_sort=\"ds\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 193,
   "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>ds</th>\n",
       "      <th>Gas</th>\n",
       "      <th>DXY</th>\n",
       "      <th>Gold(CMX)</th>\n",
       "      <th>Silver(CMX)</th>\n",
       "      <th>Platinum(NMX)</th>\n",
       "      <th>Palladium(NMX)</th>\n",
       "      <th>USBond</th>\n",
       "      <th>S&amp;P500</th>\n",
       "      <th>GoldMiners</th>\n",
       "      <th>GPRD</th>\n",
       "      <th>unique_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2024-08-05</td>\n",
       "      <td>2.3336</td>\n",
       "      <td>102.690002</td>\n",
       "      <td>2444.4</td>\n",
       "      <td>27.207</td>\n",
       "      <td>915.5</td>\n",
       "      <td>826.10</td>\n",
       "      <td>125.4062</td>\n",
       "      <td>5186.33</td>\n",
       "      <td>35.34</td>\n",
       "      <td>210.206711</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2024-08-02</td>\n",
       "      <td>2.3176</td>\n",
       "      <td>103.209999</td>\n",
       "      <td>2469.8</td>\n",
       "      <td>28.392</td>\n",
       "      <td>967.6</td>\n",
       "      <td>882.50</td>\n",
       "      <td>125.0625</td>\n",
       "      <td>5346.56</td>\n",
       "      <td>36.48</td>\n",
       "      <td>207.799438</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2024-08-01</td>\n",
       "      <td>2.3980</td>\n",
       "      <td>104.419998</td>\n",
       "      <td>2480.8</td>\n",
       "      <td>28.477</td>\n",
       "      <td>970.5</td>\n",
       "      <td>895.10</td>\n",
       "      <td>122.5312</td>\n",
       "      <td>5446.68</td>\n",
       "      <td>37.27</td>\n",
       "      <td>139.878098</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2024-07-31</td>\n",
       "      <td>2.4425</td>\n",
       "      <td>104.099998</td>\n",
       "      <td>2473.0</td>\n",
       "      <td>28.938</td>\n",
       "      <td>986.4</td>\n",
       "      <td>925.20</td>\n",
       "      <td>120.7812</td>\n",
       "      <td>5522.30</td>\n",
       "      <td>37.93</td>\n",
       "      <td>135.206848</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2024-07-30</td>\n",
       "      <td>2.3443</td>\n",
       "      <td>104.550003</td>\n",
       "      <td>2451.9</td>\n",
       "      <td>28.525</td>\n",
       "      <td>971.0</td>\n",
       "      <td>881.70</td>\n",
       "      <td>120.1875</td>\n",
       "      <td>5436.44</td>\n",
       "      <td>36.95</td>\n",
       "      <td>95.696396</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2496</th>\n",
       "      <td>2014-08-29</td>\n",
       "      <td>2.6229</td>\n",
       "      <td>82.750000</td>\n",
       "      <td>1287.4</td>\n",
       "      <td>19.492</td>\n",
       "      <td>1424.7</td>\n",
       "      <td>909.55</td>\n",
       "      <td>140.0938</td>\n",
       "      <td>2003.37</td>\n",
       "      <td>26.69</td>\n",
       "      <td>157.336334</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2497</th>\n",
       "      <td>2014-08-28</td>\n",
       "      <td>2.5908</td>\n",
       "      <td>82.480003</td>\n",
       "      <td>1290.4</td>\n",
       "      <td>19.609</td>\n",
       "      <td>1425.2</td>\n",
       "      <td>898.10</td>\n",
       "      <td>141.7188</td>\n",
       "      <td>1996.74</td>\n",
       "      <td>26.46</td>\n",
       "      <td>97.534706</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2498</th>\n",
       "      <td>2014-08-27</td>\n",
       "      <td>2.5905</td>\n",
       "      <td>82.430000</td>\n",
       "      <td>1283.4</td>\n",
       "      <td>19.475</td>\n",
       "      <td>1419.9</td>\n",
       "      <td>894.70</td>\n",
       "      <td>141.1562</td>\n",
       "      <td>2000.12</td>\n",
       "      <td>26.11</td>\n",
       "      <td>143.082108</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2499</th>\n",
       "      <td>2014-08-26</td>\n",
       "      <td>2.6001</td>\n",
       "      <td>82.650002</td>\n",
       "      <td>1285.2</td>\n",
       "      <td>19.459</td>\n",
       "      <td>1419.6</td>\n",
       "      <td>890.15</td>\n",
       "      <td>140.5312</td>\n",
       "      <td>2000.02</td>\n",
       "      <td>26.19</td>\n",
       "      <td>118.143707</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2500</th>\n",
       "      <td>2014-08-25</td>\n",
       "      <td>2.5954</td>\n",
       "      <td>82.550003</td>\n",
       "      <td>1278.9</td>\n",
       "      <td>19.431</td>\n",
       "      <td>1418.4</td>\n",
       "      <td>891.50</td>\n",
       "      <td>140.7188</td>\n",
       "      <td>1997.92</td>\n",
       "      <td>25.62</td>\n",
       "      <td>167.600159</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2501 rows × 12 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             ds     Gas         DXY  Gold(CMX)  Silver(CMX)  Platinum(NMX)  \\\n",
       "0    2024-08-05  2.3336  102.690002     2444.4       27.207          915.5   \n",
       "1    2024-08-02  2.3176  103.209999     2469.8       28.392          967.6   \n",
       "2    2024-08-01  2.3980  104.419998     2480.8       28.477          970.5   \n",
       "3    2024-07-31  2.4425  104.099998     2473.0       28.938          986.4   \n",
       "4    2024-07-30  2.3443  104.550003     2451.9       28.525          971.0   \n",
       "...         ...     ...         ...        ...          ...            ...   \n",
       "2496 2014-08-29  2.6229   82.750000     1287.4       19.492         1424.7   \n",
       "2497 2014-08-28  2.5908   82.480003     1290.4       19.609         1425.2   \n",
       "2498 2014-08-27  2.5905   82.430000     1283.4       19.475         1419.9   \n",
       "2499 2014-08-26  2.6001   82.650002     1285.2       19.459         1419.6   \n",
       "2500 2014-08-25  2.5954   82.550003     1278.9       19.431         1418.4   \n",
       "\n",
       "      Palladium(NMX)    USBond   S&P500  GoldMiners        GPRD  unique_id  \n",
       "0             826.10  125.4062  5186.33       35.34  210.206711          1  \n",
       "1             882.50  125.0625  5346.56       36.48  207.799438          1  \n",
       "2             895.10  122.5312  5446.68       37.27  139.878098          1  \n",
       "3             925.20  120.7812  5522.30       37.93  135.206848          1  \n",
       "4             881.70  120.1875  5436.44       36.95   95.696396          1  \n",
       "...              ...       ...      ...         ...         ...        ...  \n",
       "2496          909.55  140.0938  2003.37       26.69  157.336334          1  \n",
       "2497          898.10  141.7188  1996.74       26.46   97.534706          1  \n",
       "2498          894.70  141.1562  2000.12       26.11  143.082108          1  \n",
       "2499          890.15  140.5312  2000.02       26.19  118.143707          1  \n",
       "2500          891.50  140.7188  1997.92       25.62  167.600159          1  \n",
       "\n",
       "[2501 rows x 12 columns]"
      ]
     },
     "execution_count": 193,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 190,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "e:\\VM\\miniconda3\\envs\\multi-its\\lib\\site-packages\\tsfresh\\utilities\\dataframe_functions.py:198: RuntimeWarning: The columns ['Gas__query_similarity_count__query_None__threshold_0.0'\n",
      " 'DXY__query_similarity_count__query_None__threshold_0.0'\n",
      " 'Platinum(NMX)__query_similarity_count__query_None__threshold_0.0'\n",
      " 'Palladium(NMX)__query_similarity_count__query_None__threshold_0.0'\n",
      " 'Silver(CMX)__query_similarity_count__query_None__threshold_0.0'\n",
      " 'Gold(CMX)__query_similarity_count__query_None__threshold_0.0'\n",
      " 'USBond__query_similarity_count__query_None__threshold_0.0'\n",
      " 'S&P500__query_similarity_count__query_None__threshold_0.0'\n",
      " 'GoldMiners__query_similarity_count__query_None__threshold_0.0'\n",
      " 'GPRD__query_similarity_count__query_None__threshold_0.0'] did not have any finite values. Filling with zeros.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "ename": "AssertionError",
     "evalue": "X and y must contain the same number of samples.",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mAssertionError\u001b[0m                            Traceback (most recent call last)",
      "Cell \u001b[1;32mIn[190], line 5\u001b[0m\n\u001b[0;32m      2\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtsfresh\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mutilities\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mdataframe_functions\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m impute\n\u001b[0;32m      4\u001b[0m impute(extracted_features)\n\u001b[1;32m----> 5\u001b[0m features_filtered \u001b[38;5;241m=\u001b[39m \u001b[43mselect_features\u001b[49m\u001b[43m(\u001b[49m\u001b[43mextracted_features\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43my_df\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[1;32me:\\VM\\miniconda3\\envs\\multi-its\\lib\\site-packages\\tsfresh\\feature_selection\\selection.py:154\u001b[0m, in \u001b[0;36mselect_features\u001b[1;34m(X, y, test_for_binary_target_binary_feature, test_for_binary_target_real_feature, test_for_real_target_binary_feature, test_for_real_target_real_feature, fdr_level, hypotheses_independent, n_jobs, show_warnings, chunksize, ml_task, multiclass, n_significant)\u001b[0m\n\u001b[0;32m    150\u001b[0m \u001b[38;5;28;01massert\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(y, (pd\u001b[38;5;241m.\u001b[39mSeries, np\u001b[38;5;241m.\u001b[39mndarray)), (\n\u001b[0;32m    151\u001b[0m     \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mThe type of target vector y must be one of: \u001b[39m\u001b[38;5;124m\"\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mpandas.Series, numpy.ndarray\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m    152\u001b[0m )\n\u001b[0;32m    153\u001b[0m \u001b[38;5;28;01massert\u001b[39;00m \u001b[38;5;28mlen\u001b[39m(y) \u001b[38;5;241m>\u001b[39m \u001b[38;5;241m1\u001b[39m, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124my must contain at least two samples.\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m--> 154\u001b[0m \u001b[38;5;28;01massert\u001b[39;00m \u001b[38;5;28mlen\u001b[39m(X) \u001b[38;5;241m==\u001b[39m \u001b[38;5;28mlen\u001b[39m(y), \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mX and y must contain the same number of samples.\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m    155\u001b[0m \u001b[38;5;28;01massert\u001b[39;00m (\n\u001b[0;32m    156\u001b[0m     \u001b[38;5;28mlen\u001b[39m(\u001b[38;5;28mset\u001b[39m(y)) \u001b[38;5;241m>\u001b[39m \u001b[38;5;241m1\u001b[39m\n\u001b[0;32m    157\u001b[0m ), \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mFeature selection is only possible if more than 1 label/class is provided\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m    159\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(y, pd\u001b[38;5;241m.\u001b[39mSeries) \u001b[38;5;129;01mand\u001b[39;00m \u001b[38;5;28mset\u001b[39m(X\u001b[38;5;241m.\u001b[39mindex) \u001b[38;5;241m!=\u001b[39m \u001b[38;5;28mset\u001b[39m(y\u001b[38;5;241m.\u001b[39mindex):\n",
      "\u001b[1;31mAssertionError\u001b[0m: X and y must contain the same number of samples."
     ]
    }
   ],
   "source": [
    "from tsfresh import select_features\n",
    "from tsfresh.utilities.dataframe_functions import impute\n",
    "\n",
    "impute(extracted_features)\n",
    "features_filtered = select_features(extracted_features, y_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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>y__variance_larger_than_standard_deviation</th>\n",
       "      <th>y__has_duplicate_max</th>\n",
       "      <th>y__has_duplicate_min</th>\n",
       "      <th>y__has_duplicate</th>\n",
       "      <th>y__sum_values</th>\n",
       "      <th>y__abs_energy</th>\n",
       "      <th>y__mean_abs_change</th>\n",
       "      <th>y__mean_change</th>\n",
       "      <th>y__mean_second_derivative_central</th>\n",
       "      <th>y__median</th>\n",
       "      <th>...</th>\n",
       "      <th>GPRD__fourier_entropy__bins_5</th>\n",
       "      <th>GPRD__fourier_entropy__bins_10</th>\n",
       "      <th>GPRD__fourier_entropy__bins_100</th>\n",
       "      <th>GPRD__permutation_entropy__dimension_3__tau_1</th>\n",
       "      <th>GPRD__permutation_entropy__dimension_4__tau_1</th>\n",
       "      <th>GPRD__permutation_entropy__dimension_5__tau_1</th>\n",
       "      <th>GPRD__permutation_entropy__dimension_6__tau_1</th>\n",
       "      <th>GPRD__permutation_entropy__dimension_7__tau_1</th>\n",
       "      <th>GPRD__query_similarity_count__query_None__threshold_0.0</th>\n",
       "      <th>GPRD__mean_n_absolute_max__number_of_maxima_7</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Dated</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4.547474e-13</td>\n",
       "      <td>2501.0</td>\n",
       "      <td>0.062532</td>\n",
       "      <td>0.000441</td>\n",
       "      <td>-0.000012</td>\n",
       "      <td>-0.142643</td>\n",
       "      <td>...</td>\n",
       "      <td>0.170467</td>\n",
       "      <td>0.390338</td>\n",
       "      <td>1.894835</td>\n",
       "      <td>1.791383</td>\n",
       "      <td>3.171978</td>\n",
       "      <td>4.758524</td>\n",
       "      <td>6.418777</td>\n",
       "      <td>7.498423</td>\n",
       "      <td>NaN</td>\n",
       "      <td>6.830437</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1 rows × 9396 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       y__variance_larger_than_standard_deviation  y__has_duplicate_max  \\\n",
       "Dated                                         0.0                   0.0   \n",
       "\n",
       "       y__has_duplicate_min  y__has_duplicate  y__sum_values  y__abs_energy  \\\n",
       "Dated                   0.0               1.0   4.547474e-13         2501.0   \n",
       "\n",
       "       y__mean_abs_change  y__mean_change  y__mean_second_derivative_central  \\\n",
       "Dated            0.062532        0.000441                          -0.000012   \n",
       "\n",
       "       y__median  ...  GPRD__fourier_entropy__bins_5  \\\n",
       "Dated  -0.142643  ...                       0.170467   \n",
       "\n",
       "       GPRD__fourier_entropy__bins_10  GPRD__fourier_entropy__bins_100  \\\n",
       "Dated                        0.390338                         1.894835   \n",
       "\n",
       "       GPRD__permutation_entropy__dimension_3__tau_1  \\\n",
       "Dated                                       1.791383   \n",
       "\n",
       "       GPRD__permutation_entropy__dimension_4__tau_1  \\\n",
       "Dated                                       3.171978   \n",
       "\n",
       "       GPRD__permutation_entropy__dimension_5__tau_1  \\\n",
       "Dated                                       4.758524   \n",
       "\n",
       "       GPRD__permutation_entropy__dimension_6__tau_1  \\\n",
       "Dated                                       6.418777   \n",
       "\n",
       "       GPRD__permutation_entropy__dimension_7__tau_1  \\\n",
       "Dated                                       7.498423   \n",
       "\n",
       "       GPRD__query_similarity_count__query_None__threshold_0.0  \\\n",
       "Dated                                                NaN         \n",
       "\n",
       "       GPRD__mean_n_absolute_max__number_of_maxima_7  \n",
       "Dated                                       6.830437  \n",
       "\n",
       "[1 rows x 9396 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "extracted_features"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Save and load models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "metadata": {},
   "outputs": [],
   "source": [
    "nf.save(path='./checkpoints/test_run/',\n",
    "        model_index=None, \n",
    "        overwrite=True,\n",
    "        save_dataset=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 158,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "e:\\VM\\miniconda3\\envs\\multi-its\\lib\\site-packages\\neuralforecast\\common\\_base_model.py:444: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.\n",
      "  content = torch.load(f, **kwargs)\n",
      "Seed set to 8\n",
      "e:\\VM\\miniconda3\\envs\\multi-its\\lib\\site-packages\\neuralforecast\\common\\_base_model.py:444: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.\n",
      "  content = torch.load(f, **kwargs)\n",
      "Seed set to 1\n",
      "e:\\VM\\miniconda3\\envs\\multi-its\\lib\\site-packages\\torch\\storage.py:414: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.\n",
      "  return torch.load(io.BytesIO(b))\n",
      "GPU available: True (cuda), used: True\n",
      "TPU available: False, using: 0 TPU cores\n",
      "HPU available: False, using: 0 HPUs\n",
      "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predicting DataLoader 0: 100%|██████████| 1/1 [00:00<00:00, 58.57it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "GPU available: True (cuda), used: True\n",
      "TPU available: False, using: 0 TPU cores\n",
      "HPU available: False, using: 0 HPUs\n",
      "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Predicting DataLoader 0: 100%|██████████| 1/1 [00:00<00:00, 31.66it/s]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "e:\\VM\\miniconda3\\envs\\multi-its\\lib\\site-packages\\neuralforecast\\core.py:209: FutureWarning: In a future version the predictions will have the id as a column. You can set the `NIXTLA_ID_AS_COL` environment variable to adopt the new behavior and to suppress this warning.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>unique_id</th>\n",
       "      <th>ds</th>\n",
       "      <th>AutoNHITS</th>\n",
       "      <th>AutoTSMixer</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-06</td>\n",
       "      <td>71.889229</td>\n",
       "      <td>80.282181</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-07</td>\n",
       "      <td>71.929710</td>\n",
       "      <td>77.726891</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-08</td>\n",
       "      <td>71.966599</td>\n",
       "      <td>78.719620</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-09</td>\n",
       "      <td>71.985695</td>\n",
       "      <td>79.689644</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-10</td>\n",
       "      <td>71.991013</td>\n",
       "      <td>81.262947</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  unique_id         ds  AutoNHITS  AutoTSMixer\n",
       "0     Dated 2024-08-06  71.889229    80.282181\n",
       "1     Dated 2024-08-07  71.929710    77.726891\n",
       "2     Dated 2024-08-08  71.966599    78.719620\n",
       "3     Dated 2024-08-09  71.985695    79.689644\n",
       "4     Dated 2024-08-10  71.991013    81.262947"
      ]
     },
     "execution_count": 158,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nf2 = NeuralForecast.load(path='./checkpoints/test_run/')\n",
    "Y_hat_df = nf2.predict().reset_index()\n",
    "Y_hat_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 159,
   "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>unique_id</th>\n",
       "      <th>ds</th>\n",
       "      <th>AutoNHITS</th>\n",
       "      <th>AutoTSMixer</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-06</td>\n",
       "      <td>71.889229</td>\n",
       "      <td>80.282181</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-07</td>\n",
       "      <td>71.929710</td>\n",
       "      <td>77.726891</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-08</td>\n",
       "      <td>71.966599</td>\n",
       "      <td>78.719620</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-09</td>\n",
       "      <td>71.985695</td>\n",
       "      <td>79.689644</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-10</td>\n",
       "      <td>71.991013</td>\n",
       "      <td>81.262947</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-11</td>\n",
       "      <td>72.061859</td>\n",
       "      <td>81.586983</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-12</td>\n",
       "      <td>72.137817</td>\n",
       "      <td>80.715485</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-13</td>\n",
       "      <td>72.140602</td>\n",
       "      <td>80.055084</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-14</td>\n",
       "      <td>72.211769</td>\n",
       "      <td>80.489311</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-15</td>\n",
       "      <td>72.257744</td>\n",
       "      <td>81.713593</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-16</td>\n",
       "      <td>72.288536</td>\n",
       "      <td>79.390053</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Dated</td>\n",
       "      <td>2024-08-17</td>\n",
       "      <td>72.298134</td>\n",
       "      <td>81.773094</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   unique_id         ds  AutoNHITS  AutoTSMixer\n",
       "0      Dated 2024-08-06  71.889229    80.282181\n",
       "1      Dated 2024-08-07  71.929710    77.726891\n",
       "2      Dated 2024-08-08  71.966599    78.719620\n",
       "3      Dated 2024-08-09  71.985695    79.689644\n",
       "4      Dated 2024-08-10  71.991013    81.262947\n",
       "5      Dated 2024-08-11  72.061859    81.586983\n",
       "6      Dated 2024-08-12  72.137817    80.715485\n",
       "7      Dated 2024-08-13  72.140602    80.055084\n",
       "8      Dated 2024-08-14  72.211769    80.489311\n",
       "9      Dated 2024-08-15  72.257744    81.713593\n",
       "10     Dated 2024-08-16  72.288536    79.390053\n",
       "11     Dated 2024-08-17  72.298134    81.773094"
      ]
     },
     "execution_count": 159,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y_hat_df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "\n",
    "test = pd.read_csv('artifacts\\cf12e42a-5a61-4bb0-bce2-0c0b9b246e53\\dataset.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Select columns that end with 'Close' except for 'Date'\n",
    "close_columns_df = pd.concat([test['Date'], test.filter(regex=r'Close$')], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0      2014-08-25\n",
       "1      2014-08-26\n",
       "2      2014-08-27\n",
       "3      2014-08-28\n",
       "4      2014-08-29\n",
       "          ...    \n",
       "2495   2024-07-29\n",
       "2496   2024-07-30\n",
       "2497   2024-07-31\n",
       "2498   2024-08-01\n",
       "2499   2024-08-02\n",
       "Name: Date, Length: 2500, dtype: datetime64[ns]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.to_datetime(close_columns_df['Date'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "TimeSeriesDataset(n_data=2,500, n_groups=1)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pickle as pkl\n",
    "import pandas as pd\n",
    "\n",
    "object = 'artifacts/498f27cb-5e23-4503-9323-a4333f22becf/model/dataset.pkl'\n",
    "\n",
    "df = pd.read_pickle(object)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "TimeSeriesDataset(n_data=2,500, n_groups=1)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "object"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "multi-its",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.19"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}