LSTM_Price_Prediction_With_Sentiment_and_indicators.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "import torchvision\n",
    "from torch.utils.data import DataLoader,TensorDataset\n",
    "import matplotlib.pyplot as plt\n",
    "from statsmodels.tsa.seasonal import seasonal_decompose\n",
    "from sklearn.metrics import mean_squared_error, mean_absolute_error\n",
    "from sklearn.preprocessing import MinMaxScaler, StandardScaler\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Date', 'Title', 'URL', 'sentiment', 'confidence', 'Text', 'pct',\n",
       "       'pred', 'weight'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2=pd.read_csv('df2.csv')\n",
    "df2.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "765"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sms=[]\n",
    "for date, group in df2.groupby('Date'):\n",
    "  cons = list(group['confidence'])\n",
    "  wes = list(group['weight'])\n",
    "  s = 0\n",
    "  for i in range(len(cons)):\n",
    "    s += cons[i]*wes[i]\n",
    "  sms.append(s)\n",
    "\n",
    "len(sms) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "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>Gold</th>\n",
       "      <th>Coal</th>\n",
       "      <th>Closing_price</th>\n",
       "      <th>Crude_Oil</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-01-03</td>\n",
       "      <td>79035.1</td>\n",
       "      <td>305.30</td>\n",
       "      <td>498.037018</td>\n",
       "      <td>52.330002</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017-01-04</td>\n",
       "      <td>79721.3</td>\n",
       "      <td>306.15</td>\n",
       "      <td>487.865082</td>\n",
       "      <td>53.259998</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017-01-06</td>\n",
       "      <td>80321.3</td>\n",
       "      <td>306.20</td>\n",
       "      <td>491.430969</td>\n",
       "      <td>53.990002</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017-01-07</td>\n",
       "      <td>80321.3</td>\n",
       "      <td>306.20</td>\n",
       "      <td>492.688171</td>\n",
       "      <td>51.959999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2017-01-08</td>\n",
       "      <td>80321.3</td>\n",
       "      <td>306.20</td>\n",
       "      <td>492.688171</td>\n",
       "      <td>51.959999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>760</th>\n",
       "      <td>2021-09-02</td>\n",
       "      <td>133121.0</td>\n",
       "      <td>134.50</td>\n",
       "      <td>2117.730469</td>\n",
       "      <td>69.989998</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>761</th>\n",
       "      <td>2021-09-03</td>\n",
       "      <td>133146.8</td>\n",
       "      <td>141.05</td>\n",
       "      <td>2204.584717</td>\n",
       "      <td>69.290001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>762</th>\n",
       "      <td>2021-10-01</td>\n",
       "      <td>130776.3</td>\n",
       "      <td>147.60</td>\n",
       "      <td>2329.374268</td>\n",
       "      <td>75.879997</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>763</th>\n",
       "      <td>2021-10-02</td>\n",
       "      <td>130776.3</td>\n",
       "      <td>149.10</td>\n",
       "      <td>2359.325684</td>\n",
       "      <td>77.620003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>764</th>\n",
       "      <td>2021-10-03</td>\n",
       "      <td>130776.3</td>\n",
       "      <td>153.10</td>\n",
       "      <td>2359.325684</td>\n",
       "      <td>77.620003</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>765 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           Date      Gold    Coal  Closing_price  Crude_Oil\n",
       "0    2017-01-03   79035.1  305.30     498.037018  52.330002\n",
       "1    2017-01-04   79721.3  306.15     487.865082  53.259998\n",
       "2    2017-01-06   80321.3  306.20     491.430969  53.990002\n",
       "3    2017-01-07   80321.3  306.20     492.688171  51.959999\n",
       "4    2017-01-08   80321.3  306.20     492.688171  51.959999\n",
       "..          ...       ...     ...            ...        ...\n",
       "760  2021-09-02  133121.0  134.50    2117.730469  69.989998\n",
       "761  2021-09-03  133146.8  141.05    2204.584717  69.290001\n",
       "762  2021-10-01  130776.3  147.60    2329.374268  75.879997\n",
       "763  2021-10-02  130776.3  149.10    2359.325684  77.620003\n",
       "764  2021-10-03  130776.3  153.10    2359.325684  77.620003\n",
       "\n",
       "[765 rows x 5 columns]"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df=pd.read_csv('Stock.csv')\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "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>Gold</th>\n",
       "      <th>Coal</th>\n",
       "      <th>Closing_price</th>\n",
       "      <th>Crude_Oil</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-01-03</td>\n",
       "      <td>-1.086089</td>\n",
       "      <td>1.359663</td>\n",
       "      <td>0.017401</td>\n",
       "      <td>-0.003241</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017-01-04</td>\n",
       "      <td>-1.054820</td>\n",
       "      <td>1.373258</td>\n",
       "      <td>0.012031</td>\n",
       "      <td>0.074474</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017-01-06</td>\n",
       "      <td>-1.027479</td>\n",
       "      <td>1.374057</td>\n",
       "      <td>0.013913</td>\n",
       "      <td>0.135477</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017-01-07</td>\n",
       "      <td>-1.027479</td>\n",
       "      <td>1.374057</td>\n",
       "      <td>0.014577</td>\n",
       "      <td>-0.034160</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2017-01-08</td>\n",
       "      <td>-1.027479</td>\n",
       "      <td>1.374057</td>\n",
       "      <td>0.014577</td>\n",
       "      <td>-0.034160</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>760</th>\n",
       "      <td>2021-09-02</td>\n",
       "      <td>1.378535</td>\n",
       "      <td>-1.372021</td>\n",
       "      <td>0.872459</td>\n",
       "      <td>1.472518</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>761</th>\n",
       "      <td>2021-09-03</td>\n",
       "      <td>1.379711</td>\n",
       "      <td>-1.267264</td>\n",
       "      <td>0.918310</td>\n",
       "      <td>1.414022</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>762</th>\n",
       "      <td>2021-10-01</td>\n",
       "      <td>1.271690</td>\n",
       "      <td>-1.162507</td>\n",
       "      <td>0.984188</td>\n",
       "      <td>1.964716</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>763</th>\n",
       "      <td>2021-10-02</td>\n",
       "      <td>1.271690</td>\n",
       "      <td>-1.138517</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.110120</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>764</th>\n",
       "      <td>2021-10-03</td>\n",
       "      <td>1.271690</td>\n",
       "      <td>-1.074543</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.110120</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>765 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           Date      Gold      Coal  Closing_price  Crude_Oil\n",
       "0    2017-01-03 -1.086089  1.359663       0.017401  -0.003241\n",
       "1    2017-01-04 -1.054820  1.373258       0.012031   0.074474\n",
       "2    2017-01-06 -1.027479  1.374057       0.013913   0.135477\n",
       "3    2017-01-07 -1.027479  1.374057       0.014577  -0.034160\n",
       "4    2017-01-08 -1.027479  1.374057       0.014577  -0.034160\n",
       "..          ...       ...       ...            ...        ...\n",
       "760  2021-09-02  1.378535 -1.372021       0.872459   1.472518\n",
       "761  2021-09-03  1.379711 -1.267264       0.918310   1.414022\n",
       "762  2021-10-01  1.271690 -1.162507       0.984188   1.964716\n",
       "763  2021-10-02  1.271690 -1.138517       1.000000   2.110120\n",
       "764  2021-10-03  1.271690 -1.074543       1.000000   2.110120\n",
       "\n",
       "[765 rows x 5 columns]"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "min_max_scaler = MinMaxScaler()\n",
    "df['Closing_price'] = min_max_scaler.fit_transform(df[['Closing_price']])\n",
    "standard_scaler = StandardScaler()\n",
    "df['Gold']=standard_scaler.fit_transform(df[['Gold']])\n",
    "df['Coal']=standard_scaler.fit_transform(df[['Coal']])\n",
    "df['Crude_Oil']=standard_scaler.fit_transform(df[['Crude_Oil']])\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['Sentiments']=sms"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['Lag1']=df['Closing_price'].shift(1)\n",
    "df=df.dropna().reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_features(data, window,feature_cols):\n",
    "    X, y = [], []\n",
    "    # feature_cols = ['Gold', 'Coal','Crude_Oil','Lag1','pos','neg','neu','compound']  \n",
    "    target_col = ['Closing_price']            \n",
    "    max_index = len(data) - window -1\n",
    "    for i in range(max_index):\n",
    "        feature = data.loc[i:(i+window), feature_cols].values.astype('float32')\n",
    "        target = data.loc[i+window, target_col].values.astype('float32')\n",
    "        X.append(feature)\n",
    "        y.append(target)\n",
    "    X=np.array(X)\n",
    "    y=np.array(y)    \n",
    "    return X, y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "batchsize=32\n",
    "learningrate=0.001\n",
    "window_size = 35"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Model Architure"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "class LSTM(nn.Module):\n",
    "    def __init__(self, num_classes, input_size, hidden_size, num_layers):\n",
    "        super().__init__()\n",
    "        self.num_classes = num_classes \n",
    "        self.num_layers = num_layers \n",
    "        self.input_size = input_size \n",
    "        self.hidden_size = hidden_size \n",
    "        self.lstm = nn.LSTM(input_size=input_size, hidden_size=hidden_size,\n",
    "                            num_layers=num_layers, batch_first=True, dropout=0.2)\n",
    "        self.fc = nn.Sequential(\n",
    "            nn.Linear(hidden_size, 128),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(128, num_classes)\n",
    "        )\n",
    "        \n",
    "    def forward(self, x):\n",
    "        h_0 = torch.zeros(self.num_layers, x.size(0), self.hidden_size)\n",
    "        c_0 = torch.zeros(self.num_layers, x.size(0), self.hidden_size)\n",
    "        output, (hn, _) = self.lstm(x, (h_0, c_0))\n",
    "        hn = hn.view(-1, self.hidden_size)\n",
    "        out = self.fc(hn)\n",
    "        return out\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_lstm(feature_col):\n",
    "    n_epochs = 500 \n",
    "    learning_rate = 0.001 \n",
    "\n",
    "    input_size = len(feature_col)\n",
    "    hidden_size = 2 \n",
    "    num_layers = 1 \n",
    "\n",
    "    num_classes = 1  \n",
    "\n",
    "    lstm = LSTM(num_classes, \n",
    "                input_size, \n",
    "                hidden_size, \n",
    "                num_layers)\n",
    "\n",
    "    print(lstm)            \n",
    "\n",
    "    return lstm            "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [],
   "source": [
    "def training_loop(n_epochs, lstm,train_l,val_l):\n",
    "    loss_fn = torch.nn.MSELoss()    \n",
    "    optimiser = torch.optim.Adam(lstm.parameters(), lr=0.001)\n",
    "    train_loss_a=[]\n",
    "    val_loss_a=[]\n",
    "    for epoch in range(n_epochs):\n",
    "        train_loss = 0.0\n",
    "        val_loss = 0.0\n",
    "        lstm.train()\n",
    "        for inputs, targets in train_l:\n",
    "            optimiser.zero_grad()\n",
    "            outputs = lstm(inputs)\n",
    "            loss = loss_fn(outputs, targets)\n",
    "            loss.backward()\n",
    "            optimiser.step()\n",
    "            train_loss += loss.item() * inputs.size(0)\n",
    "\n",
    "        lstm.eval()\n",
    "        with torch.no_grad():\n",
    "            for inputs, targets in val_l:\n",
    "                outputs = lstm(inputs)\n",
    "                loss = loss_fn(outputs, targets)\n",
    "                val_loss += loss.item() * inputs.size(0)\n",
    "\n",
    "        train_loss /= len(train_l.dataset)\n",
    "        val_loss /= len(val_l.dataset)\n",
    "        train_loss_a.append(train_loss)\n",
    "        val_loss_a.append(val_loss)\n",
    "\n",
    "        if epoch % 100 == 0:\n",
    "            print(\"Epoch: %d, train loss: %1.5f, val loss: %1.5f\" % (epoch, train_loss, val_loss))\n",
    "\n",
    "    return train_loss_a,val_loss_a        \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot(lstm,feature_col):\n",
    "    df_X_ss, df_y_mm = create_features(df,window_size,feature_col)\n",
    "\n",
    "    df_X_ss = torch.Tensor(df_X_ss)\n",
    "    df_y_mm = torch.Tensor(df_y_mm)\n",
    "\n",
    "\n",
    "    df_X_ss = torch.reshape(df_X_ss, (df_X_ss.shape[0], window_size+1, df_X_ss.shape[2]))\n",
    "\n",
    "    train_predict = lstm(df_X_ss)\n",
    "    data_predict = train_predict.data.numpy()\n",
    "    dataY_plot = df_y_mm.data.numpy()\n",
    "\n",
    "    data_predict = min_max_scaler.inverse_transform(data_predict) \n",
    "    dataY_plot = min_max_scaler.inverse_transform(dataY_plot)\n",
    "    true, preds = [], []\n",
    "    for i in range(len(dataY_plot)):\n",
    "        true.append(dataY_plot[i][0])\n",
    "    for i in range(len(data_predict)):\n",
    "        preds.append(data_predict[i][0])\n",
    "\n",
    "    rmse = np.sqrt(mean_squared_error(true, preds))\n",
    "    print(\"Root Mean Squared Error (RMSE):\", rmse)    \n",
    "    plt.figure(figsize=(10,6)) \n",
    "    plt.axvline(x=len(df)*0.90  , c='r', linestyle='--') \n",
    "\n",
    "    plt.plot(true, label='Actual Data') \n",
    "    plt.plot(preds, label='Predicted Data') \n",
    "    plt.title('Stock Prediction')\n",
    "    plt.legend()\n",
    "\n",
    "    plt.show() "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Data Pre-Processing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [],
   "source": [
    "def making_dataloader(x, y):\n",
    "    X_tensor = torch.tensor(x)\n",
    "    Y_tensor = torch.tensor(y)\n",
    "    dataset = TensorDataset(X_tensor, Y_tensor)\n",
    "    dataloader = DataLoader(dataset, batch_size=batchsize, shuffle=False)\n",
    "    return dataloader"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training Size :- 687\n",
      "Validation Size :- 38\n",
      "Testing Size :- 39\n"
     ]
    }
   ],
   "source": [
    "\n",
    "train_size = int(len(df) * 0.90)\n",
    "val_size = int((len(df) - train_size) * 0.50)\n",
    "train = df[:train_size].reset_index(drop=True)\n",
    "val= df[train_size:train_size + val_size].reset_index(drop=True)\n",
    "test= df[train_size + val_size:].reset_index(drop=True)\n",
    "\n",
    "print(f\"Training Size :- {len(train)}\")\n",
    "print(f\"Validation Size :- {len(val)}\")\n",
    "print(f\"Testing Size :- {len(test)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [],
   "source": [
    "def Data(feature_col):\n",
    "    X_train,y_train=create_features(train,window_size,feature_col)\n",
    "    X_val,y_val=create_features(val,window_size,feature_col)\n",
    "    X_test,y_test=create_features(test,window_size,feature_col)\n",
    "\n",
    "    print(f\"Shape of X Train:- {X_train.shape}\")\n",
    "    print(f\"Shape of Y Train:- {y_train.shape}\")\n",
    "\n",
    "    train_l=making_dataloader(X_train,y_train)\n",
    "    val_l=making_dataloader(X_val,y_val)\n",
    "    test_l=making_dataloader(X_test,y_test)\n",
    "\n",
    "    return train_l,val_l,test_l\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Date', 'Gold', 'Coal', 'Closing_price', 'Crude_Oil', 'Sentiments',\n",
       "       'Lag1'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [],
   "source": [
    "feature_cols = ['Gold', 'Coal','Crude_Oil','Lag1','Sentiments']  \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of X Train:- (651, 36, 5)\n",
      "Shape of Y Train:- (651, 1)\n",
      "LSTM(\n",
      "  (lstm): LSTM(5, 2, batch_first=True, dropout=0.2)\n",
      "  (fc): Sequential(\n",
      "    (0): Linear(in_features=2, out_features=128, bias=True)\n",
      "    (1): ReLU()\n",
      "    (2): Linear(in_features=128, out_features=1, bias=True)\n",
      "  )\n",
      ")\n",
      "Epoch: 0, train loss: 0.01625, val loss: 0.02451\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 100, train loss: 0.00320, val loss: 0.01231\n",
      "Epoch: 200, train loss: 0.00239, val loss: 0.00875\n",
      "Epoch: 300, train loss: 0.00170, val loss: 0.00719\n",
      "Epoch: 400, train loss: 0.00268, val loss: 0.00580\n",
      "Epoch: 500, train loss: 0.00202, val loss: 0.00423\n",
      "Epoch: 600, train loss: 0.00087, val loss: 0.00364\n",
      "Epoch: 700, train loss: 0.00107, val loss: 0.00262\n",
      "Epoch: 800, train loss: 0.00032, val loss: 0.00350\n",
      "Epoch: 900, train loss: 0.00029, val loss: 0.00289\n"
     ]
    }
   ],
   "source": [
    "train_l,val_l,test_l=Data(feature_cols)\n",
    "lstm=get_lstm(feature_cols)\n",
    "train_loss,val_loss=training_loop(1000,lstm,train_l,val_l)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Root Mean Squared Error (RMSE): 65.24101\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot(lstm,feature_cols)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x1e73422be60>"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(train_loss, label='Training Loss') \n",
    "plt.plot(val_loss, label='Validation Loss') \n",
    "plt.legend()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}