RNN_2.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "c268b8f3",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import LSTM, Dense\n",
    "from tensorflow.keras.models import load_model\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import confusion_matrix, classification_report\n",
    "from sklearn.metrics import roc_curve, roc_auc_score, precision_recall_curve, average_precision_score, mean_squared_error, mean_absolute_error, r2_score\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "2603f416",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load the dataset\n",
    "df = pd.read_csv(\"all_stocks_5yr.csv\")\n",
    "\n",
    "# Extract the relevant features\n",
    "# Create a new column for the target and initialize it with 0\n",
    "df['target'] = 0\n",
    "\n",
    "# Define a function to compare close values and assign the target value\n",
    "def compare_close_values(group):\n",
    "    group['target'] = (group['close'].shift(-1) > group['close']).astype(int)\n",
    "    return group\n",
    "\n",
    "# Apply the function to each group within the DataFrame\n",
    "df = df.groupby('Name', group_keys=False).apply(compare_close_values)\n",
    "\n",
    "df.dropna(inplace=True)\n",
    "\n",
    "# Reset the index of the DataFrame\n",
    "df = df.reset_index(drop=True)\n",
    "\n",
    "# Print the resulting dataset with the target column\n",
    "features = df[['open', 'high', 'low', 'close', 'volume']]\n",
    "target = df['target']\n",
    "\n",
    "# Normalize the features\n",
    "scaler = MinMaxScaler(feature_range=(0, 1))\n",
    "scaled_features = scaler.fit_transform(features)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "e0d27a3a",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(scaled_features, target, test_size=0.2, shuffle=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "a4d7ffff",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Reshape the input data for LSTM\n",
    "X_train = np.reshape(X_train, (X_train.shape[0], 1, X_train.shape[1]))\n",
    "X_test = np.reshape(X_test, (X_test.shape[0], 1, X_test.shape[1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "327b86e2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "15476/15476 [==============================] - 20s 1ms/step - loss: 0.6924 - accuracy: 0.5204\n",
      "Epoch 2/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 3/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 4/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 5/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 6/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 7/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 8/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 9/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 10/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 11/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 12/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 13/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 14/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 15/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 16/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 17/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 18/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 19/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 20/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 21/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 22/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 23/100\n",
      "15476/15476 [==============================] - 26s 2ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 24/100\n",
      "15476/15476 [==============================] - 46s 3ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 25/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 26/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 27/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 28/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 29/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 30/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 31/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 32/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 33/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 34/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 35/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 36/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 37/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 38/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 39/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 40/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 41/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 42/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 43/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 44/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 45/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 46/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 47/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 48/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 49/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 50/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 51/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 52/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 53/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 54/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 55/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 56/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 57/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 58/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 59/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 60/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 61/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 62/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 63/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 64/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 65/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 66/100\n",
      "15476/15476 [==============================] - 20s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 67/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 68/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 69/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 70/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 71/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 72/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 73/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 74/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 75/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 76/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 77/100\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 78/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 79/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 80/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 81/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 82/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 83/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 84/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 85/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 86/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 87/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 88/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 89/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 90/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 91/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 92/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 93/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 94/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 95/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 96/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 97/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 98/100\n",
      "15476/15476 [==============================] - 18s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 99/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n",
      "Epoch 100/100\n",
      "15476/15476 [==============================] - 19s 1ms/step - loss: 0.6923 - accuracy: 0.5204\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x7fa7e032cc40>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = Sequential()\n",
    "model.add(LSTM(units=50, input_shape=(1, X_train.shape[2])))\n",
    "model.add(Dense(units=1, activation='sigmoid'))\n",
    "model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])\n",
    "\n",
    "model.fit(X_train, y_train, epochs=100, batch_size=32)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "d7f9f29f",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-06-06 16:09:17.624062: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  SSE4.1 SSE4.2\n",
      "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n"
     ]
    }
   ],
   "source": [
    "# model = load_model('RNN_4feature_1_model.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "071f50c6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3869/3869 [==============================] - 3s 673us/step - loss: 0.6921 - accuracy: 0.5228\n",
      "Test Loss: 0.6921271085739136\n",
      "Test Accuracy: 0.5228260159492493\n"
     ]
    }
   ],
   "source": [
    "loss, accuracy = model.evaluate(X_test, y_test)\n",
    "print(\"Test Loss:\", loss)\n",
    "print(\"Test Accuracy:\", accuracy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "37cfd9a8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# model.save(\"RNN_4feature_1_model.h5\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "fba17244",
   "metadata": {},
   "outputs": [],
   "source": [
    "loss, accuracy = Trained_model_1.evaluate(X_test, y_test)\n",
    "print(\"Test Loss:\", loss)\n",
    "print(\"Test Accuracy:\", accuracy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "c76dd01b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3869/3869 [==============================] - 2s 597us/step\n",
      "Accuracy: 0.5228260342794372\n",
      "Confusion Matrix:\n",
      "[[    0 59077]\n",
      " [    0 64729]]\n",
      "Classification Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.00      0.00      0.00     59077\n",
      "           1       0.52      1.00      0.69     64729\n",
      "\n",
      "    accuracy                           0.52    123806\n",
      "   macro avg       0.26      0.50      0.34    123806\n",
      "weighted avg       0.27      0.52      0.36    123806\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/behdadfeilizadeh/anaconda3/lib/python3.10/site-packages/sklearn/metrics/_classification.py:1344: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/Users/behdadfeilizadeh/anaconda3/lib/python3.10/site-packages/sklearn/metrics/_classification.py:1344: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/Users/behdadfeilizadeh/anaconda3/lib/python3.10/site-packages/sklearn/metrics/_classification.py:1344: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "\n",
    "#Obtain predicted probabilities for positive class (class 1)\n",
    "# X_test_4_feat = X_test[:, :, :4]\n",
    "y_prob = model.predict(X_test)[:, 0]\n",
    "\n",
    "# Convert probabilities to binary predictions (0 or 1)\n",
    "y_pred = np.where(y_prob > 0.5, 1, 0)\n",
    "\n",
    "# Accuracy\n",
    "accuracy = accuracy_score(y_test, y_pred)\n",
    "print(\"Accuracy:\", accuracy)\n",
    "\n",
    "# Confusion matrix\n",
    "cm = confusion_matrix(y_test, y_pred)\n",
    "print(\"Confusion Matrix:\")\n",
    "print(cm)\n",
    "\n",
    "# Classification report\n",
    "cr = classification_report(y_test, y_pred)\n",
    "print(\"Classification Report:\")\n",
    "print(cr)\n",
    "\n",
    "# ROC Curve\n",
    "fpr, tpr, thresholds = roc_curve(y_test, y_prob)\n",
    "roc_auc = roc_auc_score(y_test, y_prob)\n",
    "\n",
    "plt.figure()\n",
    "plt.plot(fpr, tpr, color='blue', lw=2, label='ROC curve (AUC = %0.2f)' % roc_auc)\n",
    "plt.plot([0, 1], [0, 1], color='red', lw=2, linestyle='--')\n",
    "plt.xlim([0.0, 1.0])\n",
    "plt.ylim([0.0, 1.05])\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('Receiver Operating Characteristic')\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.show()\n",
    "\n",
    "# Precision-Recall Curve\n",
    "precision, recall, thresholds = precision_recall_curve(y_test, y_prob)\n",
    "average_precision = average_precision_score(y_test, y_prob)\n",
    "\n",
    "plt.figure()\n",
    "plt.plot(recall, precision, color='blue', lw=2, label='Precision-Recall curve (AP = %0.2f)' % average_precision)\n",
    "plt.xlim([0.0, 1.0])\n",
    "plt.ylim([0.0, 1.05])\n",
    "plt.xlabel('Recall')\n",
    "plt.ylabel('Precision')\n",
    "plt.title('Precision-Recall Curve')\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "94ebb830",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE:  0.24949042053528747\n",
      "R2:  -4.5889142938460026e-05\n",
      "RMSE: 0.49949016059907275\n",
      "MAE: 0.49904181018088734\n"
     ]
    }
   ],
   "source": [
    "mse = mean_squared_error(y_test, y_prob)\n",
    "rmse = mean_squared_error(y_test, y_prob, squared=False)\n",
    "r2 = r2_score(y_test, y_prob)\n",
    "mae = mean_absolute_error(y_test, y_prob)\n",
    "print(\"MSE: \", mse)\n",
    "print(\"R2: \", r2)\n",
    "print(\"RMSE:\", rmse)\n",
    "print(\"MAE:\", mae)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5ce749ad",
   "metadata": {},
   "source": [
    "## Parameter Tuning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "87671120",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.wrappers.scikit_learn import KerasClassifier\n",
    "from sklearn.model_selection import GridSearchCV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a764acf4",
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_model(units=50, activation='relu', optimizer='adam', loss='binary_crossentropy'):\n",
    "    model = Sequential()\n",
    "    model.add(LSTM(units=units, input_shape=(1, X_train.shape[2])))\n",
    "    model.add(Dense(units=1, activation=activation))\n",
    "    model.compile(optimizer=optimizer, loss=loss, metrics=['accuracy'])\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5d8278a2",
   "metadata": {},
   "outputs": [],
   "source": [
    "param_grid = {\n",
    "    'units': [50, 100, 200],  # Number of LSTM units\n",
    "    'batch_size': [32, 64, 128],  # Batch size\n",
    "    'epochs': [25, 50, 100],  # Number of epochs\n",
    "    'activation': ['relu', 'sigmoid', 'tanh'],  # Activation functions to try\n",
    "    'optimizer': ['SGD', 'RMSprop', 'Adam'],  # Optimizers to try\n",
    "    'loss': ['binary_crossentropy', 'mean_squared_error', 'hinge']\n",
    "}\n",
    "\n",
    "# Create the model\n",
    "model = KerasClassifier(build_fn=create_model)\n",
    "\n",
    "# Perform grid search\n",
    "grid = GridSearchCV(estimator=model, param_grid=param_grid, cv=3, scoring='accuracy', verbose=1)\n",
    "grid_result = grid.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4d31ce60",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get the best hyperparameters\n",
    "best_params = grid_result.best_params_\n",
    "print(\"Best Hyperparameters:\", best_params)\n",
    "\n",
    "# Get the best model\n",
    "best_model = grid_result.best_estimator_.model\n",
    "print(\"Best Model Architecture:\")\n",
    "best_model.summary()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}