FEATURE: Implement Machine Learning Models for Financial Trend and St…

…ock Price Prediction

Fixes ayush-that#3209

Finveda/models/stock_price_prediction/__init__.py

@@ -0,0 +1,10 @@
+# Initialize the stock price prediction package
+
+# Import necessary functions for ease of access
+from .data_preprocessing import fetch_stock_data, preprocess_data
+from .lstm_model import build_lstm_model
+from .train_model import train_model
+from .evaluate_model import evaluate_model
+from .flask_api import app  # Importing Flask app for API exposure
+
+# You can also add other modules or utilities as needed

Finveda/models/stock_price_prediction/data_preprocessing.py

@@ -0,0 +1,33 @@
+import pandas as pd
+import numpy as np
+from sklearn.preprocessing import MinMaxScaler
+import yfinance as yf
+
+# Fetching historical stock data
+def fetch_stock_data(ticker, start_date='2010-01-01', end_date='2024-01-01'):
+    data = yf.download(ticker, start=start_date, end=end_date)
+    return data
+
+# Feature engineering and preprocessing
+def preprocess_data(df):
+    # Keep only relevant columns: 'Date', 'Close' (you can add more features like 'Open', 'Volume' etc.)
+    df = df[['Close']]
+
+    # Normalize the data using MinMaxScaler
+    scaler = MinMaxScaler(feature_range=(0, 1))
+    df_scaled = scaler.fit_transform(df)
+
+    # Create a dataset with 'X' as previous n-day stock prices and 'y' as next day price
+    def create_dataset(data, time_step=60):
+        X, y = [], []
+        for i in range(len(data) - time_step - 1):
+            X.append(data[i:(i + time_step), 0])
+            y.append(data[i + time_step, 0])
+        return np.array(X), np.array(y)
+
+    X, y = create_dataset(df_scaled)
+
+    # Reshaping X for LSTM (samples, time steps, features)
+    X = X.reshape(X.shape[0], X.shape[1], 1)
+
+    return X, y, scaler

Finveda/models/stock_price_prediction/evaluate_model.py

@@ -0,0 +1,12 @@
+# Evaluate the model using test data
+def evaluate_model(model, X_test, y_test, scaler):
+    predicted_stock_price = model.predict(X_test)
+
+    # Rescale the predictions and actual values back to the original scale
+    predicted_stock_price = scaler.inverse_transform(predicted_stock_price)
+    y_test_rescaled = scaler.inverse_transform(y_test.reshape(-1, 1))
+
+    # Calculate the Mean Squared Error (MSE)
+    mse = np.mean(np.square(predicted_stock_price - y_test_rescaled))
+
+    return mse, predicted_stock_price, y_test_rescaled

Finveda/models/stock_price_prediction/flask_api.py

@@ -0,0 +1,31 @@
+from flask import Flask, request, jsonify
+from models.stock_price_prediction.data_preprocessing import fetch_stock_data, preprocess_data
+from models.stock_price_prediction.lstm_model import build_lstm_model
+
+# Initialize Flask app
+app = Flask(__name__)
+
+# Load the trained model
+model = build_lstm_model((60, 1))  # Ensure the model is built
+model.load_weights('models/stock_price_prediction/model_weights.h5')  # Load saved model weights
+
+# API endpoint for stock price prediction
+@app.route('/predict', methods=['POST'])
+def predict_stock():
+    # Extract data from request (e.g., stock ticker symbol)
+    ticker = request.json['ticker']
+
+    # Fetch and preprocess the stock data
+    data = fetch_stock_data(ticker)
+    X, y, scaler = preprocess_data(data)
+
+    # Predict the stock price using the LSTM model
+    prediction = model.predict(X[-1].reshape(1, X.shape[1], 1))
+    prediction = scaler.inverse_transform(prediction)
+
+    # Return the prediction as a JSON response
+    return jsonify({'predicted_stock_price': prediction[0][0]})
+
+# Run the app
+if __name__ == "__main__":
+    app.run(debug=True)

Finveda/models/stock_price_prediction/lstm_model.py

@@ -0,0 +1,15 @@
+import tensorflow as tf
+from tensorflow.keras.models import Sequential
+from tensorflow.keras.layers import LSTM, Dense, Dropout
+
+# Build and compile the LSTM model
+def build_lstm_model(input_shape):
+    model = Sequential()
+    model.add(LSTM(units=50, return_sequences=True, input_shape=input_shape))
+    model.add(Dropout(0.2))
+    model.add(LSTM(units=50, return_sequences=False))
+    model.add(Dropout(0.2))
+    model.add(Dense(units=1))  # Output layer for stock price prediction
+
+    model.compile(optimizer='adam', loss='mean_squared_error')
+    return model

Finveda/models/stock_price_prediction/train_model.py

@@ -0,0 +1,4 @@
+# Train the LSTM model
+def train_model(model, X_train, y_train, epochs=10, batch_size=32):
+    model.fit(X_train, y_train, epochs=epochs, batch_size=batch_size)
+    return model

Finveda/static/JS/stock_predict.js

@@ -0,0 +1,21 @@
+// Function to call the Flask API and get stock price prediction
+async function getStockPrediction(ticker) {
+    const response = await fetch('http://localhost:5000/predict', {
+        method: 'POST',
+        headers: {
+            'Content-Type': 'application/json',
+        },
+        body: JSON.stringify({ ticker: ticker })
+    });
+
+    const data = await response.json();
+    console.log('Predicted Stock Price:', data.predicted_stock_price);
+    // Display the predicted stock price on the website
+    document.getElementById('stockPrediction').innerText = `Predicted Stock Price: $${data.predicted_stock_price}`;
+}
+
+// Example usage: Trigger prediction when the user enters a stock symbol
+document.getElementById('predictButton').addEventListener('click', () => {
+    const ticker = document.getElementById('stockTicker').value;
+    getStockPrediction(ticker);
+});