app.py

# Import necessary libraries
import pandas as pd
import yfinance as yf
from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestRegressor
from sklearn.metrics import mean_squared_error
import streamlit as st
from datetime import datetime, timedelta, timezone

# Function to fetch historical stock data
def fetch_stock_data(symbol, start_date, end_date):
    data = yf.download(symbol, start=start_date, end=end_date)
    return data

# Function to create features for the model
def create_features(data):
    data['Date'] = data.index
    data['Year'] = data['Date'].dt.year
    data['Month'] = data['Date'].dt.month
    data['Day'] = data['Date'].dt.day
    data['Hour'] = data['Date'].dt.hour
    data['Minute'] = data['Date'].dt.minute

    return data

# Function to train a machine learning model
def train_model(data):
    features = ['Year', 'Month', 'Day', 'Hour', 'Minute']
    target = 'Close'

    X = data[features]
    y = data[target]

    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

    model = RandomForestRegressor()
    model.fit(X_train, y_train)

    # Evaluate the model
    predictions = model.predict(X_test)
    mse = mean_squared_error(y_test, predictions)
    st.write(f"Mean Squared Error: {mse}")

    return model

# Streamlit UI
def main():
    st.title("Stock Price Prediction Tool")

    symbol = st.text_input("Enter Stock Symbol (e.g., AAPL):")
    start_date = st.date_input("Select Start Date:")
    end_date = st.date_input("Select End Date:")

    if st.button("Predict Stock Price"):
        # Fetch stock data
        stock_data = fetch_stock_data(symbol, start_date, end_date)

        # Create features
        stock_data = create_features(stock_data)

        # Train the model
        model = train_model(stock_data)

        # Predict the stock price for a specific date (e.g., the last date in the dataset)
        prediction_date = stock_data['Date'].iloc[-1]
        prediction_features = [[
            prediction_date.year,
            prediction_date.month,
            prediction_date.day,
            prediction_date.hour,
            prediction_date.minute
        ]]
        predicted_price = model.predict(prediction_features)[0]

        st.subheader(f"Predicted Stock Price on {prediction_date} (UTC):")
        st.write(f"${predicted_price:.2f}")

# Run the Streamlit app
if __name__ == '__main__':
    main()