Stock Price Predection 📈

A machine learning project that predicts whether a stock's price will go UP or DOWN the next day based on historical data. The project includes data processing, feature engineering, model training using Logistic Regression, and a Streamlit web application for interactive predictions.

Project Structure

• app.py: Streamlit application for interactive prediction.
• predict.py: Script for making predictions in the terminal.
• test.py: Script for data loading, feature engineering, model training, evaluation, saving the model, and plotting.
• stock_data.csv: Historical stock data used for training.
• requirements.txt: Python dependencies.

Installation & Setup

1. Clone the repository:

   git clone <repository_url>
   cd "Stock Price Predection"

2. Install dependencies:

   pip install -r requirements.txt

3. Train the Model: First, you need to train the model so that app.py and predict.py can load the saved models. Run the test script to generate the models:

   python test.py

   Note: This will generate scaler.pkl and stock_trend_model.pkl. Move them to a models/ directory for app.py to use them.

4. Run the Streamlit App:

   streamlit run app.py

Model Pipeline

The project performs feature engineering (Daily Return, Moving Averages, Volatility, etc.), scales features, and uses Logistic Regression to perform binary classification on the target (1 for UP, 0 for DOWN).

Author

Pavan