Stock Price Prediction App

This project is a simple web application that predicts the next day's closing stock price using a trained LSTM model. The app fetches stock price data, performs feature engineering, scales the features, and then uses the trained model to predict the stock price.

Table of Contents

• Getting Started
  • Prerequisites
  • Installation
• Usage
• Data
• Model
• Endpoints

Getting Started

Prerequisites

• Python 3.x
• Flask
• Pandas
• NumPy
• Scikit-Learn
• Keras

Installation

1. Clone the repository:

   git clone https://github.com/anuj-thakkar/Stock-Market
   cd stock-prediction-app
   
   

Usage

Run the Flask app:

  python app.py

Open your web browser and go to http://localhost:5000. The home page provides a description of the project, and the prediction page displays the latest stock price data along with the predicted closing price for the next day.

Data

The app fetches stock price data using the Yahoo Finance API. The data_loader.py script handles data fetching, updating, preprocessing, and feature engineering. The processed data is saved in the data/clean directory.

Model

The LSTM model used for prediction is trained using historical stock price data. The training process is performed in the train_model.py script, and the trained model is saved as trained_lstm_model.h5.

Endpoints

/: Displays the home page with project description.

/prediction: Displays the latest stock price data and the predicted closing price for the next day.