Grand Line: Time-Series Models Platform

Welcome to Grand Line, a platform designed for time-series modeling using LSTM models. This project combines the power of machine learning with a user-friendly interface to facilitate automatic model creation and deployment.

Features

• LSTM Models: Implemented using Keras and TensorFlow for accurate time-series forecasting.

• Automatic Model Creation and Deployment: Backend developed with Python Flask to automate the process of training models and deploying them seamlessly.

• Responsive Frontend: Built with React TSX to provide an intuitive user interface. Users can upload datasets, initiate model training, and view forecasts effortlessly.

• Database Management: MySQL is utilized for efficient data storage and management.

Components

Backend (Python Flask)

The backend handles the core functionalities of the platform:

• Model Training: Automatically trains LSTM models based on user-uploaded datasets.
• Deployment: Provides endpoints for deploying trained models to generate forecasts.
• API Integration: Supports integration with the frontend for seamless data flow and model execution.

Frontend (React TSX)

The frontend offers a responsive interface accessible via web browsers:

• Dataset Upload: Allows users to upload time-series datasets in csv format.
• Model Training: Initiates training of LSTM models with just a few clicks.
• Forecast Visualization: Displays forecasted results in an easy-to-understand format.
• User Management: User authentication and session management functionalities.

Database (MySQL)

Contributors

• Jan Haslik

Feel free to contribute to this project by submitting pull requests or reporting issues.