/PreProd-Corp---Buildathon-2024---Gyatt-Maximus

Welcome to MaximusML! This application provides a seamless experience for users to upload datasets, train models using various machine learning algorithms, and store the models in AWS S3 for future use.

Primary LanguagePythonMIT LicenseMIT

MaximusML

Welcome to MaximusML! This application provides a seamless experience for users to upload datasets, train models using various machine learning algorithms, and store the models in AWS S3 for future use.

Features

  • User Authentication: Secure login and registration system to manage user access.
  • Dataset Upload: Simple and intuitive interface for uploading datasets.
  • Model Training: Automatically train models using five different algorithms:
    • Support Vector Machine (SVM)
    • Random Forest (RF)
    • AdaBoost (ADA)
    • Gradient Boosting Regressor (GBR)
    • Multi-Layer Perceptron (MLP)
    • Decision Tree (DT)
  • Download Models: Download trained models from AWS S3 for easy deployment.

Installation

Prerequisites

  • Python 3.10+
  • AWS Account with S3 access
  • Required Python packages (listed in requirements.txt)

Steps

  1. Clone the Repository

    git clone https://github.com/arpy8/PreProd-Corp---Buildathon-2024---Gyatt-Maximus.git
cd PreProd-Corp---Buildathon-2024---Gyatt-Maximus

  2. Install Dependencies

    uv pip install -r requirements.txt

  3. Set Up Environment Variables

    Create a .env file in the root directory with the following variables:

    AWS_DEFAULT_REGION=your_aws_default_region
AWS_ACCESS_KEY_ID=your_aws_access_key_id
AWS_SECRET_ACCESS_KEY=your_aws_secret_access_key
S3_BUCKET_NAME=your_s3_bucket_name

  4. Start the Application

    streamlit run main.py

Usage

  1. Register / Log In

    Navigate to the registration or login page to create an account or log in.

  2. Upload Dataset

    Once logged in, go to the upload page and upload your dataset (CSV/XLSX format).

  3. Train Models

    After the dataset is uploaded, navigate to the model training page. Select the algorithms you want to use (SVM, RF, ADA, GBR, MLP, DT) and click "Train".

  4. Store Models

    The trained models will automatically be stored in your configured AWS S3 bucket.

Project Structure

├── static/               # Public files (Data, Models, etc.)
│   ├── data/             # Sample datasets
│   └── images/           # Images for the web interface
├── pycaret/              # Slightly tweaked version of pycaret
├── main.py               # Main application file
├── user_auth/            # User authentication and registration
├── utils.py              # Utility functions
├── train_model.py        # Model training and tuning
├── Dockerfile            # Docker file for deployment
├── requirements.txt      # List of required Python packages
└── sections.py           # Contains the three sections for the web interface

Contributing

We welcome contributions! Please fork the repository and submit pull requests for any enhancements or bug fixes.

License

This project is licensed under the MIT License.

Team Members

Contact

For any questions or suggestions, please contact [arpitsengar99@gmail.com].