Data Science Experimentation Platform

A comprehensive web-based platform for data science experimentation with educational guidance and interactive visualizations.

Features

Core Components

• Dataset Selection: Preloaded datasets (Iris, Titanic, Boston Housing) + custom CSV upload
• Phase-Specific Sandboxes: Guided workflows for each data science phase
• Educational Guidance: Explanations, best practices, and comparative analysis
• Interactive Visualizations: Real-time feedback using Chart.js
• Comparative Analysis: Before/after comparisons for all transformations

Data Science Phases

1. Data Wrangling: Imputation, outlier handling, missing value analysis
2. Data Manipulation: Pivot, melt, filtering, aggregation
3. Feature Engineering: Polynomial features, binning, encoding
4. Feature Selection: RFE, SelectKBest, variance threshold
5. Scaling/Normalization: Min-Max, Standard, Robust scaling
6. Model Building: Multiple algorithm choices with explanations
7. Hyperparameter Tuning: Grid search, random search, Bayesian optimization
8. Deployment: REST API, batch processing, feature flags

Technology Stack

Frontend

• React.js with TypeScript
• Chart.js for visualizations
• Material-UI for components
• React Router for navigation

Backend

• FastAPI with Python
• Scikit-learn for ML operations
• Pandas for data manipulation
• PostgreSQL for metadata storage

Infrastructure

• Docker for containerization
• Kubernetes for orchestration
• Auth0 for authentication
• S3 for ephemeral storage

Project Structure

DataScienceSandbox/
├── frontend/                 # React frontend application
├── backend/                  # FastAPI backend service
├── docker/                   # Docker configurations
├── k8s/                      # Kubernetes manifests
├── docs/                     # Documentation
└── datasets/                 # Sample datasets

Getting Started

Prerequisites

• Node.js 18+
• Python 3.9+
• Docker
• PostgreSQL

Installation

1. Clone the repository

   git clone <repository-url>
   cd DataScienceSandbox

2. Start the backend

   cd backend
   pip install -r requirements.txt
   uvicorn main:app --reload

3. Start the frontend

   cd frontend
   npm install
   npm start

4. Access the application

   • Frontend: http://localhost:3000
   • Backend API: http://localhost:8000
   • API Docs: http://localhost:8000/docs

Usage

1. Select a Dataset: Choose from preloaded datasets or upload your own CSV
2. Navigate Phases: Use the phase navigation to move through the workflow
3. Choose Methods: Select processing options with educational guidance
4. View Results: See real-time visualizations and comparative analysis
5. Learn: Read explanations and best practices for each method

Contributing

Please read CONTRIBUTING.md for details on our code of conduct and the process for submitting pull requests.

License

This project is licensed under the MIT License - see the LICENSE file for details.