Welcome to GossipCat’s documentation!

GossipCat is a data science project framework that simplifies the process of machine learning from data cleaning, simple feature engineering, machine learning algorithm comparison, hyper parameter tuning, model evaluation, to results output. It is designed to be efficient with following features:

1. Agile machine learning framework: designed with a lean start and continuing improvement.

2. Pipeline data preprocessing: high cohesion, low coupling.

3. Algorithms comparison: provides a overview of multiple machine learning algorithms comparison.

4. Diverse model evaluation: makes the evaluation visible and with business sense.

5. Architectural thinking: not only data science but also machine learning engineering.

Overview

• Installation Guide
  • The General Approach
  • The Latest Version
  • For Users in China
• GossipCat Tutorials
  • Data Science Projects Basics
  • Data Science Experiment
  • Model Development
• GossipCat API
  • Data Science Project Basics
  • Data Science Experiment
  • Model Development
  • Graph Data Science
• Appendix
  • Data Science Project Checklist
  • Visualizing a SQL query
• Contribute to GossipCat

Indices

• Index

• Search Page