MLBox

[se-teryoshkin]

https://github.com/AxeldeRomblay/MLBox

1. GitHub: Stars - 1.3k, Forks - 261, Contributors - 5

2. Releases: Last Release - 0.8.1 (26th August 2019), First Release - (10th July 2017)

3. Language: Python 3.5 - 3.7

4. Data Types: Structured data only

5. Execution backends: joblib

6. Distributed framework: None

7. Distributed algorithms libraries: None

8. AutoML pipeline elements:

-- Data Reading --

• Datasets merging
• Task detection
• Test/train split

-- Preprocessing --

• Data cleaning
• Feature encoding

-- Optimisation --

• Feature engineering
• Feature selection
• Model selection
• Hyper-parameter optimisation
• Ensembling

-- Prediction --

• Model interpretation
• Leak detection

9. Elements of pipeline are covered in distributed mode: None

10. Is it possible to mix distributed and non-distributed backends?: No

11. How is the pipeline converted to computing applications?: the entire pipeline is a single local application

12. Approach to ML models selection computing: Local training

13. Data processing steps pruning: None

14. Resource managers: None

15. Hyperparameter tuner: hyperopt

16. Task types: Classification, Regression

17. Tests: local-mode tests

18. Comments: --