incremental CART decision tree, based on the hoeffding tree i.e. very fast decision tree (VFDT), which is proposed in this paper "Mining High-Speed Data Streams" by Domingos & Hulten (2000). And a newly extended model "Extremely Fast Decision Tree" (EFDT) by Manapragada, Webb & Salehi (2018). Added new implementation of Random Forest
Python 100.00%
incremental_decision_tree-cart-random_forest's People
Hi, since this project has no license I'm aware the copyright laws apply and it is not legally permitted for someone to redistribute or modify the contents of this repository. Did you purposely create it without license to have the default GitHub policy implicit or do you allow the re-use and modification of your code for other purposes?
is it fast what I got for this file vfdt.py for windows 10 laptop
do you have really big data sets to run?
Total data size: 5000
Training size: 4500
Test set size: 500
Training set: 1000, ACCURACY: 0.7240
Training set: 2000, ACCURACY: 0.7500
Training set: 4500, ACCURACY: 0.7680
--- Running time: 0.936634 seconds ---