Library for online extreme-multi-label learning
Authors : Nishit Asnani, Govind Gopakumar
Affiliation : Indian Institute of Technology, Kanpur
Usage :
Use testall.py in the Tests/ folder.
Implemented so far :
- Normal Random forests
- Alternating method to optimize w's, deltas
- UCB selection of features at every node
- UCB selection of features at every level
- Thomspon sampling selection of features at every level
All datasets are taken from publicly available resources. The ones in use are: