PyTorch implementation for RF/XGB

openkinome / kinoml

Structure-informed machine learning for kinase modeling

https://openkinome.org/kinoml/

MIT License

51 stars 21 forks source link

PyTorch implementation for RF/XGB #17

Closed jaimergp closed 3 years ago

jaimergp commented 4 years ago

Some resources:

Write the model in sklearn / xgboost and convert to PyTorch: https://towardsdatascience.com/transform-your-ml-model-to-pytorch-with-hummingbird-da49665497e7 (https://github.com/microsoft/hummingbird)
Use a native PyTorch tree implementation: https://github.com/ValentinFigue/Sklearn_PyTorch

jaimergp commented 4 years ago

I have imported sklearn_pytorch in our symbols for now.

t-kimber commented 4 years ago

I created a new PR (see #25) which deals with the XBGoost package (not sure the native PyTorch tree implementation linked above is flexible enough for the observation models).

t-kimber commented 3 years ago

Closing issue since focusing on deep learning models.