ScikitLearn: how to import experimental estimator Histogram-Based Gradient Boosting?

How can I use the ScikitLearn’s experimental estimator “Histogram-Based Gradient Boosting” in Julia?

According to the ScikitLearn Documentation, first from experimental one has to import “enable_hist_gradient_boosting” and then “HistGradientBoostingClassifier” from the ensemble module (1.11. Ensemble methods — scikit-learn 1.1.2 documentation). This doesn’t work with Julia’s ScikitLearn “@sk_import” command. Error Message: “KeyError … not found” for both “enable_hist_gradient_boosting” and “HistGradientBoostingClassifier”

I’m using Julia 1.3.1 and ScikitLearn.jl 0.6.2, PyCall 1.92

This isn’t a direct answer to your question, but there is a pure Julia implementation of histogram-based gradient boosting that you might be interested in:

https://github.com/Evovest/EvoTrees.jl

3 Likes

Could you post a full REPL example (i.e. what did you type, what was the output / error message?)