I’m slowly moving all my personal projects and ML hacking over to Julia, so I’m trying to find substitutes for packages that I’m used to. On in particular I’m heavily dependent on is Python’s scikit-learn, and in particular I’m dependent on the calculations for things like precision, recall, f-score, etc. for both regression and classification problems.
Is there a JuliaML / JuliaStats / Flux implementation of these that I should be using?
My package has a way of handling this. Don’t know if I’ve tested all the validation metrics, but yea, it takes a one hot or one cold encoding and returns a big dictionary of metrics,
https://caseykneale.github.io/ChemometricsTools.jl/dev/man/classMetrics/
https://caseykneale.github.io/ChemometricsTools.jl/dev/man/regressMetrics/
Example usages:
https://github.com/caseykneale/ChemometricsTools.jl/blob/master/shootouts/ClassificationShootout.jl
https://github.com/caseykneale/ChemometricsTools.jl/blob/master/shootouts/RegressionShootout.jl
Guess I should have asked, are you working in computer vision? If so those are implemented differently. I could help with that if you need it.
Looks like they’ve come quite a long ways. I remember when I made my package there was nothing.