Open srfrew opened 10 months ago
I agree, we could definitely improve that. Along with the confusion matrix too. But I don't think its a big priority right now. Maybe we can tweak it after we submit this milestone?
For sure, we can take a look after this milestone
I realized we're using hard coded values for
RandomizedSearchCV
with an explicit list of values instead of a distribution (likeloguniform()
. Would it be worth updating the script to do this? It may allow for RandomizedSearch to further refine performance.