WIP: Resolve "Adding new datasets"

added 2 commits

• af068a00 - Better to take non zero values of list as indicated in the numpy doc
• 69f71671 - Merge branch '17-adding-new-datasets' of...

Compare with previous version

added 1 commit

• 8d911afd - Add log to find no_weight bug

Compare with previous version

added 1 commit

• 3507558e - Results for California

Compare with previous version

added 1 commit

• a955fda6 - Adding kin8nm optimal hyperparameters

Compare with previous version

added 13 commits

• 125817c1 - Add back kmeans method in the trainer (disapeared during a merge)
• 94668904 - Remove two deprecated points in the unused TODO list (uncomplete yet)
• ca5d0080 - Add missing import in trainer
• 46a4a8b0 - Add stage5 scripts for kmeans and sim
• 0363926f - Add wo_weights and similarity in stage5 loss plot
• 1194ee2f - Integrate Paolo's code of method 'Ensemble selection from libraries of models'...
• bf240b77 - Add missing self in ensemble_selection_forest_regressor
• 86c4cf10 - Fix wrong variable name in ensemble_selection_forest_regressor
• 2a265135 - Fix hyperparams bugs in base and random. Fix extracted forest size used in...
• 731cee0a - Add stage5 experiments script
• 138660cb - Update fix random strategy (wip
• 6483c0dc - Correction on random extraction
• 5f15d2a5 - Merge branch '15-integration-sota' into 17-adding-new-datasets

Compare with previous version

added 1 commit

• 802fc322 - Resultfs for california

Compare with previous version

added 1 commit

• dd5e9cde - Some new results

Compare with previous version

merged