1
0
Fork 0
QuaPy/MultiLabel/NOTES.txt

39 lines
1.0 KiB
Plaintext

Classifiers
- Classifiers binary, single-label, OneVsRest or MultiOutput:
- LR
- LinearSVC (?)
- Classifiers natively multi-label:
- from scikit-multilearn (x11)
-
Protocols:
- NPP
- APP (for each class)
Things to test:
- MultiChain for classification, MultiChain for regression...
- Reimplement stacking with sklearn.ensemble.StackingClassifier? No parece facil.
- Independent classifiers + independent quantifiers
- Stacking + independent quantifiers
- ClassifierChain + independent quantifiers
- Independent quantifiers + cross-class regression (independent?)
- Stacking + cross-class regression
- ClassifierChain + cross-class regression
- Covariates (Means, CovMatrix from samples) + multioutput regression?
- Covariates concatented with quantifiers predictions + cross-class regression?
- Model Selection for specific protocols?
TODO:
- decide methods
- decide classifiers binary
- decide classifiers multi-label
- decide quantifiers naive
- decide quantifiers multi-label
- decide datasets