get_best_not_failing_model

hcrystalball.model_selection.get_best_not_failing_model(grid_search, X, y)

Prevent situation when model incompatible data are not seen during CV (in the last split of actuals) and model which cannot be fitted on the full dataset is choseen. In such a situation the next model with the lowest test score should be selected.

Parameters

• grid_search (sklearn.model_selection.GridSearchCV) – Fitted instance compatible with Sklearn GridSearchCV

• X (pandas.DataFrame) – Input features.

• y (array_like, (1d)) – Target vector.

Returns

with keys rank and params

Return type

dict

Raises

ValueError – No available model could be fit on given data