Closed
Description
As a result of #15123, we now have a common private loss function module under sklearn._loss. In sklearn.linear_models we have 2 algorithms that need Cython version that calculate losses and gradients on single values (not on arrays), namely
_plain_sgdas used inSGDClassifier,SGDRegressorandSGDOneClassSVM; andsag_solveras used inLogisticRegression.
My plan is to break this migration into smaller steps:
- DEP deprecate loss_function_ attribute in SGDClassifier and SGDOneClassSVM #27979 Deprecate
loss_function_attribute in v1.4 which gives access to the Cython loss functions that we want to replace. - MNT replace Cython loss functions in SGD part 1 #27999 change order of Cython loss function arguments to align with the ones in
sklearn._loss. - Carry out the deprecation after release 1.5 (to be released with 1.6)
MNT remove deprecated loss_function_ in SGD #29095 - ENH replace Cython loss functions in SGD part 2 #28029 Replace Cython losses with the ones from
sklearn._loss, except multinomial one - MNT replace Cython loss functions in SGD part 3 #28037 Replace the multinomial loss
Metadata
Metadata
Assignees
Type
Projects
Status
Done