Reference Issues/PRs

Issue #31571

What does this implement/fix? Explain your changes.

I found some bugs or unclear areas that need further improvement in several versions of whats_new documentation.

v1.5.0

• "Deprecates Y in favor of y in the methods fit, transform and inverse_transform of: :class:cross_decomposition.PLSRegression, :class:cross_decomposition.PLSCanonical, :class:cross_decomposition.CCA, and :class:cross_decomposition.PLSSVD."(link) However, classcross_decomposition.PLSSVD doesn‘t seem to have the inverse_transform method (refer to class PLSSVD)
• "store_cv_values and cv_values_ are deprecated in favor of store_cv_results and cv_results_ in ~linear_model.RidgeCV and ~linear_model.RidgeClassifierCV."(link) I recommend to make it clear that the store_cv_values and cv_values_ are Parameters (like the previous item), otherwise it will be misleading to know whether they are parameters or methods.

v1.4.0

• ":func:sklearn.extmath.log_logistic is deprecated and will be removed in 1.6. Use -np.logaddexp(0, -x) instead."(link) The full qualified name of function log_logistic should be sklearn.utils.extmath.log_logistic.

v1.3.0

• "The parameter log_scale in the class :class:model_selection.LearningCurveDisplay has been deprecated in 1.3 and will be removed in 1.5."(link) The deprecated parameter log_scale is actually a parameter of the plot method of the class model_selection.LearningCurveDisplay, not the class constructor. Refer to this.

Any other comments?

@adrinjalali