8000 RidgeCV gives a different result from running Ridge with manually implemented CV · Issue #9299 · scikit-learn/scikit-learn · GitHub
[go: up one dir, main page]

Skip to content
RidgeCV gives a different result from running Ridge with manually implemented CV #9299
Closed
@pancodia

Description

@pancodia

Description

I run Ridge regression with two methods:

  1. Using RidgeCV
  2. Using KFold to prepare folds for CV and then use Ridge for each fold

The optional alpha selected from the two methods are very different.

Steps/Code to Reproduce

https://gist.github.com/pancha0/a1e76afd12b2d93af7ddabe53a55680a

Expected Results

The best alphas from the two methods should be close.

Actual Results

They are very different.

Versions

Darwin-16.6.0-x86_64-i386-64bit
Python 3.6.1 |Anaconda 4.4.0 (x86_64)| (default, May 11 2017, 13:04:09)
[GCC 4.2.1 Compatible Apple LLVM 6.0 (clang-600.0.57)]
NumPy 1.12.1
SciPy 0.19.0
Scikit-Learn 0.18.1

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions

      0