Test fixes to sparse_center_data and center_data#3
Test fixes to sparse_center_data and center_data#3MechCoder wants to merge 3 commits intojnothman:fix_centeringfrom
Conversation
Had been normalizing to unit/n_samples variance since 295dc3c
Not that I understand why this is a feature, but it's tested.
|
Never mind the prev comment, I split it across another PR |
sklearn/linear_model/base.py
Outdated
There was a problem hiding this comment.
When I said there may be stability benefits to not dividing, that wasn't because we were creating too small a number to represent. Unless you have numbers with large negative exponents, dividing by a power of two is lossless, as far as I understand (but am far from an expert). Dividing, or multiplying, by some other number may be lossy, so this change is not beneficial (except that by keeping the old functionality the tests don't need fixing)
|
You've not pushed the sparsefuncs changes. |
|
But I'm not sure if they quite belong in the same PR anyway...? |
|
I think ym statement about the current version being good for stability is On 27 March 2014 00:36, Manoj Kumar notifications@github.com wrote:
|
|
@jnothman I removed the comment about numerical stability. for now this can be merged. |
|
Assuming unit variance isn't something we want..? On 27 March 2014 14:53, Manoj Kumar notifications@github.com wrote:
|
|
Yes, as I said before, I don't still completely understand why this is there. But I suppose it is better to keep it as it is, till someone has a good explanation. |
…scikit-learn#7838) * initial commit for return_std * initial commit for return_std * adding tests, examples, ARD predict_std * adding tests, examples, ARD predict_std * a smidge more documentation * a smidge more documentation * Missed a few PEP8 issues * Changing predict_std to return_std #1 * Changing predict_std to return_std #2 * Changing predict_std to return_std #3 * Changing predict_std to return_std final * adding better plots via polynomial regression * trying to fix flake error * fix to ARD plotting issue * fixing some flakes * Two blank lines part 1 * Two blank lines part 2 * More newlines! * Even more newlines * adding info to the doc string for the two plot files * Rephrasing "polynomial" for Bayesian Ridge Regression * Updating "polynomia" for ARD * Adding more formal references * Another asked-for improvement to doc string. * Fixing flake8 errors * Cleaning up the tests a smidge. * A few more flakes * requested fixes from Andy * Mini bug fix * Final pep8 fix * pep8 fix round 2 * Fix beta_ to alpha_ in the comments
No description provided.