ENH: Stop unstable lowess iterations #9220
Open
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
jseabold
commented
Apr 18, 2024
| # std_resid[std_resid >= 0.999] = 1.0 | ||
| # std_resid[std_resid <= 0.001] = 0.0 | ||
|
|
||
| return bisquare(std_resid) |
There was a problem hiding this comment.
I just copy-pasted this into the loops, so I could check the median and break. There are other ways to do this...
|
Test failures look unrelated. |
This was referenced Apr 18, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
NumPy's guide.
I was running into some issues trying to interpolate between a lot of 0s and 1s similar to some of the old test cases and issues. I noticed that R just bails out if the MAD gets close to zero "since the algorithm is highly unstable in that case." This seems to close a few of the open lowess issue and (partially) fix the one I was having.
I removed one of the test cases that is no longer the behavior. I guess technically this is a breaking change, but the new behavior now matches R.
This partially fixes #7337, if you consider extrapolating a good fix. Only the frac=.3 test case spits out NaNs now. I'll have to go through that part, but that's probably for a different PR.