Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Am I using the best algorithm for univLogReg? #8

Closed
privefl opened this issue Oct 10, 2017 · 3 comments
Closed

Am I using the best algorithm for univLogReg? #8

privefl opened this issue Oct 10, 2017 · 3 comments
Labels
enhancement question

Comments

@privefl
Copy link
Owner

privefl commented Oct 10, 2017

For univariate logistic regression (big_univLogReg), maybe it could be improved (faster? always converge?).
See this SO question.
@privefl
Copy link
Owner Author

privefl commented Jun 30, 2018

This is equivalent to what I currently use.

@privefl privefl closed this as completed Jun 30, 2018
@privefl privefl reopened this Feb 3, 2019
@privefl privefl changed the title Am I using the best algorithm for IRLS? Am I using the best algorithm for univLogReg? Feb 3, 2019
@privefl
Copy link
Owner Author

privefl commented Feb 3, 2019

BFGS seems to be fast. See https://tminka.github.io/papers/logreg/minka-logreg.pdf.

@privefl
Copy link
Owner Author

privefl commented Feb 4, 2019

BFGS is not precise enough to determine std error (see https://github.com/privefl/bigstatsr/blob/master/tmp-tests/univLogReg-BFGS2.cpp).

@privefl privefl closed this as completed Feb 4, 2019
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
enhancement question
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@privefl