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

Check PLS (elbow) heuristic #44

Closed
fradav opened this issue Jan 7, 2020 · 1 comment
Closed

Check PLS (elbow) heuristic #44

fradav opened this issue Jan 7, 2020 · 1 comment
Assignees
@fradav
Labels
estim-param related to the parameter estimation methodology invalid This doesn't seem right
Milestone
1.2

Comments

@fradav
Copy link
Collaborator

fradav commented Jan 7, 2020

Manual exhaustive PLS computing (all axis computed) at 99% of the maximum of the explained Y-variance shows a lot more axis than our heuristic. In theory the heuristic guarantees than we should get at least more axis than the complete computation.
@fradav fradav added invalid This doesn't seem right estim-param related to the parameter estimation methodology labels Jan 7, 2020
@fradav fradav added this to the 1.2 milestone Jan 7, 2020
@fradav fradav self-assigned this Jan 7, 2020
@fradav
Copy link
Collaborator Author

fradav commented Jan 8, 2020

Ok, so the assumption we have a concave function is plain wrong.
We revert to a manual heuristic, checking a window of Y explained variance to check if we really have a sensible "elbow" (actually 15-values sized window).
Done in 8d3cee5

@fradav fradav closed this as completed Jan 8, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@fradav fradav

Labels
estim-param related to the parameter estimation methodology invalid This doesn't seem right
Projects
None yet
Milestone
1.2
Development

No branches or pull requests
1 participant
@fradav