add maxiter and maxfun optimisation arguments to LFR fit method #184
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
psattige
reviewed
Jun 17, 2020
There was a problem hiding this comment.
Thanks for the PR. Looks good to me.
@hoffmansc @nrkarthikeyan Any thoughts?
psattige
approved these changes
Jun 17, 2020
nrkarthikeyan
approved these changes
Jun 17, 2020
Illia-Kryvoviaz
pushed a commit
to Illia-Kryvoviaz/AIF360
that referenced
this pull request
Jun 7, 2023
…-fit add maxiter and maxfun optimisation arguments to LFR fit method
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Outdated suggestions cannot be applied.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.
The
fitmethod for theLFRclass has so far pre-defined arguments for the L-BFGS-B optimisation method, which the user cannot change. We hereby emphasisemaxiterandmaxfun, denoting the maximum number of iterations and function evaluations, which are pre-set to a value of5000.In some experiments, these values can be too small to allow for convergence of the optimisation method. This can lead to suboptimal solutions, hence suboptimal results for accuracy and fairness in predictions based on the learnt fair representation.
We added the arguments
maxiterandmaxfunto thefitmethod, which are set by default to the original values. However the user has now the freedom to change those values if required to ensure convergence.