Refactor modelbuilder fit #198
Merged
Conversation
This file contains hidden or 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
Added method-specific inference functions to ModelBuilder and made fit call them based on a fit_method parameter. Tests coming in the next commit. Also need to try to factor out code common to both methods.
Added fit_method as a parameter to the tests so that test models are tested with mcmc and MAP fitting methods.
When test_save_load failed, pytest complained about the temp file not being closed. It didn't seem to cause any problems, but maybe this is nicer.
Refactored the ModelBuilder.fit method again to eliminate code redundancy between the two fitting methods. Eliminated extra kwargs from being passed to scipy.optimize.minimize.
twiecki
approved these changes
Jun 13, 2023
|
Thanks @pdb5627! |
michaelraczycki
added a commit
that referenced
this pull request
Jun 19, 2023
This reverts commit 2da3c81.
michaelraczycki
added a commit
that referenced
this pull request
Jul 2, 2023
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Outdated suggestions cannot be applied.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.
In response to #196, I made an attempt at the refactoring.
The first commit naively duplicates the existing
fitfunction for both mcmc and find_MAP fitting methods. The fourth commit puts some of the setup and cleanup code back in thefitfunction and switches only the actual inference part. I didn't know which way the team might prefer, so they're both there.I considered making an
Enumfor the fitting methods, but I saw thatpymc.sampling.sample()usesstr, so that's what I did here.I don't feel great about the handling of
sampler_configarguments here. There is some overlap in args to mcmcsample_*andfind_MAP, but some args are different. At leastscipy.optimize.minimizedoesn't accept extraneous kwargs, so I pass through only whitelisted args tofind_MAP.