Don't mutate the rng hyper-parameter in fit#64
Merged
Merged
Conversation
This was referenced Jun 15, 2026
Merged
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Currently, the
rnghyper-parameter, if an AbstractRNG, is mutated byfit. This makes it harder to get reproducibility in some contexts. This PR hasfitmake a copy of therngbefore using it.The state of the
rngcopy is now cached by the random forest models, so that it can be used in theupdatemethods for warm-restarts (by handing it offDecisionTree.build_tree). The idea is that training 15 = 10 + 5 trees in two steps, with a warm restart, ought to be equivalent to training in one step with 15 trees. Turns out this already didn't work prior to this PR, and, because of the waybuild_treeworks, cannot currently be made to work within MLJDecisionTree. However, as I wrote a test for this, I've included it here and tagged it as broken, with the explanation I've just made included as a code comment.