Use log probability for distribution fitting - #85
Draft
jonathanberthias wants to merge 2 commits into
Draft
jonathanberthias wants to merge 2 commits into
jonathanberthias wants to merge 2 commits into
Conversation
jonathanberthias
force-pushed
the
migration/explicit-state-profiler
branch
from
September 27, 2026 13:51
3e0f5b9 to
6d69535
Compare
jonathanberthias
force-pushed
the
migration/log-probability
branch
2 times, most recently
from
September 27, 2026 20:27
360f181 to
7707140
Compare
jonathanberthias
force-pushed
the
migration/explicit-state-profiler
branch
2 times, most recently
from
October 2, 2026 22:25
4e1e242 to
fd2630a
Compare
jonathanberthias
force-pushed
the
migration/log-probability
branch
from
October 2, 2026 22:25
7f1ac6f to
654c6ff
Compare
jonathanberthias
force-pushed
the
migration/log-probability
branch
from
October 3, 2026 11:02
654c6ff to
7a83104
Compare
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The opt-in distribution path originally required density, CDF, and quantile methods, and likelihood fitting called
logpdf. Discrete and multivariate laws cannot always provide those operations, while a future backend should not have to implement SciPy method names to satisfy the statistical core.This PR makes pointwise
log_probthe common scoring method and uses it for likelihood fitting. Continuous density, CDF, and quantile methods stay on a univariate continuous subtype. Itspdfandlogpdfmethods are conveniences built onlog_prob; the SciPy adapter, truncation, GEV, and GPD implement the common method directly.This establishes the scoring boundary without adding a general shape or backend framework. The legacy public distribution imports remain in place.