Merged
Conversation
This is redundant, since the model is always created on fit. Further, `set_tunable_parameters` may be called before `fit` but proper initialization is only guaranteed after the first call to `fit`.
The scikit-learn estimator API stipulates that __init__ should only store its parameters and do no validation (since parameters can later be overridden in set_parameters). Parameters need to be validated on use. https://scikit-learn.org/stable/developers/develop.html#random-numbers
7 tasks
Codecov Report
@@ Coverage Diff @@
## master #117 +/- ##
==========================================
+ Coverage 55.81% 60.18% +4.37%
==========================================
Files 116 116
Lines 7643 7656 +13
==========================================
+ Hits 4266 4608 +342
+ Misses 3377 3048 -329
|
kiudee
approved these changes
May 13, 2020
Owner
kiudee
left a comment
kiudee
left a comment
There was a problem hiding this comment.
This change looks straightforward and good to me.
Closed
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Description
Part of #94 and #116.
Motivation and Context
The scikit-learn estimator API requires us to store all
__init__parameters unmodified. Validation should be delayed as late as possible. That is becauseset_paramscould override those parameters. See https://scikit-learn.org/stable/developers/develop.html#random-numbers.I'm using the pattern of setting
self.random_state_at the beginning of every fit function. In theory that is not always needed, since often the random state is only used in one function. I think its less error-prone to always do it the same way though.How Has This Been Tested?
Ran the pre-commit hooks and the test-suite.
Does this close/impact existing issues?
#94 and #116
Types of changes
Checklist: