[ML] Fix cause of SIGSEGV for classification and regression #1379
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
tveasey
commented
Jul 3, 2020
| EInitializationStage m_InitializationStage = E_NotInitialized; | ||
| std::size_t m_NumberThreads; | ||
| std::size_t m_DependentVariable = std::numeric_limits<std::size_t>::max(); | ||
| TOptionalSize m_PaddedExtraColumns; |
There was a problem hiding this comment.
Note that this doesn't need persisting because it is recalculated every time.
tveasey
added a commit
to tveasey/ml-cpp-1
that referenced
this pull request
Jul 6, 2020
tveasey
added a commit
to tveasey/ml-cpp-1
that referenced
this pull request
Jul 6, 2020
This was referenced Jul 6, 2020
tveasey
added a commit
that referenced
this pull request
Jul 6, 2020
tveasey
added a commit
that referenced
this pull request
Jul 8, 2020
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Outdated suggestions cannot be applied.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.
We weren't accounting for the padding we include since we memory aligned columns, #1142, in
CBoostedTreeFactorywhen computing the number of extra columns. This meant we were using presenting extra columns to feature selection (which would nearly always ignore them) and occasionally changing the initialisation of the learn rate.The problem we hit which caused the SIGSEGV was when there was only one feature which itself carried no information, i.e. it was a constant. We could then erroneously select an extra column as a feature.