[ML] Impose an absolute cutoff on the minimum variance #8
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
… to avoid overflow when calculating lgamma values
tveasey
approved these changes
Mar 8, 2018
lib/maths/CXMeansOnline1d.cc
Outdated
| @@ -306,7 +306,8 @@ void BICGain(maths_t::EDataType dataType, | |||
| double a = square(m + offset) / v; | |||
| double b = (m + offset) / v; | |||
|
|
|||
| double vmin = std::min(MIN_RELATIVE_VARIANCE * v, MIN_ABSOLUTE_VARIANCE); | |||
| double vmin = std::max(std::min(MIN_RELATIVE_VARIANCE * v, MIN_ABSOLUTE_VARIANCE), | |||
| square(MIN_RELATIVE_VARIANCE * offset)); | |||
There was a problem hiding this comment.
I think this is exactly the right idea to fix this issue. To be consistent we probably shouldn't square MIN_RELATIVE_VARIANCE, i.e. this should be MIN_RELATIVE_VARIANCE * square(offset) so we are saying a lower bound on the variance should be the offset^2 we apply. I think the clearest way to indicate this intention is std::min(MIN_RELATIVE_VARIANCE * std::max(v, square(offset)), MIN_ABSOLUTE_VARIANCE).
Address code review comments. fixes #488
sophiec20
changed the title
droberts195
pushed a commit
that referenced
this pull request
Apr 23, 2018
Impose an absolute cutoff on the minimum variance based on the offset to avoid overflow when calculating lgamma values fixes #488
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Outdated suggestions cannot be applied.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.
[ML] To avoid overflow when calculating lgamma values an absolute cutoff on the minimum variance, based on the offset, is imposed.
fixes #488