Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

FastMKS segfault fix #1986

Merged
merged 3 commits into from Aug 26, 2019

Conversation

@rcurtin
Copy link
Member

commented Aug 22, 2019

I dug into #1767 and found that the issue is actually not the fact that the hyperbolic tangent kernel is only conditionally positive definite. Instead, for the optdigits dataset, the issue is that the dot product between each point is sufficiently large that every kernel evaluation is identical---meaning that the distance between points is always 0.

This then means that the cover tree gets built in a weird way that is actually invalid. So I fixed this bug in e584aea.

While I was in there, I also made the FastMKS binding check for a valid value of k, if it is passed, and added a test for it. And I hid the Log::Fatal output from the one test that tries to load nonexistentfile.arff. :)

@zoq
zoq approved these changes Aug 22, 2019
Copy link
Member

left a comment

Great, no comments from my side.

@mlpack-bot
mlpack-bot bot approved these changes Aug 23, 2019
Copy link

left a comment

Second approval provided automatically after 24 hours. 👍

@rcurtin rcurtin merged commit fff2357 into mlpack:master Aug 26, 2019
6 checks passed
6 checks passed
LaTeX Documentation Checks Build finished.
Details
Memory Checks Build finished.
Details
Static Code Analysis Checks Build finished.
Details
Style Checks Build finished.
Details
continuous-integration/appveyor/pr AppVeyor build succeeded
Details
continuous-integration/travis-ci/pr The Travis CI build passed
Details
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
2 participants
You can’t perform that action at this time.