-
-
Notifications
You must be signed in to change notification settings - Fork 453
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
Cdf testing with Kolmogorov Smirnov #1494
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
There are a couple of other Rust crates for KS stat/test, but on quick examination they may not be any use to us.
I am quite happy with the code. After a review I would add the missing tests. We might have to wait until a release of |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Could you please either rebase on master or just to drop the merge commit? I don't believe anything related was changed, so there should be no need to merge master into this PR.
I like the new approach using multiple seeds and parameter sets. Lets not add additional tests into this PR though; a new PR would be better for that (I mean, you can implement it all and push to a new PR immediately if you like; it would just need rebasing and possibly adjusting later).
2f2ae89
to
f7f029c
Compare
…f everything works correctly
f7f029c
to
17f3565
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
In progress...
I got an interesting test failure on macos:
Could be a compiler bug, maybe? Maybe someone with a mac can reproduce this? |
It passed when re-run. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good; just some nits below
CHANGELOG.md
entrySummary
As addition and alternative to pdf testing, I implemented distribution testing based on cdf
Motivation
The cdf tests should be more robust and avoid having the bin-width choice problem
Details
I have two simple example tests for normal and binomial. If we like these tests we should also add some way to test on different parameter sets and the other distributions.
This can't be easily generalized to multivariate ones.