Apply the same preprocessing to the test set as to the training set #213
Conversation
This file contains 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
|
|
birm
approved these changes
May 16, 2023
There was a problem hiding this comment.
I had some issues checking this out with the binder link (seemed to freeze at step 40/63 when looking for the python dependencies; maybe it was just slow, but I came back some time later where it had just nondescriptly errored. I'll try it again next PR).
That said, this clearly fixes the issue. Hope you're doing well rcurtin!
![mlpack-bot[bot]](https://github.com/avatars/in/23934?s=60&v=4){kind=small}
|
Thanks! Yeah, doing well, just chugging away at a first release of bandicoot. :) Hope things are well for you too! |
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Outdated suggestions cannot be applied.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.
I noticed while reading the CI logs for #209 that the test accuracy on
mnist_batch_normandmnist_simplewas atrocious (9%!) even though the training and validation accuracies seemed fine. A quick look indicated that this was because we were using the wrong dimension for labels, and formnist_simplewe weren't normalizing the test data. Easy fix, and now the test accuracies for both are what is expected (95%+).