EuroSAT: use percentile normalization during plotting #2557
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
|
I don't think percentile normalization is a good way to get around whatever effects that the augmentations/normalization step adds. In some cases (e.g. I'm augmenting color or brightness) I'd like to be able to see differences between samples clearly. Having a reverse augmentation step is a better way to handle that case.
|
|
I'm also fine with a reverse normalization in the trainer/task/module/thingy. I've never actually tried it before, how straightforward is it to use? Does it have issues with cropping? Let me see how much work this is to do. |
|
Superseded by #2560 |
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Outdated suggestions cannot be applied.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.
Related to #1263
The current plotting method works great for our dataset, but fails as soon as normalization is added (e.g., via our data module). By using percentile normalization, both the dataset and data module should work.
P.S. We should do this for all datasets if anyone is ever bored and looking for an easy contribution. I only did EuroSAT because it's used throughout our tutorials.