Upgrade to tensorflow2 #3
Merged
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
Owner
atorch
commented
Mar 22, 2020
•
atorch
commented
- I couldn't get class weights to work, so I deleted them and replaced them with temporal sample weights
- I should explain the idea behind masking in the readme: CDL developed contains roads, buildings, and (at 30 meter resolution) many NAIP pixels that are neither roads nor buildings, hence the idea of masking out CDL developed when computing loss.
- Without masking, the model learns to predict developed for pixels near roads. Often, the ground truth is that these pixels are crops or pasture or forest.
- It looks like Keras - how to use class_weight with 3D data keras-team/keras#3653 still has no elegant solution, even with tensorflow2
* I couldn't get class weights to work, so I deleted them for now
atorch
commented
Mar 22, 2020
| ) | ||
| mask -= cdl_index_indicators | ||
|
|
||
| return K.categorical_crossentropy(y_true, y_pred) * mask |
There was a problem hiding this comment.
I got errors here when running fit_model.py on this branch
There was a problem hiding this comment.
I replaced this with temporal sample weights (based on class): see keras-team/keras#3653 (comment)
atorch
commented
Mar 22, 2020
|
|
||
| # TODO Unused, reimplement this? | ||
| cdl_indices_to_mask = label_encoder.transform(CDL_CLASSES_TO_MASK) |
There was a problem hiding this comment.
Explain the idea behind masking in the readme: CDL developed contains roads, buildings, and (at 30 meter resolution) many NAIP pixels that are neither roads nor buildings, hence the idea of masking out CDL developed when computing loss.
… (for temporal sample weights)
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.