Feature/sg 573 pose estimation #634
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
…into feature/SG-573-pose-estimation
# Conflicts: # src/super_gradients/common/registry/registry.py
shaydeci
requested changes
Jan 23, 2023
There was a problem hiding this comment.
Nice!
Besides some little comments, my main suggestions, detailed inline are:
- Regarding the way we handle transforms.I know we've yet to come up with a descent solution...but maybes better to handle transforms that are already implemented and can be resolved (which by itself, might impose a problem if we are allowing our users to resolve for example SegmentationTransform, but lets leave that for now..) by stating that non KeyPointsTransform are applied only on image, then we can reduce the amount of transforms we would need to reimplement simply for packing and unpacking items into the sample. See my suggestion inline, and let me know what you think (:
- Pose estimation base dataset- so I think it would be a good idea to pull at least some of the class methods of the dataset which we can probably take use in, and form a base class. If you feel that for COCO everything is very specific, we can probably skip this part and as time progresses and we have more datasets, the common methods will be more apparent and we can form a base class.
See my other comments inline.
src/super_gradients/training/datasets/pose_estimation_datasets/coco_keypoints.py
Outdated
Show resolved
Hide resolved
src/super_gradients/training/datasets/pose_estimation_datasets/coco_keypoints.py
Show resolved
Hide resolved
src/super_gradients/recipes/dataset_params/coco_pose_estimation_dataset_params.yaml
Outdated
Show resolved
Hide resolved
src/super_gradients/recipes/dataset_params/coco_pose_estimation_dataset_params.yaml
Outdated
Show resolved
Hide resolved
# Conflicts: # src/super_gradients/common/registry/registry.py
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.
No description provided.