User/rcadene/2024 10 07 vla #467
Draft
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
danaaubakirova
commented
Oct 10, 2024
| ) | ||
|
|
||
| hidden_states = llava_output.hidden_states[-1] # Use last layer's hidden state | ||
| hidden_states = hidden_states[:, -4:, :] #make 4 a config parameter |
There was a problem hiding this comment.
Taking the last 4 embeddings as an input to the action decoder. Because chunk_size is divisible by 4.
danaaubakirova
commented
Oct 10, 2024
Comment on lines
+84
to
+86
| processed_inputs = self.processor( | ||
| text=batch["prompt"], videos=list(batch["observation.images"]), return_tensors="pt", padding=True, do_rescale=False | ||
| ).to(self.device) |
There was a problem hiding this comment.
for some reasons I lose the original batch size? it should be 2. Maybe because we have a 5-dim input, not sure what to do with the camera index. I remove normalization as well, because the processor takes unnormalized PIL images/frames
danaaubakirova
commented
Oct 10, 2024
| Returns: | ||
| action_logits: Tensor of predicted actions. | ||
| """ | ||
| batch_size = hidden_states.size(0) # Ensure batch size is extracted |
There was a problem hiding this comment.
This part is the core and I think it might contain the mistakes. the way we repeat the input to match the chunk_size and also the encoder_out value needs to be checked and reviewed.
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Outdated suggestions cannot be applied.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.
What this does
Explain what this PR does. Feel free to tag your PR with the appropriate label(s).
Examples:
How it was tested
Explain/show how you tested your changes.
Examples:
test_somethingintests/test_stuff.py.new_featureand checked that training converges with policy X on dataset/environment Y.some_function, it now runs X times faster than previously.How to checkout & try? (for the reviewer)
Provide a simple way for the reviewer to try out your changes.
Examples:
SECTION TO REMOVE BEFORE SUBMITTING YOUR PR
Note: Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR. Try to avoid tagging more than 3 people.
Note: Before submitting this PR, please read the contributor guideline.