Initial pr to support dynamic shape detection #7817
Closed
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
|
rioght now if I used this feature in https://github.com/pytorch/xla/blob/master/examples/train_decoder_only_base.py it will error out because second step graph is different from the first step one(most likely due to the optimizer state). |
|
Close in favor of #7918 |
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Outdated suggestions cannot be applied.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.
The implementation in this pr is not ideal because it can only handle a single graph for a compiled program. This is usually too strict for the real world program because in training we usually at least have 2 graphs since in the step0 the optimizer state is not initialized.
I think what I need to do next is to provide a
max_dynamic_graph_allowedarguments and use a TRIE to keep tracked of all the IR being generated and when they diverage.