Skip to content

[DO NOT MERGE] Exporting dynamic version of google/vit-base-patch16-224 model #6249

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Open
wants to merge 4 commits into
base: master
Choose a base branch
from
Open

[DO NOT MERGE] Exporting dynamic version of google/vit-base-patch16-224 model #6249

wants to merge 4 commits into from

Conversation

@sdasgup3
Copy link
Collaborator

@sdasgup3 sdasgup3 commented Jan 2, 2024 •
edited
Loading

This PR tracks all the changes in PyTorch/XLA codebase to export google/vit-base-patch16-224 model with the prerequisite that

Note that the PR is based on Support of implicit broadcasting with unbounded dynamism #6219
 which is yet to be merged. Feel free to review only the followings files/methods which are implemented on top on the artifacts in #6219

  1. File: torch_xla/csrc/convolution.cpp
  2. File: torch_xla/csrc/data_ops.cpp
  3. File: torch_xla/csrc/helpers.cpp
    • XlaHelpers::DynamicReshape
    • XlaHelpers::DynamicUnboundedReshape
    • XlaHelpers::DynamicUnboundedBroadcast
@sdasgup3 sdasgup3 requested review from GleasonK, lsy323 and qihqi January 2, 2024 22:54
@sdasgup3 sdasgup3 force-pushed the sdasgup3/export_hf_vit_base_patch16_224 branch from 44c4160 to ec70d3b Compare January 3, 2024 17:52
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

2 participants

@sdasgup3