This repository was archived by the owner on Aug 21, 2025. It is now read-only.

Repository navigation

All pull requests

Labels (16)Milestones (2)

Fix bibtex entry
#1105bamos opened Jan 19, 2023
[hotfix] fix build error from FuncTorch dispatch key
#1016winnie1994 opened Sep 9, 2022
2
[POC] vmap for TensorDict
#1004vmoens opened Aug 13, 2022
1
[WIP,POC] Faster functional modules
#983vmoens opened Jul 24, 2022
8
[WIP] Create test-m1.yml
#960vfdev-5 opened Jul 13, 2022Contributor
Will that work
#920malfet opened Jun 28, 2022Contributor
Separate forward and backwad compilation and support higher order derivatives for aot_function
#856anjali411 opened Jun 7, 2022
5
Fixed issue with total_weight in nll_loss_forward_decomposition
#829vfdev-5 opened May 22, 2022Contributor
3
Add the ability to trace TensorList in-place ops
#828alanwaketan opened May 20, 2022
3
added list clearing FX codegen for backwards in AOTAutograd.
#779Chillee opened May 5, 2022Contributor
3
Add decomposition for aten.native_batch_norm_backward op
#675Shukla-Gaurav opened Apr 7, 2022
10
Fixing tensor.numpy on wrapped tensors
#627vfdev-5 opened Mar 29, 2022Contributor
2
[skip-ci] Fixed get_perf in jacobians_hessians
#598vfdev-5 opened Mar 15, 2022Contributor
[PROTOTYPE] generated batching rules for custom dispatcher ops
#578bdhirsh opened Mar 9, 2022Contributor
2
Added support for linalg.lu_factor* as _lu_with_info will be deprecated
#386vfdev-5 opened Jan 10, 2022Contributor
2
[skip ci] WIP on index_fill batch rule
#370vfdev-5 opened Dec 27, 2021Contributor
WIP: Fixing clamp tests
#225Padarn opened Oct 26, 2021Contributor1 of 2 tasks
1