Skip to content

Pull requests: pytorch/FBGEMM

New pull request New
356 Open 2,812 Closed
356 Open 2,812 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Improve FP8 BMM heuristic for large shapes and MoE E2E performance cla signed fb-exported
#3344 opened Nov 8, 2024 by jiawenliu64 Loading…
2
Add new optimizer state row_counter for Adam [Backend] cla signed fb-exported
#3342 opened Nov 8, 2024 by spcyppt Loading…
2
Remove unused-variable in /fbcode/deeplearning/fbgemm/fbgemm_gpu/src/ssd_split_embeddings_cache/kv_db_table_batched_embeddings.cpp cla signed fb-exported
#3335 opened Nov 6, 2024 by r-barnes Loading…
2
Fix global namespace pollution in ATen/Dispatch.h cla signed fb-exported
#3334 opened Nov 6, 2024 by slyfox3 Loading…
2
open-source SLL jagged_dense_bmm cla signed fb-exported
#3331 opened Nov 5, 2024 by brad-mengchi Loading…
5
Remove unused-variable in some generated code cla signed fb-exported
#3327 opened Nov 5, 2024 by r-barnes Loading…
2
Add support for int32_t indices in TBE training (2/N) cla signed fb-exported
#3326 opened Nov 5, 2024 by q10 Loading…
5
Add template info into generated files cla signed fb-exported
#3325 opened Nov 5, 2024 by q10 Loading…
2
- Kernel support for multiple buckets per rank cla signed fb-exported
#3323 opened Nov 5, 2024 by dstaay-fb Loading…
3
Add support for int32_t indices cla signed fb-exported
#3319 opened Nov 3, 2024 by q10 Loading…
4
Update benchmark test for Int32_t Indicies cla signed fb-exported
#3317 opened Nov 3, 2024 by q10 Loading…
4
Unitifed Prefetching API for CPU TBE cla signed fb-exported
#3314 opened Nov 2, 2024 by excelle08 Loading…
2
Allow registering SSD prefetcher after initiliaztion of the TBE module cla signed fb-exported
#3313 opened Nov 2, 2024 by excelle08 Loading…
2
Store the prefetching results in a queue in SSD prefetcher, retrieve later in .forward() method cla signed fb-exported
#3312 opened Nov 2, 2024 by excelle08 Loading…
2
Add manual loop unroll for rocm devices in fwd pass cla signed module: rocm
#3309 opened Nov 1, 2024 by avbokovoy Loading…
3
Do not call scalar_type cla signed fb-exported
#3308 opened Nov 1, 2024 by malfet Loading…
5
Clean up sparse bucketize cla signed fb-exported
#3302 opened Oct 31, 2024 by q10 Loading…
5
Reorganize sparse block bucketize macros cla signed fb-exported
#3296 opened Oct 30, 2024 by q10 Loading…
5
Refactor repeat code in sparse_block_bucketize cla signed fb-exported
#3295 opened Oct 30, 2024 by q10 Loading…
5
Add large my_size support in _block_bucketize_pooled_sparse_features_cuda_kernel2 cla signed fb-exported
#3294 opened Oct 30, 2024 by sryap Loading…
3
Add feature knobs to FBGEMM cla signed fb-exported
#3290 opened Oct 29, 2024 by q10 Loading…
3
Use c10::irange in deeplearning/fbgemm/BUCK +10 cla signed fb-exported
#3288 opened Oct 29, 2024 by q10 Loading…
2
Amd 2 cla signed fb-exported module: rocm
#3279 opened Oct 25, 2024 by jianyuh • Draft
3
repo cla signed fb-exported
#3274 opened Oct 24, 2024 by laithsakka Loading…
2
re-warmup cla signed fb-exported
#3271 opened Oct 23, 2024 by minhua-chen Loading…
3
ProTip! Add no:assignee to see everything that’s not assigned.