Skip to content

Pull requests: Oneflow-Inc/oneflow

New pull request New
187 Open 9,390 Closed
187 Open 9,390 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

fix bug of group norm and layer norm for npu
#10609 opened Dec 20, 2024 by crazy-JiangDongHua Loading…
1
fix autocast to support global tensor
#10605 opened Dec 18, 2024 by fpzh2011 Loading…
Dev support graph 2d parallel on npu
#10566 opened Nov 18, 2024 by Flowingsun007 • Draft
1
Remove CUDA-only restriction for multi-tensor model updates in optimizer eager enhancement graph graph mode op
#10564 opened Nov 14, 2024 by ShawnXuan Loading…
5
fix device_type2sub_tsk_gph_builder_
#10563 opened Nov 12, 2024 by Flowingsun007 Loading…
2
layer_norm_grad for npu
#10560 opened Nov 7, 2024 by fpzh2011 • Draft
Update README.md, adding python 3.12 and cuda version cu121, cu122, and things related to PEP 668
#10557 opened Sep 12, 2024 by Neutron3529 Loading…
2
add Tensor method: new and new_ones
#10553 opened Aug 21, 2024 by crazy-JiangDongHua Loading…
1
Add softmax parameter stacklevel
#10552 opened Aug 20, 2024 by Dmovic • Draft
add interpolate_like for cpu enhancement op
#10544 opened Jul 9, 2024 by woaixiaoxiao Loading…
1 task done
10
add support for xpu/KunLunXin device feature system
#10540 opened Jul 3, 2024 by fpzh2011 Loading…
3
autocast support new devices enhancement feature system
#10539 opened Jun 26, 2024 by fpzh2011 Loading…
9
profiler adapt to ascend npu
#10536 opened Jun 18, 2024 by Flowingsun007 • Draft
Support input check for pool operator enhancement system
#10532 opened May 24, 2024 by Dmovic Loading…
18
18
flash attention use mirror automerge build enhancement
#10530 opened May 22, 2024 by mosout • Draft
3
Use cpm and move flash_attention to external
#10529 opened May 22, 2024 by mosout • Draft
Use Opentelemetry
#10526 opened May 15, 2024 by mosout • Draft
[feature]add tensor folder map to graph
#10525 opened May 13, 2024 by crazy-JiangDongHua • Draft
add FusedApplyRotaryEmbGradKernel enhancement op
#10517 opened May 11, 2024 by cccddd77 Loading…
6
Update dependency constraints on cudnn build enhancement
#10515 opened May 6, 2024 by jackalcooper Loading…
2
modify clip_grad with no to_global enhancement system
#10443 opened Mar 11, 2024 by hanwen-sun Loading…
24
Update cudnn convolution kernel enhancement feature op
#10440 opened Mar 6, 2024 by linzs148 Loading…
10
[bugfix] enable faster rcnn and sd model with oneflow backend automerge bug system
#10439 opened Mar 5, 2024 by crazy-JiangDongHua Loading…
4
Add sync data for to_global function
#10433 opened Feb 6, 2024 by hanwen-sun Loading…
2
Sync oneflow pro interpolate like op
#10421 opened Jan 26, 2024 by lixiang007666 Loading…
ProTip! Type g p on any issue or pull request to go back to the pull request listing page.