关于GCU、沐曦GPGPU、MLU、0卡V100资源4月7日恢复上架的公告>>> 关于共建具身智能开源数据集的倡议>>> 关于云脑任务中统一路径访问方式的公告>>> 关于将启智集群GPU资源迁移至智算集群的公告>>>
Xin Yan thomas-yanxin
Public Activity Repositories Datasets Starred Repositories
31
 Following
24
 Followers
43
 Achievement Badge
7
Loading Heatmap…

thomas-yanxin synced commits to main at thomas-yanxin/MedicalGPT from mirror

15 hours ago

thomas-yanxin synced commits to main at thomas-yanxin/accelerate from mirror

  • 6af157ea93 Add diffusers to req (#2711)
  • 83317b3081 add distributed examples (#2672) * add distributed examples * typo * uncomment * require multigpu * add stable diffusion example * style * add copyright * style * remove tqdm * Apply suggestions from code review Co-authored-by: Zach Mueller <muellerzr@gmail.com> * add comments * remove print * More comments --------- Co-authored-by: Zach Mueller <muellerzr@gmail.com>
  • e831bcb3b1 Change dataloader send_to_device calls to non-blocking (#2685) * Change dataloader send_to_device calls to non-blocking * add non_blocking to dataloader dataclass * add dataloader non blocking option from dataclass * add handling for non blocking to accelerator * add notes on non-blocking transfers to quicktour * link to dataloaderconfiguration in docs * linting * "requires" -> "recommended" on non-blocking setting Co-authored-by: Zach Mueller <muellerzr@gmail.com> --------- Co-authored-by: drhead <a@a.a> Co-authored-by: Zach Mueller <muellerzr@gmail.com>
  • 092c3af0c4 Add version checks for the import of DeepSpeed moe utils (#2705) * fix import for moe utils * Apply suggestions from code review Co-authored-by: Zach Mueller <muellerzr@gmail.com> --------- Co-authored-by: Zach Mueller <muellerzr@gmail.com>
  • 3e944c5583 add cann version info to command accelerate env (#2689)
  • Compare 10 commits »

15 hours ago

thomas-yanxin synced commits to develop at thomas-yanxin/PaddleMIX from mirror

1 day ago

thomas-yanxin synced commits to v4.40-release at thomas-yanxin/transformers from mirror

1 day ago

thomas-yanxin synced commits to react-code-agent at thomas-yanxin/transformers from mirror

  • 4c33190c10 Merge branch 'agents' into react-code-agent
  • 90f69495e2 Add messages to agents (#5) (#30337) Add message passing format Co-authored-by: Cyril Kondratenko <kkn1993@gmail.com>
  • f8020520db Using variables (#4) (#30297) Support variable usage in ReactAgent
  • 91e50ef173 Improve logging (#3) (#30254) * Add custom errors, improve logging
  • Compare 4 commits »

1 day ago

thomas-yanxin synced commits to muellerzr-reset-state at thomas-yanxin/transformers from mirror

  • 51cd21cd47 Finish rebase
  • fdef46509d Merge branch 'main' into muellerzr-reset-state
  • 661190b44d [`research_project`] Most of the security issues come from this requirement.txt (#29977) update most of decision transformers research project
  • d0d430f14a Fix wrong indent in `utils/check_if_new_model_added.py` (#30456) fix Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
  • c9693db2fc Phi-3 (#30423) * chore(root): Initial commit of Phi-3 files. * fix(root): Fixes Phi-3 missing on readme. * fix(root): Ensures files are consistent. * fix(phi3): Fixes unit tests. * fix(tests): Fixes style of phi-3 test file. * chore(tests): Adds integration tests for Phi-3. * fix(phi3): Removes additional flash-attention usage, .e.g, swiglu and rmsnorm. * fix(phi3): Fixes incorrect docstrings. * fix(phi3): Fixes docstring typos. * fix(phi3): Adds support for Su and Yarn embeddings. * fix(phi3): Improves according first batch of reviews. * fix(phi3): Uses up_states instead of y in Phi3MLP. * fix(phi3): Uses gemma rotary embedding to support torch.compile. * fix(phi3): Improves how rotary embedding classes are defined. * fix(phi3): Fixes inv_freq not being re-computed for extended RoPE. * fix(phi3): Adds last suggestions to modeling file. * fix(phi3): Splits inv_freq calculation in two lines.
  • Compare 294 commits »

1 day ago

thomas-yanxin synced commits to muellerzr-checkpoint-callbacks at thomas-yanxin/transformers from mirror

  • e46a42e12c Fix failing tests
  • 865bf2a06c Merge branch 'main' into muellerzr-checkpoint-callbacks
  • 661190b44d [`research_project`] Most of the security issues come from this requirement.txt (#29977) update most of decision transformers research project
  • d0d430f14a Fix wrong indent in `utils/check_if_new_model_added.py` (#30456) fix Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
  • c9693db2fc Phi-3 (#30423) * chore(root): Initial commit of Phi-3 files. * fix(root): Fixes Phi-3 missing on readme. * fix(root): Ensures files are consistent. * fix(phi3): Fixes unit tests. * fix(tests): Fixes style of phi-3 test file. * chore(tests): Adds integration tests for Phi-3. * fix(phi3): Removes additional flash-attention usage, .e.g, swiglu and rmsnorm. * fix(phi3): Fixes incorrect docstrings. * fix(phi3): Fixes docstring typos. * fix(phi3): Adds support for Su and Yarn embeddings. * fix(phi3): Improves according first batch of reviews. * fix(phi3): Uses up_states instead of y in Phi3MLP. * fix(phi3): Uses gemma rotary embedding to support torch.compile. * fix(phi3): Improves how rotary embedding classes are defined. * fix(phi3): Fixes inv_freq not being re-computed for extended RoPE. * fix(phi3): Adds last suggestions to modeling file. * fix(phi3): Splits inv_freq calculation in two lines.
  • Compare 346 commits »

1 day ago

thomas-yanxin synced commits to main at thomas-yanxin/transformers from mirror

  • 5c57463bde Enable fp16 on CPU (#30459) * Check removing flag for torch * LLM oops * Getting there... * More discoveries * Change * Clean up and prettify * Logic check * Not
  • d1d94d798f Neuron: When save_safetensor=False, no need to move model to CPU (#29703) save_safetensor=True is default as of release 4.35.0, which then required TPU hotfix https://github.com/huggingface/transformers/pull/27799 (issue https://github.com/huggingface/transformers/issues/27578). However, when the flag save_safetensor is set to False (compatibility mode), moving the model to CPU causes generation of too many graphs during checkpoint https://github.com/huggingface/transformers/issues/28438. This PR disable moving of model to CPU when save_safetensor=False.
  • 661190b44d [`research_project`] Most of the security issues come from this requirement.txt (#29977) update most of decision transformers research project
  • d0d430f14a Fix wrong indent in `utils/check_if_new_model_added.py` (#30456) fix Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
  • c9693db2fc Phi-3 (#30423) * chore(root): Initial commit of Phi-3 files. * fix(root): Fixes Phi-3 missing on readme. * fix(root): Ensures files are consistent. * fix(phi3): Fixes unit tests. * fix(tests): Fixes style of phi-3 test file. * chore(tests): Adds integration tests for Phi-3. * fix(phi3): Removes additional flash-attention usage, .e.g, swiglu and rmsnorm. * fix(phi3): Fixes incorrect docstrings. * fix(phi3): Fixes docstring typos. * fix(phi3): Adds support for Su and Yarn embeddings. * fix(phi3): Improves according first batch of reviews. * fix(phi3): Uses up_states instead of y in Phi3MLP. * fix(phi3): Uses gemma rotary embedding to support torch.compile. * fix(phi3): Improves how rotary embedding classes are defined. * fix(phi3): Fixes inv_freq not being re-computed for extended RoPE. * fix(phi3): Adds last suggestions to modeling file. * fix(phi3): Splits inv_freq calculation in two lines.
  • Compare 18 commits »

1 day ago

thomas-yanxin synced commits to add-clip-sdpa at thomas-yanxin/transformers from mirror

1 day ago

thomas-yanxin synced commits to builds/2.2.3 at thomas-yanxin/lightning from mirror

1 day ago

thomas-yanxin synced commits to readme-tests at thomas-yanxin/lit-parrot from mirror

1 day ago

thomas-yanxin synced commits to phi-3-checkpoint at thomas-yanxin/lit-parrot from mirror

1 day ago

thomas-yanxin synced commits to main at thomas-yanxin/lit-parrot from mirror

1 day ago

thomas-yanxin synced commits to main at thomas-yanxin/DB-GPT from mirror

  • 6520367623 support oceanbase as an optional vector database (#1435) Signed-off-by: shanhaikang.shk <shanhaikang.shk@oceanbase.com>
  • 91c1371234 feat(RAG):add cross-encoder rerank (#1442) Co-authored-by: Fangyin Cheng <staneyffer@gmail.com>
  • Compare 2 commits »

1 day ago

thomas-yanxin synced commits to main at thomas-yanxin/Qwen from mirror

1 day ago

thomas-yanxin synced commits to dev at thomas-yanxin/Langchain-Chatchat from mirror

1 day ago

thomas-yanxin synced commits to release/2.8 at thomas-yanxin/PaddleNLP from mirror

  • 3105c18b01 fix 0f428bbe47daed3cd861f7047c3e9acbec4ea0b1 try import
  • 0f428bbe47 [Distributed] [CustomDevices] Adapt SP on lora && polish MC2 APIs (#8303) * [Distributed] adapt sequence parallel on LoRA (#8235) * [Distributed] [CustomDevices] adapt lora sp && polish MC2 APIs
  • 871070d95e bug fixer (#8314) (#8318)
  • Compare 3 commits »

1 day ago

thomas-yanxin synced commits to release/2.7 at thomas-yanxin/PaddleNLP from mirror

1 day ago

thomas-yanxin synced commits to develop at thomas-yanxin/PaddleNLP from mirror

1 day ago

thomas-yanxin synced commits to release/2.0 at thomas-yanxin/swift from mirror

1 day ago

Please read the following content carefully:

Dear OpenI User

Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.

For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》

Agree and continue
Disagree, exit