Skip to content

Issues: microsoft/DeepSpeed

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

738 Open 988 Closed
738 Open 988 Closed
Author
Filter by author
Label
Filter by label
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Milestones
Filter by milestone
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

[BUG] bs=1 mp_size=8 OPT/LLaMA inference CUDA error: an illegal memory access was encountered bug Something isn't working inference
#3758 opened Jun 15, 2023 by chhzh123
@RezaYazdaniAminabadi
3
[Question] Does deepspeed support model parallelism via API "PipeModelDataParallelTopology"? bug Something isn't working training
#3757 opened Jun 15, 2023 by chenyaofo
DeepSpeed stage 3 doesn't move weight to GPU automatically after version 0.9.3 bug Something isn't working training
#3751 opened Jun 14, 2023 by calvinzhan
[BUG] Questions about PipelineEngine._reduce_output when output is list bug Something isn't working training
#3750 opened Jun 14, 2023 by x54-729
[BUG] ZeRO is unsupported in init_inference bug Something isn't working inference
#3746 opened Jun 13, 2023 by molohov
Can you provide a startup script for deepspeed inference llama 65B MP=8? bug Something isn't working compression
#3744 opened Jun 13, 2023 by zcuuu
Uses bf16 training, there is an abnormal loss[BUG] bug Something isn't working training
#3742 opened Jun 13, 2023 by suolyer
1
[REQUEST] Dict input and output of Pipeline Parallelism enhancement New feature or request
#3738 opened Jun 12, 2023 by x54-729
[BUG] Is it right that per_device_train_batch_size = per_device_mini_train_batch_size * gradient_accumulation_steps bug Something isn't working training
#3737 opened Jun 12, 2023 by feiliya333
[BUG] RuntimeError: inflight params error when using DeepSpeed for Reinforcement Learning bug Something isn't working deepspeed-chat Related to DeepSpeed-Chat
#3735 opened Jun 12, 2023 by shyustc
3
High Peak GPU Memory with ZeRO Stage 3 bug Something isn't working training
#3734 opened Jun 11, 2023 by gnovack
@jomayeri
4
[BUG]AttributeError: 'HfDeepSpeedConfig' object has no attribute 'trainer_config_finalize' bug Something isn't working training
#3733 opened Jun 10, 2023 by 2018211801
[REQUEST] DeepSpeed Zero3, gathering, updating and repartitioning gradients enhancement New feature or request
#3732 opened Jun 10, 2023 by bhavyashahh
[REQUEST] Can the DeepSpeed support automatic selection of different types of network cards, such as Ethernet and high-speed IB network cards? enhancement New feature or request
#3729 opened Jun 9, 2023 by pengshuang
pip install deepspeed Assertion `err == hipSuccess' failed
#3725 opened Jun 9, 2023 by sirus20x6
@loadams
2
[BUG] deepspeed-chat bloom training error, raise RuntimeError "still have inflight params " after 14 steps training of step3 with offload option turned on bug Something isn't working deepspeed-chat Related to DeepSpeed-Chat
#3724 opened Jun 9, 2023 by DZ9
2
When I load model and use deepspeed with ZERO stage 3, I have a problem with high cpu memory Usage
#3722 opened Jun 9, 2023 by ZJXNEFU
[REQUEST]can deepspeed support hybrid shard enhancement New feature or request
#3721 opened Jun 9, 2023 by QingruiSun
[BUG] Getting .half() is not supported when using QLora bug Something isn't working compression
#3719 opened Jun 8, 2023 by abdulvirta
1
[BUG] KeyError in stage_1_and_2.py when training dreambooth with deepspeed (in kohya_ss) bug Something isn't working training
#3718 opened Jun 8, 2023 by me-fraud
[BUG] CPU Offloading Super Slow Even on a 60M Parameter Model? bug Something isn't working training
#3717 opened Jun 8, 2023 by rileyhun
[QUESTION] GPT2 not supported by AutoTP enhancement New feature or request
#3711 opened Jun 8, 2023 by Yejing-Lai
[BUG] DeepSpeed Inference example in the tutorial got killed for no reason. bug Something isn't working inference
#3710 opened Jun 8, 2023 by Entropy-xcy
1
[REQUEST] Mixture of Experts (MoE) Segmentation Task enhancement New feature or request
#3701 opened Jun 7, 2023 by deep-matter
chatglm-6b can not use deepspeed inference[BUG] bug Something isn't working inference
#3690 opened Jun 6, 2023 by zuocebianpingmao
ProTip! Add no:assignee to see everything that’s not assigned.