The Wayback Machine - https://web.archive.org/web/20230127164415/https://github.com/microsoft/DeepSpeed/issues
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

502 Open 664 Closed
502 Open 664 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] GPT-J inference on DeepSpeed using too much RAM and giving garbage results bug Something isn't working inference
#2755 opened Jan 27, 2023 by sridhardev07
Selecting a range of ports to try until an open port is found enhancement New feature or request
#2752 opened Jan 26, 2023 by BramVanroy
2
Question about section Ⅳ.C.(3) ”Communication Optimization“ in the paper “DeepSpeed Inference”
#2751 opened Jan 26, 2023 by Eutenacity
[BUG] OPT-66B: OOM at reasonable inference sizes bug Something isn't working inference
#2747 opened Jan 25, 2023 by aws-stdun
RuntimeError: 'weight' must be 2-D while training Flan-T5 models with stage 3 bug Something isn't working training
#2746 opened Jan 25, 2023 by smitanannaware
[BUG] message Creating fp16 ZeRO stage 3 optimizer logged while the training is a bf16's one bug Something isn't working training
#2739 opened Jan 23, 2023 by SaulLu
1
@loadams
2
[BUG] Loading MoE Checkpint with load_optimizer_states=False should tolerate missing optim_state files bug Something isn't working training
#2737 opened Jan 23, 2023 by clumsy
[BUG] RuntimeError: Tensors must be contiguous error while finetuning with deepspeed. bug Something isn't working training
#2736 opened Jan 23, 2023 by FahriBilici
@GuanhuaWang
1
[BUG] Caching op_builder kernels bug Something isn't working inference
#2735 opened Jan 23, 2023 by tchaton
@lokoppakmsft
Tests should fail indicating actual number of GPUs is below desired world_size
#2733 opened Jan 20, 2023 by clumsy
[BUG] Incorrect logits on Bloom models bug Something isn't working inference
#2730 opened Jan 20, 2023 by akamaster
@molly-smith
1
[BUG] Wrong logits/outputs when using HFOPTLayerPolicy on OPT model bug Something isn't working inference
#2729 opened Jan 20, 2023 by akamaster
@molly-smith
[BUG] Subpar time and accuracy performance when using Deepspeed compared to pure Pytorch bug Something isn't working training
#2724 opened Jan 19, 2023 by AnthoJack
@samadejacobs
CUDNN_STATUS_MAPPING_ERROR when using DeepSpeed own DataLoader
#2722 opened Jan 19, 2023 by yorickbrunet
1
[REQUEST] Building features only (instead of the whole DeepSpeed package) enhancement New feature or request
#2719 opened Jan 18, 2023 by taehyunzzz
[BUG] Make InferenceModule enable_cuda_graph more flexible. bug Something isn't working inference
#2717 opened Jan 18, 2023 by tchaton
@lekurile
[Question] What does inference_cuda_module.pad_transform_fp16 do ? bug Something isn't working inference
#2710 opened Jan 17, 2023 by tchaton
@cmikeh2
2
[BUG]A tremendous amount of errors when trying to install deepspeed with DS_BUILD_OPS=1 bug Something isn't working
#2707 opened Jan 16, 2023 by flandore
5
[REQUEST] Saving checkpoints to cloud bucket enhancement New feature or request training
#2701 opened Jan 13, 2023 by BioGeek
[BUG] Parameter CUDA alignment issue bug Something isn't working training
#2700 opened Jan 13, 2023 by achicu
@tjruwase
4
[BUG] ImportError with git build bug Something isn't working training
#2697 opened Jan 12, 2023 by brucethemoose
@HeyangQin
6
GPTJ DeepSpeed Inference Crashes After A Number Of Runs[BUG] bug Something isn't working inference
#2696 opened Jan 12, 2023 by mallorbc
@mrwyattii
2
[BUG] bug Something isn't working inference
#2693 opened Jan 12, 2023 by felifri
1
[QUESTION] Hosting\Inferencing requirements
#2692 opened Jan 12, 2023 by eli-halych
[BUG] init_inference() cannot load GPT2 from checkpoint bug Something isn't working inference
#2691 opened Jan 12, 2023 by Wenhan-Tan
ProTip! Mix and match filters to narrow down what you’re looking for.