Skip to content
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

398 Open 409 Closed
398 Open 409 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
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀
[BUG] network crashes with stage=3 & dynamically-changed network bug
#1843 opened Mar 18, 2022 by amsword
[BUG] DS_BUILD_OPS =1 does not work bug
#1840 opened Mar 16, 2022 by oborchers
@jeffra
5
Consolidate norm computation logic enhancement
#1839 opened Mar 16, 2022 by tjruwase
[REQUEST] Why ep_size should be less than world_size? enhancement
#1838 opened Mar 16, 2022 by Satan012
[BUG]AttributeError: 'DeepSpeedEngine' object has no attribute 'quantizer' bug
#1837 opened Mar 16, 2022 by AQA6666
2
Add explicit gradient_accumulation_dtype config enhancement
#1835 opened Mar 15, 2022 by jeffra
2
[BUG] Job fails with NCCLTimeout error bug
#1834 opened Mar 15, 2022 by amsword
Is "High-performance INT8 inference kernels" released?
#1833 opened Mar 15, 2022 by wanghaoshuang
2
[REQUEST] ZeRO-Infinity: GPU Memory Usage Higher Than Expected enhancement
#1831 opened Mar 14, 2022 by kiehls90
4
[BUG] Same train time with DeepSpeed (despite increased batch size) bug
#1825 opened Mar 11, 2022 by pminhyung
Can deepspeed be used to infer fairseq trained gpt models? enhancement
#1824 opened Mar 11, 2022 by dingjingzhen
[BUG]: Deepspeed build fails on rocm , error in hipify bug
#1816 opened Mar 8, 2022 by bmedishe
5
[REQUEST] High CPU memory usage at initialization enhancement
#1814 opened Mar 7, 2022 by HaokunLiu
7
[BUG] deepspeed_aio_handle_t::_stop_threads(): Assertion `0 == _num_pending_ops' failed. bug
#1813 opened Mar 6, 2022 by JIN-096
@tjruwase
1
[Potential contribution] 'AdaFamily' optimizer enhancement
#1810 opened Mar 4, 2022 by hfassold
[REQUEST] Extraneous param validation in config enhancement
#1809 opened Mar 4, 2022 by jbloxham
[BUG] lr scheduler get_last_lr() does work with fp16 enabled bug
#1808 opened Mar 4, 2022 by mbetser
1
[BUG] Problem with optimizer update step bug
#1806 opened Mar 3, 2022 by benjamin-reichman
1
[BUG] ZeRO/bf16 grad accumulation in bf16 needs higher precision accumulator bug
#1800 opened Mar 1, 2022 by stas00
[BUG] DeepSpeed Inference with GPT-J using batches with padding gives wrong outputs bug
#1797 opened Feb 27, 2022 by tomerip
2
[BUG] IndexError / Runtime Error with torch.nn.TransformerEncoder bug
#1795 opened Feb 26, 2022 by floatshadow
[BUG] zero_to_fp32.py cannot convert the model bug
#1793 opened Feb 25, 2022 by chinoll
2
[BUG] save_checkpoint() missing some params bug
#1789 opened Feb 24, 2022 by skpig
2
[BUG] CUDA error with INT 8 inference bug
#1788 opened Feb 23, 2022 by gsujankumar
8
[BUG] bf16 w/ ZeRO Stage 0 assert bug
#1784 opened Feb 19, 2022 by stas00
1
ProTip! Follow long discussions with comments:>50.