Skip to content

Issues: NVIDIA/FasterTransformer

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

Clear current search query, filters, and sorts
249 Open 376 Closed
249 Open 376 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
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

Always got the same results even enabling random seed. bug Something isn't working
#525 by zhang-ge-hao was closed Mar 29, 2023
10
update of performance data
#523 by jinluyang was closed Apr 10, 2023
5
The FT t5 beam search algorithm generates inconsistent results with HF's
#522 by shiqingzhangCSU was closed Aug 18, 2023
8
The FT t5 beam search algorithm generates inconsistent results with HF's bug Something isn't working
#521 by shiqingzhangCSU was closed Mar 25, 2023
generic_activation is using too many registers bug Something isn't working
#520 by lzhangzz was closed Aug 23, 2023
2
How to convert vit weights from pytorch to fastertransformer format for plugin use ? question Further information is requested
#519 by ywfwyht was closed May 12, 2023
9
RuntimeError: Tried to instantiate class 'FasterTransformer.ParallelGptOp', but it does not exist! Ensure that it is registered via torch::class_ bug Something isn't working
#518 by Siyuan011 was closed Mar 23, 2023
11
Accuracy of weight only quantization is abnormal bug Something isn't working
#514 by lippman1125 was closed Mar 24, 2023
1
Can I use cross-attention in gpt example?
#513 by pangsg was closed Mar 22, 2023
7
FT SmoothQuant doesn't use INT8 BMM question Further information is requested
#512 by lippman1125 was closed Mar 24, 2023
Rouge accuarcy of Large-Scale OPT weight-only quantized models heavily drop bug Something isn't working
#510 by lippman1125 was closed Mar 24, 2023
7
use lambada_task_example.py to test ft opt-125m, but get zero accuracy bug Something isn't working
#509 by lippman1125 was closed Mar 24, 2023
4
What does "mem_seq_len" mean in bert_example.py
#508 by pangsg was closed Mar 20, 2023
5
What changes in the output of a conversion if multiple DSM are set?
#507 by Chris113113 was closed Mar 27, 2023
1
There is a bug about shared_contexts_ratio bug Something isn't working
#504 by zhang-ge-hao was closed Mar 28, 2023
1
shared_contexts_ratio bug Something isn't working
#501 by zhang-ge-hao was closed Mar 16, 2023
shared_contexts_ratio bug Something isn't working
#500 by zhang-ge-hao was closed Mar 16, 2023
shared_contexts_ratio bug Something isn't working
#499 by zhang-ge-hao was closed Mar 16, 2023
shared_contexts_ratio
#498 by zhang-ge-hao was closed Mar 16, 2023
The whole vit network as a trt plugin operator?
#497 by ywfwyht was closed Mar 17, 2023
3
Build Python Version Failed bug Something isn't working
#496 by pangsg was closed Mar 16, 2023
6
What is 'internal error' on nccl_utils? bug Something isn't working
#495 by lkm2835 was closed Mar 16, 2023
4
lib/libth_bert.so: undefined symbol
#494 by pangsg was closed Mar 17, 2023
10
build FasterTransformer fail
#493 by shiqingzhangCSU was closed Apr 4, 2023
5
convert the SmoothQuant model for w8a8 inference in FT failed
#490 by Anychnn was closed Mar 15, 2023
1
ProTip! Exclude everything labeled bug with -label:bug.