This repository has been archived by the owner on Nov 1, 2024. It is now read-only.
-
Notifications
You must be signed in to change notification settings - Fork 729
Issues: facebookresearch/metaseq
Author
Label
Projects
Milestones
Assignee
Sort
Issues list
Improved memory utilization in API
api
enhancement
New feature or request
#106
opened May 18, 2022 by
stephenroller
Fine-tune OPT with my own dataset
question
Further information is requested
#105
opened May 18, 2022 by
xiaomaiaa
Add OPT to huggingface conversion guidelines/scripts
enhancement
New feature or request
#98
opened May 12, 2022 by
tbmihailov
assert key_padding_mask.size(1) == src_len in 350M model
question
Further information is requested
#89
opened May 10, 2022 by
Mrs-Hudson
Running opt-baselines requires os.environ["USER"]
bug
Something isn't working
#83
opened May 10, 2022 by
Leli1024
No clear way to load models
better-eng
Things that can help make things sane
bug
Something isn't working
#78
opened May 10, 2022 by
stephenroller
interactive_cli requires making changes to services/constants.py
better-eng
Things that can help make things sane
bug
Something isn't working
#77
opened May 10, 2022 by
stephenroller
layer_norm is fp32, can't be wrapped inside half precision layers.
bug
Something isn't working
#66
opened May 8, 2022 by
BlackSamorez
Track all of our RNG offsets to avoid collisions
enhancement
New feature or request
good first issue
Good for newcomers
#65
opened May 8, 2022 by
suchenzang
Releasing prediction logits for NLP tasks?
enhancement
New feature or request
#25
opened May 3, 2022 by
tholiao
"RuntimeError: torch.distributed is not yet initialized but process group is requested" when trying to run API
question
Further information is requested
#23
opened May 3, 2022 by
jminjie
[eval] Process hanging when eval 175B model on more than 1 nodes
bug
Something isn't working
#2
opened Apr 4, 2022 by
qingliu1111
[eval] RuntimeError: attn_batches % batches_per_block when run with certain tasks
bug
Something isn't working
#18
opened Apr 2, 2022 by
tbmihailov
Inconsistency between valid ppl from TensorBoard and eval_lm.py
bug
Something isn't working
#17
opened Apr 1, 2022 by
danielsimig
Make eval_lm.py work with .jsonl formatted data
enhancement
New feature or request
#16
opened Mar 31, 2022 by
danielsimig
Validate args to prevent RuntimeError: attn_batches % batches_per_block == 0
#13
opened Mar 29, 2022 by
GabrielPereyra
Refactor prompt generation to not require loading a model
better-eng
Things that can help make things sane
enhancement
New feature or request
#12
opened Mar 28, 2022 by
ramakanth-pasunuru
Add tests for MultiplePadDataset class
better-eng
Things that can help make things sane
enhancement
New feature or request
good first issue
Good for newcomers
#11
opened Mar 28, 2022 by
punitkoura
Defining arch from args/checkpoint/default?
better-eng
Things that can help make things sane
bug
Something isn't working
config
eval
#10
opened Mar 25, 2022 by
suchenzang
Clean up utils.py
better-eng
Things that can help make things sane
cleanup
#9
opened Mar 20, 2022 by
suchenzang
ProTip!
Find all open issues with in progress development work with linked:pr.