Skip to content

Pull requests: huggingface/accelerate

New pull request New
30 Open 1,578 Closed
30 Open 1,578 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
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

support for wrapped schedulefree optimizer when using deepspeed
#3266 opened Nov 26, 2024 by winglian Loading…
5 tasks
Update deferring_execution.md
#3262 opened Nov 26, 2024 by max-yue Loading…
5 tasks
1
Fix: Resolve #3257
#3261 opened Nov 26, 2024 by as12138 Loading…
5 tasks done
1
2
Replaced set/check breakpoint with set/check trigger in the troubleshooting documentation
#3259 opened Nov 25, 2024 by relh Loading…
1 of 5 tasks
2
Select the DeepSpeedCPUOptimizer based on the original optimizer class.
#3255 opened Nov 24, 2024 by eljandoubi Loading…
6
Revert default behavior of get_state_dict_from_offload
#3253 opened Nov 23, 2024 by kylesayrs Loading…
4
Fix: Resolve #3060, preload_module_classes is lost for nested modules
#3248 opened Nov 22, 2024 by wejoncy Loading…
1
6
Fix : get_balanced_memory when using multi gpus with small models or quantized models with a large vocabulary
#3244 opened Nov 18, 2024 by MekkCyber Loading…
1 of 5 tasks
1
5
[RFC] Support FSDP2
#3231 opened Nov 8, 2024 by kmehant Loading…
1 of 5 tasks
1
1
Fix slurm multinode example
#3229 opened Nov 8, 2024 by ffrancesco94 Loading…
3 of 5 tasks
1
[docs] update set-seed
#3228 opened Nov 8, 2024 by faaany Loading…
7
[docs] fix typo
#3221 opened Nov 5, 2024 by faaany Loading…
3
[docs] use real path for checkpoint
#3220 opened Nov 5, 2024 by faaany Loading…
3
Ensure explicit output dtype for pad_across_processes
#3219 opened Nov 5, 2024 by mariusarvinte Loading…
4 of 5 tasks
1
3
create _preprare_fsdp to pre- prepare fsdp model training
#3213 opened Nov 3, 2024 by eljandoubi Loading…
1
10
fix load_state_dict for npu
#3211 opened Nov 2, 2024 by statelesshz Loading…
1 of 5 tasks
1
Give example on how to handle gradient accumulation with cross-entropy
#3193 opened Oct 24, 2024 by ylacombe Loading…
26
[MLU] update deepspeed-mlu dependency
#3192 opened Oct 24, 2024 by Andy666G Loading…
1 of 5 tasks
6
Distributed inference example for llava_next
#3179 opened Oct 20, 2024 by VladOS95-cyber Loading…
3 of 5 tasks
36
feat: support tensor parallel using Pytorch 2.0 & Data loader
#3173 opened Oct 16, 2024 by kmehant Loading…
1 of 5 tasks
26
add use_all_gather for option
#3164 opened Oct 14, 2024 by SangbumChoi Loading…
5 tasks
1
7
🚀 [Feature] Add deprecated Decorator
#3161 opened Oct 12, 2024 by yhna940 Loading…
14
🚑 [HotFix] Fix dev environment for CPU Docker
#3160 opened Oct 12, 2024 by yhna940 Loading…
4
add support for custom function for reducing the batch size wip Work in progress
#3071 opened Sep 3, 2024 by winglian Loading…
5 tasks
7
Some adjustment for supporting Deepspeed-Ulysses wip Work in progress
#2877 opened Jun 20, 2024 by zeyugao Loading…
1 of 5 tasks
12
ProTip! Adding no:label will show everything without a label.