issues
search
huggingface
/
accelerate
🚀 A simple way to launch, train, and use PyTorch models on almost any device and distributed configuration, automatic mixed precision (including fp8), and easy-to-configure FSDP and DeepSpeed support
https://huggingface.co/docs/accelerate
Apache License 2.0
7.31k
stars
870
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
False device placement when use with quantization_config
#2905
xinghaow99
opened
19 hours ago
0
How to merge Qlora FSDP weights with an LLM and save model.
#2904
Minami-su
opened
1 day ago
0
With the same epoch, the result of multiple Gpus is much lower than that of a single gpu，why？
#2903
xiuguangLi
opened
3 days ago
0
Added a MultiCPU SLURM example using Accelerate Launch and MPIRun
#2902
okhleif-IL
opened
3 days ago
0
Problem on custom device_map
#2901
wonkyoc
opened
3 days ago
1
about run glm4 demo error
#2900
leizhu1989
opened
3 days ago
3
training loop freezes after first step on TPU
#2899
drimeF0
opened
3 days ago
2
Move to cpu takes extra memory usage after .gather()
#2898
xinghaow99
closed
1 day ago
4
Accelerate load_checkpoint_and_dispatch - RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cuda:1
#2897
adarsh-ks
opened
4 days ago
9
Feature Request: Pipeline multiple batches together for Llama3 70B distributed inference
#2896
ishan-gaur
closed
2 days ago
4
Add early support for `torchdata.stateful_dataloader.StatefulDataLoader` within the `Accelerator`
#2895
byi8220
opened
4 days ago
0
Importing `torchdata.stateful_dataloader` causes the test `check_seedable_sampler` to fail
#2894
byi8220
opened
4 days ago
1
notebook_launcher on kaggle tpu
#2893
lhiqwj173
opened
5 days ago
3
Add XLA Dynamo backends for training and inference
#2892
johnsutor
opened
5 days ago
1
How to set a custom Config in python code using Accelerate?
#2891
konstantinator
opened
5 days ago
1
More than 10 times slowdown between version 0.26.1 and version 0.31.0, EDIT: It was a data loading issue with Hugginface Datasets
#2890
marhlder
closed
5 days ago
6
Hotfix PyTorch Version Installation in CI Workflow for Minimum Version Matrix
#2889
yhna940
opened
6 days ago
4
Remove `log_line_prefix_template` argument from LaunchConfig to ensure compatibility with supported PyTorch versions
#2888
yhna940
opened
6 days ago
3
fix mlu device longTensor bugs
#2887
huismiling
opened
6 days ago
0
Can't apply LoRA's PiSSA weight init when using DeepSpeed ZeRO3 + LoRA to finetune!
#2886
ANYMS-A
closed
5 days ago
4
The saved model with deepspeed zero3 can not be correctly loaded
#2885
rubickkcibur
closed
3 days ago
2
Why is there a double fetch in the first batch when using accelerate?"
#2884
qsunyuan
opened
1 week ago
1
Add Profiler Support for Performance Analysis
#2883
yhna940
opened
1 week ago
3
accelerator.prepare just can be run jus once ?
#2882
DavideHe
opened
1 week ago
2
typo in examples/slurm/submit_multinode.sh script
#2881
hubutui
closed
1 week ago
2
Add ignore_unexpected_keys arg to load_checkpoint_in_model()
#2880
Qubitium
closed
1 week ago
1
fix `load_state_dict` for xpu and refine xpu safetensor version check
#2879
faaany
opened
1 week ago
1
add `require_triton` and enable `test_dynamo` work on xpu
#2878
faaany
opened
1 week ago
3
Some adjustment for supporting Deepspeed-Ulysses
#2877
zeyugao
opened
1 week ago
1
make more cuda-only tests device-agnostic
#2876
faaany
opened
1 week ago
3
Correct loading of models with shared tensors when using accelerator.load_state()
#2875
jkuntzer
opened
1 week ago
1
fix bug when getting the real accelerator's device number
#2874
faaany
opened
1 week ago
3
Plan to support FSDP2?
#2873
ByronHsu
opened
1 week ago
4
Accelerate test fails: Exception: Could not find the transformer layer class to wrap in the model
#2872
MikaSie
opened
1 week ago
8
Cannot free VRAM after loading a quantized model
#2871
lstein
opened
1 week ago
1
Support for Torch XLA Dynamo Backend
#2870
johnsutor
opened
1 week ago
1
Incorrect Argument Default for DeepSpeed Multi-node Training
#2869
jomayeri
opened
1 week ago
1
RuntimeError: Storage size calculation overflowed with sizes=[1, 4623015400198258675]
#2868
artkpv
opened
1 week ago
1
🍻 Add static typing
#2867
julien-blanchon
opened
1 week ago
4
Accelerate 0.31.0 gradient accumulation bug.
#2866
nikitabalabin
opened
1 week ago
1
Dataloader WeightedRandomSampler + Distributed Training
#2865
FrsECM
opened
1 week ago
4
[tests] enable XPU backend for `test_zero3_integration`
#2864
faaany
closed
4 days ago
2
[tests] fix bug in `test_tracking.ClearMLTest`
#2863
faaany
closed
1 week ago
1
Potentially fix tests
#2862
muellerzr
closed
1 week ago
2
[tests] use `torch_device` instead of `0` for device check
#2861
faaany
closed
1 week ago
2
[tests] skip bnb-related tests instead of failing on xpu
#2860
faaany
closed
1 week ago
2
Improve `skip_first_batches` method to efficiently support `IterableDataset` and `StatefulDataloader`
#2859
yzhangcs
opened
2 weeks ago
7
gather objects in TPU is not supported
#2858
carlesoctav
opened
2 weeks ago
1
Fix get_backend bug and add clear_device_cache function
#2857
NurmaU
opened
2 weeks ago
1
Drop torch re-imports in npu and mlu paths
#2856
dvrogozh
closed
2 weeks ago
1
Next