When I run: import torch.nn as nn from accelerate import Accelerator if __name__ == "__main__": accelerator = Accelerator() model = nn.Conv2d(10, 20, 3, 1, 1) print("prepare") model = accelerator.prepare(model) print("done") with config: compute_environment: LOCAL_MACHINE di…

Accelerate.prepare hang on single machine multiple gpu

kaimeng July 16, 2023, 1:13am 5

Hi, any updates?

Topic		Replies	Views
`Accelerator.prepare` utilize only one GPU instead of all the 8 available GPUs and raises "CUDA out of memory" 🤗Accelerate	3	2913	July 19, 2024
Accelerate on 1 GPU 🤗Accelerate	2	1918	April 8, 2022
What does "--multi_gpu" do under the hood? (and how to use it) 🤗Accelerate	7	6910	May 31, 2023
Multi-GPU Training sometimes working with 2GPU, but never more than 2 🤗Accelerate	5	3113	August 8, 2024
No GPUs found in distributed mode 🤗Accelerate	0	958	March 1, 2023