-
https://github.com/ddps-lab/research-issues/issues/554 이슈와 관련하여 멀티노드 DDD가 필요합니다.
그러나 현재 코드로는 멀티노드 DDD가 불가능하여 해당 이슈를 해결합니다.
-
- The issue has been described here-> https://github.com/DalgoT4D/DDP_backend/issues/738
-
I use this setting below to train flux lora:
```
accelerate launch --gpu_ids 0,1 --main_process_port 29502 --mixed_precision bf16 --num_cpu_threads_per_process=2 \
flux_train_network.py --pr…
-
Hi @chrockey, great work!
Can you guide me on how to set up multigpu training? I have only 20GB gpus available, and when using batch size of 2 I obtain poor performance (~6% lower mIoU and mAcc; pr…
-
## 🚀 Feature
Make DDP/FSDP a regular transform (to a large part including making transforms flexible enough to support this).
### Motivation
Currently DDP/FSDP is not a regular transform, lea…
t-vi updated
4 months ago
-
Thank you for the work on the DDP component, so far so good with working with WLED and DDP RGB.
I noticed that only RGB is currently supported.
WLED now has support for DDP RGBW, have you consider…
-
Because config is instantiated only on the main process and then DDP is initialized, objects that are initialized in the config are not DDP aware.
Maybe there is a way to fix this by instantiating…
-
使用两张卡训练的时候,一直卡在Load backbone weights,差不多要三四分钟才能继续
-
How to run ns-render in ddp mode for faster inference?
-
## Instructions To Reproduce the Issue:
to speedup training, I add torch.compile operation after DistributedDataParallel in detectron2/engine/defaults.py:
```
ddp = DistributedDataParallel(mo…