Cannot finetune on custom dataset - Githubissues

tinyvision / DAMO-YOLO

DAMO-YOLO: a fast and accurate object detection method with some new techs, including NAS backbones, efficient RepGFPN, ZeroHead, AlignedOTA, and distillation enhancement.

Apache License 2.0

3.75k stars 470 forks source link

Cannot finetune on custom dataset #108

Open doubleK29 opened 1 year ago

doubleK29 commented 1 year ago

Before Asking

[X] I have read the README carefully. 我已经仔细阅读了README上的操作指引。
[X] I want to train my custom dataset, and I have read the tutorials for finetune on your data carefully and organize my dataset correctly; 我想训练自定义数据集，我已经仔细阅读了训练自定义数据的教程，以及按照正确的目录结构存放数据集。
[X] I have pulled the latest code of main branch to run again and the problem still existed. 我已经拉取了主分支上最新的代码，重新运行之后，问题仍不能解决。

Search before asking

[X] I have searched the DAMO-YOLO issues and found no similar questions.

Question

Set up:

I have already organize my dataset and configs correctly.
My GPU is a RTX 3090 24GB with cuda 11.7.
Issue:

I have a command like this to train the model: python -m torch.distributed.launch --nproc_per_node=1 tools/train.py -f configs/damoyolo_tinynasL18_Ns.py BUT here is what I got:

P/S: I tried the solution that adding python path from issue 11 but it didn't work with my case. Please drop me a message if you think you can help. Thank you in advance!

Additional

No response

ksv87 commented 5 months ago

try torchrun tools/train.py -f configs/damoyolo_tinynasL18_Ns.py