wenshihao1
wenshihao1
I want to ask if anyone has solved it?
> [https://github.com/WongKinYiu/yolov9/pull/13/files](url) 它的工作原理，你可以尝试一下。 > > ``` > with dt[2]: > preds = preds[0][1] if isinstance(preds[0], list) else preds[0]#add > ``` Where to add it?
But I had a problem when I was training. ---- 回复的原邮件 ---- | 发件人 | ***@***.***> | | 日期 | 2024年05月21日 19:29 | | 收件人 | ***@***.***> | | 抄送至...
Didn't find a place to add in val.py ---- Replied Message ---- | From | ***@***.***> | | Date | 05/21/2024 19:40 | | To | ***@***.***> | | Cc...
ok, I will try. Thanks. ---- Replied Message ---- | From | ***@***.***> | | Date | 05/21/2024 19:54 | | To | ***@***.***> | | Cc | ***@***.***>***@***.***> |...
export DATA_ROOT=path/to/coco_dataset # with pretrained object detector on 4 datasets python train_caption.py exp.name=caption_4ds model.detector.checkpoint=4ds_detector_path # with pretrained object detector on Visual Genome python train_caption.py exp.name=caption_4ds model.detector.checkpoint=vg_detector_path this i can't unsderstand....
#vi_caption_path = "/media/local_workspace/quang/datasets/vietcap/vi_captions.json" This means all data in json format?
Sorry, i can't find the example to train my dataset. I just can find three files. Thanks for your reply. Good luck with your work At 2023-08-01 21:07:27, "Van-Quang Nguyen"...