Jinwang Wang

Results 11 comments of Jinwang Wang

@Shashikant255 I use two **same classes** to train SqueezeDet, it works.

@Shashikant255 As far as I know, you should delete these label files. File reader does not care about images without labels files.

Hi, The annotation format of AI-TOD is COCO. Best On Dec 14, 2021, at 18:02, PeterPham ***@***.***> wrote:  Hi @jwwangchn which label format dataset for training (coco,voc, yolo...)? Best...

@Sharpiless [mmcv-nwd](https://github.com/jwwangchn/mmcv-nwd) is now updated to 1.3.5

@lx709 @starsky0426 I'm sorry that I didn't fully verify the codes before uploading them. The missing code has been completed. Please verify it. In addition, since I don't have a...

The annotations of the BONAI dataset are annotated by SenseTime's internal annotation tool, without any open-source software.

Since the validation code is deeply integrated into another toolkit, I need some time to sort it out. It is expected to be completed in the next two weeks.

Can you share the accuracy comparison results between CLIP-ResNet50 and ImageNet-ResNet50 on the COCO dataset?