New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Zero mAP and no detections on custom dataset #61
Comments
What is the mAP if without TensorRT? Also did you set the classes to |
It is in the range of 70-80 without TensorRT. Also, the classes are set as ('person', ) label map is = { 1: 1,} |
If you try evaluate with FP16 TensorRT, will the AP improve? |
Unfortunately it is same with FP16 - zero mAP |
Please try the latest code base in haotian-dev branch. |
Hi @oidpsv ,I also encountered this problem. Have you found a solution now? |
@haotian-liu sorry to bother you. It can evaluate normally when I disable tensorrt. First of all:
The setting format about Mobilenetv3 and Rexnet refer the code. When I use model with rexnet, the error showed. And I changed the line 1512:
The error will not showed , but it still got zero mAP. I think maybe the problem is caused by my setting about Could you tell me how to set the Thanks for your reading~ Have a nice day. BTW, I have already used the command: |
@ntut108318099 You should put in |
@haotian-liu Thanks for your reply. Sorry, Thanks for your help. |
@ntut108318099 In each module, e.g. |
Depending on your advice, I convert the trt module successfully. Have you ever met this error before ? |
@haotian-liu Hi,Sorry to bother you. I have fixed the problem . The problem was caused by "PredictionModuleTRTWrapper" . I didn't notice that have some parameter also need to set . The result looks great !! Have a nice day . |
@ntut108318099 Glad you have the problem solved! |
Dear Haotian Liu,
currently I'm trying the yolact_edge trained on custom coco-like dataset for one class ('person'). Contrary to coco the image resolution is 512.
When I run evaluation with TensorRT conversion I get the following error during the protonet conversion:
which can be fixed by changing line 1563 in yolact.py from x = torch.ones((1, 256, 69, 69)).cuda()
to x = torch.ones((1, 256, 64, 64)).cuda().
The problem is that in this case further evaluation with TensorRT conversion gives zero mAP and processing of images provides empty result (no masks or boxes). Could you please help me? Many thanks.
The text was updated successfully, but these errors were encountered: