ZhengMengbin

Results 6 issues of ZhengMengbin

As mentioned in title, have you used giou/diou/ciou to train yolox, and who is better?

Can convert HRFormer backbone into TensorRT? I want to speed it, but the structure is so complex, especially 'class MHA_(MultiheadAttention)', can you give some advice?

I use the configs/pretrain/seg_rec_poly_fuse_feature.yaml to train icdar_2013 dataset, and use default config(Use two 1080ti GPUs): ![image](https://user-images.githubusercontent.com/26650043/97465527-1a6eaf80-197d-11eb-9cb8-636ac4258c30.png) But, I meet the Nan loss. Is the learning rate too high, can you...

When I bulit tf_mesh_renderer using runtests.sh, it failed: ![46724C3F-CAF7-4B4C-B1A7-C2D1BFBC5384_4_5005_c](https://user-images.githubusercontent.com/26650043/76982830-6a14e380-6977-11ea-818c-96ce835c8615.jpeg) Can you give me some help? I use two codes to build bazel-2.2.0 as follows: `chmod +x bazel--installer-linux-x86_64.sh ` `./bazel--installer-linux-x86_64.sh --user...

Thanks very much for your work! I only read the paper(I do not read the code), and I want to know how to classify pixels in occlusion, using multi-label classification(i.e....

Can you provide the training loss log? And how to evaluate pretrain stage's performance ?