search

speedinghzl / CCNet

CCNet: Criss-Cross Attention for Semantic Segmentation (TPAMI 2020 & ICCV 2019).
MIT License
1.43k stars 276 forks source link

about INF #115

Open Thatboy7 opened 2 years ago

Thatboy7 commented 2 years ago

excuse me . Is there an another method? the torch.diag methon isn't supported by onnx version and when i use torch.eye ,my tensorrt doesn‘t support it either.

speedinghzl commented 2 years ago

@Thatboy7 You could drop the INF if you train the model again. The overlapping pixels have a minor effect on the final performance.