flyingmrwang comments

Repositories
Issues
Comments

Results 5 comments of


                                            flyingmrwang

assert len(set(num_channels_list)) == 1

the same for speedup pruned ghostnet, and the num channels are 16 and 1

是否有可能增加landmark 點

@610265158 "switch to pytorch, HAHA... don't worry it wont happen." 🤔

Face alignment speed up and GPU usage

@Wenlong0913 I am facing the same problem. Could you give some hints on how much it is faster than this one? And do you think the key point of low...

'Tensor' object has no attribute 'uint8'

The same for torch 1.9 After change *.uint8() to *.to(torch.uint8), this part passed. However face another error: ![image](https://user-images.githubusercontent.com/18489060/127983522-023848a9-9fa1-4d04-860b-c4cc5e7c5f50.png)

'Tensor' object has no attribute 'uint8'

after wrapping export with torch.no_grad(), previous one passed.. While here comes another one: ![image](https://user-images.githubusercontent.com/18489060/128003313-5737e4f0-696e-47d2-ae6b-a809331d0ed2.png)