Seyyed Hossein Hasanpour

There are different versions, which the difference between them is how they are cleaned. Use the `ms1m-retinaface-t1.zip` this is the one that was used to train these models.

@snahmad could you find anything for server side?

Its been around 2~3 months since the last comment here, Is there any update in this regard? should we or can we be expecting something in near future(even possibly the...

I noticed the channel numbers are governed by the [torch_pruning/prune/structured.py](https://github.com/VainF/Torch-Pruning/blob/master/torch_pruning/prune/structured.py) and simply using the `abs()` around them would solve this. However, after doing so, I get errors complaining about the...

@VainF Thanks a lot for the prompt response, really appreciate it. lowering the pruning rate for the last 2 layers, actually solved this. but could you kindly tell me whats...

Thanks, really appreciate your time and professionalism. Best regards

@Blue-Eagle-10 I faced this on windows, you need to be on linux I guess in order not to face this error.

For those of you that are looking for alignment and how its done, you can have a look [here](https://github.com/foamliu/InsightFace-v2/blob/master/align_faces.py) basically there are many ways you can do it, but the...

I'm facing this on the latest version of opencv (4.8.0) as well. is this still expected or am I dealing with sth entirely different? I noticed the conda version has...

Hi, Yes BCE usually performs better it seems. I included this because it was also in the keras/tensorflow official repository, I just added it for the sake of completeness and...