Xuan Lin
Xuan Lin
Usually nan problem occurs when the init weight of some layers are set too large. And I've tested the current initialization is ok. Are you using the latest version of...
There's no python scripts for these ops. You can find the c++ scripts on mxnet/src/operator. The output of linear regression op is identical to its input
I've tested the code on ubuntu14.04 + mxnet v0.9.3 but I didn't encounter the nan problem you mentioned. @peyer @huynhthedang
I'm sorry but I have no idea because I didn't implement landmark localization task in my code.
Something like `AttributeError: 'NoneType' object has no attribute 'copy'` ? That means pnet didn't detect any faces
@Cv9527 Copy mxnet_diff.patch to your mxnet path and run `git apply mxnet_diff.patch `
@dtivger No it's not necessary to make clean, just git apply then run make.
No, landmark localization is not implemented in the code.
@dtivger CelebA faces only contribute to landmark localization task so the bounding box is useless. Use logging.basicConfig() to specify the log file name in example/train.py
@so-as That would be lots of things to do. Firstly you should crop, resize and save the faces from celebA dataset, and generate imglist file which contains these faces. The...