Luyue Lin

Guangzhou, Guangdong Province, China My interests include deep learning and machine learning for computer vision such as VAE-based networks, CNN, SVM-based classification, and Bayesian method.

Results 3 issues of


                                            Luyue Lin

waiting for supporting cython

1

enhancement

help wanted

Would you please reimplement this source code in Pytorch, since the theano is no longer supported.

Why should we train NetL separately in train_L_step_w

1

In the function `train_L_step_w`, line 192, in `train.py`, we the gradients of loss are firstly propagated to NetL with Doc3D data set. Then in the gradients are propagated to NetL...