yang-ruixin

Much lower loss but wrong output

1

Hi authors, Thanks for your great work. I did an experiment: 1. I trained a model successfully, with loss around 0.6, and the output were meaningful. 2. I did gamma...

feature: export head for moby

1

Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand...

请问如何得到自监督MoBY算法最终predictor的输出结果，直接用于以图搜图

1

**Describe the feature** 作者您好，当前推理时只能得到backbone的输出或者projector_q的输出，我希望可以得到最终predictor的输出，直接用于以图搜图。 在模型导出时，当前将MoBY导出为一个分类模型，舍弃了MoBY的head（即predictor），添加了一个dummy classification head。 我也可以自己实现，但是因为EasyCV包了很多层，可以实现的方式比较多，你们会官方来实现吗，或者给些建议，最elegant的实现方式。 如果我实现了可以pull request吗？ 谢谢

In wav2lip project, if use argument --face_enhancement, GPEN-512.pdparams will be downloaded automatically. Where does this weights come from? How can I train this weights?

1

Dear authors, When I try [wav2lip](https://github.com/PaddlePaddle/PaddleGAN/blob/develop/docs/en_US/tutorials/wav2lip.md), if use argument --face_enhancement, GPEN-512.pdparams will be downloaded automatically. Where does this weights come from? How can I train this weights? Thanks in advance.