Pretraining AE for UKT imporve experiment

UBCDingXin / improved_CcGAN

Continuous Conditional Generative Adversarial Networks (CcGAN)

https://arxiv.org/abs/2011.07466

MIT License

121 stars 35 forks source link

Pretraining AE for UKT imporve experiment #4

Closed ZeWang95 closed 3 years ago

ZeWang95 commented 3 years ago

Dear authors,

I found this no file and command for pretraining AE under the UTKFace-improved folder. However, I believe this file is required in order to perform the evaluation according to the main.py.

May I ask can I just use the pretrainining script and command under the UTKFace folder in order to reproduce the results?

Thanks

UBCDingXin commented 3 years ago

Hi @ZeWang95,

You can use the pre-trained AE for UTKFace to do the evaluation. https://github.com/UBCDingXin/improved_CcGAN/blob/master/UTKFace/pretrain_AE.py

UTKFace and UTKFace-improved are evaluated by using the same AE.