jweihe comments

Results 15 comments of


                                            jweihe

finetune experience or code

> I have finetuned CLIP on PASCAL VOC2012 dataset using this [implementation](https://github.com/openai/CLIP/issues/83). When I changed the weight decay from 0.2 to 0.001, it works well on my task. Could you...

finetune experience or code

> My code was same with [the implementation](https://github.com/openai/CLIP/issues/83) ok， thank u

使用StyleGAN训练自己的数据集.md | 数据与算法

good job

Would you be able to share the training params for style space latent mapper?

hi，gays~ have you found why this happened？i met same problem

stylegan是否可以进行人脸正面化的操作

似乎并不可以 clip对人脸姿态不敏感

stylegan是否可以进行人脸正面化的操作

我试了一下 source是side face，target是ID photo，但是得到的结果并不对

关于test.py中的entityVector.txt文件

这个是程序运行生成的放在哪里都可以

pytorch_lightning.utilities.exceptions.MisconfigurationException: No `test_dataloader()` method defined to run `Trainer.test`

hi bro,I think you can try to use '--no_test'

How to train dataset with node feature ?

@dongZheX @ahxt Have you resolved it? I'm encountering the same issue.

[BUG] <title>Lora训练的时候想同时打开其他部分进行非lora训练，该怎么做

最后用modules_to_save实现的