LOOKCC

@KeremTurgutlu I found the same question when I want to train CLIP with 8 GPUs. I found that if my total batch size is 128, then pytorch will split it...

@KeremTurgutlu It's no doubt that using queues is a good way if whose GPUs are not very good. I know the way contrastive learning(such as MoCo) do. If train CLIP...

you can find the model_deploy.py which used here 'from deployment import model_deploy' in this link : https://github.com/tensorflow/models/blob/master/research/slim/deployment/model_deploy.py @twinsyssy1018

我安装了mathjax插件，配饰了站点文件，在主题配置文件加入了cdn，但是并没有渲染出来公式。

@Helianus 我在我的网站F12，在js控制台发现是我找的cdn不能用，我去了mathjax的官网手册，这里： http://docs.mathjax.org/en/latest/start.html 按照推荐的换了cdn，就可以用了。谢了

我在网上找了一系列视频测试，非常多的real被识别成了fake，感觉误报率有点高

My results on c23 are very similar to yours. Reporting the results of c23 in the paper are the consensus in the field of Deepfake Detection, and the author did...

> Have you ever run the code on C40 and get any results? I did some experiments on c23 and the results were not very satisfying to me.