SuperYang

Results 7 comments of SuperYang

@Casi11as 你好,我这边是这样的, ![image](https://user-images.githubusercontent.com/28568741/141261694-d31a2d23-9bf3-4f70-804e-809571ae2d8f.png) 然后卡在这一步 ![image](https://user-images.githubusercontent.com/28568741/141261794-b6e9939c-2e8f-44b4-8f5a-d49d957cca5e.png) 请问你的shape参数也是?吗?为什么我这里不行呢

@Casi11as 我是在红线这一行 直接退出程序了,不反馈信息 ![image](https://user-images.githubusercontent.com/28568741/141263469-87b08cae-f0e1-41b9-b233-48ec2a3dc13c.png)

不是,我是因为启动训练脚本过一会就报错进程已退出,然后一点点打断点到这里,这一步走不通,不确定是不是数据的问题,所以我在重新生成tfrecord,因此现在没法演示,生成完我明天给你演示一下,感谢回复。

> hi, shouldn't the data be like ./Data/Train/Fear/013736360/ 1.jpg 2.jpg 3.jpg (these are the frames). > > I have starred. how u convert this video into img? frame by frame?

> You can use ffmpeg our Opencv to convert video into image. Many instructions online. Give me a star, thanks. @LeiiY so after convert to img,should i detect faces and...

> I think you might be using Windows -- is this issue related? [pytorch/pytorch#18384](https://github.com/pytorch/pytorch/issues/18384) yeah,windows10, so ur system version?Ubuntu 16.04 or other

download the checkpoint wget to report CCC on the validation set python3 eval.py --test_on_val --checkpoint m3t_mtl-vox2.pt python3 get_smoothed_ccc predictions_val.pt to generate test set predictions python3 eval.py --checkpoint m3t_mtl-vox2.pt the weget...