bks-abs

from I know, the "seq2seq" model output is variable, but when i print the pred_out shape, I found that all the predict out is (batch, 11, 3000), can u help...

hi, I love what you do and open source work. and I have a question about the loss function. I read your paper, the paper said loss function is MFE...

can anybody share the predict time cost?

can u provide the InsightFace Project.

1

大佬，我从csdn看到到到， 可以开源下 InsightFace这个么， 看你博客说这个效果比较好（可能是facenet不太适用亚洲人脸=_=）

感谢大佬无私的分享， 我看到w&d模型，但是没有对应数据集， 可以分享下么，~~

what does num_frame_per_clib mean in train.py and test.py

2

I noticed that the values of num_frame_per_clib parameters in training and prediction are different. The example is RGB flow: why num_frame_per_clib is 16 when predicted, and when training is 64....

if we train from scratch on kinetics-600 dataset to get a model. how long it cost? can anybody give a reference?