YZYLHR

> 为什么这个输出通道的conv的多个kernel_size操作？ >  Do you know why?

yes！You can use openpose. py to gendata. json file for each video. Then you can use kinetics_gendata.py to get the. npy file.  ---Original--- From: "Akhil"

Hi! Maybe your dataset is too small.You can try using the pre-training model. ------------------ 原始邮件 ------------------ 发件人: "luoluo-gif"