ddlau

您好，非常感谢回复和指导！有一些细节想请教一下： 目前我这边测试用的数据，仅有一个含15万个顶点和5万个面片的clothed的mesh；在调用optimize_offsets时这个mesh应该是用来初始化all_posed_cloth_meshes吧？那init_smpl_meshes该怎么办呢？可以用这个mesh算吗？ 再次感谢！

@googlebot I signed it!

This is a Categorical-DQN legacy, which was just "transposed" in QR-DQN. In contrast to Categorical-DQN, QR-DQN set fixed probabilites and learnable locations (a.k.a diracs).

> rightly Yeah, I noticed it too, after "uptraced" to its caller, and found the strange thing is not used. To be a little strict, isn't this sort of impreciseness...