search

Mathux / TEMOS

Official PyTorch implementation of the paper "TEMOS: Generating diverse human motions from textual descriptions", ECCV 2022 (Oral)
https://mathis.petrovich.fr/temos/
MIT License
378 stars 25 forks source link

APE and AVE_pose metric calculation #13

Closed angelacast135 closed 2 years ago

angelacast135 commented 2 years ago

Hey, Thank you so much for sharing this project. I was trying to run the training code and saw what I think is a little bug here https://github.com/Mathux/TEMOS/blob/ea12cf6b22122aa5be95bbd75fcc374c0f42398a/temos/model/base.py#L45

Should it have a .mean()? dico.update({f"Metrics/{metric}": value.mean() for metric, value in metrics_dict.items()}) Otherwise, I get this error

image

Please let me know if I'm right or not :)

Thanks!

Mathux commented 2 years ago

Hello @angelacast135,

Thanks for the pointer, I just got the error myself when upgrading the packages. I just fixed this issue by removing the logging of the arrays (APE_pose, APE_joints, AVE_pose, AVE_joints) as it is not necessary (and now not supported apparently). I am already logging the means. It solves the problem.

Commit: https://github.com/Mathux/TEMOS/commit/40410d925600182fd20395ebe439c432f0e17fe4