search

tr3e / InterGen

[IJCV 2024] InterGen: Diffusion-based Multi-human Motion Generation under Complex Interactions
https://tr3e.github.io/intergen-page/
192 stars 10 forks source link

Missing data in motion_processed #4

Open zhenzhiwang opened 1 year ago

zhenzhiwang commented 1 year ago

Hi, thanks for sharing your great work and your code!

When I run the training scripts, interhuman.py reports a lot of '.npy' file missing in person2 folder. I also find a lot of xxxx(1).npy in both person1 and person2 folder where xxxx is numbers, which seems to be errors related to the pre-processing scripts to converting original '.pkl' files to be '.npy' files. Could you fix this issue or could you provide the scripts for converting the data in "motions" to be the format in "motions_processd"?

Looking for your reply! Thanks.

tr3e commented 1 year ago

Sorry for that. This is a Google Drive error and I am resubmitting them as soon as possible.

zhenzhiwang commented 1 year ago

Thanks for the quick reply! By the way, I am wondering whether this network really need 2000 epochs to converge? Although training one epoch is fast, 2000 epochs still seems to be too large. Did you try a larger lr and fewer epochs with a good lr drop strategy (e.g., the cosine annealing in the code but with other proper parameters)?