storyteller comments

Repositories
Issues
Comments

Results 4 comments of


                                            storyteller

question about run python main.py demo_offline

> @KrisLee512 you should fix /home/llb/codes/ST_source/st-gcn/processor/demo_offline.py", line 137 :opWrapper.emplaceAndPop([datum]) is opWrapper.emplaceAndPop(op.VectorDatum([datum])). thanks，it work for me.

Nan

我换成involution，结果参数好像不能进行优化。train loss一直下降，但是val loss一直保持一个值没变。有大佬知道这是不是过拟合造成的，还是代码错误。我感觉不是过拟，因为train loss下降，val loss基本没变。还没有解决这个问题

Nan

> The loss of mine in the training set is fine, while in cv set, some batches are nan. > It's definitely not gradient explosion. I don't know how to...

AMASS Evaluation

hi I should choose that data to download, AMASS 1. SMPL+H G 2. SMPL+x G 3. SMPL+x N 4. Render https://amass.is.tue.mpg.de/download.php I should choose