jzh

Bad result in reproducing horse2zebra

2

Thank you for sharing this projects. BUT I use the default setting of CUT, by simply run "python train.py", after 400 epoches, I use pytorch_fid to calculate FID, but I...

Result FID for horse2zebra

1

Thank you for your reproduce of CUT Do you calculate the FID for horse2zebra? My result is about 65, not as well as shown in the paper.

Teacher forcing on TIMIT and GRID dataset

2

Hi, I want to know how to set teacher forcing in GRID and TCDTIMIT dataset. The same as lip2wav dataset? teacher forcing decay from 29000 steps?

why we need this line here?

1

https://github.com/lucidrains/soundstorm-pytorch/blob/fbd83331b7051c26460cfbc7c54542c29e685eda/soundstorm_pytorch/soundstorm.py#L945

Something wrong in training Clevr

1

It is a really impressive work, however I meet the following problem, it seems that there is something with the size, but I just follow the README. nohup ./scripts/train_clevr.sh /home/hjz/hjzfolder/ReferringRelationships-master/checkpoints...