PyTorch code for training MM-DistillNet for multimodal knowledge distillation. http://rl.uni-freiburg.de/research/multimodal-distill
GNU General Public License v3.0
59
stars
14
forks
source link
Source codes and datasets are missing #1
Closed
as4mz closed 3 years ago
Hi,
I have tried to run the following code:
python train.py --config ./configs
Then, got the following error: ModuleNotFoundError: No module named 'src.fullcnn_net'.
After checking the utils.py, I feel the followings source files are missing:
src.fullcnn_net src.loss.ABLoss src.loss.MTALoss src.loss.KLLoss src.loss.GroupAttentionLoss src.loss.MultiTeacherPairWiseSimilarityLoss src.loss.MultiTeacherPairWiseSimilarityLoss src.loss.MultiTeacherContrastiveAttentionLoss src.loss. MultiTeacherContrastiveAttentionLoss src.loss.MultiTeacherTrippletAttentionLoss src.loss. MultiTeacherTrippletAttentionLoss src.loss.CRDLoss import CRDLoss src.loss.NSTLoss src.loss.PKTLoss src.loss.SimilarityLoss src.loss.RankingLoss
Also, I am not sure how to download/access the datasets. I do not see any binary files downloaded inside the folder.
Thanks for your help.