Justin1904 Low-rank-Multimodal-Fusion issues

Justin1904 / Low-rank-Multimodal-Fusion

This is the repository for "Efficient Low-rank Multimodal Fusion with Modality-Specific Factors", Liu and Shen, et. al. ACL 2018

254 stars 47 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

The dataset link is invalid.

#27 fhcgdtxnbhc opened 5 months ago
0
IEMOCAP labels

#26 calvin0205 opened 1 year ago
0
when I run "train_mosi.py", there is the below results with "UserWarning: dropout option adds dropout after all but last recurrent layer, so non-zero dropout expects num_layers greater than 1, but got dropout=0.3 and num_layers=1 "num_layers={}".format(dropout, num_layers))"

#25 xiajili opened 2 years ago
1
iemocap dataset labels

#24 padisarala opened 2 years ago
0
Can not pickle data

#23 Weiyinyin11 opened 2 years ago
1
IEMOCAP datasets question

#22 yuanjiechen opened 3 years ago
0
iemocap no key

#21 songyh111 opened 3 years ago
2
Some questions about mosi dataset.

#20 Junge718 opened 3 years ago
1
i can't save model on colab

#19 komalanadkat opened 3 years ago
1
a question about train_pom.py

#18 qq291078365 opened 3 years ago
2
A question about formula (6)

#17 oyss1996 closed 3 years ago
1
AttributeError

#16 frallisland closed 3 years ago
3
Cannot pickle dataset?

#15 252057052 opened 4 years ago
9
is the implementation consistent with the math?

#14 shalijiang closed 3 years ago
1
what is z hat Equation (8)

#13 shalijiang closed 3 years ago
1
can we understand this method as a generalization of factorization machine?

#12 shalijiang closed 4 years ago
1
About the labels of IEMOCAP

#11 yuki9965 closed 4 years ago
5
Training got into NaN values...

#10 hbzhang closed 4 years ago
2
about the learning rate

#9 srtianxia closed 4 years ago
4
Epoch 0 complete! Average Training loss: nan

#8 xiaoiker closed 5 years ago
4
Used for classification or to generate a tensor that reduces the amount of calculation？

#7 Paidalin closed 5 years ago
4
can you share the train/val/test video name in each dataset?

#6 hugddygff closed 5 years ago
8
How many times did you use the 8895952 combination? How much time does each data set need to train?

#5 Paidalin closed 5 years ago
4
about the original face features and audio features

#4 hugddygff closed 5 years ago
3
Dimensions of data

#3 kushagrajain11 closed 6 years ago
2
Update README.md

#2 Justin1904 closed 6 years ago
0
Best hyper-parameters settings

#1 lan-qing closed 5 years ago
4