issues
search
Sachin19
/
seq2seq-con
Implementation of "Von Mises-Fisher Loss for Training Sequence to Sequence Models with Continuous Outputs"
Other
76
stars
19
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Got strange result while training translation from zh to en.
#12
EuphoriaYan
opened
3 years ago
3
Hyper-parameters to train word2vec/FastText
#11
toshohirasawa
closed
3 years ago
1
Why don't you experiment VMF loss on Transformer?
#10
HuihuiChyan
closed
3 years ago
1
weird value of LogCmk
#9
xxchauncey
closed
4 years ago
13
Hyperparameter in NLL-VMF loss
#8
kingofspace0wzz
closed
5 years ago
1
error running translate.py
#7
yiulau
opened
5 years ago
5
pickle error when saving checkpoints
#6
yiulau
closed
5 years ago
2
Regarding the training speed of the model
#5
kalyangvs
closed
5 years ago
2
function returns nothing
#4
serjtroshin
closed
4 years ago
3
How to use evaluate.sh?
#3
jianqiaol
closed
5 years ago
2
VMF loss problems
#2
Islanna
closed
5 years ago
1
Extra term in `LogCmk` forward
#1
amanjitsk
closed
4 years ago
10