HadoopIt rnn-nlu issues

HadoopIt / rnn-nlu

A TensorFlow implementation of Recurrent Neural Networks for Sequence Classification and Sequence Labeling

483 stars 171 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

how to use pre-train word embedding

#22 seeledu opened 6 years ago
0
load into tensorflow.js

#21 nalbion opened 6 years ago
0
License

#20 nalbion opened 6 years ago
0
update TF version to 1.9

#19 poyuwu opened 6 years ago
0
I can't find any code about the [red line] in figure3.

#18 ttslr opened 6 years ago
1
Slow spinup time

#17 anthnyprschka opened 6 years ago
0
May I Expect to see how I do prediction?

#16 jageshmaharjan opened 6 years ago
0
If you have so many addresses, will it cause a very big vocabulary? How to address it?

#15 wanghongling123 opened 6 years ago
0
Why not use dynamic rnn?

#14 Ailln opened 6 years ago
0
About the models.

#13 Peydon opened 6 years ago
1
need the dataset

#12 bringtree closed 6 years ago
0
when i run it in docker, it throws an err.

#11 bringtree closed 6 years ago
0
Restoring model from checkpoint

#10 dianamurgulet opened 7 years ago
7
How to reproduce the results of the paper?

#9 pfllo opened 7 years ago
9
some error about code

#8 billy800413 closed 7 years ago
1
Update with the recent changes in Tensorflow

#7 zuxfoucault closed 7 years ago
3
The logit in #L44 set shape from itself, fix this with zero_logit

#6 elias-1 closed 7 years ago
0
some error in the function _step of seq_labeling.py

#5 elias-1 closed 7 years ago
1
Linear is expecting 2D arguments: [[2, None, 128]] with num_layers=2

#4 D-X-Y closed 7 years ago
1
How to turn on proper bucket support?

#3 MtDersvan closed 7 years ago
1
how to impove predict result?

#2 liaicheng opened 7 years ago
9
Linear is expecting 2D arguments: [[2, None, 128]] with --bidirectionnal_rnn=True argument

#1 pldelisle closed 7 years ago
2