Skip to content

Latest commit

 

History

History
 
 

neural-machine-translation

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

How-to

  1. Run any notebook using Jupyter Notebook.

Accuracy, not sorted

Based on 20 epochs accuracy. The results will be different on different dataset. Trained on a GTX 960, 4GB VRAM.

name accuracy
1.basic-seq2seq-manual 0.915255
2.lstm-seq2seq-manual 0.917009
3.gru-seq2seq-manual 0.920200
4.basic-seq2seq-api-greedy 0.960998
5.lstm-seq2seq-api-greedy 0.202590
6.gru-seq2seq-greedy 0.408099
7.basic-birnn-seq2seq-manual 0.919491
8.lstm-birnn-seq2seq-manual 0.918473
9.gru-birnn-seq2seq-manual 0.922818
10.basic-birnn-seq2seq-greedy 0.957355
11.lstm-birnn-seq2seq-greedy 0.202628
12.gru-birnn-seq2seq-greedy 0.484461
13.basic-seq2seq-luong 0.916100
14.lstm-seq2seq-luong 0.917736
15.gru-seq2seq-luong 0.919482
16.basic-seq2seq-bahdanau 0.915700
17.lstm-seq2seq-bahdanau 0.721833
18.gru-seq2seq-bahdanau 0.919218
19.lstm-birnn-seq2seq-luong 0.918555
20.gru-birnn-seq2seq-luong 0.919445
21.lstm-birnn-seq2seq-bahdanau 0.917655
22.gru-birnn-seq2seq-bahdanau 0.920555
23.lstm-birnn-seq2seq-bahdanau-luong 0.918182
24.gru-birnn-seq2seq-bahdanau-luong 0.920045
25.lstm-seq2seq-greedy-luong 0.364322
26.gru-seq2seq-greedy-luong 0.627814
27.lstm-seq2seq-greedy-bahdanau 0.378199
28.gru-seq2seq-greedy-bahdanau 0.470696
29.lstm-seq2seq-beam 0.122135
30.gru-seq2seq-beam 0.163046
31.lstm-birnn-seq2seq-beam-luong 0.171741
32.gru-birnn-seq2seq-beam-luong 0.189787
33.lstm-birnn-seq2seq-luong-bahdanau-stack-beam 0.098961
34.gru-birnn-seq2seq-luong-bahdanau-stack-beam 0.091473
35.byte-net 1.022409
36.estimator
37.capsule-lstm-seq2seq-greedy
38.capsule-lstm-seq2seq-luong-beam
39.lstm-birnn-seq2seq-luong-bahdanau-stack-beam-dropout-l2 0.066305
40.dnc-seq2seq-bahdanau-greedy 0.711184
41.lstm-birnn-seq2seq-beam-luongmonotic 0.624756
42.lstm-birnn-seq2seq-beam-bahdanaumonotic 0.624756
43.memory-network-basic 0.965700
44.memory-network-lstm 0.942591
45.attention-is-all-you-need 0.170279
46.transformer-xl 0.114907
47.attention-is-all-you-need-beam-search 0.158205
48.conv-encoder-conv-decoder 0.462655
49.conv-encoder-lstm 0.438702