Skip to content

Instantly share code, notes, and snippets.

@kylegao91
Created July 18, 2017 13:59
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save kylegao91/53dea90b3f3572a28318d8eb72d4ec8d to your computer and use it in GitHub Desktop.
Save kylegao91/53dea90b3f3572a28318d8eb72d4ec8d to your computer and use it in GitHub Desktop.
OpenNMT-py v.s. pytorch-seq2seq on newstest2013
Namespace(batch_size=64, brnn=False, brnn_merge='concat', curriculum=False, data='data/demo.train.pt', dropout=0, encoder_type='text', epochs=50, extra_shuffle=False, gpus=[0], input_feed=1, layers=1, learning_rate=1.0, learning_rate_decay=0.5, log_interval=10, max_generator_batches=32, max_grad_norm=5, optim='sgd', param_init=0.1, pre_word_vecs_dec=None, pre_word_vecs_enc=None, rnn_size=128, save_model='data/demo-model', start_decay_at=8, start_epoch=1, train_from='', train_from_state_dict='', word_vec_size=128)
Loading data from 'data/demo.train.pt'
* vocabulary size. source = 15832; target = 15832
* number of training sentences. 2959
* maximum batch size. 64
Building model...
* number of parameters: 6474200
NMTModel (
(encoder): Encoder (
(word_lut): Embedding(15832, 128, padding_idx=0)
(rnn): LSTM(128, 128)
)
(decoder): Decoder (
(word_lut): Embedding(15832, 128, padding_idx=0)
(rnn): StackedLSTM (
(dropout): Dropout (p = 0)
(layers): ModuleList (
(0): LSTMCell(256, 128)
)
)
(attn): GlobalAttention (
(linear_in): Linear (128 -> 128)
(sm): Softmax ()
(linear_out): Linear (256 -> 128)
(tanh): Tanh ()
)
(dropout): Dropout (p = 0)
)
(generator): Sequential (
(0): Linear (128 -> 15832)
(1): LogSoftmax ()
)
)
Epoch 1, 10/ 47; acc: 3.11; ppl: 59733.06;9581 src tok/s; 10063 tgt tok/s; 1 s elapsed
Epoch 1, 20/ 47; acc: 4.01; ppl: 28122.81;15531 src tok/s; 16647 tgt tok/s; 2 s elapsed
Epoch 1, 30/ 47; acc: 4.77; ppl: 26965.76;15650 src tok/s; 16761 tgt tok/s; 2 s elapsed
Epoch 1, 40/ 47; acc: 2.50; ppl: 12775.80;16428 src tok/s; 17271 tgt tok/s; 3 s elapsed
Train perplexity: 30250.4
Train accuracy: 3.43708
Validation perplexity: 16644
Validation accuracy: 2.67597
Epoch 2, 10/ 47; acc: 3.60; ppl: 11081.88;14039 src tok/s; 14808 tgt tok/s; 6 s elapsed
Epoch 2, 20/ 47; acc: 3.09; ppl: 21771.08;15707 src tok/s; 16758 tgt tok/s; 7 s elapsed
Epoch 2, 30/ 47; acc: 4.14; ppl: 3929.06;15590 src tok/s; 16767 tgt tok/s; 7 s elapsed
Epoch 2, 40/ 47; acc: 3.46; ppl: 3466.95;16768 src tok/s; 17592 tgt tok/s; 8 s elapsed
Train perplexity: 6278.18
Train accuracy: 3.5586
Validation perplexity: 66325.3
Validation accuracy: 5.25052
Decaying learning rate to 0.5
Epoch 3, 10/ 47; acc: 4.13; ppl: 2454.30;16851 src tok/s; 17679 tgt tok/s; 11 s elapsed
Epoch 3, 20/ 47; acc: 4.90; ppl: 2171.04;13929 src tok/s; 14700 tgt tok/s; 12 s elapsed
Epoch 3, 30/ 47; acc: 5.62; ppl: 2464.46;15769 src tok/s; 16664 tgt tok/s; 13 s elapsed
Epoch 3, 40/ 47; acc: 4.66; ppl: 2021.92;15959 src tok/s; 17040 tgt tok/s; 13 s elapsed
Train perplexity: 2256.1
Train accuracy: 4.80466
Validation perplexity: 25163.6
Validation accuracy: 0
Decaying learning rate to 0.25
Epoch 4, 10/ 47; acc: 5.12; ppl: 1814.99;12971 src tok/s; 13874 tgt tok/s; 16 s elapsed
Epoch 4, 20/ 47; acc: 4.99; ppl: 1902.57;16457 src tok/s; 17338 tgt tok/s; 17 s elapsed
Epoch 4, 30/ 47; acc: 5.46; ppl: 1868.35;16318 src tok/s; 17161 tgt tok/s; 17 s elapsed
Epoch 4, 40/ 47; acc: 5.35; ppl: 1780.79;16220 src tok/s; 17144 tgt tok/s; 18 s elapsed
Train perplexity: 1843.93
Train accuracy: 5.35604
Validation perplexity: 20548.4
Validation accuracy: 5.14198
Decaying learning rate to 0.125
Epoch 5, 10/ 47; acc: 6.81; ppl: 1614.78;16113 src tok/s; 17088 tgt tok/s; 21 s elapsed
Epoch 5, 20/ 47; acc: 6.17; ppl: 1713.93;16183 src tok/s; 16983 tgt tok/s; 22 s elapsed
Epoch 5, 30/ 47; acc: 6.91; ppl: 1702.43;15843 src tok/s; 16713 tgt tok/s; 22 s elapsed
Epoch 5, 40/ 47; acc: 5.53; ppl: 1627.52;16242 src tok/s; 17405 tgt tok/s; 23 s elapsed
Train perplexity: 1684.91
Train accuracy: 6.2067
Validation perplexity: 22562.8
Validation accuracy: 4.64202
Decaying learning rate to 0.0625
Epoch 6, 10/ 47; acc: 6.32; ppl: 1616.80;16268 src tok/s; 17116 tgt tok/s; 26 s elapsed
Epoch 6, 20/ 47; acc: 7.66; ppl: 1500.72;16497 src tok/s; 17547 tgt tok/s; 27 s elapsed
Epoch 6, 30/ 47; acc: 6.00; ppl: 1802.97;15678 src tok/s; 16483 tgt tok/s; 27 s elapsed
Epoch 6, 40/ 47; acc: 6.22; ppl: 1608.61;16836 src tok/s; 17762 tgt tok/s; 28 s elapsed
Train perplexity: 1630.56
Train accuracy: 6.60754
Validation perplexity: 24598.2
Validation accuracy: 4.80571
Decaying learning rate to 0.03125
Epoch 7, 10/ 47; acc: 9.18; ppl: 1414.71;15970 src tok/s; 17222 tgt tok/s; 31 s elapsed
Epoch 7, 20/ 47; acc: 6.88; ppl: 1507.67;13155 src tok/s; 14013 tgt tok/s; 31 s elapsed
Epoch 7, 30/ 47; acc: 7.68; ppl: 1556.39;16025 src tok/s; 17002 tgt tok/s; 32 s elapsed
Epoch 7, 40/ 47; acc: 5.40; ppl: 1732.87;16882 src tok/s; 17529 tgt tok/s; 33 s elapsed
Train perplexity: 1581.12
Train accuracy: 7.0247
Validation perplexity: 24190.1
Validation accuracy: 4.60288
Decaying learning rate to 0.015625
Epoch 8, 10/ 47; acc: 7.49; ppl: 1581.48;16195 src tok/s; 17143 tgt tok/s; 36 s elapsed
Epoch 8, 20/ 47; acc: 7.20; ppl: 1549.01;16283 src tok/s; 17223 tgt tok/s; 36 s elapsed
Epoch 8, 30/ 47; acc: 5.89; ppl: 1699.87;13809 src tok/s; 14405 tgt tok/s; 37 s elapsed
Epoch 8, 40/ 47; acc: 6.74; ppl: 1509.28;16262 src tok/s; 17324 tgt tok/s; 38 s elapsed
Train perplexity: 1567.37
Train accuracy: 6.86872
Validation perplexity: 24866
Validation accuracy: 4.88399
Decaying learning rate to 0.0078125
Epoch 9, 10/ 47; acc: 6.83; ppl: 1556.39;15696 src tok/s; 16652 tgt tok/s; 41 s elapsed
Epoch 9, 20/ 47; acc: 6.23; ppl: 1550.88;16496 src tok/s; 17245 tgt tok/s; 42 s elapsed
Epoch 9, 30/ 47; acc: 7.25; ppl: 1568.98;13660 src tok/s; 14507 tgt tok/s; 42 s elapsed
Epoch 9, 40/ 47; acc: 7.52; ppl: 1512.56;16560 src tok/s; 17533 tgt tok/s; 43 s elapsed
Train perplexity: 1556
Train accuracy: 6.86328
Validation perplexity: 24765.4
Validation accuracy: 4.9089
Decaying learning rate to 0.00390625
Epoch 10, 10/ 47; acc: 6.96; ppl: 1532.61;15936 src tok/s; 17011 tgt tok/s; 46 s elapsed
Epoch 10, 20/ 47; acc: 6.90; ppl: 1564.03;16056 src tok/s; 16913 tgt tok/s; 46 s elapsed
Epoch 10, 30/ 47; acc: 6.72; ppl: 1565.30;16351 src tok/s; 17234 tgt tok/s; 47 s elapsed
Epoch 10, 40/ 47; acc: 7.15; ppl: 1569.59;16151 src tok/s; 17051 tgt tok/s; 48 s elapsed
Train perplexity: 1551.4
Train accuracy: 6.97392
Validation perplexity: 25091
Validation accuracy: 4.81282
Decaying learning rate to 0.00195312
Epoch 11, 10/ 47; acc: 7.66; ppl: 1471.31;17082 src tok/s; 18059 tgt tok/s; 51 s elapsed
Epoch 11, 20/ 47; acc: 7.26; ppl: 1495.86;16954 src tok/s; 17927 tgt tok/s; 51 s elapsed
Epoch 11, 30/ 47; acc: 6.44; ppl: 1638.06;15264 src tok/s; 16177 tgt tok/s; 52 s elapsed
Epoch 11, 40/ 47; acc: 8.17; ppl: 1417.79;15437 src tok/s; 16552 tgt tok/s; 52 s elapsed
Train perplexity: 1548.65
Train accuracy: 6.96122
Validation perplexity: 24977.9
Validation accuracy: 4.81282
Decaying learning rate to 0.000976562
Epoch 12, 10/ 47; acc: 6.80; ppl: 1502.19;15297 src tok/s; 16396 tgt tok/s; 55 s elapsed
Epoch 12, 20/ 47; acc: 7.36; ppl: 1534.94;16918 src tok/s; 17826 tgt tok/s; 56 s elapsed
Epoch 12, 30/ 47; acc: 7.99; ppl: 1488.45;15980 src tok/s; 16960 tgt tok/s; 57 s elapsed
Epoch 12, 40/ 47; acc: 6.13; ppl: 1596.41;14290 src tok/s; 14941 tgt tok/s; 58 s elapsed
Train perplexity: 1547.37
Train accuracy: 6.97936
Validation perplexity: 24885.7
Validation accuracy: 4.80927
Decaying learning rate to 0.000488281
Epoch 13, 10/ 47; acc: 6.81; ppl: 1515.86;15948 src tok/s; 16964 tgt tok/s; 60 s elapsed
Epoch 13, 20/ 47; acc: 6.73; ppl: 1519.07;13744 src tok/s; 14524 tgt tok/s; 61 s elapsed
Epoch 13, 30/ 47; acc: 7.13; ppl: 1593.96;16005 src tok/s; 16873 tgt tok/s; 62 s elapsed
Epoch 13, 40/ 47; acc: 7.23; ppl: 1520.57;16853 src tok/s; 17736 tgt tok/s; 63 s elapsed
Train perplexity: 1546.66
Train accuracy: 6.98843
Validation perplexity: 24939.8
Validation accuracy: 4.81282
Decaying learning rate to 0.000244141
Epoch 14, 10/ 47; acc: 7.11; ppl: 1483.49;15979 src tok/s; 16988 tgt tok/s; 65 s elapsed
Epoch 14, 20/ 47; acc: 5.97; ppl: 1686.96;16001 src tok/s; 16721 tgt tok/s; 66 s elapsed
Epoch 14, 30/ 47; acc: 6.70; ppl: 1551.63;14256 src tok/s; 14949 tgt tok/s; 67 s elapsed
Epoch 14, 40/ 47; acc: 8.87; ppl: 1420.19;15897 src tok/s; 17118 tgt tok/s; 68 s elapsed
Train perplexity: 1546.33
Train accuracy: 6.98299
Validation perplexity: 24954.8
Validation accuracy: 4.8146
Decaying learning rate to 0.00012207
Epoch 15, 10/ 47; acc: 9.45; ppl: 1348.87;15331 src tok/s; 16667 tgt tok/s; 70 s elapsed
Epoch 15, 20/ 47; acc: 6.68; ppl: 1506.52;12999 src tok/s; 13829 tgt tok/s; 71 s elapsed
Epoch 15, 30/ 47; acc: 6.93; ppl: 1613.50;16229 src tok/s; 17124 tgt tok/s; 72 s elapsed
Epoch 15, 40/ 47; acc: 5.98; ppl: 1629.69;16999 src tok/s; 17713 tgt tok/s; 72 s elapsed
Train perplexity: 1546.17
Train accuracy: 6.97755
Validation perplexity: 24958.7
Validation accuracy: 4.8146
Decaying learning rate to 6.10352e-05
Epoch 16, 10/ 47; acc: 6.85; ppl: 1581.12;15770 src tok/s; 16754 tgt tok/s; 75 s elapsed
Epoch 16, 20/ 47; acc: 7.89; ppl: 1459.89;15421 src tok/s; 16547 tgt tok/s; 76 s elapsed
Epoch 16, 30/ 47; acc: 6.02; ppl: 1584.08;14348 src tok/s; 15022 tgt tok/s; 77 s elapsed
Epoch 16, 40/ 47; acc: 8.22; ppl: 1428.70;16807 src tok/s; 17819 tgt tok/s; 77 s elapsed
Train perplexity: 1546.08
Train accuracy: 6.97936
Validation perplexity: 24960.7
Validation accuracy: 4.8146
Decaying learning rate to 3.05176e-05
Epoch 17, 10/ 47; acc: 6.50; ppl: 1599.80;15710 src tok/s; 16631 tgt tok/s; 80 s elapsed
Epoch 17, 20/ 47; acc: 6.27; ppl: 1649.77;13956 src tok/s; 14604 tgt tok/s; 81 s elapsed
Epoch 17, 30/ 47; acc: 7.44; ppl: 1476.45;16734 src tok/s; 17757 tgt tok/s; 82 s elapsed
Epoch 17, 40/ 47; acc: 7.42; ppl: 1487.92;16655 src tok/s; 17566 tgt tok/s; 82 s elapsed
Train perplexity: 1546.04
Train accuracy: 6.98117
Validation perplexity: 24961.6
Validation accuracy: 4.8146
Decaying learning rate to 1.52588e-05
Epoch 18, 10/ 47; acc: 5.84; ppl: 1611.19;15893 src tok/s; 16676 tgt tok/s; 85 s elapsed
Epoch 18, 20/ 47; acc: 6.38; ppl: 1669.62;16398 src tok/s; 17164 tgt tok/s; 86 s elapsed
Epoch 18, 30/ 47; acc: 8.33; ppl: 1456.01;16389 src tok/s; 17497 tgt tok/s; 87 s elapsed
Epoch 18, 40/ 47; acc: 8.79; ppl: 1409.50;16482 src tok/s; 17624 tgt tok/s; 87 s elapsed
Train perplexity: 1546.02
Train accuracy: 6.97755
Validation perplexity: 24962
Validation accuracy: 4.8146
Decaying learning rate to 7.62939e-06
Epoch 19, 10/ 47; acc: 6.65; ppl: 1606.52;16586 src tok/s; 17399 tgt tok/s; 90 s elapsed
Epoch 19, 20/ 47; acc: 6.37; ppl: 1544.08;16589 src tok/s; 17447 tgt tok/s; 91 s elapsed
Epoch 19, 30/ 47; acc: 9.22; ppl: 1406.84;16262 src tok/s; 17356 tgt tok/s; 92 s elapsed
Epoch 19, 40/ 47; acc: 5.90; ppl: 1666.93;15629 src tok/s; 16415 tgt tok/s; 92 s elapsed
Train perplexity: 1546.01
Train accuracy: 6.97936
Validation perplexity: 24962.2
Validation accuracy: 4.8146
Decaying learning rate to 3.8147e-06
Epoch 20, 10/ 47; acc: 6.36; ppl: 1618.63;13975 src tok/s; 14696 tgt tok/s; 95 s elapsed
Epoch 20, 20/ 47; acc: 8.27; ppl: 1341.78;15632 src tok/s; 16986 tgt tok/s; 96 s elapsed
Epoch 20, 30/ 47; acc: 7.08; ppl: 1600.65;16242 src tok/s; 17115 tgt tok/s; 96 s elapsed
Epoch 20, 40/ 47; acc: 6.21; ppl: 1648.74;16548 src tok/s; 17317 tgt tok/s; 97 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.3
Validation accuracy: 4.8146
Decaying learning rate to 1.90735e-06
Epoch 21, 10/ 47; acc: 8.10; ppl: 1430.21;15987 src tok/s; 17094 tgt tok/s; 100 s elapsed
Epoch 21, 20/ 47; acc: 5.10; ppl: 1706.06;15972 src tok/s; 16613 tgt tok/s; 101 s elapsed
Epoch 21, 30/ 47; acc: 6.43; ppl: 1589.48;14537 src tok/s; 15215 tgt tok/s; 102 s elapsed
Epoch 21, 40/ 47; acc: 8.71; ppl: 1441.91;15637 src tok/s; 16824 tgt tok/s; 102 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.3
Validation accuracy: 4.8146
Decaying learning rate to 9.53674e-07
Epoch 22, 10/ 47; acc: 7.16; ppl: 1506.50;16679 src tok/s; 17639 tgt tok/s; 105 s elapsed
Epoch 22, 20/ 47; acc: 6.27; ppl: 1631.51;13942 src tok/s; 14621 tgt tok/s; 106 s elapsed
Epoch 22, 30/ 47; acc: 9.10; ppl: 1381.02;16053 src tok/s; 17173 tgt tok/s; 106 s elapsed
Epoch 22, 40/ 47; acc: 6.52; ppl: 1573.50;15404 src tok/s; 16310 tgt tok/s; 107 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.3
Validation accuracy: 4.8146
Decaying learning rate to 4.76837e-07
Epoch 23, 10/ 47; acc: 8.52; ppl: 1468.01;15966 src tok/s; 17087 tgt tok/s; 110 s elapsed
Epoch 23, 20/ 47; acc: 6.99; ppl: 1474.39;13375 src tok/s; 14211 tgt tok/s; 110 s elapsed
Epoch 23, 30/ 47; acc: 7.59; ppl: 1558.33;15398 src tok/s; 16381 tgt tok/s; 111 s elapsed
Epoch 23, 40/ 47; acc: 5.80; ppl: 1672.05;16257 src tok/s; 16929 tgt tok/s; 112 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 2.38419e-07
Epoch 24, 10/ 47; acc: 7.26; ppl: 1494.87;15897 src tok/s; 16852 tgt tok/s; 115 s elapsed
Epoch 24, 20/ 47; acc: 5.76; ppl: 1625.41;16330 src tok/s; 17077 tgt tok/s; 116 s elapsed
Epoch 24, 30/ 47; acc: 7.42; ppl: 1470.34;16470 src tok/s; 17479 tgt tok/s; 116 s elapsed
Epoch 24, 40/ 47; acc: 6.33; ppl: 1637.63;13325 src tok/s; 14048 tgt tok/s; 117 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 1.19209e-07
Epoch 25, 10/ 47; acc: 6.48; ppl: 1596.36;16124 src tok/s; 16927 tgt tok/s; 120 s elapsed
Epoch 25, 20/ 47; acc: 8.10; ppl: 1478.57;16224 src tok/s; 17249 tgt tok/s; 121 s elapsed
Epoch 25, 30/ 47; acc: 7.38; ppl: 1415.22;12613 src tok/s; 13603 tgt tok/s; 121 s elapsed
Epoch 25, 40/ 47; acc: 7.49; ppl: 1525.14;16007 src tok/s; 16941 tgt tok/s; 122 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 5.96046e-08
Epoch 26, 10/ 47; acc: 6.84; ppl: 1563.25;16658 src tok/s; 17491 tgt tok/s; 125 s elapsed
Epoch 26, 20/ 47; acc: 7.37; ppl: 1574.94;15613 src tok/s; 16479 tgt tok/s; 126 s elapsed
Epoch 26, 30/ 47; acc: 6.33; ppl: 1606.50;16328 src tok/s; 17093 tgt tok/s; 126 s elapsed
Epoch 26, 40/ 47; acc: 7.76; ppl: 1478.63;15471 src tok/s; 16539 tgt tok/s; 127 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 2.98023e-08
Epoch 27, 10/ 47; acc: 6.22; ppl: 1619.13;14418 src tok/s; 15061 tgt tok/s; 130 s elapsed
Epoch 27, 20/ 47; acc: 8.93; ppl: 1409.69;15351 src tok/s; 16686 tgt tok/s; 131 s elapsed
Epoch 27, 30/ 47; acc: 8.14; ppl: 1447.18;16414 src tok/s; 17436 tgt tok/s; 131 s elapsed
Epoch 27, 40/ 47; acc: 6.12; ppl: 1601.19;16065 src tok/s; 16859 tgt tok/s; 132 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 1.49012e-08
Epoch 28, 10/ 47; acc: 6.88; ppl: 1451.11;15587 src tok/s; 16682 tgt tok/s; 135 s elapsed
Epoch 28, 20/ 47; acc: 7.35; ppl: 1519.48;14154 src tok/s; 14924 tgt tok/s; 135 s elapsed
Epoch 28, 30/ 47; acc: 6.74; ppl: 1535.54;16643 src tok/s; 17497 tgt tok/s; 136 s elapsed
Epoch 28, 40/ 47; acc: 6.57; ppl: 1653.25;15574 src tok/s; 16445 tgt tok/s; 137 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 7.45058e-09
Epoch 29, 10/ 47; acc: 7.80; ppl: 1489.01;16000 src tok/s; 16982 tgt tok/s; 140 s elapsed
Epoch 29, 20/ 47; acc: 7.11; ppl: 1535.23;16951 src tok/s; 17847 tgt tok/s; 140 s elapsed
Epoch 29, 30/ 47; acc: 6.07; ppl: 1642.33;13722 src tok/s; 14378 tgt tok/s; 141 s elapsed
Epoch 29, 40/ 47; acc: 7.30; ppl: 1519.73;15617 src tok/s; 16677 tgt tok/s; 142 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 3.72529e-09
Epoch 30, 10/ 47; acc: 8.02; ppl: 1480.45;15908 src tok/s; 16911 tgt tok/s; 145 s elapsed
Epoch 30, 20/ 47; acc: 7.55; ppl: 1482.16;16169 src tok/s; 17114 tgt tok/s; 145 s elapsed
Epoch 30, 30/ 47; acc: 4.82; ppl: 1677.55;13909 src tok/s; 14576 tgt tok/s; 146 s elapsed
Epoch 30, 40/ 47; acc: 9.30; ppl: 1421.38;15777 src tok/s; 16996 tgt tok/s; 147 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 1.86265e-09
Epoch 31, 10/ 47; acc: 6.68; ppl: 1458.77;16538 src tok/s; 17576 tgt tok/s; 150 s elapsed
Epoch 31, 20/ 47; acc: 7.48; ppl: 1526.60;13546 src tok/s; 14318 tgt tok/s; 150 s elapsed
Epoch 31, 30/ 47; acc: 7.49; ppl: 1478.26;16758 src tok/s; 17713 tgt tok/s; 151 s elapsed
Epoch 31, 40/ 47; acc: 6.77; ppl: 1607.50;16013 src tok/s; 16855 tgt tok/s; 152 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 9.31323e-10
Epoch 32, 10/ 47; acc: 6.98; ppl: 1546.92;16553 src tok/s; 17407 tgt tok/s; 155 s elapsed
Epoch 32, 20/ 47; acc: 6.41; ppl: 1604.94;13678 src tok/s; 14332 tgt tok/s; 156 s elapsed
Epoch 32, 30/ 47; acc: 6.64; ppl: 1581.39;15781 src tok/s; 16729 tgt tok/s; 156 s elapsed
Epoch 32, 40/ 47; acc: 7.68; ppl: 1506.75;16124 src tok/s; 17094 tgt tok/s; 157 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 4.65661e-10
Epoch 33, 10/ 47; acc: 6.96; ppl: 1539.23;16123 src tok/s; 17027 tgt tok/s; 160 s elapsed
Epoch 33, 20/ 47; acc: 6.60; ppl: 1605.66;16313 src tok/s; 17098 tgt tok/s; 160 s elapsed
Epoch 33, 30/ 47; acc: 6.24; ppl: 1594.36;16091 src tok/s; 16965 tgt tok/s; 161 s elapsed
Epoch 33, 40/ 47; acc: 7.59; ppl: 1479.42;12984 src tok/s; 13816 tgt tok/s; 162 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 2.32831e-10
Epoch 34, 10/ 47; acc: 6.59; ppl: 1520.51;16378 src tok/s; 17278 tgt tok/s; 165 s elapsed
Epoch 34, 20/ 47; acc: 6.38; ppl: 1662.76;13899 src tok/s; 14556 tgt tok/s; 165 s elapsed
Epoch 34, 30/ 47; acc: 6.48; ppl: 1596.60;15526 src tok/s; 16459 tgt tok/s; 166 s elapsed
Epoch 34, 40/ 47; acc: 7.95; ppl: 1479.05;16622 src tok/s; 17577 tgt tok/s; 167 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 1.16415e-10
Epoch 35, 10/ 47; acc: 7.15; ppl: 1558.14;16333 src tok/s; 17275 tgt tok/s; 170 s elapsed
Epoch 35, 20/ 47; acc: 7.80; ppl: 1456.51;13824 src tok/s; 14647 tgt tok/s; 170 s elapsed
Epoch 35, 30/ 47; acc: 8.34; ppl: 1475.30;15782 src tok/s; 16777 tgt tok/s; 171 s elapsed
Epoch 35, 40/ 47; acc: 5.38; ppl: 1672.59;15525 src tok/s; 16281 tgt tok/s; 172 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 5.82077e-11
Epoch 36, 10/ 47; acc: 6.87; ppl: 1484.26;15925 src tok/s; 16891 tgt tok/s; 174 s elapsed
Epoch 36, 20/ 47; acc: 7.67; ppl: 1511.85;13914 src tok/s; 14692 tgt tok/s; 175 s elapsed
Epoch 36, 30/ 47; acc: 7.48; ppl: 1487.91;16054 src tok/s; 17036 tgt tok/s; 176 s elapsed
Epoch 36, 40/ 47; acc: 5.29; ppl: 1714.35;15639 src tok/s; 16358 tgt tok/s; 177 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 2.91038e-11
Epoch 37, 10/ 47; acc: 7.76; ppl: 1500.33;16128 src tok/s; 17041 tgt tok/s; 180 s elapsed
Epoch 37, 20/ 47; acc: 7.98; ppl: 1534.90;15728 src tok/s; 16697 tgt tok/s; 180 s elapsed
Epoch 37, 30/ 47; acc: 7.46; ppl: 1459.75;16500 src tok/s; 17516 tgt tok/s; 181 s elapsed
Epoch 37, 40/ 47; acc: 5.48; ppl: 1633.41;13457 src tok/s; 14127 tgt tok/s; 182 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 1.45519e-11
Epoch 38, 10/ 47; acc: 7.62; ppl: 1547.49;15459 src tok/s; 16383 tgt tok/s; 184 s elapsed
Epoch 38, 20/ 47; acc: 6.36; ppl: 1628.51;13758 src tok/s; 14454 tgt tok/s; 185 s elapsed
Epoch 38, 30/ 47; acc: 8.42; ppl: 1407.17;15449 src tok/s; 16660 tgt tok/s; 186 s elapsed
Epoch 38, 40/ 47; acc: 6.50; ppl: 1546.96;16690 src tok/s; 17476 tgt tok/s; 187 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 7.27596e-12
Epoch 39, 10/ 47; acc: 6.99; ppl: 1559.94;16131 src tok/s; 17059 tgt tok/s; 189 s elapsed
Epoch 39, 20/ 47; acc: 7.28; ppl: 1477.87;15866 src tok/s; 16830 tgt tok/s; 190 s elapsed
Epoch 39, 30/ 47; acc: 7.93; ppl: 1503.61;15737 src tok/s; 16763 tgt tok/s; 191 s elapsed
Epoch 39, 40/ 47; acc: 5.93; ppl: 1654.52;16083 src tok/s; 16856 tgt tok/s; 192 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 3.63798e-12
Epoch 40, 10/ 47; acc: 5.87; ppl: 1699.68;16054 src tok/s; 16768 tgt tok/s; 195 s elapsed
Epoch 40, 20/ 47; acc: 7.47; ppl: 1399.04;14673 src tok/s; 15811 tgt tok/s; 195 s elapsed
Epoch 40, 30/ 47; acc: 7.76; ppl: 1524.73;13936 src tok/s; 14727 tgt tok/s; 196 s elapsed
Epoch 40, 40/ 47; acc: 7.36; ppl: 1513.68;16011 src tok/s; 16923 tgt tok/s; 197 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 1.81899e-12
Epoch 41, 10/ 47; acc: 8.24; ppl: 1437.12;15965 src tok/s; 17095 tgt tok/s; 199 s elapsed
Epoch 41, 20/ 47; acc: 6.56; ppl: 1558.53;15789 src tok/s; 16554 tgt tok/s; 200 s elapsed
Epoch 41, 30/ 47; acc: 6.28; ppl: 1650.33;15604 src tok/s; 16453 tgt tok/s; 201 s elapsed
Epoch 41, 40/ 47; acc: 6.57; ppl: 1547.31;16205 src tok/s; 17125 tgt tok/s; 202 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 9.09495e-13
Epoch 42, 10/ 47; acc: 6.15; ppl: 1642.70;16193 src tok/s; 16977 tgt tok/s; 205 s elapsed
Epoch 42, 20/ 47; acc: 7.28; ppl: 1524.89;15794 src tok/s; 16743 tgt tok/s; 205 s elapsed
Epoch 42, 30/ 47; acc: 8.71; ppl: 1377.65;15947 src tok/s; 17123 tgt tok/s; 206 s elapsed
Epoch 42, 40/ 47; acc: 6.81; ppl: 1544.22;16303 src tok/s; 17195 tgt tok/s; 207 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 4.54747e-13
Epoch 43, 10/ 47; acc: 6.26; ppl: 1579.77;15620 src tok/s; 16437 tgt tok/s; 210 s elapsed
Epoch 43, 20/ 47; acc: 6.95; ppl: 1561.67;15617 src tok/s; 16603 tgt tok/s; 210 s elapsed
Epoch 43, 30/ 47; acc: 7.00; ppl: 1544.48;16543 src tok/s; 17516 tgt tok/s; 211 s elapsed
Epoch 43, 40/ 47; acc: 7.17; ppl: 1539.66;13916 src tok/s; 14656 tgt tok/s; 212 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 2.27374e-13
Epoch 44, 10/ 47; acc: 5.81; ppl: 1622.87;16423 src tok/s; 17176 tgt tok/s; 215 s elapsed
Epoch 44, 20/ 47; acc: 7.93; ppl: 1442.15;15938 src tok/s; 16914 tgt tok/s; 215 s elapsed
Epoch 44, 30/ 47; acc: 6.23; ppl: 1634.25;14047 src tok/s; 14685 tgt tok/s; 216 s elapsed
Epoch 44, 40/ 47; acc: 8.49; ppl: 1486.86;16124 src tok/s; 17164 tgt tok/s; 217 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 1.13687e-13
Epoch 45, 10/ 47; acc: 8.03; ppl: 1417.67;13546 src tok/s; 14436 tgt tok/s; 219 s elapsed
Epoch 45, 20/ 47; acc: 7.75; ppl: 1579.37;15703 src tok/s; 16681 tgt tok/s; 220 s elapsed
Epoch 45, 30/ 47; acc: 7.52; ppl: 1509.11;16614 src tok/s; 17603 tgt tok/s; 221 s elapsed
Epoch 45, 40/ 47; acc: 6.44; ppl: 1536.83;15989 src tok/s; 16957 tgt tok/s; 221 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 5.68434e-14
Epoch 46, 10/ 47; acc: 7.03; ppl: 1491.96;15985 src tok/s; 16940 tgt tok/s; 224 s elapsed
Epoch 46, 20/ 47; acc: 8.72; ppl: 1394.98;15058 src tok/s; 16195 tgt tok/s; 225 s elapsed
Epoch 46, 30/ 47; acc: 5.97; ppl: 1638.22;14592 src tok/s; 15228 tgt tok/s; 226 s elapsed
Epoch 46, 40/ 47; acc: 7.76; ppl: 1502.18;15615 src tok/s; 16726 tgt tok/s; 226 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 2.84217e-14
Epoch 47, 10/ 47; acc: 6.56; ppl: 1519.94;12974 src tok/s; 13756 tgt tok/s; 229 s elapsed
Epoch 47, 20/ 47; acc: 7.16; ppl: 1585.43;15845 src tok/s; 16726 tgt tok/s; 230 s elapsed
Epoch 47, 30/ 47; acc: 6.73; ppl: 1580.12;15783 src tok/s; 16662 tgt tok/s; 231 s elapsed
Epoch 47, 40/ 47; acc: 7.11; ppl: 1510.53;16566 src tok/s; 17428 tgt tok/s; 232 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 1.42109e-14
Epoch 48, 10/ 47; acc: 7.15; ppl: 1512.65;16892 src tok/s; 17796 tgt tok/s; 234 s elapsed
Epoch 48, 20/ 47; acc: 7.52; ppl: 1450.48;15391 src tok/s; 16465 tgt tok/s; 235 s elapsed
Epoch 48, 30/ 47; acc: 7.23; ppl: 1503.69;15894 src tok/s; 16900 tgt tok/s; 236 s elapsed
Epoch 48, 40/ 47; acc: 6.06; ppl: 1683.12;15569 src tok/s; 16241 tgt tok/s; 237 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 7.10543e-15
Epoch 49, 10/ 47; acc: 5.90; ppl: 1563.62;15802 src tok/s; 16577 tgt tok/s; 240 s elapsed
Epoch 49, 20/ 47; acc: 6.66; ppl: 1549.07;16273 src tok/s; 17285 tgt tok/s; 240 s elapsed
Epoch 49, 30/ 47; acc: 7.04; ppl: 1500.97;13740 src tok/s; 14511 tgt tok/s; 241 s elapsed
Epoch 49, 40/ 47; acc: 7.51; ppl: 1590.54;15778 src tok/s; 16675 tgt tok/s; 242 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 3.55271e-15
Epoch 50, 10/ 47; acc: 7.80; ppl: 1484.19;15798 src tok/s; 16692 tgt tok/s; 244 s elapsed
Epoch 50, 20/ 47; acc: 7.15; ppl: 1503.73;16056 src tok/s; 17087 tgt tok/s; 245 s elapsed
Epoch 50, 30/ 47; acc: 6.44; ppl: 1534.46;15922 src tok/s; 16718 tgt tok/s; 246 s elapsed
Epoch 50, 40/ 47; acc: 7.64; ppl: 1500.51;12384 src tok/s; 13280 tgt tok/s; 246 s elapsed
Train perplexity: 1546
Train accuracy: 6.97936
Validation perplexity: 24962.4
Validation accuracy: 4.8146
Decaying learning rate to 1.77636e-15
Time elapsed: 5s, Progress: 0%, Train Perplexity: 12064.0190
Time elapsed: 10s, Progress: 0%, Train Perplexity: 7062.5039
Time elapsed: 15s, Progress: 0%, Train Perplexity: 2220.6536
Time elapsed: 18s, Progress: 0%, Train Perplexity: 4136.7274
Time elapsed: 20s, Progress: 1%, Train Perplexity: 3540.9369
Time elapsed: 23s, Progress: 1%, Train Perplexity: 2793.5759
Time elapsed: 25s, Progress: 1%, Train Perplexity: 2451.3224
Time elapsed: 28s, Progress: 1%, Train Perplexity: 2018.3823
Time elapsed: 33s, Progress: 1%, Train Perplexity: 2406.7421
Finished epoch 1, Dev Perplexity: 2262.6731
Time elapsed: 51s, Progress: 2%, Train Perplexity: 2017.9869
Time elapsed: 53s, Progress: 2%, Train Perplexity: 2897.6919
Time elapsed: 55s, Progress: 2%, Train Perplexity: 1772.3356
Time elapsed: 57s, Progress: 2%, Train Perplexity: 1189.4782
Time elapsed: 1m 0s, Progress: 3%, Train Perplexity: 1412.2238
Time elapsed: 1m 2s, Progress: 3%, Train Perplexity: 1338.7009
Time elapsed: 1m 6s, Progress: 3%, Train Perplexity: 1880.7948
Time elapsed: 1m 11s, Progress: 3%, Train Perplexity: 1759.5221
Time elapsed: 1m 14s, Progress: 3%, Train Perplexity: 1819.2658
Finished epoch 2, Dev Perplexity: 1635.8633
Time elapsed: 1m 32s, Progress: 4%, Train Perplexity: 1827.5220
Time elapsed: 1m 36s, Progress: 4%, Train Perplexity: 1507.1703
Time elapsed: 1m 40s, Progress: 4%, Train Perplexity: 1706.8688
Time elapsed: 1m 44s, Progress: 4%, Train Perplexity: 1980.5540
Time elapsed: 1m 48s, Progress: 4%, Train Perplexity: 1726.4512
Time elapsed: 1m 51s, Progress: 5%, Train Perplexity: 1865.5049
Time elapsed: 1m 56s, Progress: 5%, Train Perplexity: 1799.2255
Time elapsed: 2m 0s, Progress: 5%, Train Perplexity: 1785.6693
Time elapsed: 2m 3s, Progress: 5%, Train Perplexity: 1704.5307
Finished epoch 3, Dev Perplexity: 1421.9510
Time elapsed: 2m 22s, Progress: 6%, Train Perplexity: 1676.1614
Time elapsed: 2m 25s, Progress: 6%, Train Perplexity: 1340.5262
Time elapsed: 2m 29s, Progress: 6%, Train Perplexity: 1473.0865
Time elapsed: 2m 33s, Progress: 6%, Train Perplexity: 1468.2574
Time elapsed: 2m 36s, Progress: 6%, Train Perplexity: 1600.0709
Time elapsed: 2m 40s, Progress: 7%, Train Perplexity: 1454.7108
Time elapsed: 2m 43s, Progress: 7%, Train Perplexity: 1568.7890
Time elapsed: 2m 47s, Progress: 7%, Train Perplexity: 1462.6514
Time elapsed: 2m 50s, Progress: 7%, Train Perplexity: 1467.1608
Time elapsed: 2m 54s, Progress: 7%, Train Perplexity: 1370.1677
Finished epoch 4, Dev Perplexity: 1589.4471
Time elapsed: 3m 11s, Progress: 8%, Train Perplexity: 1146.5988
Time elapsed: 3m 14s, Progress: 8%, Train Perplexity: 1078.6142
Time elapsed: 3m 18s, Progress: 8%, Train Perplexity: 1160.8384
Time elapsed: 3m 21s, Progress: 8%, Train Perplexity: 1125.1799
Time elapsed: 3m 25s, Progress: 9%, Train Perplexity: 1183.6633
Time elapsed: 3m 28s, Progress: 9%, Train Perplexity: 1232.4102
Time elapsed: 3m 31s, Progress: 9%, Train Perplexity: 1138.2386
Time elapsed: 3m 35s, Progress: 9%, Train Perplexity: 1185.8167
Time elapsed: 3m 38s, Progress: 9%, Train Perplexity: 1137.3155
Finished epoch 5, Dev Perplexity: 2147.5124
Time elapsed: 3m 57s, Progress: 10%, Train Perplexity: 1117.3546
Time elapsed: 4m 1s, Progress: 10%, Train Perplexity: 893.3994
Time elapsed: 4m 4s, Progress: 10%, Train Perplexity: 946.7700
Time elapsed: 4m 8s, Progress: 10%, Train Perplexity: 960.5643
Time elapsed: 4m 11s, Progress: 10%, Train Perplexity: 969.0492
Time elapsed: 4m 15s, Progress: 11%, Train Perplexity: 884.6977
Time elapsed: 4m 18s, Progress: 11%, Train Perplexity: 971.7974
Time elapsed: 4m 22s, Progress: 11%, Train Perplexity: 949.3997
Time elapsed: 4m 25s, Progress: 11%, Train Perplexity: 963.0147
Finished epoch 6, Dev Perplexity: 2402.0038
Time elapsed: 4m 44s, Progress: 12%, Train Perplexity: 851.1640
Time elapsed: 4m 48s, Progress: 12%, Train Perplexity: 737.3465
Time elapsed: 4m 52s, Progress: 12%, Train Perplexity: 793.4397
Time elapsed: 4m 56s, Progress: 12%, Train Perplexity: 804.6245
Time elapsed: 4m 59s, Progress: 12%, Train Perplexity: 819.2354
Time elapsed: 5m 3s, Progress: 13%, Train Perplexity: 829.0880
Time elapsed: 5m 7s, Progress: 13%, Train Perplexity: 783.8333
Time elapsed: 5m 10s, Progress: 13%, Train Perplexity: 800.1800
Time elapsed: 5m 14s, Progress: 13%, Train Perplexity: 739.0168
Time elapsed: 5m 17s, Progress: 13%, Train Perplexity: 795.8957
Finished epoch 7, Dev Perplexity: 3099.7776
Time elapsed: 5m 36s, Progress: 14%, Train Perplexity: 667.3761
Time elapsed: 5m 40s, Progress: 14%, Train Perplexity: 593.3539
Time elapsed: 5m 44s, Progress: 14%, Train Perplexity: 664.5755
Time elapsed: 5m 47s, Progress: 14%, Train Perplexity: 616.4736
Time elapsed: 5m 51s, Progress: 15%, Train Perplexity: 638.4265
Time elapsed: 5m 55s, Progress: 15%, Train Perplexity: 625.6891
Time elapsed: 5m 58s, Progress: 15%, Train Perplexity: 640.4083
Time elapsed: 6m 2s, Progress: 15%, Train Perplexity: 666.7520
Time elapsed: 6m 6s, Progress: 15%, Train Perplexity: 674.9899
Finished epoch 8, Dev Perplexity: 3365.2423
Time elapsed: 6m 24s, Progress: 16%, Train Perplexity: 515.1513
Time elapsed: 6m 28s, Progress: 16%, Train Perplexity: 486.3229
Time elapsed: 6m 32s, Progress: 16%, Train Perplexity: 510.2389
Time elapsed: 6m 36s, Progress: 16%, Train Perplexity: 499.5748
Time elapsed: 6m 39s, Progress: 16%, Train Perplexity: 533.6954
Time elapsed: 6m 43s, Progress: 17%, Train Perplexity: 491.6925
Time elapsed: 6m 47s, Progress: 17%, Train Perplexity: 531.1896
Time elapsed: 6m 50s, Progress: 17%, Train Perplexity: 519.4428
Time elapsed: 6m 54s, Progress: 17%, Train Perplexity: 545.8580
Finished epoch 9, Dev Perplexity: 4393.5124
Time elapsed: 7m 13s, Progress: 18%, Train Perplexity: 514.3702
Time elapsed: 7m 16s, Progress: 18%, Train Perplexity: 378.1452
Time elapsed: 7m 20s, Progress: 18%, Train Perplexity: 419.2095
Time elapsed: 7m 24s, Progress: 18%, Train Perplexity: 405.1825
Time elapsed: 7m 27s, Progress: 18%, Train Perplexity: 393.3341
Time elapsed: 7m 31s, Progress: 19%, Train Perplexity: 418.1897
Time elapsed: 7m 35s, Progress: 19%, Train Perplexity: 433.1738
Time elapsed: 7m 38s, Progress: 19%, Train Perplexity: 414.4442
Time elapsed: 7m 42s, Progress: 19%, Train Perplexity: 457.9027
Time elapsed: 7m 46s, Progress: 20%, Train Perplexity: 438.0495
Finished epoch 10, Dev Perplexity: 5189.0745
Time elapsed: 8m 5s, Progress: 20%, Train Perplexity: 336.7984
Time elapsed: 8m 9s, Progress: 20%, Train Perplexity: 321.8272
Time elapsed: 8m 12s, Progress: 20%, Train Perplexity: 310.5734
Time elapsed: 8m 16s, Progress: 20%, Train Perplexity: 333.9349
Time elapsed: 8m 20s, Progress: 21%, Train Perplexity: 356.2809
Time elapsed: 8m 24s, Progress: 21%, Train Perplexity: 345.0352
Time elapsed: 8m 27s, Progress: 21%, Train Perplexity: 343.4847
Time elapsed: 8m 31s, Progress: 21%, Train Perplexity: 353.0172
Time elapsed: 8m 35s, Progress: 21%, Train Perplexity: 363.0145
Finished epoch 11, Dev Perplexity: 5589.3480
Time elapsed: 8m 53s, Progress: 22%, Train Perplexity: 296.7774
Time elapsed: 8m 57s, Progress: 22%, Train Perplexity: 259.6640
Time elapsed: 9m 1s, Progress: 22%, Train Perplexity: 265.0358
Time elapsed: 9m 5s, Progress: 22%, Train Perplexity: 256.8508
Time elapsed: 9m 9s, Progress: 23%, Train Perplexity: 261.4596
Time elapsed: 9m 13s, Progress: 23%, Train Perplexity: 291.3458
Time elapsed: 9m 16s, Progress: 23%, Train Perplexity: 293.0510
Time elapsed: 9m 20s, Progress: 23%, Train Perplexity: 299.6770
Time elapsed: 9m 24s, Progress: 23%, Train Perplexity: 292.2070
Finished epoch 12, Dev Perplexity: 5039.5653
Time elapsed: 9m 43s, Progress: 24%, Train Perplexity: 264.8841
Time elapsed: 9m 46s, Progress: 24%, Train Perplexity: 202.2874
Time elapsed: 9m 50s, Progress: 24%, Train Perplexity: 214.2224
Time elapsed: 9m 54s, Progress: 24%, Train Perplexity: 208.0391
Time elapsed: 9m 58s, Progress: 24%, Train Perplexity: 217.2690
Time elapsed: 10m 2s, Progress: 25%, Train Perplexity: 242.3922
Time elapsed: 10m 5s, Progress: 25%, Train Perplexity: 212.9597
Time elapsed: 10m 9s, Progress: 25%, Train Perplexity: 238.3820
Time elapsed: 10m 13s, Progress: 25%, Train Perplexity: 250.2861
Finished epoch 13, Dev Perplexity: 4728.6293
Time elapsed: 10m 31s, Progress: 26%, Train Perplexity: 228.6048
Time elapsed: 10m 35s, Progress: 26%, Train Perplexity: 165.2087
Time elapsed: 10m 39s, Progress: 26%, Train Perplexity: 172.7357
Time elapsed: 10m 43s, Progress: 26%, Train Perplexity: 174.3654
Time elapsed: 10m 47s, Progress: 26%, Train Perplexity: 177.7761
Time elapsed: 10m 51s, Progress: 27%, Train Perplexity: 192.0120
Time elapsed: 10m 54s, Progress: 27%, Train Perplexity: 190.1174
Time elapsed: 10m 58s, Progress: 27%, Train Perplexity: 189.6029
Time elapsed: 11m 2s, Progress: 27%, Train Perplexity: 201.6829
Time elapsed: 11m 6s, Progress: 27%, Train Perplexity: 192.1468
Finished epoch 14, Dev Perplexity: 5094.4192
Time elapsed: 11m 24s, Progress: 28%, Train Perplexity: 160.1107
Time elapsed: 11m 28s, Progress: 28%, Train Perplexity: 139.4926
Time elapsed: 11m 32s, Progress: 28%, Train Perplexity: 144.3353
Time elapsed: 11m 36s, Progress: 28%, Train Perplexity: 138.7662
Time elapsed: 11m 39s, Progress: 29%, Train Perplexity: 146.2865
Time elapsed: 11m 43s, Progress: 29%, Train Perplexity: 157.6710
Time elapsed: 11m 47s, Progress: 29%, Train Perplexity: 161.8209
Time elapsed: 11m 51s, Progress: 29%, Train Perplexity: 157.2941
Time elapsed: 11m 55s, Progress: 29%, Train Perplexity: 160.4086
Finished epoch 15, Dev Perplexity: 5472.6506
Time elapsed: 12m 13s, Progress: 30%, Train Perplexity: 133.6996
Time elapsed: 12m 17s, Progress: 30%, Train Perplexity: 115.9270
Time elapsed: 12m 21s, Progress: 30%, Train Perplexity: 115.0015
Time elapsed: 12m 25s, Progress: 30%, Train Perplexity: 121.3867
Time elapsed: 12m 28s, Progress: 30%, Train Perplexity: 128.4904
Time elapsed: 12m 32s, Progress: 31%, Train Perplexity: 117.3177
Time elapsed: 12m 36s, Progress: 31%, Train Perplexity: 123.9350
Time elapsed: 12m 40s, Progress: 31%, Train Perplexity: 129.9500
Time elapsed: 12m 44s, Progress: 31%, Train Perplexity: 129.6720
Finished epoch 16, Dev Perplexity: 5988.7686
Time elapsed: 13m 2s, Progress: 32%, Train Perplexity: 131.6853
Time elapsed: 13m 6s, Progress: 32%, Train Perplexity: 96.7661
Time elapsed: 13m 10s, Progress: 32%, Train Perplexity: 94.4887
Time elapsed: 13m 14s, Progress: 32%, Train Perplexity: 100.8311
Time elapsed: 13m 17s, Progress: 32%, Train Perplexity: 101.2741
Time elapsed: 13m 21s, Progress: 33%, Train Perplexity: 104.8473
Time elapsed: 13m 25s, Progress: 33%, Train Perplexity: 106.9645
Time elapsed: 13m 29s, Progress: 33%, Train Perplexity: 105.3404
Time elapsed: 13m 33s, Progress: 33%, Train Perplexity: 109.0339
Time elapsed: 13m 37s, Progress: 33%, Train Perplexity: 109.8217
Finished epoch 17, Dev Perplexity: 6617.7469
Time elapsed: 13m 56s, Progress: 34%, Train Perplexity: 82.1143
Time elapsed: 13m 59s, Progress: 34%, Train Perplexity: 77.0516
Time elapsed: 14m 3s, Progress: 34%, Train Perplexity: 77.7184
Time elapsed: 14m 7s, Progress: 34%, Train Perplexity: 85.6571
Time elapsed: 14m 11s, Progress: 35%, Train Perplexity: 87.7678
Time elapsed: 14m 14s, Progress: 35%, Train Perplexity: 85.6980
Time elapsed: 14m 18s, Progress: 35%, Train Perplexity: 86.5760
Time elapsed: 14m 22s, Progress: 35%, Train Perplexity: 92.3441
Time elapsed: 14m 26s, Progress: 35%, Train Perplexity: 98.1120
Finished epoch 18, Dev Perplexity: 6593.7602
Time elapsed: 14m 45s, Progress: 36%, Train Perplexity: 79.2255
Time elapsed: 14m 49s, Progress: 36%, Train Perplexity: 62.3503
Time elapsed: 14m 52s, Progress: 36%, Train Perplexity: 67.6677
Time elapsed: 14m 56s, Progress: 36%, Train Perplexity: 69.3157
Time elapsed: 15m 0s, Progress: 36%, Train Perplexity: 72.7170
Time elapsed: 15m 4s, Progress: 37%, Train Perplexity: 74.7740
Time elapsed: 15m 8s, Progress: 37%, Train Perplexity: 78.6687
Time elapsed: 15m 11s, Progress: 37%, Train Perplexity: 78.9518
Time elapsed: 15m 15s, Progress: 37%, Train Perplexity: 78.4187
Finished epoch 19, Dev Perplexity: 7656.3505
Time elapsed: 15m 34s, Progress: 38%, Train Perplexity: 71.0876
Time elapsed: 15m 38s, Progress: 38%, Train Perplexity: 57.0178
Time elapsed: 15m 42s, Progress: 38%, Train Perplexity: 58.1548
Time elapsed: 15m 46s, Progress: 38%, Train Perplexity: 59.0265
Time elapsed: 15m 49s, Progress: 38%, Train Perplexity: 60.3169
Time elapsed: 15m 53s, Progress: 39%, Train Perplexity: 63.1183
Time elapsed: 15m 57s, Progress: 39%, Train Perplexity: 62.7434
Time elapsed: 16m 1s, Progress: 39%, Train Perplexity: 60.2808
Time elapsed: 16m 5s, Progress: 39%, Train Perplexity: 67.7314
Time elapsed: 16m 8s, Progress: 40%, Train Perplexity: 67.5243
Finished epoch 20, Dev Perplexity: 7686.8316
Time elapsed: 16m 27s, Progress: 40%, Train Perplexity: 47.9937
Time elapsed: 16m 31s, Progress: 40%, Train Perplexity: 47.3105
Time elapsed: 16m 35s, Progress: 40%, Train Perplexity: 50.0563
Time elapsed: 16m 39s, Progress: 40%, Train Perplexity: 51.9930
Time elapsed: 16m 43s, Progress: 41%, Train Perplexity: 55.2752
Time elapsed: 16m 46s, Progress: 41%, Train Perplexity: 54.0303
Time elapsed: 16m 50s, Progress: 41%, Train Perplexity: 54.5946
Time elapsed: 16m 54s, Progress: 41%, Train Perplexity: 55.2999
Time elapsed: 16m 58s, Progress: 41%, Train Perplexity: 54.6257
Finished epoch 21, Dev Perplexity: 8796.1873
Time elapsed: 17m 17s, Progress: 42%, Train Perplexity: 44.2810
Time elapsed: 17m 21s, Progress: 42%, Train Perplexity: 40.5151
Time elapsed: 17m 24s, Progress: 42%, Train Perplexity: 43.4625
Time elapsed: 17m 28s, Progress: 42%, Train Perplexity: 45.0047
Time elapsed: 17m 32s, Progress: 43%, Train Perplexity: 43.6498
Time elapsed: 17m 36s, Progress: 43%, Train Perplexity: 45.1757
Time elapsed: 17m 39s, Progress: 43%, Train Perplexity: 44.9244
Time elapsed: 17m 43s, Progress: 43%, Train Perplexity: 46.8944
Time elapsed: 17m 47s, Progress: 43%, Train Perplexity: 47.4511
Finished epoch 22, Dev Perplexity: 9337.7158
Time elapsed: 18m 6s, Progress: 44%, Train Perplexity: 41.3676
Time elapsed: 18m 10s, Progress: 44%, Train Perplexity: 36.3214
Time elapsed: 18m 13s, Progress: 44%, Train Perplexity: 33.6637
Time elapsed: 18m 17s, Progress: 44%, Train Perplexity: 37.8450
Time elapsed: 18m 21s, Progress: 44%, Train Perplexity: 36.7766
Time elapsed: 18m 25s, Progress: 45%, Train Perplexity: 37.9718
Time elapsed: 18m 29s, Progress: 45%, Train Perplexity: 40.2940
Time elapsed: 18m 32s, Progress: 45%, Train Perplexity: 40.5989
Time elapsed: 18m 36s, Progress: 45%, Train Perplexity: 39.2418
Finished epoch 23, Dev Perplexity: 10008.7606
Time elapsed: 18m 55s, Progress: 46%, Train Perplexity: 43.1701
Time elapsed: 18m 59s, Progress: 46%, Train Perplexity: 29.7233
Time elapsed: 19m 3s, Progress: 46%, Train Perplexity: 30.4189
Time elapsed: 19m 7s, Progress: 46%, Train Perplexity: 31.1094
Time elapsed: 19m 11s, Progress: 46%, Train Perplexity: 31.3974
Time elapsed: 19m 14s, Progress: 47%, Train Perplexity: 31.9606
Time elapsed: 19m 18s, Progress: 47%, Train Perplexity: 33.8395
Time elapsed: 19m 22s, Progress: 47%, Train Perplexity: 36.5964
Time elapsed: 19m 26s, Progress: 47%, Train Perplexity: 34.9556
Time elapsed: 19m 30s, Progress: 47%, Train Perplexity: 35.6673
Finished epoch 24, Dev Perplexity: 9873.9333
Time elapsed: 19m 49s, Progress: 48%, Train Perplexity: 26.1621
Time elapsed: 19m 52s, Progress: 48%, Train Perplexity: 26.4737
Time elapsed: 19m 56s, Progress: 48%, Train Perplexity: 25.0622
Time elapsed: 20m 0s, Progress: 48%, Train Perplexity: 27.6286
Time elapsed: 20m 4s, Progress: 49%, Train Perplexity: 29.1854
Time elapsed: 20m 8s, Progress: 49%, Train Perplexity: 28.3941
Time elapsed: 20m 12s, Progress: 49%, Train Perplexity: 29.9404
Time elapsed: 20m 15s, Progress: 49%, Train Perplexity: 29.5269
Time elapsed: 20m 19s, Progress: 49%, Train Perplexity: 31.4228
Finished epoch 25, Dev Perplexity: 10213.2826
Time elapsed: 20m 38s, Progress: 50%, Train Perplexity: 26.2301
Time elapsed: 20m 42s, Progress: 50%, Train Perplexity: 23.2701
Time elapsed: 20m 46s, Progress: 50%, Train Perplexity: 23.1606
Time elapsed: 20m 49s, Progress: 50%, Train Perplexity: 24.0501
Time elapsed: 20m 53s, Progress: 50%, Train Perplexity: 24.1646
Time elapsed: 20m 57s, Progress: 51%, Train Perplexity: 26.0877
Time elapsed: 21m 1s, Progress: 51%, Train Perplexity: 24.9941
Time elapsed: 21m 5s, Progress: 51%, Train Perplexity: 25.1861
Time elapsed: 21m 8s, Progress: 51%, Train Perplexity: 26.0110
Finished epoch 26, Dev Perplexity: 10490.4581
Time elapsed: 21m 27s, Progress: 52%, Train Perplexity: 24.9415
Time elapsed: 21m 31s, Progress: 52%, Train Perplexity: 18.8586
Time elapsed: 21m 35s, Progress: 52%, Train Perplexity: 20.8179
Time elapsed: 21m 39s, Progress: 52%, Train Perplexity: 20.6850
Time elapsed: 21m 42s, Progress: 52%, Train Perplexity: 20.8885
Time elapsed: 21m 46s, Progress: 53%, Train Perplexity: 20.5334
Time elapsed: 21m 50s, Progress: 53%, Train Perplexity: 21.4359
Time elapsed: 21m 54s, Progress: 53%, Train Perplexity: 22.5277
Time elapsed: 21m 58s, Progress: 53%, Train Perplexity: 22.4506
Time elapsed: 22m 1s, Progress: 53%, Train Perplexity: 23.2554
Finished epoch 27, Dev Perplexity: 11950.9353
Time elapsed: 22m 20s, Progress: 54%, Train Perplexity: 17.1774
Time elapsed: 22m 24s, Progress: 54%, Train Perplexity: 17.8211
Time elapsed: 22m 28s, Progress: 54%, Train Perplexity: 17.3503
Time elapsed: 22m 32s, Progress: 54%, Train Perplexity: 18.6644
Time elapsed: 22m 36s, Progress: 55%, Train Perplexity: 18.1188
Time elapsed: 22m 39s, Progress: 55%, Train Perplexity: 19.3299
Time elapsed: 22m 43s, Progress: 55%, Train Perplexity: 20.6036
Time elapsed: 22m 47s, Progress: 55%, Train Perplexity: 20.0424
Time elapsed: 22m 51s, Progress: 55%, Train Perplexity: 19.4339
Finished epoch 28, Dev Perplexity: 11792.7203
Time elapsed: 23m 10s, Progress: 56%, Train Perplexity: 16.2640
Time elapsed: 23m 13s, Progress: 56%, Train Perplexity: 14.6226
Time elapsed: 23m 17s, Progress: 56%, Train Perplexity: 16.0145
Time elapsed: 23m 21s, Progress: 56%, Train Perplexity: 15.6083
Time elapsed: 23m 25s, Progress: 56%, Train Perplexity: 16.6502
Time elapsed: 23m 29s, Progress: 57%, Train Perplexity: 16.1890
Time elapsed: 23m 32s, Progress: 57%, Train Perplexity: 17.0283
Time elapsed: 23m 36s, Progress: 57%, Train Perplexity: 17.4576
Time elapsed: 23m 40s, Progress: 57%, Train Perplexity: 17.7882
Finished epoch 29, Dev Perplexity: 11629.2361
Time elapsed: 23m 59s, Progress: 58%, Train Perplexity: 17.3465
Time elapsed: 24m 3s, Progress: 58%, Train Perplexity: 12.9778
Time elapsed: 24m 6s, Progress: 58%, Train Perplexity: 13.9080
Time elapsed: 24m 10s, Progress: 58%, Train Perplexity: 14.0678
Time elapsed: 24m 14s, Progress: 58%, Train Perplexity: 14.2753
Time elapsed: 24m 18s, Progress: 59%, Train Perplexity: 14.0517
Time elapsed: 24m 21s, Progress: 59%, Train Perplexity: 15.2584
Time elapsed: 24m 25s, Progress: 59%, Train Perplexity: 15.5233
Time elapsed: 24m 29s, Progress: 59%, Train Perplexity: 15.4929
Time elapsed: 24m 33s, Progress: 60%, Train Perplexity: 15.3491
Finished epoch 30, Dev Perplexity: 11522.2030
Time elapsed: 24m 52s, Progress: 60%, Train Perplexity: 12.1112
Time elapsed: 24m 56s, Progress: 60%, Train Perplexity: 12.0041
Time elapsed: 25m 0s, Progress: 60%, Train Perplexity: 12.0163
Time elapsed: 25m 3s, Progress: 60%, Train Perplexity: 12.0481
Time elapsed: 25m 7s, Progress: 61%, Train Perplexity: 13.1407
Time elapsed: 25m 11s, Progress: 61%, Train Perplexity: 12.6385
Time elapsed: 25m 15s, Progress: 61%, Train Perplexity: 13.2440
Time elapsed: 25m 18s, Progress: 61%, Train Perplexity: 13.5025
Time elapsed: 25m 22s, Progress: 61%, Train Perplexity: 14.0686
Finished epoch 31, Dev Perplexity: 11549.1875
Time elapsed: 25m 41s, Progress: 62%, Train Perplexity: 11.5817
Time elapsed: 25m 45s, Progress: 62%, Train Perplexity: 10.5352
Time elapsed: 25m 49s, Progress: 62%, Train Perplexity: 10.7854
Time elapsed: 25m 53s, Progress: 62%, Train Perplexity: 10.7445
Time elapsed: 25m 57s, Progress: 63%, Train Perplexity: 11.8582
Time elapsed: 26m 0s, Progress: 63%, Train Perplexity: 11.6747
Time elapsed: 26m 4s, Progress: 63%, Train Perplexity: 11.5853
Time elapsed: 26m 8s, Progress: 63%, Train Perplexity: 11.6326
Time elapsed: 26m 12s, Progress: 63%, Train Perplexity: 11.7204
Finished epoch 32, Dev Perplexity: 11358.9430
Time elapsed: 26m 31s, Progress: 64%, Train Perplexity: 11.2109
Time elapsed: 26m 34s, Progress: 64%, Train Perplexity: 9.5465
Time elapsed: 26m 38s, Progress: 64%, Train Perplexity: 9.4609
Time elapsed: 26m 42s, Progress: 64%, Train Perplexity: 10.3601
Time elapsed: 26m 46s, Progress: 64%, Train Perplexity: 9.6112
Time elapsed: 26m 50s, Progress: 65%, Train Perplexity: 9.8225
Time elapsed: 26m 53s, Progress: 65%, Train Perplexity: 10.2594
Time elapsed: 26m 57s, Progress: 65%, Train Perplexity: 10.7516
Time elapsed: 27m 1s, Progress: 65%, Train Perplexity: 10.7159
Finished epoch 33, Dev Perplexity: 11370.3338
Time elapsed: 27m 20s, Progress: 66%, Train Perplexity: 10.6552
Time elapsed: 27m 24s, Progress: 66%, Train Perplexity: 7.9246
Time elapsed: 27m 27s, Progress: 66%, Train Perplexity: 8.4909
Time elapsed: 27m 31s, Progress: 66%, Train Perplexity: 8.5834
Time elapsed: 27m 35s, Progress: 66%, Train Perplexity: 9.0125
Time elapsed: 27m 39s, Progress: 67%, Train Perplexity: 8.9479
Time elapsed: 27m 43s, Progress: 67%, Train Perplexity: 9.0304
Time elapsed: 27m 47s, Progress: 67%, Train Perplexity: 9.8543
Time elapsed: 27m 50s, Progress: 67%, Train Perplexity: 9.9726
Time elapsed: 27m 54s, Progress: 67%, Train Perplexity: 9.2310
Finished epoch 34, Dev Perplexity: 11198.6791
Time elapsed: 28m 13s, Progress: 68%, Train Perplexity: 8.3145
Time elapsed: 28m 17s, Progress: 68%, Train Perplexity: 7.8392
Time elapsed: 28m 20s, Progress: 68%, Train Perplexity: 7.6300
Time elapsed: 28m 24s, Progress: 68%, Train Perplexity: 8.4804
Time elapsed: 28m 28s, Progress: 69%, Train Perplexity: 7.9653
Time elapsed: 28m 32s, Progress: 69%, Train Perplexity: 7.8190
Time elapsed: 28m 36s, Progress: 69%, Train Perplexity: 8.3633
Time elapsed: 28m 40s, Progress: 69%, Train Perplexity: 8.7110
Time elapsed: 28m 43s, Progress: 69%, Train Perplexity: 8.6737
Finished epoch 35, Dev Perplexity: 11143.3359
Time elapsed: 29m 2s, Progress: 70%, Train Perplexity: 7.3700
Time elapsed: 29m 6s, Progress: 70%, Train Perplexity: 6.8435
Time elapsed: 29m 10s, Progress: 70%, Train Perplexity: 7.2606
Time elapsed: 29m 14s, Progress: 70%, Train Perplexity: 7.0924
Time elapsed: 29m 17s, Progress: 70%, Train Perplexity: 7.5923
Time elapsed: 29m 21s, Progress: 71%, Train Perplexity: 7.6828
Time elapsed: 29m 25s, Progress: 71%, Train Perplexity: 7.2870
Time elapsed: 29m 29s, Progress: 71%, Train Perplexity: 7.5019
Time elapsed: 29m 33s, Progress: 71%, Train Perplexity: 7.6082
Finished epoch 36, Dev Perplexity: 10011.4628
Time elapsed: 29m 51s, Progress: 72%, Train Perplexity: 7.5647
Time elapsed: 29m 55s, Progress: 72%, Train Perplexity: 6.1894
Time elapsed: 29m 59s, Progress: 72%, Train Perplexity: 6.2246
Time elapsed: 30m 3s, Progress: 72%, Train Perplexity: 6.4897
Time elapsed: 30m 7s, Progress: 72%, Train Perplexity: 6.5516
Time elapsed: 30m 10s, Progress: 73%, Train Perplexity: 6.9181
Time elapsed: 30m 14s, Progress: 73%, Train Perplexity: 6.6139
Time elapsed: 30m 18s, Progress: 73%, Train Perplexity: 6.7642
Time elapsed: 30m 22s, Progress: 73%, Train Perplexity: 7.2023
Time elapsed: 30m 26s, Progress: 73%, Train Perplexity: 6.9217
Finished epoch 37, Dev Perplexity: 9266.2099
Time elapsed: 30m 44s, Progress: 74%, Train Perplexity: 6.1606
Time elapsed: 30m 48s, Progress: 74%, Train Perplexity: 5.5370
Time elapsed: 30m 52s, Progress: 74%, Train Perplexity: 5.7062
Time elapsed: 30m 56s, Progress: 74%, Train Perplexity: 6.0862
Time elapsed: 30m 59s, Progress: 75%, Train Perplexity: 5.9932
Time elapsed: 31m 3s, Progress: 75%, Train Perplexity: 6.3440
Time elapsed: 31m 7s, Progress: 75%, Train Perplexity: 6.4011
Time elapsed: 31m 11s, Progress: 75%, Train Perplexity: 6.2331
Time elapsed: 31m 15s, Progress: 75%, Train Perplexity: 6.0724
Finished epoch 38, Dev Perplexity: 8547.2899
Time elapsed: 31m 34s, Progress: 76%, Train Perplexity: 5.9103
Time elapsed: 31m 37s, Progress: 76%, Train Perplexity: 5.5005
Time elapsed: 31m 41s, Progress: 76%, Train Perplexity: 5.2085
Time elapsed: 31m 45s, Progress: 76%, Train Perplexity: 5.5622
Time elapsed: 31m 49s, Progress: 76%, Train Perplexity: 5.1874
Time elapsed: 31m 53s, Progress: 77%, Train Perplexity: 5.5293
Time elapsed: 31m 56s, Progress: 77%, Train Perplexity: 5.7165
Time elapsed: 32m 0s, Progress: 77%, Train Perplexity: 5.7619
Time elapsed: 32m 4s, Progress: 77%, Train Perplexity: 5.5658
Finished epoch 39, Dev Perplexity: 8631.1280
Time elapsed: 32m 23s, Progress: 78%, Train Perplexity: 5.4598
Time elapsed: 32m 27s, Progress: 78%, Train Perplexity: 4.8408
Time elapsed: 32m 31s, Progress: 78%, Train Perplexity: 4.7511
Time elapsed: 32m 34s, Progress: 78%, Train Perplexity: 4.8847
Time elapsed: 32m 38s, Progress: 78%, Train Perplexity: 4.7489
Time elapsed: 32m 42s, Progress: 79%, Train Perplexity: 5.2488
Time elapsed: 32m 46s, Progress: 79%, Train Perplexity: 5.6200
Time elapsed: 32m 50s, Progress: 79%, Train Perplexity: 5.3802
Time elapsed: 32m 53s, Progress: 79%, Train Perplexity: 5.2254
Time elapsed: 32m 57s, Progress: 80%, Train Perplexity: 5.2549
Finished epoch 40, Dev Perplexity: 7665.9225
Time elapsed: 33m 16s, Progress: 80%, Train Perplexity: 4.3930
Time elapsed: 33m 20s, Progress: 80%, Train Perplexity: 4.5870
Time elapsed: 33m 23s, Progress: 80%, Train Perplexity: 4.7919
Time elapsed: 33m 27s, Progress: 80%, Train Perplexity: 4.7183
Time elapsed: 33m 31s, Progress: 81%, Train Perplexity: 4.6285
Time elapsed: 33m 35s, Progress: 81%, Train Perplexity: 4.6956
Time elapsed: 33m 39s, Progress: 81%, Train Perplexity: 4.5199
Time elapsed: 33m 42s, Progress: 81%, Train Perplexity: 4.9708
Time elapsed: 33m 46s, Progress: 81%, Train Perplexity: 4.8370
Finished epoch 41, Dev Perplexity: 6729.7290
Time elapsed: 34m 5s, Progress: 82%, Train Perplexity: 4.4102
Time elapsed: 34m 9s, Progress: 82%, Train Perplexity: 4.0517
Time elapsed: 34m 13s, Progress: 82%, Train Perplexity: 4.1922
Time elapsed: 34m 16s, Progress: 82%, Train Perplexity: 4.0434
Time elapsed: 34m 20s, Progress: 83%, Train Perplexity: 4.4846
Time elapsed: 34m 24s, Progress: 83%, Train Perplexity: 4.3298
Time elapsed: 34m 28s, Progress: 83%, Train Perplexity: 4.4902
Time elapsed: 34m 32s, Progress: 83%, Train Perplexity: 4.3659
Time elapsed: 34m 35s, Progress: 83%, Train Perplexity: 4.5699
Finished epoch 42, Dev Perplexity: 5833.8743
Time elapsed: 34m 54s, Progress: 84%, Train Perplexity: 4.4380
Time elapsed: 34m 58s, Progress: 84%, Train Perplexity: 3.8771
Time elapsed: 35m 2s, Progress: 84%, Train Perplexity: 3.7632
Time elapsed: 35m 6s, Progress: 84%, Train Perplexity: 3.7720
Time elapsed: 35m 10s, Progress: 84%, Train Perplexity: 3.9608
Time elapsed: 35m 13s, Progress: 85%, Train Perplexity: 4.0236
Time elapsed: 35m 17s, Progress: 85%, Train Perplexity: 4.0233
Time elapsed: 35m 21s, Progress: 85%, Train Perplexity: 4.1530
Time elapsed: 35m 25s, Progress: 85%, Train Perplexity: 4.3651
Finished epoch 43, Dev Perplexity: 5401.5466
Time elapsed: 35m 43s, Progress: 86%, Train Perplexity: 4.1281
Time elapsed: 35m 47s, Progress: 86%, Train Perplexity: 3.5266
Time elapsed: 35m 51s, Progress: 86%, Train Perplexity: 3.6296
Time elapsed: 35m 55s, Progress: 86%, Train Perplexity: 3.6615
Time elapsed: 35m 59s, Progress: 86%, Train Perplexity: 3.6745
Time elapsed: 36m 3s, Progress: 87%, Train Perplexity: 4.0258
Time elapsed: 36m 6s, Progress: 87%, Train Perplexity: 3.8741
Time elapsed: 36m 10s, Progress: 87%, Train Perplexity: 3.8995
Time elapsed: 36m 14s, Progress: 87%, Train Perplexity: 3.8560
Time elapsed: 36m 18s, Progress: 87%, Train Perplexity: 4.0269
Finished epoch 44, Dev Perplexity: 4151.4733
Time elapsed: 36m 36s, Progress: 88%, Train Perplexity: 3.2780
Time elapsed: 36m 40s, Progress: 88%, Train Perplexity: 3.4263
Time elapsed: 36m 44s, Progress: 88%, Train Perplexity: 3.5530
Time elapsed: 36m 48s, Progress: 88%, Train Perplexity: 3.4061
Time elapsed: 36m 52s, Progress: 89%, Train Perplexity: 3.5413
Time elapsed: 36m 55s, Progress: 89%, Train Perplexity: 3.5634
Time elapsed: 36m 59s, Progress: 89%, Train Perplexity: 3.6677
Time elapsed: 37m 3s, Progress: 89%, Train Perplexity: 3.5598
Time elapsed: 37m 7s, Progress: 89%, Train Perplexity: 3.6993
Finished epoch 45, Dev Perplexity: 4150.4539
Time elapsed: 37m 26s, Progress: 90%, Train Perplexity: 3.5982
Time elapsed: 37m 29s, Progress: 90%, Train Perplexity: 3.1689
Time elapsed: 37m 33s, Progress: 90%, Train Perplexity: 3.2560
Time elapsed: 37m 37s, Progress: 90%, Train Perplexity: 3.3895
Time elapsed: 37m 41s, Progress: 90%, Train Perplexity: 3.3471
Time elapsed: 37m 45s, Progress: 91%, Train Perplexity: 3.2439
Time elapsed: 37m 48s, Progress: 91%, Train Perplexity: 3.3473
Time elapsed: 37m 52s, Progress: 91%, Train Perplexity: 3.3848
Time elapsed: 37m 56s, Progress: 91%, Train Perplexity: 3.5320
Finished epoch 46, Dev Perplexity: 3119.4339
Time elapsed: 38m 15s, Progress: 92%, Train Perplexity: 3.2386
Time elapsed: 38m 19s, Progress: 92%, Train Perplexity: 2.9342
Time elapsed: 38m 22s, Progress: 92%, Train Perplexity: 3.0095
Time elapsed: 38m 26s, Progress: 92%, Train Perplexity: 3.1227
Time elapsed: 38m 30s, Progress: 92%, Train Perplexity: 3.0577
Time elapsed: 38m 34s, Progress: 93%, Train Perplexity: 3.0439
Time elapsed: 38m 38s, Progress: 93%, Train Perplexity: 3.2033
Time elapsed: 38m 41s, Progress: 93%, Train Perplexity: 3.2394
Time elapsed: 38m 45s, Progress: 93%, Train Perplexity: 3.1924
Time elapsed: 38m 49s, Progress: 93%, Train Perplexity: 3.3048
Finished epoch 47, Dev Perplexity: 3018.1231
Time elapsed: 39m 8s, Progress: 94%, Train Perplexity: 2.8237
Time elapsed: 39m 11s, Progress: 94%, Train Perplexity: 2.7675
Time elapsed: 39m 15s, Progress: 94%, Train Perplexity: 2.9253
Time elapsed: 39m 19s, Progress: 94%, Train Perplexity: 2.9398
Time elapsed: 39m 23s, Progress: 95%, Train Perplexity: 2.9498
Time elapsed: 39m 27s, Progress: 95%, Train Perplexity: 2.9798
Time elapsed: 39m 30s, Progress: 95%, Train Perplexity: 2.8721
Time elapsed: 39m 34s, Progress: 95%, Train Perplexity: 3.0421
Time elapsed: 39m 38s, Progress: 95%, Train Perplexity: 2.9975
Finished epoch 48, Dev Perplexity: 2429.0407
Time elapsed: 39m 57s, Progress: 96%, Train Perplexity: 2.8337
Time elapsed: 40m 1s, Progress: 96%, Train Perplexity: 2.6175
Time elapsed: 40m 4s, Progress: 96%, Train Perplexity: 2.8246
Time elapsed: 40m 8s, Progress: 96%, Train Perplexity: 2.7964
Time elapsed: 40m 12s, Progress: 96%, Train Perplexity: 2.5768
Time elapsed: 40m 16s, Progress: 97%, Train Perplexity: 2.8509
Time elapsed: 40m 19s, Progress: 97%, Train Perplexity: 2.8014
Time elapsed: 40m 23s, Progress: 97%, Train Perplexity: 2.7515
Time elapsed: 40m 27s, Progress: 97%, Train Perplexity: 2.9301
Finished epoch 49, Dev Perplexity: 2032.1917
Time elapsed: 40m 46s, Progress: 98%, Train Perplexity: 2.7389
Time elapsed: 40m 49s, Progress: 98%, Train Perplexity: 2.4749
Time elapsed: 40m 53s, Progress: 98%, Train Perplexity: 2.6321
Time elapsed: 40m 57s, Progress: 98%, Train Perplexity: 2.6808
Time elapsed: 41m 1s, Progress: 98%, Train Perplexity: 2.6913
Time elapsed: 41m 5s, Progress: 99%, Train Perplexity: 2.6219
Time elapsed: 41m 8s, Progress: 99%, Train Perplexity: 2.5986
Time elapsed: 41m 12s, Progress: 99%, Train Perplexity: 2.5875
Time elapsed: 41m 16s, Progress: 99%, Train Perplexity: 2.7806
Time elapsed: 41m 20s, Progress: 100%, Train Perplexity: 2.8456
Finished epoch 50, Dev Perplexity: 1695.8370
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment