DL之Attention-ED:基于TF NMT利用带有Attention的 ED模型训练、测试(中英文平行语料库)实现将英文翻译为中文的LSTM翻译模型过程全记录

本文涉及的产品
图片翻译,图片翻译 100张
语种识别,语种识别 100万字符
文档翻译,文档翻译 1千页
简介: DL之Attention-ED:基于TF NMT利用带有Attention的 ED模型训练、测试(中英文平行语料库)实现将英文翻译为中文的LSTM翻译模型过程全记录

 

目录

测试输出结果

模型监控

训练过程全记录


 

 

 

测试输出结果

轻轻的我走了, 正如我轻轻的来; 我轻轻的招手, 作别...

 

 模型监控

1、SCALARS

image.png


image.png

image.png

image.png

image.png

2、IMAGES

attention_images_1/image/0   step 6,000

20181020114941609.gif

3、GRAPHS

image.png

 

训练过程全记录

1. 开始训练
2. # Job id 0
3. # Loading hparams from tmp/nmt_model_zh\hparams
4.   saving hparams to tmp/nmt_model_zh\hparams
5.   saving hparams to tmp/nmt_model_zh\best_bleu\hparams
6.   attention=scaled_luong
7.   attention_architecture=standard
8.   batch_size=128
9.   beam_width=0
10.   best_bleu=0
11.   best_bleu_dir=tmp/nmt_model_zh\best_bleu
12.   bpe_delimiter=None
13.   colocate_gradients_with_ops=True
14.   decay_factor=0.98
15.   decay_steps=10000
16.   dev_prefix=tmp/nmt_zh/dev
17.   dropout=0.2
18.   encoder_type=uni
19.   eos=</s>
20.   epoch_step=0
21.   forget_bias=1.0
22.   infer_batch_size=32
23.   init_op=uniform
24.   init_weight=0.1
25.   learning_rate=1.0
26.   length_penalty_weight=0.0
27.   log_device_placement=False
28.   max_gradient_norm=5.0
29.   max_train=0
30.   metrics=['bleu']
31.   num_buckets=5
32.   num_embeddings_partitions=0
33.   num_gpus=1
34.   num_layers=3
35.   num_residual_layers=0
36.   num_train_steps=200000
37.   num_units=256
38.   optimizer=sgd
39.   out_dir=tmp/nmt_model_zh
40.   pass_hidden_state=True
41.   random_seed=None
42.   residual=False
43.   share_vocab=False
44.   sos=<s>
45.   source_reverse=False
46.   src=en
47.   src_max_len=50
48.   src_max_len_infer=None
49.   src_vocab_file=tmp/nmt_zh/vocab.en
50.   src_vocab_size=35028
51.   start_decay_step=0
52.   steps_per_external_eval=None
53.   steps_per_stats=100
54.   test_prefix=tmp/nmt_zh/test
55.   tgt=zh
56.   tgt_max_len=50
57.   tgt_max_len_infer=None
58.   tgt_vocab_file=tmp/nmt_zh/vocab.zh
59.   tgt_vocab_size=53712
60.   time_major=True
61.   train_prefix=tmp/nmt_zh/train
62.   unit_type=lstm
63.   vocab_prefix=tmp/nmt_zh/vocab
64. # creating train graph ...
65.   num_layers = 3, num_residual_layers=0
66.   cell 0  LSTM, forget_bias=1WARNING:tensorflow:From 
67. Instructions for updating:
68. This class is deprecated, please use tf.nn.rnn_cell.LSTMCell, which supports all the feature this cell currently has. Please replace the existing code with tf.nn.rnn_cell.LSTMCell(name='basic_lstm_cell').
69.   DropoutWrapper, dropout=0.2   DeviceWrapper, device=/gpu:0
70.   cell 1  LSTM, forget_bias=1  DropoutWrapper, dropout=0.2   DeviceWrapper, device=/gpu:0
71.   cell 2  LSTM, forget_bias=1  DropoutWrapper, dropout=0.2   DeviceWrapper, device=/gpu:0
72.   cell 0  LSTM, forget_bias=1  DropoutWrapper, dropout=0.2   DeviceWrapper, device=/gpu:0
73.   cell 1  LSTM, forget_bias=1  DropoutWrapper, dropout=0.2   DeviceWrapper, device=/gpu:0
74.   cell 2  LSTM, forget_bias=1  DropoutWrapper, dropout=0.2   DeviceWrapper, device=/gpu:0
75.   start_decay_step=0, learning_rate=1, decay_steps 10000,decay_factor 0.98
76. # Trainable variables
77.   embeddings/encoder/embedding_encoder:0, (35028, 256),
78.   embeddings/decoder/embedding_decoder:0, (53712, 256),
79.   dynamic_seq2seq/encoder/rnn/multi_rnn_cell/cell_0/basic_lstm_cell/kernel:0, (512, 1024), /device:GPU:0
80.   dynamic_seq2seq/encoder/rnn/multi_rnn_cell/cell_0/basic_lstm_cell/bias:0, (1024,), /device:GPU:0
81.   dynamic_seq2seq/encoder/rnn/multi_rnn_cell/cell_1/basic_lstm_cell/kernel:0, (512, 1024), /device:GPU:0
82.   dynamic_seq2seq/encoder/rnn/multi_rnn_cell/cell_1/basic_lstm_cell/bias:0, (1024,), /device:GPU:0
83.   dynamic_seq2seq/encoder/rnn/multi_rnn_cell/cell_2/basic_lstm_cell/kernel:0, (512, 1024), /device:GPU:0
84.   dynamic_seq2seq/encoder/rnn/multi_rnn_cell/cell_2/basic_lstm_cell/bias:0, (1024,), /device:GPU:0
85.   dynamic_seq2seq/decoder/memory_layer/kernel:0, (256, 256),
86.   dynamic_seq2seq/decoder/attention/multi_rnn_cell/cell_0/basic_lstm_cell/kernel:0, (768, 1024), /device:GPU:0
87.   dynamic_seq2seq/decoder/attention/multi_rnn_cell/cell_0/basic_lstm_cell/bias:0, (1024,), /device:GPU:0
88.   dynamic_seq2seq/decoder/attention/multi_rnn_cell/cell_1/basic_lstm_cell/kernel:0, (512, 1024), /device:GPU:0
89.   dynamic_seq2seq/decoder/attention/multi_rnn_cell/cell_1/basic_lstm_cell/bias:0, (1024,), /device:GPU:0
90.   dynamic_seq2seq/decoder/attention/multi_rnn_cell/cell_2/basic_lstm_cell/kernel:0, (512, 1024), /device:GPU:0
91.   dynamic_seq2seq/decoder/attention/multi_rnn_cell/cell_2/basic_lstm_cell/bias:0, (1024,), /device:GPU:0
92.   dynamic_seq2seq/decoder/attention/luong_attention/attention_g:0, (), /device:GPU:0
93.   dynamic_seq2seq/decoder/attention/attention_layer/kernel:0, (512, 256), /device:GPU:0
94.   dynamic_seq2seq/decoder/output_projection/kernel:0, (256, 53712), /device:GPU:0
95. # creating eval graph ...
96.   num_layers = 3, num_residual_layers=0
97.   cell 0  LSTM, forget_bias=1  DeviceWrapper, device=/gpu:0
98.   cell 1  LSTM, forget_bias=1  DeviceWrapper, device=/gpu:0
99.   cell 2  LSTM, forget_bias=1  DeviceWrapper, device=/gpu:0
100.   cell 0  LSTM, forget_bias=1  DeviceWrapper, device=/gpu:0
101.   cell 1  LSTM, forget_bias=1  DeviceWrapper, device=/gpu:0
102.   cell 2  LSTM, forget_bias=1  DeviceWrapper, device=/gpu:0
103.   start_decay_step=0, learning_rate=1, decay_steps 10000,decay_factor 0.98
104. # Trainable variables
105.   embeddings/encoder/embedding_encoder:0, (35028, 256),
106.   embeddings/decoder/embedding_decoder:0, (53712, 256),
107.   dynamic_seq2seq/encoder/rnn/multi_rnn_cell/cell_0/basic_lstm_cell/kernel:0, (512, 1024), /device:GPU:0
108.   dynamic_seq2seq/encoder/rnn/multi_rnn_cell/cell_0/basic_lstm_cell/bias:0, (1024,), /device:GPU:0
109.   dynamic_seq2seq/encoder/rnn/multi_rnn_cell/cell_1/basic_lstm_cell/kernel:0, (512, 1024), /device:GPU:0
110.   dynamic_seq2seq/encoder/rnn/multi_rnn_cell/cell_1/basic_lstm_cell/bias:0, (1024,), /device:GPU:0
111.   dynamic_seq2seq/encoder/rnn/multi_rnn_cell/cell_2/basic_lstm_cell/kernel:0, (512, 1024), /device:GPU:0
112.   dynamic_seq2seq/encoder/rnn/multi_rnn_cell/cell_2/basic_lstm_cell/bias:0, (1024,), /device:GPU:0
113.   dynamic_seq2seq/decoder/memory_layer/kernel:0, (256, 256),
114.   dynamic_seq2seq/decoder/attention/multi_rnn_cell/cell_0/basic_lstm_cell/kernel:0, (768, 1024), /device:GPU:0
115.   dynamic_seq2seq/decoder/attention/multi_rnn_cell/cell_0/basic_lstm_cell/bias:0, (1024,), /device:GPU:0
116.   dynamic_seq2seq/decoder/attention/multi_rnn_cell/cell_1/basic_lstm_cell/kernel:0, (512, 1024), /device:GPU:0
117.   dynamic_seq2seq/decoder/attention/multi_rnn_cell/cell_1/basic_lstm_cell/bias:0, (1024,), /device:GPU:0
118.   dynamic_seq2seq/decoder/attention/multi_rnn_cell/cell_2/basic_lstm_cell/kernel:0, (512, 1024), /device:GPU:0
119.   dynamic_seq2seq/decoder/attention/multi_rnn_cell/cell_2/basic_lstm_cell/bias:0, (1024,), /device:GPU:0
120.   dynamic_seq2seq/decoder/attention/luong_attention/attention_g:0, (), /device:GPU:0
121.   dynamic_seq2seq/decoder/attention/attention_layer/kernel:0, (512, 256), /device:GPU:0
122.   dynamic_seq2seq/decoder/output_projection/kernel:0, (256, 53712), /device:GPU:0
123. # creating infer graph ...
124.   num_layers = 3, num_residual_layers=0
125.   cell 0  LSTM, forget_bias=1  DeviceWrapper, device=/gpu:0
126.   cell 1  LSTM, forget_bias=1  DeviceWrapper, device=/gpu:0
127.   cell 2  LSTM, forget_bias=1  DeviceWrapper, device=/gpu:0
128.   cell 0  LSTM, forget_bias=1  DeviceWrapper, device=/gpu:0
129.   cell 1  LSTM, forget_bias=1  DeviceWrapper, device=/gpu:0
130.   cell 2  LSTM, forget_bias=1  DeviceWrapper, device=/gpu:0
131.   start_decay_step=0, learning_rate=1, decay_steps 10000,decay_factor 0.98
132. # Trainable variables
133.   embeddings/encoder/embedding_encoder:0, (35028, 256),
134.   embeddings/decoder/embedding_decoder:0, (53712, 256),
135.   dynamic_seq2seq/encoder/rnn/multi_rnn_cell/cell_0/basic_lstm_cell/kernel:0, (512, 1024), /device:GPU:0
136.   dynamic_seq2seq/encoder/rnn/multi_rnn_cell/cell_0/basic_lstm_cell/bias:0, (1024,), /device:GPU:0
137.   dynamic_seq2seq/encoder/rnn/multi_rnn_cell/cell_1/basic_lstm_cell/kernel:0, (512, 1024), /device:GPU:0
138.   dynamic_seq2seq/encoder/rnn/multi_rnn_cell/cell_1/basic_lstm_cell/bias:0, (1024,), /device:GPU:0
139.   dynamic_seq2seq/encoder/rnn/multi_rnn_cell/cell_2/basic_lstm_cell/kernel:0, (512, 1024), /device:GPU:0
140.   dynamic_seq2seq/encoder/rnn/multi_rnn_cell/cell_2/basic_lstm_cell/bias:0, (1024,), /device:GPU:0
141.   dynamic_seq2seq/decoder/memory_layer/kernel:0, (256, 256),
142.   dynamic_seq2seq/decoder/attention/multi_rnn_cell/cell_0/basic_lstm_cell/kernel:0, (768, 1024), /device:GPU:0
143.   dynamic_seq2seq/decoder/attention/multi_rnn_cell/cell_0/basic_lstm_cell/bias:0, (1024,), /device:GPU:0
144.   dynamic_seq2seq/decoder/attention/multi_rnn_cell/cell_1/basic_lstm_cell/kernel:0, (512, 1024), /device:GPU:0
145.   dynamic_seq2seq/decoder/attention/multi_rnn_cell/cell_1/basic_lstm_cell/bias:0, (1024,), /device:GPU:0
146.   dynamic_seq2seq/decoder/attention/multi_rnn_cell/cell_2/basic_lstm_cell/kernel:0, (512, 1024), /device:GPU:0
147.   dynamic_seq2seq/decoder/attention/multi_rnn_cell/cell_2/basic_lstm_cell/bias:0, (1024,), /device:GPU:0
148.   dynamic_seq2seq/decoder/attention/luong_attention/attention_g:0, (), /device:GPU:0
149.   dynamic_seq2seq/decoder/attention/attention_layer/kernel:0, (512, 256), /device:GPU:0
150.   dynamic_seq2seq/decoder/output_projection/kernel:0, (256, 53712),
151. # log_file=tmp/nmt_model_zh\log_1539923931
152. 2018-10-19 12:38:51.109178: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2
153.   created train model with fresh parameters, time 0.94s
154.   created infer model with fresh parameters, time 0.50s
155. # 215
156.     src: second , the human capital , knowledge reserves , and other high - grade economic elements have formed the comparative advantages of the province in terms of technical innovation and the development of new high - tech industries .
157.     ref: 其二 , 人力 资本 , 知识 储备 等 高级 经济 要素 , 形成 了 吉林省 技术 创新 和 发展 高新技术 产业 的 比较 优势 .
158.     nmt: b'\xe9\xaa\x87 \xe6\x9d\x83\xe8\xa1\xa1 \xe6\x9d\x83\xe8\xa1\xa1 \xe6\x9d\x83\xe8\xa1\xa1 \xe6\x9d\x83\xe8\xa1\xa1 \xe6\x9d\x83\xe8\xa1\xa1 \xe5\x8d\x8a\xe8\xbe\xb9\xe5\xa4\xa9 \xe5\x8d\x8a\xe8\xbe\xb9\xe5\xa4\xa9 \xe5\x8d\x8a\xe8\xbe\xb9\xe5\xa4\xa9 \xe8\xb4\xa7\xe6\xac\xbe \xe8\xb4\xa7\xe6\xac\xbe \xe8\xb4\xa7\xe6\xac\xbe \xe8\xb4\xa7\xe6\xac\xbe \xe8\xb4\xa7\xe6\xac\xbe \xe6\xb1\x9f\xe5\x8d\xab\xe5\x9b\xbd \xe6\xb1\x9f\xe5\x8d\xab\xe5\x9b\xbd \xe6\xb1\x9f\xe5\x8d\xab\xe5\x9b\xbd \xe6\xb1\x9f\xe5\x8d\xab\xe5\x9b\xbd \xe6\xa1\x91\xe4\xb9\x85\xe7\xbe\x8e \xe6\xa1\x91\xe4\xb9\x85\xe7\xbe\x8e \xe9\x93\xb8\xe6\x88\x90\xe5\xa4\xa7\xe9\x94\x99 \xe9\x93\xb8\xe6\x88\x90\xe5\xa4\xa7\xe9\x94\x99 \xe9\x93\xb8\xe6\x88\x90\xe5\xa4\xa7\xe9\x94\x99 \xe9\x93\xb8\xe6\x88\x90\xe5\xa4\xa7\xe9\x94\x99 \xe9\x93\xb8\xe6\x88\x90\xe5\xa4\xa7\xe9\x94\x99 \xe9\x93\xb8\xe6\x88\x90\xe5\xa4\xa7\xe9\x94\x99 \xe9\x93\xb8\xe6\x88\x90\xe5\xa4\xa7\xe9\x94\x99 \xe9\x93\xb8\xe6\x88\x90\xe5\xa4\xa7\xe9\x94\x99 \xe9\x93\xb8\xe6\x88\x90\xe5\xa4\xa7\xe9\x94\x99 \xe9\x93\xb8\xe6\x88\x90\xe5\xa4\xa7\xe9\x94\x99 \xe4\xbe\x9b\xe9\x94\x80 \xe6\x9d\x8e\xe4\xb8\xbd\xe8\xbe\x89 \xe6\x9d\x8e\xe4\xb8\xbd\xe8\xbe\x89 \xe6\x9d\x8e\xe4\xb8\xbd\xe8\xbe\x89 \xe6\x97\xa5\xe6\x9c\xac\xe6\xb5\xb7 \xe6\x97\xa5\xe6\x9c\xac\xe6\xb5\xb7 \xe6\x97\xa5\xe6\x9c\xac\xe6\xb5\xb7 \xe6\x97\xa5\xe6\x9c\xac\xe6\xb5\xb7 \xe6\x97\xa5\xe6\x9c\xac\xe6\xb5\xb7 \xe6\x97\xa5\xe6\x9c\xac\xe6\xb5\xb7 \xe6\x97\xa5\xe6\x9c\xac\xe6\xb5\xb7 \xe6\x9d\x8e\xe6\xb0\xb8\xe5\x88\x9d \xe6\x9d\x8e\xe6\xb0\xb8\xe5\x88\x9d \xe6\x9d\x8e\xe6\xb0\xb8\xe5\x88\x9d \xe6\x9d\x8e\xe6\xb0\xb8\xe5\x88\x9d \xe6\x9d\x8e\xe6\xb0\xb8\xe5\x88\x9d \xe5\x9b\xb4\xe6\xad\xbc \xe5\x9b\xb4\xe6\xad\xbc \xe5\x9b\xb4\xe6\xad\xbc \xe5\x9b\xb4\xe6\xad\xbc \xe5\x9b\xb4\xe6\xad\xbc \xe6\x9d\x8e\xe5\xae\x89\xe5\xb9\xb3 \xe6\x9d\x8e\xe5\xae\x89\xe5\xb9\xb3 \xe5\xae\x8b\xe8\xb1\xab \xe5\xae\x8b\xe8\xb1\xab \xe5\xae\x8b\xe8\xb1\xab \xe5\xae\x8b\xe8\xb1\xab \xe5\xae\x8b\xe8\xb1\xab \xe5\xae\x8b\xe8\xb1\xab \xe5\xae\x8b\xe8\xb1\xab \xe6\xaf\x8f\xe5\x86\xb5\xe6\x84\x88\xe4\xb8\x8b \xe5\xbd\xb1\xe5\x93\x8d\xe5\x88\xb0 \xe5\xbd\xb1\xe5\x93\x8d\xe5\x88\xb0 \xe5\xbd\xb1\xe5\x93\x8d\xe5\x88\xb0 \xe5\xbd\xb1\xe5\x93\x8d\xe5\x88\xb0 \xe5\xbd\xb1\xe5\x93\x8d\xe5\x88\xb0 \xe7\xbb\xa7\xe4\xbd\x8d \xe4\xbe\x9d\xe9\x82\xa3\xe5\x90\x90\xe6\x8b\x89 \xe4\xbe\x9d\xe9\x82\xa3\xe5\x90\x90\xe6\x8b\x89 \xe4\xbe\x9d\xe9\x82\xa3\xe5\x90\x90\xe6\x8b\x89 \xe4\xbe\x9d\xe9\x82\xa3\xe5\x90\x90\xe6\x8b\x89 \xe4\xbe\x9d\xe9\x82\xa3\xe5\x90\x90\xe6\x8b\x89 \xe4\xbe\x9d\xe9\x82\xa3\xe5\x90\x90\xe6\x8b\x89 \xe4\xbe\x9d\xe9\x82\xa3\xe5\x90\x90\xe6\x8b\x89 \xe4\xbe\x9d\xe9\x82\xa3\xe5\x90\x90\xe6\x8b\x89 \xe4\xbe\x9d\xe9\x82\xa3\xe5\x90\x90\xe6\x8b\x89 \xe4\xbe\x9d\xe9\x82\xa3\xe5\x90\x90\xe6\x8b\x89 \xe4\xbb\xa5\xe9\x82\xbb\xe4\xb8\xba\xe5\xa3\x91'
159.   created eval model with fresh parameters, time 0.61s
160. 2018-10-19 12:38:57.783584: W tensorflow/core/framework/allocator.cc:113] Allocation of 641750976 exceeds 10% of system memory.
161. 2018-10-19 12:39:01.312543: W tensorflow/core/framework/allocator.cc:113] Allocation of 812769984 exceeds 10% of system memory.
162. 2018-10-19 12:39:05.244676: W tensorflow/core/framework/allocator.cc:113] Allocation of 660013056 exceeds 10% of system memory.
163. 2018-10-19 12:39:08.661067: W tensorflow/core/framework/allocator.cc:113] Allocation of 800523648 exceeds 10% of system memory.
164. eval dev: perplexity 53779.34, time 16s, Fri Oct 19 12:39:11 2018.
165. 2018-10-19 12:39:14.135415: W tensorflow/core/framework/allocator.cc:113] Allocation of 641750976 exceeds 10% of system memory.
166. eval test: perplexity 53779.35, time 17s, Fri Oct 19 12:39:29 2018.
167. 2018-10-19 12:39:29.544562: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
168. 2018-10-19 12:39:29.544589: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.en is already initialized.
169. 2018-10-19 12:39:29.544778: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
170.   created infer model with fresh parameters, time 0.50s
171. # Start step 0, lr 1, Fri Oct 19 12:39:29 2018
172. # Init train iterator, skipping 0 elements
173.   global step 100 lr 1 step-time 12.18s wps 0.61K ppl 370383.80 bleu 0.00
174.   global step 200 lr 1 step-time 11.44s wps 0.64K ppl 69384.04 bleu 0.00
175.   global step 300 lr 1 step-time 11.49s wps 0.65K ppl 22598.76 bleu 0.00
176.   global step 400 lr 1 step-time 11.55s wps 0.64K ppl 14178.53 bleu 0.00
177.   global step 500 lr 1 step-time 11.50s wps 0.65K ppl 10184.07 bleu 0.00
178.   global step 600 lr 1 step-time 12.13s wps 0.61K ppl 6656.94 bleu 0.00
179.   global step 700 lr 1 step-time 13.51s wps 0.55K ppl 2673.24 bleu 0.00
180. # Finished an epoch, step 785. Perform external evaluation
181. 2018-10-19 15:16:44.846427: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
182. 2018-10-19 15:16:44.846427: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
183. 2018-10-19 15:16:44.867191: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.en is already initialized.
184.   created infer model with fresh parameters, time 1.07s
185. # 64
186.     src: under the leadership of the state civil air defense office , the center will assume overall work related to china 's civil air defense engineering as well as nuclear and chemical protection .
187.     ref: 这个 中心 将 在 人民 办公室  下 , 担负 我国 人防 工程 化 防护 的 全面 工作 .
188.     nmt: b'\xe6\x96\xb9\xe5\x85\xb3\xe6\x96\xbc 2020\xe5\xb9\xb4 \xe9\x9f\xa6\xe5\xa4\x9a \xe9\x9f\xa6\xe5\xa4\x9a 19-02-2000 19-02-2000 19-02-2000 19-02-2000 \xe7\xb3\xbb\xe7\xbb\x9f\xe5\x9c\xb0 \xe7\xb3\xbb\xe7\xbb\x9f\xe5\x9c\xb0 \xe5\x97\x85\xe5\x87\xba \xe5\x97\x85\xe5\x87\xba \xe5\x97\x85\xe5\x87\xba \xe6\xac\xa1\xe4\xbe\x8d \xe6\xac\xa1\xe4\xbe\x8d \xe6\xac\xa1\xe4\xbe\x8d \xe6\xac\xa1\xe4\xbe\x8d \xe6\xac\xa1\xe4\xbe\x8d \xe5\x8f\x8d\xe8\xb4\xaa\xe5\xb1\x80 \xe5\x8f\x8d\xe8\xb4\xaa\xe5\xb1\x80 \xe5\x8f\x8d\xe8\xb4\xaa\xe5\xb1\x80 \xe8\x96\x84\xe2\x85\xb0\xe5\x82\xb2 \xe8\x96\x84\xe2\x85\xb0\xe5\x82\xb2 \xe8\x96\x84\xe2\x85\xb0\xe5\x82\xb2 \xe8\x96\x84\xe2\x85\xb0\xe5\x82\xb2 \xe5\x8f\x8c\xe5\xba\xa7 \xe5\x8f\x8c\xe5\xba\xa7 \xe6\x96\xbd\xe5\x90\x9b\xe7\x8e\x89 \xe6\x96\xbd\xe5\x90\x9b\xe7\x8e\x89 \xe6\x96\xbd\xe5\x90\x9b\xe7\x8e\x89 \xe6\x96\xbd\xe5\x90\x9b\xe7\x8e\x89 \xe6\x96\xbd\xe5\x90\x9b\xe7\x8e\x89 \xe6\x96\xbd\xe5\x90\x9b\xe7\x8e\x89 \xe6\x96\xbd\xe5\x90\x9b\xe7\x8e\x89 \xe6\x96\xbd\xe5\x90\x9b\xe7\x8e\x89 \xe9\x99\xa2 \xe9\x99\xa2 \xe9\x99\xa2 \xe9\x99\xa2 \xe9\x99\xa2 \xe9\x99\xa2 \xe9\x99\xa2 \xe9\x99\xa2 \xe9\x99\xa2 \xe9\x99\xa2 \xe9\x99\xa2 \xe9\x99\xa2 \xe9\x99\xa2 \xe5\xa4\xa7\xe4\xbd\xbf\xe7\xba\xa7 \xe5\xa4\xa7\xe4\xbd\xbf\xe7\xba\xa7 \xe5\xa4\xa7\xe4\xbd\xbf\xe7\xba\xa7 \xe5\xa4\xa7\xe4\xbd\xbf\xe7\xba\xa7 \xe5\xa4\xa7\xe4\xbd\xbf\xe7\xba\xa7 \xe5\xa4\xa7\xe4\xbd\xbf\xe7\xba\xa7 \xe5\xa4\xa7\xe4\xbd\xbf\xe7\xba\xa7 \xe5\xa4\xa7\xe4\xbd\xbf\xe7\xba\xa7 \xe5\x98\x89 \xe5\x98\x89 \xe5\x98\x89 \xe5\x98\x89 \xe5\x98\x89 \xe5\x98\x89 \xe5\x98\x89 \xe6\xb1\x82\xe6\x95\x99 \xe6\xb1\x82\xe6\x95\x99 \xe6\xb1\x82\xe6\x95\x99'
189. 2018-10-19 15:16:46.110426: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.en is already initialized.
190. 2018-10-19 15:16:46.110445: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
191. 2018-10-19 15:16:46.110444: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
192.   created infer model with fresh parameters, time 0.88s
193.   global step 800 lr 1 step-time 12.50s wps 0.57K ppl 2048.36 bleu 0.00
194.   global step 900 lr 1 step-time 13.74s wps 0.54K ppl 1883.18 bleu 0.00
195.   global step 1000 lr 1 step-time 14.76s wps 0.50K ppl 1583.37 bleu 0.00
196. # Save eval, global step 1000
197. 2018-10-19 16:07:39.926588: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
198. 2018-10-19 16:07:39.926612: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.en is already initialized.
199. 2018-10-19 16:07:39.926588: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
200.   loaded infer model parameters from tmp/nmt_model_zh\translate.ckpt-1000, time 1.62s
201.   # 125
202.     src: the united states suffered heavy casualties in the korean war in the 1950 's and the vietnam war of the 1960 's , especially the latter , and this directly caused strong antiwar sentiment within the united states and played a very great driving role in the ultimate and undoubted us defeat in the war .
203.     ref: 最终 战败 无疑 起到 了 巨大 的 推动 作用 .
204.     nmt: b'\xe4\xbb\x96 \xe8\xaf\xb4 , \xe4\xbb\x96 \xe5\x9c\xa8 \xe4\xbb\x96 , \xe4\xbb\x96 \xe7\x9a\x84 \xe7\x9a\x84 \xe7\x9a\x84 \xe7\x9a\x84 \xe7\x9a\x84 \xe7\x9a\x84 \xe7\x9a\x84 \xe7\x9a\x84 \xe7\x9a\x84 \xe7\x9a\x84 \xe7\x9a\x84 \xe7\x9a\x84 \xe7\x9a\x84 \xe7\x9a\x84 \xe7\x9a\x84 \xe7\x9a\x84 \xe7\x9a\x84 \xe7\x9a\x84 \xe7\x9a\x84 \xe7\x9a\x84 . </s>'
205. 2018-10-19 16:07:41.873322: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.en is already initialized.
206. 2018-10-19 16:07:41.873376: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
207.   loaded eval model parameters from tmp/nmt_model_zh\translate.ckpt-1000, time 1.54s
208.   eval dev: perplexity 1055.18, time 19s, Fri Oct 19 16:08:01 2018.
209.   eval test: perplexity 1055.18, time 22s, Fri Oct 19 16:08:24 2018.
210.   global step 1100 lr 1 step-time 14.13s wps 0.53K ppl 1435.07 bleu 0.00
211.   global step 1200 lr 1 step-time 13.73s wps 0.54K ppl 1276.74 bleu 0.00
212.   global step 1300 lr 1 step-time 12.87s wps 0.58K ppl 1170.78 bleu 0.00
213.   global step 1400 lr 1 step-time 14.89s wps 0.50K ppl 1066.19 bleu 0.00
214.   global step 1500 lr 1 step-time 15.34s wps 0.48K ppl 1046.93 bleu 0.00
215. # Finished an epoch, step 1570. Perform external evaluation
216. 2018-10-19 18:26:54.761219: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.en is already initialized.
217. 2018-10-19 18:26:54.784158: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
218. 2018-10-19 18:26:54.784232: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
219.   loaded infer model parameters from tmp/nmt_model_zh\translate.ckpt-1000, time 3.16s
220.   # 21
221.     src: this saying figuratively shows that this is indeed an absurd cycle .
222.     ref: 这个 说法 以 拟人化 手法 生动 地 表明 , 这 的确 是 个 怪圈 .
223.     nmt: b'\xe4\xbb\x96 \xe8\xaf\xb4 , \xe4\xbb\x96 \xe7\x9a\x84 \xe7\x9a\x84 \xe7\x9a\x84 \xe7\x9a\x84 . </s>'
224. 2018-10-19 18:26:55.501112: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
225. 2018-10-19 18:26:55.501112: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
226. 2018-10-19 18:26:55.501131: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.en is already initialized.
227.   loaded infer model parameters from tmp/nmt_model_zh\translate.ckpt-1000, time 0.59s
228. # External evaluation, global step 1000
229.   decoding to output tmp/nmt_model_zh\output_dev.
230.   done, num sentences 400, time 34s, Fri Oct 19 18:27:29 2018.
231.   bleu dev: 0.0
232.   saving hparams to tmp/nmt_model_zh\hparams
233. # External evaluation, global step 1000
234.   decoding to output tmp/nmt_model_zh\output_test.
235.   done, num sentences 400, time 33s, Fri Oct 19 18:28:04 2018.
236.   bleu test: 0.0
237.   saving hparams to tmp/nmt_model_zh\hparams
238.   global step 1600 lr 1 step-time 17.33s wps 0.41K ppl 935.22 bleu 0.00
239.   global step 1700 lr 1 step-time 17.57s wps 0.43K ppl 914.06 bleu 0.00
240.   global step 1800 lr 1 step-time 15.53s wps 0.48K ppl 833.33 bleu 0.00
241.   global step 1900 lr 1 step-time 14.46s wps 0.51K ppl 784.47 bleu 0.00
242.   global step 2000 lr 1 step-time 15.27s wps 0.49K ppl 730.18 bleu 0.00
243. # Save eval, global step 2000
244. 2018-10-19 20:21:36.887237: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.en is already initialized.
245. 2018-10-19 20:21:36.887234: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
246. 2018-10-19 20:21:36.887234: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
247.   loaded infer model parameters from tmp/nmt_model_zh\translate.ckpt-2000, time 2.08s
248.   # 171
249.     src: the sino - dprk traditional friendship created and nurtured personally by chairman mao zedong , premier zhou enlai , comrade deng xiaoping , president kim il - song and other leaders of the older generation have withstood historical tests and have taken deep roots in the hearts of peoples of the two countries .
250.     ref: 友谊 经受 了 历史 的 考验 , 已 深深 扎根 於 人民 心中 .
251.     nmt: b'\xe4\xb8\xad\xe5\x9b\xbd \xe6\x98\xaf \xe4\xb8\xad\xe5\x9b\xbd \xe7\x9a\x84 \xe5\x8f\x91\xe5\xb1\x95 \xe7\x9a\x84 \xe4\xb8\xad\xe5\x9b\xbd \xe7\x9a\x84 \xe5\x8f\x91\xe5\xb1\x95 \xe7\x9a\x84 \xe4\xb8\xad\xe5\x9b\xbd \xe7\x9a\x84 \xe5\x8f\x91\xe5\xb1\x95 \xe7\x9a\x84 \xe4\xb8\xad\xe5\x9b\xbd \xe7\x9a\x84 \xe5\x8f\x91\xe5\xb1\x95 \xe7\x9a\x84 \xe5\x8f\x91\xe5\xb1\x95 \xe5\x92\x8c \xe4\xb8\xad\xe5\x9b\xbd \xe7\x9a\x84 \xe5\x8f\x91\xe5\xb1\x95 \xe7\x9a\x84 \xe5\x8f\x91\xe5\xb1\x95 \xe5\x92\x8c \xe5\x8f\x91\xe5\xb1\x95 \xe7\x9a\x84 \xe5\x8f\x91\xe5\xb1\x95 \xe5\x92\x8c \xe5\x8f\x91\xe5\xb1\x95 \xe7\x9a\x84 \xe5\x8f\x91\xe5\xb1\x95 \xe5\x92\x8c \xe5\x8f\x91\xe5\xb1\x95 \xe7\x9a\x84 \xe5\x8f\x91\xe5\xb1\x95 \xe5\x92\x8c \xe5\x8f\x91\xe5\xb1\x95 \xe7\x9a\x84 \xe5\x8f\x91\xe5\xb1\x95 \xe7\x9a\x84 \xe5\x8f\x91\xe5\xb1\x95 . </s>'
252. 2018-10-19 20:21:37.898203: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.en is already initialized.
253. 2018-10-19 20:21:37.898825: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
254.   loaded eval model parameters from tmp/nmt_model_zh\translate.ckpt-2000, time 0.49s
255.   eval dev: perplexity 694.89, time 26s, Fri Oct 19 20:22:04 2018.
256.   eval test: perplexity 694.89, time 28s, Fri Oct 19 20:22:32 2018.
257.   global step 2100 lr 1 step-time 14.39s wps 0.51K ppl 689.72 bleu 0.00
258.   global step 2200 lr 1 step-time 13.70s wps 0.54K ppl 649.35 bleu 0.00
259.   global step 2300 lr 1 step-time 13.78s wps 0.54K ppl 609.52 bleu 0.00
260. # Finished an epoch, step 2355. Perform external evaluation
261. 2018-10-19 21:44:15.494641: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.en is already initialized.
262. 2018-10-19 21:44:15.494641: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
263. 2018-10-19 21:44:15.494654: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
264.   loaded infer model parameters from tmp/nmt_model_zh\translate.ckpt-2000, time 0.75s
265.   # 277
266.     src: the complete reunification of the motherland is the trend of the times .
267.     ref: 统一 是 大势所趋 .
268.     nmt: b'\xe4\xb8\xad\xe5\x9b\xbd \xe6\x98\xaf \xe4\xb8\xad\xe5\x9b\xbd \xe7\x9a\x84 \xe4\xb8\xad\xe5\x9b\xbd \xe7\x9a\x84 \xe5\x8f\x91\xe5\xb1\x95 \xe7\x9a\x84 \xe5\x8f\x91\xe5\xb1\x95 . </s>'
269. 2018-10-19 21:44:16.136067: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
270. 2018-10-19 21:44:16.136100: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.en is already initialized.
271. 2018-10-19 21:44:16.136067: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
272.   loaded infer model parameters from tmp/nmt_model_zh\translate.ckpt-2000, time 0.57s
273. # External evaluation, global step 2000
274.   decoding to output tmp/nmt_model_zh\output_dev.
275.   done, num sentences 400, time 44s, Fri Oct 19 21:45:00 2018.
276.   bleu dev: 0.0
277.   saving hparams to tmp/nmt_model_zh\hparams
278. # External evaluation, global step 2000
279.   decoding to output tmp/nmt_model_zh\output_test.
280.   done, num sentences 400, time 42s, Fri Oct 19 21:45:43 2018.
281.   bleu test: 0.0
282.   saving hparams to tmp/nmt_model_zh\hparams
283.   global step 2400 lr 1 step-time 13.68s wps 0.52K ppl 578.41 bleu 0.00
284.   global step 2500 lr 1 step-time 14.21s wps 0.52K ppl 532.20 bleu 0.00
285.   global step 2600 lr 1 step-time 14.41s wps 0.52K ppl 522.94 bleu 0.00
286.   global step 2700 lr 1 step-time 13.97s wps 0.53K ppl 477.39 bleu 0.00
287.   global step 2800 lr 1 step-time 11.82s wps 0.63K ppl 458.58 bleu 0.00
288.   global step 2900 lr 1 step-time 11.21s wps 0.66K ppl 426.22 bleu 0.00
289.   global step 3000 lr 1 step-time 11.19s wps 0.66K ppl 408.34 bleu 0.00
290. # Save eval, global step 3000
291. 2018-10-20 00:04:43.141117: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
292. 2018-10-20 00:04:43.141117: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
293. 2018-10-20 00:04:43.162239: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.en is already initialized.
294.   loaded infer model parameters from tmp/nmt_model_zh\translate.ckpt-3000, time 1.25s
295.   # 22
296.     src: we will seize the historic opportunity of the development of the western region and earnestly study new ideas , mechanism , and measures suited to tibet 's characteristics .
297.     ref: 我们 要 抓住大 开发 的 历史 机遇 , 认真 研究 适应 特点 的 新 思路 , 新 机制 , 新 措施 .
298.     nmt: b'\xe5\x9c\xa8 \xe8\xbf\x99 \xe4\xb8\x80 \xe9\x97\xae\xe9\xa2\x98 \xe4\xb8\x8a , \xe5\x9c\xa8 \xe7\xbb\x8f\xe6\xb5\x8e \xe4\xb8\x8a , \xe7\xbb\x8f\xe6\xb5\x8e \xe5\x8f\x91\xe5\xb1\x95 , \xe5\x8f\x91\xe5\xb1\x95 \xe5\x8f\x91\xe5\xb1\x95 , \xe5\x8f\x91\xe5\xb1\x95 \xe5\x8f\x91\xe5\xb1\x95 , \xe5\x8f\x91\xe5\xb1\x95 \xe5\x8f\x91\xe5\xb1\x95 . </s>'
299. 2018-10-20 00:04:43.760353: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.en is already initialized.
300. 2018-10-20 00:04:43.760378: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
301.   loaded eval model parameters from tmp/nmt_model_zh\translate.ckpt-3000, time 0.45s
302. eval dev: perplexity 387.54, time 15s, Sat Oct 20 00:04:59 2018.
303. eval test: perplexity 387.54, time 17s, Sat Oct 20 00:05:17 2018.
304.   global step 3100 lr 1 step-time 11.16s wps 0.66K ppl 385.81 bleu 0.00
305. # Finished an epoch, step 3140. Perform external evaluation
306. 2018-10-20 00:30:34.679555: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.en is already initialized.
307. 2018-10-20 00:30:34.680522: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
308. 2018-10-20 00:30:34.680586: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
309.   loaded infer model parameters from tmp/nmt_model_zh\translate.ckpt-3000, time 0.56s
310. # 232
311.     src: because taiwan 's new leader deliberately obscured the one - china principle and denied the " 1992 consensus " of the association for relations across the taiwan strait and the strait exchange foundation , the two sides have been unable to ease their tense relations and achieve a breakthrough in their political impasse .
312.     ref: 过去 一 年 , .
313.     nmt: b'\xe5\x9c\xa8 \xe8\xbf\x99 \xe4\xb8\x80 \xe9\x97\xae\xe9\xa2\x98 \xe4\xb8\x8a , \xe5\x9c\xa8 \xe7\xbb\x8f\xe6\xb5\x8e \xe4\xb8\x8a , \xe4\xb8\xad\xe5\x9b\xbd \xe4\xba\xba\xe6\xb0\x91 , \xe4\xba\xba\xe6\xb0\x91 , \xe4\xba\xba\xe6\xb0\x91 , \xe4\xba\xba\xe6\xb0\x91 , \xe4\xba\xba\xe6\xb0\x91 , \xe4\xba\xba\xe6\xb0\x91 , \xe4\xba\xba\xe6\xb0\x91 , \xe4\xba\xba\xe6\xb0\x91 , \xe4\xba\xba\xe6\xb0\x91 \xe7\xad\x89 \xe7\xbb\x8f\xe6\xb5\x8e \xe5\x90\x88\xe4\xbd\x9c , \xe4\xbf\x83\xe8\xbf\x9b \xe7\xbb\x8f\xe6\xb5\x8e \xe5\x8f\x91\xe5\xb1\x95 , \xe4\xbf\x83\xe8\xbf\x9b \xe7\xbb\x8f\xe6\xb5\x8e \xe5\x8f\x91\xe5\xb1\x95 . </s>'
314. 2018-10-20 00:30:35.313256: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
315. 2018-10-20 00:30:35.313274: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.en is already initialized.
316. 2018-10-20 00:30:35.313289: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
317.   loaded infer model parameters from tmp/nmt_model_zh\translate.ckpt-3000, time 0.38s
318. # External evaluation, global step 3000
319.   decoding to output tmp/nmt_model_zh\output_dev.
320.   done, num sentences 400, time 30s, Sat Oct 20 00:31:05 2018.
321.   bleu dev: 0.2
322.   saving hparams to tmp/nmt_model_zh\hparams
323. # External evaluation, global step 3000
324.   decoding to output tmp/nmt_model_zh\output_test.
325.   done, num sentences 400, time 28s, Sat Oct 20 00:31:37 2018.
326.   bleu test: 0.2
327.   saving hparams to tmp/nmt_model_zh\hparams
328.   global step 3200 lr 1 step-time 10.83s wps 0.66K ppl 357.57 bleu 0.17
329.   global step 3300 lr 1 step-time 11.15s wps 0.66K ppl 342.12 bleu 0.17
330.   global step 3400 lr 1 step-time 11.37s wps 0.66K ppl 325.57 bleu 0.17
331.   global step 3500 lr 1 step-time 11.14s wps 0.66K ppl 312.20 bleu 0.17
332.   global step 3600 lr 1 step-time 11.14s wps 0.66K ppl 307.63 bleu 0.17
333.   global step 3700 lr 1 step-time 11.28s wps 0.66K ppl 296.23 bleu 0.17
334.   global step 3800 lr 1 step-time 11.14s wps 0.66K ppl 278.77 bleu 0.17
335.   global step 3900 lr 1 step-time 11.11s wps 0.67K ppl 275.24 bleu 0.17
336. # Finished an epoch, step 3925. Perform external evaluation
337. 2018-10-20 02:57:30.646778: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
338. 2018-10-20 02:57:30.646780: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
339. 2018-10-20 02:57:30.646874: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.en is already initialized.
340.   loaded infer model parameters from tmp/nmt_model_zh\translate.ckpt-3000, time 0.54s
341.   # 120
342.     src: in china , socialism has succeeded in greatly emancipating and developing productive forces in society .
343.     ref: 生产力 得到 极大 解放 和 发展 .
344.     nmt: b'\xe5\x9c\xa8 \xe8\xbf\x99 \xe4\xb8\x80 \xe9\x97\xae\xe9\xa2\x98 \xe4\xb8\x8a , \xe6\x88\x91\xe4\xbb\xac \xe8\xa6\x81 \xe5\x8a\xa0\xe5\xbc\xba \xe5\x8f\x91\xe5\xb1\x95 \xe5\x92\x8c \xe5\x8f\x91\xe5\xb1\x95 . </s>'
345. 2018-10-20 02:57:31.074731: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.en is already initialized.
346. 2018-10-20 02:57:31.074738: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
347. 2018-10-20 02:57:31.074738: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
348.   loaded infer model parameters from tmp/nmt_model_zh\translate.ckpt-3000, time 0.35s
349. # External evaluation, global step 3000
350.   decoding to output tmp/nmt_model_zh\output_dev.
351.   done, num sentences 400, time 29s, Sat Oct 20 02:58:01 2018.
352.   bleu dev: 0.2
353.   saving hparams to tmp/nmt_model_zh\hparams
354. # External evaluation, global step 3000
355.   decoding to output tmp/nmt_model_zh\output_test.
356.   done, num sentences 400, time 30s, Sat Oct 20 02:58:31 2018.
357.   bleu test: 0.2
358.   saving hparams to tmp/nmt_model_zh\hparams
359.   global step 4000 lr 1 step-time 10.69s wps 0.66K ppl 252.99 bleu 0.17
360. # Save eval, global step 4000
361. 2018-10-20 03:12:28.024198: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.en is already initialized.
362. 2018-10-20 03:12:28.024218: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
363. 2018-10-20 03:12:28.024223: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
364.   loaded infer model parameters from tmp/nmt_model_zh\translate.ckpt-4000, time 0.36s
365.   # 224
366.     src: the ninth guangdong provincial people 's congress concluded its meeting this morning . pursuant to the proposal of guangdong provincial higher people 's court president lu botao , ling qiman and li yifeng were appointed vice presidents of the provincial higher people 's court .
367.     ref: 今天 上午 闭幕 的  .
368.     nmt: b'\xe5\x9c\xa8 \xe8\xb0\x88\xe5\x88\xb0 \xe4\xb8\xad\xe5\x9b\xbd \xe5\x85\xb1\xe4\xba\xa7\xe5\x85\x9a , \xe5\x9b\xbd\xe5\x8a\xa1\xe9\x99\xa2 \xe6\x80\xbb\xe7\x90\x86 \xe6\x9c\xb1\xe9\x8e\x94\xe5\x9f\xba \xe4\xbb\x8a\xe5\xa4\xa9 \xe4\xb8\x8a\xe5\x8d\x88 , \xe5\x9b\xbd\xe5\x8a\xa1\xe9\x99\xa2 \xe5\x89\xaf\xe6\x80\xbb\xe7\x90\x86 \xe9\x92\xb1\xe5\x85\xb6\xe7\x90\x9b , \xe5\x9b\xbd\xe5\x8a\xa1\xe9\x99\xa2 , \xe4\xb9\xa6\xe8\xae\xb0\xe5\xa4\x84 , \xe4\xb9\xa6\xe8\xae\xb0\xe5\xa4\x84 , \xe4\xb9\xa6\xe8\xae\xb0\xe5\xa4\x84 , \xe6\x9b\xbe\xe5\xba\x86\xe7\xba\xa2 , \xe6\x9b\xbe\xe5\xba\x86\xe7\xba\xa2 , \xe6\x9b\xbe\xe5\xba\x86\xe7\xba\xa2 , \xe6\x9b\xbe\xe5\xba\x86\xe7\xba\xa2 , \xe6\x9b\xbe\xe5\xba\x86\xe7\xba\xa2 , \xe6\x9b\xbe\xe5\xba\x86\xe7\xba\xa2 , \xe6\x9b\xbe\xe5\xba\x86\xe7\xba\xa2 , \xe6\x9b\xbe\xe5\xba\x86\xe7\xba\xa2 , \xe6\x9b\xbe\xe5\xba\x86\xe7\xba\xa2 , \xe6\x9b\xbe\xe5\xba\x86\xe7\xba\xa2 . </s>'
369. 2018-10-20 03:12:28.632761: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
370. 2018-10-20 03:12:28.632820: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.en is already initialized.
371.   loaded eval model parameters from tmp/nmt_model_zh\translate.ckpt-4000, time 0.36s
372. eval dev: perplexity 252.28, time 15s, Sat Oct 20 03:12:44 2018.
373. eval test: perplexity 252.28, time 16s, Sat Oct 20 03:13:00 2018.
374.   global step 4100 lr 1 step-time 11.56s wps 0.64K ppl 246.02 bleu 0.17
375.   global step 4200 lr 1 step-time 11.29s wps 0.66K ppl 237.87 bleu 0.17
376.   global step 4300 lr 1 step-time 11.32s wps 0.66K ppl 233.55 bleu 0.17
377.   global step 4400 lr 1 step-time 11.36s wps 0.66K ppl 221.90 bleu 0.17
378.   global step 4500 lr 1 step-time 11.06s wps 0.66K ppl 224.83 bleu 0.17
379.   global step 4600 lr 1 step-time 11.32s wps 0.66K ppl 211.41 bleu 0.17
380.   global step 4700 lr 1 step-time 11.21s wps 0.66K ppl 213.31 bleu 0.17
381. # Finished an epoch, step 4710. Perform external evaluation
382. 2018-10-20 05:26:06.693124: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
383. 2018-10-20 05:26:06.693124: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.en is already initialized.
384. 2018-10-20 05:26:06.693124: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
385.   loaded infer model parameters from tmp/nmt_model_zh\translate.ckpt-4000, time 0.84s
386. # 33
387.     src: peng peiyun and he luli respectively attended the plenary meetings of the zhejiang and beijing delegations .
388.     ref: 云 参加 了 浙江 代表团 的 全体 会议 . 何鲁丽 参加 了 北京 代表团 的 全体 会议 .
389.     nmt: b'\xe4\xb8\xad\xe5\x9b\xbd \xe6\x94\xbf\xe5\xba\x9c \xe5\x92\x8c \xe4\xba\xba\xe6\xb0\x91 , \xe4\xba\xba\xe6\xb0\x91 , \xe7\x9b\xb4\xe8\xbe\x96\xe5\xb8\x82 , \xe7\x9b\xb4\xe8\xbe\x96\xe5\xb8\x82 , \xe7\x9b\xb4\xe8\xbe\x96\xe5\xb8\x82 . </s>'
390. 2018-10-20 05:26:07.132553: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.en is already initialized.
391. 2018-10-20 05:26:07.132553: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
392. 2018-10-20 05:26:07.132618: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
393.   loaded infer model parameters from tmp/nmt_model_zh\translate.ckpt-4000, time 0.37s
394. # External evaluation, global step 4000
395.   decoding to output tmp/nmt_model_zh\output_dev.
396. done, num sentences 400, time 45s, Sat Oct 20 05:26:52 2018.
397.   bleu dev: 0.2
398.   saving hparams to tmp/nmt_model_zh\hparams
399. # External evaluation, global step 4000
400.   decoding to output tmp/nmt_model_zh\output_test.
401. done, num sentences 400, time 43s, Sat Oct 20 05:27:39 2018.
402.   bleu test: 0.2
403.   saving hparams to tmp/nmt_model_zh\hparams
404.   global step 4800 lr 1 step-time 10.88s wps 0.66K ppl 199.44 bleu 0.18
405.   global step 4900 lr 1 step-time 11.33s wps 0.66K ppl 192.33 bleu 0.18
406.   global step 5000 lr 1 step-time 11.41s wps 0.65K ppl 186.30 bleu 0.18
407. # Save eval, global step 5000
408. 2018-10-20 06:22:32.135143: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.en is already initialized.
409. 2018-10-20 06:22:32.135150: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
410. 2018-10-20 06:22:32.135223: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
411.   loaded infer model parameters from tmp/nmt_model_zh\translate.ckpt-5000, time 0.76s
412. # 119
413.     src: as a result they could only avoid the differences of opinion and delay the solutions .
414.     ref: 结果 只能 是 回避 分歧 , 推迟 解决 .
415.     nmt: b'\xe4\xbb\x8e \xe6\xa0\xb9\xe6\x9c\xac\xe4\xb8\x8a \xe7\x9c\x8b , \xe6\x88\x91\xe4\xbb\xac \xe5\xbf\x85\xe9\xa1\xbb \xe5\x9d\x9a\xe6\x8c\x81 \xe4\xb8\x80\xe4\xb8\xaa \xe4\xb8\xad\xe5\x9b\xbd \xe7\x9a\x84 \xe5\x8e\x9f\xe5\x88\x99 . </s>'
416. 2018-10-20 06:22:32.937075: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
417. 2018-10-20 06:22:32.937075: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.en is already initialized.
418.   loaded eval model parameters from tmp/nmt_model_zh\translate.ckpt-5000, time 0.94s
419. eval dev: perplexity 188.49, time 15s, Sat Oct 20 06:22:49 2018.
420. eval test: perplexity 188.49, time 17s, Sat Oct 20 06:23:06 2018.
421. 2018-10-20 06:23:09.323110: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
422. 2018-10-20 06:23:09.323133: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.en is already initialized.
423. 2018-10-20 06:23:09.323159: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
424.   loaded infer model parameters from tmp/nmt_model_zh\translate.ckpt-5000, time 0.36s
425. # 208
426.     src: by the middle of this century , we should build the western region into a new region enjoying economic prosperity , social progress , a good living environment , and nationality unity , and having green mountains and clean rivers .
427.     ref: 到 21世纪中叶 , 将 西部 地区 建成 一个 经济 繁荣 , 社会 进步 , 生活 安定 , 民族团结 , 山川 秀美 的 新 西部 .
428.     nmt: b'\xe4\xbb\x8e \xe4\xbb\x8a\xe5\xb9\xb4 \xe4\xb8\x8b\xe5\x8d\x8a\xe5\xb9\xb4 \xe5\xbc\x80\xe5\xa7\x8b , \xe6\x88\x91\xe5\x9b\xbd \xe5\xb0\x86 \xe7\xbb\xa7\xe7\xbb\xad \xe5\x8a\xa0\xe5\xbc\xba \xe5\xaf\xb9 \xe8\xa5\xbf\xe9\x83\xa8 \xe5\x9c\xb0\xe5\x8c\xba \xe7\x9a\x84 \xe5\x8f\x91\xe5\xb1\x95 , \xe5\xb9\xb6 \xe8\xbf\x9b\xe4\xb8\x80\xe6\xad\xa5 \xe6\x8e\xa8\xe5\x8a\xa8 \xe4\xb8\xad\xe5\x9b\xbd \xe6\x94\xbf\xe5\xba\x9c \xe5\xaf\xb9 \xe4\xb8\xad\xe5\x9b\xbd \xe4\xba\xba\xe6\xb0\x91 \xe7\x9a\x84 \xe5\x88\xa9\xe7\x9b\x8a . </s>'
429. 2018-10-20 06:23:09.887178: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.en is already initialized.
430. 2018-10-20 06:23:09.887162: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
431. 2018-10-20 06:23:09.887366: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
432.   loaded infer model parameters from tmp/nmt_model_zh\translate.ckpt-5000, time 0.77s
433. # External evaluation, global step 5000
434.   decoding to output tmp/nmt_model_zh\output_dev.
435. done, num sentences 400, time 29s, Sat Oct 20 06:23:39 2018.
436.   bleu dev: 0.4
437.   saving hparams to tmp/nmt_model_zh\hparams
438. # External evaluation, global step 5000
439.   decoding to output tmp/nmt_model_zh\output_test.
440. done, num sentences 400, time 29s, Sat Oct 20 06:24:12 2018.
441.   bleu test: 0.4
442.   saving hparams to tmp/nmt_model_zh\hparams
443.   global step 5100 lr 1 step-time 11.38s wps 0.65K ppl 185.57 bleu 0.39
444.   global step 5200 lr 1 step-time 11.30s wps 0.65K ppl 178.95 bleu 0.39
445.   global step 5300 lr 1 step-time 11.18s wps 0.66K ppl 174.15 bleu 0.39
446.   global step 5400 lr 1 step-time 11.39s wps 0.66K ppl 173.02 bleu 0.39
447. # Finished an epoch, step 5495. Perform external evaluation
448. 2018-10-20 07:57:03.000782: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
449. 2018-10-20 07:57:03.000804: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.en is already initialized.
450. 2018-10-20 07:57:03.000921: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
451.   loaded infer model parameters from tmp/nmt_model_zh\translate.ckpt-5000, time 0.82s
452. # 6
453.     src: we should truly realize " six unifications , " especially the unification of fees to improve efficiency . we should never allow the practice of " ostensibly combining the three types of inspections while failing to do so in private , " and should truly combine the " three types of inspections . "
454.     ref: 真正 实现率 , 决不能 出现 那种 " 明合暗 不合 , 面 合心 不合 " 的 现 像 , 把真正 落实 下去 .
455.     nmt: b'\xe6\x88\x91\xe4\xbb\xac \xe8\xa6\x81 \xe6\x8a\x8a " \xe4\xb8\x89 \xe4\xbb\xa3\xe8\xa1\xa8 " \xe4\xb8\xba \xe6\xa0\xb8\xe5\xbf\x83 \xe7\x9a\x84 \xe6\x80\x9d\xe6\x83\xb3 \xe6\x94\xbf\xe6\xb2\xbb \xe5\xbb\xba\xe8\xae\xbe \xe4\xbd\x9c\xe4\xb8\xba \xe6\x96\xb0 \xe7\x9a\x84 \xe4\xbc\x9f\xe5\xa4\xa7 \xe7\x9a\x84 \xe4\xbc\x9f\xe5\xa4\xa7 \xe4\xbb\xbb\xe5\x8a\xa1 , \xe6\x88\x91\xe4\xbb\xac \xe5\x85\x9a \xe7\x9a\x84 \xe5\xbb\xba\xe8\xae\xbe \xe6\x98\xaf \xe4\xb8\x80\xe4\xb8\xaa \xe6\x96\xb0 \xe7\x9a\x84 \xe5\x8f\x91\xe5\xb1\x95 \xe6\x97\xb6\xe6\x9c\x9f , \xe6\x98\xaf \xe6\x88\x91\xe4\xbb\xac \xe5\x85\x9a \xe7\x9a\x84 \xe5\xbb\xba\xe8\xae\xbe \xe7\x9a\x84 \xe4\xbc\x9f\xe5\xa4\xa7 \xe4\xbb\xbb\xe5\x8a\xa1 . </s>'
456. 2018-10-20 07:57:04.047105: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.en is already initialized.
457. 2018-10-20 07:57:04.047147: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
458. 2018-10-20 07:57:04.047113: I tensorflow/core/kernels/lookup_util.cc:376] Table trying to initialize from file tmp/nmt_zh/vocab.zh is already initialized.
459.   loaded infer model parameters from tmp/nmt_model_zh\translate.ckpt-5000, time 0.79s
460. # External evaluation, global step 5000
461.   decoding to output tmp/nmt_model_zh\output_dev.
462. done, num sentences 400, time 30s, Sat Oct 20 07:57:35 2018.
463.   bleu dev: 0.4
464.   saving hparams to tmp/nmt_model_zh\hparams
465. # External evaluation, global step 5000
466.   decoding to output tmp/nmt_model_zh\output_test.
467. done, num sentences 400, time 30s, Sat Oct 20 07:58:06 2018.
468.   bleu test: 0.4
469.   saving hparams to tmp/nmt_model_zh\hparams
470.   global step 5500 lr 1 step-time 10.96s wps 0.64K ppl 170.63 bleu 0.39
471.   global step 5600 lr 1 step-time 11.52s wps 0.65K ppl 154.09 bleu 0.39
472. 
473. ……
474. 
475.   global step 6100


相关文章
|
21天前
|
人工智能 JSON 算法
Qwen2.5-Coder 系列模型在 PAI-QuickStart 的训练、评测、压缩及部署实践
阿里云的人工智能平台 PAI,作为一站式、 AI Native 的大模型与 AIGC 工程平台,为开发者和企业客户提供了 Qwen2.5-Coder 系列模型的全链路最佳实践。本文以Qwen2.5-Coder-32B为例,详细介绍在 PAI-QuickStart 完成 Qwen2.5-Coder 的训练、评测和快速部署。
Qwen2.5-Coder 系列模型在 PAI-QuickStart 的训练、评测、压缩及部署实践
|
5天前
|
编解码 机器人 测试技术
技术实践 | 使用 PAI+LLaMA Factory 微调 Qwen2-VL 模型快速搭建专业领域知识问答机器人
Qwen2-VL是一款具备高级图像和视频理解能力的多模态模型,支持多种语言,适用于多模态应用开发。通过PAI和LLaMA Factory框架,用户可以轻松微调Qwen2-VL模型,快速构建文旅领域的知识问答机器人。本教程详细介绍了从模型部署、微调到对话测试的全过程,帮助开发者高效实现定制化多模态应用。
|
25天前
|
机器学习/深度学习 PyTorch API
优化注意力层提升 Transformer 模型效率:通过改进注意力机制降低机器学习成本
Transformer架构自2017年被Vaswani等人提出以来,凭借其核心的注意力机制,已成为AI领域的重大突破。该机制允许模型根据任务需求灵活聚焦于输入的不同部分,极大地增强了对复杂语言和结构的理解能力。起初主要应用于自然语言处理,Transformer迅速扩展至语音识别、计算机视觉等多领域,展现出强大的跨学科应用潜力。然而,随着模型规模的增长,注意力层的高计算复杂度成为发展瓶颈。为此,本文探讨了在PyTorch生态系统中优化注意力层的各种技术,
49 6
优化注意力层提升 Transformer 模型效率:通过改进注意力机制降低机器学习成本
|
14天前
|
机器学习/深度学习 人工智能 算法
人工智能浪潮下的编程实践:构建你的第一个机器学习模型
在人工智能的巨浪中,每个人都有机会成为弄潮儿。本文将带你一探究竟,从零基础开始,用最易懂的语言和步骤,教你如何构建属于自己的第一个机器学习模型。不需要复杂的数学公式,也不必担心编程难题,只需跟随我们的步伐,一起探索这个充满魔力的AI世界。
33 12
|
21天前
|
机器学习/深度学习 Python
机器学习中评估模型性能的重要工具——混淆矩阵和ROC曲线。混淆矩阵通过真正例、假正例等指标展示模型预测情况
本文介绍了机器学习中评估模型性能的重要工具——混淆矩阵和ROC曲线。混淆矩阵通过真正例、假正例等指标展示模型预测情况,而ROC曲线则通过假正率和真正率评估二分类模型性能。文章还提供了Python中的具体实现示例,展示了如何计算和使用这两种工具来评估模型。
41 8
|
21天前
|
机器学习/深度学习 Python
机器学习中模型选择和优化的关键技术——交叉验证与网格搜索
本文深入探讨了机器学习中模型选择和优化的关键技术——交叉验证与网格搜索。介绍了K折交叉验证、留一交叉验证等方法,以及网格搜索的原理和步骤,展示了如何结合两者在Python中实现模型参数的优化,并强调了使用时需注意的计算成本、过拟合风险等问题。
41 6
|
23天前
|
机器学习/深度学习 数据采集 算法
从零到一:构建高效机器学习模型的旅程####
在探索技术深度与广度的征途中,我深刻体会到技术创新既在于理论的飞跃,更在于实践的积累。本文将通过一个具体案例,分享我在构建高效机器学习模型过程中的实战经验,包括数据预处理、特征工程、模型选择与优化等关键环节,旨在为读者提供一个从零开始构建并优化机器学习模型的实用指南。 ####
|
28天前
|
人工智能 边缘计算 JSON
DistilQwen2 蒸馏小模型在 PAI-QuickStart 的训练、评测、压缩及部署实践
本文详细介绍在 PAI 平台使用 DistilQwen2 蒸馏小模型的全链路最佳实践。
|
25天前
|
机器学习/深度学习 人工智能 算法
探索机器学习中的线性回归模型
本文深入探讨了机器学习中广泛使用的线性回归模型,从其基本概念和数学原理出发,逐步引导读者理解模型的构建、训练及评估过程。通过实例分析与代码演示,本文旨在为初学者提供一个清晰的学习路径,帮助他们在实践中更好地应用线性回归模型解决实际问题。
|
1月前
|
机器学习/深度学习 数据采集 监控
如何使用机器学习模型来自动化评估数据质量?
如何使用机器学习模型来自动化评估数据质量?