File size: 23,951 Bytes
e94235f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
2023-10-16 12:45:52,143 ----------------------------------------------------------------------------------------------------
2023-10-16 12:45:52,144 Model: "SequenceTagger(
  (embeddings): TransformerWordEmbeddings(
    (model): BertModel(
      (embeddings): BertEmbeddings(
        (word_embeddings): Embedding(32001, 768)
        (position_embeddings): Embedding(512, 768)
        (token_type_embeddings): Embedding(2, 768)
        (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
        (dropout): Dropout(p=0.1, inplace=False)
      )
      (encoder): BertEncoder(
        (layer): ModuleList(
          (0-11): 12 x BertLayer(
            (attention): BertAttention(
              (self): BertSelfAttention(
                (query): Linear(in_features=768, out_features=768, bias=True)
                (key): Linear(in_features=768, out_features=768, bias=True)
                (value): Linear(in_features=768, out_features=768, bias=True)
                (dropout): Dropout(p=0.1, inplace=False)
              )
              (output): BertSelfOutput(
                (dense): Linear(in_features=768, out_features=768, bias=True)
                (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
                (dropout): Dropout(p=0.1, inplace=False)
              )
            )
            (intermediate): BertIntermediate(
              (dense): Linear(in_features=768, out_features=3072, bias=True)
              (intermediate_act_fn): GELUActivation()
            )
            (output): BertOutput(
              (dense): Linear(in_features=3072, out_features=768, bias=True)
              (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
              (dropout): Dropout(p=0.1, inplace=False)
            )
          )
        )
      )
      (pooler): BertPooler(
        (dense): Linear(in_features=768, out_features=768, bias=True)
        (activation): Tanh()
      )
    )
  )
  (locked_dropout): LockedDropout(p=0.5)
  (linear): Linear(in_features=768, out_features=17, bias=True)
  (loss_function): CrossEntropyLoss()
)"
2023-10-16 12:45:52,144 ----------------------------------------------------------------------------------------------------
2023-10-16 12:45:52,145 MultiCorpus: 7142 train + 698 dev + 2570 test sentences
 - NER_HIPE_2022 Corpus: 7142 train + 698 dev + 2570 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fr/with_doc_seperator
2023-10-16 12:45:52,145 ----------------------------------------------------------------------------------------------------
2023-10-16 12:45:52,145 Train:  7142 sentences
2023-10-16 12:45:52,145         (train_with_dev=False, train_with_test=False)
2023-10-16 12:45:52,145 ----------------------------------------------------------------------------------------------------
2023-10-16 12:45:52,145 Training Params:
2023-10-16 12:45:52,145  - learning_rate: "3e-05" 
2023-10-16 12:45:52,145  - mini_batch_size: "8"
2023-10-16 12:45:52,145  - max_epochs: "10"
2023-10-16 12:45:52,145  - shuffle: "True"
2023-10-16 12:45:52,145 ----------------------------------------------------------------------------------------------------
2023-10-16 12:45:52,145 Plugins:
2023-10-16 12:45:52,145  - LinearScheduler | warmup_fraction: '0.1'
2023-10-16 12:45:52,145 ----------------------------------------------------------------------------------------------------
2023-10-16 12:45:52,145 Final evaluation on model from best epoch (best-model.pt)
2023-10-16 12:45:52,145  - metric: "('micro avg', 'f1-score')"
2023-10-16 12:45:52,145 ----------------------------------------------------------------------------------------------------
2023-10-16 12:45:52,145 Computation:
2023-10-16 12:45:52,145  - compute on device: cuda:0
2023-10-16 12:45:52,145  - embedding storage: none
2023-10-16 12:45:52,145 ----------------------------------------------------------------------------------------------------
2023-10-16 12:45:52,145 Model training base path: "hmbench-newseye/fr-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4"
2023-10-16 12:45:52,145 ----------------------------------------------------------------------------------------------------
2023-10-16 12:45:52,145 ----------------------------------------------------------------------------------------------------
2023-10-16 12:46:00,431 epoch 1 - iter 89/893 - loss 2.55511339 - time (sec): 8.28 - samples/sec: 2909.27 - lr: 0.000003 - momentum: 0.000000
2023-10-16 12:46:07,327 epoch 1 - iter 178/893 - loss 1.61235074 - time (sec): 15.18 - samples/sec: 3224.15 - lr: 0.000006 - momentum: 0.000000
2023-10-16 12:46:14,109 epoch 1 - iter 267/893 - loss 1.21408701 - time (sec): 21.96 - samples/sec: 3363.25 - lr: 0.000009 - momentum: 0.000000
2023-10-16 12:46:21,076 epoch 1 - iter 356/893 - loss 0.98196317 - time (sec): 28.93 - samples/sec: 3425.09 - lr: 0.000012 - momentum: 0.000000
2023-10-16 12:46:27,573 epoch 1 - iter 445/893 - loss 0.84813604 - time (sec): 35.43 - samples/sec: 3461.42 - lr: 0.000015 - momentum: 0.000000
2023-10-16 12:46:34,458 epoch 1 - iter 534/893 - loss 0.73801609 - time (sec): 42.31 - samples/sec: 3505.89 - lr: 0.000018 - momentum: 0.000000
2023-10-16 12:46:41,293 epoch 1 - iter 623/893 - loss 0.65814262 - time (sec): 49.15 - samples/sec: 3539.49 - lr: 0.000021 - momentum: 0.000000
2023-10-16 12:46:48,032 epoch 1 - iter 712/893 - loss 0.60020660 - time (sec): 55.89 - samples/sec: 3546.66 - lr: 0.000024 - momentum: 0.000000
2023-10-16 12:46:54,998 epoch 1 - iter 801/893 - loss 0.55223454 - time (sec): 62.85 - samples/sec: 3546.82 - lr: 0.000027 - momentum: 0.000000
2023-10-16 12:47:01,870 epoch 1 - iter 890/893 - loss 0.51136141 - time (sec): 69.72 - samples/sec: 3557.52 - lr: 0.000030 - momentum: 0.000000
2023-10-16 12:47:02,083 ----------------------------------------------------------------------------------------------------
2023-10-16 12:47:02,083 EPOCH 1 done: loss 0.5100 - lr: 0.000030
2023-10-16 12:47:04,728 DEV : loss 0.11952730268239975 - f1-score (micro avg)  0.6946
2023-10-16 12:47:04,745 saving best model
2023-10-16 12:47:05,151 ----------------------------------------------------------------------------------------------------
2023-10-16 12:47:12,143 epoch 2 - iter 89/893 - loss 0.11614387 - time (sec): 6.99 - samples/sec: 3765.39 - lr: 0.000030 - momentum: 0.000000
2023-10-16 12:47:18,713 epoch 2 - iter 178/893 - loss 0.11703984 - time (sec): 13.56 - samples/sec: 3705.32 - lr: 0.000029 - momentum: 0.000000
2023-10-16 12:47:25,911 epoch 2 - iter 267/893 - loss 0.11593650 - time (sec): 20.76 - samples/sec: 3665.76 - lr: 0.000029 - momentum: 0.000000
2023-10-16 12:47:33,003 epoch 2 - iter 356/893 - loss 0.11898584 - time (sec): 27.85 - samples/sec: 3575.78 - lr: 0.000029 - momentum: 0.000000
2023-10-16 12:47:40,167 epoch 2 - iter 445/893 - loss 0.11380050 - time (sec): 35.01 - samples/sec: 3595.39 - lr: 0.000028 - momentum: 0.000000
2023-10-16 12:47:46,495 epoch 2 - iter 534/893 - loss 0.11094488 - time (sec): 41.34 - samples/sec: 3622.50 - lr: 0.000028 - momentum: 0.000000
2023-10-16 12:47:53,145 epoch 2 - iter 623/893 - loss 0.11041525 - time (sec): 47.99 - samples/sec: 3621.92 - lr: 0.000028 - momentum: 0.000000
2023-10-16 12:47:59,812 epoch 2 - iter 712/893 - loss 0.10964599 - time (sec): 54.66 - samples/sec: 3624.04 - lr: 0.000027 - momentum: 0.000000
2023-10-16 12:48:06,674 epoch 2 - iter 801/893 - loss 0.10815829 - time (sec): 61.52 - samples/sec: 3626.54 - lr: 0.000027 - momentum: 0.000000
2023-10-16 12:48:13,580 epoch 2 - iter 890/893 - loss 0.10593895 - time (sec): 68.43 - samples/sec: 3619.64 - lr: 0.000027 - momentum: 0.000000
2023-10-16 12:48:13,861 ----------------------------------------------------------------------------------------------------
2023-10-16 12:48:13,861 EPOCH 2 done: loss 0.1057 - lr: 0.000027
2023-10-16 12:48:18,043 DEV : loss 0.10355333238840103 - f1-score (micro avg)  0.7514
2023-10-16 12:48:18,059 saving best model
2023-10-16 12:48:18,614 ----------------------------------------------------------------------------------------------------
2023-10-16 12:48:25,574 epoch 3 - iter 89/893 - loss 0.06161232 - time (sec): 6.96 - samples/sec: 3465.77 - lr: 0.000026 - momentum: 0.000000
2023-10-16 12:48:33,414 epoch 3 - iter 178/893 - loss 0.05931401 - time (sec): 14.80 - samples/sec: 3535.00 - lr: 0.000026 - momentum: 0.000000
2023-10-16 12:48:40,156 epoch 3 - iter 267/893 - loss 0.06373319 - time (sec): 21.54 - samples/sec: 3554.90 - lr: 0.000026 - momentum: 0.000000
2023-10-16 12:48:46,889 epoch 3 - iter 356/893 - loss 0.06176266 - time (sec): 28.27 - samples/sec: 3583.15 - lr: 0.000025 - momentum: 0.000000
2023-10-16 12:48:53,244 epoch 3 - iter 445/893 - loss 0.06271250 - time (sec): 34.63 - samples/sec: 3585.52 - lr: 0.000025 - momentum: 0.000000
2023-10-16 12:48:59,780 epoch 3 - iter 534/893 - loss 0.06428874 - time (sec): 41.16 - samples/sec: 3599.68 - lr: 0.000025 - momentum: 0.000000
2023-10-16 12:49:06,534 epoch 3 - iter 623/893 - loss 0.06294223 - time (sec): 47.92 - samples/sec: 3638.07 - lr: 0.000024 - momentum: 0.000000
2023-10-16 12:49:12,998 epoch 3 - iter 712/893 - loss 0.06425486 - time (sec): 54.38 - samples/sec: 3652.85 - lr: 0.000024 - momentum: 0.000000
2023-10-16 12:49:19,311 epoch 3 - iter 801/893 - loss 0.06474582 - time (sec): 60.70 - samples/sec: 3655.96 - lr: 0.000024 - momentum: 0.000000
2023-10-16 12:49:26,634 epoch 3 - iter 890/893 - loss 0.06378154 - time (sec): 68.02 - samples/sec: 3646.66 - lr: 0.000023 - momentum: 0.000000
2023-10-16 12:49:26,907 ----------------------------------------------------------------------------------------------------
2023-10-16 12:49:26,907 EPOCH 3 done: loss 0.0640 - lr: 0.000023
2023-10-16 12:49:31,625 DEV : loss 0.12550972402095795 - f1-score (micro avg)  0.7835
2023-10-16 12:49:31,640 saving best model
2023-10-16 12:49:32,212 ----------------------------------------------------------------------------------------------------
2023-10-16 12:49:39,003 epoch 4 - iter 89/893 - loss 0.04612929 - time (sec): 6.79 - samples/sec: 3721.21 - lr: 0.000023 - momentum: 0.000000
2023-10-16 12:49:45,662 epoch 4 - iter 178/893 - loss 0.04670224 - time (sec): 13.45 - samples/sec: 3748.05 - lr: 0.000023 - momentum: 0.000000
2023-10-16 12:49:52,483 epoch 4 - iter 267/893 - loss 0.04673840 - time (sec): 20.27 - samples/sec: 3689.22 - lr: 0.000022 - momentum: 0.000000
2023-10-16 12:49:59,270 epoch 4 - iter 356/893 - loss 0.04627007 - time (sec): 27.06 - samples/sec: 3678.67 - lr: 0.000022 - momentum: 0.000000
2023-10-16 12:50:06,421 epoch 4 - iter 445/893 - loss 0.04530410 - time (sec): 34.21 - samples/sec: 3675.49 - lr: 0.000022 - momentum: 0.000000
2023-10-16 12:50:13,138 epoch 4 - iter 534/893 - loss 0.04603225 - time (sec): 40.92 - samples/sec: 3680.66 - lr: 0.000021 - momentum: 0.000000
2023-10-16 12:50:19,943 epoch 4 - iter 623/893 - loss 0.04600487 - time (sec): 47.73 - samples/sec: 3649.96 - lr: 0.000021 - momentum: 0.000000
2023-10-16 12:50:26,861 epoch 4 - iter 712/893 - loss 0.04696416 - time (sec): 54.65 - samples/sec: 3624.44 - lr: 0.000021 - momentum: 0.000000
2023-10-16 12:50:33,465 epoch 4 - iter 801/893 - loss 0.04647850 - time (sec): 61.25 - samples/sec: 3635.01 - lr: 0.000020 - momentum: 0.000000
2023-10-16 12:50:40,628 epoch 4 - iter 890/893 - loss 0.04696665 - time (sec): 68.41 - samples/sec: 3626.05 - lr: 0.000020 - momentum: 0.000000
2023-10-16 12:50:40,819 ----------------------------------------------------------------------------------------------------
2023-10-16 12:50:40,819 EPOCH 4 done: loss 0.0469 - lr: 0.000020
2023-10-16 12:50:45,113 DEV : loss 0.14134925603866577 - f1-score (micro avg)  0.7836
2023-10-16 12:50:45,130 saving best model
2023-10-16 12:50:45,700 ----------------------------------------------------------------------------------------------------
2023-10-16 12:50:52,682 epoch 5 - iter 89/893 - loss 0.02860295 - time (sec): 6.98 - samples/sec: 3752.60 - lr: 0.000020 - momentum: 0.000000
2023-10-16 12:50:59,524 epoch 5 - iter 178/893 - loss 0.03232203 - time (sec): 13.82 - samples/sec: 3696.27 - lr: 0.000019 - momentum: 0.000000
2023-10-16 12:51:06,414 epoch 5 - iter 267/893 - loss 0.03327893 - time (sec): 20.71 - samples/sec: 3549.51 - lr: 0.000019 - momentum: 0.000000
2023-10-16 12:51:13,137 epoch 5 - iter 356/893 - loss 0.03240572 - time (sec): 27.44 - samples/sec: 3555.85 - lr: 0.000019 - momentum: 0.000000
2023-10-16 12:51:20,796 epoch 5 - iter 445/893 - loss 0.03319621 - time (sec): 35.09 - samples/sec: 3560.42 - lr: 0.000018 - momentum: 0.000000
2023-10-16 12:51:27,811 epoch 5 - iter 534/893 - loss 0.03409730 - time (sec): 42.11 - samples/sec: 3572.12 - lr: 0.000018 - momentum: 0.000000
2023-10-16 12:51:34,927 epoch 5 - iter 623/893 - loss 0.03454180 - time (sec): 49.23 - samples/sec: 3592.93 - lr: 0.000018 - momentum: 0.000000
2023-10-16 12:51:41,334 epoch 5 - iter 712/893 - loss 0.03417524 - time (sec): 55.63 - samples/sec: 3589.84 - lr: 0.000017 - momentum: 0.000000
2023-10-16 12:51:48,137 epoch 5 - iter 801/893 - loss 0.03472578 - time (sec): 62.44 - samples/sec: 3578.49 - lr: 0.000017 - momentum: 0.000000
2023-10-16 12:51:54,801 epoch 5 - iter 890/893 - loss 0.03490357 - time (sec): 69.10 - samples/sec: 3588.45 - lr: 0.000017 - momentum: 0.000000
2023-10-16 12:51:55,044 ----------------------------------------------------------------------------------------------------
2023-10-16 12:51:55,044 EPOCH 5 done: loss 0.0348 - lr: 0.000017
2023-10-16 12:51:59,372 DEV : loss 0.17483924329280853 - f1-score (micro avg)  0.7945
2023-10-16 12:51:59,392 saving best model
2023-10-16 12:51:59,954 ----------------------------------------------------------------------------------------------------
2023-10-16 12:52:06,819 epoch 6 - iter 89/893 - loss 0.02626343 - time (sec): 6.86 - samples/sec: 3670.10 - lr: 0.000016 - momentum: 0.000000
2023-10-16 12:52:13,623 epoch 6 - iter 178/893 - loss 0.03047495 - time (sec): 13.67 - samples/sec: 3610.14 - lr: 0.000016 - momentum: 0.000000
2023-10-16 12:52:20,194 epoch 6 - iter 267/893 - loss 0.03214194 - time (sec): 20.24 - samples/sec: 3673.48 - lr: 0.000016 - momentum: 0.000000
2023-10-16 12:52:27,056 epoch 6 - iter 356/893 - loss 0.03185153 - time (sec): 27.10 - samples/sec: 3671.57 - lr: 0.000015 - momentum: 0.000000
2023-10-16 12:52:34,130 epoch 6 - iter 445/893 - loss 0.02932483 - time (sec): 34.17 - samples/sec: 3672.16 - lr: 0.000015 - momentum: 0.000000
2023-10-16 12:52:40,891 epoch 6 - iter 534/893 - loss 0.02993883 - time (sec): 40.93 - samples/sec: 3659.20 - lr: 0.000015 - momentum: 0.000000
2023-10-16 12:52:47,512 epoch 6 - iter 623/893 - loss 0.02884411 - time (sec): 47.56 - samples/sec: 3683.06 - lr: 0.000014 - momentum: 0.000000
2023-10-16 12:52:54,195 epoch 6 - iter 712/893 - loss 0.02850536 - time (sec): 54.24 - samples/sec: 3668.78 - lr: 0.000014 - momentum: 0.000000
2023-10-16 12:53:01,056 epoch 6 - iter 801/893 - loss 0.02830873 - time (sec): 61.10 - samples/sec: 3657.82 - lr: 0.000014 - momentum: 0.000000
2023-10-16 12:53:08,041 epoch 6 - iter 890/893 - loss 0.02857723 - time (sec): 68.08 - samples/sec: 3645.85 - lr: 0.000013 - momentum: 0.000000
2023-10-16 12:53:08,211 ----------------------------------------------------------------------------------------------------
2023-10-16 12:53:08,212 EPOCH 6 done: loss 0.0287 - lr: 0.000013
2023-10-16 12:53:12,859 DEV : loss 0.19940702617168427 - f1-score (micro avg)  0.7727
2023-10-16 12:53:12,875 ----------------------------------------------------------------------------------------------------
2023-10-16 12:53:19,596 epoch 7 - iter 89/893 - loss 0.02500460 - time (sec): 6.72 - samples/sec: 3541.40 - lr: 0.000013 - momentum: 0.000000
2023-10-16 12:53:26,259 epoch 7 - iter 178/893 - loss 0.02277466 - time (sec): 13.38 - samples/sec: 3591.82 - lr: 0.000013 - momentum: 0.000000
2023-10-16 12:53:33,085 epoch 7 - iter 267/893 - loss 0.02090527 - time (sec): 20.21 - samples/sec: 3616.67 - lr: 0.000012 - momentum: 0.000000
2023-10-16 12:53:39,862 epoch 7 - iter 356/893 - loss 0.02011475 - time (sec): 26.99 - samples/sec: 3636.71 - lr: 0.000012 - momentum: 0.000000
2023-10-16 12:53:46,751 epoch 7 - iter 445/893 - loss 0.02069675 - time (sec): 33.88 - samples/sec: 3629.54 - lr: 0.000012 - momentum: 0.000000
2023-10-16 12:53:53,473 epoch 7 - iter 534/893 - loss 0.02079766 - time (sec): 40.60 - samples/sec: 3626.98 - lr: 0.000011 - momentum: 0.000000
2023-10-16 12:54:00,256 epoch 7 - iter 623/893 - loss 0.02128472 - time (sec): 47.38 - samples/sec: 3641.22 - lr: 0.000011 - momentum: 0.000000
2023-10-16 12:54:07,283 epoch 7 - iter 712/893 - loss 0.02254890 - time (sec): 54.41 - samples/sec: 3648.20 - lr: 0.000011 - momentum: 0.000000
2023-10-16 12:54:13,922 epoch 7 - iter 801/893 - loss 0.02269062 - time (sec): 61.05 - samples/sec: 3639.12 - lr: 0.000010 - momentum: 0.000000
2023-10-16 12:54:20,953 epoch 7 - iter 890/893 - loss 0.02194304 - time (sec): 68.08 - samples/sec: 3640.63 - lr: 0.000010 - momentum: 0.000000
2023-10-16 12:54:21,171 ----------------------------------------------------------------------------------------------------
2023-10-16 12:54:21,171 EPOCH 7 done: loss 0.0220 - lr: 0.000010
2023-10-16 12:54:25,732 DEV : loss 0.19999033212661743 - f1-score (micro avg)  0.7912
2023-10-16 12:54:25,755 ----------------------------------------------------------------------------------------------------
2023-10-16 12:54:32,512 epoch 8 - iter 89/893 - loss 0.01168160 - time (sec): 6.76 - samples/sec: 3588.80 - lr: 0.000010 - momentum: 0.000000
2023-10-16 12:54:39,559 epoch 8 - iter 178/893 - loss 0.01427644 - time (sec): 13.80 - samples/sec: 3476.32 - lr: 0.000009 - momentum: 0.000000
2023-10-16 12:54:46,333 epoch 8 - iter 267/893 - loss 0.01591158 - time (sec): 20.58 - samples/sec: 3543.59 - lr: 0.000009 - momentum: 0.000000
2023-10-16 12:54:53,376 epoch 8 - iter 356/893 - loss 0.01732768 - time (sec): 27.62 - samples/sec: 3605.76 - lr: 0.000009 - momentum: 0.000000
2023-10-16 12:55:00,019 epoch 8 - iter 445/893 - loss 0.01967539 - time (sec): 34.26 - samples/sec: 3629.85 - lr: 0.000008 - momentum: 0.000000
2023-10-16 12:55:07,011 epoch 8 - iter 534/893 - loss 0.01853002 - time (sec): 41.25 - samples/sec: 3637.34 - lr: 0.000008 - momentum: 0.000000
2023-10-16 12:55:13,914 epoch 8 - iter 623/893 - loss 0.01808121 - time (sec): 48.16 - samples/sec: 3620.58 - lr: 0.000008 - momentum: 0.000000
2023-10-16 12:55:20,357 epoch 8 - iter 712/893 - loss 0.01781687 - time (sec): 54.60 - samples/sec: 3625.12 - lr: 0.000007 - momentum: 0.000000
2023-10-16 12:55:27,362 epoch 8 - iter 801/893 - loss 0.01794196 - time (sec): 61.61 - samples/sec: 3612.50 - lr: 0.000007 - momentum: 0.000000
2023-10-16 12:55:34,282 epoch 8 - iter 890/893 - loss 0.01754005 - time (sec): 68.53 - samples/sec: 3621.46 - lr: 0.000007 - momentum: 0.000000
2023-10-16 12:55:34,488 ----------------------------------------------------------------------------------------------------
2023-10-16 12:55:34,489 EPOCH 8 done: loss 0.0175 - lr: 0.000007
2023-10-16 12:55:38,684 DEV : loss 0.20922400057315826 - f1-score (micro avg)  0.8038
2023-10-16 12:55:38,700 saving best model
2023-10-16 12:55:39,186 ----------------------------------------------------------------------------------------------------
2023-10-16 12:55:45,857 epoch 9 - iter 89/893 - loss 0.00982904 - time (sec): 6.67 - samples/sec: 3621.30 - lr: 0.000006 - momentum: 0.000000
2023-10-16 12:55:52,816 epoch 9 - iter 178/893 - loss 0.01033872 - time (sec): 13.63 - samples/sec: 3702.90 - lr: 0.000006 - momentum: 0.000000
2023-10-16 12:56:00,091 epoch 9 - iter 267/893 - loss 0.01214149 - time (sec): 20.90 - samples/sec: 3649.27 - lr: 0.000006 - momentum: 0.000000
2023-10-16 12:56:06,660 epoch 9 - iter 356/893 - loss 0.01195331 - time (sec): 27.47 - samples/sec: 3654.09 - lr: 0.000005 - momentum: 0.000000
2023-10-16 12:56:13,921 epoch 9 - iter 445/893 - loss 0.01203697 - time (sec): 34.73 - samples/sec: 3632.41 - lr: 0.000005 - momentum: 0.000000
2023-10-16 12:56:20,404 epoch 9 - iter 534/893 - loss 0.01189900 - time (sec): 41.22 - samples/sec: 3644.72 - lr: 0.000005 - momentum: 0.000000
2023-10-16 12:56:27,229 epoch 9 - iter 623/893 - loss 0.01250389 - time (sec): 48.04 - samples/sec: 3649.49 - lr: 0.000004 - momentum: 0.000000
2023-10-16 12:56:34,221 epoch 9 - iter 712/893 - loss 0.01230678 - time (sec): 55.03 - samples/sec: 3640.52 - lr: 0.000004 - momentum: 0.000000
2023-10-16 12:56:40,816 epoch 9 - iter 801/893 - loss 0.01224281 - time (sec): 61.63 - samples/sec: 3637.30 - lr: 0.000004 - momentum: 0.000000
2023-10-16 12:56:47,523 epoch 9 - iter 890/893 - loss 0.01274520 - time (sec): 68.34 - samples/sec: 3631.08 - lr: 0.000003 - momentum: 0.000000
2023-10-16 12:56:47,696 ----------------------------------------------------------------------------------------------------
2023-10-16 12:56:47,696 EPOCH 9 done: loss 0.0128 - lr: 0.000003
2023-10-16 12:56:52,288 DEV : loss 0.21604780852794647 - f1-score (micro avg)  0.7879
2023-10-16 12:56:52,314 ----------------------------------------------------------------------------------------------------
2023-10-16 12:56:59,282 epoch 10 - iter 89/893 - loss 0.00670445 - time (sec): 6.97 - samples/sec: 3448.15 - lr: 0.000003 - momentum: 0.000000
2023-10-16 12:57:05,865 epoch 10 - iter 178/893 - loss 0.00793855 - time (sec): 13.55 - samples/sec: 3566.79 - lr: 0.000003 - momentum: 0.000000
2023-10-16 12:57:12,745 epoch 10 - iter 267/893 - loss 0.00788264 - time (sec): 20.43 - samples/sec: 3594.85 - lr: 0.000002 - momentum: 0.000000
2023-10-16 12:57:19,300 epoch 10 - iter 356/893 - loss 0.00725310 - time (sec): 26.98 - samples/sec: 3573.51 - lr: 0.000002 - momentum: 0.000000
2023-10-16 12:57:26,244 epoch 10 - iter 445/893 - loss 0.00808299 - time (sec): 33.93 - samples/sec: 3599.24 - lr: 0.000002 - momentum: 0.000000
2023-10-16 12:57:32,839 epoch 10 - iter 534/893 - loss 0.00893934 - time (sec): 40.52 - samples/sec: 3616.94 - lr: 0.000001 - momentum: 0.000000
2023-10-16 12:57:39,669 epoch 10 - iter 623/893 - loss 0.00976410 - time (sec): 47.35 - samples/sec: 3625.68 - lr: 0.000001 - momentum: 0.000000
2023-10-16 12:57:46,950 epoch 10 - iter 712/893 - loss 0.01000539 - time (sec): 54.63 - samples/sec: 3628.40 - lr: 0.000001 - momentum: 0.000000
2023-10-16 12:57:54,074 epoch 10 - iter 801/893 - loss 0.00983205 - time (sec): 61.76 - samples/sec: 3622.71 - lr: 0.000000 - momentum: 0.000000
2023-10-16 12:58:00,863 epoch 10 - iter 890/893 - loss 0.00969370 - time (sec): 68.55 - samples/sec: 3621.54 - lr: 0.000000 - momentum: 0.000000
2023-10-16 12:58:01,015 ----------------------------------------------------------------------------------------------------
2023-10-16 12:58:01,015 EPOCH 10 done: loss 0.0097 - lr: 0.000000
2023-10-16 12:58:05,669 DEV : loss 0.2217082679271698 - f1-score (micro avg)  0.7952
2023-10-16 12:58:06,080 ----------------------------------------------------------------------------------------------------
2023-10-16 12:58:06,081 Loading model from best epoch ...
2023-10-16 12:58:07,923 SequenceTagger predicts: Dictionary with 17 tags: O, S-PER, B-PER, E-PER, I-PER, S-LOC, B-LOC, E-LOC, I-LOC, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
2023-10-16 12:58:17,287 
Results:
- F-score (micro) 0.6906
- F-score (macro) 0.6137
- Accuracy 0.5455

By class:
              precision    recall  f1-score   support

         LOC     0.7176    0.6868    0.7018      1095
         PER     0.7696    0.7658    0.7677      1012
         ORG     0.4312    0.5882    0.4976       357
   HumanProd     0.4082    0.6061    0.4878        33

   micro avg     0.6781    0.7036    0.6906      2497
   macro avg     0.5816    0.6617    0.6137      2497
weighted avg     0.6936    0.7036    0.6965      2497

2023-10-16 12:58:17,287 ----------------------------------------------------------------------------------------------------