File size: 23,936 Bytes
dd58e09
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
244
245
2023-10-13 10:44:10,449 ----------------------------------------------------------------------------------------------------
2023-10-13 10:44:10,450 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=25, bias=True)
  (loss_function): CrossEntropyLoss()
)"
2023-10-13 10:44:10,450 ----------------------------------------------------------------------------------------------------
2023-10-13 10:44:10,450 MultiCorpus: 966 train + 219 dev + 204 test sentences
 - NER_HIPE_2022 Corpus: 966 train + 219 dev + 204 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/fr/with_doc_seperator
2023-10-13 10:44:10,450 ----------------------------------------------------------------------------------------------------
2023-10-13 10:44:10,450 Train:  966 sentences
2023-10-13 10:44:10,450         (train_with_dev=False, train_with_test=False)
2023-10-13 10:44:10,450 ----------------------------------------------------------------------------------------------------
2023-10-13 10:44:10,450 Training Params:
2023-10-13 10:44:10,450  - learning_rate: "3e-05" 
2023-10-13 10:44:10,450  - mini_batch_size: "8"
2023-10-13 10:44:10,450  - max_epochs: "10"
2023-10-13 10:44:10,450  - shuffle: "True"
2023-10-13 10:44:10,450 ----------------------------------------------------------------------------------------------------
2023-10-13 10:44:10,450 Plugins:
2023-10-13 10:44:10,450  - LinearScheduler | warmup_fraction: '0.1'
2023-10-13 10:44:10,451 ----------------------------------------------------------------------------------------------------
2023-10-13 10:44:10,451 Final evaluation on model from best epoch (best-model.pt)
2023-10-13 10:44:10,451  - metric: "('micro avg', 'f1-score')"
2023-10-13 10:44:10,451 ----------------------------------------------------------------------------------------------------
2023-10-13 10:44:10,451 Computation:
2023-10-13 10:44:10,451  - compute on device: cuda:0
2023-10-13 10:44:10,451  - embedding storage: none
2023-10-13 10:44:10,451 ----------------------------------------------------------------------------------------------------
2023-10-13 10:44:10,451 Model training base path: "hmbench-ajmc/fr-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2"
2023-10-13 10:44:10,451 ----------------------------------------------------------------------------------------------------
2023-10-13 10:44:10,451 ----------------------------------------------------------------------------------------------------
2023-10-13 10:44:11,132 epoch 1 - iter 12/121 - loss 3.51321597 - time (sec): 0.68 - samples/sec: 3523.65 - lr: 0.000003 - momentum: 0.000000
2023-10-13 10:44:11,834 epoch 1 - iter 24/121 - loss 3.29488879 - time (sec): 1.38 - samples/sec: 3581.80 - lr: 0.000006 - momentum: 0.000000
2023-10-13 10:44:12,512 epoch 1 - iter 36/121 - loss 2.99548579 - time (sec): 2.06 - samples/sec: 3394.31 - lr: 0.000009 - momentum: 0.000000
2023-10-13 10:44:13,299 epoch 1 - iter 48/121 - loss 2.41287355 - time (sec): 2.85 - samples/sec: 3445.12 - lr: 0.000012 - momentum: 0.000000
2023-10-13 10:44:14,009 epoch 1 - iter 60/121 - loss 2.04321003 - time (sec): 3.56 - samples/sec: 3458.41 - lr: 0.000015 - momentum: 0.000000
2023-10-13 10:44:14,676 epoch 1 - iter 72/121 - loss 1.79618228 - time (sec): 4.22 - samples/sec: 3498.27 - lr: 0.000018 - momentum: 0.000000
2023-10-13 10:44:15,433 epoch 1 - iter 84/121 - loss 1.62352073 - time (sec): 4.98 - samples/sec: 3452.27 - lr: 0.000021 - momentum: 0.000000
2023-10-13 10:44:16,153 epoch 1 - iter 96/121 - loss 1.48154791 - time (sec): 5.70 - samples/sec: 3470.15 - lr: 0.000024 - momentum: 0.000000
2023-10-13 10:44:16,929 epoch 1 - iter 108/121 - loss 1.36723369 - time (sec): 6.48 - samples/sec: 3430.89 - lr: 0.000027 - momentum: 0.000000
2023-10-13 10:44:17,617 epoch 1 - iter 120/121 - loss 1.27088232 - time (sec): 7.16 - samples/sec: 3446.16 - lr: 0.000030 - momentum: 0.000000
2023-10-13 10:44:17,666 ----------------------------------------------------------------------------------------------------
2023-10-13 10:44:17,666 EPOCH 1 done: loss 1.2702 - lr: 0.000030
2023-10-13 10:44:18,571 DEV : loss 0.3508635461330414 - f1-score (micro avg)  0.4129
2023-10-13 10:44:18,575 saving best model
2023-10-13 10:44:18,959 ----------------------------------------------------------------------------------------------------
2023-10-13 10:44:19,661 epoch 2 - iter 12/121 - loss 0.34872943 - time (sec): 0.70 - samples/sec: 3702.46 - lr: 0.000030 - momentum: 0.000000
2023-10-13 10:44:20,357 epoch 2 - iter 24/121 - loss 0.32940314 - time (sec): 1.40 - samples/sec: 3499.83 - lr: 0.000029 - momentum: 0.000000
2023-10-13 10:44:21,042 epoch 2 - iter 36/121 - loss 0.31951398 - time (sec): 2.08 - samples/sec: 3464.17 - lr: 0.000029 - momentum: 0.000000
2023-10-13 10:44:21,756 epoch 2 - iter 48/121 - loss 0.29887519 - time (sec): 2.80 - samples/sec: 3542.35 - lr: 0.000029 - momentum: 0.000000
2023-10-13 10:44:22,543 epoch 2 - iter 60/121 - loss 0.29679138 - time (sec): 3.58 - samples/sec: 3476.77 - lr: 0.000028 - momentum: 0.000000
2023-10-13 10:44:23,260 epoch 2 - iter 72/121 - loss 0.28969882 - time (sec): 4.30 - samples/sec: 3440.26 - lr: 0.000028 - momentum: 0.000000
2023-10-13 10:44:23,980 epoch 2 - iter 84/121 - loss 0.27870861 - time (sec): 5.02 - samples/sec: 3409.28 - lr: 0.000028 - momentum: 0.000000
2023-10-13 10:44:24,704 epoch 2 - iter 96/121 - loss 0.27071546 - time (sec): 5.74 - samples/sec: 3390.97 - lr: 0.000027 - momentum: 0.000000
2023-10-13 10:44:25,448 epoch 2 - iter 108/121 - loss 0.26477634 - time (sec): 6.49 - samples/sec: 3389.90 - lr: 0.000027 - momentum: 0.000000
2023-10-13 10:44:26,188 epoch 2 - iter 120/121 - loss 0.25613493 - time (sec): 7.23 - samples/sec: 3400.07 - lr: 0.000027 - momentum: 0.000000
2023-10-13 10:44:26,236 ----------------------------------------------------------------------------------------------------
2023-10-13 10:44:26,236 EPOCH 2 done: loss 0.2562 - lr: 0.000027
2023-10-13 10:44:26,979 DEV : loss 0.16044385731220245 - f1-score (micro avg)  0.723
2023-10-13 10:44:26,984 saving best model
2023-10-13 10:44:27,448 ----------------------------------------------------------------------------------------------------
2023-10-13 10:44:28,185 epoch 3 - iter 12/121 - loss 0.14606844 - time (sec): 0.73 - samples/sec: 3114.53 - lr: 0.000026 - momentum: 0.000000
2023-10-13 10:44:28,881 epoch 3 - iter 24/121 - loss 0.14489480 - time (sec): 1.43 - samples/sec: 3285.70 - lr: 0.000026 - momentum: 0.000000
2023-10-13 10:44:29,605 epoch 3 - iter 36/121 - loss 0.14732529 - time (sec): 2.15 - samples/sec: 3318.56 - lr: 0.000026 - momentum: 0.000000
2023-10-13 10:44:30,356 epoch 3 - iter 48/121 - loss 0.14501133 - time (sec): 2.90 - samples/sec: 3269.44 - lr: 0.000025 - momentum: 0.000000
2023-10-13 10:44:31,043 epoch 3 - iter 60/121 - loss 0.14185856 - time (sec): 3.59 - samples/sec: 3255.97 - lr: 0.000025 - momentum: 0.000000
2023-10-13 10:44:31,792 epoch 3 - iter 72/121 - loss 0.14156836 - time (sec): 4.34 - samples/sec: 3332.27 - lr: 0.000025 - momentum: 0.000000
2023-10-13 10:44:32,468 epoch 3 - iter 84/121 - loss 0.13930022 - time (sec): 5.02 - samples/sec: 3344.01 - lr: 0.000024 - momentum: 0.000000
2023-10-13 10:44:33,197 epoch 3 - iter 96/121 - loss 0.13453425 - time (sec): 5.75 - samples/sec: 3387.72 - lr: 0.000024 - momentum: 0.000000
2023-10-13 10:44:33,915 epoch 3 - iter 108/121 - loss 0.13486151 - time (sec): 6.46 - samples/sec: 3379.71 - lr: 0.000024 - momentum: 0.000000
2023-10-13 10:44:34,740 epoch 3 - iter 120/121 - loss 0.13199486 - time (sec): 7.29 - samples/sec: 3369.40 - lr: 0.000023 - momentum: 0.000000
2023-10-13 10:44:34,795 ----------------------------------------------------------------------------------------------------
2023-10-13 10:44:34,796 EPOCH 3 done: loss 0.1318 - lr: 0.000023
2023-10-13 10:44:35,547 DEV : loss 0.1352299004793167 - f1-score (micro avg)  0.7884
2023-10-13 10:44:35,552 saving best model
2023-10-13 10:44:36,012 ----------------------------------------------------------------------------------------------------
2023-10-13 10:44:36,811 epoch 4 - iter 12/121 - loss 0.07294851 - time (sec): 0.79 - samples/sec: 3234.72 - lr: 0.000023 - momentum: 0.000000
2023-10-13 10:44:37,552 epoch 4 - iter 24/121 - loss 0.07174414 - time (sec): 1.53 - samples/sec: 3466.33 - lr: 0.000023 - momentum: 0.000000
2023-10-13 10:44:38,288 epoch 4 - iter 36/121 - loss 0.08161430 - time (sec): 2.26 - samples/sec: 3388.07 - lr: 0.000022 - momentum: 0.000000
2023-10-13 10:44:38,982 epoch 4 - iter 48/121 - loss 0.08164795 - time (sec): 2.96 - samples/sec: 3429.61 - lr: 0.000022 - momentum: 0.000000
2023-10-13 10:44:39,666 epoch 4 - iter 60/121 - loss 0.08391292 - time (sec): 3.64 - samples/sec: 3433.24 - lr: 0.000022 - momentum: 0.000000
2023-10-13 10:44:40,399 epoch 4 - iter 72/121 - loss 0.08008740 - time (sec): 4.38 - samples/sec: 3375.57 - lr: 0.000021 - momentum: 0.000000
2023-10-13 10:44:41,111 epoch 4 - iter 84/121 - loss 0.08382287 - time (sec): 5.09 - samples/sec: 3391.95 - lr: 0.000021 - momentum: 0.000000
2023-10-13 10:44:41,929 epoch 4 - iter 96/121 - loss 0.08503397 - time (sec): 5.91 - samples/sec: 3365.88 - lr: 0.000021 - momentum: 0.000000
2023-10-13 10:44:42,633 epoch 4 - iter 108/121 - loss 0.08523333 - time (sec): 6.61 - samples/sec: 3356.44 - lr: 0.000020 - momentum: 0.000000
2023-10-13 10:44:43,351 epoch 4 - iter 120/121 - loss 0.08617500 - time (sec): 7.33 - samples/sec: 3356.76 - lr: 0.000020 - momentum: 0.000000
2023-10-13 10:44:43,398 ----------------------------------------------------------------------------------------------------
2023-10-13 10:44:43,398 EPOCH 4 done: loss 0.0857 - lr: 0.000020
2023-10-13 10:44:44,144 DEV : loss 0.12280040979385376 - f1-score (micro avg)  0.8232
2023-10-13 10:44:44,148 saving best model
2023-10-13 10:44:44,604 ----------------------------------------------------------------------------------------------------
2023-10-13 10:44:45,380 epoch 5 - iter 12/121 - loss 0.08088457 - time (sec): 0.77 - samples/sec: 3184.56 - lr: 0.000020 - momentum: 0.000000
2023-10-13 10:44:46,150 epoch 5 - iter 24/121 - loss 0.06668373 - time (sec): 1.54 - samples/sec: 3319.15 - lr: 0.000019 - momentum: 0.000000
2023-10-13 10:44:46,881 epoch 5 - iter 36/121 - loss 0.06423692 - time (sec): 2.27 - samples/sec: 3309.82 - lr: 0.000019 - momentum: 0.000000
2023-10-13 10:44:47,662 epoch 5 - iter 48/121 - loss 0.06070212 - time (sec): 3.05 - samples/sec: 3322.60 - lr: 0.000019 - momentum: 0.000000
2023-10-13 10:44:48,333 epoch 5 - iter 60/121 - loss 0.06194933 - time (sec): 3.72 - samples/sec: 3333.25 - lr: 0.000018 - momentum: 0.000000
2023-10-13 10:44:49,100 epoch 5 - iter 72/121 - loss 0.06304159 - time (sec): 4.49 - samples/sec: 3306.08 - lr: 0.000018 - momentum: 0.000000
2023-10-13 10:44:49,782 epoch 5 - iter 84/121 - loss 0.06360609 - time (sec): 5.17 - samples/sec: 3331.77 - lr: 0.000018 - momentum: 0.000000
2023-10-13 10:44:50,526 epoch 5 - iter 96/121 - loss 0.06367570 - time (sec): 5.91 - samples/sec: 3322.57 - lr: 0.000017 - momentum: 0.000000
2023-10-13 10:44:51,275 epoch 5 - iter 108/121 - loss 0.06269414 - time (sec): 6.66 - samples/sec: 3347.64 - lr: 0.000017 - momentum: 0.000000
2023-10-13 10:44:51,987 epoch 5 - iter 120/121 - loss 0.06168175 - time (sec): 7.38 - samples/sec: 3336.12 - lr: 0.000017 - momentum: 0.000000
2023-10-13 10:44:52,047 ----------------------------------------------------------------------------------------------------
2023-10-13 10:44:52,047 EPOCH 5 done: loss 0.0614 - lr: 0.000017
2023-10-13 10:44:52,821 DEV : loss 0.12571021914482117 - f1-score (micro avg)  0.8123
2023-10-13 10:44:52,826 ----------------------------------------------------------------------------------------------------
2023-10-13 10:44:53,502 epoch 6 - iter 12/121 - loss 0.06029715 - time (sec): 0.68 - samples/sec: 3405.79 - lr: 0.000016 - momentum: 0.000000
2023-10-13 10:44:54,238 epoch 6 - iter 24/121 - loss 0.04547772 - time (sec): 1.41 - samples/sec: 3318.54 - lr: 0.000016 - momentum: 0.000000
2023-10-13 10:44:55,072 epoch 6 - iter 36/121 - loss 0.04428078 - time (sec): 2.25 - samples/sec: 3264.89 - lr: 0.000016 - momentum: 0.000000
2023-10-13 10:44:55,796 epoch 6 - iter 48/121 - loss 0.04916090 - time (sec): 2.97 - samples/sec: 3366.87 - lr: 0.000015 - momentum: 0.000000
2023-10-13 10:44:56,487 epoch 6 - iter 60/121 - loss 0.04902978 - time (sec): 3.66 - samples/sec: 3328.14 - lr: 0.000015 - momentum: 0.000000
2023-10-13 10:44:57,276 epoch 6 - iter 72/121 - loss 0.04833321 - time (sec): 4.45 - samples/sec: 3337.54 - lr: 0.000015 - momentum: 0.000000
2023-10-13 10:44:58,011 epoch 6 - iter 84/121 - loss 0.04368248 - time (sec): 5.18 - samples/sec: 3365.14 - lr: 0.000014 - momentum: 0.000000
2023-10-13 10:44:58,714 epoch 6 - iter 96/121 - loss 0.04329465 - time (sec): 5.89 - samples/sec: 3385.53 - lr: 0.000014 - momentum: 0.000000
2023-10-13 10:44:59,407 epoch 6 - iter 108/121 - loss 0.04378744 - time (sec): 6.58 - samples/sec: 3377.78 - lr: 0.000014 - momentum: 0.000000
2023-10-13 10:45:00,098 epoch 6 - iter 120/121 - loss 0.04492980 - time (sec): 7.27 - samples/sec: 3384.12 - lr: 0.000013 - momentum: 0.000000
2023-10-13 10:45:00,147 ----------------------------------------------------------------------------------------------------
2023-10-13 10:45:00,147 EPOCH 6 done: loss 0.0447 - lr: 0.000013
2023-10-13 10:45:01,065 DEV : loss 0.14645931124687195 - f1-score (micro avg)  0.8302
2023-10-13 10:45:01,070 saving best model
2023-10-13 10:45:01,558 ----------------------------------------------------------------------------------------------------
2023-10-13 10:45:02,241 epoch 7 - iter 12/121 - loss 0.04212327 - time (sec): 0.68 - samples/sec: 3463.91 - lr: 0.000013 - momentum: 0.000000
2023-10-13 10:45:02,956 epoch 7 - iter 24/121 - loss 0.03030730 - time (sec): 1.40 - samples/sec: 3467.42 - lr: 0.000013 - momentum: 0.000000
2023-10-13 10:45:03,664 epoch 7 - iter 36/121 - loss 0.02753375 - time (sec): 2.10 - samples/sec: 3384.82 - lr: 0.000012 - momentum: 0.000000
2023-10-13 10:45:04,436 epoch 7 - iter 48/121 - loss 0.02907635 - time (sec): 2.88 - samples/sec: 3381.23 - lr: 0.000012 - momentum: 0.000000
2023-10-13 10:45:05,204 epoch 7 - iter 60/121 - loss 0.02750698 - time (sec): 3.64 - samples/sec: 3429.92 - lr: 0.000012 - momentum: 0.000000
2023-10-13 10:45:05,925 epoch 7 - iter 72/121 - loss 0.02826464 - time (sec): 4.36 - samples/sec: 3420.28 - lr: 0.000011 - momentum: 0.000000
2023-10-13 10:45:06,684 epoch 7 - iter 84/121 - loss 0.03040430 - time (sec): 5.12 - samples/sec: 3383.53 - lr: 0.000011 - momentum: 0.000000
2023-10-13 10:45:07,367 epoch 7 - iter 96/121 - loss 0.03095690 - time (sec): 5.81 - samples/sec: 3389.27 - lr: 0.000011 - momentum: 0.000000
2023-10-13 10:45:08,055 epoch 7 - iter 108/121 - loss 0.03036297 - time (sec): 6.50 - samples/sec: 3422.44 - lr: 0.000010 - momentum: 0.000000
2023-10-13 10:45:08,809 epoch 7 - iter 120/121 - loss 0.03149892 - time (sec): 7.25 - samples/sec: 3390.98 - lr: 0.000010 - momentum: 0.000000
2023-10-13 10:45:08,860 ----------------------------------------------------------------------------------------------------
2023-10-13 10:45:08,860 EPOCH 7 done: loss 0.0317 - lr: 0.000010
2023-10-13 10:45:09,616 DEV : loss 0.1589440405368805 - f1-score (micro avg)  0.8136
2023-10-13 10:45:09,621 ----------------------------------------------------------------------------------------------------
2023-10-13 10:45:10,338 epoch 8 - iter 12/121 - loss 0.03037428 - time (sec): 0.72 - samples/sec: 3324.50 - lr: 0.000010 - momentum: 0.000000
2023-10-13 10:45:11,043 epoch 8 - iter 24/121 - loss 0.03173288 - time (sec): 1.42 - samples/sec: 3154.50 - lr: 0.000009 - momentum: 0.000000
2023-10-13 10:45:11,804 epoch 8 - iter 36/121 - loss 0.02729142 - time (sec): 2.18 - samples/sec: 3174.68 - lr: 0.000009 - momentum: 0.000000
2023-10-13 10:45:12,625 epoch 8 - iter 48/121 - loss 0.02623794 - time (sec): 3.00 - samples/sec: 3228.35 - lr: 0.000009 - momentum: 0.000000
2023-10-13 10:45:13,340 epoch 8 - iter 60/121 - loss 0.02517167 - time (sec): 3.72 - samples/sec: 3266.24 - lr: 0.000008 - momentum: 0.000000
2023-10-13 10:45:14,055 epoch 8 - iter 72/121 - loss 0.02431651 - time (sec): 4.43 - samples/sec: 3309.11 - lr: 0.000008 - momentum: 0.000000
2023-10-13 10:45:14,740 epoch 8 - iter 84/121 - loss 0.02713066 - time (sec): 5.12 - samples/sec: 3316.82 - lr: 0.000008 - momentum: 0.000000
2023-10-13 10:45:15,467 epoch 8 - iter 96/121 - loss 0.02632913 - time (sec): 5.85 - samples/sec: 3286.93 - lr: 0.000008 - momentum: 0.000000
2023-10-13 10:45:16,283 epoch 8 - iter 108/121 - loss 0.02462711 - time (sec): 6.66 - samples/sec: 3302.97 - lr: 0.000007 - momentum: 0.000000
2023-10-13 10:45:17,035 epoch 8 - iter 120/121 - loss 0.02425663 - time (sec): 7.41 - samples/sec: 3319.99 - lr: 0.000007 - momentum: 0.000000
2023-10-13 10:45:17,086 ----------------------------------------------------------------------------------------------------
2023-10-13 10:45:17,086 EPOCH 8 done: loss 0.0241 - lr: 0.000007
2023-10-13 10:45:17,843 DEV : loss 0.16912458837032318 - f1-score (micro avg)  0.8396
2023-10-13 10:45:17,848 saving best model
2023-10-13 10:45:18,331 ----------------------------------------------------------------------------------------------------
2023-10-13 10:45:19,090 epoch 9 - iter 12/121 - loss 0.01197683 - time (sec): 0.75 - samples/sec: 3262.19 - lr: 0.000006 - momentum: 0.000000
2023-10-13 10:45:19,850 epoch 9 - iter 24/121 - loss 0.01853147 - time (sec): 1.51 - samples/sec: 3248.80 - lr: 0.000006 - momentum: 0.000000
2023-10-13 10:45:20,652 epoch 9 - iter 36/121 - loss 0.01813611 - time (sec): 2.31 - samples/sec: 3259.97 - lr: 0.000006 - momentum: 0.000000
2023-10-13 10:45:21,412 epoch 9 - iter 48/121 - loss 0.01789886 - time (sec): 3.07 - samples/sec: 3326.63 - lr: 0.000006 - momentum: 0.000000
2023-10-13 10:45:22,117 epoch 9 - iter 60/121 - loss 0.01892584 - time (sec): 3.77 - samples/sec: 3312.04 - lr: 0.000005 - momentum: 0.000000
2023-10-13 10:45:22,843 epoch 9 - iter 72/121 - loss 0.01887653 - time (sec): 4.50 - samples/sec: 3300.25 - lr: 0.000005 - momentum: 0.000000
2023-10-13 10:45:23,579 epoch 9 - iter 84/121 - loss 0.01813914 - time (sec): 5.24 - samples/sec: 3330.45 - lr: 0.000005 - momentum: 0.000000
2023-10-13 10:45:24,347 epoch 9 - iter 96/121 - loss 0.02178354 - time (sec): 6.00 - samples/sec: 3318.56 - lr: 0.000004 - momentum: 0.000000
2023-10-13 10:45:25,045 epoch 9 - iter 108/121 - loss 0.02094768 - time (sec): 6.70 - samples/sec: 3331.57 - lr: 0.000004 - momentum: 0.000000
2023-10-13 10:45:25,797 epoch 9 - iter 120/121 - loss 0.01977182 - time (sec): 7.45 - samples/sec: 3306.07 - lr: 0.000004 - momentum: 0.000000
2023-10-13 10:45:25,842 ----------------------------------------------------------------------------------------------------
2023-10-13 10:45:25,843 EPOCH 9 done: loss 0.0197 - lr: 0.000004
2023-10-13 10:45:26,597 DEV : loss 0.1732056736946106 - f1-score (micro avg)  0.8407
2023-10-13 10:45:26,603 saving best model
2023-10-13 10:45:27,097 ----------------------------------------------------------------------------------------------------
2023-10-13 10:45:27,828 epoch 10 - iter 12/121 - loss 0.02930384 - time (sec): 0.72 - samples/sec: 3381.93 - lr: 0.000003 - momentum: 0.000000
2023-10-13 10:45:28,620 epoch 10 - iter 24/121 - loss 0.01725666 - time (sec): 1.51 - samples/sec: 3402.96 - lr: 0.000003 - momentum: 0.000000
2023-10-13 10:45:29,395 epoch 10 - iter 36/121 - loss 0.01442001 - time (sec): 2.29 - samples/sec: 3239.62 - lr: 0.000003 - momentum: 0.000000
2023-10-13 10:45:30,116 epoch 10 - iter 48/121 - loss 0.01419669 - time (sec): 3.01 - samples/sec: 3290.84 - lr: 0.000002 - momentum: 0.000000
2023-10-13 10:45:30,918 epoch 10 - iter 60/121 - loss 0.01458065 - time (sec): 3.81 - samples/sec: 3246.16 - lr: 0.000002 - momentum: 0.000000
2023-10-13 10:45:31,687 epoch 10 - iter 72/121 - loss 0.01542659 - time (sec): 4.58 - samples/sec: 3255.17 - lr: 0.000002 - momentum: 0.000000
2023-10-13 10:45:32,415 epoch 10 - iter 84/121 - loss 0.01524820 - time (sec): 5.31 - samples/sec: 3242.25 - lr: 0.000001 - momentum: 0.000000
2023-10-13 10:45:33,159 epoch 10 - iter 96/121 - loss 0.01681722 - time (sec): 6.05 - samples/sec: 3244.42 - lr: 0.000001 - momentum: 0.000000
2023-10-13 10:45:33,897 epoch 10 - iter 108/121 - loss 0.01701795 - time (sec): 6.79 - samples/sec: 3232.95 - lr: 0.000001 - momentum: 0.000000
2023-10-13 10:45:34,692 epoch 10 - iter 120/121 - loss 0.01627520 - time (sec): 7.59 - samples/sec: 3244.26 - lr: 0.000000 - momentum: 0.000000
2023-10-13 10:45:34,742 ----------------------------------------------------------------------------------------------------
2023-10-13 10:45:34,742 EPOCH 10 done: loss 0.0162 - lr: 0.000000
2023-10-13 10:45:35,596 DEV : loss 0.17861199378967285 - f1-score (micro avg)  0.8354
2023-10-13 10:45:35,974 ----------------------------------------------------------------------------------------------------
2023-10-13 10:45:35,976 Loading model from best epoch ...
2023-10-13 10:45:37,432 SequenceTagger predicts: Dictionary with 25 tags: O, S-scope, B-scope, E-scope, I-scope, S-pers, B-pers, E-pers, I-pers, S-work, B-work, E-work, I-work, S-loc, B-loc, E-loc, I-loc, S-object, B-object, E-object, I-object, S-date, B-date, E-date, I-date
2023-10-13 10:45:38,311 
Results:
- F-score (micro) 0.8125
- F-score (macro) 0.5575
- Accuracy 0.7035

By class:
              precision    recall  f1-score   support

        pers     0.8243    0.8777    0.8502       139
       scope     0.8456    0.8915    0.8679       129
        work     0.6667    0.7500    0.7059        80
         loc     1.0000    0.2222    0.3636         9
        date     0.0000    0.0000    0.0000         3

   micro avg     0.7952    0.8306    0.8125       360
   macro avg     0.6673    0.5483    0.5575       360
weighted avg     0.7944    0.8306    0.8052       360

2023-10-13 10:45:38,312 ----------------------------------------------------------------------------------------------------