File size: 24,185 Bytes
f3c5a65
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
246
247
248
249
2023-10-23 15:47:11,119 ----------------------------------------------------------------------------------------------------
2023-10-23 15:47:11,120 Model: "SequenceTagger(
  (embeddings): TransformerWordEmbeddings(
    (model): BertModel(
      (embeddings): BertEmbeddings(
        (word_embeddings): Embedding(64001, 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()
)"
2023-10-23 15:47:11,120 ----------------------------------------------------------------------------------------------------
2023-10-23 15:47:11,120 MultiCorpus: 1100 train + 206 dev + 240 test sentences
 - NER_HIPE_2022 Corpus: 1100 train + 206 dev + 240 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/de/with_doc_seperator
2023-10-23 15:47:11,120 ----------------------------------------------------------------------------------------------------
2023-10-23 15:47:11,120 Train:  1100 sentences
2023-10-23 15:47:11,120         (train_with_dev=False, train_with_test=False)
2023-10-23 15:47:11,121 ----------------------------------------------------------------------------------------------------
2023-10-23 15:47:11,121 Training Params:
2023-10-23 15:47:11,121  - learning_rate: "5e-05" 
2023-10-23 15:47:11,121  - mini_batch_size: "8"
2023-10-23 15:47:11,121  - max_epochs: "10"
2023-10-23 15:47:11,121  - shuffle: "True"
2023-10-23 15:47:11,121 ----------------------------------------------------------------------------------------------------
2023-10-23 15:47:11,121 Plugins:
2023-10-23 15:47:11,121  - TensorboardLogger
2023-10-23 15:47:11,121  - LinearScheduler | warmup_fraction: '0.1'
2023-10-23 15:47:11,121 ----------------------------------------------------------------------------------------------------
2023-10-23 15:47:11,121 Final evaluation on model from best epoch (best-model.pt)
2023-10-23 15:47:11,121  - metric: "('micro avg', 'f1-score')"
2023-10-23 15:47:11,121 ----------------------------------------------------------------------------------------------------
2023-10-23 15:47:11,121 Computation:
2023-10-23 15:47:11,121  - compute on device: cuda:0
2023-10-23 15:47:11,121  - embedding storage: none
2023-10-23 15:47:11,121 ----------------------------------------------------------------------------------------------------
2023-10-23 15:47:11,121 Model training base path: "hmbench-ajmc/de-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4"
2023-10-23 15:47:11,121 ----------------------------------------------------------------------------------------------------
2023-10-23 15:47:11,121 ----------------------------------------------------------------------------------------------------
2023-10-23 15:47:11,121 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-23 15:47:11,830 epoch 1 - iter 13/138 - loss 3.11505708 - time (sec): 0.71 - samples/sec: 2973.30 - lr: 0.000004 - momentum: 0.000000
2023-10-23 15:47:12,567 epoch 1 - iter 26/138 - loss 2.46744691 - time (sec): 1.44 - samples/sec: 2948.36 - lr: 0.000009 - momentum: 0.000000
2023-10-23 15:47:13,294 epoch 1 - iter 39/138 - loss 2.00064272 - time (sec): 2.17 - samples/sec: 2848.83 - lr: 0.000014 - momentum: 0.000000
2023-10-23 15:47:14,059 epoch 1 - iter 52/138 - loss 1.65199002 - time (sec): 2.94 - samples/sec: 2947.01 - lr: 0.000018 - momentum: 0.000000
2023-10-23 15:47:14,778 epoch 1 - iter 65/138 - loss 1.48169540 - time (sec): 3.66 - samples/sec: 2901.54 - lr: 0.000023 - momentum: 0.000000
2023-10-23 15:47:15,519 epoch 1 - iter 78/138 - loss 1.29756776 - time (sec): 4.40 - samples/sec: 2920.72 - lr: 0.000028 - momentum: 0.000000
2023-10-23 15:47:16,270 epoch 1 - iter 91/138 - loss 1.16855120 - time (sec): 5.15 - samples/sec: 2865.24 - lr: 0.000033 - momentum: 0.000000
2023-10-23 15:47:17,053 epoch 1 - iter 104/138 - loss 1.03837897 - time (sec): 5.93 - samples/sec: 2921.34 - lr: 0.000037 - momentum: 0.000000
2023-10-23 15:47:17,867 epoch 1 - iter 117/138 - loss 0.94938763 - time (sec): 6.75 - samples/sec: 2886.32 - lr: 0.000042 - momentum: 0.000000
2023-10-23 15:47:18,654 epoch 1 - iter 130/138 - loss 0.88540734 - time (sec): 7.53 - samples/sec: 2878.25 - lr: 0.000047 - momentum: 0.000000
2023-10-23 15:47:19,132 ----------------------------------------------------------------------------------------------------
2023-10-23 15:47:19,132 EPOCH 1 done: loss 0.8523 - lr: 0.000047
2023-10-23 15:47:19,554 DEV : loss 0.2265705019235611 - f1-score (micro avg)  0.6643
2023-10-23 15:47:19,560 saving best model
2023-10-23 15:47:19,955 ----------------------------------------------------------------------------------------------------
2023-10-23 15:47:20,667 epoch 2 - iter 13/138 - loss 0.26476026 - time (sec): 0.71 - samples/sec: 2982.33 - lr: 0.000050 - momentum: 0.000000
2023-10-23 15:47:21,385 epoch 2 - iter 26/138 - loss 0.21464055 - time (sec): 1.43 - samples/sec: 3104.30 - lr: 0.000049 - momentum: 0.000000
2023-10-23 15:47:22,113 epoch 2 - iter 39/138 - loss 0.20326127 - time (sec): 2.16 - samples/sec: 3134.81 - lr: 0.000048 - momentum: 0.000000
2023-10-23 15:47:22,857 epoch 2 - iter 52/138 - loss 0.18189690 - time (sec): 2.90 - samples/sec: 3124.57 - lr: 0.000048 - momentum: 0.000000
2023-10-23 15:47:23,666 epoch 2 - iter 65/138 - loss 0.18181936 - time (sec): 3.71 - samples/sec: 3005.53 - lr: 0.000047 - momentum: 0.000000
2023-10-23 15:47:24,458 epoch 2 - iter 78/138 - loss 0.17939422 - time (sec): 4.50 - samples/sec: 2972.13 - lr: 0.000047 - momentum: 0.000000
2023-10-23 15:47:25,250 epoch 2 - iter 91/138 - loss 0.17730142 - time (sec): 5.29 - samples/sec: 2935.32 - lr: 0.000046 - momentum: 0.000000
2023-10-23 15:47:26,046 epoch 2 - iter 104/138 - loss 0.17262996 - time (sec): 6.09 - samples/sec: 2894.92 - lr: 0.000046 - momentum: 0.000000
2023-10-23 15:47:26,839 epoch 2 - iter 117/138 - loss 0.16894373 - time (sec): 6.88 - samples/sec: 2840.58 - lr: 0.000045 - momentum: 0.000000
2023-10-23 15:47:27,618 epoch 2 - iter 130/138 - loss 0.16407567 - time (sec): 7.66 - samples/sec: 2819.51 - lr: 0.000045 - momentum: 0.000000
2023-10-23 15:47:28,102 ----------------------------------------------------------------------------------------------------
2023-10-23 15:47:28,102 EPOCH 2 done: loss 0.1596 - lr: 0.000045
2023-10-23 15:47:28,638 DEV : loss 0.11940671503543854 - f1-score (micro avg)  0.8465
2023-10-23 15:47:28,644 saving best model
2023-10-23 15:47:29,202 ----------------------------------------------------------------------------------------------------
2023-10-23 15:47:30,003 epoch 3 - iter 13/138 - loss 0.07988993 - time (sec): 0.80 - samples/sec: 2593.24 - lr: 0.000044 - momentum: 0.000000
2023-10-23 15:47:30,798 epoch 3 - iter 26/138 - loss 0.10888248 - time (sec): 1.59 - samples/sec: 2556.50 - lr: 0.000043 - momentum: 0.000000
2023-10-23 15:47:31,608 epoch 3 - iter 39/138 - loss 0.09016497 - time (sec): 2.40 - samples/sec: 2621.42 - lr: 0.000043 - momentum: 0.000000
2023-10-23 15:47:32,396 epoch 3 - iter 52/138 - loss 0.08999953 - time (sec): 3.19 - samples/sec: 2617.65 - lr: 0.000042 - momentum: 0.000000
2023-10-23 15:47:33,173 epoch 3 - iter 65/138 - loss 0.09109642 - time (sec): 3.96 - samples/sec: 2592.58 - lr: 0.000042 - momentum: 0.000000
2023-10-23 15:47:33,952 epoch 3 - iter 78/138 - loss 0.08731252 - time (sec): 4.74 - samples/sec: 2642.90 - lr: 0.000041 - momentum: 0.000000
2023-10-23 15:47:34,766 epoch 3 - iter 91/138 - loss 0.09236665 - time (sec): 5.56 - samples/sec: 2646.20 - lr: 0.000041 - momentum: 0.000000
2023-10-23 15:47:35,556 epoch 3 - iter 104/138 - loss 0.08968405 - time (sec): 6.35 - samples/sec: 2661.88 - lr: 0.000040 - momentum: 0.000000
2023-10-23 15:47:36,367 epoch 3 - iter 117/138 - loss 0.09102394 - time (sec): 7.16 - samples/sec: 2676.05 - lr: 0.000040 - momentum: 0.000000
2023-10-23 15:47:37,174 epoch 3 - iter 130/138 - loss 0.09080745 - time (sec): 7.97 - samples/sec: 2709.68 - lr: 0.000039 - momentum: 0.000000
2023-10-23 15:47:37,664 ----------------------------------------------------------------------------------------------------
2023-10-23 15:47:37,664 EPOCH 3 done: loss 0.0903 - lr: 0.000039
2023-10-23 15:47:38,358 DEV : loss 0.1325010061264038 - f1-score (micro avg)  0.839
2023-10-23 15:47:38,364 ----------------------------------------------------------------------------------------------------
2023-10-23 15:47:39,107 epoch 4 - iter 13/138 - loss 0.04646751 - time (sec): 0.74 - samples/sec: 2994.83 - lr: 0.000038 - momentum: 0.000000
2023-10-23 15:47:39,891 epoch 4 - iter 26/138 - loss 0.05610819 - time (sec): 1.53 - samples/sec: 2739.53 - lr: 0.000038 - momentum: 0.000000
2023-10-23 15:47:40,684 epoch 4 - iter 39/138 - loss 0.05405158 - time (sec): 2.32 - samples/sec: 2691.38 - lr: 0.000037 - momentum: 0.000000
2023-10-23 15:47:41,500 epoch 4 - iter 52/138 - loss 0.05580227 - time (sec): 3.14 - samples/sec: 2655.31 - lr: 0.000037 - momentum: 0.000000
2023-10-23 15:47:42,296 epoch 4 - iter 65/138 - loss 0.04970762 - time (sec): 3.93 - samples/sec: 2609.10 - lr: 0.000036 - momentum: 0.000000
2023-10-23 15:47:43,110 epoch 4 - iter 78/138 - loss 0.05294991 - time (sec): 4.75 - samples/sec: 2629.20 - lr: 0.000036 - momentum: 0.000000
2023-10-23 15:47:43,916 epoch 4 - iter 91/138 - loss 0.05649414 - time (sec): 5.55 - samples/sec: 2658.90 - lr: 0.000035 - momentum: 0.000000
2023-10-23 15:47:44,734 epoch 4 - iter 104/138 - loss 0.05731326 - time (sec): 6.37 - samples/sec: 2672.85 - lr: 0.000035 - momentum: 0.000000
2023-10-23 15:47:45,535 epoch 4 - iter 117/138 - loss 0.05895584 - time (sec): 7.17 - samples/sec: 2691.66 - lr: 0.000034 - momentum: 0.000000
2023-10-23 15:47:46,328 epoch 4 - iter 130/138 - loss 0.06194870 - time (sec): 7.96 - samples/sec: 2686.50 - lr: 0.000034 - momentum: 0.000000
2023-10-23 15:47:46,822 ----------------------------------------------------------------------------------------------------
2023-10-23 15:47:46,822 EPOCH 4 done: loss 0.0612 - lr: 0.000034
2023-10-23 15:47:47,354 DEV : loss 0.12988846004009247 - f1-score (micro avg)  0.8554
2023-10-23 15:47:47,360 saving best model
2023-10-23 15:47:47,915 ----------------------------------------------------------------------------------------------------
2023-10-23 15:47:48,628 epoch 5 - iter 13/138 - loss 0.04927867 - time (sec): 0.71 - samples/sec: 2938.01 - lr: 0.000033 - momentum: 0.000000
2023-10-23 15:47:49,431 epoch 5 - iter 26/138 - loss 0.03675946 - time (sec): 1.51 - samples/sec: 2794.05 - lr: 0.000032 - momentum: 0.000000
2023-10-23 15:47:50,208 epoch 5 - iter 39/138 - loss 0.04209356 - time (sec): 2.29 - samples/sec: 2798.76 - lr: 0.000032 - momentum: 0.000000
2023-10-23 15:47:50,987 epoch 5 - iter 52/138 - loss 0.04151665 - time (sec): 3.07 - samples/sec: 2811.44 - lr: 0.000031 - momentum: 0.000000
2023-10-23 15:47:51,704 epoch 5 - iter 65/138 - loss 0.03815920 - time (sec): 3.78 - samples/sec: 2832.74 - lr: 0.000031 - momentum: 0.000000
2023-10-23 15:47:52,432 epoch 5 - iter 78/138 - loss 0.04408756 - time (sec): 4.51 - samples/sec: 2896.67 - lr: 0.000030 - momentum: 0.000000
2023-10-23 15:47:53,147 epoch 5 - iter 91/138 - loss 0.04668946 - time (sec): 5.23 - samples/sec: 2880.16 - lr: 0.000030 - momentum: 0.000000
2023-10-23 15:47:53,836 epoch 5 - iter 104/138 - loss 0.04823073 - time (sec): 5.92 - samples/sec: 2872.66 - lr: 0.000029 - momentum: 0.000000
2023-10-23 15:47:54,576 epoch 5 - iter 117/138 - loss 0.04762912 - time (sec): 6.66 - samples/sec: 2917.07 - lr: 0.000029 - momentum: 0.000000
2023-10-23 15:47:55,286 epoch 5 - iter 130/138 - loss 0.04742132 - time (sec): 7.37 - samples/sec: 2933.92 - lr: 0.000028 - momentum: 0.000000
2023-10-23 15:47:55,736 ----------------------------------------------------------------------------------------------------
2023-10-23 15:47:55,736 EPOCH 5 done: loss 0.0454 - lr: 0.000028
2023-10-23 15:47:56,268 DEV : loss 0.15986910462379456 - f1-score (micro avg)  0.8606
2023-10-23 15:47:56,274 saving best model
2023-10-23 15:47:56,818 ----------------------------------------------------------------------------------------------------
2023-10-23 15:47:57,521 epoch 6 - iter 13/138 - loss 0.04161822 - time (sec): 0.70 - samples/sec: 3012.41 - lr: 0.000027 - momentum: 0.000000
2023-10-23 15:47:58,239 epoch 6 - iter 26/138 - loss 0.05124527 - time (sec): 1.42 - samples/sec: 3101.86 - lr: 0.000027 - momentum: 0.000000
2023-10-23 15:47:58,975 epoch 6 - iter 39/138 - loss 0.04212149 - time (sec): 2.15 - samples/sec: 3019.44 - lr: 0.000026 - momentum: 0.000000
2023-10-23 15:47:59,690 epoch 6 - iter 52/138 - loss 0.03547030 - time (sec): 2.87 - samples/sec: 2992.94 - lr: 0.000026 - momentum: 0.000000
2023-10-23 15:48:00,386 epoch 6 - iter 65/138 - loss 0.03220072 - time (sec): 3.56 - samples/sec: 3004.40 - lr: 0.000025 - momentum: 0.000000
2023-10-23 15:48:01,097 epoch 6 - iter 78/138 - loss 0.02840167 - time (sec): 4.27 - samples/sec: 3103.60 - lr: 0.000025 - momentum: 0.000000
2023-10-23 15:48:01,795 epoch 6 - iter 91/138 - loss 0.02684088 - time (sec): 4.97 - samples/sec: 3106.79 - lr: 0.000024 - momentum: 0.000000
2023-10-23 15:48:02,516 epoch 6 - iter 104/138 - loss 0.03197223 - time (sec): 5.69 - samples/sec: 3058.56 - lr: 0.000024 - momentum: 0.000000
2023-10-23 15:48:03,217 epoch 6 - iter 117/138 - loss 0.03171220 - time (sec): 6.39 - samples/sec: 3068.40 - lr: 0.000023 - momentum: 0.000000
2023-10-23 15:48:03,901 epoch 6 - iter 130/138 - loss 0.03129985 - time (sec): 7.08 - samples/sec: 3033.14 - lr: 0.000023 - momentum: 0.000000
2023-10-23 15:48:04,322 ----------------------------------------------------------------------------------------------------
2023-10-23 15:48:04,323 EPOCH 6 done: loss 0.0310 - lr: 0.000023
2023-10-23 15:48:04,853 DEV : loss 0.1651483029127121 - f1-score (micro avg)  0.8617
2023-10-23 15:48:04,859 saving best model
2023-10-23 15:48:05,386 ----------------------------------------------------------------------------------------------------
2023-10-23 15:48:06,114 epoch 7 - iter 13/138 - loss 0.00533927 - time (sec): 0.72 - samples/sec: 2942.35 - lr: 0.000022 - momentum: 0.000000
2023-10-23 15:48:06,817 epoch 7 - iter 26/138 - loss 0.01162164 - time (sec): 1.43 - samples/sec: 3030.67 - lr: 0.000021 - momentum: 0.000000
2023-10-23 15:48:07,522 epoch 7 - iter 39/138 - loss 0.02842342 - time (sec): 2.13 - samples/sec: 3046.05 - lr: 0.000021 - momentum: 0.000000
2023-10-23 15:48:08,284 epoch 7 - iter 52/138 - loss 0.02496210 - time (sec): 2.89 - samples/sec: 3099.66 - lr: 0.000020 - momentum: 0.000000
2023-10-23 15:48:09,011 epoch 7 - iter 65/138 - loss 0.02352598 - time (sec): 3.62 - samples/sec: 3028.85 - lr: 0.000020 - momentum: 0.000000
2023-10-23 15:48:09,752 epoch 7 - iter 78/138 - loss 0.02146960 - time (sec): 4.36 - samples/sec: 3021.91 - lr: 0.000019 - momentum: 0.000000
2023-10-23 15:48:10,477 epoch 7 - iter 91/138 - loss 0.02311809 - time (sec): 5.09 - samples/sec: 2983.84 - lr: 0.000019 - momentum: 0.000000
2023-10-23 15:48:11,224 epoch 7 - iter 104/138 - loss 0.02189737 - time (sec): 5.83 - samples/sec: 2999.38 - lr: 0.000018 - momentum: 0.000000
2023-10-23 15:48:11,967 epoch 7 - iter 117/138 - loss 0.02004301 - time (sec): 6.58 - samples/sec: 2995.24 - lr: 0.000018 - momentum: 0.000000
2023-10-23 15:48:12,695 epoch 7 - iter 130/138 - loss 0.01962060 - time (sec): 7.30 - samples/sec: 2966.80 - lr: 0.000017 - momentum: 0.000000
2023-10-23 15:48:13,127 ----------------------------------------------------------------------------------------------------
2023-10-23 15:48:13,127 EPOCH 7 done: loss 0.0200 - lr: 0.000017
2023-10-23 15:48:13,671 DEV : loss 0.1657860279083252 - f1-score (micro avg)  0.8644
2023-10-23 15:48:13,678 saving best model
2023-10-23 15:48:14,230 ----------------------------------------------------------------------------------------------------
2023-10-23 15:48:14,928 epoch 8 - iter 13/138 - loss 0.00589471 - time (sec): 0.69 - samples/sec: 2934.05 - lr: 0.000016 - momentum: 0.000000
2023-10-23 15:48:15,622 epoch 8 - iter 26/138 - loss 0.00765983 - time (sec): 1.39 - samples/sec: 3195.34 - lr: 0.000016 - momentum: 0.000000
2023-10-23 15:48:16,335 epoch 8 - iter 39/138 - loss 0.00720803 - time (sec): 2.10 - samples/sec: 3132.26 - lr: 0.000015 - momentum: 0.000000
2023-10-23 15:48:17,042 epoch 8 - iter 52/138 - loss 0.01376241 - time (sec): 2.81 - samples/sec: 3114.13 - lr: 0.000015 - momentum: 0.000000
2023-10-23 15:48:17,757 epoch 8 - iter 65/138 - loss 0.01729594 - time (sec): 3.52 - samples/sec: 3040.45 - lr: 0.000014 - momentum: 0.000000
2023-10-23 15:48:18,467 epoch 8 - iter 78/138 - loss 0.01653361 - time (sec): 4.23 - samples/sec: 3061.06 - lr: 0.000014 - momentum: 0.000000
2023-10-23 15:48:19,184 epoch 8 - iter 91/138 - loss 0.01734789 - time (sec): 4.95 - samples/sec: 3064.51 - lr: 0.000013 - momentum: 0.000000
2023-10-23 15:48:19,884 epoch 8 - iter 104/138 - loss 0.01590998 - time (sec): 5.65 - samples/sec: 3029.29 - lr: 0.000013 - momentum: 0.000000
2023-10-23 15:48:20,592 epoch 8 - iter 117/138 - loss 0.01576961 - time (sec): 6.36 - samples/sec: 3025.51 - lr: 0.000012 - momentum: 0.000000
2023-10-23 15:48:21,291 epoch 8 - iter 130/138 - loss 0.01454689 - time (sec): 7.06 - samples/sec: 3046.09 - lr: 0.000012 - momentum: 0.000000
2023-10-23 15:48:21,722 ----------------------------------------------------------------------------------------------------
2023-10-23 15:48:21,722 EPOCH 8 done: loss 0.0141 - lr: 0.000012
2023-10-23 15:48:22,253 DEV : loss 0.1713750660419464 - f1-score (micro avg)  0.8809
2023-10-23 15:48:22,259 saving best model
2023-10-23 15:48:22,785 ----------------------------------------------------------------------------------------------------
2023-10-23 15:48:23,573 epoch 9 - iter 13/138 - loss 0.00064167 - time (sec): 0.78 - samples/sec: 2662.24 - lr: 0.000011 - momentum: 0.000000
2023-10-23 15:48:24,379 epoch 9 - iter 26/138 - loss 0.00090612 - time (sec): 1.59 - samples/sec: 2591.74 - lr: 0.000010 - momentum: 0.000000
2023-10-23 15:48:25,156 epoch 9 - iter 39/138 - loss 0.00356920 - time (sec): 2.37 - samples/sec: 2661.44 - lr: 0.000010 - momentum: 0.000000
2023-10-23 15:48:25,905 epoch 9 - iter 52/138 - loss 0.00694166 - time (sec): 3.12 - samples/sec: 2703.28 - lr: 0.000009 - momentum: 0.000000
2023-10-23 15:48:26,662 epoch 9 - iter 65/138 - loss 0.00634685 - time (sec): 3.87 - samples/sec: 2750.75 - lr: 0.000009 - momentum: 0.000000
2023-10-23 15:48:27,371 epoch 9 - iter 78/138 - loss 0.00630311 - time (sec): 4.58 - samples/sec: 2811.30 - lr: 0.000008 - momentum: 0.000000
2023-10-23 15:48:28,085 epoch 9 - iter 91/138 - loss 0.00702119 - time (sec): 5.29 - samples/sec: 2857.61 - lr: 0.000008 - momentum: 0.000000
2023-10-23 15:48:28,793 epoch 9 - iter 104/138 - loss 0.00633296 - time (sec): 6.00 - samples/sec: 2881.34 - lr: 0.000007 - momentum: 0.000000
2023-10-23 15:48:29,494 epoch 9 - iter 117/138 - loss 0.00628241 - time (sec): 6.70 - samples/sec: 2893.94 - lr: 0.000007 - momentum: 0.000000
2023-10-23 15:48:30,188 epoch 9 - iter 130/138 - loss 0.00742424 - time (sec): 7.40 - samples/sec: 2912.60 - lr: 0.000006 - momentum: 0.000000
2023-10-23 15:48:30,618 ----------------------------------------------------------------------------------------------------
2023-10-23 15:48:30,619 EPOCH 9 done: loss 0.0081 - lr: 0.000006
2023-10-23 15:48:31,149 DEV : loss 0.168824702501297 - f1-score (micro avg)  0.8878
2023-10-23 15:48:31,155 saving best model
2023-10-23 15:48:31,685 ----------------------------------------------------------------------------------------------------
2023-10-23 15:48:32,447 epoch 10 - iter 13/138 - loss 0.00118623 - time (sec): 0.76 - samples/sec: 2828.23 - lr: 0.000005 - momentum: 0.000000
2023-10-23 15:48:33,213 epoch 10 - iter 26/138 - loss 0.00065035 - time (sec): 1.53 - samples/sec: 2780.14 - lr: 0.000005 - momentum: 0.000000
2023-10-23 15:48:33,980 epoch 10 - iter 39/138 - loss 0.00146356 - time (sec): 2.29 - samples/sec: 2887.96 - lr: 0.000004 - momentum: 0.000000
2023-10-23 15:48:34,716 epoch 10 - iter 52/138 - loss 0.00148726 - time (sec): 3.03 - samples/sec: 2858.71 - lr: 0.000004 - momentum: 0.000000
2023-10-23 15:48:35,468 epoch 10 - iter 65/138 - loss 0.00601163 - time (sec): 3.78 - samples/sec: 2811.73 - lr: 0.000003 - momentum: 0.000000
2023-10-23 15:48:36,217 epoch 10 - iter 78/138 - loss 0.00550576 - time (sec): 4.53 - samples/sec: 2841.53 - lr: 0.000003 - momentum: 0.000000
2023-10-23 15:48:36,937 epoch 10 - iter 91/138 - loss 0.00571920 - time (sec): 5.25 - samples/sec: 2850.58 - lr: 0.000002 - momentum: 0.000000
2023-10-23 15:48:37,659 epoch 10 - iter 104/138 - loss 0.00554833 - time (sec): 5.97 - samples/sec: 2870.23 - lr: 0.000002 - momentum: 0.000000
2023-10-23 15:48:38,411 epoch 10 - iter 117/138 - loss 0.00516556 - time (sec): 6.72 - samples/sec: 2864.43 - lr: 0.000001 - momentum: 0.000000
2023-10-23 15:48:39,166 epoch 10 - iter 130/138 - loss 0.00497822 - time (sec): 7.48 - samples/sec: 2875.52 - lr: 0.000000 - momentum: 0.000000
2023-10-23 15:48:39,615 ----------------------------------------------------------------------------------------------------
2023-10-23 15:48:39,615 EPOCH 10 done: loss 0.0052 - lr: 0.000000
2023-10-23 15:48:40,154 DEV : loss 0.16836421191692352 - f1-score (micro avg)  0.8892
2023-10-23 15:48:40,160 saving best model
2023-10-23 15:48:41,107 ----------------------------------------------------------------------------------------------------
2023-10-23 15:48:41,109 Loading model from best epoch ...
2023-10-23 15:48:42,759 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
2023-10-23 15:48:43,434 
Results:
- F-score (micro) 0.9098
- F-score (macro) 0.8239
- Accuracy 0.8426

By class:
              precision    recall  f1-score   support

       scope     0.8883    0.9034    0.8958       176
        pers     0.9840    0.9609    0.9723       128
        work     0.8514    0.8514    0.8514        74
         loc     0.3333    0.5000    0.4000         2
      object     1.0000    1.0000    1.0000         2

   micro avg     0.9086    0.9110    0.9098       382
   macro avg     0.8114    0.8431    0.8239       382
weighted avg     0.9109    0.9110    0.9108       382

2023-10-23 15:48:43,435 ----------------------------------------------------------------------------------------------------