Hugging Face's logo

stefan-it
/
flair-co-funer-german_bert_base-bs8-e10-lr3e-05-2

Token Classification
Flair
PyTorch
TensorBoard
German
sequence-tagger-model
hetzner
hetzner-gex44
hetzner-gpu
Model card Files Metrics Community
2
flair-co-funer-german_bert_base-bs8-e10-lr3e-05-2
File size: 23,821 Bytes 
18223ce
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
 
2024-03-26 11:21:23,580 ----------------------------------------------------------------------------------------------------
2024-03-26 11:21:23,580 Model: "SequenceTagger(
  (embeddings): TransformerWordEmbeddings(
    (model): BertModel(
      (embeddings): BertEmbeddings(
        (word_embeddings): Embedding(30001, 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()
)"
2024-03-26 11:21:23,580 ----------------------------------------------------------------------------------------------------
2024-03-26 11:21:23,580 Corpus: 758 train + 94 dev + 96 test sentences
2024-03-26 11:21:23,580 ----------------------------------------------------------------------------------------------------
2024-03-26 11:21:23,580 Train:  758 sentences
2024-03-26 11:21:23,581         (train_with_dev=False, train_with_test=False)
2024-03-26 11:21:23,581 ----------------------------------------------------------------------------------------------------
2024-03-26 11:21:23,581 Training Params:
2024-03-26 11:21:23,581  - learning_rate: "3e-05" 
2024-03-26 11:21:23,581  - mini_batch_size: "8"
2024-03-26 11:21:23,581  - max_epochs: "10"
2024-03-26 11:21:23,581  - shuffle: "True"
2024-03-26 11:21:23,581 ----------------------------------------------------------------------------------------------------
2024-03-26 11:21:23,581 Plugins:
2024-03-26 11:21:23,581  - TensorboardLogger
2024-03-26 11:21:23,581  - LinearScheduler | warmup_fraction: '0.1'
2024-03-26 11:21:23,581 ----------------------------------------------------------------------------------------------------
2024-03-26 11:21:23,581 Final evaluation on model from best epoch (best-model.pt)
2024-03-26 11:21:23,581  - metric: "('micro avg', 'f1-score')"
2024-03-26 11:21:23,581 ----------------------------------------------------------------------------------------------------
2024-03-26 11:21:23,581 Computation:
2024-03-26 11:21:23,581  - compute on device: cuda:0
2024-03-26 11:21:23,581  - embedding storage: none
2024-03-26 11:21:23,581 ----------------------------------------------------------------------------------------------------
2024-03-26 11:21:23,581 Model training base path: "flair-co-funer-german_bert_base-bs8-e10-lr3e-05-2"
2024-03-26 11:21:23,581 ----------------------------------------------------------------------------------------------------
2024-03-26 11:21:23,581 ----------------------------------------------------------------------------------------------------
2024-03-26 11:21:23,581 Logging anything other than scalars to TensorBoard is currently not supported.
2024-03-26 11:21:25,474 epoch 1 - iter 9/95 - loss 3.11557072 - time (sec): 1.89 - samples/sec: 1861.65 - lr: 0.000003 - momentum: 0.000000
2024-03-26 11:21:27,689 epoch 1 - iter 18/95 - loss 3.07674999 - time (sec): 4.11 - samples/sec: 1754.51 - lr: 0.000005 - momentum: 0.000000
2024-03-26 11:21:29,285 epoch 1 - iter 27/95 - loss 2.92205455 - time (sec): 5.70 - samples/sec: 1767.69 - lr: 0.000008 - momentum: 0.000000
2024-03-26 11:21:31,266 epoch 1 - iter 36/95 - loss 2.71661912 - time (sec): 7.69 - samples/sec: 1795.94 - lr: 0.000011 - momentum: 0.000000
2024-03-26 11:21:33,385 epoch 1 - iter 45/95 - loss 2.53581484 - time (sec): 9.80 - samples/sec: 1739.09 - lr: 0.000014 - momentum: 0.000000
2024-03-26 11:21:35,431 epoch 1 - iter 54/95 - loss 2.36834659 - time (sec): 11.85 - samples/sec: 1713.63 - lr: 0.000017 - momentum: 0.000000
2024-03-26 11:21:37,009 epoch 1 - iter 63/95 - loss 2.23517454 - time (sec): 13.43 - samples/sec: 1722.79 - lr: 0.000020 - momentum: 0.000000
2024-03-26 11:21:38,320 epoch 1 - iter 72/95 - loss 2.10452943 - time (sec): 14.74 - samples/sec: 1774.24 - lr: 0.000022 - momentum: 0.000000
2024-03-26 11:21:39,914 epoch 1 - iter 81/95 - loss 1.97687171 - time (sec): 16.33 - samples/sec: 1801.00 - lr: 0.000025 - momentum: 0.000000
2024-03-26 11:21:41,943 epoch 1 - iter 90/95 - loss 1.85329090 - time (sec): 18.36 - samples/sec: 1775.88 - lr: 0.000028 - momentum: 0.000000
2024-03-26 11:21:43,059 ----------------------------------------------------------------------------------------------------
2024-03-26 11:21:43,059 EPOCH 1 done: loss 1.7886 - lr: 0.000028
2024-03-26 11:21:43,993 DEV : loss 0.5206313133239746 - f1-score (micro avg)  0.6482
2024-03-26 11:21:43,994 saving best model
2024-03-26 11:21:44,253 ----------------------------------------------------------------------------------------------------
2024-03-26 11:21:45,596 epoch 2 - iter 9/95 - loss 0.67823346 - time (sec): 1.34 - samples/sec: 2417.26 - lr: 0.000030 - momentum: 0.000000
2024-03-26 11:21:47,479 epoch 2 - iter 18/95 - loss 0.54760123 - time (sec): 3.22 - samples/sec: 2130.38 - lr: 0.000029 - momentum: 0.000000
2024-03-26 11:21:50,351 epoch 2 - iter 27/95 - loss 0.46607815 - time (sec): 6.10 - samples/sec: 1895.87 - lr: 0.000029 - momentum: 0.000000
2024-03-26 11:21:52,507 epoch 2 - iter 36/95 - loss 0.44210057 - time (sec): 8.25 - samples/sec: 1801.44 - lr: 0.000029 - momentum: 0.000000
2024-03-26 11:21:54,309 epoch 2 - iter 45/95 - loss 0.41849252 - time (sec): 10.06 - samples/sec: 1787.61 - lr: 0.000028 - momentum: 0.000000
2024-03-26 11:21:56,470 epoch 2 - iter 54/95 - loss 0.40354492 - time (sec): 12.22 - samples/sec: 1741.72 - lr: 0.000028 - momentum: 0.000000
2024-03-26 11:21:58,102 epoch 2 - iter 63/95 - loss 0.40666529 - time (sec): 13.85 - samples/sec: 1757.53 - lr: 0.000028 - momentum: 0.000000
2024-03-26 11:21:59,637 epoch 2 - iter 72/95 - loss 0.40026895 - time (sec): 15.38 - samples/sec: 1783.27 - lr: 0.000028 - momentum: 0.000000
2024-03-26 11:22:00,816 epoch 2 - iter 81/95 - loss 0.39659236 - time (sec): 16.56 - samples/sec: 1819.08 - lr: 0.000027 - momentum: 0.000000
2024-03-26 11:22:02,117 epoch 2 - iter 90/95 - loss 0.38998382 - time (sec): 17.86 - samples/sec: 1841.73 - lr: 0.000027 - momentum: 0.000000
2024-03-26 11:22:03,109 ----------------------------------------------------------------------------------------------------
2024-03-26 11:22:03,109 EPOCH 2 done: loss 0.3799 - lr: 0.000027
2024-03-26 11:22:04,036 DEV : loss 0.2723105847835541 - f1-score (micro avg)  0.8308
2024-03-26 11:22:04,037 saving best model
2024-03-26 11:22:04,454 ----------------------------------------------------------------------------------------------------
2024-03-26 11:22:06,471 epoch 3 - iter 9/95 - loss 0.21258499 - time (sec): 2.01 - samples/sec: 1652.16 - lr: 0.000026 - momentum: 0.000000
2024-03-26 11:22:08,586 epoch 3 - iter 18/95 - loss 0.22203275 - time (sec): 4.13 - samples/sec: 1758.63 - lr: 0.000026 - momentum: 0.000000
2024-03-26 11:22:09,564 epoch 3 - iter 27/95 - loss 0.22177471 - time (sec): 5.11 - samples/sec: 1883.80 - lr: 0.000026 - momentum: 0.000000
2024-03-26 11:22:11,324 epoch 3 - iter 36/95 - loss 0.21771092 - time (sec): 6.87 - samples/sec: 1847.68 - lr: 0.000025 - momentum: 0.000000
2024-03-26 11:22:12,606 epoch 3 - iter 45/95 - loss 0.23043711 - time (sec): 8.15 - samples/sec: 1889.83 - lr: 0.000025 - momentum: 0.000000
2024-03-26 11:22:14,664 epoch 3 - iter 54/95 - loss 0.22935549 - time (sec): 10.21 - samples/sec: 1831.51 - lr: 0.000025 - momentum: 0.000000
2024-03-26 11:22:16,315 epoch 3 - iter 63/95 - loss 0.22730811 - time (sec): 11.86 - samples/sec: 1840.55 - lr: 0.000025 - momentum: 0.000000
2024-03-26 11:22:17,888 epoch 3 - iter 72/95 - loss 0.22301907 - time (sec): 13.43 - samples/sec: 1844.66 - lr: 0.000024 - momentum: 0.000000
2024-03-26 11:22:19,703 epoch 3 - iter 81/95 - loss 0.21456530 - time (sec): 15.25 - samples/sec: 1833.54 - lr: 0.000024 - momentum: 0.000000
2024-03-26 11:22:22,391 epoch 3 - iter 90/95 - loss 0.19523106 - time (sec): 17.94 - samples/sec: 1823.40 - lr: 0.000024 - momentum: 0.000000
2024-03-26 11:22:23,574 ----------------------------------------------------------------------------------------------------
2024-03-26 11:22:23,574 EPOCH 3 done: loss 0.1913 - lr: 0.000024
2024-03-26 11:22:24,510 DEV : loss 0.240545853972435 - f1-score (micro avg)  0.8746
2024-03-26 11:22:24,511 saving best model
2024-03-26 11:22:24,945 ----------------------------------------------------------------------------------------------------
2024-03-26 11:22:26,686 epoch 4 - iter 9/95 - loss 0.17933388 - time (sec): 1.74 - samples/sec: 1848.00 - lr: 0.000023 - momentum: 0.000000
2024-03-26 11:22:28,783 epoch 4 - iter 18/95 - loss 0.14876494 - time (sec): 3.84 - samples/sec: 1755.97 - lr: 0.000023 - momentum: 0.000000
2024-03-26 11:22:30,025 epoch 4 - iter 27/95 - loss 0.13968538 - time (sec): 5.08 - samples/sec: 1853.38 - lr: 0.000022 - momentum: 0.000000
2024-03-26 11:22:31,729 epoch 4 - iter 36/95 - loss 0.13484846 - time (sec): 6.78 - samples/sec: 1830.28 - lr: 0.000022 - momentum: 0.000000
2024-03-26 11:22:33,950 epoch 4 - iter 45/95 - loss 0.13656782 - time (sec): 9.00 - samples/sec: 1773.50 - lr: 0.000022 - momentum: 0.000000
2024-03-26 11:22:35,519 epoch 4 - iter 54/95 - loss 0.14503066 - time (sec): 10.57 - samples/sec: 1785.26 - lr: 0.000022 - momentum: 0.000000
2024-03-26 11:22:38,032 epoch 4 - iter 63/95 - loss 0.13934337 - time (sec): 13.08 - samples/sec: 1742.44 - lr: 0.000021 - momentum: 0.000000
2024-03-26 11:22:40,579 epoch 4 - iter 72/95 - loss 0.12999178 - time (sec): 15.63 - samples/sec: 1711.27 - lr: 0.000021 - momentum: 0.000000
2024-03-26 11:22:42,044 epoch 4 - iter 81/95 - loss 0.12819267 - time (sec): 17.10 - samples/sec: 1718.80 - lr: 0.000021 - momentum: 0.000000
2024-03-26 11:22:43,874 epoch 4 - iter 90/95 - loss 0.12899945 - time (sec): 18.93 - samples/sec: 1718.76 - lr: 0.000020 - momentum: 0.000000
2024-03-26 11:22:45,041 ----------------------------------------------------------------------------------------------------
2024-03-26 11:22:45,041 EPOCH 4 done: loss 0.1254 - lr: 0.000020
2024-03-26 11:22:45,970 DEV : loss 0.20030681788921356 - f1-score (micro avg)  0.8864
2024-03-26 11:22:45,971 saving best model
2024-03-26 11:22:46,388 ----------------------------------------------------------------------------------------------------
2024-03-26 11:22:47,357 epoch 5 - iter 9/95 - loss 0.06968794 - time (sec): 0.97 - samples/sec: 2130.62 - lr: 0.000020 - momentum: 0.000000
2024-03-26 11:22:48,987 epoch 5 - iter 18/95 - loss 0.08422334 - time (sec): 2.60 - samples/sec: 2049.60 - lr: 0.000019 - momentum: 0.000000
2024-03-26 11:22:51,634 epoch 5 - iter 27/95 - loss 0.08687954 - time (sec): 5.24 - samples/sec: 1738.57 - lr: 0.000019 - momentum: 0.000000
2024-03-26 11:22:53,530 epoch 5 - iter 36/95 - loss 0.08428394 - time (sec): 7.14 - samples/sec: 1737.79 - lr: 0.000019 - momentum: 0.000000
2024-03-26 11:22:55,536 epoch 5 - iter 45/95 - loss 0.08119957 - time (sec): 9.15 - samples/sec: 1709.01 - lr: 0.000019 - momentum: 0.000000
2024-03-26 11:22:57,188 epoch 5 - iter 54/95 - loss 0.08334333 - time (sec): 10.80 - samples/sec: 1745.18 - lr: 0.000018 - momentum: 0.000000
2024-03-26 11:22:59,620 epoch 5 - iter 63/95 - loss 0.08575493 - time (sec): 13.23 - samples/sec: 1731.33 - lr: 0.000018 - momentum: 0.000000
2024-03-26 11:23:01,050 epoch 5 - iter 72/95 - loss 0.09325589 - time (sec): 14.66 - samples/sec: 1752.28 - lr: 0.000018 - momentum: 0.000000
2024-03-26 11:23:02,976 epoch 5 - iter 81/95 - loss 0.08887188 - time (sec): 16.59 - samples/sec: 1731.37 - lr: 0.000017 - momentum: 0.000000
2024-03-26 11:23:04,876 epoch 5 - iter 90/95 - loss 0.09027369 - time (sec): 18.49 - samples/sec: 1733.99 - lr: 0.000017 - momentum: 0.000000
2024-03-26 11:23:06,246 ----------------------------------------------------------------------------------------------------
2024-03-26 11:23:06,246 EPOCH 5 done: loss 0.0912 - lr: 0.000017
2024-03-26 11:23:07,179 DEV : loss 0.21548356115818024 - f1-score (micro avg)  0.9065
2024-03-26 11:23:07,180 saving best model
2024-03-26 11:23:07,618 ----------------------------------------------------------------------------------------------------
2024-03-26 11:23:09,068 epoch 6 - iter 9/95 - loss 0.06991012 - time (sec): 1.45 - samples/sec: 1987.50 - lr: 0.000016 - momentum: 0.000000
2024-03-26 11:23:11,278 epoch 6 - iter 18/95 - loss 0.06426227 - time (sec): 3.66 - samples/sec: 1959.95 - lr: 0.000016 - momentum: 0.000000
2024-03-26 11:23:12,875 epoch 6 - iter 27/95 - loss 0.06315170 - time (sec): 5.26 - samples/sec: 1910.70 - lr: 0.000016 - momentum: 0.000000
2024-03-26 11:23:14,916 epoch 6 - iter 36/95 - loss 0.06999208 - time (sec): 7.30 - samples/sec: 1850.41 - lr: 0.000016 - momentum: 0.000000
2024-03-26 11:23:17,134 epoch 6 - iter 45/95 - loss 0.08174634 - time (sec): 9.51 - samples/sec: 1867.15 - lr: 0.000015 - momentum: 0.000000
2024-03-26 11:23:18,345 epoch 6 - iter 54/95 - loss 0.07922099 - time (sec): 10.73 - samples/sec: 1884.83 - lr: 0.000015 - momentum: 0.000000
2024-03-26 11:23:19,439 epoch 6 - iter 63/95 - loss 0.07926444 - time (sec): 11.82 - samples/sec: 1904.88 - lr: 0.000015 - momentum: 0.000000
2024-03-26 11:23:21,006 epoch 6 - iter 72/95 - loss 0.07362739 - time (sec): 13.39 - samples/sec: 1907.35 - lr: 0.000014 - momentum: 0.000000
2024-03-26 11:23:23,060 epoch 6 - iter 81/95 - loss 0.07358180 - time (sec): 15.44 - samples/sec: 1893.48 - lr: 0.000014 - momentum: 0.000000
2024-03-26 11:23:25,114 epoch 6 - iter 90/95 - loss 0.07329242 - time (sec): 17.49 - samples/sec: 1880.01 - lr: 0.000014 - momentum: 0.000000
2024-03-26 11:23:26,077 ----------------------------------------------------------------------------------------------------
2024-03-26 11:23:26,077 EPOCH 6 done: loss 0.0711 - lr: 0.000014
2024-03-26 11:23:27,018 DEV : loss 0.19396518170833588 - f1-score (micro avg)  0.9188
2024-03-26 11:23:27,020 saving best model
2024-03-26 11:23:27,446 ----------------------------------------------------------------------------------------------------
2024-03-26 11:23:28,896 epoch 7 - iter 9/95 - loss 0.03784945 - time (sec): 1.45 - samples/sec: 1835.43 - lr: 0.000013 - momentum: 0.000000
2024-03-26 11:23:30,765 epoch 7 - iter 18/95 - loss 0.05259004 - time (sec): 3.32 - samples/sec: 1747.20 - lr: 0.000013 - momentum: 0.000000
2024-03-26 11:23:32,415 epoch 7 - iter 27/95 - loss 0.04797867 - time (sec): 4.97 - samples/sec: 1836.54 - lr: 0.000013 - momentum: 0.000000
2024-03-26 11:23:34,192 epoch 7 - iter 36/95 - loss 0.04941735 - time (sec): 6.74 - samples/sec: 1783.27 - lr: 0.000012 - momentum: 0.000000
2024-03-26 11:23:35,573 epoch 7 - iter 45/95 - loss 0.04881865 - time (sec): 8.13 - samples/sec: 1804.48 - lr: 0.000012 - momentum: 0.000000
2024-03-26 11:23:37,685 epoch 7 - iter 54/95 - loss 0.04949831 - time (sec): 10.24 - samples/sec: 1751.01 - lr: 0.000012 - momentum: 0.000000
2024-03-26 11:23:39,986 epoch 7 - iter 63/95 - loss 0.04927595 - time (sec): 12.54 - samples/sec: 1702.76 - lr: 0.000011 - momentum: 0.000000
2024-03-26 11:23:42,684 epoch 7 - iter 72/95 - loss 0.05611880 - time (sec): 15.24 - samples/sec: 1691.98 - lr: 0.000011 - momentum: 0.000000
2024-03-26 11:23:44,691 epoch 7 - iter 81/95 - loss 0.06216960 - time (sec): 17.24 - samples/sec: 1699.79 - lr: 0.000011 - momentum: 0.000000
2024-03-26 11:23:46,711 epoch 7 - iter 90/95 - loss 0.06210809 - time (sec): 19.26 - samples/sec: 1701.56 - lr: 0.000010 - momentum: 0.000000
2024-03-26 11:23:47,666 ----------------------------------------------------------------------------------------------------
2024-03-26 11:23:47,666 EPOCH 7 done: loss 0.0607 - lr: 0.000010
2024-03-26 11:23:48,604 DEV : loss 0.18384647369384766 - f1-score (micro avg)  0.9245
2024-03-26 11:23:48,605 saving best model
2024-03-26 11:23:49,027 ----------------------------------------------------------------------------------------------------
2024-03-26 11:23:51,337 epoch 8 - iter 9/95 - loss 0.05530984 - time (sec): 2.31 - samples/sec: 1640.94 - lr: 0.000010 - momentum: 0.000000
2024-03-26 11:23:52,924 epoch 8 - iter 18/95 - loss 0.05076397 - time (sec): 3.90 - samples/sec: 1769.13 - lr: 0.000010 - momentum: 0.000000
2024-03-26 11:23:55,113 epoch 8 - iter 27/95 - loss 0.05844996 - time (sec): 6.08 - samples/sec: 1737.74 - lr: 0.000009 - momentum: 0.000000
2024-03-26 11:23:56,689 epoch 8 - iter 36/95 - loss 0.05317296 - time (sec): 7.66 - samples/sec: 1761.32 - lr: 0.000009 - momentum: 0.000000
2024-03-26 11:23:58,620 epoch 8 - iter 45/95 - loss 0.04686965 - time (sec): 9.59 - samples/sec: 1735.73 - lr: 0.000009 - momentum: 0.000000
2024-03-26 11:24:00,380 epoch 8 - iter 54/95 - loss 0.04922465 - time (sec): 11.35 - samples/sec: 1740.26 - lr: 0.000008 - momentum: 0.000000
2024-03-26 11:24:02,193 epoch 8 - iter 63/95 - loss 0.04837573 - time (sec): 13.16 - samples/sec: 1744.40 - lr: 0.000008 - momentum: 0.000000
2024-03-26 11:24:03,520 epoch 8 - iter 72/95 - loss 0.04744390 - time (sec): 14.49 - samples/sec: 1765.34 - lr: 0.000008 - momentum: 0.000000
2024-03-26 11:24:05,405 epoch 8 - iter 81/95 - loss 0.04838397 - time (sec): 16.38 - samples/sec: 1787.88 - lr: 0.000007 - momentum: 0.000000
2024-03-26 11:24:07,937 epoch 8 - iter 90/95 - loss 0.04553123 - time (sec): 18.91 - samples/sec: 1744.50 - lr: 0.000007 - momentum: 0.000000
2024-03-26 11:24:08,789 ----------------------------------------------------------------------------------------------------
2024-03-26 11:24:08,789 EPOCH 8 done: loss 0.0458 - lr: 0.000007
2024-03-26 11:24:09,737 DEV : loss 0.20191706717014313 - f1-score (micro avg)  0.9334
2024-03-26 11:24:09,738 saving best model
2024-03-26 11:24:10,138 ----------------------------------------------------------------------------------------------------
2024-03-26 11:24:11,917 epoch 9 - iter 9/95 - loss 0.04832010 - time (sec): 1.78 - samples/sec: 1911.27 - lr: 0.000007 - momentum: 0.000000
2024-03-26 11:24:14,236 epoch 9 - iter 18/95 - loss 0.03665657 - time (sec): 4.10 - samples/sec: 1692.76 - lr: 0.000006 - momentum: 0.000000
2024-03-26 11:24:16,172 epoch 9 - iter 27/95 - loss 0.04227514 - time (sec): 6.03 - samples/sec: 1729.90 - lr: 0.000006 - momentum: 0.000000
2024-03-26 11:24:17,756 epoch 9 - iter 36/95 - loss 0.04242815 - time (sec): 7.62 - samples/sec: 1743.09 - lr: 0.000006 - momentum: 0.000000
2024-03-26 11:24:19,202 epoch 9 - iter 45/95 - loss 0.03653640 - time (sec): 9.06 - samples/sec: 1780.07 - lr: 0.000005 - momentum: 0.000000
2024-03-26 11:24:20,629 epoch 9 - iter 54/95 - loss 0.03452892 - time (sec): 10.49 - samples/sec: 1833.64 - lr: 0.000005 - momentum: 0.000000
2024-03-26 11:24:22,536 epoch 9 - iter 63/95 - loss 0.03920444 - time (sec): 12.40 - samples/sec: 1833.45 - lr: 0.000005 - momentum: 0.000000
2024-03-26 11:24:24,590 epoch 9 - iter 72/95 - loss 0.04021237 - time (sec): 14.45 - samples/sec: 1807.68 - lr: 0.000004 - momentum: 0.000000
2024-03-26 11:24:26,931 epoch 9 - iter 81/95 - loss 0.04095813 - time (sec): 16.79 - samples/sec: 1767.62 - lr: 0.000004 - momentum: 0.000000
2024-03-26 11:24:28,720 epoch 9 - iter 90/95 - loss 0.04018029 - time (sec): 18.58 - samples/sec: 1781.36 - lr: 0.000004 - momentum: 0.000000
2024-03-26 11:24:29,321 ----------------------------------------------------------------------------------------------------
2024-03-26 11:24:29,321 EPOCH 9 done: loss 0.0401 - lr: 0.000004
2024-03-26 11:24:30,254 DEV : loss 0.19197042286396027 - f1-score (micro avg)  0.9391
2024-03-26 11:24:30,255 saving best model
2024-03-26 11:24:30,673 ----------------------------------------------------------------------------------------------------
2024-03-26 11:24:32,792 epoch 10 - iter 9/95 - loss 0.00978685 - time (sec): 2.12 - samples/sec: 1824.59 - lr: 0.000003 - momentum: 0.000000
2024-03-26 11:24:34,600 epoch 10 - iter 18/95 - loss 0.02419535 - time (sec): 3.92 - samples/sec: 1810.09 - lr: 0.000003 - momentum: 0.000000
2024-03-26 11:24:35,713 epoch 10 - iter 27/95 - loss 0.02107737 - time (sec): 5.04 - samples/sec: 1890.85 - lr: 0.000003 - momentum: 0.000000
2024-03-26 11:24:37,186 epoch 10 - iter 36/95 - loss 0.02692307 - time (sec): 6.51 - samples/sec: 1923.00 - lr: 0.000002 - momentum: 0.000000
2024-03-26 11:24:39,247 epoch 10 - iter 45/95 - loss 0.03594101 - time (sec): 8.57 - samples/sec: 1841.70 - lr: 0.000002 - momentum: 0.000000
2024-03-26 11:24:40,359 epoch 10 - iter 54/95 - loss 0.03803968 - time (sec): 9.68 - samples/sec: 1890.32 - lr: 0.000002 - momentum: 0.000000
2024-03-26 11:24:41,627 epoch 10 - iter 63/95 - loss 0.03512331 - time (sec): 10.95 - samples/sec: 1915.96 - lr: 0.000001 - momentum: 0.000000
2024-03-26 11:24:43,578 epoch 10 - iter 72/95 - loss 0.03520742 - time (sec): 12.90 - samples/sec: 1913.85 - lr: 0.000001 - momentum: 0.000000
2024-03-26 11:24:46,230 epoch 10 - iter 81/95 - loss 0.03414263 - time (sec): 15.55 - samples/sec: 1883.19 - lr: 0.000001 - momentum: 0.000000
2024-03-26 11:24:48,329 epoch 10 - iter 90/95 - loss 0.03430259 - time (sec): 17.65 - samples/sec: 1858.65 - lr: 0.000000 - momentum: 0.000000
2024-03-26 11:24:49,281 ----------------------------------------------------------------------------------------------------
2024-03-26 11:24:49,281 EPOCH 10 done: loss 0.0340 - lr: 0.000000
2024-03-26 11:24:50,220 DEV : loss 0.20091010630130768 - f1-score (micro avg)  0.9305
2024-03-26 11:24:50,504 ----------------------------------------------------------------------------------------------------
2024-03-26 11:24:50,504 Loading model from best epoch ...
2024-03-26 11:24:51,344 SequenceTagger predicts: Dictionary with 17 tags: O, S-Unternehmen, B-Unternehmen, E-Unternehmen, I-Unternehmen, S-Auslagerung, B-Auslagerung, E-Auslagerung, I-Auslagerung, S-Ort, B-Ort, E-Ort, I-Ort, S-Software, B-Software, E-Software, I-Software
2024-03-26 11:24:52,097 
Results:
- F-score (micro) 0.9126
- F-score (macro) 0.6925
- Accuracy 0.8416

By class:
              precision    recall  f1-score   support

 Unternehmen     0.9091    0.9023    0.9057       266
 Auslagerung     0.8849    0.8956    0.8902       249
         Ort     0.9635    0.9851    0.9742       134
    Software     0.0000    0.0000    0.0000         0

   micro avg     0.9084    0.9168    0.9126       649
   macro avg     0.6894    0.6957    0.6925       649
weighted avg     0.9111    0.9168    0.9139       649

2024-03-26 11:24:52,097 ----------------------------------------------------------------------------------------------------