Hugging Face's logo

stefan-it
/
flair-co-funer-german_bert_base-bs8-e10-lr5e-05-4

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-lr5e-05-4
File size: 23,737 Bytes 
d000a19
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
 
2024-03-26 11:59:09,057 ----------------------------------------------------------------------------------------------------
2024-03-26 11:59:09,057 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 11:59:09,057 ----------------------------------------------------------------------------------------------------
2024-03-26 11:59:09,057 Corpus: 758 train + 94 dev + 96 test sentences
2024-03-26 11:59:09,057 ----------------------------------------------------------------------------------------------------
2024-03-26 11:59:09,057 Train:  758 sentences
2024-03-26 11:59:09,057         (train_with_dev=False, train_with_test=False)
2024-03-26 11:59:09,057 ----------------------------------------------------------------------------------------------------
2024-03-26 11:59:09,057 Training Params:
2024-03-26 11:59:09,057  - learning_rate: "5e-05" 
2024-03-26 11:59:09,057  - mini_batch_size: "8"
2024-03-26 11:59:09,057  - max_epochs: "10"
2024-03-26 11:59:09,057  - shuffle: "True"
2024-03-26 11:59:09,057 ----------------------------------------------------------------------------------------------------
2024-03-26 11:59:09,057 Plugins:
2024-03-26 11:59:09,057  - TensorboardLogger
2024-03-26 11:59:09,057  - LinearScheduler | warmup_fraction: '0.1'
2024-03-26 11:59:09,057 ----------------------------------------------------------------------------------------------------
2024-03-26 11:59:09,057 Final evaluation on model from best epoch (best-model.pt)
2024-03-26 11:59:09,057  - metric: "('micro avg', 'f1-score')"
2024-03-26 11:59:09,057 ----------------------------------------------------------------------------------------------------
2024-03-26 11:59:09,057 Computation:
2024-03-26 11:59:09,057  - compute on device: cuda:0
2024-03-26 11:59:09,057  - embedding storage: none
2024-03-26 11:59:09,057 ----------------------------------------------------------------------------------------------------
2024-03-26 11:59:09,057 Model training base path: "flair-co-funer-german_bert_base-bs8-e10-lr5e-05-4"
2024-03-26 11:59:09,057 ----------------------------------------------------------------------------------------------------
2024-03-26 11:59:09,057 ----------------------------------------------------------------------------------------------------
2024-03-26 11:59:09,058 Logging anything other than scalars to TensorBoard is currently not supported.
2024-03-26 11:59:10,418 epoch 1 - iter 9/95 - loss 3.11551508 - time (sec): 1.36 - samples/sec: 2132.11 - lr: 0.000004 - momentum: 0.000000
2024-03-26 11:59:11,850 epoch 1 - iter 18/95 - loss 2.91592715 - time (sec): 2.79 - samples/sec: 1975.14 - lr: 0.000009 - momentum: 0.000000
2024-03-26 11:59:13,523 epoch 1 - iter 27/95 - loss 2.63491605 - time (sec): 4.46 - samples/sec: 1920.85 - lr: 0.000014 - momentum: 0.000000
2024-03-26 11:59:15,532 epoch 1 - iter 36/95 - loss 2.39303542 - time (sec): 6.47 - samples/sec: 1843.82 - lr: 0.000018 - momentum: 0.000000
2024-03-26 11:59:17,444 epoch 1 - iter 45/95 - loss 2.19445986 - time (sec): 8.39 - samples/sec: 1859.55 - lr: 0.000023 - momentum: 0.000000
2024-03-26 11:59:19,715 epoch 1 - iter 54/95 - loss 2.01807178 - time (sec): 10.66 - samples/sec: 1796.53 - lr: 0.000028 - momentum: 0.000000
2024-03-26 11:59:21,763 epoch 1 - iter 63/95 - loss 1.85928116 - time (sec): 12.71 - samples/sec: 1778.94 - lr: 0.000033 - momentum: 0.000000
2024-03-26 11:59:22,764 epoch 1 - iter 72/95 - loss 1.76634872 - time (sec): 13.71 - samples/sec: 1822.57 - lr: 0.000037 - momentum: 0.000000
2024-03-26 11:59:25,096 epoch 1 - iter 81/95 - loss 1.62019393 - time (sec): 16.04 - samples/sec: 1771.09 - lr: 0.000042 - momentum: 0.000000
2024-03-26 11:59:26,515 epoch 1 - iter 90/95 - loss 1.49289861 - time (sec): 17.46 - samples/sec: 1830.77 - lr: 0.000047 - momentum: 0.000000
2024-03-26 11:59:27,817 ----------------------------------------------------------------------------------------------------
2024-03-26 11:59:27,817 EPOCH 1 done: loss 1.4311 - lr: 0.000047
2024-03-26 11:59:28,673 DEV : loss 0.3974745273590088 - f1-score (micro avg)  0.7207
2024-03-26 11:59:28,674 saving best model
2024-03-26 11:59:28,954 ----------------------------------------------------------------------------------------------------
2024-03-26 11:59:30,567 epoch 2 - iter 9/95 - loss 0.51132915 - time (sec): 1.61 - samples/sec: 1787.31 - lr: 0.000050 - momentum: 0.000000
2024-03-26 11:59:32,312 epoch 2 - iter 18/95 - loss 0.41287581 - time (sec): 3.36 - samples/sec: 1840.74 - lr: 0.000049 - momentum: 0.000000
2024-03-26 11:59:34,170 epoch 2 - iter 27/95 - loss 0.38177177 - time (sec): 5.22 - samples/sec: 1813.09 - lr: 0.000048 - momentum: 0.000000
2024-03-26 11:59:36,623 epoch 2 - iter 36/95 - loss 0.33454736 - time (sec): 7.67 - samples/sec: 1701.24 - lr: 0.000048 - momentum: 0.000000
2024-03-26 11:59:38,634 epoch 2 - iter 45/95 - loss 0.32462192 - time (sec): 9.68 - samples/sec: 1704.50 - lr: 0.000047 - momentum: 0.000000
2024-03-26 11:59:40,436 epoch 2 - iter 54/95 - loss 0.33859423 - time (sec): 11.48 - samples/sec: 1727.29 - lr: 0.000047 - momentum: 0.000000
2024-03-26 11:59:42,905 epoch 2 - iter 63/95 - loss 0.32027748 - time (sec): 13.95 - samples/sec: 1713.90 - lr: 0.000046 - momentum: 0.000000
2024-03-26 11:59:44,796 epoch 2 - iter 72/95 - loss 0.31809533 - time (sec): 15.84 - samples/sec: 1707.05 - lr: 0.000046 - momentum: 0.000000
2024-03-26 11:59:47,076 epoch 2 - iter 81/95 - loss 0.30946687 - time (sec): 18.12 - samples/sec: 1688.25 - lr: 0.000045 - momentum: 0.000000
2024-03-26 11:59:48,400 epoch 2 - iter 90/95 - loss 0.30584654 - time (sec): 19.45 - samples/sec: 1713.26 - lr: 0.000045 - momentum: 0.000000
2024-03-26 11:59:48,844 ----------------------------------------------------------------------------------------------------
2024-03-26 11:59:48,845 EPOCH 2 done: loss 0.3017 - lr: 0.000045
2024-03-26 11:59:49,798 DEV : loss 0.24508343636989594 - f1-score (micro avg)  0.8649
2024-03-26 11:59:49,800 saving best model
2024-03-26 11:59:50,246 ----------------------------------------------------------------------------------------------------
2024-03-26 11:59:51,810 epoch 3 - iter 9/95 - loss 0.17662022 - time (sec): 1.56 - samples/sec: 1694.57 - lr: 0.000044 - momentum: 0.000000
2024-03-26 11:59:53,538 epoch 3 - iter 18/95 - loss 0.14946269 - time (sec): 3.29 - samples/sec: 1697.74 - lr: 0.000043 - momentum: 0.000000
2024-03-26 11:59:55,440 epoch 3 - iter 27/95 - loss 0.17150757 - time (sec): 5.19 - samples/sec: 1714.83 - lr: 0.000043 - momentum: 0.000000
2024-03-26 11:59:57,385 epoch 3 - iter 36/95 - loss 0.17605043 - time (sec): 7.14 - samples/sec: 1690.97 - lr: 0.000042 - momentum: 0.000000
2024-03-26 11:59:59,448 epoch 3 - iter 45/95 - loss 0.17196591 - time (sec): 9.20 - samples/sec: 1715.95 - lr: 0.000042 - momentum: 0.000000
2024-03-26 12:00:01,784 epoch 3 - iter 54/95 - loss 0.16879250 - time (sec): 11.54 - samples/sec: 1678.78 - lr: 0.000041 - momentum: 0.000000
2024-03-26 12:00:03,576 epoch 3 - iter 63/95 - loss 0.16109343 - time (sec): 13.33 - samples/sec: 1678.31 - lr: 0.000041 - momentum: 0.000000
2024-03-26 12:00:05,589 epoch 3 - iter 72/95 - loss 0.15752743 - time (sec): 15.34 - samples/sec: 1687.28 - lr: 0.000040 - momentum: 0.000000
2024-03-26 12:00:07,623 epoch 3 - iter 81/95 - loss 0.16724057 - time (sec): 17.38 - samples/sec: 1704.42 - lr: 0.000040 - momentum: 0.000000
2024-03-26 12:00:09,915 epoch 3 - iter 90/95 - loss 0.16469999 - time (sec): 19.67 - samples/sec: 1686.10 - lr: 0.000039 - momentum: 0.000000
2024-03-26 12:00:10,534 ----------------------------------------------------------------------------------------------------
2024-03-26 12:00:10,534 EPOCH 3 done: loss 0.1671 - lr: 0.000039
2024-03-26 12:00:11,475 DEV : loss 0.20654602348804474 - f1-score (micro avg)  0.866
2024-03-26 12:00:11,476 saving best model
2024-03-26 12:00:11,925 ----------------------------------------------------------------------------------------------------
2024-03-26 12:00:14,379 epoch 4 - iter 9/95 - loss 0.06959102 - time (sec): 2.45 - samples/sec: 1612.63 - lr: 0.000039 - momentum: 0.000000
2024-03-26 12:00:15,556 epoch 4 - iter 18/95 - loss 0.07924386 - time (sec): 3.63 - samples/sec: 1763.18 - lr: 0.000038 - momentum: 0.000000
2024-03-26 12:00:17,785 epoch 4 - iter 27/95 - loss 0.09207853 - time (sec): 5.86 - samples/sec: 1774.88 - lr: 0.000037 - momentum: 0.000000
2024-03-26 12:00:19,293 epoch 4 - iter 36/95 - loss 0.09786047 - time (sec): 7.37 - samples/sec: 1810.88 - lr: 0.000037 - momentum: 0.000000
2024-03-26 12:00:20,634 epoch 4 - iter 45/95 - loss 0.09845056 - time (sec): 8.71 - samples/sec: 1842.74 - lr: 0.000036 - momentum: 0.000000
2024-03-26 12:00:22,742 epoch 4 - iter 54/95 - loss 0.09663956 - time (sec): 10.82 - samples/sec: 1789.42 - lr: 0.000036 - momentum: 0.000000
2024-03-26 12:00:25,103 epoch 4 - iter 63/95 - loss 0.10785088 - time (sec): 13.18 - samples/sec: 1757.00 - lr: 0.000035 - momentum: 0.000000
2024-03-26 12:00:26,614 epoch 4 - iter 72/95 - loss 0.10490869 - time (sec): 14.69 - samples/sec: 1789.10 - lr: 0.000035 - momentum: 0.000000
2024-03-26 12:00:28,246 epoch 4 - iter 81/95 - loss 0.10385446 - time (sec): 16.32 - samples/sec: 1819.91 - lr: 0.000034 - momentum: 0.000000
2024-03-26 12:00:29,896 epoch 4 - iter 90/95 - loss 0.10468697 - time (sec): 17.97 - samples/sec: 1847.82 - lr: 0.000034 - momentum: 0.000000
2024-03-26 12:00:30,524 ----------------------------------------------------------------------------------------------------
2024-03-26 12:00:30,524 EPOCH 4 done: loss 0.1058 - lr: 0.000034
2024-03-26 12:00:31,464 DEV : loss 0.20639045536518097 - f1-score (micro avg)  0.9061
2024-03-26 12:00:31,465 saving best model
2024-03-26 12:00:31,899 ----------------------------------------------------------------------------------------------------
2024-03-26 12:00:33,151 epoch 5 - iter 9/95 - loss 0.11849622 - time (sec): 1.25 - samples/sec: 2367.15 - lr: 0.000033 - momentum: 0.000000
2024-03-26 12:00:34,644 epoch 5 - iter 18/95 - loss 0.09990788 - time (sec): 2.74 - samples/sec: 2125.68 - lr: 0.000032 - momentum: 0.000000
2024-03-26 12:00:36,707 epoch 5 - iter 27/95 - loss 0.08403115 - time (sec): 4.80 - samples/sec: 1913.94 - lr: 0.000032 - momentum: 0.000000
2024-03-26 12:00:39,162 epoch 5 - iter 36/95 - loss 0.08445462 - time (sec): 7.26 - samples/sec: 1749.59 - lr: 0.000031 - momentum: 0.000000
2024-03-26 12:00:40,406 epoch 5 - iter 45/95 - loss 0.09122409 - time (sec): 8.50 - samples/sec: 1796.12 - lr: 0.000031 - momentum: 0.000000
2024-03-26 12:00:42,378 epoch 5 - iter 54/95 - loss 0.08649690 - time (sec): 10.48 - samples/sec: 1827.16 - lr: 0.000030 - momentum: 0.000000
2024-03-26 12:00:44,512 epoch 5 - iter 63/95 - loss 0.08178669 - time (sec): 12.61 - samples/sec: 1812.39 - lr: 0.000030 - momentum: 0.000000
2024-03-26 12:00:45,811 epoch 5 - iter 72/95 - loss 0.08147182 - time (sec): 13.91 - samples/sec: 1840.43 - lr: 0.000029 - momentum: 0.000000
2024-03-26 12:00:48,448 epoch 5 - iter 81/95 - loss 0.07532693 - time (sec): 16.55 - samples/sec: 1775.15 - lr: 0.000029 - momentum: 0.000000
2024-03-26 12:00:50,528 epoch 5 - iter 90/95 - loss 0.07517831 - time (sec): 18.63 - samples/sec: 1757.85 - lr: 0.000028 - momentum: 0.000000
2024-03-26 12:00:51,418 ----------------------------------------------------------------------------------------------------
2024-03-26 12:00:51,418 EPOCH 5 done: loss 0.0778 - lr: 0.000028
2024-03-26 12:00:52,372 DEV : loss 0.1956116110086441 - f1-score (micro avg)  0.9178
2024-03-26 12:00:52,374 saving best model
2024-03-26 12:00:52,849 ----------------------------------------------------------------------------------------------------
2024-03-26 12:00:54,564 epoch 6 - iter 9/95 - loss 0.09104755 - time (sec): 1.71 - samples/sec: 1934.88 - lr: 0.000027 - momentum: 0.000000
2024-03-26 12:00:56,671 epoch 6 - iter 18/95 - loss 0.06055749 - time (sec): 3.82 - samples/sec: 1773.75 - lr: 0.000027 - momentum: 0.000000
2024-03-26 12:00:58,142 epoch 6 - iter 27/95 - loss 0.06306007 - time (sec): 5.29 - samples/sec: 1805.78 - lr: 0.000026 - momentum: 0.000000
2024-03-26 12:01:00,604 epoch 6 - iter 36/95 - loss 0.05149544 - time (sec): 7.75 - samples/sec: 1656.27 - lr: 0.000026 - momentum: 0.000000
2024-03-26 12:01:02,418 epoch 6 - iter 45/95 - loss 0.04808451 - time (sec): 9.57 - samples/sec: 1681.48 - lr: 0.000025 - momentum: 0.000000
2024-03-26 12:01:04,989 epoch 6 - iter 54/95 - loss 0.05677027 - time (sec): 12.14 - samples/sec: 1658.67 - lr: 0.000025 - momentum: 0.000000
2024-03-26 12:01:06,572 epoch 6 - iter 63/95 - loss 0.05658119 - time (sec): 13.72 - samples/sec: 1672.98 - lr: 0.000024 - momentum: 0.000000
2024-03-26 12:01:08,132 epoch 6 - iter 72/95 - loss 0.05676127 - time (sec): 15.28 - samples/sec: 1697.53 - lr: 0.000024 - momentum: 0.000000
2024-03-26 12:01:10,274 epoch 6 - iter 81/95 - loss 0.05615297 - time (sec): 17.42 - samples/sec: 1692.00 - lr: 0.000023 - momentum: 0.000000
2024-03-26 12:01:11,487 epoch 6 - iter 90/95 - loss 0.05920230 - time (sec): 18.64 - samples/sec: 1735.54 - lr: 0.000023 - momentum: 0.000000
2024-03-26 12:01:12,892 ----------------------------------------------------------------------------------------------------
2024-03-26 12:01:12,892 EPOCH 6 done: loss 0.0576 - lr: 0.000023
2024-03-26 12:01:13,859 DEV : loss 0.19573496282100677 - f1-score (micro avg)  0.9042
2024-03-26 12:01:13,861 ----------------------------------------------------------------------------------------------------
2024-03-26 12:01:15,247 epoch 7 - iter 9/95 - loss 0.03915603 - time (sec): 1.39 - samples/sec: 2290.04 - lr: 0.000022 - momentum: 0.000000
2024-03-26 12:01:17,434 epoch 7 - iter 18/95 - loss 0.03185381 - time (sec): 3.57 - samples/sec: 1887.66 - lr: 0.000021 - momentum: 0.000000
2024-03-26 12:01:19,422 epoch 7 - iter 27/95 - loss 0.03946797 - time (sec): 5.56 - samples/sec: 1757.96 - lr: 0.000021 - momentum: 0.000000
2024-03-26 12:01:20,775 epoch 7 - iter 36/95 - loss 0.03799302 - time (sec): 6.91 - samples/sec: 1809.47 - lr: 0.000020 - momentum: 0.000000
2024-03-26 12:01:22,488 epoch 7 - iter 45/95 - loss 0.03831134 - time (sec): 8.63 - samples/sec: 1824.07 - lr: 0.000020 - momentum: 0.000000
2024-03-26 12:01:24,868 epoch 7 - iter 54/95 - loss 0.03448802 - time (sec): 11.01 - samples/sec: 1780.19 - lr: 0.000019 - momentum: 0.000000
2024-03-26 12:01:26,952 epoch 7 - iter 63/95 - loss 0.03711925 - time (sec): 13.09 - samples/sec: 1739.36 - lr: 0.000019 - momentum: 0.000000
2024-03-26 12:01:29,235 epoch 7 - iter 72/95 - loss 0.03678211 - time (sec): 15.37 - samples/sec: 1707.00 - lr: 0.000018 - momentum: 0.000000
2024-03-26 12:01:30,770 epoch 7 - iter 81/95 - loss 0.04143827 - time (sec): 16.91 - samples/sec: 1717.57 - lr: 0.000018 - momentum: 0.000000
2024-03-26 12:01:32,742 epoch 7 - iter 90/95 - loss 0.04455468 - time (sec): 18.88 - samples/sec: 1742.11 - lr: 0.000017 - momentum: 0.000000
2024-03-26 12:01:33,439 ----------------------------------------------------------------------------------------------------
2024-03-26 12:01:33,439 EPOCH 7 done: loss 0.0439 - lr: 0.000017
2024-03-26 12:01:34,398 DEV : loss 0.20444811880588531 - f1-score (micro avg)  0.9106
2024-03-26 12:01:34,400 ----------------------------------------------------------------------------------------------------
2024-03-26 12:01:36,096 epoch 8 - iter 9/95 - loss 0.01074191 - time (sec): 1.70 - samples/sec: 1735.19 - lr: 0.000016 - momentum: 0.000000
2024-03-26 12:01:38,281 epoch 8 - iter 18/95 - loss 0.01421139 - time (sec): 3.88 - samples/sec: 1708.18 - lr: 0.000016 - momentum: 0.000000
2024-03-26 12:01:40,155 epoch 8 - iter 27/95 - loss 0.02248889 - time (sec): 5.75 - samples/sec: 1686.55 - lr: 0.000015 - momentum: 0.000000
2024-03-26 12:01:42,191 epoch 8 - iter 36/95 - loss 0.02397763 - time (sec): 7.79 - samples/sec: 1690.53 - lr: 0.000015 - momentum: 0.000000
2024-03-26 12:01:43,228 epoch 8 - iter 45/95 - loss 0.03170376 - time (sec): 8.83 - samples/sec: 1776.94 - lr: 0.000014 - momentum: 0.000000
2024-03-26 12:01:45,207 epoch 8 - iter 54/95 - loss 0.03599503 - time (sec): 10.81 - samples/sec: 1767.59 - lr: 0.000014 - momentum: 0.000000
2024-03-26 12:01:47,502 epoch 8 - iter 63/95 - loss 0.03927120 - time (sec): 13.10 - samples/sec: 1749.17 - lr: 0.000013 - momentum: 0.000000
2024-03-26 12:01:49,770 epoch 8 - iter 72/95 - loss 0.03887791 - time (sec): 15.37 - samples/sec: 1739.28 - lr: 0.000013 - momentum: 0.000000
2024-03-26 12:01:51,507 epoch 8 - iter 81/95 - loss 0.03775282 - time (sec): 17.11 - samples/sec: 1744.68 - lr: 0.000012 - momentum: 0.000000
2024-03-26 12:01:53,554 epoch 8 - iter 90/95 - loss 0.03465067 - time (sec): 19.15 - samples/sec: 1733.30 - lr: 0.000012 - momentum: 0.000000
2024-03-26 12:01:54,156 ----------------------------------------------------------------------------------------------------
2024-03-26 12:01:54,156 EPOCH 8 done: loss 0.0348 - lr: 0.000012
2024-03-26 12:01:55,092 DEV : loss 0.22374196350574493 - f1-score (micro avg)  0.9219
2024-03-26 12:01:55,093 saving best model
2024-03-26 12:01:55,540 ----------------------------------------------------------------------------------------------------
2024-03-26 12:01:57,115 epoch 9 - iter 9/95 - loss 0.02255766 - time (sec): 1.57 - samples/sec: 2020.25 - lr: 0.000011 - momentum: 0.000000
2024-03-26 12:01:59,514 epoch 9 - iter 18/95 - loss 0.02441884 - time (sec): 3.97 - samples/sec: 1714.10 - lr: 0.000010 - momentum: 0.000000
2024-03-26 12:02:01,166 epoch 9 - iter 27/95 - loss 0.02100220 - time (sec): 5.62 - samples/sec: 1732.76 - lr: 0.000010 - momentum: 0.000000
2024-03-26 12:02:03,532 epoch 9 - iter 36/95 - loss 0.02224059 - time (sec): 7.99 - samples/sec: 1697.76 - lr: 0.000009 - momentum: 0.000000
2024-03-26 12:02:05,456 epoch 9 - iter 45/95 - loss 0.02327845 - time (sec): 9.91 - samples/sec: 1678.15 - lr: 0.000009 - momentum: 0.000000
2024-03-26 12:02:06,885 epoch 9 - iter 54/95 - loss 0.02693223 - time (sec): 11.34 - samples/sec: 1720.22 - lr: 0.000008 - momentum: 0.000000
2024-03-26 12:02:09,040 epoch 9 - iter 63/95 - loss 0.02451513 - time (sec): 13.50 - samples/sec: 1698.43 - lr: 0.000008 - momentum: 0.000000
2024-03-26 12:02:10,272 epoch 9 - iter 72/95 - loss 0.02812439 - time (sec): 14.73 - samples/sec: 1735.32 - lr: 0.000007 - momentum: 0.000000
2024-03-26 12:02:13,095 epoch 9 - iter 81/95 - loss 0.02633959 - time (sec): 17.55 - samples/sec: 1691.71 - lr: 0.000007 - momentum: 0.000000
2024-03-26 12:02:14,755 epoch 9 - iter 90/95 - loss 0.02469000 - time (sec): 19.21 - samples/sec: 1714.60 - lr: 0.000006 - momentum: 0.000000
2024-03-26 12:02:15,429 ----------------------------------------------------------------------------------------------------
2024-03-26 12:02:15,429 EPOCH 9 done: loss 0.0262 - lr: 0.000006
2024-03-26 12:02:16,387 DEV : loss 0.22455130517482758 - f1-score (micro avg)  0.9291
2024-03-26 12:02:16,388 saving best model
2024-03-26 12:02:16,842 ----------------------------------------------------------------------------------------------------
2024-03-26 12:02:18,756 epoch 10 - iter 9/95 - loss 0.02413491 - time (sec): 1.91 - samples/sec: 1621.19 - lr: 0.000005 - momentum: 0.000000
2024-03-26 12:02:20,996 epoch 10 - iter 18/95 - loss 0.02195416 - time (sec): 4.15 - samples/sec: 1605.14 - lr: 0.000005 - momentum: 0.000000
2024-03-26 12:02:22,415 epoch 10 - iter 27/95 - loss 0.02161648 - time (sec): 5.57 - samples/sec: 1760.59 - lr: 0.000004 - momentum: 0.000000
2024-03-26 12:02:24,160 epoch 10 - iter 36/95 - loss 0.01921928 - time (sec): 7.32 - samples/sec: 1803.60 - lr: 0.000004 - momentum: 0.000000
2024-03-26 12:02:25,592 epoch 10 - iter 45/95 - loss 0.01857663 - time (sec): 8.75 - samples/sec: 1835.65 - lr: 0.000003 - momentum: 0.000000
2024-03-26 12:02:26,637 epoch 10 - iter 54/95 - loss 0.01707115 - time (sec): 9.79 - samples/sec: 1905.49 - lr: 0.000003 - momentum: 0.000000
2024-03-26 12:02:28,484 epoch 10 - iter 63/95 - loss 0.01548965 - time (sec): 11.64 - samples/sec: 1877.87 - lr: 0.000002 - momentum: 0.000000
2024-03-26 12:02:30,794 epoch 10 - iter 72/95 - loss 0.02073655 - time (sec): 13.95 - samples/sec: 1827.57 - lr: 0.000002 - momentum: 0.000000
2024-03-26 12:02:32,482 epoch 10 - iter 81/95 - loss 0.02161353 - time (sec): 15.64 - samples/sec: 1816.89 - lr: 0.000001 - momentum: 0.000000
2024-03-26 12:02:34,943 epoch 10 - iter 90/95 - loss 0.02064998 - time (sec): 18.10 - samples/sec: 1795.54 - lr: 0.000001 - momentum: 0.000000
2024-03-26 12:02:36,200 ----------------------------------------------------------------------------------------------------
2024-03-26 12:02:36,200 EPOCH 10 done: loss 0.0209 - lr: 0.000001
2024-03-26 12:02:37,134 DEV : loss 0.24257376790046692 - f1-score (micro avg)  0.9269
2024-03-26 12:02:37,396 ----------------------------------------------------------------------------------------------------
2024-03-26 12:02:37,397 Loading model from best epoch ...
2024-03-26 12:02:38,241 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 12:02:39,079 
Results:
- F-score (micro) 0.9062
- F-score (macro) 0.6887
- Accuracy 0.8319

By class:
              precision    recall  f1-score   support

 Unternehmen     0.8902    0.8835    0.8868       266
 Auslagerung     0.8764    0.9116    0.8937       249
         Ort     0.9635    0.9851    0.9742       134
    Software     0.0000    0.0000    0.0000         0

   micro avg     0.8973    0.9153    0.9062       649
   macro avg     0.6825    0.6950    0.6887       649
weighted avg     0.9000    0.9153    0.9075       649

2024-03-26 12:02:39,079 ----------------------------------------------------------------------------------------------------