Hugging Face's logo

stefan-it
/
flair-co-funer-german_dbmdz_bert_base-bs16-e10-lr5e-05-1

Token Classification
Flair
PyTorch
TensorBoard
German
sequence-tagger-model
hetzner
hetzner-gex44
hetzner-gpu
Model card Files Metrics Community
2
flair-co-funer-german_dbmdz_bert_base-bs16-e10-lr5e-05-1
File size: 26,663 Bytes 
ef98539
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
 
2024-03-26 15:19:10,879 ----------------------------------------------------------------------------------------------------
2024-03-26 15:19:10,880 Model: "SequenceTagger(
  (embeddings): TransformerWordEmbeddings(
    (model): BertModel(
      (embeddings): BertEmbeddings(
        (word_embeddings): Embedding(31103, 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 15:19:10,880 ----------------------------------------------------------------------------------------------------
2024-03-26 15:19:10,880 Corpus: 758 train + 94 dev + 96 test sentences
2024-03-26 15:19:10,880 ----------------------------------------------------------------------------------------------------
2024-03-26 15:19:10,880 Train:  758 sentences
2024-03-26 15:19:10,880         (train_with_dev=False, train_with_test=False)
2024-03-26 15:19:10,880 ----------------------------------------------------------------------------------------------------
2024-03-26 15:19:10,880 Training Params:
2024-03-26 15:19:10,880  - learning_rate: "5e-05" 
2024-03-26 15:19:10,880  - mini_batch_size: "16"
2024-03-26 15:19:10,880  - max_epochs: "10"
2024-03-26 15:19:10,880  - shuffle: "True"
2024-03-26 15:19:10,880 ----------------------------------------------------------------------------------------------------
2024-03-26 15:19:10,880 Plugins:
2024-03-26 15:19:10,880  - TensorboardLogger
2024-03-26 15:19:10,880  - LinearScheduler | warmup_fraction: '0.1'
2024-03-26 15:19:10,880 ----------------------------------------------------------------------------------------------------
2024-03-26 15:19:10,880 Final evaluation on model from best epoch (best-model.pt)
2024-03-26 15:19:10,880  - metric: "('micro avg', 'f1-score')"
2024-03-26 15:19:10,880 ----------------------------------------------------------------------------------------------------
2024-03-26 15:19:10,880 Computation:
2024-03-26 15:19:10,880  - compute on device: cuda:0
2024-03-26 15:19:10,880  - embedding storage: none
2024-03-26 15:19:10,880 ----------------------------------------------------------------------------------------------------
2024-03-26 15:19:10,880 Model training base path: "flair-co-funer-german_dbmdz_bert_base-bs16-e10-lr5e-05-1"
2024-03-26 15:19:10,880 ----------------------------------------------------------------------------------------------------
2024-03-26 15:19:10,880 ----------------------------------------------------------------------------------------------------
2024-03-26 15:19:10,880 Logging anything other than scalars to TensorBoard is currently not supported.
2024-03-26 15:19:12,847 epoch 1 - iter 4/48 - loss 3.18710618 - time (sec): 1.97 - samples/sec: 1380.24 - lr: 0.000003 - momentum: 0.000000
2024-03-26 15:19:14,124 epoch 1 - iter 8/48 - loss 3.09303232 - time (sec): 3.24 - samples/sec: 1661.43 - lr: 0.000007 - momentum: 0.000000
2024-03-26 15:19:17,035 epoch 1 - iter 12/48 - loss 2.92685396 - time (sec): 6.15 - samples/sec: 1413.88 - lr: 0.000011 - momentum: 0.000000
2024-03-26 15:19:20,178 epoch 1 - iter 16/48 - loss 2.77394981 - time (sec): 9.30 - samples/sec: 1311.46 - lr: 0.000016 - momentum: 0.000000
2024-03-26 15:19:22,625 epoch 1 - iter 20/48 - loss 2.60483538 - time (sec): 11.74 - samples/sec: 1309.74 - lr: 0.000020 - momentum: 0.000000
2024-03-26 15:19:24,324 epoch 1 - iter 24/48 - loss 2.45562033 - time (sec): 13.44 - samples/sec: 1356.58 - lr: 0.000024 - momentum: 0.000000
2024-03-26 15:19:25,898 epoch 1 - iter 28/48 - loss 2.33984829 - time (sec): 15.02 - samples/sec: 1378.49 - lr: 0.000028 - momentum: 0.000000
2024-03-26 15:19:27,958 epoch 1 - iter 32/48 - loss 2.22907451 - time (sec): 17.08 - samples/sec: 1384.04 - lr: 0.000032 - momentum: 0.000000
2024-03-26 15:19:28,922 epoch 1 - iter 36/48 - loss 2.13980267 - time (sec): 18.04 - samples/sec: 1443.80 - lr: 0.000036 - momentum: 0.000000
2024-03-26 15:19:30,847 epoch 1 - iter 40/48 - loss 2.03505922 - time (sec): 19.97 - samples/sec: 1458.26 - lr: 0.000041 - momentum: 0.000000
2024-03-26 15:19:32,831 epoch 1 - iter 44/48 - loss 1.93976338 - time (sec): 21.95 - samples/sec: 1443.26 - lr: 0.000045 - momentum: 0.000000
2024-03-26 15:19:34,302 epoch 1 - iter 48/48 - loss 1.83562410 - time (sec): 23.42 - samples/sec: 1471.82 - lr: 0.000049 - momentum: 0.000000
2024-03-26 15:19:34,302 ----------------------------------------------------------------------------------------------------
2024-03-26 15:19:34,302 EPOCH 1 done: loss 1.8356 - lr: 0.000049
2024-03-26 15:19:35,110 DEV : loss 0.5811887979507446 - f1-score (micro avg)  0.5756
2024-03-26 15:19:35,111 saving best model
2024-03-26 15:19:35,406 ----------------------------------------------------------------------------------------------------
2024-03-26 15:19:37,866 epoch 2 - iter 4/48 - loss 0.58486496 - time (sec): 2.46 - samples/sec: 1261.37 - lr: 0.000050 - momentum: 0.000000
2024-03-26 15:19:39,917 epoch 2 - iter 8/48 - loss 0.58476379 - time (sec): 4.51 - samples/sec: 1466.21 - lr: 0.000049 - momentum: 0.000000
2024-03-26 15:19:42,156 epoch 2 - iter 12/48 - loss 0.55560922 - time (sec): 6.75 - samples/sec: 1371.90 - lr: 0.000049 - momentum: 0.000000
2024-03-26 15:19:44,192 epoch 2 - iter 16/48 - loss 0.52680501 - time (sec): 8.78 - samples/sec: 1356.44 - lr: 0.000048 - momentum: 0.000000
2024-03-26 15:19:46,297 epoch 2 - iter 20/48 - loss 0.50334021 - time (sec): 10.89 - samples/sec: 1377.00 - lr: 0.000048 - momentum: 0.000000
2024-03-26 15:19:49,439 epoch 2 - iter 24/48 - loss 0.46270809 - time (sec): 14.03 - samples/sec: 1318.68 - lr: 0.000047 - momentum: 0.000000
2024-03-26 15:19:51,796 epoch 2 - iter 28/48 - loss 0.45349179 - time (sec): 16.39 - samples/sec: 1314.34 - lr: 0.000047 - momentum: 0.000000
2024-03-26 15:19:53,512 epoch 2 - iter 32/48 - loss 0.44049362 - time (sec): 18.11 - samples/sec: 1332.61 - lr: 0.000046 - momentum: 0.000000
2024-03-26 15:19:54,537 epoch 2 - iter 36/48 - loss 0.43314393 - time (sec): 19.13 - samples/sec: 1382.81 - lr: 0.000046 - momentum: 0.000000
2024-03-26 15:19:56,393 epoch 2 - iter 40/48 - loss 0.42530707 - time (sec): 20.99 - samples/sec: 1400.87 - lr: 0.000046 - momentum: 0.000000
2024-03-26 15:19:58,389 epoch 2 - iter 44/48 - loss 0.41740411 - time (sec): 22.98 - samples/sec: 1397.08 - lr: 0.000045 - momentum: 0.000000
2024-03-26 15:19:59,838 epoch 2 - iter 48/48 - loss 0.40956046 - time (sec): 24.43 - samples/sec: 1410.97 - lr: 0.000045 - momentum: 0.000000
2024-03-26 15:19:59,839 ----------------------------------------------------------------------------------------------------
2024-03-26 15:19:59,839 EPOCH 2 done: loss 0.4096 - lr: 0.000045
2024-03-26 15:20:00,755 DEV : loss 0.272553026676178 - f1-score (micro avg)  0.8298
2024-03-26 15:20:00,756 saving best model
2024-03-26 15:20:01,196 ----------------------------------------------------------------------------------------------------
2024-03-26 15:20:03,756 epoch 3 - iter 4/48 - loss 0.31989894 - time (sec): 2.56 - samples/sec: 1192.67 - lr: 0.000044 - momentum: 0.000000
2024-03-26 15:20:05,595 epoch 3 - iter 8/48 - loss 0.27331228 - time (sec): 4.40 - samples/sec: 1333.91 - lr: 0.000044 - momentum: 0.000000
2024-03-26 15:20:07,412 epoch 3 - iter 12/48 - loss 0.27445132 - time (sec): 6.22 - samples/sec: 1413.77 - lr: 0.000043 - momentum: 0.000000
2024-03-26 15:20:09,808 epoch 3 - iter 16/48 - loss 0.24654164 - time (sec): 8.61 - samples/sec: 1418.25 - lr: 0.000043 - momentum: 0.000000
2024-03-26 15:20:11,265 epoch 3 - iter 20/48 - loss 0.25313421 - time (sec): 10.07 - samples/sec: 1469.72 - lr: 0.000042 - momentum: 0.000000
2024-03-26 15:20:14,197 epoch 3 - iter 24/48 - loss 0.23543316 - time (sec): 13.00 - samples/sec: 1454.62 - lr: 0.000042 - momentum: 0.000000
2024-03-26 15:20:14,966 epoch 3 - iter 28/48 - loss 0.22728542 - time (sec): 13.77 - samples/sec: 1528.13 - lr: 0.000041 - momentum: 0.000000
2024-03-26 15:20:17,516 epoch 3 - iter 32/48 - loss 0.21669512 - time (sec): 16.32 - samples/sec: 1472.22 - lr: 0.000041 - momentum: 0.000000
2024-03-26 15:20:19,519 epoch 3 - iter 36/48 - loss 0.20670519 - time (sec): 18.32 - samples/sec: 1465.84 - lr: 0.000040 - momentum: 0.000000
2024-03-26 15:20:21,407 epoch 3 - iter 40/48 - loss 0.20928493 - time (sec): 20.21 - samples/sec: 1455.87 - lr: 0.000040 - momentum: 0.000000
2024-03-26 15:20:23,547 epoch 3 - iter 44/48 - loss 0.20368097 - time (sec): 22.35 - samples/sec: 1459.35 - lr: 0.000040 - momentum: 0.000000
2024-03-26 15:20:24,773 epoch 3 - iter 48/48 - loss 0.20349750 - time (sec): 23.58 - samples/sec: 1462.14 - lr: 0.000039 - momentum: 0.000000
2024-03-26 15:20:24,773 ----------------------------------------------------------------------------------------------------
2024-03-26 15:20:24,774 EPOCH 3 done: loss 0.2035 - lr: 0.000039
2024-03-26 15:20:25,672 DEV : loss 0.23047742247581482 - f1-score (micro avg)  0.8644
2024-03-26 15:20:25,673 saving best model
2024-03-26 15:20:26,144 ----------------------------------------------------------------------------------------------------
2024-03-26 15:20:27,630 epoch 4 - iter 4/48 - loss 0.14469592 - time (sec): 1.49 - samples/sec: 1836.29 - lr: 0.000039 - momentum: 0.000000
2024-03-26 15:20:30,029 epoch 4 - iter 8/48 - loss 0.14637825 - time (sec): 3.88 - samples/sec: 1477.43 - lr: 0.000038 - momentum: 0.000000
2024-03-26 15:20:32,121 epoch 4 - iter 12/48 - loss 0.15299926 - time (sec): 5.98 - samples/sec: 1462.06 - lr: 0.000038 - momentum: 0.000000
2024-03-26 15:20:34,281 epoch 4 - iter 16/48 - loss 0.13823076 - time (sec): 8.14 - samples/sec: 1471.76 - lr: 0.000037 - momentum: 0.000000
2024-03-26 15:20:37,290 epoch 4 - iter 20/48 - loss 0.13133648 - time (sec): 11.14 - samples/sec: 1389.87 - lr: 0.000037 - momentum: 0.000000
2024-03-26 15:20:38,715 epoch 4 - iter 24/48 - loss 0.13273192 - time (sec): 12.57 - samples/sec: 1433.84 - lr: 0.000036 - momentum: 0.000000
2024-03-26 15:20:40,234 epoch 4 - iter 28/48 - loss 0.13029406 - time (sec): 14.09 - samples/sec: 1472.01 - lr: 0.000036 - momentum: 0.000000
2024-03-26 15:20:42,695 epoch 4 - iter 32/48 - loss 0.13574662 - time (sec): 16.55 - samples/sec: 1464.53 - lr: 0.000035 - momentum: 0.000000
2024-03-26 15:20:43,678 epoch 4 - iter 36/48 - loss 0.13547460 - time (sec): 17.53 - samples/sec: 1515.41 - lr: 0.000035 - momentum: 0.000000
2024-03-26 15:20:46,040 epoch 4 - iter 40/48 - loss 0.13133812 - time (sec): 19.90 - samples/sec: 1467.93 - lr: 0.000034 - momentum: 0.000000
2024-03-26 15:20:47,826 epoch 4 - iter 44/48 - loss 0.13151623 - time (sec): 21.68 - samples/sec: 1488.17 - lr: 0.000034 - momentum: 0.000000
2024-03-26 15:20:49,177 epoch 4 - iter 48/48 - loss 0.13003788 - time (sec): 23.03 - samples/sec: 1496.71 - lr: 0.000034 - momentum: 0.000000
2024-03-26 15:20:49,177 ----------------------------------------------------------------------------------------------------
2024-03-26 15:20:49,177 EPOCH 4 done: loss 0.1300 - lr: 0.000034
2024-03-26 15:20:50,080 DEV : loss 0.19684486091136932 - f1-score (micro avg)  0.8822
2024-03-26 15:20:50,081 saving best model
2024-03-26 15:20:50,536 ----------------------------------------------------------------------------------------------------
2024-03-26 15:20:52,431 epoch 5 - iter 4/48 - loss 0.11101026 - time (sec): 1.89 - samples/sec: 1472.44 - lr: 0.000033 - momentum: 0.000000
2024-03-26 15:20:54,844 epoch 5 - iter 8/48 - loss 0.09405511 - time (sec): 4.31 - samples/sec: 1378.35 - lr: 0.000033 - momentum: 0.000000
2024-03-26 15:20:56,780 epoch 5 - iter 12/48 - loss 0.09868726 - time (sec): 6.24 - samples/sec: 1372.16 - lr: 0.000032 - momentum: 0.000000
2024-03-26 15:20:58,770 epoch 5 - iter 16/48 - loss 0.09987471 - time (sec): 8.23 - samples/sec: 1403.14 - lr: 0.000032 - momentum: 0.000000
2024-03-26 15:21:00,667 epoch 5 - iter 20/48 - loss 0.10225445 - time (sec): 10.13 - samples/sec: 1412.64 - lr: 0.000031 - momentum: 0.000000
2024-03-26 15:21:02,174 epoch 5 - iter 24/48 - loss 0.10614739 - time (sec): 11.64 - samples/sec: 1461.65 - lr: 0.000031 - momentum: 0.000000
2024-03-26 15:21:04,347 epoch 5 - iter 28/48 - loss 0.10649413 - time (sec): 13.81 - samples/sec: 1458.67 - lr: 0.000030 - momentum: 0.000000
2024-03-26 15:21:06,936 epoch 5 - iter 32/48 - loss 0.10514418 - time (sec): 16.40 - samples/sec: 1443.49 - lr: 0.000030 - momentum: 0.000000
2024-03-26 15:21:09,285 epoch 5 - iter 36/48 - loss 0.09964481 - time (sec): 18.75 - samples/sec: 1447.45 - lr: 0.000029 - momentum: 0.000000
2024-03-26 15:21:10,168 epoch 5 - iter 40/48 - loss 0.10020545 - time (sec): 19.63 - samples/sec: 1490.45 - lr: 0.000029 - momentum: 0.000000
2024-03-26 15:21:12,771 epoch 5 - iter 44/48 - loss 0.09600844 - time (sec): 22.23 - samples/sec: 1456.49 - lr: 0.000029 - momentum: 0.000000
2024-03-26 15:21:14,236 epoch 5 - iter 48/48 - loss 0.09625165 - time (sec): 23.70 - samples/sec: 1454.55 - lr: 0.000028 - momentum: 0.000000
2024-03-26 15:21:14,236 ----------------------------------------------------------------------------------------------------
2024-03-26 15:21:14,237 EPOCH 5 done: loss 0.0963 - lr: 0.000028
2024-03-26 15:21:15,149 DEV : loss 0.18545496463775635 - f1-score (micro avg)  0.9018
2024-03-26 15:21:15,151 saving best model
2024-03-26 15:21:15,614 ----------------------------------------------------------------------------------------------------
2024-03-26 15:21:17,602 epoch 6 - iter 4/48 - loss 0.06257767 - time (sec): 1.99 - samples/sec: 1330.90 - lr: 0.000028 - momentum: 0.000000
2024-03-26 15:21:19,694 epoch 6 - iter 8/48 - loss 0.08686096 - time (sec): 4.08 - samples/sec: 1356.29 - lr: 0.000027 - momentum: 0.000000
2024-03-26 15:21:21,500 epoch 6 - iter 12/48 - loss 0.08442024 - time (sec): 5.88 - samples/sec: 1468.53 - lr: 0.000027 - momentum: 0.000000
2024-03-26 15:21:23,710 epoch 6 - iter 16/48 - loss 0.08112975 - time (sec): 8.09 - samples/sec: 1419.29 - lr: 0.000026 - momentum: 0.000000
2024-03-26 15:21:25,455 epoch 6 - iter 20/48 - loss 0.08423255 - time (sec): 9.84 - samples/sec: 1426.20 - lr: 0.000026 - momentum: 0.000000
2024-03-26 15:21:27,886 epoch 6 - iter 24/48 - loss 0.07972403 - time (sec): 12.27 - samples/sec: 1403.28 - lr: 0.000025 - momentum: 0.000000
2024-03-26 15:21:29,707 epoch 6 - iter 28/48 - loss 0.08176774 - time (sec): 14.09 - samples/sec: 1403.53 - lr: 0.000025 - momentum: 0.000000
2024-03-26 15:21:32,151 epoch 6 - iter 32/48 - loss 0.08182395 - time (sec): 16.54 - samples/sec: 1382.24 - lr: 0.000024 - momentum: 0.000000
2024-03-26 15:21:35,535 epoch 6 - iter 36/48 - loss 0.07761302 - time (sec): 19.92 - samples/sec: 1338.70 - lr: 0.000024 - momentum: 0.000000
2024-03-26 15:21:37,124 epoch 6 - iter 40/48 - loss 0.07583205 - time (sec): 21.51 - samples/sec: 1373.90 - lr: 0.000023 - momentum: 0.000000
2024-03-26 15:21:38,926 epoch 6 - iter 44/48 - loss 0.07472669 - time (sec): 23.31 - samples/sec: 1377.69 - lr: 0.000023 - momentum: 0.000000
2024-03-26 15:21:40,201 epoch 6 - iter 48/48 - loss 0.07681090 - time (sec): 24.59 - samples/sec: 1402.13 - lr: 0.000023 - momentum: 0.000000
2024-03-26 15:21:40,201 ----------------------------------------------------------------------------------------------------
2024-03-26 15:21:40,201 EPOCH 6 done: loss 0.0768 - lr: 0.000023
2024-03-26 15:21:41,112 DEV : loss 0.18713432550430298 - f1-score (micro avg)  0.9134
2024-03-26 15:21:41,114 saving best model
2024-03-26 15:21:41,539 ----------------------------------------------------------------------------------------------------
2024-03-26 15:21:43,160 epoch 7 - iter 4/48 - loss 0.09293975 - time (sec): 1.62 - samples/sec: 1696.85 - lr: 0.000022 - momentum: 0.000000
2024-03-26 15:21:45,287 epoch 7 - iter 8/48 - loss 0.07180060 - time (sec): 3.75 - samples/sec: 1435.30 - lr: 0.000022 - momentum: 0.000000
2024-03-26 15:21:47,573 epoch 7 - iter 12/48 - loss 0.07251455 - time (sec): 6.03 - samples/sec: 1376.28 - lr: 0.000021 - momentum: 0.000000
2024-03-26 15:21:50,141 epoch 7 - iter 16/48 - loss 0.06294037 - time (sec): 8.60 - samples/sec: 1342.06 - lr: 0.000021 - momentum: 0.000000
2024-03-26 15:21:52,401 epoch 7 - iter 20/48 - loss 0.06103248 - time (sec): 10.86 - samples/sec: 1345.79 - lr: 0.000020 - momentum: 0.000000
2024-03-26 15:21:53,748 epoch 7 - iter 24/48 - loss 0.05931107 - time (sec): 12.21 - samples/sec: 1401.44 - lr: 0.000020 - momentum: 0.000000
2024-03-26 15:21:55,138 epoch 7 - iter 28/48 - loss 0.05692812 - time (sec): 13.60 - samples/sec: 1466.97 - lr: 0.000019 - momentum: 0.000000
2024-03-26 15:21:57,107 epoch 7 - iter 32/48 - loss 0.05413519 - time (sec): 15.57 - samples/sec: 1457.06 - lr: 0.000019 - momentum: 0.000000
2024-03-26 15:21:59,243 epoch 7 - iter 36/48 - loss 0.05205198 - time (sec): 17.70 - samples/sec: 1446.84 - lr: 0.000018 - momentum: 0.000000
2024-03-26 15:22:01,686 epoch 7 - iter 40/48 - loss 0.05290442 - time (sec): 20.15 - samples/sec: 1426.73 - lr: 0.000018 - momentum: 0.000000
2024-03-26 15:22:03,506 epoch 7 - iter 44/48 - loss 0.05308150 - time (sec): 21.97 - samples/sec: 1443.20 - lr: 0.000017 - momentum: 0.000000
2024-03-26 15:22:05,405 epoch 7 - iter 48/48 - loss 0.05261033 - time (sec): 23.86 - samples/sec: 1444.48 - lr: 0.000017 - momentum: 0.000000
2024-03-26 15:22:05,405 ----------------------------------------------------------------------------------------------------
2024-03-26 15:22:05,405 EPOCH 7 done: loss 0.0526 - lr: 0.000017
2024-03-26 15:22:06,353 DEV : loss 0.1793615221977234 - f1-score (micro avg)  0.911
2024-03-26 15:22:06,355 ----------------------------------------------------------------------------------------------------
2024-03-26 15:22:08,300 epoch 8 - iter 4/48 - loss 0.04398425 - time (sec): 1.94 - samples/sec: 1390.21 - lr: 0.000017 - momentum: 0.000000
2024-03-26 15:22:11,067 epoch 8 - iter 8/48 - loss 0.04003061 - time (sec): 4.71 - samples/sec: 1179.46 - lr: 0.000016 - momentum: 0.000000
2024-03-26 15:22:12,332 epoch 8 - iter 12/48 - loss 0.04018561 - time (sec): 5.98 - samples/sec: 1335.91 - lr: 0.000016 - momentum: 0.000000
2024-03-26 15:22:14,718 epoch 8 - iter 16/48 - loss 0.04398152 - time (sec): 8.36 - samples/sec: 1346.18 - lr: 0.000015 - momentum: 0.000000
2024-03-26 15:22:17,196 epoch 8 - iter 20/48 - loss 0.03752477 - time (sec): 10.84 - samples/sec: 1390.07 - lr: 0.000015 - momentum: 0.000000
2024-03-26 15:22:18,470 epoch 8 - iter 24/48 - loss 0.03831492 - time (sec): 12.11 - samples/sec: 1469.13 - lr: 0.000014 - momentum: 0.000000
2024-03-26 15:22:21,703 epoch 8 - iter 28/48 - loss 0.03971932 - time (sec): 15.35 - samples/sec: 1421.87 - lr: 0.000014 - momentum: 0.000000
2024-03-26 15:22:23,681 epoch 8 - iter 32/48 - loss 0.04079558 - time (sec): 17.32 - samples/sec: 1423.06 - lr: 0.000013 - momentum: 0.000000
2024-03-26 15:22:24,735 epoch 8 - iter 36/48 - loss 0.04136868 - time (sec): 18.38 - samples/sec: 1460.99 - lr: 0.000013 - momentum: 0.000000
2024-03-26 15:22:26,389 epoch 8 - iter 40/48 - loss 0.04156890 - time (sec): 20.03 - samples/sec: 1458.98 - lr: 0.000012 - momentum: 0.000000
2024-03-26 15:22:27,955 epoch 8 - iter 44/48 - loss 0.04044990 - time (sec): 21.60 - samples/sec: 1479.33 - lr: 0.000012 - momentum: 0.000000
2024-03-26 15:22:29,881 epoch 8 - iter 48/48 - loss 0.04185402 - time (sec): 23.53 - samples/sec: 1465.33 - lr: 0.000011 - momentum: 0.000000
2024-03-26 15:22:29,881 ----------------------------------------------------------------------------------------------------
2024-03-26 15:22:29,881 EPOCH 8 done: loss 0.0419 - lr: 0.000011
2024-03-26 15:22:30,781 DEV : loss 0.1833573579788208 - f1-score (micro avg)  0.9202
2024-03-26 15:22:30,782 saving best model
2024-03-26 15:22:31,220 ----------------------------------------------------------------------------------------------------
2024-03-26 15:22:33,067 epoch 9 - iter 4/48 - loss 0.01983911 - time (sec): 1.84 - samples/sec: 1452.18 - lr: 0.000011 - momentum: 0.000000
2024-03-26 15:22:36,225 epoch 9 - iter 8/48 - loss 0.01479519 - time (sec): 5.00 - samples/sec: 1248.34 - lr: 0.000011 - momentum: 0.000000
2024-03-26 15:22:37,878 epoch 9 - iter 12/48 - loss 0.02269400 - time (sec): 6.66 - samples/sec: 1305.81 - lr: 0.000010 - momentum: 0.000000
2024-03-26 15:22:40,119 epoch 9 - iter 16/48 - loss 0.02357261 - time (sec): 8.90 - samples/sec: 1295.05 - lr: 0.000010 - momentum: 0.000000
2024-03-26 15:22:42,385 epoch 9 - iter 20/48 - loss 0.02814877 - time (sec): 11.16 - samples/sec: 1325.41 - lr: 0.000009 - momentum: 0.000000
2024-03-26 15:22:44,548 epoch 9 - iter 24/48 - loss 0.02932997 - time (sec): 13.33 - samples/sec: 1342.11 - lr: 0.000009 - momentum: 0.000000
2024-03-26 15:22:46,915 epoch 9 - iter 28/48 - loss 0.02799787 - time (sec): 15.69 - samples/sec: 1335.53 - lr: 0.000008 - momentum: 0.000000
2024-03-26 15:22:49,248 epoch 9 - iter 32/48 - loss 0.02763381 - time (sec): 18.03 - samples/sec: 1332.11 - lr: 0.000008 - momentum: 0.000000
2024-03-26 15:22:51,038 epoch 9 - iter 36/48 - loss 0.03042036 - time (sec): 19.82 - samples/sec: 1351.06 - lr: 0.000007 - momentum: 0.000000
2024-03-26 15:22:53,218 epoch 9 - iter 40/48 - loss 0.03212191 - time (sec): 22.00 - samples/sec: 1340.54 - lr: 0.000007 - momentum: 0.000000
2024-03-26 15:22:55,349 epoch 9 - iter 44/48 - loss 0.03178939 - time (sec): 24.13 - samples/sec: 1351.29 - lr: 0.000006 - momentum: 0.000000
2024-03-26 15:22:56,102 epoch 9 - iter 48/48 - loss 0.03247208 - time (sec): 24.88 - samples/sec: 1385.53 - lr: 0.000006 - momentum: 0.000000
2024-03-26 15:22:56,102 ----------------------------------------------------------------------------------------------------
2024-03-26 15:22:56,103 EPOCH 9 done: loss 0.0325 - lr: 0.000006
2024-03-26 15:22:57,024 DEV : loss 0.18409405648708344 - f1-score (micro avg)  0.922
2024-03-26 15:22:57,025 saving best model
2024-03-26 15:22:57,468 ----------------------------------------------------------------------------------------------------
2024-03-26 15:22:59,218 epoch 10 - iter 4/48 - loss 0.02465374 - time (sec): 1.75 - samples/sec: 1502.78 - lr: 0.000006 - momentum: 0.000000
2024-03-26 15:23:01,162 epoch 10 - iter 8/48 - loss 0.02043026 - time (sec): 3.69 - samples/sec: 1500.48 - lr: 0.000005 - momentum: 0.000000
2024-03-26 15:23:03,770 epoch 10 - iter 12/48 - loss 0.02341113 - time (sec): 6.30 - samples/sec: 1385.00 - lr: 0.000005 - momentum: 0.000000
2024-03-26 15:23:05,677 epoch 10 - iter 16/48 - loss 0.02494848 - time (sec): 8.21 - samples/sec: 1397.68 - lr: 0.000004 - momentum: 0.000000
2024-03-26 15:23:07,497 epoch 10 - iter 20/48 - loss 0.02414231 - time (sec): 10.03 - samples/sec: 1442.51 - lr: 0.000004 - momentum: 0.000000
2024-03-26 15:23:09,127 epoch 10 - iter 24/48 - loss 0.02884870 - time (sec): 11.66 - samples/sec: 1454.73 - lr: 0.000003 - momentum: 0.000000
2024-03-26 15:23:10,862 epoch 10 - iter 28/48 - loss 0.02687785 - time (sec): 13.39 - samples/sec: 1477.39 - lr: 0.000003 - momentum: 0.000000
2024-03-26 15:23:12,055 epoch 10 - iter 32/48 - loss 0.02669034 - time (sec): 14.59 - samples/sec: 1509.99 - lr: 0.000002 - momentum: 0.000000
2024-03-26 15:23:15,009 epoch 10 - iter 36/48 - loss 0.02386780 - time (sec): 17.54 - samples/sec: 1460.35 - lr: 0.000002 - momentum: 0.000000
2024-03-26 15:23:17,775 epoch 10 - iter 40/48 - loss 0.02684189 - time (sec): 20.31 - samples/sec: 1432.15 - lr: 0.000001 - momentum: 0.000000
2024-03-26 15:23:20,553 epoch 10 - iter 44/48 - loss 0.02557963 - time (sec): 23.08 - samples/sec: 1398.52 - lr: 0.000001 - momentum: 0.000000
2024-03-26 15:23:22,144 epoch 10 - iter 48/48 - loss 0.02482446 - time (sec): 24.67 - samples/sec: 1397.05 - lr: 0.000000 - momentum: 0.000000
2024-03-26 15:23:22,144 ----------------------------------------------------------------------------------------------------
2024-03-26 15:23:22,144 EPOCH 10 done: loss 0.0248 - lr: 0.000000
2024-03-26 15:23:23,081 DEV : loss 0.183772012591362 - f1-score (micro avg)  0.927
2024-03-26 15:23:23,083 saving best model
2024-03-26 15:23:23,957 ----------------------------------------------------------------------------------------------------
2024-03-26 15:23:23,957 Loading model from best epoch ...
2024-03-26 15:23:24,860 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 15:23:25,631 
Results:
- F-score (micro) 0.9052
- F-score (macro) 0.689
- Accuracy 0.8303

By class:
              precision    recall  f1-score   support

 Unternehmen     0.8902    0.8835    0.8868       266
 Auslagerung     0.8577    0.9197    0.8876       249
         Ort     0.9708    0.9925    0.9815       134
    Software     0.0000    0.0000    0.0000         0

   micro avg     0.8910    0.9199    0.9052       649
   macro avg     0.6797    0.6989    0.6890       649
weighted avg     0.8943    0.9199    0.9067       649

2024-03-26 15:23:25,631 ----------------------------------------------------------------------------------------------------