Hugging Face's logo

stefan-it
/
flair-co-funer-german_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_bert_base-bs16-e10-lr5e-05-1
File size: 26,663 Bytes 
95c5527
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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 10:59:50,722 ----------------------------------------------------------------------------------------------------
2024-03-26 10:59:50,722 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 10:59:50,722 ----------------------------------------------------------------------------------------------------
2024-03-26 10:59:50,722 Corpus: 758 train + 94 dev + 96 test sentences
2024-03-26 10:59:50,722 ----------------------------------------------------------------------------------------------------
2024-03-26 10:59:50,722 Train:  758 sentences
2024-03-26 10:59:50,722         (train_with_dev=False, train_with_test=False)
2024-03-26 10:59:50,722 ----------------------------------------------------------------------------------------------------
2024-03-26 10:59:50,722 Training Params:
2024-03-26 10:59:50,722  - learning_rate: "5e-05" 
2024-03-26 10:59:50,722  - mini_batch_size: "16"
2024-03-26 10:59:50,722  - max_epochs: "10"
2024-03-26 10:59:50,722  - shuffle: "True"
2024-03-26 10:59:50,722 ----------------------------------------------------------------------------------------------------
2024-03-26 10:59:50,722 Plugins:
2024-03-26 10:59:50,722  - TensorboardLogger
2024-03-26 10:59:50,722  - LinearScheduler | warmup_fraction: '0.1'
2024-03-26 10:59:50,723 ----------------------------------------------------------------------------------------------------
2024-03-26 10:59:50,723 Final evaluation on model from best epoch (best-model.pt)
2024-03-26 10:59:50,723  - metric: "('micro avg', 'f1-score')"
2024-03-26 10:59:50,723 ----------------------------------------------------------------------------------------------------
2024-03-26 10:59:50,723 Computation:
2024-03-26 10:59:50,723  - compute on device: cuda:0
2024-03-26 10:59:50,723  - embedding storage: none
2024-03-26 10:59:50,723 ----------------------------------------------------------------------------------------------------
2024-03-26 10:59:50,723 Model training base path: "flair-co-funer-german_bert_base-bs16-e10-lr5e-05-1"
2024-03-26 10:59:50,723 ----------------------------------------------------------------------------------------------------
2024-03-26 10:59:50,723 ----------------------------------------------------------------------------------------------------
2024-03-26 10:59:50,723 Logging anything other than scalars to TensorBoard is currently not supported.
2024-03-26 10:59:52,827 epoch 1 - iter 4/48 - loss 3.17719565 - time (sec): 2.10 - samples/sec: 1290.23 - lr: 0.000003 - momentum: 0.000000
2024-03-26 10:59:54,174 epoch 1 - iter 8/48 - loss 3.22940502 - time (sec): 3.45 - samples/sec: 1561.49 - lr: 0.000007 - momentum: 0.000000
2024-03-26 10:59:57,198 epoch 1 - iter 12/48 - loss 3.11554189 - time (sec): 6.48 - samples/sec: 1343.86 - lr: 0.000011 - momentum: 0.000000
2024-03-26 11:00:00,546 epoch 1 - iter 16/48 - loss 2.96962598 - time (sec): 9.82 - samples/sec: 1241.33 - lr: 0.000016 - momentum: 0.000000
2024-03-26 11:00:03,131 epoch 1 - iter 20/48 - loss 2.80109533 - time (sec): 12.41 - samples/sec: 1239.73 - lr: 0.000020 - momentum: 0.000000
2024-03-26 11:00:04,908 epoch 1 - iter 24/48 - loss 2.64646526 - time (sec): 14.18 - samples/sec: 1285.69 - lr: 0.000024 - momentum: 0.000000
2024-03-26 11:00:06,561 epoch 1 - iter 28/48 - loss 2.52516425 - time (sec): 15.84 - samples/sec: 1307.09 - lr: 0.000028 - momentum: 0.000000
2024-03-26 11:00:08,744 epoch 1 - iter 32/48 - loss 2.41186996 - time (sec): 18.02 - samples/sec: 1311.61 - lr: 0.000032 - momentum: 0.000000
2024-03-26 11:00:09,739 epoch 1 - iter 36/48 - loss 2.32081588 - time (sec): 19.02 - samples/sec: 1369.83 - lr: 0.000036 - momentum: 0.000000
2024-03-26 11:00:11,786 epoch 1 - iter 40/48 - loss 2.21383836 - time (sec): 21.06 - samples/sec: 1382.32 - lr: 0.000041 - momentum: 0.000000
2024-03-26 11:00:13,867 epoch 1 - iter 44/48 - loss 2.11699490 - time (sec): 23.14 - samples/sec: 1368.86 - lr: 0.000045 - momentum: 0.000000
2024-03-26 11:00:15,419 epoch 1 - iter 48/48 - loss 2.01053611 - time (sec): 24.70 - samples/sec: 1395.88 - lr: 0.000049 - momentum: 0.000000
2024-03-26 11:00:15,419 ----------------------------------------------------------------------------------------------------
2024-03-26 11:00:15,419 EPOCH 1 done: loss 2.0105 - lr: 0.000049
2024-03-26 11:00:16,279 DEV : loss 0.6016283631324768 - f1-score (micro avg)  0.5787
2024-03-26 11:00:16,281 saving best model
2024-03-26 11:00:16,555 ----------------------------------------------------------------------------------------------------
2024-03-26 11:00:19,245 epoch 2 - iter 4/48 - loss 0.70959951 - time (sec): 2.69 - samples/sec: 1153.63 - lr: 0.000050 - momentum: 0.000000
2024-03-26 11:00:21,418 epoch 2 - iter 8/48 - loss 0.70407287 - time (sec): 4.86 - samples/sec: 1360.00 - lr: 0.000049 - momentum: 0.000000
2024-03-26 11:00:23,773 epoch 2 - iter 12/48 - loss 0.65597292 - time (sec): 7.22 - samples/sec: 1282.88 - lr: 0.000049 - momentum: 0.000000
2024-03-26 11:00:25,873 epoch 2 - iter 16/48 - loss 0.61448768 - time (sec): 9.32 - samples/sec: 1278.96 - lr: 0.000048 - momentum: 0.000000
2024-03-26 11:00:28,044 epoch 2 - iter 20/48 - loss 0.57629162 - time (sec): 11.49 - samples/sec: 1305.28 - lr: 0.000048 - momentum: 0.000000
2024-03-26 11:00:31,219 epoch 2 - iter 24/48 - loss 0.52019540 - time (sec): 14.66 - samples/sec: 1261.90 - lr: 0.000047 - momentum: 0.000000
2024-03-26 11:00:33,666 epoch 2 - iter 28/48 - loss 0.50568930 - time (sec): 17.11 - samples/sec: 1258.99 - lr: 0.000047 - momentum: 0.000000
2024-03-26 11:00:35,456 epoch 2 - iter 32/48 - loss 0.48837358 - time (sec): 18.90 - samples/sec: 1276.55 - lr: 0.000046 - momentum: 0.000000
2024-03-26 11:00:36,527 epoch 2 - iter 36/48 - loss 0.47541041 - time (sec): 19.97 - samples/sec: 1324.58 - lr: 0.000046 - momentum: 0.000000
2024-03-26 11:00:38,421 epoch 2 - iter 40/48 - loss 0.46034741 - time (sec): 21.87 - samples/sec: 1344.51 - lr: 0.000046 - momentum: 0.000000
2024-03-26 11:00:40,494 epoch 2 - iter 44/48 - loss 0.45159252 - time (sec): 23.94 - samples/sec: 1341.26 - lr: 0.000045 - momentum: 0.000000
2024-03-26 11:00:41,950 epoch 2 - iter 48/48 - loss 0.44143919 - time (sec): 25.39 - samples/sec: 1357.48 - lr: 0.000045 - momentum: 0.000000
2024-03-26 11:00:41,950 ----------------------------------------------------------------------------------------------------
2024-03-26 11:00:41,950 EPOCH 2 done: loss 0.4414 - lr: 0.000045
2024-03-26 11:00:42,973 DEV : loss 0.3147665560245514 - f1-score (micro avg)  0.8072
2024-03-26 11:00:42,974 saving best model
2024-03-26 11:00:43,402 ----------------------------------------------------------------------------------------------------
2024-03-26 11:00:45,991 epoch 3 - iter 4/48 - loss 0.28698034 - time (sec): 2.59 - samples/sec: 1179.06 - lr: 0.000044 - momentum: 0.000000
2024-03-26 11:00:47,895 epoch 3 - iter 8/48 - loss 0.25658164 - time (sec): 4.49 - samples/sec: 1305.97 - lr: 0.000044 - momentum: 0.000000
2024-03-26 11:00:49,761 epoch 3 - iter 12/48 - loss 0.26858709 - time (sec): 6.36 - samples/sec: 1381.81 - lr: 0.000043 - momentum: 0.000000
2024-03-26 11:00:52,180 epoch 3 - iter 16/48 - loss 0.24605106 - time (sec): 8.78 - samples/sec: 1391.39 - lr: 0.000043 - momentum: 0.000000
2024-03-26 11:00:53,687 epoch 3 - iter 20/48 - loss 0.25301403 - time (sec): 10.29 - samples/sec: 1438.66 - lr: 0.000042 - momentum: 0.000000
2024-03-26 11:00:56,688 epoch 3 - iter 24/48 - loss 0.24057070 - time (sec): 13.29 - samples/sec: 1423.43 - lr: 0.000042 - momentum: 0.000000
2024-03-26 11:00:57,456 epoch 3 - iter 28/48 - loss 0.22678079 - time (sec): 14.05 - samples/sec: 1497.16 - lr: 0.000041 - momentum: 0.000000
2024-03-26 11:01:00,172 epoch 3 - iter 32/48 - loss 0.22006429 - time (sec): 16.77 - samples/sec: 1432.64 - lr: 0.000041 - momentum: 0.000000
2024-03-26 11:01:02,238 epoch 3 - iter 36/48 - loss 0.20958441 - time (sec): 18.84 - samples/sec: 1425.88 - lr: 0.000040 - momentum: 0.000000
2024-03-26 11:01:04,210 epoch 3 - iter 40/48 - loss 0.20985973 - time (sec): 20.81 - samples/sec: 1414.06 - lr: 0.000040 - momentum: 0.000000
2024-03-26 11:01:06,442 epoch 3 - iter 44/48 - loss 0.20453707 - time (sec): 23.04 - samples/sec: 1415.64 - lr: 0.000040 - momentum: 0.000000
2024-03-26 11:01:07,725 epoch 3 - iter 48/48 - loss 0.20307138 - time (sec): 24.32 - samples/sec: 1417.29 - lr: 0.000039 - momentum: 0.000000
2024-03-26 11:01:07,725 ----------------------------------------------------------------------------------------------------
2024-03-26 11:01:07,725 EPOCH 3 done: loss 0.2031 - lr: 0.000039
2024-03-26 11:01:08,671 DEV : loss 0.25678685307502747 - f1-score (micro avg)  0.8519
2024-03-26 11:01:08,672 saving best model
2024-03-26 11:01:09,109 ----------------------------------------------------------------------------------------------------
2024-03-26 11:01:10,611 epoch 4 - iter 4/48 - loss 0.12080968 - time (sec): 1.50 - samples/sec: 1816.16 - lr: 0.000039 - momentum: 0.000000
2024-03-26 11:01:13,125 epoch 4 - iter 8/48 - loss 0.15034122 - time (sec): 4.01 - samples/sec: 1429.16 - lr: 0.000038 - momentum: 0.000000
2024-03-26 11:01:15,244 epoch 4 - iter 12/48 - loss 0.16078746 - time (sec): 6.13 - samples/sec: 1424.34 - lr: 0.000038 - momentum: 0.000000
2024-03-26 11:01:17,467 epoch 4 - iter 16/48 - loss 0.14298013 - time (sec): 8.36 - samples/sec: 1432.75 - lr: 0.000037 - momentum: 0.000000
2024-03-26 11:01:20,491 epoch 4 - iter 20/48 - loss 0.13519036 - time (sec): 11.38 - samples/sec: 1360.97 - lr: 0.000037 - momentum: 0.000000
2024-03-26 11:01:21,975 epoch 4 - iter 24/48 - loss 0.13663630 - time (sec): 12.87 - samples/sec: 1400.91 - lr: 0.000036 - momentum: 0.000000
2024-03-26 11:01:23,514 epoch 4 - iter 28/48 - loss 0.13658070 - time (sec): 14.40 - samples/sec: 1439.83 - lr: 0.000036 - momentum: 0.000000
2024-03-26 11:01:26,144 epoch 4 - iter 32/48 - loss 0.13666749 - time (sec): 17.03 - samples/sec: 1422.91 - lr: 0.000035 - momentum: 0.000000
2024-03-26 11:01:27,183 epoch 4 - iter 36/48 - loss 0.13693895 - time (sec): 18.07 - samples/sec: 1470.16 - lr: 0.000035 - momentum: 0.000000
2024-03-26 11:01:29,617 epoch 4 - iter 40/48 - loss 0.13278247 - time (sec): 20.51 - samples/sec: 1424.10 - lr: 0.000034 - momentum: 0.000000
2024-03-26 11:01:31,480 epoch 4 - iter 44/48 - loss 0.13260496 - time (sec): 22.37 - samples/sec: 1442.33 - lr: 0.000034 - momentum: 0.000000
2024-03-26 11:01:32,883 epoch 4 - iter 48/48 - loss 0.13127512 - time (sec): 23.77 - samples/sec: 1450.02 - lr: 0.000034 - momentum: 0.000000
2024-03-26 11:01:32,884 ----------------------------------------------------------------------------------------------------
2024-03-26 11:01:32,884 EPOCH 4 done: loss 0.1313 - lr: 0.000034
2024-03-26 11:01:33,832 DEV : loss 0.21583451330661774 - f1-score (micro avg)  0.8756
2024-03-26 11:01:33,833 saving best model
2024-03-26 11:01:34,279 ----------------------------------------------------------------------------------------------------
2024-03-26 11:01:36,221 epoch 5 - iter 4/48 - loss 0.10870825 - time (sec): 1.94 - samples/sec: 1437.93 - lr: 0.000033 - momentum: 0.000000
2024-03-26 11:01:38,764 epoch 5 - iter 8/48 - loss 0.09524973 - time (sec): 4.48 - samples/sec: 1324.28 - lr: 0.000033 - momentum: 0.000000
2024-03-26 11:01:40,833 epoch 5 - iter 12/48 - loss 0.09712497 - time (sec): 6.55 - samples/sec: 1307.49 - lr: 0.000032 - momentum: 0.000000
2024-03-26 11:01:42,897 epoch 5 - iter 16/48 - loss 0.09416056 - time (sec): 8.62 - samples/sec: 1340.70 - lr: 0.000032 - momentum: 0.000000
2024-03-26 11:01:44,870 epoch 5 - iter 20/48 - loss 0.09859046 - time (sec): 10.59 - samples/sec: 1351.47 - lr: 0.000031 - momentum: 0.000000
2024-03-26 11:01:46,427 epoch 5 - iter 24/48 - loss 0.10256071 - time (sec): 12.15 - samples/sec: 1400.37 - lr: 0.000031 - momentum: 0.000000
2024-03-26 11:01:48,743 epoch 5 - iter 28/48 - loss 0.10312729 - time (sec): 14.46 - samples/sec: 1392.86 - lr: 0.000030 - momentum: 0.000000
2024-03-26 11:01:51,403 epoch 5 - iter 32/48 - loss 0.10114194 - time (sec): 17.12 - samples/sec: 1382.56 - lr: 0.000030 - momentum: 0.000000
2024-03-26 11:01:53,777 epoch 5 - iter 36/48 - loss 0.09523322 - time (sec): 19.50 - samples/sec: 1391.93 - lr: 0.000029 - momentum: 0.000000
2024-03-26 11:01:54,702 epoch 5 - iter 40/48 - loss 0.09494868 - time (sec): 20.42 - samples/sec: 1432.80 - lr: 0.000029 - momentum: 0.000000
2024-03-26 11:01:57,363 epoch 5 - iter 44/48 - loss 0.09114674 - time (sec): 23.08 - samples/sec: 1402.97 - lr: 0.000029 - momentum: 0.000000
2024-03-26 11:01:58,852 epoch 5 - iter 48/48 - loss 0.09130317 - time (sec): 24.57 - samples/sec: 1402.93 - lr: 0.000028 - momentum: 0.000000
2024-03-26 11:01:58,853 ----------------------------------------------------------------------------------------------------
2024-03-26 11:01:58,853 EPOCH 5 done: loss 0.0913 - lr: 0.000028
2024-03-26 11:01:59,809 DEV : loss 0.19779938459396362 - f1-score (micro avg)  0.8914
2024-03-26 11:01:59,812 saving best model
2024-03-26 11:02:00,262 ----------------------------------------------------------------------------------------------------
2024-03-26 11:02:02,268 epoch 6 - iter 4/48 - loss 0.05285658 - time (sec): 2.01 - samples/sec: 1318.38 - lr: 0.000028 - momentum: 0.000000
2024-03-26 11:02:04,480 epoch 6 - iter 8/48 - loss 0.07094565 - time (sec): 4.22 - samples/sec: 1311.76 - lr: 0.000027 - momentum: 0.000000
2024-03-26 11:02:06,334 epoch 6 - iter 12/48 - loss 0.07133832 - time (sec): 6.07 - samples/sec: 1423.40 - lr: 0.000027 - momentum: 0.000000
2024-03-26 11:02:08,617 epoch 6 - iter 16/48 - loss 0.07133402 - time (sec): 8.35 - samples/sec: 1375.26 - lr: 0.000026 - momentum: 0.000000
2024-03-26 11:02:10,436 epoch 6 - iter 20/48 - loss 0.07725707 - time (sec): 10.17 - samples/sec: 1379.46 - lr: 0.000026 - momentum: 0.000000
2024-03-26 11:02:12,944 epoch 6 - iter 24/48 - loss 0.07565556 - time (sec): 12.68 - samples/sec: 1357.85 - lr: 0.000025 - momentum: 0.000000
2024-03-26 11:02:14,899 epoch 6 - iter 28/48 - loss 0.07510084 - time (sec): 14.64 - samples/sec: 1351.33 - lr: 0.000025 - momentum: 0.000000
2024-03-26 11:02:17,426 epoch 6 - iter 32/48 - loss 0.07424382 - time (sec): 17.16 - samples/sec: 1331.71 - lr: 0.000024 - momentum: 0.000000
2024-03-26 11:02:20,980 epoch 6 - iter 36/48 - loss 0.06879978 - time (sec): 20.72 - samples/sec: 1287.17 - lr: 0.000024 - momentum: 0.000000
2024-03-26 11:02:22,685 epoch 6 - iter 40/48 - loss 0.06691408 - time (sec): 22.42 - samples/sec: 1317.93 - lr: 0.000023 - momentum: 0.000000
2024-03-26 11:02:24,577 epoch 6 - iter 44/48 - loss 0.06499311 - time (sec): 24.31 - samples/sec: 1320.84 - lr: 0.000023 - momentum: 0.000000
2024-03-26 11:02:25,873 epoch 6 - iter 48/48 - loss 0.06648827 - time (sec): 25.61 - samples/sec: 1346.03 - lr: 0.000023 - momentum: 0.000000
2024-03-26 11:02:25,873 ----------------------------------------------------------------------------------------------------
2024-03-26 11:02:25,873 EPOCH 6 done: loss 0.0665 - lr: 0.000023
2024-03-26 11:02:26,828 DEV : loss 0.1999371349811554 - f1-score (micro avg)  0.9161
2024-03-26 11:02:26,830 saving best model
2024-03-26 11:02:27,281 ----------------------------------------------------------------------------------------------------
2024-03-26 11:02:28,971 epoch 7 - iter 4/48 - loss 0.06036740 - time (sec): 1.69 - samples/sec: 1629.41 - lr: 0.000022 - momentum: 0.000000
2024-03-26 11:02:31,071 epoch 7 - iter 8/48 - loss 0.06185033 - time (sec): 3.79 - samples/sec: 1420.34 - lr: 0.000022 - momentum: 0.000000
2024-03-26 11:02:33,357 epoch 7 - iter 12/48 - loss 0.06731893 - time (sec): 6.07 - samples/sec: 1367.20 - lr: 0.000021 - momentum: 0.000000
2024-03-26 11:02:35,999 epoch 7 - iter 16/48 - loss 0.05862319 - time (sec): 8.71 - samples/sec: 1324.52 - lr: 0.000021 - momentum: 0.000000
2024-03-26 11:02:38,385 epoch 7 - iter 20/48 - loss 0.05493743 - time (sec): 11.10 - samples/sec: 1316.65 - lr: 0.000020 - momentum: 0.000000
2024-03-26 11:02:39,763 epoch 7 - iter 24/48 - loss 0.05293912 - time (sec): 12.48 - samples/sec: 1370.98 - lr: 0.000020 - momentum: 0.000000
2024-03-26 11:02:41,198 epoch 7 - iter 28/48 - loss 0.05212759 - time (sec): 13.91 - samples/sec: 1433.60 - lr: 0.000019 - momentum: 0.000000
2024-03-26 11:02:43,213 epoch 7 - iter 32/48 - loss 0.05081756 - time (sec): 15.93 - samples/sec: 1423.96 - lr: 0.000019 - momentum: 0.000000
2024-03-26 11:02:45,413 epoch 7 - iter 36/48 - loss 0.04903245 - time (sec): 18.13 - samples/sec: 1412.86 - lr: 0.000018 - momentum: 0.000000
2024-03-26 11:02:47,992 epoch 7 - iter 40/48 - loss 0.05091923 - time (sec): 20.71 - samples/sec: 1388.02 - lr: 0.000018 - momentum: 0.000000
2024-03-26 11:02:49,872 epoch 7 - iter 44/48 - loss 0.05031687 - time (sec): 22.59 - samples/sec: 1403.46 - lr: 0.000017 - momentum: 0.000000
2024-03-26 11:02:51,872 epoch 7 - iter 48/48 - loss 0.04903771 - time (sec): 24.59 - samples/sec: 1402.03 - lr: 0.000017 - momentum: 0.000000
2024-03-26 11:02:51,872 ----------------------------------------------------------------------------------------------------
2024-03-26 11:02:51,872 EPOCH 7 done: loss 0.0490 - lr: 0.000017
2024-03-26 11:02:52,826 DEV : loss 0.19628769159317017 - f1-score (micro avg)  0.91
2024-03-26 11:02:52,829 ----------------------------------------------------------------------------------------------------
2024-03-26 11:02:54,902 epoch 8 - iter 4/48 - loss 0.03574904 - time (sec): 2.07 - samples/sec: 1304.28 - lr: 0.000017 - momentum: 0.000000
2024-03-26 11:02:57,726 epoch 8 - iter 8/48 - loss 0.03594155 - time (sec): 4.90 - samples/sec: 1134.80 - lr: 0.000016 - momentum: 0.000000
2024-03-26 11:02:59,057 epoch 8 - iter 12/48 - loss 0.03729970 - time (sec): 6.23 - samples/sec: 1281.91 - lr: 0.000016 - momentum: 0.000000
2024-03-26 11:03:01,523 epoch 8 - iter 16/48 - loss 0.04291151 - time (sec): 8.69 - samples/sec: 1294.85 - lr: 0.000015 - momentum: 0.000000
2024-03-26 11:03:04,125 epoch 8 - iter 20/48 - loss 0.03659224 - time (sec): 11.30 - samples/sec: 1333.96 - lr: 0.000015 - momentum: 0.000000
2024-03-26 11:03:05,469 epoch 8 - iter 24/48 - loss 0.03664124 - time (sec): 12.64 - samples/sec: 1408.00 - lr: 0.000014 - momentum: 0.000000
2024-03-26 11:03:08,835 epoch 8 - iter 28/48 - loss 0.03886089 - time (sec): 16.01 - samples/sec: 1363.36 - lr: 0.000014 - momentum: 0.000000
2024-03-26 11:03:10,860 epoch 8 - iter 32/48 - loss 0.03830054 - time (sec): 18.03 - samples/sec: 1367.36 - lr: 0.000013 - momentum: 0.000000
2024-03-26 11:03:11,922 epoch 8 - iter 36/48 - loss 0.03798648 - time (sec): 19.09 - samples/sec: 1406.43 - lr: 0.000013 - momentum: 0.000000
2024-03-26 11:03:13,634 epoch 8 - iter 40/48 - loss 0.03809773 - time (sec): 20.80 - samples/sec: 1404.82 - lr: 0.000012 - momentum: 0.000000
2024-03-26 11:03:15,288 epoch 8 - iter 44/48 - loss 0.03863525 - time (sec): 22.46 - samples/sec: 1422.70 - lr: 0.000012 - momentum: 0.000000
2024-03-26 11:03:17,312 epoch 8 - iter 48/48 - loss 0.03975429 - time (sec): 24.48 - samples/sec: 1408.02 - lr: 0.000011 - momentum: 0.000000
2024-03-26 11:03:17,312 ----------------------------------------------------------------------------------------------------
2024-03-26 11:03:17,312 EPOCH 8 done: loss 0.0398 - lr: 0.000011
2024-03-26 11:03:18,283 DEV : loss 0.19287405908107758 - f1-score (micro avg)  0.9269
2024-03-26 11:03:18,284 saving best model
2024-03-26 11:03:18,696 ----------------------------------------------------------------------------------------------------
2024-03-26 11:03:20,629 epoch 9 - iter 4/48 - loss 0.01911719 - time (sec): 1.93 - samples/sec: 1386.69 - lr: 0.000011 - momentum: 0.000000
2024-03-26 11:03:23,880 epoch 9 - iter 8/48 - loss 0.01246336 - time (sec): 5.18 - samples/sec: 1204.95 - lr: 0.000011 - momentum: 0.000000
2024-03-26 11:03:25,618 epoch 9 - iter 12/48 - loss 0.01953089 - time (sec): 6.92 - samples/sec: 1255.75 - lr: 0.000010 - momentum: 0.000000
2024-03-26 11:03:27,891 epoch 9 - iter 16/48 - loss 0.02215764 - time (sec): 9.19 - samples/sec: 1253.18 - lr: 0.000010 - momentum: 0.000000
2024-03-26 11:03:30,272 epoch 9 - iter 20/48 - loss 0.02954547 - time (sec): 11.57 - samples/sec: 1278.31 - lr: 0.000009 - momentum: 0.000000
2024-03-26 11:03:32,531 epoch 9 - iter 24/48 - loss 0.03008158 - time (sec): 13.83 - samples/sec: 1292.82 - lr: 0.000009 - momentum: 0.000000
2024-03-26 11:03:34,919 epoch 9 - iter 28/48 - loss 0.02806782 - time (sec): 16.22 - samples/sec: 1292.06 - lr: 0.000008 - momentum: 0.000000
2024-03-26 11:03:37,371 epoch 9 - iter 32/48 - loss 0.02822729 - time (sec): 18.67 - samples/sec: 1285.90 - lr: 0.000008 - momentum: 0.000000
2024-03-26 11:03:39,285 epoch 9 - iter 36/48 - loss 0.03078142 - time (sec): 20.59 - samples/sec: 1300.40 - lr: 0.000007 - momentum: 0.000000
2024-03-26 11:03:41,552 epoch 9 - iter 40/48 - loss 0.03147072 - time (sec): 22.86 - samples/sec: 1290.12 - lr: 0.000007 - momentum: 0.000000
2024-03-26 11:03:43,705 epoch 9 - iter 44/48 - loss 0.03087065 - time (sec): 25.01 - samples/sec: 1303.67 - lr: 0.000006 - momentum: 0.000000
2024-03-26 11:03:44,519 epoch 9 - iter 48/48 - loss 0.03185010 - time (sec): 25.82 - samples/sec: 1335.01 - lr: 0.000006 - momentum: 0.000000
2024-03-26 11:03:44,519 ----------------------------------------------------------------------------------------------------
2024-03-26 11:03:44,519 EPOCH 9 done: loss 0.0319 - lr: 0.000006
2024-03-26 11:03:45,484 DEV : loss 0.19984783232212067 - f1-score (micro avg)  0.9272
2024-03-26 11:03:45,485 saving best model
2024-03-26 11:03:46,030 ----------------------------------------------------------------------------------------------------
2024-03-26 11:03:47,843 epoch 10 - iter 4/48 - loss 0.01174329 - time (sec): 1.81 - samples/sec: 1450.44 - lr: 0.000006 - momentum: 0.000000
2024-03-26 11:03:49,911 epoch 10 - iter 8/48 - loss 0.02116348 - time (sec): 3.88 - samples/sec: 1427.87 - lr: 0.000005 - momentum: 0.000000
2024-03-26 11:03:52,585 epoch 10 - iter 12/48 - loss 0.02314468 - time (sec): 6.55 - samples/sec: 1331.47 - lr: 0.000005 - momentum: 0.000000
2024-03-26 11:03:54,534 epoch 10 - iter 16/48 - loss 0.02792261 - time (sec): 8.50 - samples/sec: 1349.10 - lr: 0.000004 - momentum: 0.000000
2024-03-26 11:03:56,483 epoch 10 - iter 20/48 - loss 0.02603172 - time (sec): 10.45 - samples/sec: 1384.04 - lr: 0.000004 - momentum: 0.000000
2024-03-26 11:03:58,175 epoch 10 - iter 24/48 - loss 0.03120182 - time (sec): 12.14 - samples/sec: 1396.51 - lr: 0.000003 - momentum: 0.000000
2024-03-26 11:04:00,003 epoch 10 - iter 28/48 - loss 0.02955069 - time (sec): 13.97 - samples/sec: 1416.07 - lr: 0.000003 - momentum: 0.000000
2024-03-26 11:04:01,231 epoch 10 - iter 32/48 - loss 0.02961233 - time (sec): 15.20 - samples/sec: 1448.94 - lr: 0.000002 - momentum: 0.000000
2024-03-26 11:04:04,310 epoch 10 - iter 36/48 - loss 0.02651203 - time (sec): 18.28 - samples/sec: 1401.24 - lr: 0.000002 - momentum: 0.000000
2024-03-26 11:04:07,199 epoch 10 - iter 40/48 - loss 0.02941146 - time (sec): 21.17 - samples/sec: 1373.80 - lr: 0.000001 - momentum: 0.000000
2024-03-26 11:04:10,009 epoch 10 - iter 44/48 - loss 0.02745505 - time (sec): 23.98 - samples/sec: 1346.36 - lr: 0.000001 - momentum: 0.000000
2024-03-26 11:04:11,645 epoch 10 - iter 48/48 - loss 0.02670516 - time (sec): 25.61 - samples/sec: 1345.79 - lr: 0.000000 - momentum: 0.000000
2024-03-26 11:04:11,646 ----------------------------------------------------------------------------------------------------
2024-03-26 11:04:11,646 EPOCH 10 done: loss 0.0267 - lr: 0.000000
2024-03-26 11:04:12,623 DEV : loss 0.19916442036628723 - f1-score (micro avg)  0.9316
2024-03-26 11:04:12,624 saving best model
2024-03-26 11:04:13,361 ----------------------------------------------------------------------------------------------------
2024-03-26 11:04:13,362 Loading model from best epoch ...
2024-03-26 11:04:14,266 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 11:04:15,056 
Results:
- F-score (micro) 0.9173
- F-score (macro) 0.697
- Accuracy 0.8496

By class:
              precision    recall  f1-score   support

 Unternehmen     0.9115    0.8910    0.9011       266
 Auslagerung     0.8842    0.9197    0.9016       249
         Ort     0.9779    0.9925    0.9852       134
    Software     0.0000    0.0000    0.0000         0

   micro avg     0.9117    0.9230    0.9173       649
   macro avg     0.6934    0.7008    0.6970       649
weighted avg     0.9147    0.9230    0.9187       649

2024-03-26 11:04:15,056 ----------------------------------------------------------------------------------------------------