Hugging Face's logo

stefan-it
/
flair-co-funer-german_bert_base-bs16-e10-lr3e-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-lr3e-05-1
File size: 26,708 Bytes 
4ce6349
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
268
 
2024-03-26 10:55:09,597 ----------------------------------------------------------------------------------------------------
2024-03-26 10:55:09,597 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 10:55:09,597 ----------------------------------------------------------------------------------------------------
2024-03-26 10:55:09,597 Corpus: 758 train + 94 dev + 96 test sentences
2024-03-26 10:55:09,597 ----------------------------------------------------------------------------------------------------
2024-03-26 10:55:09,597 Train:  758 sentences
2024-03-26 10:55:09,597         (train_with_dev=False, train_with_test=False)
2024-03-26 10:55:09,597 ----------------------------------------------------------------------------------------------------
2024-03-26 10:55:09,598 Training Params:
2024-03-26 10:55:09,598  - learning_rate: "3e-05" 
2024-03-26 10:55:09,598  - mini_batch_size: "16"
2024-03-26 10:55:09,598  - max_epochs: "10"
2024-03-26 10:55:09,598  - shuffle: "True"
2024-03-26 10:55:09,598 ----------------------------------------------------------------------------------------------------
2024-03-26 10:55:09,598 Plugins:
2024-03-26 10:55:09,598  - TensorboardLogger
2024-03-26 10:55:09,598  - LinearScheduler | warmup_fraction: '0.1'
2024-03-26 10:55:09,598 ----------------------------------------------------------------------------------------------------
2024-03-26 10:55:09,598 Final evaluation on model from best epoch (best-model.pt)
2024-03-26 10:55:09,598  - metric: "('micro avg', 'f1-score')"
2024-03-26 10:55:09,598 ----------------------------------------------------------------------------------------------------
2024-03-26 10:55:09,598 Computation:
2024-03-26 10:55:09,598  - compute on device: cuda:0
2024-03-26 10:55:09,598  - embedding storage: none
2024-03-26 10:55:09,598 ----------------------------------------------------------------------------------------------------
2024-03-26 10:55:09,598 Model training base path: "flair-co-funer-german_bert_base-bs16-e10-lr3e-05-1"
2024-03-26 10:55:09,598 ----------------------------------------------------------------------------------------------------
2024-03-26 10:55:09,598 ----------------------------------------------------------------------------------------------------
2024-03-26 10:55:09,598 Logging anything other than scalars to TensorBoard is currently not supported.
2024-03-26 10:55:11,752 epoch 1 - iter 4/48 - loss 3.18136204 - time (sec): 2.15 - samples/sec: 1260.53 - lr: 0.000002 - momentum: 0.000000
2024-03-26 10:55:13,064 epoch 1 - iter 8/48 - loss 3.26104187 - time (sec): 3.47 - samples/sec: 1555.09 - lr: 0.000004 - momentum: 0.000000
2024-03-26 10:55:16,152 epoch 1 - iter 12/48 - loss 3.20211529 - time (sec): 6.55 - samples/sec: 1327.87 - lr: 0.000007 - momentum: 0.000000
2024-03-26 10:55:19,396 epoch 1 - iter 16/48 - loss 3.11692024 - time (sec): 9.80 - samples/sec: 1244.50 - lr: 0.000009 - momentum: 0.000000
2024-03-26 10:55:21,894 epoch 1 - iter 20/48 - loss 2.98646881 - time (sec): 12.30 - samples/sec: 1251.03 - lr: 0.000012 - momentum: 0.000000
2024-03-26 10:55:23,619 epoch 1 - iter 24/48 - loss 2.85768210 - time (sec): 14.02 - samples/sec: 1300.72 - lr: 0.000014 - momentum: 0.000000
2024-03-26 10:55:25,219 epoch 1 - iter 28/48 - loss 2.74291622 - time (sec): 15.62 - samples/sec: 1325.19 - lr: 0.000017 - momentum: 0.000000
2024-03-26 10:55:27,326 epoch 1 - iter 32/48 - loss 2.63361563 - time (sec): 17.73 - samples/sec: 1333.24 - lr: 0.000019 - momentum: 0.000000
2024-03-26 10:55:28,289 epoch 1 - iter 36/48 - loss 2.55063314 - time (sec): 18.69 - samples/sec: 1393.65 - lr: 0.000022 - momentum: 0.000000
2024-03-26 10:55:30,274 epoch 1 - iter 40/48 - loss 2.46376363 - time (sec): 20.68 - samples/sec: 1408.22 - lr: 0.000024 - momentum: 0.000000
2024-03-26 10:55:32,292 epoch 1 - iter 44/48 - loss 2.37711797 - time (sec): 22.69 - samples/sec: 1396.02 - lr: 0.000027 - momentum: 0.000000
2024-03-26 10:55:33,804 epoch 1 - iter 48/48 - loss 2.27943829 - time (sec): 24.21 - samples/sec: 1424.14 - lr: 0.000029 - momentum: 0.000000
2024-03-26 10:55:33,804 ----------------------------------------------------------------------------------------------------
2024-03-26 10:55:33,804 EPOCH 1 done: loss 2.2794 - lr: 0.000029
2024-03-26 10:55:34,638 DEV : loss 0.8841983079910278 - f1-score (micro avg)  0.3292
2024-03-26 10:55:34,639 saving best model
2024-03-26 10:55:34,944 ----------------------------------------------------------------------------------------------------
2024-03-26 10:55:37,559 epoch 2 - iter 4/48 - loss 1.06963551 - time (sec): 2.61 - samples/sec: 1186.48 - lr: 0.000030 - momentum: 0.000000
2024-03-26 10:55:39,674 epoch 2 - iter 8/48 - loss 1.02472331 - time (sec): 4.73 - samples/sec: 1397.79 - lr: 0.000030 - momentum: 0.000000
2024-03-26 10:55:41,963 epoch 2 - iter 12/48 - loss 0.95761277 - time (sec): 7.02 - samples/sec: 1319.21 - lr: 0.000029 - momentum: 0.000000
2024-03-26 10:55:44,022 epoch 2 - iter 16/48 - loss 0.90790118 - time (sec): 9.08 - samples/sec: 1312.57 - lr: 0.000029 - momentum: 0.000000
2024-03-26 10:55:46,126 epoch 2 - iter 20/48 - loss 0.85036938 - time (sec): 11.18 - samples/sec: 1341.05 - lr: 0.000029 - momentum: 0.000000
2024-03-26 10:55:49,236 epoch 2 - iter 24/48 - loss 0.77726093 - time (sec): 14.29 - samples/sec: 1294.67 - lr: 0.000028 - momentum: 0.000000
2024-03-26 10:55:51,622 epoch 2 - iter 28/48 - loss 0.75313006 - time (sec): 16.68 - samples/sec: 1291.59 - lr: 0.000028 - momentum: 0.000000
2024-03-26 10:55:53,372 epoch 2 - iter 32/48 - loss 0.72383530 - time (sec): 18.43 - samples/sec: 1309.23 - lr: 0.000028 - momentum: 0.000000
2024-03-26 10:55:54,428 epoch 2 - iter 36/48 - loss 0.70149258 - time (sec): 19.48 - samples/sec: 1357.69 - lr: 0.000028 - momentum: 0.000000
2024-03-26 10:55:56,284 epoch 2 - iter 40/48 - loss 0.67432397 - time (sec): 21.34 - samples/sec: 1377.62 - lr: 0.000027 - momentum: 0.000000
2024-03-26 10:55:58,311 epoch 2 - iter 44/48 - loss 0.65523371 - time (sec): 23.37 - samples/sec: 1374.06 - lr: 0.000027 - momentum: 0.000000
2024-03-26 10:55:59,735 epoch 2 - iter 48/48 - loss 0.63761461 - time (sec): 24.79 - samples/sec: 1390.52 - lr: 0.000027 - momentum: 0.000000
2024-03-26 10:55:59,735 ----------------------------------------------------------------------------------------------------
2024-03-26 10:55:59,735 EPOCH 2 done: loss 0.6376 - lr: 0.000027
2024-03-26 10:56:00,749 DEV : loss 0.3372988700866699 - f1-score (micro avg)  0.7573
2024-03-26 10:56:00,750 saving best model
2024-03-26 10:56:01,223 ----------------------------------------------------------------------------------------------------
2024-03-26 10:56:03,751 epoch 3 - iter 4/48 - loss 0.38597010 - time (sec): 2.53 - samples/sec: 1208.26 - lr: 0.000026 - momentum: 0.000000
2024-03-26 10:56:05,605 epoch 3 - iter 8/48 - loss 0.34811982 - time (sec): 4.38 - samples/sec: 1339.46 - lr: 0.000026 - momentum: 0.000000
2024-03-26 10:56:07,422 epoch 3 - iter 12/48 - loss 0.35445530 - time (sec): 6.20 - samples/sec: 1417.83 - lr: 0.000026 - momentum: 0.000000
2024-03-26 10:56:09,791 epoch 3 - iter 16/48 - loss 0.33399667 - time (sec): 8.57 - samples/sec: 1425.65 - lr: 0.000026 - momentum: 0.000000
2024-03-26 10:56:11,255 epoch 3 - iter 20/48 - loss 0.34563570 - time (sec): 10.03 - samples/sec: 1475.18 - lr: 0.000025 - momentum: 0.000000
2024-03-26 10:56:14,190 epoch 3 - iter 24/48 - loss 0.32843371 - time (sec): 12.96 - samples/sec: 1458.63 - lr: 0.000025 - momentum: 0.000000
2024-03-26 10:56:14,942 epoch 3 - iter 28/48 - loss 0.31446138 - time (sec): 13.72 - samples/sec: 1533.97 - lr: 0.000025 - momentum: 0.000000
2024-03-26 10:56:17,608 epoch 3 - iter 32/48 - loss 0.29864600 - time (sec): 16.38 - samples/sec: 1466.40 - lr: 0.000025 - momentum: 0.000000
2024-03-26 10:56:19,631 epoch 3 - iter 36/48 - loss 0.28700422 - time (sec): 18.41 - samples/sec: 1459.20 - lr: 0.000024 - momentum: 0.000000
2024-03-26 10:56:21,564 epoch 3 - iter 40/48 - loss 0.28584502 - time (sec): 20.34 - samples/sec: 1446.69 - lr: 0.000024 - momentum: 0.000000
2024-03-26 10:56:23,776 epoch 3 - iter 44/48 - loss 0.27720640 - time (sec): 22.55 - samples/sec: 1446.34 - lr: 0.000024 - momentum: 0.000000
2024-03-26 10:56:25,049 epoch 3 - iter 48/48 - loss 0.27449823 - time (sec): 23.82 - samples/sec: 1446.95 - lr: 0.000023 - momentum: 0.000000
2024-03-26 10:56:25,049 ----------------------------------------------------------------------------------------------------
2024-03-26 10:56:25,049 EPOCH 3 done: loss 0.2745 - lr: 0.000023
2024-03-26 10:56:25,977 DEV : loss 0.26073935627937317 - f1-score (micro avg)  0.8484
2024-03-26 10:56:25,979 saving best model
2024-03-26 10:56:26,464 ----------------------------------------------------------------------------------------------------
2024-03-26 10:56:27,937 epoch 4 - iter 4/48 - loss 0.19093305 - time (sec): 1.47 - samples/sec: 1852.02 - lr: 0.000023 - momentum: 0.000000
2024-03-26 10:56:30,418 epoch 4 - iter 8/48 - loss 0.18898170 - time (sec): 3.95 - samples/sec: 1451.37 - lr: 0.000023 - momentum: 0.000000
2024-03-26 10:56:32,511 epoch 4 - iter 12/48 - loss 0.20110181 - time (sec): 6.05 - samples/sec: 1445.00 - lr: 0.000023 - momentum: 0.000000
2024-03-26 10:56:34,730 epoch 4 - iter 16/48 - loss 0.18320119 - time (sec): 8.27 - samples/sec: 1448.64 - lr: 0.000022 - momentum: 0.000000
2024-03-26 10:56:37,704 epoch 4 - iter 20/48 - loss 0.17234055 - time (sec): 11.24 - samples/sec: 1378.13 - lr: 0.000022 - momentum: 0.000000
2024-03-26 10:56:39,167 epoch 4 - iter 24/48 - loss 0.17634685 - time (sec): 12.70 - samples/sec: 1418.82 - lr: 0.000022 - momentum: 0.000000
2024-03-26 10:56:40,684 epoch 4 - iter 28/48 - loss 0.17459195 - time (sec): 14.22 - samples/sec: 1458.49 - lr: 0.000022 - momentum: 0.000000
2024-03-26 10:56:43,284 epoch 4 - iter 32/48 - loss 0.17854116 - time (sec): 16.82 - samples/sec: 1441.08 - lr: 0.000021 - momentum: 0.000000
2024-03-26 10:56:44,303 epoch 4 - iter 36/48 - loss 0.18073101 - time (sec): 17.84 - samples/sec: 1489.43 - lr: 0.000021 - momentum: 0.000000
2024-03-26 10:56:46,702 epoch 4 - iter 40/48 - loss 0.17591819 - time (sec): 20.24 - samples/sec: 1443.13 - lr: 0.000021 - momentum: 0.000000
2024-03-26 10:56:48,542 epoch 4 - iter 44/48 - loss 0.17572825 - time (sec): 22.08 - samples/sec: 1461.47 - lr: 0.000020 - momentum: 0.000000
2024-03-26 10:56:49,925 epoch 4 - iter 48/48 - loss 0.17431063 - time (sec): 23.46 - samples/sec: 1469.37 - lr: 0.000020 - momentum: 0.000000
2024-03-26 10:56:49,925 ----------------------------------------------------------------------------------------------------
2024-03-26 10:56:49,925 EPOCH 4 done: loss 0.1743 - lr: 0.000020
2024-03-26 10:56:50,845 DEV : loss 0.22804546356201172 - f1-score (micro avg)  0.8739
2024-03-26 10:56:50,846 saving best model
2024-03-26 10:56:51,332 ----------------------------------------------------------------------------------------------------
2024-03-26 10:56:53,239 epoch 5 - iter 4/48 - loss 0.14175185 - time (sec): 1.91 - samples/sec: 1463.66 - lr: 0.000020 - momentum: 0.000000
2024-03-26 10:56:55,735 epoch 5 - iter 8/48 - loss 0.13481281 - time (sec): 4.40 - samples/sec: 1348.74 - lr: 0.000020 - momentum: 0.000000
2024-03-26 10:56:57,772 epoch 5 - iter 12/48 - loss 0.13837671 - time (sec): 6.44 - samples/sec: 1330.44 - lr: 0.000019 - momentum: 0.000000
2024-03-26 10:56:59,801 epoch 5 - iter 16/48 - loss 0.13341679 - time (sec): 8.47 - samples/sec: 1364.24 - lr: 0.000019 - momentum: 0.000000
2024-03-26 10:57:01,739 epoch 5 - iter 20/48 - loss 0.13770967 - time (sec): 10.41 - samples/sec: 1375.23 - lr: 0.000019 - momentum: 0.000000
2024-03-26 10:57:03,284 epoch 5 - iter 24/48 - loss 0.14293971 - time (sec): 11.95 - samples/sec: 1423.20 - lr: 0.000018 - momentum: 0.000000
2024-03-26 10:57:05,585 epoch 5 - iter 28/48 - loss 0.14527389 - time (sec): 14.25 - samples/sec: 1413.39 - lr: 0.000018 - momentum: 0.000000
2024-03-26 10:57:08,217 epoch 5 - iter 32/48 - loss 0.14233404 - time (sec): 16.88 - samples/sec: 1402.10 - lr: 0.000018 - momentum: 0.000000
2024-03-26 10:57:10,561 epoch 5 - iter 36/48 - loss 0.13484154 - time (sec): 19.23 - samples/sec: 1411.34 - lr: 0.000018 - momentum: 0.000000
2024-03-26 10:57:11,473 epoch 5 - iter 40/48 - loss 0.13458303 - time (sec): 20.14 - samples/sec: 1452.79 - lr: 0.000017 - momentum: 0.000000
2024-03-26 10:57:14,103 epoch 5 - iter 44/48 - loss 0.12839504 - time (sec): 22.77 - samples/sec: 1422.22 - lr: 0.000017 - momentum: 0.000000
2024-03-26 10:57:15,569 epoch 5 - iter 48/48 - loss 0.12741136 - time (sec): 24.24 - samples/sec: 1422.38 - lr: 0.000017 - momentum: 0.000000
2024-03-26 10:57:15,569 ----------------------------------------------------------------------------------------------------
2024-03-26 10:57:15,569 EPOCH 5 done: loss 0.1274 - lr: 0.000017
2024-03-26 10:57:16,515 DEV : loss 0.20272430777549744 - f1-score (micro avg)  0.8804
2024-03-26 10:57:16,516 saving best model
2024-03-26 10:57:16,984 ----------------------------------------------------------------------------------------------------
2024-03-26 10:57:18,992 epoch 6 - iter 4/48 - loss 0.07592408 - time (sec): 2.01 - samples/sec: 1317.35 - lr: 0.000017 - momentum: 0.000000
2024-03-26 10:57:21,186 epoch 6 - iter 8/48 - loss 0.10267334 - time (sec): 4.20 - samples/sec: 1316.74 - lr: 0.000016 - momentum: 0.000000
2024-03-26 10:57:23,020 epoch 6 - iter 12/48 - loss 0.10704529 - time (sec): 6.04 - samples/sec: 1431.77 - lr: 0.000016 - momentum: 0.000000
2024-03-26 10:57:25,277 epoch 6 - iter 16/48 - loss 0.10464018 - time (sec): 8.29 - samples/sec: 1385.41 - lr: 0.000016 - momentum: 0.000000
2024-03-26 10:57:27,081 epoch 6 - iter 20/48 - loss 0.11108015 - time (sec): 10.10 - samples/sec: 1389.85 - lr: 0.000015 - momentum: 0.000000
2024-03-26 10:57:29,574 epoch 6 - iter 24/48 - loss 0.10630629 - time (sec): 12.59 - samples/sec: 1367.78 - lr: 0.000015 - momentum: 0.000000
2024-03-26 10:57:31,519 epoch 6 - iter 28/48 - loss 0.10378445 - time (sec): 14.53 - samples/sec: 1360.83 - lr: 0.000015 - momentum: 0.000000
2024-03-26 10:57:34,025 epoch 6 - iter 32/48 - loss 0.10357046 - time (sec): 17.04 - samples/sec: 1341.33 - lr: 0.000015 - momentum: 0.000000
2024-03-26 10:57:37,488 epoch 6 - iter 36/48 - loss 0.09818676 - time (sec): 20.50 - samples/sec: 1300.63 - lr: 0.000014 - momentum: 0.000000
2024-03-26 10:57:39,188 epoch 6 - iter 40/48 - loss 0.09582576 - time (sec): 22.20 - samples/sec: 1330.90 - lr: 0.000014 - momentum: 0.000000
2024-03-26 10:57:41,073 epoch 6 - iter 44/48 - loss 0.09400591 - time (sec): 24.09 - samples/sec: 1333.22 - lr: 0.000014 - momentum: 0.000000
2024-03-26 10:57:42,362 epoch 6 - iter 48/48 - loss 0.09775463 - time (sec): 25.38 - samples/sec: 1358.38 - lr: 0.000014 - momentum: 0.000000
2024-03-26 10:57:42,362 ----------------------------------------------------------------------------------------------------
2024-03-26 10:57:42,362 EPOCH 6 done: loss 0.0978 - lr: 0.000014
2024-03-26 10:57:43,416 DEV : loss 0.20098459720611572 - f1-score (micro avg)  0.8917
2024-03-26 10:57:43,417 saving best model
2024-03-26 10:57:43,882 ----------------------------------------------------------------------------------------------------
2024-03-26 10:57:45,548 epoch 7 - iter 4/48 - loss 0.09646501 - time (sec): 1.67 - samples/sec: 1650.04 - lr: 0.000013 - momentum: 0.000000
2024-03-26 10:57:47,647 epoch 7 - iter 8/48 - loss 0.08084940 - time (sec): 3.76 - samples/sec: 1428.49 - lr: 0.000013 - momentum: 0.000000
2024-03-26 10:57:49,930 epoch 7 - iter 12/48 - loss 0.08590358 - time (sec): 6.05 - samples/sec: 1372.76 - lr: 0.000013 - momentum: 0.000000
2024-03-26 10:57:52,559 epoch 7 - iter 16/48 - loss 0.08260125 - time (sec): 8.68 - samples/sec: 1330.22 - lr: 0.000012 - momentum: 0.000000
2024-03-26 10:57:54,936 epoch 7 - iter 20/48 - loss 0.08060122 - time (sec): 11.05 - samples/sec: 1322.25 - lr: 0.000012 - momentum: 0.000000
2024-03-26 10:57:56,303 epoch 7 - iter 24/48 - loss 0.07855381 - time (sec): 12.42 - samples/sec: 1377.33 - lr: 0.000012 - momentum: 0.000000
2024-03-26 10:57:57,723 epoch 7 - iter 28/48 - loss 0.07878124 - time (sec): 13.84 - samples/sec: 1441.14 - lr: 0.000012 - momentum: 0.000000
2024-03-26 10:57:59,728 epoch 7 - iter 32/48 - loss 0.07627449 - time (sec): 15.85 - samples/sec: 1431.35 - lr: 0.000011 - momentum: 0.000000
2024-03-26 10:58:01,913 epoch 7 - iter 36/48 - loss 0.07425843 - time (sec): 18.03 - samples/sec: 1420.51 - lr: 0.000011 - momentum: 0.000000
2024-03-26 10:58:04,470 epoch 7 - iter 40/48 - loss 0.07509344 - time (sec): 20.59 - samples/sec: 1396.05 - lr: 0.000011 - momentum: 0.000000
2024-03-26 10:58:06,346 epoch 7 - iter 44/48 - loss 0.07606956 - time (sec): 22.46 - samples/sec: 1411.21 - lr: 0.000010 - momentum: 0.000000
2024-03-26 10:58:08,338 epoch 7 - iter 48/48 - loss 0.07594269 - time (sec): 24.46 - samples/sec: 1409.57 - lr: 0.000010 - momentum: 0.000000
2024-03-26 10:58:08,338 ----------------------------------------------------------------------------------------------------
2024-03-26 10:58:08,338 EPOCH 7 done: loss 0.0759 - lr: 0.000010
2024-03-26 10:58:09,277 DEV : loss 0.18951921164989471 - f1-score (micro avg)  0.8989
2024-03-26 10:58:09,278 saving best model
2024-03-26 10:58:09,759 ----------------------------------------------------------------------------------------------------
2024-03-26 10:58:11,825 epoch 8 - iter 4/48 - loss 0.06179609 - time (sec): 2.07 - samples/sec: 1308.88 - lr: 0.000010 - momentum: 0.000000
2024-03-26 10:58:14,623 epoch 8 - iter 8/48 - loss 0.05406802 - time (sec): 4.86 - samples/sec: 1142.42 - lr: 0.000010 - momentum: 0.000000
2024-03-26 10:58:15,947 epoch 8 - iter 12/48 - loss 0.05651994 - time (sec): 6.19 - samples/sec: 1290.20 - lr: 0.000009 - momentum: 0.000000
2024-03-26 10:58:18,398 epoch 8 - iter 16/48 - loss 0.06729804 - time (sec): 8.64 - samples/sec: 1303.19 - lr: 0.000009 - momentum: 0.000000
2024-03-26 10:58:20,988 epoch 8 - iter 20/48 - loss 0.06041800 - time (sec): 11.23 - samples/sec: 1341.99 - lr: 0.000009 - momentum: 0.000000
2024-03-26 10:58:22,321 epoch 8 - iter 24/48 - loss 0.06144765 - time (sec): 12.56 - samples/sec: 1416.83 - lr: 0.000009 - momentum: 0.000000
2024-03-26 10:58:25,656 epoch 8 - iter 28/48 - loss 0.06053862 - time (sec): 15.90 - samples/sec: 1372.70 - lr: 0.000008 - momentum: 0.000000
2024-03-26 10:58:27,673 epoch 8 - iter 32/48 - loss 0.06204024 - time (sec): 17.91 - samples/sec: 1376.29 - lr: 0.000008 - momentum: 0.000000
2024-03-26 10:58:28,730 epoch 8 - iter 36/48 - loss 0.06119637 - time (sec): 18.97 - samples/sec: 1415.54 - lr: 0.000008 - momentum: 0.000000
2024-03-26 10:58:30,430 epoch 8 - iter 40/48 - loss 0.06095629 - time (sec): 20.67 - samples/sec: 1413.97 - lr: 0.000007 - momentum: 0.000000
2024-03-26 10:58:32,073 epoch 8 - iter 44/48 - loss 0.06097632 - time (sec): 22.31 - samples/sec: 1431.98 - lr: 0.000007 - momentum: 0.000000
2024-03-26 10:58:34,072 epoch 8 - iter 48/48 - loss 0.06172598 - time (sec): 24.31 - samples/sec: 1417.89 - lr: 0.000007 - momentum: 0.000000
2024-03-26 10:58:34,073 ----------------------------------------------------------------------------------------------------
2024-03-26 10:58:34,073 EPOCH 8 done: loss 0.0617 - lr: 0.000007
2024-03-26 10:58:35,009 DEV : loss 0.19442327320575714 - f1-score (micro avg)  0.9037
2024-03-26 10:58:35,010 saving best model
2024-03-26 10:58:35,492 ----------------------------------------------------------------------------------------------------
2024-03-26 10:58:37,420 epoch 9 - iter 4/48 - loss 0.03787709 - time (sec): 1.93 - samples/sec: 1389.85 - lr: 0.000007 - momentum: 0.000000
2024-03-26 10:58:40,657 epoch 9 - iter 8/48 - loss 0.02721893 - time (sec): 5.16 - samples/sec: 1209.21 - lr: 0.000006 - momentum: 0.000000
2024-03-26 10:58:42,375 epoch 9 - iter 12/48 - loss 0.03793841 - time (sec): 6.88 - samples/sec: 1262.76 - lr: 0.000006 - momentum: 0.000000
2024-03-26 10:58:44,628 epoch 9 - iter 16/48 - loss 0.04341471 - time (sec): 9.13 - samples/sec: 1261.32 - lr: 0.000006 - momentum: 0.000000
2024-03-26 10:58:46,982 epoch 9 - iter 20/48 - loss 0.05014430 - time (sec): 11.49 - samples/sec: 1287.78 - lr: 0.000006 - momentum: 0.000000
2024-03-26 10:58:49,212 epoch 9 - iter 24/48 - loss 0.05183345 - time (sec): 13.72 - samples/sec: 1303.60 - lr: 0.000005 - momentum: 0.000000
2024-03-26 10:58:51,578 epoch 9 - iter 28/48 - loss 0.05030911 - time (sec): 16.09 - samples/sec: 1302.97 - lr: 0.000005 - momentum: 0.000000
2024-03-26 10:58:54,009 epoch 9 - iter 32/48 - loss 0.05050612 - time (sec): 18.52 - samples/sec: 1296.84 - lr: 0.000005 - momentum: 0.000000
2024-03-26 10:58:55,907 epoch 9 - iter 36/48 - loss 0.05285281 - time (sec): 20.41 - samples/sec: 1311.41 - lr: 0.000004 - momentum: 0.000000
2024-03-26 10:58:58,155 epoch 9 - iter 40/48 - loss 0.05425255 - time (sec): 22.66 - samples/sec: 1301.12 - lr: 0.000004 - momentum: 0.000000
2024-03-26 10:59:00,300 epoch 9 - iter 44/48 - loss 0.05319438 - time (sec): 24.81 - samples/sec: 1314.23 - lr: 0.000004 - momentum: 0.000000
2024-03-26 10:59:01,098 epoch 9 - iter 48/48 - loss 0.05376715 - time (sec): 25.61 - samples/sec: 1346.29 - lr: 0.000004 - momentum: 0.000000
2024-03-26 10:59:01,098 ----------------------------------------------------------------------------------------------------
2024-03-26 10:59:01,098 EPOCH 9 done: loss 0.0538 - lr: 0.000004
2024-03-26 10:59:02,047 DEV : loss 0.18321390450000763 - f1-score (micro avg)  0.9084
2024-03-26 10:59:02,048 saving best model
2024-03-26 10:59:02,515 ----------------------------------------------------------------------------------------------------
2024-03-26 10:59:04,383 epoch 10 - iter 4/48 - loss 0.03045480 - time (sec): 1.87 - samples/sec: 1407.80 - lr: 0.000003 - momentum: 0.000000
2024-03-26 10:59:06,446 epoch 10 - iter 8/48 - loss 0.03749981 - time (sec): 3.93 - samples/sec: 1409.71 - lr: 0.000003 - momentum: 0.000000
2024-03-26 10:59:09,113 epoch 10 - iter 12/48 - loss 0.03813115 - time (sec): 6.60 - samples/sec: 1322.73 - lr: 0.000003 - momentum: 0.000000
2024-03-26 10:59:11,061 epoch 10 - iter 16/48 - loss 0.04613621 - time (sec): 8.55 - samples/sec: 1342.43 - lr: 0.000002 - momentum: 0.000000
2024-03-26 10:59:12,999 epoch 10 - iter 20/48 - loss 0.04639203 - time (sec): 10.48 - samples/sec: 1379.82 - lr: 0.000002 - momentum: 0.000000
2024-03-26 10:59:14,686 epoch 10 - iter 24/48 - loss 0.05611524 - time (sec): 12.17 - samples/sec: 1393.39 - lr: 0.000002 - momentum: 0.000000
2024-03-26 10:59:16,506 epoch 10 - iter 28/48 - loss 0.05319972 - time (sec): 13.99 - samples/sec: 1414.20 - lr: 0.000002 - momentum: 0.000000
2024-03-26 10:59:17,731 epoch 10 - iter 32/48 - loss 0.05157465 - time (sec): 15.22 - samples/sec: 1447.46 - lr: 0.000001 - momentum: 0.000000
2024-03-26 10:59:20,788 epoch 10 - iter 36/48 - loss 0.04699753 - time (sec): 18.27 - samples/sec: 1401.78 - lr: 0.000001 - momentum: 0.000000
2024-03-26 10:59:23,662 epoch 10 - iter 40/48 - loss 0.05009458 - time (sec): 21.15 - samples/sec: 1375.18 - lr: 0.000001 - momentum: 0.000000
2024-03-26 10:59:26,468 epoch 10 - iter 44/48 - loss 0.04782682 - time (sec): 23.95 - samples/sec: 1347.81 - lr: 0.000001 - momentum: 0.000000
2024-03-26 10:59:28,113 epoch 10 - iter 48/48 - loss 0.04668607 - time (sec): 25.60 - samples/sec: 1346.68 - lr: 0.000000 - momentum: 0.000000
2024-03-26 10:59:28,114 ----------------------------------------------------------------------------------------------------
2024-03-26 10:59:28,114 EPOCH 10 done: loss 0.0467 - lr: 0.000000
2024-03-26 10:59:29,068 DEV : loss 0.18393239378929138 - f1-score (micro avg)  0.9136
2024-03-26 10:59:29,069 saving best model
2024-03-26 10:59:29,855 ----------------------------------------------------------------------------------------------------
2024-03-26 10:59:29,855 Loading model from best epoch ...
2024-03-26 10:59:30,803 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 10:59:31,572 
Results:
- F-score (micro) 0.9003
- F-score (macro) 0.6853
- Accuracy 0.821

By class:
              precision    recall  f1-score   support

 Unternehmen     0.8923    0.8722    0.8821       266
 Auslagerung     0.8677    0.8956    0.8814       249
         Ort     0.9706    0.9851    0.9778       134
    Software     0.0000    0.0000    0.0000         0

   micro avg     0.8962    0.9045    0.9003       649
   macro avg     0.6827    0.6882    0.6853       649
weighted avg     0.8990    0.9045    0.9016       649

2024-03-26 10:59:31,572 ----------------------------------------------------------------------------------------------------