Hugging Face's logo

stefan-it
/
flair-co-funer-german_bert_base-bs16-e10-lr3e-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-bs16-e10-lr3e-05-4
File size: 26,658 Bytes 
829a558
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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 11:45:43,177 ----------------------------------------------------------------------------------------------------
2024-03-26 11:45:43,177 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 11:45:43,178 ----------------------------------------------------------------------------------------------------
2024-03-26 11:45:43,178 Corpus: 758 train + 94 dev + 96 test sentences
2024-03-26 11:45:43,178 ----------------------------------------------------------------------------------------------------
2024-03-26 11:45:43,178 Train:  758 sentences
2024-03-26 11:45:43,178         (train_with_dev=False, train_with_test=False)
2024-03-26 11:45:43,178 ----------------------------------------------------------------------------------------------------
2024-03-26 11:45:43,178 Training Params:
2024-03-26 11:45:43,178  - learning_rate: "3e-05" 
2024-03-26 11:45:43,178  - mini_batch_size: "16"
2024-03-26 11:45:43,178  - max_epochs: "10"
2024-03-26 11:45:43,178  - shuffle: "True"
2024-03-26 11:45:43,178 ----------------------------------------------------------------------------------------------------
2024-03-26 11:45:43,178 Plugins:
2024-03-26 11:45:43,178  - TensorboardLogger
2024-03-26 11:45:43,178  - LinearScheduler | warmup_fraction: '0.1'
2024-03-26 11:45:43,178 ----------------------------------------------------------------------------------------------------
2024-03-26 11:45:43,178 Final evaluation on model from best epoch (best-model.pt)
2024-03-26 11:45:43,178  - metric: "('micro avg', 'f1-score')"
2024-03-26 11:45:43,178 ----------------------------------------------------------------------------------------------------
2024-03-26 11:45:43,178 Computation:
2024-03-26 11:45:43,178  - compute on device: cuda:0
2024-03-26 11:45:43,178  - embedding storage: none
2024-03-26 11:45:43,178 ----------------------------------------------------------------------------------------------------
2024-03-26 11:45:43,178 Model training base path: "flair-co-funer-german_bert_base-bs16-e10-lr3e-05-4"
2024-03-26 11:45:43,178 ----------------------------------------------------------------------------------------------------
2024-03-26 11:45:43,178 ----------------------------------------------------------------------------------------------------
2024-03-26 11:45:43,178 Logging anything other than scalars to TensorBoard is currently not supported.
2024-03-26 11:45:44,651 epoch 1 - iter 4/48 - loss 3.09549930 - time (sec): 1.47 - samples/sec: 1771.64 - lr: 0.000002 - momentum: 0.000000
2024-03-26 11:45:46,599 epoch 1 - iter 8/48 - loss 3.05860497 - time (sec): 3.42 - samples/sec: 1497.54 - lr: 0.000004 - momentum: 0.000000
2024-03-26 11:45:47,991 epoch 1 - iter 12/48 - loss 2.97301780 - time (sec): 4.81 - samples/sec: 1517.11 - lr: 0.000007 - momentum: 0.000000
2024-03-26 11:45:50,619 epoch 1 - iter 16/48 - loss 2.81904351 - time (sec): 7.44 - samples/sec: 1437.84 - lr: 0.000009 - momentum: 0.000000
2024-03-26 11:45:52,789 epoch 1 - iter 20/48 - loss 2.69072216 - time (sec): 9.61 - samples/sec: 1425.29 - lr: 0.000012 - momentum: 0.000000
2024-03-26 11:45:55,470 epoch 1 - iter 24/48 - loss 2.55781908 - time (sec): 12.29 - samples/sec: 1376.96 - lr: 0.000014 - momentum: 0.000000
2024-03-26 11:45:58,068 epoch 1 - iter 28/48 - loss 2.43219709 - time (sec): 14.89 - samples/sec: 1362.47 - lr: 0.000017 - momentum: 0.000000
2024-03-26 11:45:59,951 epoch 1 - iter 32/48 - loss 2.34102141 - time (sec): 16.77 - samples/sec: 1363.77 - lr: 0.000019 - momentum: 0.000000
2024-03-26 11:46:00,897 epoch 1 - iter 36/48 - loss 2.26816176 - time (sec): 17.72 - samples/sec: 1409.81 - lr: 0.000022 - momentum: 0.000000
2024-03-26 11:46:02,804 epoch 1 - iter 40/48 - loss 2.17382454 - time (sec): 19.63 - samples/sec: 1419.19 - lr: 0.000024 - momentum: 0.000000
2024-03-26 11:46:04,923 epoch 1 - iter 44/48 - loss 2.06590701 - time (sec): 21.74 - samples/sec: 1436.16 - lr: 0.000027 - momentum: 0.000000
2024-03-26 11:46:06,685 epoch 1 - iter 48/48 - loss 1.97932119 - time (sec): 23.51 - samples/sec: 1466.46 - lr: 0.000029 - momentum: 0.000000
2024-03-26 11:46:06,686 ----------------------------------------------------------------------------------------------------
2024-03-26 11:46:06,686 EPOCH 1 done: loss 1.9793 - lr: 0.000029
2024-03-26 11:46:07,557 DEV : loss 0.7610657215118408 - f1-score (micro avg)  0.523
2024-03-26 11:46:07,559 saving best model
2024-03-26 11:46:07,873 ----------------------------------------------------------------------------------------------------
2024-03-26 11:46:09,182 epoch 2 - iter 4/48 - loss 1.00588389 - time (sec): 1.31 - samples/sec: 1809.27 - lr: 0.000030 - momentum: 0.000000
2024-03-26 11:46:11,520 epoch 2 - iter 8/48 - loss 0.79773619 - time (sec): 3.65 - samples/sec: 1496.17 - lr: 0.000030 - momentum: 0.000000
2024-03-26 11:46:13,365 epoch 2 - iter 12/48 - loss 0.75438594 - time (sec): 5.49 - samples/sec: 1551.86 - lr: 0.000029 - momentum: 0.000000
2024-03-26 11:46:15,862 epoch 2 - iter 16/48 - loss 0.69088754 - time (sec): 7.99 - samples/sec: 1412.25 - lr: 0.000029 - momentum: 0.000000
2024-03-26 11:46:19,281 epoch 2 - iter 20/48 - loss 0.63021194 - time (sec): 11.41 - samples/sec: 1293.37 - lr: 0.000029 - momentum: 0.000000
2024-03-26 11:46:20,807 epoch 2 - iter 24/48 - loss 0.62522090 - time (sec): 12.93 - samples/sec: 1348.63 - lr: 0.000028 - momentum: 0.000000
2024-03-26 11:46:23,493 epoch 2 - iter 28/48 - loss 0.60290004 - time (sec): 15.62 - samples/sec: 1325.52 - lr: 0.000028 - momentum: 0.000000
2024-03-26 11:46:26,244 epoch 2 - iter 32/48 - loss 0.57675820 - time (sec): 18.37 - samples/sec: 1328.72 - lr: 0.000028 - momentum: 0.000000
2024-03-26 11:46:28,390 epoch 2 - iter 36/48 - loss 0.56861350 - time (sec): 20.52 - samples/sec: 1318.12 - lr: 0.000028 - momentum: 0.000000
2024-03-26 11:46:30,965 epoch 2 - iter 40/48 - loss 0.54982812 - time (sec): 23.09 - samples/sec: 1307.11 - lr: 0.000027 - momentum: 0.000000
2024-03-26 11:46:32,094 epoch 2 - iter 44/48 - loss 0.53992976 - time (sec): 24.22 - samples/sec: 1338.69 - lr: 0.000027 - momentum: 0.000000
2024-03-26 11:46:33,278 epoch 2 - iter 48/48 - loss 0.52605631 - time (sec): 25.40 - samples/sec: 1356.93 - lr: 0.000027 - momentum: 0.000000
2024-03-26 11:46:33,278 ----------------------------------------------------------------------------------------------------
2024-03-26 11:46:33,278 EPOCH 2 done: loss 0.5261 - lr: 0.000027
2024-03-26 11:46:34,230 DEV : loss 0.3057578504085541 - f1-score (micro avg)  0.8021
2024-03-26 11:46:34,232 saving best model
2024-03-26 11:46:34,699 ----------------------------------------------------------------------------------------------------
2024-03-26 11:46:36,914 epoch 3 - iter 4/48 - loss 0.30145275 - time (sec): 2.21 - samples/sec: 1108.67 - lr: 0.000026 - momentum: 0.000000
2024-03-26 11:46:38,470 epoch 3 - iter 8/48 - loss 0.25277524 - time (sec): 3.77 - samples/sec: 1270.27 - lr: 0.000026 - momentum: 0.000000
2024-03-26 11:46:41,075 epoch 3 - iter 12/48 - loss 0.26402459 - time (sec): 6.38 - samples/sec: 1220.17 - lr: 0.000026 - momentum: 0.000000
2024-03-26 11:46:43,227 epoch 3 - iter 16/48 - loss 0.26444662 - time (sec): 8.53 - samples/sec: 1250.49 - lr: 0.000026 - momentum: 0.000000
2024-03-26 11:46:45,257 epoch 3 - iter 20/48 - loss 0.26227190 - time (sec): 10.56 - samples/sec: 1311.31 - lr: 0.000025 - momentum: 0.000000
2024-03-26 11:46:47,499 epoch 3 - iter 24/48 - loss 0.25621355 - time (sec): 12.80 - samples/sec: 1335.61 - lr: 0.000025 - momentum: 0.000000
2024-03-26 11:46:50,084 epoch 3 - iter 28/48 - loss 0.24877474 - time (sec): 15.38 - samples/sec: 1296.51 - lr: 0.000025 - momentum: 0.000000
2024-03-26 11:46:52,756 epoch 3 - iter 32/48 - loss 0.24041754 - time (sec): 18.06 - samples/sec: 1273.36 - lr: 0.000025 - momentum: 0.000000
2024-03-26 11:46:54,934 epoch 3 - iter 36/48 - loss 0.23832241 - time (sec): 20.23 - samples/sec: 1279.35 - lr: 0.000024 - momentum: 0.000000
2024-03-26 11:46:57,306 epoch 3 - iter 40/48 - loss 0.24630100 - time (sec): 22.61 - samples/sec: 1296.33 - lr: 0.000024 - momentum: 0.000000
2024-03-26 11:46:59,921 epoch 3 - iter 44/48 - loss 0.24001606 - time (sec): 25.22 - samples/sec: 1280.99 - lr: 0.000024 - momentum: 0.000000
2024-03-26 11:47:01,444 epoch 3 - iter 48/48 - loss 0.24260838 - time (sec): 26.74 - samples/sec: 1288.93 - lr: 0.000023 - momentum: 0.000000
2024-03-26 11:47:01,444 ----------------------------------------------------------------------------------------------------
2024-03-26 11:47:01,444 EPOCH 3 done: loss 0.2426 - lr: 0.000023
2024-03-26 11:47:02,396 DEV : loss 0.24360495805740356 - f1-score (micro avg)  0.8595
2024-03-26 11:47:02,398 saving best model
2024-03-26 11:47:02,856 ----------------------------------------------------------------------------------------------------
2024-03-26 11:47:05,914 epoch 4 - iter 4/48 - loss 0.12795448 - time (sec): 3.06 - samples/sec: 1192.54 - lr: 0.000023 - momentum: 0.000000
2024-03-26 11:47:07,229 epoch 4 - iter 8/48 - loss 0.14859439 - time (sec): 4.37 - samples/sec: 1345.42 - lr: 0.000023 - momentum: 0.000000
2024-03-26 11:47:09,386 epoch 4 - iter 12/48 - loss 0.16381198 - time (sec): 6.53 - samples/sec: 1412.84 - lr: 0.000023 - momentum: 0.000000
2024-03-26 11:47:12,101 epoch 4 - iter 16/48 - loss 0.16311786 - time (sec): 9.24 - samples/sec: 1317.99 - lr: 0.000022 - momentum: 0.000000
2024-03-26 11:47:13,115 epoch 4 - iter 20/48 - loss 0.16463006 - time (sec): 10.26 - samples/sec: 1400.38 - lr: 0.000022 - momentum: 0.000000
2024-03-26 11:47:14,580 epoch 4 - iter 24/48 - loss 0.16449895 - time (sec): 11.72 - samples/sec: 1441.92 - lr: 0.000022 - momentum: 0.000000
2024-03-26 11:47:17,752 epoch 4 - iter 28/48 - loss 0.15737998 - time (sec): 14.89 - samples/sec: 1355.04 - lr: 0.000022 - momentum: 0.000000
2024-03-26 11:47:20,326 epoch 4 - iter 32/48 - loss 0.16722060 - time (sec): 17.47 - samples/sec: 1346.27 - lr: 0.000021 - momentum: 0.000000
2024-03-26 11:47:21,933 epoch 4 - iter 36/48 - loss 0.16726320 - time (sec): 19.08 - samples/sec: 1377.55 - lr: 0.000021 - momentum: 0.000000
2024-03-26 11:47:23,982 epoch 4 - iter 40/48 - loss 0.16404076 - time (sec): 21.12 - samples/sec: 1391.74 - lr: 0.000021 - momentum: 0.000000
2024-03-26 11:47:25,933 epoch 4 - iter 44/48 - loss 0.16294098 - time (sec): 23.08 - samples/sec: 1405.54 - lr: 0.000020 - momentum: 0.000000
2024-03-26 11:47:27,014 epoch 4 - iter 48/48 - loss 0.16439185 - time (sec): 24.16 - samples/sec: 1427.00 - lr: 0.000020 - momentum: 0.000000
2024-03-26 11:47:27,015 ----------------------------------------------------------------------------------------------------
2024-03-26 11:47:27,015 EPOCH 4 done: loss 0.1644 - lr: 0.000020
2024-03-26 11:47:27,966 DEV : loss 0.23152601718902588 - f1-score (micro avg)  0.8802
2024-03-26 11:47:27,968 saving best model
2024-03-26 11:47:28,459 ----------------------------------------------------------------------------------------------------
2024-03-26 11:47:29,577 epoch 5 - iter 4/48 - loss 0.19052126 - time (sec): 1.11 - samples/sec: 2287.23 - lr: 0.000020 - momentum: 0.000000
2024-03-26 11:47:31,576 epoch 5 - iter 8/48 - loss 0.17349732 - time (sec): 3.11 - samples/sec: 1665.56 - lr: 0.000020 - momentum: 0.000000
2024-03-26 11:47:33,746 epoch 5 - iter 12/48 - loss 0.15486907 - time (sec): 5.28 - samples/sec: 1515.53 - lr: 0.000019 - momentum: 0.000000
2024-03-26 11:47:36,149 epoch 5 - iter 16/48 - loss 0.14690282 - time (sec): 7.68 - samples/sec: 1443.38 - lr: 0.000019 - momentum: 0.000000
2024-03-26 11:47:38,420 epoch 5 - iter 20/48 - loss 0.14271247 - time (sec): 9.95 - samples/sec: 1374.60 - lr: 0.000019 - momentum: 0.000000
2024-03-26 11:47:40,647 epoch 5 - iter 24/48 - loss 0.13760899 - time (sec): 12.18 - samples/sec: 1394.70 - lr: 0.000018 - momentum: 0.000000
2024-03-26 11:47:42,374 epoch 5 - iter 28/48 - loss 0.13435191 - time (sec): 13.91 - samples/sec: 1415.02 - lr: 0.000018 - momentum: 0.000000
2024-03-26 11:47:44,537 epoch 5 - iter 32/48 - loss 0.12594520 - time (sec): 16.07 - samples/sec: 1437.44 - lr: 0.000018 - momentum: 0.000000
2024-03-26 11:47:46,006 epoch 5 - iter 36/48 - loss 0.12460472 - time (sec): 17.54 - samples/sec: 1459.35 - lr: 0.000018 - momentum: 0.000000
2024-03-26 11:47:48,705 epoch 5 - iter 40/48 - loss 0.11980810 - time (sec): 20.24 - samples/sec: 1424.31 - lr: 0.000017 - momentum: 0.000000
2024-03-26 11:47:51,717 epoch 5 - iter 44/48 - loss 0.11945876 - time (sec): 23.25 - samples/sec: 1377.39 - lr: 0.000017 - momentum: 0.000000
2024-03-26 11:47:53,286 epoch 5 - iter 48/48 - loss 0.12163729 - time (sec): 24.82 - samples/sec: 1388.84 - lr: 0.000017 - momentum: 0.000000
2024-03-26 11:47:53,286 ----------------------------------------------------------------------------------------------------
2024-03-26 11:47:53,286 EPOCH 5 done: loss 0.1216 - lr: 0.000017
2024-03-26 11:47:54,267 DEV : loss 0.18433180451393127 - f1-score (micro avg)  0.8884
2024-03-26 11:47:54,269 saving best model
2024-03-26 11:47:54,756 ----------------------------------------------------------------------------------------------------
2024-03-26 11:47:56,703 epoch 6 - iter 4/48 - loss 0.13051442 - time (sec): 1.95 - samples/sec: 1509.64 - lr: 0.000017 - momentum: 0.000000
2024-03-26 11:47:58,475 epoch 6 - iter 8/48 - loss 0.10157641 - time (sec): 3.72 - samples/sec: 1558.95 - lr: 0.000016 - momentum: 0.000000
2024-03-26 11:48:00,861 epoch 6 - iter 12/48 - loss 0.10287141 - time (sec): 6.10 - samples/sec: 1444.47 - lr: 0.000016 - momentum: 0.000000
2024-03-26 11:48:02,495 epoch 6 - iter 16/48 - loss 0.09403836 - time (sec): 7.74 - samples/sec: 1464.25 - lr: 0.000016 - momentum: 0.000000
2024-03-26 11:48:05,189 epoch 6 - iter 20/48 - loss 0.08683212 - time (sec): 10.43 - samples/sec: 1377.07 - lr: 0.000015 - momentum: 0.000000
2024-03-26 11:48:07,268 epoch 6 - iter 24/48 - loss 0.08891366 - time (sec): 12.51 - samples/sec: 1397.92 - lr: 0.000015 - momentum: 0.000000
2024-03-26 11:48:10,008 epoch 6 - iter 28/48 - loss 0.08855013 - time (sec): 15.25 - samples/sec: 1372.63 - lr: 0.000015 - momentum: 0.000000
2024-03-26 11:48:12,127 epoch 6 - iter 32/48 - loss 0.08678004 - time (sec): 17.37 - samples/sec: 1352.96 - lr: 0.000015 - momentum: 0.000000
2024-03-26 11:48:13,293 epoch 6 - iter 36/48 - loss 0.08791381 - time (sec): 18.54 - samples/sec: 1399.47 - lr: 0.000014 - momentum: 0.000000
2024-03-26 11:48:15,565 epoch 6 - iter 40/48 - loss 0.08879619 - time (sec): 20.81 - samples/sec: 1389.61 - lr: 0.000014 - momentum: 0.000000
2024-03-26 11:48:17,262 epoch 6 - iter 44/48 - loss 0.09174772 - time (sec): 22.50 - samples/sec: 1410.84 - lr: 0.000014 - momentum: 0.000000
2024-03-26 11:48:19,176 epoch 6 - iter 48/48 - loss 0.08921771 - time (sec): 24.42 - samples/sec: 1411.69 - lr: 0.000014 - momentum: 0.000000
2024-03-26 11:48:19,176 ----------------------------------------------------------------------------------------------------
2024-03-26 11:48:19,176 EPOCH 6 done: loss 0.0892 - lr: 0.000014
2024-03-26 11:48:20,141 DEV : loss 0.1853693574666977 - f1-score (micro avg)  0.9087
2024-03-26 11:48:20,143 saving best model
2024-03-26 11:48:20,630 ----------------------------------------------------------------------------------------------------
2024-03-26 11:48:22,266 epoch 7 - iter 4/48 - loss 0.06603823 - time (sec): 1.64 - samples/sec: 1709.98 - lr: 0.000013 - momentum: 0.000000
2024-03-26 11:48:24,417 epoch 7 - iter 8/48 - loss 0.05619435 - time (sec): 3.79 - samples/sec: 1616.26 - lr: 0.000013 - momentum: 0.000000
2024-03-26 11:48:26,762 epoch 7 - iter 12/48 - loss 0.05569263 - time (sec): 6.13 - samples/sec: 1436.63 - lr: 0.000013 - momentum: 0.000000
2024-03-26 11:48:28,025 epoch 7 - iter 16/48 - loss 0.06293055 - time (sec): 7.39 - samples/sec: 1522.02 - lr: 0.000012 - momentum: 0.000000
2024-03-26 11:48:30,201 epoch 7 - iter 20/48 - loss 0.06307589 - time (sec): 9.57 - samples/sec: 1498.89 - lr: 0.000012 - momentum: 0.000000
2024-03-26 11:48:31,777 epoch 7 - iter 24/48 - loss 0.06017691 - time (sec): 11.15 - samples/sec: 1544.12 - lr: 0.000012 - momentum: 0.000000
2024-03-26 11:48:33,986 epoch 7 - iter 28/48 - loss 0.06081892 - time (sec): 13.36 - samples/sec: 1503.01 - lr: 0.000012 - momentum: 0.000000
2024-03-26 11:48:36,785 epoch 7 - iter 32/48 - loss 0.06314787 - time (sec): 16.16 - samples/sec: 1441.57 - lr: 0.000011 - momentum: 0.000000
2024-03-26 11:48:38,902 epoch 7 - iter 36/48 - loss 0.06250353 - time (sec): 18.27 - samples/sec: 1436.17 - lr: 0.000011 - momentum: 0.000000
2024-03-26 11:48:40,055 epoch 7 - iter 40/48 - loss 0.06618433 - time (sec): 19.43 - samples/sec: 1467.11 - lr: 0.000011 - momentum: 0.000000
2024-03-26 11:48:42,708 epoch 7 - iter 44/48 - loss 0.06737787 - time (sec): 22.08 - samples/sec: 1452.17 - lr: 0.000010 - momentum: 0.000000
2024-03-26 11:48:43,879 epoch 7 - iter 48/48 - loss 0.06792919 - time (sec): 23.25 - samples/sec: 1482.72 - lr: 0.000010 - momentum: 0.000000
2024-03-26 11:48:43,880 ----------------------------------------------------------------------------------------------------
2024-03-26 11:48:43,880 EPOCH 7 done: loss 0.0679 - lr: 0.000010
2024-03-26 11:48:44,827 DEV : loss 0.18232010304927826 - f1-score (micro avg)  0.91
2024-03-26 11:48:44,829 saving best model
2024-03-26 11:48:45,300 ----------------------------------------------------------------------------------------------------
2024-03-26 11:48:47,467 epoch 8 - iter 4/48 - loss 0.03714663 - time (sec): 2.16 - samples/sec: 1280.39 - lr: 0.000010 - momentum: 0.000000
2024-03-26 11:48:50,167 epoch 8 - iter 8/48 - loss 0.03361336 - time (sec): 4.87 - samples/sec: 1241.21 - lr: 0.000010 - momentum: 0.000000
2024-03-26 11:48:51,850 epoch 8 - iter 12/48 - loss 0.03532371 - time (sec): 6.55 - samples/sec: 1295.03 - lr: 0.000009 - momentum: 0.000000
2024-03-26 11:48:54,494 epoch 8 - iter 16/48 - loss 0.04333594 - time (sec): 9.19 - samples/sec: 1252.38 - lr: 0.000009 - momentum: 0.000000
2024-03-26 11:48:56,236 epoch 8 - iter 20/48 - loss 0.04632785 - time (sec): 10.93 - samples/sec: 1299.41 - lr: 0.000009 - momentum: 0.000000
2024-03-26 11:48:57,768 epoch 8 - iter 24/48 - loss 0.05319808 - time (sec): 12.47 - samples/sec: 1363.58 - lr: 0.000009 - momentum: 0.000000
2024-03-26 11:48:59,674 epoch 8 - iter 28/48 - loss 0.05657064 - time (sec): 14.37 - samples/sec: 1387.59 - lr: 0.000008 - momentum: 0.000000
2024-03-26 11:49:02,385 epoch 8 - iter 32/48 - loss 0.05777782 - time (sec): 17.08 - samples/sec: 1375.19 - lr: 0.000008 - momentum: 0.000000
2024-03-26 11:49:04,885 epoch 8 - iter 36/48 - loss 0.05931562 - time (sec): 19.58 - samples/sec: 1365.05 - lr: 0.000008 - momentum: 0.000000
2024-03-26 11:49:07,154 epoch 8 - iter 40/48 - loss 0.05882360 - time (sec): 21.85 - samples/sec: 1346.48 - lr: 0.000007 - momentum: 0.000000
2024-03-26 11:49:09,580 epoch 8 - iter 44/48 - loss 0.05703173 - time (sec): 24.28 - samples/sec: 1330.58 - lr: 0.000007 - momentum: 0.000000
2024-03-26 11:49:11,182 epoch 8 - iter 48/48 - loss 0.05700452 - time (sec): 25.88 - samples/sec: 1331.98 - lr: 0.000007 - momentum: 0.000000
2024-03-26 11:49:11,182 ----------------------------------------------------------------------------------------------------
2024-03-26 11:49:11,182 EPOCH 8 done: loss 0.0570 - lr: 0.000007
2024-03-26 11:49:12,150 DEV : loss 0.1904592365026474 - f1-score (micro avg)  0.9126
2024-03-26 11:49:12,153 saving best model
2024-03-26 11:49:12,624 ----------------------------------------------------------------------------------------------------
2024-03-26 11:49:14,556 epoch 9 - iter 4/48 - loss 0.05954110 - time (sec): 1.93 - samples/sec: 1495.61 - lr: 0.000007 - momentum: 0.000000
2024-03-26 11:49:17,841 epoch 9 - iter 8/48 - loss 0.05659017 - time (sec): 5.22 - samples/sec: 1206.66 - lr: 0.000006 - momentum: 0.000000
2024-03-26 11:49:19,553 epoch 9 - iter 12/48 - loss 0.04714393 - time (sec): 6.93 - samples/sec: 1248.33 - lr: 0.000006 - momentum: 0.000000
2024-03-26 11:49:21,495 epoch 9 - iter 16/48 - loss 0.05412922 - time (sec): 8.87 - samples/sec: 1288.76 - lr: 0.000006 - momentum: 0.000000
2024-03-26 11:49:24,385 epoch 9 - iter 20/48 - loss 0.04842308 - time (sec): 11.76 - samples/sec: 1262.88 - lr: 0.000006 - momentum: 0.000000
2024-03-26 11:49:25,948 epoch 9 - iter 24/48 - loss 0.04794347 - time (sec): 13.32 - samples/sec: 1309.55 - lr: 0.000005 - momentum: 0.000000
2024-03-26 11:49:27,975 epoch 9 - iter 28/48 - loss 0.05034416 - time (sec): 15.35 - samples/sec: 1330.81 - lr: 0.000005 - momentum: 0.000000
2024-03-26 11:49:30,408 epoch 9 - iter 32/48 - loss 0.04906075 - time (sec): 17.78 - samples/sec: 1306.28 - lr: 0.000005 - momentum: 0.000000
2024-03-26 11:49:31,739 epoch 9 - iter 36/48 - loss 0.05395636 - time (sec): 19.11 - samples/sec: 1337.37 - lr: 0.000004 - momentum: 0.000000
2024-03-26 11:49:35,032 epoch 9 - iter 40/48 - loss 0.05096711 - time (sec): 22.41 - samples/sec: 1291.46 - lr: 0.000004 - momentum: 0.000000
2024-03-26 11:49:37,174 epoch 9 - iter 44/48 - loss 0.04799635 - time (sec): 24.55 - samples/sec: 1315.65 - lr: 0.000004 - momentum: 0.000000
2024-03-26 11:49:38,168 epoch 9 - iter 48/48 - loss 0.04898665 - time (sec): 25.54 - samples/sec: 1349.54 - lr: 0.000004 - momentum: 0.000000
2024-03-26 11:49:38,168 ----------------------------------------------------------------------------------------------------
2024-03-26 11:49:38,168 EPOCH 9 done: loss 0.0490 - lr: 0.000004
2024-03-26 11:49:39,113 DEV : loss 0.18231666088104248 - f1-score (micro avg)  0.9251
2024-03-26 11:49:39,114 saving best model
2024-03-26 11:49:39,563 ----------------------------------------------------------------------------------------------------
2024-03-26 11:49:41,526 epoch 10 - iter 4/48 - loss 0.05551802 - time (sec): 1.96 - samples/sec: 1317.05 - lr: 0.000003 - momentum: 0.000000
2024-03-26 11:49:44,390 epoch 10 - iter 8/48 - loss 0.03666070 - time (sec): 4.83 - samples/sec: 1198.44 - lr: 0.000003 - momentum: 0.000000
2024-03-26 11:49:46,447 epoch 10 - iter 12/48 - loss 0.04190061 - time (sec): 6.88 - samples/sec: 1265.75 - lr: 0.000003 - momentum: 0.000000
2024-03-26 11:49:48,559 epoch 10 - iter 16/48 - loss 0.04315321 - time (sec): 9.00 - samples/sec: 1352.32 - lr: 0.000002 - momentum: 0.000000
2024-03-26 11:49:49,427 epoch 10 - iter 20/48 - loss 0.04114819 - time (sec): 9.86 - samples/sec: 1431.18 - lr: 0.000002 - momentum: 0.000000
2024-03-26 11:49:51,170 epoch 10 - iter 24/48 - loss 0.04045979 - time (sec): 11.61 - samples/sec: 1457.39 - lr: 0.000002 - momentum: 0.000000
2024-03-26 11:49:52,116 epoch 10 - iter 28/48 - loss 0.03995345 - time (sec): 12.55 - samples/sec: 1522.22 - lr: 0.000002 - momentum: 0.000000
2024-03-26 11:49:54,497 epoch 10 - iter 32/48 - loss 0.03875589 - time (sec): 14.93 - samples/sec: 1489.79 - lr: 0.000001 - momentum: 0.000000
2024-03-26 11:49:57,098 epoch 10 - iter 36/48 - loss 0.04305678 - time (sec): 17.53 - samples/sec: 1453.98 - lr: 0.000001 - momentum: 0.000000
2024-03-26 11:49:59,095 epoch 10 - iter 40/48 - loss 0.04511330 - time (sec): 19.53 - samples/sec: 1444.56 - lr: 0.000001 - momentum: 0.000000
2024-03-26 11:50:01,791 epoch 10 - iter 44/48 - loss 0.04429458 - time (sec): 22.23 - samples/sec: 1430.59 - lr: 0.000001 - momentum: 0.000000
2024-03-26 11:50:03,422 epoch 10 - iter 48/48 - loss 0.04405461 - time (sec): 23.86 - samples/sec: 1444.82 - lr: 0.000000 - momentum: 0.000000
2024-03-26 11:50:03,423 ----------------------------------------------------------------------------------------------------
2024-03-26 11:50:03,423 EPOCH 10 done: loss 0.0441 - lr: 0.000000
2024-03-26 11:50:04,381 DEV : loss 0.18572011590003967 - f1-score (micro avg)  0.9186
2024-03-26 11:50:04,707 ----------------------------------------------------------------------------------------------------
2024-03-26 11:50:04,708 Loading model from best epoch ...
2024-03-26 11:50:05,626 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 11:50:06,416 
Results:
- F-score (micro) 0.9084
- F-score (macro) 0.6914
- Accuracy 0.8345

By class:
              precision    recall  f1-score   support

 Unternehmen     0.9144    0.8835    0.8987       266
 Auslagerung     0.8534    0.9116    0.8816       249
         Ort     0.9779    0.9925    0.9852       134
    Software     0.0000    0.0000    0.0000         0

   micro avg     0.9002    0.9168    0.9084       649
   macro avg     0.6864    0.6969    0.6914       649
weighted avg     0.9041    0.9168    0.9100       649

2024-03-26 11:50:06,416 ----------------------------------------------------------------------------------------------------