Hugging Face's logo

stefan-it
/
flair-co-funer-german_bert_base-bs8-e10-lr3e-05-3

Token Classification
Flair
PyTorch
TensorBoard
German
sequence-tagger-model
hetzner
hetzner-gex44
hetzner-gpu
Model card Files Metrics Community
2
flair-co-funer-german_bert_base-bs8-e10-lr3e-05-3
File size: 23,859 Bytes 
8145aa3
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
 
2024-03-26 11:38:10,623 ----------------------------------------------------------------------------------------------------
2024-03-26 11:38:10,623 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 11:38:10,623 ----------------------------------------------------------------------------------------------------
2024-03-26 11:38:10,623 Corpus: 758 train + 94 dev + 96 test sentences
2024-03-26 11:38:10,624 ----------------------------------------------------------------------------------------------------
2024-03-26 11:38:10,624 Train:  758 sentences
2024-03-26 11:38:10,624         (train_with_dev=False, train_with_test=False)
2024-03-26 11:38:10,624 ----------------------------------------------------------------------------------------------------
2024-03-26 11:38:10,624 Training Params:
2024-03-26 11:38:10,624  - learning_rate: "3e-05" 
2024-03-26 11:38:10,624  - mini_batch_size: "8"
2024-03-26 11:38:10,624  - max_epochs: "10"
2024-03-26 11:38:10,624  - shuffle: "True"
2024-03-26 11:38:10,624 ----------------------------------------------------------------------------------------------------
2024-03-26 11:38:10,624 Plugins:
2024-03-26 11:38:10,624  - TensorboardLogger
2024-03-26 11:38:10,624  - LinearScheduler | warmup_fraction: '0.1'
2024-03-26 11:38:10,624 ----------------------------------------------------------------------------------------------------
2024-03-26 11:38:10,624 Final evaluation on model from best epoch (best-model.pt)
2024-03-26 11:38:10,624  - metric: "('micro avg', 'f1-score')"
2024-03-26 11:38:10,624 ----------------------------------------------------------------------------------------------------
2024-03-26 11:38:10,624 Computation:
2024-03-26 11:38:10,624  - compute on device: cuda:0
2024-03-26 11:38:10,624  - embedding storage: none
2024-03-26 11:38:10,624 ----------------------------------------------------------------------------------------------------
2024-03-26 11:38:10,624 Model training base path: "flair-co-funer-german_bert_base-bs8-e10-lr3e-05-3"
2024-03-26 11:38:10,624 ----------------------------------------------------------------------------------------------------
2024-03-26 11:38:10,624 ----------------------------------------------------------------------------------------------------
2024-03-26 11:38:10,624 Logging anything other than scalars to TensorBoard is currently not supported.
2024-03-26 11:38:12,056 epoch 1 - iter 9/95 - loss 3.00468469 - time (sec): 1.43 - samples/sec: 2227.30 - lr: 0.000003 - momentum: 0.000000
2024-03-26 11:38:13,987 epoch 1 - iter 18/95 - loss 2.92881161 - time (sec): 3.36 - samples/sec: 1877.95 - lr: 0.000005 - momentum: 0.000000
2024-03-26 11:38:15,931 epoch 1 - iter 27/95 - loss 2.76773785 - time (sec): 5.31 - samples/sec: 1861.61 - lr: 0.000008 - momentum: 0.000000
2024-03-26 11:38:17,338 epoch 1 - iter 36/95 - loss 2.60379851 - time (sec): 6.71 - samples/sec: 1884.54 - lr: 0.000011 - momentum: 0.000000
2024-03-26 11:38:19,280 epoch 1 - iter 45/95 - loss 2.47709620 - time (sec): 8.66 - samples/sec: 1874.16 - lr: 0.000014 - momentum: 0.000000
2024-03-26 11:38:20,693 epoch 1 - iter 54/95 - loss 2.35481628 - time (sec): 10.07 - samples/sec: 1895.68 - lr: 0.000017 - momentum: 0.000000
2024-03-26 11:38:21,973 epoch 1 - iter 63/95 - loss 2.24785391 - time (sec): 11.35 - samples/sec: 1924.02 - lr: 0.000020 - momentum: 0.000000
2024-03-26 11:38:23,942 epoch 1 - iter 72/95 - loss 2.09969141 - time (sec): 13.32 - samples/sec: 1916.20 - lr: 0.000022 - momentum: 0.000000
2024-03-26 11:38:25,948 epoch 1 - iter 81/95 - loss 1.95569193 - time (sec): 15.32 - samples/sec: 1905.21 - lr: 0.000025 - momentum: 0.000000
2024-03-26 11:38:27,481 epoch 1 - iter 90/95 - loss 1.84081212 - time (sec): 16.86 - samples/sec: 1927.03 - lr: 0.000028 - momentum: 0.000000
2024-03-26 11:38:28,567 ----------------------------------------------------------------------------------------------------
2024-03-26 11:38:28,567 EPOCH 1 done: loss 1.7687 - lr: 0.000028
2024-03-26 11:38:29,421 DEV : loss 0.5348557829856873 - f1-score (micro avg)  0.6468
2024-03-26 11:38:29,423 saving best model
2024-03-26 11:38:29,692 ----------------------------------------------------------------------------------------------------
2024-03-26 11:38:31,120 epoch 2 - iter 9/95 - loss 0.56636333 - time (sec): 1.43 - samples/sec: 1919.48 - lr: 0.000030 - momentum: 0.000000
2024-03-26 11:38:33,005 epoch 2 - iter 18/95 - loss 0.48310807 - time (sec): 3.31 - samples/sec: 1844.25 - lr: 0.000029 - momentum: 0.000000
2024-03-26 11:38:34,210 epoch 2 - iter 27/95 - loss 0.47598804 - time (sec): 4.52 - samples/sec: 1899.34 - lr: 0.000029 - momentum: 0.000000
2024-03-26 11:38:36,520 epoch 2 - iter 36/95 - loss 0.45006817 - time (sec): 6.83 - samples/sec: 1857.02 - lr: 0.000029 - momentum: 0.000000
2024-03-26 11:38:38,497 epoch 2 - iter 45/95 - loss 0.43584834 - time (sec): 8.80 - samples/sec: 1867.64 - lr: 0.000028 - momentum: 0.000000
2024-03-26 11:38:40,759 epoch 2 - iter 54/95 - loss 0.42240852 - time (sec): 11.07 - samples/sec: 1833.29 - lr: 0.000028 - momentum: 0.000000
2024-03-26 11:38:42,823 epoch 2 - iter 63/95 - loss 0.40358698 - time (sec): 13.13 - samples/sec: 1790.32 - lr: 0.000028 - momentum: 0.000000
2024-03-26 11:38:44,393 epoch 2 - iter 72/95 - loss 0.40170925 - time (sec): 14.70 - samples/sec: 1797.04 - lr: 0.000028 - momentum: 0.000000
2024-03-26 11:38:45,876 epoch 2 - iter 81/95 - loss 0.40641165 - time (sec): 16.18 - samples/sec: 1820.31 - lr: 0.000027 - momentum: 0.000000
2024-03-26 11:38:48,250 epoch 2 - iter 90/95 - loss 0.39195044 - time (sec): 18.56 - samples/sec: 1787.46 - lr: 0.000027 - momentum: 0.000000
2024-03-26 11:38:48,909 ----------------------------------------------------------------------------------------------------
2024-03-26 11:38:48,909 EPOCH 2 done: loss 0.3885 - lr: 0.000027
2024-03-26 11:38:49,844 DEV : loss 0.2869340181350708 - f1-score (micro avg)  0.8096
2024-03-26 11:38:49,845 saving best model
2024-03-26 11:38:50,309 ----------------------------------------------------------------------------------------------------
2024-03-26 11:38:51,988 epoch 3 - iter 9/95 - loss 0.21703660 - time (sec): 1.68 - samples/sec: 1779.77 - lr: 0.000026 - momentum: 0.000000
2024-03-26 11:38:53,823 epoch 3 - iter 18/95 - loss 0.20441662 - time (sec): 3.51 - samples/sec: 1802.25 - lr: 0.000026 - momentum: 0.000000
2024-03-26 11:38:55,060 epoch 3 - iter 27/95 - loss 0.21475983 - time (sec): 4.75 - samples/sec: 1966.73 - lr: 0.000026 - momentum: 0.000000
2024-03-26 11:38:56,653 epoch 3 - iter 36/95 - loss 0.20563935 - time (sec): 6.34 - samples/sec: 1958.60 - lr: 0.000025 - momentum: 0.000000
2024-03-26 11:38:58,141 epoch 3 - iter 45/95 - loss 0.21041674 - time (sec): 7.83 - samples/sec: 1957.44 - lr: 0.000025 - momentum: 0.000000
2024-03-26 11:39:00,188 epoch 3 - iter 54/95 - loss 0.20554863 - time (sec): 9.88 - samples/sec: 1913.23 - lr: 0.000025 - momentum: 0.000000
2024-03-26 11:39:02,236 epoch 3 - iter 63/95 - loss 0.20061276 - time (sec): 11.93 - samples/sec: 1866.68 - lr: 0.000025 - momentum: 0.000000
2024-03-26 11:39:04,124 epoch 3 - iter 72/95 - loss 0.20369030 - time (sec): 13.81 - samples/sec: 1851.62 - lr: 0.000024 - momentum: 0.000000
2024-03-26 11:39:06,179 epoch 3 - iter 81/95 - loss 0.19527871 - time (sec): 15.87 - samples/sec: 1826.89 - lr: 0.000024 - momentum: 0.000000
2024-03-26 11:39:08,208 epoch 3 - iter 90/95 - loss 0.20474082 - time (sec): 17.90 - samples/sec: 1826.75 - lr: 0.000024 - momentum: 0.000000
2024-03-26 11:39:09,343 ----------------------------------------------------------------------------------------------------
2024-03-26 11:39:09,343 EPOCH 3 done: loss 0.2006 - lr: 0.000024
2024-03-26 11:39:10,283 DEV : loss 0.22436510026454926 - f1-score (micro avg)  0.8601
2024-03-26 11:39:10,284 saving best model
2024-03-26 11:39:10,721 ----------------------------------------------------------------------------------------------------
2024-03-26 11:39:12,045 epoch 4 - iter 9/95 - loss 0.15904831 - time (sec): 1.32 - samples/sec: 2097.94 - lr: 0.000023 - momentum: 0.000000
2024-03-26 11:39:13,966 epoch 4 - iter 18/95 - loss 0.13585153 - time (sec): 3.24 - samples/sec: 1894.27 - lr: 0.000023 - momentum: 0.000000
2024-03-26 11:39:16,032 epoch 4 - iter 27/95 - loss 0.14408454 - time (sec): 5.31 - samples/sec: 1818.35 - lr: 0.000022 - momentum: 0.000000
2024-03-26 11:39:17,579 epoch 4 - iter 36/95 - loss 0.13786917 - time (sec): 6.86 - samples/sec: 1829.31 - lr: 0.000022 - momentum: 0.000000
2024-03-26 11:39:20,120 epoch 4 - iter 45/95 - loss 0.13153200 - time (sec): 9.40 - samples/sec: 1758.29 - lr: 0.000022 - momentum: 0.000000
2024-03-26 11:39:22,024 epoch 4 - iter 54/95 - loss 0.12653534 - time (sec): 11.30 - samples/sec: 1746.17 - lr: 0.000022 - momentum: 0.000000
2024-03-26 11:39:24,049 epoch 4 - iter 63/95 - loss 0.12344140 - time (sec): 13.33 - samples/sec: 1725.65 - lr: 0.000021 - momentum: 0.000000
2024-03-26 11:39:25,973 epoch 4 - iter 72/95 - loss 0.12630286 - time (sec): 15.25 - samples/sec: 1744.48 - lr: 0.000021 - momentum: 0.000000
2024-03-26 11:39:28,035 epoch 4 - iter 81/95 - loss 0.13168865 - time (sec): 17.31 - samples/sec: 1746.30 - lr: 0.000021 - momentum: 0.000000
2024-03-26 11:39:29,038 epoch 4 - iter 90/95 - loss 0.13058662 - time (sec): 18.32 - samples/sec: 1785.53 - lr: 0.000020 - momentum: 0.000000
2024-03-26 11:39:30,083 ----------------------------------------------------------------------------------------------------
2024-03-26 11:39:30,083 EPOCH 4 done: loss 0.1294 - lr: 0.000020
2024-03-26 11:39:31,013 DEV : loss 0.19798463582992554 - f1-score (micro avg)  0.8867
2024-03-26 11:39:31,014 saving best model
2024-03-26 11:39:31,447 ----------------------------------------------------------------------------------------------------
2024-03-26 11:39:33,430 epoch 5 - iter 9/95 - loss 0.10970783 - time (sec): 1.98 - samples/sec: 1733.87 - lr: 0.000020 - momentum: 0.000000
2024-03-26 11:39:34,886 epoch 5 - iter 18/95 - loss 0.10381087 - time (sec): 3.44 - samples/sec: 1820.09 - lr: 0.000019 - momentum: 0.000000
2024-03-26 11:39:36,269 epoch 5 - iter 27/95 - loss 0.10324233 - time (sec): 4.82 - samples/sec: 1873.54 - lr: 0.000019 - momentum: 0.000000
2024-03-26 11:39:38,219 epoch 5 - iter 36/95 - loss 0.10862119 - time (sec): 6.77 - samples/sec: 1812.61 - lr: 0.000019 - momentum: 0.000000
2024-03-26 11:39:40,531 epoch 5 - iter 45/95 - loss 0.10336276 - time (sec): 9.08 - samples/sec: 1792.35 - lr: 0.000019 - momentum: 0.000000
2024-03-26 11:39:43,050 epoch 5 - iter 54/95 - loss 0.09777919 - time (sec): 11.60 - samples/sec: 1750.74 - lr: 0.000018 - momentum: 0.000000
2024-03-26 11:39:44,772 epoch 5 - iter 63/95 - loss 0.09605003 - time (sec): 13.32 - samples/sec: 1743.65 - lr: 0.000018 - momentum: 0.000000
2024-03-26 11:39:46,623 epoch 5 - iter 72/95 - loss 0.09405670 - time (sec): 15.17 - samples/sec: 1742.78 - lr: 0.000018 - momentum: 0.000000
2024-03-26 11:39:48,974 epoch 5 - iter 81/95 - loss 0.09408093 - time (sec): 17.53 - samples/sec: 1721.72 - lr: 0.000017 - momentum: 0.000000
2024-03-26 11:39:50,415 epoch 5 - iter 90/95 - loss 0.09570225 - time (sec): 18.97 - samples/sec: 1737.72 - lr: 0.000017 - momentum: 0.000000
2024-03-26 11:39:51,218 ----------------------------------------------------------------------------------------------------
2024-03-26 11:39:51,218 EPOCH 5 done: loss 0.0938 - lr: 0.000017
2024-03-26 11:39:52,179 DEV : loss 0.1974947303533554 - f1-score (micro avg)  0.9042
2024-03-26 11:39:52,180 saving best model
2024-03-26 11:39:52,617 ----------------------------------------------------------------------------------------------------
2024-03-26 11:39:54,589 epoch 6 - iter 9/95 - loss 0.07217050 - time (sec): 1.97 - samples/sec: 1769.07 - lr: 0.000016 - momentum: 0.000000
2024-03-26 11:39:56,183 epoch 6 - iter 18/95 - loss 0.07321513 - time (sec): 3.57 - samples/sec: 1783.75 - lr: 0.000016 - momentum: 0.000000
2024-03-26 11:39:58,140 epoch 6 - iter 27/95 - loss 0.07237570 - time (sec): 5.52 - samples/sec: 1790.40 - lr: 0.000016 - momentum: 0.000000
2024-03-26 11:39:59,762 epoch 6 - iter 36/95 - loss 0.07303326 - time (sec): 7.14 - samples/sec: 1784.81 - lr: 0.000016 - momentum: 0.000000
2024-03-26 11:40:01,242 epoch 6 - iter 45/95 - loss 0.07181887 - time (sec): 8.62 - samples/sec: 1823.24 - lr: 0.000015 - momentum: 0.000000
2024-03-26 11:40:02,728 epoch 6 - iter 54/95 - loss 0.06802721 - time (sec): 10.11 - samples/sec: 1821.72 - lr: 0.000015 - momentum: 0.000000
2024-03-26 11:40:04,048 epoch 6 - iter 63/95 - loss 0.06640286 - time (sec): 11.43 - samples/sec: 1880.83 - lr: 0.000015 - momentum: 0.000000
2024-03-26 11:40:06,394 epoch 6 - iter 72/95 - loss 0.07238628 - time (sec): 13.78 - samples/sec: 1843.19 - lr: 0.000014 - momentum: 0.000000
2024-03-26 11:40:08,029 epoch 6 - iter 81/95 - loss 0.06982111 - time (sec): 15.41 - samples/sec: 1859.26 - lr: 0.000014 - momentum: 0.000000
2024-03-26 11:40:09,787 epoch 6 - iter 90/95 - loss 0.07240808 - time (sec): 17.17 - samples/sec: 1877.24 - lr: 0.000014 - momentum: 0.000000
2024-03-26 11:40:11,168 ----------------------------------------------------------------------------------------------------
2024-03-26 11:40:11,168 EPOCH 6 done: loss 0.0733 - lr: 0.000014
2024-03-26 11:40:12,142 DEV : loss 0.20535056293010712 - f1-score (micro avg)  0.9062
2024-03-26 11:40:12,143 saving best model
2024-03-26 11:40:12,585 ----------------------------------------------------------------------------------------------------
2024-03-26 11:40:14,631 epoch 7 - iter 9/95 - loss 0.06137997 - time (sec): 2.05 - samples/sec: 1553.04 - lr: 0.000013 - momentum: 0.000000
2024-03-26 11:40:16,692 epoch 7 - iter 18/95 - loss 0.04583919 - time (sec): 4.11 - samples/sec: 1596.53 - lr: 0.000013 - momentum: 0.000000
2024-03-26 11:40:18,293 epoch 7 - iter 27/95 - loss 0.04102055 - time (sec): 5.71 - samples/sec: 1714.12 - lr: 0.000013 - momentum: 0.000000
2024-03-26 11:40:20,332 epoch 7 - iter 36/95 - loss 0.04230142 - time (sec): 7.75 - samples/sec: 1701.02 - lr: 0.000012 - momentum: 0.000000
2024-03-26 11:40:22,736 epoch 7 - iter 45/95 - loss 0.04534105 - time (sec): 10.15 - samples/sec: 1708.38 - lr: 0.000012 - momentum: 0.000000
2024-03-26 11:40:24,299 epoch 7 - iter 54/95 - loss 0.04383378 - time (sec): 11.71 - samples/sec: 1717.39 - lr: 0.000012 - momentum: 0.000000
2024-03-26 11:40:26,549 epoch 7 - iter 63/95 - loss 0.04715345 - time (sec): 13.96 - samples/sec: 1724.72 - lr: 0.000011 - momentum: 0.000000
2024-03-26 11:40:28,393 epoch 7 - iter 72/95 - loss 0.05266698 - time (sec): 15.81 - samples/sec: 1731.80 - lr: 0.000011 - momentum: 0.000000
2024-03-26 11:40:29,865 epoch 7 - iter 81/95 - loss 0.04998609 - time (sec): 17.28 - samples/sec: 1743.30 - lr: 0.000011 - momentum: 0.000000
2024-03-26 11:40:31,962 epoch 7 - iter 90/95 - loss 0.05320010 - time (sec): 19.38 - samples/sec: 1719.17 - lr: 0.000010 - momentum: 0.000000
2024-03-26 11:40:32,440 ----------------------------------------------------------------------------------------------------
2024-03-26 11:40:32,441 EPOCH 7 done: loss 0.0533 - lr: 0.000010
2024-03-26 11:40:33,404 DEV : loss 0.1789586842060089 - f1-score (micro avg)  0.9121
2024-03-26 11:40:33,405 saving best model
2024-03-26 11:40:33,834 ----------------------------------------------------------------------------------------------------
2024-03-26 11:40:35,732 epoch 8 - iter 9/95 - loss 0.03467893 - time (sec): 1.90 - samples/sec: 1690.13 - lr: 0.000010 - momentum: 0.000000
2024-03-26 11:40:38,308 epoch 8 - iter 18/95 - loss 0.02789460 - time (sec): 4.47 - samples/sec: 1656.17 - lr: 0.000010 - momentum: 0.000000
2024-03-26 11:40:40,169 epoch 8 - iter 27/95 - loss 0.02442571 - time (sec): 6.33 - samples/sec: 1680.13 - lr: 0.000009 - momentum: 0.000000
2024-03-26 11:40:41,782 epoch 8 - iter 36/95 - loss 0.02717416 - time (sec): 7.95 - samples/sec: 1668.45 - lr: 0.000009 - momentum: 0.000000
2024-03-26 11:40:43,339 epoch 8 - iter 45/95 - loss 0.02552008 - time (sec): 9.50 - samples/sec: 1699.36 - lr: 0.000009 - momentum: 0.000000
2024-03-26 11:40:45,064 epoch 8 - iter 54/95 - loss 0.02682554 - time (sec): 11.23 - samples/sec: 1717.91 - lr: 0.000008 - momentum: 0.000000
2024-03-26 11:40:47,330 epoch 8 - iter 63/95 - loss 0.03473284 - time (sec): 13.50 - samples/sec: 1714.85 - lr: 0.000008 - momentum: 0.000000
2024-03-26 11:40:49,691 epoch 8 - iter 72/95 - loss 0.04004587 - time (sec): 15.86 - samples/sec: 1692.57 - lr: 0.000008 - momentum: 0.000000
2024-03-26 11:40:51,432 epoch 8 - iter 81/95 - loss 0.04672366 - time (sec): 17.60 - samples/sec: 1692.79 - lr: 0.000007 - momentum: 0.000000
2024-03-26 11:40:52,756 epoch 8 - iter 90/95 - loss 0.04625643 - time (sec): 18.92 - samples/sec: 1735.25 - lr: 0.000007 - momentum: 0.000000
2024-03-26 11:40:53,691 ----------------------------------------------------------------------------------------------------
2024-03-26 11:40:53,691 EPOCH 8 done: loss 0.0449 - lr: 0.000007
2024-03-26 11:40:54,627 DEV : loss 0.20146656036376953 - f1-score (micro avg)  0.913
2024-03-26 11:40:54,628 saving best model
2024-03-26 11:40:55,073 ----------------------------------------------------------------------------------------------------
2024-03-26 11:40:57,083 epoch 9 - iter 9/95 - loss 0.01238365 - time (sec): 2.01 - samples/sec: 1755.11 - lr: 0.000007 - momentum: 0.000000
2024-03-26 11:40:58,881 epoch 9 - iter 18/95 - loss 0.02759052 - time (sec): 3.81 - samples/sec: 1758.22 - lr: 0.000006 - momentum: 0.000000
2024-03-26 11:41:00,905 epoch 9 - iter 27/95 - loss 0.02758322 - time (sec): 5.83 - samples/sec: 1752.05 - lr: 0.000006 - momentum: 0.000000
2024-03-26 11:41:02,853 epoch 9 - iter 36/95 - loss 0.02846974 - time (sec): 7.78 - samples/sec: 1746.04 - lr: 0.000006 - momentum: 0.000000
2024-03-26 11:41:05,177 epoch 9 - iter 45/95 - loss 0.02677424 - time (sec): 10.10 - samples/sec: 1677.81 - lr: 0.000005 - momentum: 0.000000
2024-03-26 11:41:07,176 epoch 9 - iter 54/95 - loss 0.03171239 - time (sec): 12.10 - samples/sec: 1668.11 - lr: 0.000005 - momentum: 0.000000
2024-03-26 11:41:09,138 epoch 9 - iter 63/95 - loss 0.03114385 - time (sec): 14.06 - samples/sec: 1678.90 - lr: 0.000005 - momentum: 0.000000
2024-03-26 11:41:11,164 epoch 9 - iter 72/95 - loss 0.03333336 - time (sec): 16.09 - samples/sec: 1674.88 - lr: 0.000004 - momentum: 0.000000
2024-03-26 11:41:12,456 epoch 9 - iter 81/95 - loss 0.03525355 - time (sec): 17.38 - samples/sec: 1698.51 - lr: 0.000004 - momentum: 0.000000
2024-03-26 11:41:13,929 epoch 9 - iter 90/95 - loss 0.03895801 - time (sec): 18.86 - samples/sec: 1718.89 - lr: 0.000004 - momentum: 0.000000
2024-03-26 11:41:14,915 ----------------------------------------------------------------------------------------------------
2024-03-26 11:41:14,915 EPOCH 9 done: loss 0.0384 - lr: 0.000004
2024-03-26 11:41:15,903 DEV : loss 0.21197910606861115 - f1-score (micro avg)  0.92
2024-03-26 11:41:15,904 saving best model
2024-03-26 11:41:16,343 ----------------------------------------------------------------------------------------------------
2024-03-26 11:41:18,547 epoch 10 - iter 9/95 - loss 0.01499061 - time (sec): 2.20 - samples/sec: 1730.24 - lr: 0.000003 - momentum: 0.000000
2024-03-26 11:41:19,860 epoch 10 - iter 18/95 - loss 0.01541849 - time (sec): 3.52 - samples/sec: 1841.02 - lr: 0.000003 - momentum: 0.000000
2024-03-26 11:41:21,230 epoch 10 - iter 27/95 - loss 0.03534289 - time (sec): 4.89 - samples/sec: 1940.76 - lr: 0.000003 - momentum: 0.000000
2024-03-26 11:41:22,652 epoch 10 - iter 36/95 - loss 0.03443000 - time (sec): 6.31 - samples/sec: 1955.29 - lr: 0.000002 - momentum: 0.000000
2024-03-26 11:41:24,582 epoch 10 - iter 45/95 - loss 0.02965463 - time (sec): 8.24 - samples/sec: 1933.09 - lr: 0.000002 - momentum: 0.000000
2024-03-26 11:41:26,272 epoch 10 - iter 54/95 - loss 0.02812516 - time (sec): 9.93 - samples/sec: 1909.42 - lr: 0.000002 - momentum: 0.000000
2024-03-26 11:41:28,862 epoch 10 - iter 63/95 - loss 0.02903030 - time (sec): 12.52 - samples/sec: 1836.54 - lr: 0.000001 - momentum: 0.000000
2024-03-26 11:41:30,168 epoch 10 - iter 72/95 - loss 0.02845424 - time (sec): 13.82 - samples/sec: 1844.54 - lr: 0.000001 - momentum: 0.000000
2024-03-26 11:41:32,604 epoch 10 - iter 81/95 - loss 0.02695920 - time (sec): 16.26 - samples/sec: 1791.40 - lr: 0.000001 - momentum: 0.000000
2024-03-26 11:41:34,895 epoch 10 - iter 90/95 - loss 0.03168106 - time (sec): 18.55 - samples/sec: 1772.73 - lr: 0.000000 - momentum: 0.000000
2024-03-26 11:41:35,996 ----------------------------------------------------------------------------------------------------
2024-03-26 11:41:35,996 EPOCH 10 done: loss 0.0322 - lr: 0.000000
2024-03-26 11:41:36,940 DEV : loss 0.2103956937789917 - f1-score (micro avg)  0.9207
2024-03-26 11:41:36,942 saving best model
2024-03-26 11:41:37,656 ----------------------------------------------------------------------------------------------------
2024-03-26 11:41:37,657 Loading model from best epoch ...
2024-03-26 11:41:38,536 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 11:41:39,309 
Results:
- F-score (micro) 0.9048
- F-score (macro) 0.6892
- Accuracy 0.8285

By class:
              precision    recall  f1-score   support

 Unternehmen     0.8792    0.8759    0.8776       266
 Auslagerung     0.8736    0.9157    0.8941       249
         Ort     0.9779    0.9925    0.9852       134
    Software     0.0000    0.0000    0.0000         0

   micro avg     0.8946    0.9153    0.9048       649
   macro avg     0.6827    0.6960    0.6892       649
weighted avg     0.8974    0.9153    0.9061       649

2024-03-26 11:41:39,309 ----------------------------------------------------------------------------------------------------