Hugging Face's logo

stefan-it
/
flair-co-funer-german_bert_base-bs8-e10-lr5e-05-5

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-lr5e-05-5
File size: 23,779 Bytes 
f5b5899
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
 
2024-03-26 12:16:08,752 ----------------------------------------------------------------------------------------------------
2024-03-26 12:16:08,752 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 12:16:08,752 ----------------------------------------------------------------------------------------------------
2024-03-26 12:16:08,752 Corpus: 758 train + 94 dev + 96 test sentences
2024-03-26 12:16:08,752 ----------------------------------------------------------------------------------------------------
2024-03-26 12:16:08,752 Train:  758 sentences
2024-03-26 12:16:08,752         (train_with_dev=False, train_with_test=False)
2024-03-26 12:16:08,752 ----------------------------------------------------------------------------------------------------
2024-03-26 12:16:08,752 Training Params:
2024-03-26 12:16:08,752  - learning_rate: "5e-05" 
2024-03-26 12:16:08,752  - mini_batch_size: "8"
2024-03-26 12:16:08,752  - max_epochs: "10"
2024-03-26 12:16:08,752  - shuffle: "True"
2024-03-26 12:16:08,752 ----------------------------------------------------------------------------------------------------
2024-03-26 12:16:08,752 Plugins:
2024-03-26 12:16:08,752  - TensorboardLogger
2024-03-26 12:16:08,752  - LinearScheduler | warmup_fraction: '0.1'
2024-03-26 12:16:08,752 ----------------------------------------------------------------------------------------------------
2024-03-26 12:16:08,752 Final evaluation on model from best epoch (best-model.pt)
2024-03-26 12:16:08,752  - metric: "('micro avg', 'f1-score')"
2024-03-26 12:16:08,752 ----------------------------------------------------------------------------------------------------
2024-03-26 12:16:08,752 Computation:
2024-03-26 12:16:08,752  - compute on device: cuda:0
2024-03-26 12:16:08,752  - embedding storage: none
2024-03-26 12:16:08,752 ----------------------------------------------------------------------------------------------------
2024-03-26 12:16:08,752 Model training base path: "flair-co-funer-german_bert_base-bs8-e10-lr5e-05-5"
2024-03-26 12:16:08,752 ----------------------------------------------------------------------------------------------------
2024-03-26 12:16:08,752 ----------------------------------------------------------------------------------------------------
2024-03-26 12:16:08,753 Logging anything other than scalars to TensorBoard is currently not supported.
2024-03-26 12:16:10,658 epoch 1 - iter 9/95 - loss 3.11198415 - time (sec): 1.91 - samples/sec: 1645.31 - lr: 0.000004 - momentum: 0.000000
2024-03-26 12:16:12,565 epoch 1 - iter 18/95 - loss 2.93717664 - time (sec): 3.81 - samples/sec: 1739.78 - lr: 0.000009 - momentum: 0.000000
2024-03-26 12:16:14,989 epoch 1 - iter 27/95 - loss 2.69525162 - time (sec): 6.24 - samples/sec: 1663.01 - lr: 0.000014 - momentum: 0.000000
2024-03-26 12:16:16,504 epoch 1 - iter 36/95 - loss 2.50040802 - time (sec): 7.75 - samples/sec: 1742.82 - lr: 0.000018 - momentum: 0.000000
2024-03-26 12:16:18,693 epoch 1 - iter 45/95 - loss 2.31776846 - time (sec): 9.94 - samples/sec: 1728.92 - lr: 0.000023 - momentum: 0.000000
2024-03-26 12:16:20,306 epoch 1 - iter 54/95 - loss 2.14521065 - time (sec): 11.55 - samples/sec: 1751.33 - lr: 0.000028 - momentum: 0.000000
2024-03-26 12:16:21,980 epoch 1 - iter 63/95 - loss 1.99301232 - time (sec): 13.23 - samples/sec: 1769.35 - lr: 0.000033 - momentum: 0.000000
2024-03-26 12:16:23,916 epoch 1 - iter 72/95 - loss 1.84828066 - time (sec): 15.16 - samples/sec: 1760.81 - lr: 0.000037 - momentum: 0.000000
2024-03-26 12:16:26,005 epoch 1 - iter 81/95 - loss 1.70543233 - time (sec): 17.25 - samples/sec: 1747.54 - lr: 0.000042 - momentum: 0.000000
2024-03-26 12:16:27,656 epoch 1 - iter 90/95 - loss 1.60560457 - time (sec): 18.90 - samples/sec: 1741.84 - lr: 0.000047 - momentum: 0.000000
2024-03-26 12:16:28,437 ----------------------------------------------------------------------------------------------------
2024-03-26 12:16:28,437 EPOCH 1 done: loss 1.5561 - lr: 0.000047
2024-03-26 12:16:29,291 DEV : loss 0.3953661024570465 - f1-score (micro avg)  0.7299
2024-03-26 12:16:29,292 saving best model
2024-03-26 12:16:29,557 ----------------------------------------------------------------------------------------------------
2024-03-26 12:16:31,857 epoch 2 - iter 9/95 - loss 0.42411772 - time (sec): 2.30 - samples/sec: 1659.61 - lr: 0.000050 - momentum: 0.000000
2024-03-26 12:16:33,790 epoch 2 - iter 18/95 - loss 0.40210576 - time (sec): 4.23 - samples/sec: 1652.57 - lr: 0.000049 - momentum: 0.000000
2024-03-26 12:16:36,173 epoch 2 - iter 27/95 - loss 0.37000509 - time (sec): 6.62 - samples/sec: 1616.94 - lr: 0.000048 - momentum: 0.000000
2024-03-26 12:16:37,537 epoch 2 - iter 36/95 - loss 0.36304993 - time (sec): 7.98 - samples/sec: 1731.82 - lr: 0.000048 - momentum: 0.000000
2024-03-26 12:16:39,517 epoch 2 - iter 45/95 - loss 0.34645812 - time (sec): 9.96 - samples/sec: 1694.75 - lr: 0.000047 - momentum: 0.000000
2024-03-26 12:16:40,851 epoch 2 - iter 54/95 - loss 0.34203299 - time (sec): 11.29 - samples/sec: 1740.46 - lr: 0.000047 - momentum: 0.000000
2024-03-26 12:16:42,447 epoch 2 - iter 63/95 - loss 0.33030418 - time (sec): 12.89 - samples/sec: 1755.81 - lr: 0.000046 - momentum: 0.000000
2024-03-26 12:16:44,546 epoch 2 - iter 72/95 - loss 0.32731874 - time (sec): 14.99 - samples/sec: 1748.69 - lr: 0.000046 - momentum: 0.000000
2024-03-26 12:16:46,462 epoch 2 - iter 81/95 - loss 0.33534075 - time (sec): 16.90 - samples/sec: 1750.65 - lr: 0.000045 - momentum: 0.000000
2024-03-26 12:16:48,417 epoch 2 - iter 90/95 - loss 0.32353294 - time (sec): 18.86 - samples/sec: 1753.84 - lr: 0.000045 - momentum: 0.000000
2024-03-26 12:16:48,995 ----------------------------------------------------------------------------------------------------
2024-03-26 12:16:48,995 EPOCH 2 done: loss 0.3240 - lr: 0.000045
2024-03-26 12:16:49,916 DEV : loss 0.2869855463504791 - f1-score (micro avg)  0.8389
2024-03-26 12:16:49,917 saving best model
2024-03-26 12:16:50,346 ----------------------------------------------------------------------------------------------------
2024-03-26 12:16:51,570 epoch 3 - iter 9/95 - loss 0.27417263 - time (sec): 1.22 - samples/sec: 2120.71 - lr: 0.000044 - momentum: 0.000000
2024-03-26 12:16:53,842 epoch 3 - iter 18/95 - loss 0.21977563 - time (sec): 3.49 - samples/sec: 1837.05 - lr: 0.000043 - momentum: 0.000000
2024-03-26 12:16:55,561 epoch 3 - iter 27/95 - loss 0.22232998 - time (sec): 5.21 - samples/sec: 1871.79 - lr: 0.000043 - momentum: 0.000000
2024-03-26 12:16:57,399 epoch 3 - iter 36/95 - loss 0.21205041 - time (sec): 7.05 - samples/sec: 1867.67 - lr: 0.000042 - momentum: 0.000000
2024-03-26 12:16:58,861 epoch 3 - iter 45/95 - loss 0.19668558 - time (sec): 8.51 - samples/sec: 1865.50 - lr: 0.000042 - momentum: 0.000000
2024-03-26 12:17:01,066 epoch 3 - iter 54/95 - loss 0.19132023 - time (sec): 10.72 - samples/sec: 1802.40 - lr: 0.000041 - momentum: 0.000000
2024-03-26 12:17:02,798 epoch 3 - iter 63/95 - loss 0.19343591 - time (sec): 12.45 - samples/sec: 1787.69 - lr: 0.000041 - momentum: 0.000000
2024-03-26 12:17:05,134 epoch 3 - iter 72/95 - loss 0.18645392 - time (sec): 14.79 - samples/sec: 1754.33 - lr: 0.000040 - momentum: 0.000000
2024-03-26 12:17:07,392 epoch 3 - iter 81/95 - loss 0.18633466 - time (sec): 17.04 - samples/sec: 1746.40 - lr: 0.000040 - momentum: 0.000000
2024-03-26 12:17:09,149 epoch 3 - iter 90/95 - loss 0.18210534 - time (sec): 18.80 - samples/sec: 1740.35 - lr: 0.000039 - momentum: 0.000000
2024-03-26 12:17:10,046 ----------------------------------------------------------------------------------------------------
2024-03-26 12:17:10,046 EPOCH 3 done: loss 0.1812 - lr: 0.000039
2024-03-26 12:17:10,972 DEV : loss 0.23773233592510223 - f1-score (micro avg)  0.874
2024-03-26 12:17:10,973 saving best model
2024-03-26 12:17:11,406 ----------------------------------------------------------------------------------------------------
2024-03-26 12:17:14,332 epoch 4 - iter 9/95 - loss 0.09110161 - time (sec): 2.92 - samples/sec: 1458.90 - lr: 0.000039 - momentum: 0.000000
2024-03-26 12:17:15,394 epoch 4 - iter 18/95 - loss 0.11946256 - time (sec): 3.99 - samples/sec: 1669.22 - lr: 0.000038 - momentum: 0.000000
2024-03-26 12:17:18,006 epoch 4 - iter 27/95 - loss 0.11225660 - time (sec): 6.60 - samples/sec: 1612.05 - lr: 0.000037 - momentum: 0.000000
2024-03-26 12:17:20,657 epoch 4 - iter 36/95 - loss 0.10969523 - time (sec): 9.25 - samples/sec: 1568.83 - lr: 0.000037 - momentum: 0.000000
2024-03-26 12:17:22,378 epoch 4 - iter 45/95 - loss 0.10414682 - time (sec): 10.97 - samples/sec: 1605.33 - lr: 0.000036 - momentum: 0.000000
2024-03-26 12:17:24,113 epoch 4 - iter 54/95 - loss 0.10683855 - time (sec): 12.71 - samples/sec: 1621.28 - lr: 0.000036 - momentum: 0.000000
2024-03-26 12:17:26,080 epoch 4 - iter 63/95 - loss 0.10871462 - time (sec): 14.67 - samples/sec: 1646.34 - lr: 0.000035 - momentum: 0.000000
2024-03-26 12:17:27,836 epoch 4 - iter 72/95 - loss 0.11330020 - time (sec): 16.43 - samples/sec: 1689.44 - lr: 0.000035 - momentum: 0.000000
2024-03-26 12:17:28,871 epoch 4 - iter 81/95 - loss 0.11517013 - time (sec): 17.46 - samples/sec: 1730.13 - lr: 0.000034 - momentum: 0.000000
2024-03-26 12:17:30,331 epoch 4 - iter 90/95 - loss 0.11512696 - time (sec): 18.92 - samples/sec: 1753.74 - lr: 0.000034 - momentum: 0.000000
2024-03-26 12:17:30,885 ----------------------------------------------------------------------------------------------------
2024-03-26 12:17:30,885 EPOCH 4 done: loss 0.1159 - lr: 0.000034
2024-03-26 12:17:31,816 DEV : loss 0.1792612224817276 - f1-score (micro avg)  0.8988
2024-03-26 12:17:31,817 saving best model
2024-03-26 12:17:32,249 ----------------------------------------------------------------------------------------------------
2024-03-26 12:17:33,905 epoch 5 - iter 9/95 - loss 0.10687482 - time (sec): 1.65 - samples/sec: 1979.76 - lr: 0.000033 - momentum: 0.000000
2024-03-26 12:17:35,911 epoch 5 - iter 18/95 - loss 0.08268843 - time (sec): 3.66 - samples/sec: 1945.00 - lr: 0.000032 - momentum: 0.000000
2024-03-26 12:17:38,085 epoch 5 - iter 27/95 - loss 0.07045579 - time (sec): 5.83 - samples/sec: 1814.52 - lr: 0.000032 - momentum: 0.000000
2024-03-26 12:17:39,453 epoch 5 - iter 36/95 - loss 0.08169072 - time (sec): 7.20 - samples/sec: 1866.69 - lr: 0.000031 - momentum: 0.000000
2024-03-26 12:17:41,576 epoch 5 - iter 45/95 - loss 0.07912089 - time (sec): 9.33 - samples/sec: 1827.05 - lr: 0.000031 - momentum: 0.000000
2024-03-26 12:17:42,773 epoch 5 - iter 54/95 - loss 0.08168215 - time (sec): 10.52 - samples/sec: 1860.75 - lr: 0.000030 - momentum: 0.000000
2024-03-26 12:17:44,288 epoch 5 - iter 63/95 - loss 0.08776931 - time (sec): 12.04 - samples/sec: 1872.91 - lr: 0.000030 - momentum: 0.000000
2024-03-26 12:17:46,330 epoch 5 - iter 72/95 - loss 0.08910227 - time (sec): 14.08 - samples/sec: 1834.21 - lr: 0.000029 - momentum: 0.000000
2024-03-26 12:17:48,138 epoch 5 - iter 81/95 - loss 0.08562803 - time (sec): 15.89 - samples/sec: 1823.20 - lr: 0.000029 - momentum: 0.000000
2024-03-26 12:17:50,604 epoch 5 - iter 90/95 - loss 0.08505748 - time (sec): 18.35 - samples/sec: 1791.38 - lr: 0.000028 - momentum: 0.000000
2024-03-26 12:17:51,604 ----------------------------------------------------------------------------------------------------
2024-03-26 12:17:51,604 EPOCH 5 done: loss 0.0832 - lr: 0.000028
2024-03-26 12:17:52,535 DEV : loss 0.19528663158416748 - f1-score (micro avg)  0.902
2024-03-26 12:17:52,536 saving best model
2024-03-26 12:17:52,967 ----------------------------------------------------------------------------------------------------
2024-03-26 12:17:54,978 epoch 6 - iter 9/95 - loss 0.07127974 - time (sec): 2.01 - samples/sec: 1623.06 - lr: 0.000027 - momentum: 0.000000
2024-03-26 12:17:57,485 epoch 6 - iter 18/95 - loss 0.07270535 - time (sec): 4.52 - samples/sec: 1641.86 - lr: 0.000027 - momentum: 0.000000
2024-03-26 12:17:58,659 epoch 6 - iter 27/95 - loss 0.08746512 - time (sec): 5.69 - samples/sec: 1737.38 - lr: 0.000026 - momentum: 0.000000
2024-03-26 12:18:00,368 epoch 6 - iter 36/95 - loss 0.07826196 - time (sec): 7.40 - samples/sec: 1743.94 - lr: 0.000026 - momentum: 0.000000
2024-03-26 12:18:02,358 epoch 6 - iter 45/95 - loss 0.07323678 - time (sec): 9.39 - samples/sec: 1740.04 - lr: 0.000025 - momentum: 0.000000
2024-03-26 12:18:04,568 epoch 6 - iter 54/95 - loss 0.06698128 - time (sec): 11.60 - samples/sec: 1706.05 - lr: 0.000025 - momentum: 0.000000
2024-03-26 12:18:06,266 epoch 6 - iter 63/95 - loss 0.06973624 - time (sec): 13.30 - samples/sec: 1727.33 - lr: 0.000024 - momentum: 0.000000
2024-03-26 12:18:07,867 epoch 6 - iter 72/95 - loss 0.07049901 - time (sec): 14.90 - samples/sec: 1748.07 - lr: 0.000024 - momentum: 0.000000
2024-03-26 12:18:09,132 epoch 6 - iter 81/95 - loss 0.06883005 - time (sec): 16.16 - samples/sec: 1778.42 - lr: 0.000023 - momentum: 0.000000
2024-03-26 12:18:11,040 epoch 6 - iter 90/95 - loss 0.06491973 - time (sec): 18.07 - samples/sec: 1778.06 - lr: 0.000023 - momentum: 0.000000
2024-03-26 12:18:12,595 ----------------------------------------------------------------------------------------------------
2024-03-26 12:18:12,595 EPOCH 6 done: loss 0.0621 - lr: 0.000023
2024-03-26 12:18:13,530 DEV : loss 0.21062487363815308 - f1-score (micro avg)  0.9138
2024-03-26 12:18:13,531 saving best model
2024-03-26 12:18:13,967 ----------------------------------------------------------------------------------------------------
2024-03-26 12:18:15,657 epoch 7 - iter 9/95 - loss 0.03135025 - time (sec): 1.69 - samples/sec: 1864.58 - lr: 0.000022 - momentum: 0.000000
2024-03-26 12:18:17,178 epoch 7 - iter 18/95 - loss 0.04612086 - time (sec): 3.21 - samples/sec: 1832.91 - lr: 0.000021 - momentum: 0.000000
2024-03-26 12:18:18,494 epoch 7 - iter 27/95 - loss 0.06246375 - time (sec): 4.53 - samples/sec: 1871.28 - lr: 0.000021 - momentum: 0.000000
2024-03-26 12:18:20,844 epoch 7 - iter 36/95 - loss 0.05586240 - time (sec): 6.87 - samples/sec: 1848.73 - lr: 0.000020 - momentum: 0.000000
2024-03-26 12:18:22,814 epoch 7 - iter 45/95 - loss 0.05942577 - time (sec): 8.85 - samples/sec: 1840.86 - lr: 0.000020 - momentum: 0.000000
2024-03-26 12:18:24,526 epoch 7 - iter 54/95 - loss 0.05670271 - time (sec): 10.56 - samples/sec: 1836.88 - lr: 0.000019 - momentum: 0.000000
2024-03-26 12:18:26,122 epoch 7 - iter 63/95 - loss 0.05494130 - time (sec): 12.15 - samples/sec: 1854.57 - lr: 0.000019 - momentum: 0.000000
2024-03-26 12:18:27,660 epoch 7 - iter 72/95 - loss 0.05485412 - time (sec): 13.69 - samples/sec: 1846.14 - lr: 0.000018 - momentum: 0.000000
2024-03-26 12:18:30,467 epoch 7 - iter 81/95 - loss 0.05275127 - time (sec): 16.50 - samples/sec: 1780.62 - lr: 0.000018 - momentum: 0.000000
2024-03-26 12:18:32,128 epoch 7 - iter 90/95 - loss 0.05187601 - time (sec): 18.16 - samples/sec: 1788.35 - lr: 0.000017 - momentum: 0.000000
2024-03-26 12:18:33,304 ----------------------------------------------------------------------------------------------------
2024-03-26 12:18:33,304 EPOCH 7 done: loss 0.0513 - lr: 0.000017
2024-03-26 12:18:34,241 DEV : loss 0.20660282671451569 - f1-score (micro avg)  0.9272
2024-03-26 12:18:34,242 saving best model
2024-03-26 12:18:34,679 ----------------------------------------------------------------------------------------------------
2024-03-26 12:18:36,879 epoch 8 - iter 9/95 - loss 0.03834922 - time (sec): 2.20 - samples/sec: 1537.49 - lr: 0.000016 - momentum: 0.000000
2024-03-26 12:18:38,447 epoch 8 - iter 18/95 - loss 0.02899613 - time (sec): 3.77 - samples/sec: 1620.37 - lr: 0.000016 - momentum: 0.000000
2024-03-26 12:18:40,508 epoch 8 - iter 27/95 - loss 0.03237573 - time (sec): 5.83 - samples/sec: 1683.99 - lr: 0.000015 - momentum: 0.000000
2024-03-26 12:18:42,512 epoch 8 - iter 36/95 - loss 0.02944665 - time (sec): 7.83 - samples/sec: 1719.11 - lr: 0.000015 - momentum: 0.000000
2024-03-26 12:18:43,936 epoch 8 - iter 45/95 - loss 0.02812948 - time (sec): 9.26 - samples/sec: 1778.26 - lr: 0.000014 - momentum: 0.000000
2024-03-26 12:18:45,426 epoch 8 - iter 54/95 - loss 0.02849860 - time (sec): 10.74 - samples/sec: 1849.54 - lr: 0.000014 - momentum: 0.000000
2024-03-26 12:18:47,076 epoch 8 - iter 63/95 - loss 0.03008751 - time (sec): 12.39 - samples/sec: 1834.71 - lr: 0.000013 - momentum: 0.000000
2024-03-26 12:18:49,219 epoch 8 - iter 72/95 - loss 0.02843088 - time (sec): 14.54 - samples/sec: 1800.89 - lr: 0.000013 - momentum: 0.000000
2024-03-26 12:18:50,819 epoch 8 - iter 81/95 - loss 0.03172865 - time (sec): 16.14 - samples/sec: 1824.66 - lr: 0.000012 - momentum: 0.000000
2024-03-26 12:18:52,888 epoch 8 - iter 90/95 - loss 0.03341049 - time (sec): 18.21 - samples/sec: 1805.85 - lr: 0.000012 - momentum: 0.000000
2024-03-26 12:18:53,534 ----------------------------------------------------------------------------------------------------
2024-03-26 12:18:53,534 EPOCH 8 done: loss 0.0346 - lr: 0.000012
2024-03-26 12:18:54,465 DEV : loss 0.20071536302566528 - f1-score (micro avg)  0.9346
2024-03-26 12:18:54,466 saving best model
2024-03-26 12:18:54,944 ----------------------------------------------------------------------------------------------------
2024-03-26 12:18:57,548 epoch 9 - iter 9/95 - loss 0.01191280 - time (sec): 2.60 - samples/sec: 1656.95 - lr: 0.000011 - momentum: 0.000000
2024-03-26 12:18:59,147 epoch 9 - iter 18/95 - loss 0.01847100 - time (sec): 4.20 - samples/sec: 1721.14 - lr: 0.000010 - momentum: 0.000000
2024-03-26 12:19:01,758 epoch 9 - iter 27/95 - loss 0.02360116 - time (sec): 6.81 - samples/sec: 1659.11 - lr: 0.000010 - momentum: 0.000000
2024-03-26 12:19:03,626 epoch 9 - iter 36/95 - loss 0.02717013 - time (sec): 8.68 - samples/sec: 1669.48 - lr: 0.000009 - momentum: 0.000000
2024-03-26 12:19:04,806 epoch 9 - iter 45/95 - loss 0.02510231 - time (sec): 9.86 - samples/sec: 1729.78 - lr: 0.000009 - momentum: 0.000000
2024-03-26 12:19:06,575 epoch 9 - iter 54/95 - loss 0.02242460 - time (sec): 11.63 - samples/sec: 1725.14 - lr: 0.000008 - momentum: 0.000000
2024-03-26 12:19:07,996 epoch 9 - iter 63/95 - loss 0.02704971 - time (sec): 13.05 - samples/sec: 1769.91 - lr: 0.000008 - momentum: 0.000000
2024-03-26 12:19:09,192 epoch 9 - iter 72/95 - loss 0.02678838 - time (sec): 14.25 - samples/sec: 1818.51 - lr: 0.000007 - momentum: 0.000000
2024-03-26 12:19:10,741 epoch 9 - iter 81/95 - loss 0.02504760 - time (sec): 15.80 - samples/sec: 1818.68 - lr: 0.000007 - momentum: 0.000000
2024-03-26 12:19:13,539 epoch 9 - iter 90/95 - loss 0.02802835 - time (sec): 18.59 - samples/sec: 1773.37 - lr: 0.000006 - momentum: 0.000000
2024-03-26 12:19:14,336 ----------------------------------------------------------------------------------------------------
2024-03-26 12:19:14,336 EPOCH 9 done: loss 0.0276 - lr: 0.000006
2024-03-26 12:19:15,294 DEV : loss 0.2257862389087677 - f1-score (micro avg)  0.9343
2024-03-26 12:19:15,295 ----------------------------------------------------------------------------------------------------
2024-03-26 12:19:17,811 epoch 10 - iter 9/95 - loss 0.02415602 - time (sec): 2.52 - samples/sec: 1604.40 - lr: 0.000005 - momentum: 0.000000
2024-03-26 12:19:19,392 epoch 10 - iter 18/95 - loss 0.02153363 - time (sec): 4.10 - samples/sec: 1702.07 - lr: 0.000005 - momentum: 0.000000
2024-03-26 12:19:21,402 epoch 10 - iter 27/95 - loss 0.01906523 - time (sec): 6.11 - samples/sec: 1650.45 - lr: 0.000004 - momentum: 0.000000
2024-03-26 12:19:23,503 epoch 10 - iter 36/95 - loss 0.01928902 - time (sec): 8.21 - samples/sec: 1662.79 - lr: 0.000004 - momentum: 0.000000
2024-03-26 12:19:25,399 epoch 10 - iter 45/95 - loss 0.01845826 - time (sec): 10.10 - samples/sec: 1678.97 - lr: 0.000003 - momentum: 0.000000
2024-03-26 12:19:26,550 epoch 10 - iter 54/95 - loss 0.01897577 - time (sec): 11.25 - samples/sec: 1740.80 - lr: 0.000003 - momentum: 0.000000
2024-03-26 12:19:28,216 epoch 10 - iter 63/95 - loss 0.02489403 - time (sec): 12.92 - samples/sec: 1761.30 - lr: 0.000002 - momentum: 0.000000
2024-03-26 12:19:30,071 epoch 10 - iter 72/95 - loss 0.02391188 - time (sec): 14.78 - samples/sec: 1750.83 - lr: 0.000002 - momentum: 0.000000
2024-03-26 12:19:31,803 epoch 10 - iter 81/95 - loss 0.02463096 - time (sec): 16.51 - samples/sec: 1758.98 - lr: 0.000001 - momentum: 0.000000
2024-03-26 12:19:34,650 epoch 10 - iter 90/95 - loss 0.02255689 - time (sec): 19.35 - samples/sec: 1722.51 - lr: 0.000001 - momentum: 0.000000
2024-03-26 12:19:35,202 ----------------------------------------------------------------------------------------------------
2024-03-26 12:19:35,202 EPOCH 10 done: loss 0.0221 - lr: 0.000001
2024-03-26 12:19:36,133 DEV : loss 0.22455048561096191 - f1-score (micro avg)  0.9336
2024-03-26 12:19:36,401 ----------------------------------------------------------------------------------------------------
2024-03-26 12:19:36,402 Loading model from best epoch ...
2024-03-26 12:19:37,342 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 12:19:38,103 
Results:
- F-score (micro) 0.9105
- F-score (macro) 0.6913
- Accuracy 0.8368

By class:
              precision    recall  f1-score   support

 Unternehmen     0.9151    0.8910    0.9029       266
 Auslagerung     0.8692    0.9076    0.8880       249
         Ort     0.9635    0.9851    0.9742       134
    Software     0.0000    0.0000    0.0000         0

   micro avg     0.9043    0.9168    0.9105       649
   macro avg     0.6869    0.6959    0.6913       649
weighted avg     0.9075    0.9168    0.9119       649

2024-03-26 12:19:38,103 ----------------------------------------------------------------------------------------------------