Hugging Face's logo

stefan-it
/
flair-co-funer-german_bert_base-bs8-e10-lr5e-05-1

Token Classification
Flair
PyTorch
TensorBoard
German
sequence-tagger-model
hetzner
hetzner-gex44
hetzner-gpu
Model card Files Metrics Community
2
flair-co-funer-german_bert_base-bs8-e10-lr5e-05-1
File size: 23,826 Bytes 
e582aa4
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
 
2024-03-26 11:08:21,762 ----------------------------------------------------------------------------------------------------
2024-03-26 11:08:21,762 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 11:08:21,763 ----------------------------------------------------------------------------------------------------
2024-03-26 11:08:21,763 Corpus: 758 train + 94 dev + 96 test sentences
2024-03-26 11:08:21,763 ----------------------------------------------------------------------------------------------------
2024-03-26 11:08:21,763 Train:  758 sentences
2024-03-26 11:08:21,763         (train_with_dev=False, train_with_test=False)
2024-03-26 11:08:21,763 ----------------------------------------------------------------------------------------------------
2024-03-26 11:08:21,763 Training Params:
2024-03-26 11:08:21,763  - learning_rate: "5e-05" 
2024-03-26 11:08:21,763  - mini_batch_size: "8"
2024-03-26 11:08:21,763  - max_epochs: "10"
2024-03-26 11:08:21,763  - shuffle: "True"
2024-03-26 11:08:21,763 ----------------------------------------------------------------------------------------------------
2024-03-26 11:08:21,763 Plugins:
2024-03-26 11:08:21,763  - TensorboardLogger
2024-03-26 11:08:21,763  - LinearScheduler | warmup_fraction: '0.1'
2024-03-26 11:08:21,763 ----------------------------------------------------------------------------------------------------
2024-03-26 11:08:21,763 Final evaluation on model from best epoch (best-model.pt)
2024-03-26 11:08:21,763  - metric: "('micro avg', 'f1-score')"
2024-03-26 11:08:21,763 ----------------------------------------------------------------------------------------------------
2024-03-26 11:08:21,763 Computation:
2024-03-26 11:08:21,763  - compute on device: cuda:0
2024-03-26 11:08:21,763  - embedding storage: none
2024-03-26 11:08:21,763 ----------------------------------------------------------------------------------------------------
2024-03-26 11:08:21,763 Model training base path: "flair-co-funer-german_bert_base-bs8-e10-lr5e-05-1"
2024-03-26 11:08:21,763 ----------------------------------------------------------------------------------------------------
2024-03-26 11:08:21,763 ----------------------------------------------------------------------------------------------------
2024-03-26 11:08:21,763 Logging anything other than scalars to TensorBoard is currently not supported.
2024-03-26 11:08:23,411 epoch 1 - iter 9/95 - loss 3.39183266 - time (sec): 1.65 - samples/sec: 1868.90 - lr: 0.000004 - momentum: 0.000000
2024-03-26 11:08:24,991 epoch 1 - iter 18/95 - loss 3.17654028 - time (sec): 3.23 - samples/sec: 1936.66 - lr: 0.000009 - momentum: 0.000000
2024-03-26 11:08:27,506 epoch 1 - iter 27/95 - loss 2.89741196 - time (sec): 5.74 - samples/sec: 1783.20 - lr: 0.000014 - momentum: 0.000000
2024-03-26 11:08:29,798 epoch 1 - iter 36/95 - loss 2.63684990 - time (sec): 8.03 - samples/sec: 1739.97 - lr: 0.000018 - momentum: 0.000000
2024-03-26 11:08:31,776 epoch 1 - iter 45/95 - loss 2.43099939 - time (sec): 10.01 - samples/sec: 1744.14 - lr: 0.000023 - momentum: 0.000000
2024-03-26 11:08:33,049 epoch 1 - iter 54/95 - loss 2.27969388 - time (sec): 11.29 - samples/sec: 1784.54 - lr: 0.000028 - momentum: 0.000000
2024-03-26 11:08:34,823 epoch 1 - iter 63/95 - loss 2.12336427 - time (sec): 13.06 - samples/sec: 1781.41 - lr: 0.000033 - momentum: 0.000000
2024-03-26 11:08:36,187 epoch 1 - iter 72/95 - loss 1.99792100 - time (sec): 14.42 - samples/sec: 1806.03 - lr: 0.000037 - momentum: 0.000000
2024-03-26 11:08:38,261 epoch 1 - iter 81/95 - loss 1.84728397 - time (sec): 16.50 - samples/sec: 1795.33 - lr: 0.000042 - momentum: 0.000000
2024-03-26 11:08:39,631 epoch 1 - iter 90/95 - loss 1.74305657 - time (sec): 17.87 - samples/sec: 1815.53 - lr: 0.000047 - momentum: 0.000000
2024-03-26 11:08:40,922 ----------------------------------------------------------------------------------------------------
2024-03-26 11:08:40,922 EPOCH 1 done: loss 1.6644 - lr: 0.000047
2024-03-26 11:08:41,790 DEV : loss 0.42400190234184265 - f1-score (micro avg)  0.6944
2024-03-26 11:08:41,791 saving best model
2024-03-26 11:08:42,079 ----------------------------------------------------------------------------------------------------
2024-03-26 11:08:44,282 epoch 2 - iter 9/95 - loss 0.38737862 - time (sec): 2.20 - samples/sec: 1677.01 - lr: 0.000050 - momentum: 0.000000
2024-03-26 11:08:46,057 epoch 2 - iter 18/95 - loss 0.42372755 - time (sec): 3.98 - samples/sec: 1824.46 - lr: 0.000049 - momentum: 0.000000
2024-03-26 11:08:47,935 epoch 2 - iter 27/95 - loss 0.40373047 - time (sec): 5.86 - samples/sec: 1760.63 - lr: 0.000048 - momentum: 0.000000
2024-03-26 11:08:49,750 epoch 2 - iter 36/95 - loss 0.37995884 - time (sec): 7.67 - samples/sec: 1743.35 - lr: 0.000048 - momentum: 0.000000
2024-03-26 11:08:51,700 epoch 2 - iter 45/95 - loss 0.35544085 - time (sec): 9.62 - samples/sec: 1760.43 - lr: 0.000047 - momentum: 0.000000
2024-03-26 11:08:53,971 epoch 2 - iter 54/95 - loss 0.32784977 - time (sec): 11.89 - samples/sec: 1736.30 - lr: 0.000047 - momentum: 0.000000
2024-03-26 11:08:55,341 epoch 2 - iter 63/95 - loss 0.33315303 - time (sec): 13.26 - samples/sec: 1776.34 - lr: 0.000046 - momentum: 0.000000
2024-03-26 11:08:56,724 epoch 2 - iter 72/95 - loss 0.32465532 - time (sec): 14.64 - samples/sec: 1806.46 - lr: 0.000046 - momentum: 0.000000
2024-03-26 11:08:58,560 epoch 2 - iter 81/95 - loss 0.31690141 - time (sec): 16.48 - samples/sec: 1796.14 - lr: 0.000045 - momentum: 0.000000
2024-03-26 11:09:00,242 epoch 2 - iter 90/95 - loss 0.31275220 - time (sec): 18.16 - samples/sec: 1796.89 - lr: 0.000045 - momentum: 0.000000
2024-03-26 11:09:01,174 ----------------------------------------------------------------------------------------------------
2024-03-26 11:09:01,174 EPOCH 2 done: loss 0.3089 - lr: 0.000045
2024-03-26 11:09:02,115 DEV : loss 0.2924253046512604 - f1-score (micro avg)  0.8294
2024-03-26 11:09:02,117 saving best model
2024-03-26 11:09:02,583 ----------------------------------------------------------------------------------------------------
2024-03-26 11:09:04,607 epoch 3 - iter 9/95 - loss 0.27193024 - time (sec): 2.02 - samples/sec: 1659.21 - lr: 0.000044 - momentum: 0.000000
2024-03-26 11:09:06,619 epoch 3 - iter 18/95 - loss 0.23216075 - time (sec): 4.03 - samples/sec: 1668.18 - lr: 0.000043 - momentum: 0.000000
2024-03-26 11:09:08,007 epoch 3 - iter 27/95 - loss 0.21629269 - time (sec): 5.42 - samples/sec: 1764.26 - lr: 0.000043 - momentum: 0.000000
2024-03-26 11:09:10,521 epoch 3 - iter 36/95 - loss 0.20920144 - time (sec): 7.94 - samples/sec: 1703.67 - lr: 0.000042 - momentum: 0.000000
2024-03-26 11:09:12,802 epoch 3 - iter 45/95 - loss 0.19430274 - time (sec): 10.22 - samples/sec: 1737.61 - lr: 0.000042 - momentum: 0.000000
2024-03-26 11:09:14,009 epoch 3 - iter 54/95 - loss 0.18883045 - time (sec): 11.42 - samples/sec: 1793.55 - lr: 0.000041 - momentum: 0.000000
2024-03-26 11:09:16,050 epoch 3 - iter 63/95 - loss 0.17703435 - time (sec): 13.47 - samples/sec: 1769.02 - lr: 0.000041 - momentum: 0.000000
2024-03-26 11:09:17,690 epoch 3 - iter 72/95 - loss 0.16863551 - time (sec): 15.10 - samples/sec: 1778.10 - lr: 0.000040 - momentum: 0.000000
2024-03-26 11:09:19,511 epoch 3 - iter 81/95 - loss 0.17069625 - time (sec): 16.93 - samples/sec: 1767.29 - lr: 0.000040 - momentum: 0.000000
2024-03-26 11:09:21,771 epoch 3 - iter 90/95 - loss 0.16374446 - time (sec): 19.19 - samples/sec: 1736.18 - lr: 0.000039 - momentum: 0.000000
2024-03-26 11:09:22,261 ----------------------------------------------------------------------------------------------------
2024-03-26 11:09:22,261 EPOCH 3 done: loss 0.1647 - lr: 0.000039
2024-03-26 11:09:23,216 DEV : loss 0.24790360033512115 - f1-score (micro avg)  0.8685
2024-03-26 11:09:23,217 saving best model
2024-03-26 11:09:23,699 ----------------------------------------------------------------------------------------------------
2024-03-26 11:09:25,328 epoch 4 - iter 9/95 - loss 0.13145670 - time (sec): 1.63 - samples/sec: 1978.66 - lr: 0.000039 - momentum: 0.000000
2024-03-26 11:09:27,422 epoch 4 - iter 18/95 - loss 0.11936859 - time (sec): 3.72 - samples/sec: 1732.78 - lr: 0.000038 - momentum: 0.000000
2024-03-26 11:09:29,243 epoch 4 - iter 27/95 - loss 0.12783695 - time (sec): 5.54 - samples/sec: 1757.58 - lr: 0.000037 - momentum: 0.000000
2024-03-26 11:09:31,850 epoch 4 - iter 36/95 - loss 0.10550090 - time (sec): 8.15 - samples/sec: 1690.57 - lr: 0.000037 - momentum: 0.000000
2024-03-26 11:09:33,570 epoch 4 - iter 45/95 - loss 0.11156077 - time (sec): 9.87 - samples/sec: 1710.58 - lr: 0.000036 - momentum: 0.000000
2024-03-26 11:09:35,164 epoch 4 - iter 54/95 - loss 0.11735065 - time (sec): 11.46 - samples/sec: 1759.37 - lr: 0.000036 - momentum: 0.000000
2024-03-26 11:09:37,147 epoch 4 - iter 63/95 - loss 0.11856926 - time (sec): 13.45 - samples/sec: 1770.39 - lr: 0.000035 - momentum: 0.000000
2024-03-26 11:09:38,460 epoch 4 - iter 72/95 - loss 0.11769596 - time (sec): 14.76 - samples/sec: 1800.16 - lr: 0.000035 - momentum: 0.000000
2024-03-26 11:09:40,244 epoch 4 - iter 81/95 - loss 0.11529168 - time (sec): 16.54 - samples/sec: 1788.76 - lr: 0.000034 - momentum: 0.000000
2024-03-26 11:09:41,804 epoch 4 - iter 90/95 - loss 0.11252805 - time (sec): 18.10 - samples/sec: 1807.02 - lr: 0.000034 - momentum: 0.000000
2024-03-26 11:09:42,749 ----------------------------------------------------------------------------------------------------
2024-03-26 11:09:42,749 EPOCH 4 done: loss 0.1122 - lr: 0.000034
2024-03-26 11:09:43,783 DEV : loss 0.23124928772449493 - f1-score (micro avg)  0.8929
2024-03-26 11:09:43,784 saving best model
2024-03-26 11:09:44,262 ----------------------------------------------------------------------------------------------------
2024-03-26 11:09:45,949 epoch 5 - iter 9/95 - loss 0.08974360 - time (sec): 1.69 - samples/sec: 1877.38 - lr: 0.000033 - momentum: 0.000000
2024-03-26 11:09:48,225 epoch 5 - iter 18/95 - loss 0.09542597 - time (sec): 3.96 - samples/sec: 1691.57 - lr: 0.000032 - momentum: 0.000000
2024-03-26 11:09:49,848 epoch 5 - iter 27/95 - loss 0.08770892 - time (sec): 5.59 - samples/sec: 1736.78 - lr: 0.000032 - momentum: 0.000000
2024-03-26 11:09:51,589 epoch 5 - iter 36/95 - loss 0.08472340 - time (sec): 7.33 - samples/sec: 1721.93 - lr: 0.000031 - momentum: 0.000000
2024-03-26 11:09:53,361 epoch 5 - iter 45/95 - loss 0.09940044 - time (sec): 9.10 - samples/sec: 1763.27 - lr: 0.000031 - momentum: 0.000000
2024-03-26 11:09:55,038 epoch 5 - iter 54/95 - loss 0.10342254 - time (sec): 10.77 - samples/sec: 1805.05 - lr: 0.000030 - momentum: 0.000000
2024-03-26 11:09:56,931 epoch 5 - iter 63/95 - loss 0.09734399 - time (sec): 12.67 - samples/sec: 1788.49 - lr: 0.000030 - momentum: 0.000000
2024-03-26 11:09:59,168 epoch 5 - iter 72/95 - loss 0.08767242 - time (sec): 14.91 - samples/sec: 1820.62 - lr: 0.000029 - momentum: 0.000000
2024-03-26 11:10:00,467 epoch 5 - iter 81/95 - loss 0.08717347 - time (sec): 16.20 - samples/sec: 1837.73 - lr: 0.000029 - momentum: 0.000000
2024-03-26 11:10:02,677 epoch 5 - iter 90/95 - loss 0.08332544 - time (sec): 18.41 - samples/sec: 1798.30 - lr: 0.000028 - momentum: 0.000000
2024-03-26 11:10:03,321 ----------------------------------------------------------------------------------------------------
2024-03-26 11:10:03,321 EPOCH 5 done: loss 0.0838 - lr: 0.000028
2024-03-26 11:10:04,289 DEV : loss 0.22174565494060516 - f1-score (micro avg)  0.9143
2024-03-26 11:10:04,290 saving best model
2024-03-26 11:10:04,783 ----------------------------------------------------------------------------------------------------
2024-03-26 11:10:06,386 epoch 6 - iter 9/95 - loss 0.02955144 - time (sec): 1.60 - samples/sec: 1803.68 - lr: 0.000027 - momentum: 0.000000
2024-03-26 11:10:08,469 epoch 6 - iter 18/95 - loss 0.05567457 - time (sec): 3.69 - samples/sec: 1779.90 - lr: 0.000027 - momentum: 0.000000
2024-03-26 11:10:10,201 epoch 6 - iter 27/95 - loss 0.05951230 - time (sec): 5.42 - samples/sec: 1814.33 - lr: 0.000026 - momentum: 0.000000
2024-03-26 11:10:11,901 epoch 6 - iter 36/95 - loss 0.05959520 - time (sec): 7.12 - samples/sec: 1780.39 - lr: 0.000026 - momentum: 0.000000
2024-03-26 11:10:13,521 epoch 6 - iter 45/95 - loss 0.06840248 - time (sec): 8.74 - samples/sec: 1799.66 - lr: 0.000025 - momentum: 0.000000
2024-03-26 11:10:15,611 epoch 6 - iter 54/95 - loss 0.06861246 - time (sec): 10.83 - samples/sec: 1775.60 - lr: 0.000025 - momentum: 0.000000
2024-03-26 11:10:17,257 epoch 6 - iter 63/95 - loss 0.06935697 - time (sec): 12.47 - samples/sec: 1772.26 - lr: 0.000024 - momentum: 0.000000
2024-03-26 11:10:20,161 epoch 6 - iter 72/95 - loss 0.06391025 - time (sec): 15.38 - samples/sec: 1734.12 - lr: 0.000024 - momentum: 0.000000
2024-03-26 11:10:22,121 epoch 6 - iter 81/95 - loss 0.06207131 - time (sec): 17.34 - samples/sec: 1736.90 - lr: 0.000023 - momentum: 0.000000
2024-03-26 11:10:23,856 epoch 6 - iter 90/95 - loss 0.06175948 - time (sec): 19.07 - samples/sec: 1730.88 - lr: 0.000023 - momentum: 0.000000
2024-03-26 11:10:24,466 ----------------------------------------------------------------------------------------------------
2024-03-26 11:10:24,466 EPOCH 6 done: loss 0.0608 - lr: 0.000023
2024-03-26 11:10:25,421 DEV : loss 0.21966709196567535 - f1-score (micro avg)  0.9118
2024-03-26 11:10:25,422 ----------------------------------------------------------------------------------------------------
2024-03-26 11:10:26,783 epoch 7 - iter 9/95 - loss 0.06461415 - time (sec): 1.36 - samples/sec: 2172.97 - lr: 0.000022 - momentum: 0.000000
2024-03-26 11:10:28,422 epoch 7 - iter 18/95 - loss 0.05971675 - time (sec): 3.00 - samples/sec: 1957.99 - lr: 0.000021 - momentum: 0.000000
2024-03-26 11:10:30,235 epoch 7 - iter 27/95 - loss 0.06004325 - time (sec): 4.81 - samples/sec: 1899.48 - lr: 0.000021 - momentum: 0.000000
2024-03-26 11:10:32,162 epoch 7 - iter 36/95 - loss 0.05314726 - time (sec): 6.74 - samples/sec: 1856.34 - lr: 0.000020 - momentum: 0.000000
2024-03-26 11:10:34,574 epoch 7 - iter 45/95 - loss 0.05080729 - time (sec): 9.15 - samples/sec: 1790.78 - lr: 0.000020 - momentum: 0.000000
2024-03-26 11:10:35,577 epoch 7 - iter 54/95 - loss 0.04997814 - time (sec): 10.15 - samples/sec: 1865.64 - lr: 0.000019 - momentum: 0.000000
2024-03-26 11:10:37,466 epoch 7 - iter 63/95 - loss 0.04724952 - time (sec): 12.04 - samples/sec: 1867.82 - lr: 0.000019 - momentum: 0.000000
2024-03-26 11:10:39,426 epoch 7 - iter 72/95 - loss 0.04535015 - time (sec): 14.00 - samples/sec: 1829.09 - lr: 0.000018 - momentum: 0.000000
2024-03-26 11:10:41,457 epoch 7 - iter 81/95 - loss 0.04623420 - time (sec): 16.03 - samples/sec: 1820.89 - lr: 0.000018 - momentum: 0.000000
2024-03-26 11:10:43,465 epoch 7 - iter 90/95 - loss 0.04648151 - time (sec): 18.04 - samples/sec: 1822.31 - lr: 0.000017 - momentum: 0.000000
2024-03-26 11:10:44,329 ----------------------------------------------------------------------------------------------------
2024-03-26 11:10:44,329 EPOCH 7 done: loss 0.0457 - lr: 0.000017
2024-03-26 11:10:45,284 DEV : loss 0.19319681823253632 - f1-score (micro avg)  0.9232
2024-03-26 11:10:45,286 saving best model
2024-03-26 11:10:45,770 ----------------------------------------------------------------------------------------------------
2024-03-26 11:10:47,432 epoch 8 - iter 9/95 - loss 0.02840275 - time (sec): 1.66 - samples/sec: 1799.98 - lr: 0.000016 - momentum: 0.000000
2024-03-26 11:10:49,486 epoch 8 - iter 18/95 - loss 0.02912869 - time (sec): 3.72 - samples/sec: 1636.62 - lr: 0.000016 - momentum: 0.000000
2024-03-26 11:10:51,083 epoch 8 - iter 27/95 - loss 0.02837540 - time (sec): 5.31 - samples/sec: 1732.91 - lr: 0.000015 - momentum: 0.000000
2024-03-26 11:10:52,851 epoch 8 - iter 36/95 - loss 0.03120678 - time (sec): 7.08 - samples/sec: 1778.30 - lr: 0.000015 - momentum: 0.000000
2024-03-26 11:10:55,272 epoch 8 - iter 45/95 - loss 0.02729828 - time (sec): 9.50 - samples/sec: 1749.70 - lr: 0.000014 - momentum: 0.000000
2024-03-26 11:10:57,672 epoch 8 - iter 54/95 - loss 0.03232762 - time (sec): 11.90 - samples/sec: 1749.78 - lr: 0.000014 - momentum: 0.000000
2024-03-26 11:10:59,655 epoch 8 - iter 63/95 - loss 0.03291277 - time (sec): 13.88 - samples/sec: 1758.50 - lr: 0.000013 - momentum: 0.000000
2024-03-26 11:11:00,752 epoch 8 - iter 72/95 - loss 0.03264611 - time (sec): 14.98 - samples/sec: 1792.38 - lr: 0.000013 - momentum: 0.000000
2024-03-26 11:11:02,462 epoch 8 - iter 81/95 - loss 0.03256358 - time (sec): 16.69 - samples/sec: 1778.20 - lr: 0.000012 - momentum: 0.000000
2024-03-26 11:11:03,888 epoch 8 - iter 90/95 - loss 0.03270593 - time (sec): 18.12 - samples/sec: 1791.63 - lr: 0.000012 - momentum: 0.000000
2024-03-26 11:11:05,149 ----------------------------------------------------------------------------------------------------
2024-03-26 11:11:05,149 EPOCH 8 done: loss 0.0342 - lr: 0.000012
2024-03-26 11:11:06,105 DEV : loss 0.21725119650363922 - f1-score (micro avg)  0.9285
2024-03-26 11:11:06,106 saving best model
2024-03-26 11:11:06,583 ----------------------------------------------------------------------------------------------------
2024-03-26 11:11:08,416 epoch 9 - iter 9/95 - loss 0.01917477 - time (sec): 1.83 - samples/sec: 1895.30 - lr: 0.000011 - momentum: 0.000000
2024-03-26 11:11:10,459 epoch 9 - iter 18/95 - loss 0.01353751 - time (sec): 3.88 - samples/sec: 1743.54 - lr: 0.000010 - momentum: 0.000000
2024-03-26 11:11:12,350 epoch 9 - iter 27/95 - loss 0.01386675 - time (sec): 5.77 - samples/sec: 1703.87 - lr: 0.000010 - momentum: 0.000000
2024-03-26 11:11:14,294 epoch 9 - iter 36/95 - loss 0.02163211 - time (sec): 7.71 - samples/sec: 1746.03 - lr: 0.000009 - momentum: 0.000000
2024-03-26 11:11:16,225 epoch 9 - iter 45/95 - loss 0.02236524 - time (sec): 9.64 - samples/sec: 1729.33 - lr: 0.000009 - momentum: 0.000000
2024-03-26 11:11:18,110 epoch 9 - iter 54/95 - loss 0.02296002 - time (sec): 11.53 - samples/sec: 1764.80 - lr: 0.000008 - momentum: 0.000000
2024-03-26 11:11:20,041 epoch 9 - iter 63/95 - loss 0.02435314 - time (sec): 13.46 - samples/sec: 1764.55 - lr: 0.000008 - momentum: 0.000000
2024-03-26 11:11:21,707 epoch 9 - iter 72/95 - loss 0.02761300 - time (sec): 15.12 - samples/sec: 1770.25 - lr: 0.000007 - momentum: 0.000000
2024-03-26 11:11:23,465 epoch 9 - iter 81/95 - loss 0.02828915 - time (sec): 16.88 - samples/sec: 1761.41 - lr: 0.000007 - momentum: 0.000000
2024-03-26 11:11:25,270 epoch 9 - iter 90/95 - loss 0.02665807 - time (sec): 18.69 - samples/sec: 1778.92 - lr: 0.000006 - momentum: 0.000000
2024-03-26 11:11:25,788 ----------------------------------------------------------------------------------------------------
2024-03-26 11:11:25,788 EPOCH 9 done: loss 0.0283 - lr: 0.000006
2024-03-26 11:11:26,755 DEV : loss 0.22617875039577484 - f1-score (micro avg)  0.9332
2024-03-26 11:11:26,756 saving best model
2024-03-26 11:11:27,237 ----------------------------------------------------------------------------------------------------
2024-03-26 11:11:28,752 epoch 10 - iter 9/95 - loss 0.01044236 - time (sec): 1.51 - samples/sec: 1834.26 - lr: 0.000005 - momentum: 0.000000
2024-03-26 11:11:30,640 epoch 10 - iter 18/95 - loss 0.01347928 - time (sec): 3.40 - samples/sec: 1777.08 - lr: 0.000005 - momentum: 0.000000
2024-03-26 11:11:32,812 epoch 10 - iter 27/95 - loss 0.02026926 - time (sec): 5.57 - samples/sec: 1734.57 - lr: 0.000004 - momentum: 0.000000
2024-03-26 11:11:34,757 epoch 10 - iter 36/95 - loss 0.02504874 - time (sec): 7.52 - samples/sec: 1744.52 - lr: 0.000004 - momentum: 0.000000
2024-03-26 11:11:35,961 epoch 10 - iter 45/95 - loss 0.02354637 - time (sec): 8.72 - samples/sec: 1796.91 - lr: 0.000003 - momentum: 0.000000
2024-03-26 11:11:37,938 epoch 10 - iter 54/95 - loss 0.02573327 - time (sec): 10.70 - samples/sec: 1779.53 - lr: 0.000003 - momentum: 0.000000
2024-03-26 11:11:39,373 epoch 10 - iter 63/95 - loss 0.02906790 - time (sec): 12.13 - samples/sec: 1790.70 - lr: 0.000002 - momentum: 0.000000
2024-03-26 11:11:41,670 epoch 10 - iter 72/95 - loss 0.02563533 - time (sec): 14.43 - samples/sec: 1774.89 - lr: 0.000002 - momentum: 0.000000
2024-03-26 11:11:44,044 epoch 10 - iter 81/95 - loss 0.02653331 - time (sec): 16.81 - samples/sec: 1756.99 - lr: 0.000001 - momentum: 0.000000
2024-03-26 11:11:45,898 epoch 10 - iter 90/95 - loss 0.02469707 - time (sec): 18.66 - samples/sec: 1754.34 - lr: 0.000001 - momentum: 0.000000
2024-03-26 11:11:46,944 ----------------------------------------------------------------------------------------------------
2024-03-26 11:11:46,944 EPOCH 10 done: loss 0.0236 - lr: 0.000001
2024-03-26 11:11:47,914 DEV : loss 0.22828947007656097 - f1-score (micro avg)  0.9346
2024-03-26 11:11:47,915 saving best model
2024-03-26 11:11:48,710 ----------------------------------------------------------------------------------------------------
2024-03-26 11:11:48,711 Loading model from best epoch ...
2024-03-26 11:11:49,658 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 11:11:50,433 
Results:
- F-score (micro) 0.9141
- F-score (macro) 0.6954
- Accuracy 0.8442

By class:
              precision    recall  f1-score   support

 Unternehmen     0.9008    0.8872    0.8939       266
 Auslagerung     0.8867    0.9116    0.8990       249
         Ort     0.9852    0.9925    0.9888       134
    Software     0.0000    0.0000    0.0000         0

   micro avg     0.9099    0.9183    0.9141       649
   macro avg     0.6932    0.6979    0.6954       649
weighted avg     0.9128    0.9183    0.9155       649

2024-03-26 11:11:50,434 ----------------------------------------------------------------------------------------------------