Hugging Face's logo

stefan-it
/
flair-co-funer-german_bert_base-bs16-e10-lr5e-05-2

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-lr5e-05-2
File size: 26,612 Bytes 
2122bc8
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
 
2024-03-26 11:16:46,211 ----------------------------------------------------------------------------------------------------
2024-03-26 11:16:46,212 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 11:16:46,212 ----------------------------------------------------------------------------------------------------
2024-03-26 11:16:46,212 Corpus: 758 train + 94 dev + 96 test sentences
2024-03-26 11:16:46,212 ----------------------------------------------------------------------------------------------------
2024-03-26 11:16:46,212 Train:  758 sentences
2024-03-26 11:16:46,212         (train_with_dev=False, train_with_test=False)
2024-03-26 11:16:46,212 ----------------------------------------------------------------------------------------------------
2024-03-26 11:16:46,212 Training Params:
2024-03-26 11:16:46,212  - learning_rate: "5e-05" 
2024-03-26 11:16:46,212  - mini_batch_size: "16"
2024-03-26 11:16:46,212  - max_epochs: "10"
2024-03-26 11:16:46,212  - shuffle: "True"
2024-03-26 11:16:46,212 ----------------------------------------------------------------------------------------------------
2024-03-26 11:16:46,212 Plugins:
2024-03-26 11:16:46,212  - TensorboardLogger
2024-03-26 11:16:46,212  - LinearScheduler | warmup_fraction: '0.1'
2024-03-26 11:16:46,212 ----------------------------------------------------------------------------------------------------
2024-03-26 11:16:46,212 Final evaluation on model from best epoch (best-model.pt)
2024-03-26 11:16:46,212  - metric: "('micro avg', 'f1-score')"
2024-03-26 11:16:46,212 ----------------------------------------------------------------------------------------------------
2024-03-26 11:16:46,212 Computation:
2024-03-26 11:16:46,212  - compute on device: cuda:0
2024-03-26 11:16:46,212  - embedding storage: none
2024-03-26 11:16:46,212 ----------------------------------------------------------------------------------------------------
2024-03-26 11:16:46,212 Model training base path: "flair-co-funer-german_bert_base-bs16-e10-lr5e-05-2"
2024-03-26 11:16:46,212 ----------------------------------------------------------------------------------------------------
2024-03-26 11:16:46,212 ----------------------------------------------------------------------------------------------------
2024-03-26 11:16:46,212 Logging anything other than scalars to TensorBoard is currently not supported.
2024-03-26 11:16:47,993 epoch 1 - iter 4/48 - loss 3.15943466 - time (sec): 1.78 - samples/sec: 1696.29 - lr: 0.000003 - momentum: 0.000000
2024-03-26 11:16:50,199 epoch 1 - iter 8/48 - loss 3.05235430 - time (sec): 3.99 - samples/sec: 1557.38 - lr: 0.000007 - momentum: 0.000000
2024-03-26 11:16:52,117 epoch 1 - iter 12/48 - loss 2.94913909 - time (sec): 5.90 - samples/sec: 1509.63 - lr: 0.000011 - momentum: 0.000000
2024-03-26 11:16:54,122 epoch 1 - iter 16/48 - loss 2.77403858 - time (sec): 7.91 - samples/sec: 1534.06 - lr: 0.000016 - momentum: 0.000000
2024-03-26 11:16:56,377 epoch 1 - iter 20/48 - loss 2.62217028 - time (sec): 10.16 - samples/sec: 1503.44 - lr: 0.000020 - momentum: 0.000000
2024-03-26 11:16:59,474 epoch 1 - iter 24/48 - loss 2.48065772 - time (sec): 13.26 - samples/sec: 1370.87 - lr: 0.000024 - momentum: 0.000000
2024-03-26 11:17:01,976 epoch 1 - iter 28/48 - loss 2.34149056 - time (sec): 15.76 - samples/sec: 1352.03 - lr: 0.000028 - momentum: 0.000000
2024-03-26 11:17:02,812 epoch 1 - iter 32/48 - loss 2.25298393 - time (sec): 16.60 - samples/sec: 1406.67 - lr: 0.000032 - momentum: 0.000000
2024-03-26 11:17:04,135 epoch 1 - iter 36/48 - loss 2.14698057 - time (sec): 17.92 - samples/sec: 1459.09 - lr: 0.000036 - momentum: 0.000000
2024-03-26 11:17:06,067 epoch 1 - iter 40/48 - loss 2.04219894 - time (sec): 19.85 - samples/sec: 1465.20 - lr: 0.000041 - momentum: 0.000000
2024-03-26 11:17:08,030 epoch 1 - iter 44/48 - loss 1.93332136 - time (sec): 21.82 - samples/sec: 1464.45 - lr: 0.000045 - momentum: 0.000000
2024-03-26 11:17:09,428 epoch 1 - iter 48/48 - loss 1.84795953 - time (sec): 23.22 - samples/sec: 1484.86 - lr: 0.000049 - momentum: 0.000000
2024-03-26 11:17:09,428 ----------------------------------------------------------------------------------------------------
2024-03-26 11:17:09,428 EPOCH 1 done: loss 1.8480 - lr: 0.000049
2024-03-26 11:17:10,350 DEV : loss 0.5429823398590088 - f1-score (micro avg)  0.633
2024-03-26 11:17:10,351 saving best model
2024-03-26 11:17:10,630 ----------------------------------------------------------------------------------------------------
2024-03-26 11:17:11,954 epoch 2 - iter 4/48 - loss 0.75282481 - time (sec): 1.32 - samples/sec: 2192.70 - lr: 0.000050 - momentum: 0.000000
2024-03-26 11:17:13,819 epoch 2 - iter 8/48 - loss 0.62087530 - time (sec): 3.19 - samples/sec: 1912.95 - lr: 0.000049 - momentum: 0.000000
2024-03-26 11:17:17,322 epoch 2 - iter 12/48 - loss 0.52501189 - time (sec): 6.69 - samples/sec: 1520.86 - lr: 0.000049 - momentum: 0.000000
2024-03-26 11:17:19,881 epoch 2 - iter 16/48 - loss 0.49391674 - time (sec): 9.25 - samples/sec: 1439.80 - lr: 0.000048 - momentum: 0.000000
2024-03-26 11:17:22,688 epoch 2 - iter 20/48 - loss 0.46323174 - time (sec): 12.06 - samples/sec: 1377.78 - lr: 0.000048 - momentum: 0.000000
2024-03-26 11:17:24,686 epoch 2 - iter 24/48 - loss 0.44126141 - time (sec): 14.06 - samples/sec: 1371.64 - lr: 0.000047 - momentum: 0.000000
2024-03-26 11:17:26,494 epoch 2 - iter 28/48 - loss 0.44609745 - time (sec): 15.86 - samples/sec: 1382.57 - lr: 0.000047 - momentum: 0.000000
2024-03-26 11:17:28,311 epoch 2 - iter 32/48 - loss 0.44004278 - time (sec): 17.68 - samples/sec: 1391.63 - lr: 0.000046 - momentum: 0.000000
2024-03-26 11:17:30,239 epoch 2 - iter 36/48 - loss 0.42823178 - time (sec): 19.61 - samples/sec: 1399.00 - lr: 0.000046 - momentum: 0.000000
2024-03-26 11:17:31,269 epoch 2 - iter 40/48 - loss 0.42345263 - time (sec): 20.64 - samples/sec: 1446.43 - lr: 0.000046 - momentum: 0.000000
2024-03-26 11:17:32,746 epoch 2 - iter 44/48 - loss 0.42198128 - time (sec): 22.11 - samples/sec: 1465.79 - lr: 0.000045 - momentum: 0.000000
2024-03-26 11:17:34,323 epoch 2 - iter 48/48 - loss 0.40881526 - time (sec): 23.69 - samples/sec: 1454.96 - lr: 0.000045 - momentum: 0.000000
2024-03-26 11:17:34,324 ----------------------------------------------------------------------------------------------------
2024-03-26 11:17:34,324 EPOCH 2 done: loss 0.4088 - lr: 0.000045
2024-03-26 11:17:35,254 DEV : loss 0.28016945719718933 - f1-score (micro avg)  0.8134
2024-03-26 11:17:35,255 saving best model
2024-03-26 11:17:35,702 ----------------------------------------------------------------------------------------------------
2024-03-26 11:17:38,267 epoch 3 - iter 4/48 - loss 0.23345348 - time (sec): 2.56 - samples/sec: 1173.72 - lr: 0.000044 - momentum: 0.000000
2024-03-26 11:17:40,465 epoch 3 - iter 8/48 - loss 0.23145403 - time (sec): 4.76 - samples/sec: 1333.40 - lr: 0.000044 - momentum: 0.000000
2024-03-26 11:17:42,063 epoch 3 - iter 12/48 - loss 0.23780941 - time (sec): 6.36 - samples/sec: 1394.93 - lr: 0.000043 - momentum: 0.000000
2024-03-26 11:17:43,845 epoch 3 - iter 16/48 - loss 0.22351506 - time (sec): 8.14 - samples/sec: 1395.99 - lr: 0.000043 - momentum: 0.000000
2024-03-26 11:17:45,035 epoch 3 - iter 20/48 - loss 0.22899955 - time (sec): 9.33 - samples/sec: 1466.21 - lr: 0.000042 - momentum: 0.000000
2024-03-26 11:17:46,912 epoch 3 - iter 24/48 - loss 0.23890119 - time (sec): 11.21 - samples/sec: 1467.99 - lr: 0.000042 - momentum: 0.000000
2024-03-26 11:17:49,413 epoch 3 - iter 28/48 - loss 0.23771145 - time (sec): 13.71 - samples/sec: 1411.16 - lr: 0.000041 - momentum: 0.000000
2024-03-26 11:17:51,287 epoch 3 - iter 32/48 - loss 0.23968050 - time (sec): 15.58 - samples/sec: 1420.58 - lr: 0.000041 - momentum: 0.000000
2024-03-26 11:17:52,761 epoch 3 - iter 36/48 - loss 0.23251473 - time (sec): 17.06 - samples/sec: 1452.67 - lr: 0.000040 - momentum: 0.000000
2024-03-26 11:17:55,069 epoch 3 - iter 40/48 - loss 0.22549522 - time (sec): 19.37 - samples/sec: 1425.67 - lr: 0.000040 - momentum: 0.000000
2024-03-26 11:17:58,436 epoch 3 - iter 44/48 - loss 0.20910800 - time (sec): 22.73 - samples/sec: 1417.46 - lr: 0.000040 - momentum: 0.000000
2024-03-26 11:17:59,781 epoch 3 - iter 48/48 - loss 0.20549323 - time (sec): 24.08 - samples/sec: 1431.65 - lr: 0.000039 - momentum: 0.000000
2024-03-26 11:17:59,782 ----------------------------------------------------------------------------------------------------
2024-03-26 11:17:59,782 EPOCH 3 done: loss 0.2055 - lr: 0.000039
2024-03-26 11:18:00,714 DEV : loss 0.2106706202030182 - f1-score (micro avg)  0.8663
2024-03-26 11:18:00,715 saving best model
2024-03-26 11:18:01,155 ----------------------------------------------------------------------------------------------------
2024-03-26 11:18:02,763 epoch 4 - iter 4/48 - loss 0.21264590 - time (sec): 1.61 - samples/sec: 1588.44 - lr: 0.000039 - momentum: 0.000000
2024-03-26 11:18:05,112 epoch 4 - iter 8/48 - loss 0.16186058 - time (sec): 3.95 - samples/sec: 1515.65 - lr: 0.000038 - momentum: 0.000000
2024-03-26 11:18:06,364 epoch 4 - iter 12/48 - loss 0.15033662 - time (sec): 5.21 - samples/sec: 1605.29 - lr: 0.000038 - momentum: 0.000000
2024-03-26 11:18:08,647 epoch 4 - iter 16/48 - loss 0.14703161 - time (sec): 7.49 - samples/sec: 1505.41 - lr: 0.000037 - momentum: 0.000000
2024-03-26 11:18:11,277 epoch 4 - iter 20/48 - loss 0.13607905 - time (sec): 10.12 - samples/sec: 1381.72 - lr: 0.000037 - momentum: 0.000000
2024-03-26 11:18:13,388 epoch 4 - iter 24/48 - loss 0.14432807 - time (sec): 12.23 - samples/sec: 1376.38 - lr: 0.000036 - momentum: 0.000000
2024-03-26 11:18:15,527 epoch 4 - iter 28/48 - loss 0.14018550 - time (sec): 14.37 - samples/sec: 1384.49 - lr: 0.000036 - momentum: 0.000000
2024-03-26 11:18:18,192 epoch 4 - iter 32/48 - loss 0.13898599 - time (sec): 17.03 - samples/sec: 1353.72 - lr: 0.000035 - momentum: 0.000000
2024-03-26 11:18:21,031 epoch 4 - iter 36/48 - loss 0.13184037 - time (sec): 19.87 - samples/sec: 1346.00 - lr: 0.000035 - momentum: 0.000000
2024-03-26 11:18:22,799 epoch 4 - iter 40/48 - loss 0.12873724 - time (sec): 21.64 - samples/sec: 1344.37 - lr: 0.000034 - momentum: 0.000000
2024-03-26 11:18:24,883 epoch 4 - iter 44/48 - loss 0.12670940 - time (sec): 23.73 - samples/sec: 1345.51 - lr: 0.000034 - momentum: 0.000000
2024-03-26 11:18:26,610 epoch 4 - iter 48/48 - loss 0.12560460 - time (sec): 25.45 - samples/sec: 1354.39 - lr: 0.000034 - momentum: 0.000000
2024-03-26 11:18:26,610 ----------------------------------------------------------------------------------------------------
2024-03-26 11:18:26,610 EPOCH 4 done: loss 0.1256 - lr: 0.000034
2024-03-26 11:18:27,551 DEV : loss 0.1935855746269226 - f1-score (micro avg)  0.8846
2024-03-26 11:18:27,552 saving best model
2024-03-26 11:18:27,984 ----------------------------------------------------------------------------------------------------
2024-03-26 11:18:28,825 epoch 5 - iter 4/48 - loss 0.05091057 - time (sec): 0.84 - samples/sec: 2181.44 - lr: 0.000033 - momentum: 0.000000
2024-03-26 11:18:30,239 epoch 5 - iter 8/48 - loss 0.07272383 - time (sec): 2.25 - samples/sec: 1973.48 - lr: 0.000033 - momentum: 0.000000
2024-03-26 11:18:33,183 epoch 5 - iter 12/48 - loss 0.07617273 - time (sec): 5.20 - samples/sec: 1535.24 - lr: 0.000032 - momentum: 0.000000
2024-03-26 11:18:36,272 epoch 5 - iter 16/48 - loss 0.07780937 - time (sec): 8.29 - samples/sec: 1361.74 - lr: 0.000032 - momentum: 0.000000
2024-03-26 11:18:37,687 epoch 5 - iter 20/48 - loss 0.08570020 - time (sec): 9.70 - samples/sec: 1414.94 - lr: 0.000031 - momentum: 0.000000
2024-03-26 11:18:40,231 epoch 5 - iter 24/48 - loss 0.08408737 - time (sec): 12.25 - samples/sec: 1368.21 - lr: 0.000031 - momentum: 0.000000
2024-03-26 11:18:42,356 epoch 5 - iter 28/48 - loss 0.08255117 - time (sec): 14.37 - samples/sec: 1359.59 - lr: 0.000030 - momentum: 0.000000
2024-03-26 11:18:44,712 epoch 5 - iter 32/48 - loss 0.08552872 - time (sec): 16.73 - samples/sec: 1384.72 - lr: 0.000030 - momentum: 0.000000
2024-03-26 11:18:46,223 epoch 5 - iter 36/48 - loss 0.08821291 - time (sec): 18.24 - samples/sec: 1408.45 - lr: 0.000029 - momentum: 0.000000
2024-03-26 11:18:48,787 epoch 5 - iter 40/48 - loss 0.08411875 - time (sec): 20.80 - samples/sec: 1365.75 - lr: 0.000029 - momentum: 0.000000
2024-03-26 11:18:50,922 epoch 5 - iter 44/48 - loss 0.08515196 - time (sec): 22.94 - samples/sec: 1379.22 - lr: 0.000029 - momentum: 0.000000
2024-03-26 11:18:52,886 epoch 5 - iter 48/48 - loss 0.08656351 - time (sec): 24.90 - samples/sec: 1384.33 - lr: 0.000028 - momentum: 0.000000
2024-03-26 11:18:52,887 ----------------------------------------------------------------------------------------------------
2024-03-26 11:18:52,887 EPOCH 5 done: loss 0.0866 - lr: 0.000028
2024-03-26 11:18:53,817 DEV : loss 0.19532091915607452 - f1-score (micro avg)  0.9016
2024-03-26 11:18:53,819 saving best model
2024-03-26 11:18:54,268 ----------------------------------------------------------------------------------------------------
2024-03-26 11:18:55,970 epoch 6 - iter 4/48 - loss 0.08548813 - time (sec): 1.70 - samples/sec: 1464.57 - lr: 0.000028 - momentum: 0.000000
2024-03-26 11:18:58,418 epoch 6 - iter 8/48 - loss 0.07937192 - time (sec): 4.15 - samples/sec: 1542.85 - lr: 0.000027 - momentum: 0.000000
2024-03-26 11:19:00,401 epoch 6 - iter 12/48 - loss 0.07110090 - time (sec): 6.13 - samples/sec: 1477.40 - lr: 0.000027 - momentum: 0.000000
2024-03-26 11:19:02,537 epoch 6 - iter 16/48 - loss 0.06863509 - time (sec): 8.27 - samples/sec: 1466.76 - lr: 0.000026 - momentum: 0.000000
2024-03-26 11:19:05,304 epoch 6 - iter 20/48 - loss 0.06768250 - time (sec): 11.03 - samples/sec: 1447.96 - lr: 0.000026 - momentum: 0.000000
2024-03-26 11:19:06,872 epoch 6 - iter 24/48 - loss 0.07508927 - time (sec): 12.60 - samples/sec: 1468.79 - lr: 0.000025 - momentum: 0.000000
2024-03-26 11:19:08,298 epoch 6 - iter 28/48 - loss 0.07565420 - time (sec): 14.03 - samples/sec: 1473.36 - lr: 0.000025 - momentum: 0.000000
2024-03-26 11:19:09,509 epoch 6 - iter 32/48 - loss 0.07301484 - time (sec): 15.24 - samples/sec: 1492.92 - lr: 0.000024 - momentum: 0.000000
2024-03-26 11:19:11,034 epoch 6 - iter 36/48 - loss 0.06942201 - time (sec): 16.76 - samples/sec: 1523.08 - lr: 0.000024 - momentum: 0.000000
2024-03-26 11:19:13,012 epoch 6 - iter 40/48 - loss 0.07133967 - time (sec): 18.74 - samples/sec: 1512.02 - lr: 0.000023 - momentum: 0.000000
2024-03-26 11:19:15,300 epoch 6 - iter 44/48 - loss 0.06908886 - time (sec): 21.03 - samples/sec: 1528.97 - lr: 0.000023 - momentum: 0.000000
2024-03-26 11:19:17,041 epoch 6 - iter 48/48 - loss 0.06967724 - time (sec): 22.77 - samples/sec: 1513.89 - lr: 0.000023 - momentum: 0.000000
2024-03-26 11:19:17,041 ----------------------------------------------------------------------------------------------------
2024-03-26 11:19:17,041 EPOCH 6 done: loss 0.0697 - lr: 0.000023
2024-03-26 11:19:17,975 DEV : loss 0.1747245490550995 - f1-score (micro avg)  0.916
2024-03-26 11:19:17,976 saving best model
2024-03-26 11:19:18,419 ----------------------------------------------------------------------------------------------------
2024-03-26 11:19:20,065 epoch 7 - iter 4/48 - loss 0.04385190 - time (sec): 1.64 - samples/sec: 1481.61 - lr: 0.000022 - momentum: 0.000000
2024-03-26 11:19:21,725 epoch 7 - iter 8/48 - loss 0.05737850 - time (sec): 3.30 - samples/sec: 1499.05 - lr: 0.000022 - momentum: 0.000000
2024-03-26 11:19:23,900 epoch 7 - iter 12/48 - loss 0.05225209 - time (sec): 5.48 - samples/sec: 1436.43 - lr: 0.000021 - momentum: 0.000000
2024-03-26 11:19:25,967 epoch 7 - iter 16/48 - loss 0.04753228 - time (sec): 7.55 - samples/sec: 1476.37 - lr: 0.000021 - momentum: 0.000000
2024-03-26 11:19:26,626 epoch 7 - iter 20/48 - loss 0.04517211 - time (sec): 8.21 - samples/sec: 1579.39 - lr: 0.000020 - momentum: 0.000000
2024-03-26 11:19:28,231 epoch 7 - iter 24/48 - loss 0.04582161 - time (sec): 9.81 - samples/sec: 1561.85 - lr: 0.000020 - momentum: 0.000000
2024-03-26 11:19:31,160 epoch 7 - iter 28/48 - loss 0.04440788 - time (sec): 12.74 - samples/sec: 1461.94 - lr: 0.000019 - momentum: 0.000000
2024-03-26 11:19:33,988 epoch 7 - iter 32/48 - loss 0.04388928 - time (sec): 15.57 - samples/sec: 1391.71 - lr: 0.000019 - momentum: 0.000000
2024-03-26 11:19:36,840 epoch 7 - iter 36/48 - loss 0.04910196 - time (sec): 18.42 - samples/sec: 1399.69 - lr: 0.000018 - momentum: 0.000000
2024-03-26 11:19:38,832 epoch 7 - iter 40/48 - loss 0.05279163 - time (sec): 20.41 - samples/sec: 1408.42 - lr: 0.000018 - momentum: 0.000000
2024-03-26 11:19:41,428 epoch 7 - iter 44/48 - loss 0.05326335 - time (sec): 23.01 - samples/sec: 1384.52 - lr: 0.000017 - momentum: 0.000000
2024-03-26 11:19:43,269 epoch 7 - iter 48/48 - loss 0.05213195 - time (sec): 24.85 - samples/sec: 1387.33 - lr: 0.000017 - momentum: 0.000000
2024-03-26 11:19:43,269 ----------------------------------------------------------------------------------------------------
2024-03-26 11:19:43,269 EPOCH 7 done: loss 0.0521 - lr: 0.000017
2024-03-26 11:19:44,202 DEV : loss 0.1832708865404129 - f1-score (micro avg)  0.911
2024-03-26 11:19:44,203 ----------------------------------------------------------------------------------------------------
2024-03-26 11:19:46,904 epoch 8 - iter 4/48 - loss 0.04675538 - time (sec): 2.70 - samples/sec: 1223.02 - lr: 0.000017 - momentum: 0.000000
2024-03-26 11:19:49,044 epoch 8 - iter 8/48 - loss 0.03493802 - time (sec): 4.84 - samples/sec: 1212.36 - lr: 0.000016 - momentum: 0.000000
2024-03-26 11:19:52,242 epoch 8 - iter 12/48 - loss 0.03381936 - time (sec): 8.04 - samples/sec: 1205.64 - lr: 0.000016 - momentum: 0.000000
2024-03-26 11:19:54,226 epoch 8 - iter 16/48 - loss 0.04243199 - time (sec): 10.02 - samples/sec: 1231.53 - lr: 0.000015 - momentum: 0.000000
2024-03-26 11:19:55,720 epoch 8 - iter 20/48 - loss 0.03854979 - time (sec): 11.52 - samples/sec: 1275.09 - lr: 0.000015 - momentum: 0.000000
2024-03-26 11:19:58,301 epoch 8 - iter 24/48 - loss 0.03855591 - time (sec): 14.10 - samples/sec: 1265.80 - lr: 0.000014 - momentum: 0.000000
2024-03-26 11:20:00,098 epoch 8 - iter 28/48 - loss 0.04124147 - time (sec): 15.89 - samples/sec: 1300.94 - lr: 0.000014 - momentum: 0.000000
2024-03-26 11:20:01,734 epoch 8 - iter 32/48 - loss 0.03962775 - time (sec): 17.53 - samples/sec: 1327.09 - lr: 0.000013 - momentum: 0.000000
2024-03-26 11:20:03,026 epoch 8 - iter 36/48 - loss 0.03946872 - time (sec): 18.82 - samples/sec: 1359.20 - lr: 0.000013 - momentum: 0.000000
2024-03-26 11:20:05,440 epoch 8 - iter 40/48 - loss 0.03972045 - time (sec): 21.24 - samples/sec: 1365.27 - lr: 0.000012 - momentum: 0.000000
2024-03-26 11:20:08,341 epoch 8 - iter 44/48 - loss 0.03737704 - time (sec): 24.14 - samples/sec: 1334.75 - lr: 0.000012 - momentum: 0.000000
2024-03-26 11:20:10,427 epoch 8 - iter 48/48 - loss 0.03797592 - time (sec): 26.22 - samples/sec: 1314.57 - lr: 0.000011 - momentum: 0.000000
2024-03-26 11:20:10,427 ----------------------------------------------------------------------------------------------------
2024-03-26 11:20:10,427 EPOCH 8 done: loss 0.0380 - lr: 0.000011
2024-03-26 11:20:11,369 DEV : loss 0.1888364851474762 - f1-score (micro avg)  0.9299
2024-03-26 11:20:11,370 saving best model
2024-03-26 11:20:11,812 ----------------------------------------------------------------------------------------------------
2024-03-26 11:20:13,741 epoch 9 - iter 4/48 - loss 0.03350915 - time (sec): 1.93 - samples/sec: 1476.17 - lr: 0.000011 - momentum: 0.000000
2024-03-26 11:20:16,180 epoch 9 - iter 8/48 - loss 0.02728296 - time (sec): 4.37 - samples/sec: 1404.75 - lr: 0.000011 - momentum: 0.000000
2024-03-26 11:20:18,591 epoch 9 - iter 12/48 - loss 0.03564997 - time (sec): 6.78 - samples/sec: 1362.34 - lr: 0.000010 - momentum: 0.000000
2024-03-26 11:20:20,699 epoch 9 - iter 16/48 - loss 0.03661492 - time (sec): 8.88 - samples/sec: 1361.41 - lr: 0.000010 - momentum: 0.000000
2024-03-26 11:20:22,204 epoch 9 - iter 20/48 - loss 0.03193217 - time (sec): 10.39 - samples/sec: 1418.57 - lr: 0.000009 - momentum: 0.000000
2024-03-26 11:20:23,438 epoch 9 - iter 24/48 - loss 0.02928490 - time (sec): 11.62 - samples/sec: 1465.75 - lr: 0.000009 - momentum: 0.000000
2024-03-26 11:20:25,136 epoch 9 - iter 28/48 - loss 0.02830002 - time (sec): 13.32 - samples/sec: 1484.58 - lr: 0.000008 - momentum: 0.000000
2024-03-26 11:20:27,488 epoch 9 - iter 32/48 - loss 0.03365236 - time (sec): 15.67 - samples/sec: 1467.13 - lr: 0.000008 - momentum: 0.000000
2024-03-26 11:20:30,228 epoch 9 - iter 36/48 - loss 0.03477106 - time (sec): 18.41 - samples/sec: 1418.66 - lr: 0.000007 - momentum: 0.000000
2024-03-26 11:20:33,201 epoch 9 - iter 40/48 - loss 0.03467439 - time (sec): 21.39 - samples/sec: 1378.21 - lr: 0.000007 - momentum: 0.000000
2024-03-26 11:20:35,067 epoch 9 - iter 44/48 - loss 0.03450948 - time (sec): 23.25 - samples/sec: 1392.58 - lr: 0.000006 - momentum: 0.000000
2024-03-26 11:20:36,124 epoch 9 - iter 48/48 - loss 0.03400891 - time (sec): 24.31 - samples/sec: 1418.07 - lr: 0.000006 - momentum: 0.000000
2024-03-26 11:20:36,124 ----------------------------------------------------------------------------------------------------
2024-03-26 11:20:36,124 EPOCH 9 done: loss 0.0340 - lr: 0.000006
2024-03-26 11:20:37,072 DEV : loss 0.1788191944360733 - f1-score (micro avg)  0.9328
2024-03-26 11:20:37,073 saving best model
2024-03-26 11:20:37,537 ----------------------------------------------------------------------------------------------------
2024-03-26 11:20:39,897 epoch 10 - iter 4/48 - loss 0.01213481 - time (sec): 2.36 - samples/sec: 1400.54 - lr: 0.000006 - momentum: 0.000000
2024-03-26 11:20:42,071 epoch 10 - iter 8/48 - loss 0.01599539 - time (sec): 4.53 - samples/sec: 1363.25 - lr: 0.000005 - momentum: 0.000000
2024-03-26 11:20:43,992 epoch 10 - iter 12/48 - loss 0.01796113 - time (sec): 6.45 - samples/sec: 1367.47 - lr: 0.000005 - momentum: 0.000000
2024-03-26 11:20:45,227 epoch 10 - iter 16/48 - loss 0.01897141 - time (sec): 7.69 - samples/sec: 1433.43 - lr: 0.000004 - momentum: 0.000000
2024-03-26 11:20:47,232 epoch 10 - iter 20/48 - loss 0.02538412 - time (sec): 9.69 - samples/sec: 1414.27 - lr: 0.000004 - momentum: 0.000000
2024-03-26 11:20:49,589 epoch 10 - iter 24/48 - loss 0.03308522 - time (sec): 12.05 - samples/sec: 1378.16 - lr: 0.000003 - momentum: 0.000000
2024-03-26 11:20:50,495 epoch 10 - iter 28/48 - loss 0.03246447 - time (sec): 12.96 - samples/sec: 1450.20 - lr: 0.000003 - momentum: 0.000000
2024-03-26 11:20:51,826 epoch 10 - iter 32/48 - loss 0.03099225 - time (sec): 14.29 - samples/sec: 1486.63 - lr: 0.000002 - momentum: 0.000000
2024-03-26 11:20:54,658 epoch 10 - iter 36/48 - loss 0.02893431 - time (sec): 17.12 - samples/sec: 1442.43 - lr: 0.000002 - momentum: 0.000000
2024-03-26 11:20:57,051 epoch 10 - iter 40/48 - loss 0.02897645 - time (sec): 19.51 - samples/sec: 1473.63 - lr: 0.000001 - momentum: 0.000000
2024-03-26 11:20:59,691 epoch 10 - iter 44/48 - loss 0.02902513 - time (sec): 22.15 - samples/sec: 1447.93 - lr: 0.000001 - momentum: 0.000000
2024-03-26 11:21:01,695 epoch 10 - iter 48/48 - loss 0.02812175 - time (sec): 24.16 - samples/sec: 1427.09 - lr: 0.000000 - momentum: 0.000000
2024-03-26 11:21:01,695 ----------------------------------------------------------------------------------------------------
2024-03-26 11:21:01,695 EPOCH 10 done: loss 0.0281 - lr: 0.000000
2024-03-26 11:21:02,635 DEV : loss 0.1868205964565277 - f1-score (micro avg)  0.9321
2024-03-26 11:21:02,915 ----------------------------------------------------------------------------------------------------
2024-03-26 11:21:02,915 Loading model from best epoch ...
2024-03-26 11:21:03,816 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 11:21:04,576 
Results:
- F-score (micro) 0.9095
- F-score (macro) 0.6907
- Accuracy 0.8364

By class:
              precision    recall  f1-score   support

 Unternehmen     0.9109    0.8835    0.8969       266
 Auslagerung     0.8760    0.9076    0.8915       249
         Ort     0.9635    0.9851    0.9742       134
    Software     0.0000    0.0000    0.0000         0

   micro avg     0.9053    0.9137    0.9095       649
   macro avg     0.6876    0.6940    0.6907       649
weighted avg     0.9083    0.9137    0.9108       649

2024-03-26 11:21:04,576 ----------------------------------------------------------------------------------------------------