Hugging Face's logo

stefan-it
/
flair-co-funer-gbert_base-bs8-e10-lr5e-05-4

Token Classification
Flair
PyTorch
TensorBoard
German
sequence-tagger-model
hetzner
hetzner-gex44
hetzner-gpu
Model card Files Metrics Community
3
flair-co-funer-gbert_base-bs8-e10-lr5e-05-4
File size: 23,765 Bytes 
83cadb9
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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 10:28:35,728 ----------------------------------------------------------------------------------------------------
2024-03-26 10:28:35,729 Model: "SequenceTagger(
  (embeddings): TransformerWordEmbeddings(
    (model): BertModel(
      (embeddings): BertEmbeddings(
        (word_embeddings): Embedding(31103, 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 10:28:35,729 ----------------------------------------------------------------------------------------------------
2024-03-26 10:28:35,729 Corpus: 758 train + 94 dev + 96 test sentences
2024-03-26 10:28:35,729 ----------------------------------------------------------------------------------------------------
2024-03-26 10:28:35,729 Train:  758 sentences
2024-03-26 10:28:35,729         (train_with_dev=False, train_with_test=False)
2024-03-26 10:28:35,729 ----------------------------------------------------------------------------------------------------
2024-03-26 10:28:35,729 Training Params:
2024-03-26 10:28:35,729  - learning_rate: "5e-05" 
2024-03-26 10:28:35,729  - mini_batch_size: "8"
2024-03-26 10:28:35,729  - max_epochs: "10"
2024-03-26 10:28:35,729  - shuffle: "True"
2024-03-26 10:28:35,729 ----------------------------------------------------------------------------------------------------
2024-03-26 10:28:35,729 Plugins:
2024-03-26 10:28:35,729  - TensorboardLogger
2024-03-26 10:28:35,729  - LinearScheduler | warmup_fraction: '0.1'
2024-03-26 10:28:35,729 ----------------------------------------------------------------------------------------------------
2024-03-26 10:28:35,729 Final evaluation on model from best epoch (best-model.pt)
2024-03-26 10:28:35,729  - metric: "('micro avg', 'f1-score')"
2024-03-26 10:28:35,729 ----------------------------------------------------------------------------------------------------
2024-03-26 10:28:35,729 Computation:
2024-03-26 10:28:35,729  - compute on device: cuda:0
2024-03-26 10:28:35,729  - embedding storage: none
2024-03-26 10:28:35,729 ----------------------------------------------------------------------------------------------------
2024-03-26 10:28:35,729 Model training base path: "flair-co-funer-gbert_base-bs8-e10-lr5e-05-4"
2024-03-26 10:28:35,729 ----------------------------------------------------------------------------------------------------
2024-03-26 10:28:35,729 ----------------------------------------------------------------------------------------------------
2024-03-26 10:28:35,729 Logging anything other than scalars to TensorBoard is currently not supported.
2024-03-26 10:28:37,074 epoch 1 - iter 9/95 - loss 3.29095728 - time (sec): 1.34 - samples/sec: 2157.52 - lr: 0.000004 - momentum: 0.000000
2024-03-26 10:28:38,456 epoch 1 - iter 18/95 - loss 3.08915397 - time (sec): 2.73 - samples/sec: 2022.41 - lr: 0.000009 - momentum: 0.000000
2024-03-26 10:28:40,075 epoch 1 - iter 27/95 - loss 2.78831297 - time (sec): 4.35 - samples/sec: 1973.60 - lr: 0.000014 - momentum: 0.000000
2024-03-26 10:28:42,038 epoch 1 - iter 36/95 - loss 2.55176526 - time (sec): 6.31 - samples/sec: 1892.23 - lr: 0.000018 - momentum: 0.000000
2024-03-26 10:28:43,886 epoch 1 - iter 45/95 - loss 2.34472767 - time (sec): 8.16 - samples/sec: 1911.70 - lr: 0.000023 - momentum: 0.000000
2024-03-26 10:28:46,102 epoch 1 - iter 54/95 - loss 2.18435379 - time (sec): 10.37 - samples/sec: 1845.74 - lr: 0.000028 - momentum: 0.000000
2024-03-26 10:28:48,110 epoch 1 - iter 63/95 - loss 2.02400115 - time (sec): 12.38 - samples/sec: 1825.52 - lr: 0.000033 - momentum: 0.000000
2024-03-26 10:28:49,077 epoch 1 - iter 72/95 - loss 1.92796284 - time (sec): 13.35 - samples/sec: 1871.48 - lr: 0.000037 - momentum: 0.000000
2024-03-26 10:28:51,349 epoch 1 - iter 81/95 - loss 1.78146040 - time (sec): 15.62 - samples/sec: 1818.51 - lr: 0.000042 - momentum: 0.000000
2024-03-26 10:28:52,676 epoch 1 - iter 90/95 - loss 1.64759700 - time (sec): 16.95 - samples/sec: 1885.89 - lr: 0.000047 - momentum: 0.000000
2024-03-26 10:28:53,938 ----------------------------------------------------------------------------------------------------
2024-03-26 10:28:53,938 EPOCH 1 done: loss 1.5782 - lr: 0.000047
2024-03-26 10:28:54,771 DEV : loss 0.4267578125 - f1-score (micro avg)  0.6895
2024-03-26 10:28:54,773 saving best model
2024-03-26 10:28:55,033 ----------------------------------------------------------------------------------------------------
2024-03-26 10:28:56,620 epoch 2 - iter 9/95 - loss 0.54697289 - time (sec): 1.59 - samples/sec: 1816.17 - lr: 0.000050 - momentum: 0.000000
2024-03-26 10:28:58,258 epoch 2 - iter 18/95 - loss 0.45713150 - time (sec): 3.22 - samples/sec: 1916.68 - lr: 0.000049 - momentum: 0.000000
2024-03-26 10:29:00,032 epoch 2 - iter 27/95 - loss 0.41563849 - time (sec): 5.00 - samples/sec: 1891.77 - lr: 0.000048 - momentum: 0.000000
2024-03-26 10:29:02,403 epoch 2 - iter 36/95 - loss 0.36132141 - time (sec): 7.37 - samples/sec: 1770.32 - lr: 0.000048 - momentum: 0.000000
2024-03-26 10:29:04,360 epoch 2 - iter 45/95 - loss 0.34829906 - time (sec): 9.33 - samples/sec: 1769.00 - lr: 0.000047 - momentum: 0.000000
2024-03-26 10:29:06,122 epoch 2 - iter 54/95 - loss 0.35634407 - time (sec): 11.09 - samples/sec: 1788.56 - lr: 0.000047 - momentum: 0.000000
2024-03-26 10:29:08,529 epoch 2 - iter 63/95 - loss 0.33840681 - time (sec): 13.50 - samples/sec: 1771.74 - lr: 0.000046 - momentum: 0.000000
2024-03-26 10:29:10,339 epoch 2 - iter 72/95 - loss 0.33790562 - time (sec): 15.31 - samples/sec: 1766.89 - lr: 0.000046 - momentum: 0.000000
2024-03-26 10:29:12,498 epoch 2 - iter 81/95 - loss 0.33273931 - time (sec): 17.46 - samples/sec: 1751.70 - lr: 0.000045 - momentum: 0.000000
2024-03-26 10:29:13,779 epoch 2 - iter 90/95 - loss 0.32834277 - time (sec): 18.74 - samples/sec: 1777.33 - lr: 0.000045 - momentum: 0.000000
2024-03-26 10:29:14,221 ----------------------------------------------------------------------------------------------------
2024-03-26 10:29:14,221 EPOCH 2 done: loss 0.3237 - lr: 0.000045
2024-03-26 10:29:15,109 DEV : loss 0.2464773952960968 - f1-score (micro avg)  0.8605
2024-03-26 10:29:15,110 saving best model
2024-03-26 10:29:15,549 ----------------------------------------------------------------------------------------------------
2024-03-26 10:29:17,114 epoch 3 - iter 9/95 - loss 0.20825595 - time (sec): 1.56 - samples/sec: 1694.91 - lr: 0.000044 - momentum: 0.000000
2024-03-26 10:29:18,827 epoch 3 - iter 18/95 - loss 0.17689281 - time (sec): 3.28 - samples/sec: 1705.87 - lr: 0.000043 - momentum: 0.000000
2024-03-26 10:29:20,654 epoch 3 - iter 27/95 - loss 0.17979153 - time (sec): 5.10 - samples/sec: 1745.24 - lr: 0.000043 - momentum: 0.000000
2024-03-26 10:29:22,428 epoch 3 - iter 36/95 - loss 0.18580580 - time (sec): 6.88 - samples/sec: 1755.08 - lr: 0.000042 - momentum: 0.000000
2024-03-26 10:29:24,423 epoch 3 - iter 45/95 - loss 0.17998091 - time (sec): 8.87 - samples/sec: 1779.47 - lr: 0.000042 - momentum: 0.000000
2024-03-26 10:29:26,639 epoch 3 - iter 54/95 - loss 0.17698755 - time (sec): 11.09 - samples/sec: 1746.72 - lr: 0.000041 - momentum: 0.000000
2024-03-26 10:29:28,331 epoch 3 - iter 63/95 - loss 0.17392405 - time (sec): 12.78 - samples/sec: 1750.43 - lr: 0.000041 - momentum: 0.000000
2024-03-26 10:29:30,291 epoch 3 - iter 72/95 - loss 0.17133371 - time (sec): 14.74 - samples/sec: 1756.22 - lr: 0.000040 - momentum: 0.000000
2024-03-26 10:29:32,243 epoch 3 - iter 81/95 - loss 0.17832679 - time (sec): 16.69 - samples/sec: 1774.24 - lr: 0.000040 - momentum: 0.000000
2024-03-26 10:29:34,459 epoch 3 - iter 90/95 - loss 0.17162912 - time (sec): 18.91 - samples/sec: 1753.84 - lr: 0.000039 - momentum: 0.000000
2024-03-26 10:29:35,067 ----------------------------------------------------------------------------------------------------
2024-03-26 10:29:35,067 EPOCH 3 done: loss 0.1731 - lr: 0.000039
2024-03-26 10:29:35,977 DEV : loss 0.1760903000831604 - f1-score (micro avg)  0.8827
2024-03-26 10:29:35,978 saving best model
2024-03-26 10:29:36,404 ----------------------------------------------------------------------------------------------------
2024-03-26 10:29:38,783 epoch 4 - iter 9/95 - loss 0.06788233 - time (sec): 2.38 - samples/sec: 1662.93 - lr: 0.000039 - momentum: 0.000000
2024-03-26 10:29:39,946 epoch 4 - iter 18/95 - loss 0.08548344 - time (sec): 3.54 - samples/sec: 1807.39 - lr: 0.000038 - momentum: 0.000000
2024-03-26 10:29:42,047 epoch 4 - iter 27/95 - loss 0.09484554 - time (sec): 5.64 - samples/sec: 1843.15 - lr: 0.000037 - momentum: 0.000000
2024-03-26 10:29:43,510 epoch 4 - iter 36/95 - loss 0.10345344 - time (sec): 7.11 - samples/sec: 1877.49 - lr: 0.000037 - momentum: 0.000000
2024-03-26 10:29:44,804 epoch 4 - iter 45/95 - loss 0.10546483 - time (sec): 8.40 - samples/sec: 1910.57 - lr: 0.000036 - momentum: 0.000000
2024-03-26 10:29:46,854 epoch 4 - iter 54/95 - loss 0.10315337 - time (sec): 10.45 - samples/sec: 1852.22 - lr: 0.000036 - momentum: 0.000000
2024-03-26 10:29:49,095 epoch 4 - iter 63/95 - loss 0.11633996 - time (sec): 12.69 - samples/sec: 1824.42 - lr: 0.000035 - momentum: 0.000000
2024-03-26 10:29:50,531 epoch 4 - iter 72/95 - loss 0.11408630 - time (sec): 14.13 - samples/sec: 1860.23 - lr: 0.000035 - momentum: 0.000000
2024-03-26 10:29:52,116 epoch 4 - iter 81/95 - loss 0.11337976 - time (sec): 15.71 - samples/sec: 1890.41 - lr: 0.000034 - momentum: 0.000000
2024-03-26 10:29:53,706 epoch 4 - iter 90/95 - loss 0.11156393 - time (sec): 17.30 - samples/sec: 1919.30 - lr: 0.000034 - momentum: 0.000000
2024-03-26 10:29:54,325 ----------------------------------------------------------------------------------------------------
2024-03-26 10:29:54,325 EPOCH 4 done: loss 0.1120 - lr: 0.000034
2024-03-26 10:29:55,226 DEV : loss 0.17519471049308777 - f1-score (micro avg)  0.909
2024-03-26 10:29:55,227 saving best model
2024-03-26 10:29:55,670 ----------------------------------------------------------------------------------------------------
2024-03-26 10:29:56,867 epoch 5 - iter 9/95 - loss 0.12043110 - time (sec): 1.20 - samples/sec: 2473.59 - lr: 0.000033 - momentum: 0.000000
2024-03-26 10:29:58,296 epoch 5 - iter 18/95 - loss 0.10960541 - time (sec): 2.63 - samples/sec: 2220.06 - lr: 0.000032 - momentum: 0.000000
2024-03-26 10:30:00,275 epoch 5 - iter 27/95 - loss 0.09760046 - time (sec): 4.60 - samples/sec: 1997.58 - lr: 0.000032 - momentum: 0.000000
2024-03-26 10:30:02,673 epoch 5 - iter 36/95 - loss 0.09358827 - time (sec): 7.00 - samples/sec: 1813.85 - lr: 0.000031 - momentum: 0.000000
2024-03-26 10:30:03,889 epoch 5 - iter 45/95 - loss 0.09683649 - time (sec): 8.22 - samples/sec: 1858.53 - lr: 0.000031 - momentum: 0.000000
2024-03-26 10:30:05,744 epoch 5 - iter 54/95 - loss 0.09239902 - time (sec): 10.07 - samples/sec: 1900.20 - lr: 0.000030 - momentum: 0.000000
2024-03-26 10:30:07,795 epoch 5 - iter 63/95 - loss 0.08595866 - time (sec): 12.12 - samples/sec: 1884.90 - lr: 0.000030 - momentum: 0.000000
2024-03-26 10:30:09,052 epoch 5 - iter 72/95 - loss 0.08467307 - time (sec): 13.38 - samples/sec: 1913.09 - lr: 0.000029 - momentum: 0.000000
2024-03-26 10:30:11,594 epoch 5 - iter 81/95 - loss 0.07917529 - time (sec): 15.92 - samples/sec: 1844.62 - lr: 0.000029 - momentum: 0.000000
2024-03-26 10:30:13,611 epoch 5 - iter 90/95 - loss 0.07825247 - time (sec): 17.94 - samples/sec: 1825.05 - lr: 0.000028 - momentum: 0.000000
2024-03-26 10:30:14,473 ----------------------------------------------------------------------------------------------------
2024-03-26 10:30:14,473 EPOCH 5 done: loss 0.0797 - lr: 0.000028
2024-03-26 10:30:15,378 DEV : loss 0.14438194036483765 - f1-score (micro avg)  0.9224
2024-03-26 10:30:15,379 saving best model
2024-03-26 10:30:15,816 ----------------------------------------------------------------------------------------------------
2024-03-26 10:30:17,459 epoch 6 - iter 9/95 - loss 0.09244981 - time (sec): 1.64 - samples/sec: 2018.64 - lr: 0.000027 - momentum: 0.000000
2024-03-26 10:30:19,514 epoch 6 - iter 18/95 - loss 0.06354727 - time (sec): 3.70 - samples/sec: 1832.96 - lr: 0.000027 - momentum: 0.000000
2024-03-26 10:30:20,945 epoch 6 - iter 27/95 - loss 0.06574351 - time (sec): 5.13 - samples/sec: 1863.31 - lr: 0.000026 - momentum: 0.000000
2024-03-26 10:30:23,266 epoch 6 - iter 36/95 - loss 0.05330838 - time (sec): 7.45 - samples/sec: 1723.80 - lr: 0.000026 - momentum: 0.000000
2024-03-26 10:30:25,027 epoch 6 - iter 45/95 - loss 0.05097494 - time (sec): 9.21 - samples/sec: 1746.61 - lr: 0.000025 - momentum: 0.000000
2024-03-26 10:30:27,502 epoch 6 - iter 54/95 - loss 0.05867496 - time (sec): 11.69 - samples/sec: 1722.90 - lr: 0.000025 - momentum: 0.000000
2024-03-26 10:30:29,011 epoch 6 - iter 63/95 - loss 0.06045661 - time (sec): 13.19 - samples/sec: 1739.77 - lr: 0.000024 - momentum: 0.000000
2024-03-26 10:30:30,546 epoch 6 - iter 72/95 - loss 0.06179262 - time (sec): 14.73 - samples/sec: 1761.21 - lr: 0.000024 - momentum: 0.000000
2024-03-26 10:30:32,622 epoch 6 - iter 81/95 - loss 0.06226128 - time (sec): 16.81 - samples/sec: 1754.24 - lr: 0.000023 - momentum: 0.000000
2024-03-26 10:30:33,799 epoch 6 - iter 90/95 - loss 0.06477171 - time (sec): 17.98 - samples/sec: 1798.68 - lr: 0.000023 - momentum: 0.000000
2024-03-26 10:30:35,131 ----------------------------------------------------------------------------------------------------
2024-03-26 10:30:35,131 EPOCH 6 done: loss 0.0628 - lr: 0.000023
2024-03-26 10:30:36,044 DEV : loss 0.16837507486343384 - f1-score (micro avg)  0.912
2024-03-26 10:30:36,047 ----------------------------------------------------------------------------------------------------
2024-03-26 10:30:37,419 epoch 7 - iter 9/95 - loss 0.04042555 - time (sec): 1.37 - samples/sec: 2313.60 - lr: 0.000022 - momentum: 0.000000
2024-03-26 10:30:39,544 epoch 7 - iter 18/95 - loss 0.04069942 - time (sec): 3.50 - samples/sec: 1928.39 - lr: 0.000021 - momentum: 0.000000
2024-03-26 10:30:41,509 epoch 7 - iter 27/95 - loss 0.04506976 - time (sec): 5.46 - samples/sec: 1789.58 - lr: 0.000021 - momentum: 0.000000
2024-03-26 10:30:42,799 epoch 7 - iter 36/95 - loss 0.04308197 - time (sec): 6.75 - samples/sec: 1852.73 - lr: 0.000020 - momentum: 0.000000
2024-03-26 10:30:44,491 epoch 7 - iter 45/95 - loss 0.04045458 - time (sec): 8.44 - samples/sec: 1863.40 - lr: 0.000020 - momentum: 0.000000
2024-03-26 10:30:46,696 epoch 7 - iter 54/95 - loss 0.03804365 - time (sec): 10.65 - samples/sec: 1839.96 - lr: 0.000019 - momentum: 0.000000
2024-03-26 10:30:48,735 epoch 7 - iter 63/95 - loss 0.04117399 - time (sec): 12.69 - samples/sec: 1794.55 - lr: 0.000019 - momentum: 0.000000
2024-03-26 10:30:50,858 epoch 7 - iter 72/95 - loss 0.03975813 - time (sec): 14.81 - samples/sec: 1771.80 - lr: 0.000018 - momentum: 0.000000
2024-03-26 10:30:52,373 epoch 7 - iter 81/95 - loss 0.04421200 - time (sec): 16.32 - samples/sec: 1778.87 - lr: 0.000018 - momentum: 0.000000
2024-03-26 10:30:54,245 epoch 7 - iter 90/95 - loss 0.05043406 - time (sec): 18.20 - samples/sec: 1807.50 - lr: 0.000017 - momentum: 0.000000
2024-03-26 10:30:54,925 ----------------------------------------------------------------------------------------------------
2024-03-26 10:30:54,925 EPOCH 7 done: loss 0.0503 - lr: 0.000017
2024-03-26 10:30:55,831 DEV : loss 0.16629257798194885 - f1-score (micro avg)  0.9369
2024-03-26 10:30:55,833 saving best model
2024-03-26 10:30:56,261 ----------------------------------------------------------------------------------------------------
2024-03-26 10:30:57,904 epoch 8 - iter 9/95 - loss 0.01558551 - time (sec): 1.64 - samples/sec: 1791.91 - lr: 0.000016 - momentum: 0.000000
2024-03-26 10:31:00,035 epoch 8 - iter 18/95 - loss 0.02309291 - time (sec): 3.77 - samples/sec: 1757.14 - lr: 0.000016 - momentum: 0.000000
2024-03-26 10:31:01,878 epoch 8 - iter 27/95 - loss 0.03236659 - time (sec): 5.62 - samples/sec: 1728.05 - lr: 0.000015 - momentum: 0.000000
2024-03-26 10:31:03,856 epoch 8 - iter 36/95 - loss 0.03080217 - time (sec): 7.59 - samples/sec: 1734.43 - lr: 0.000015 - momentum: 0.000000
2024-03-26 10:31:04,880 epoch 8 - iter 45/95 - loss 0.03971592 - time (sec): 8.62 - samples/sec: 1820.33 - lr: 0.000014 - momentum: 0.000000
2024-03-26 10:31:06,819 epoch 8 - iter 54/95 - loss 0.04108818 - time (sec): 10.56 - samples/sec: 1809.42 - lr: 0.000014 - momentum: 0.000000
2024-03-26 10:31:09,038 epoch 8 - iter 63/95 - loss 0.04377676 - time (sec): 12.78 - samples/sec: 1793.88 - lr: 0.000013 - momentum: 0.000000
2024-03-26 10:31:11,243 epoch 8 - iter 72/95 - loss 0.04319877 - time (sec): 14.98 - samples/sec: 1784.40 - lr: 0.000013 - momentum: 0.000000
2024-03-26 10:31:12,951 epoch 8 - iter 81/95 - loss 0.04251063 - time (sec): 16.69 - samples/sec: 1788.33 - lr: 0.000012 - momentum: 0.000000
2024-03-26 10:31:14,865 epoch 8 - iter 90/95 - loss 0.03992474 - time (sec): 18.60 - samples/sec: 1784.67 - lr: 0.000012 - momentum: 0.000000
2024-03-26 10:31:15,459 ----------------------------------------------------------------------------------------------------
2024-03-26 10:31:15,459 EPOCH 8 done: loss 0.0404 - lr: 0.000012
2024-03-26 10:31:16,369 DEV : loss 0.15785594284534454 - f1-score (micro avg)  0.9382
2024-03-26 10:31:16,370 saving best model
2024-03-26 10:31:16,801 ----------------------------------------------------------------------------------------------------
2024-03-26 10:31:18,339 epoch 9 - iter 9/95 - loss 0.02870197 - time (sec): 1.54 - samples/sec: 2070.29 - lr: 0.000011 - momentum: 0.000000
2024-03-26 10:31:20,627 epoch 9 - iter 18/95 - loss 0.02268037 - time (sec): 3.82 - samples/sec: 1781.03 - lr: 0.000010 - momentum: 0.000000
2024-03-26 10:31:22,218 epoch 9 - iter 27/95 - loss 0.01915786 - time (sec): 5.41 - samples/sec: 1799.85 - lr: 0.000010 - momentum: 0.000000
2024-03-26 10:31:24,521 epoch 9 - iter 36/95 - loss 0.02449214 - time (sec): 7.72 - samples/sec: 1757.80 - lr: 0.000009 - momentum: 0.000000
2024-03-26 10:31:26,418 epoch 9 - iter 45/95 - loss 0.02330329 - time (sec): 9.62 - samples/sec: 1730.41 - lr: 0.000009 - momentum: 0.000000
2024-03-26 10:31:27,794 epoch 9 - iter 54/95 - loss 0.02848105 - time (sec): 10.99 - samples/sec: 1775.35 - lr: 0.000008 - momentum: 0.000000
2024-03-26 10:31:29,866 epoch 9 - iter 63/95 - loss 0.02663824 - time (sec): 13.06 - samples/sec: 1755.13 - lr: 0.000008 - momentum: 0.000000
2024-03-26 10:31:31,084 epoch 9 - iter 72/95 - loss 0.03016438 - time (sec): 14.28 - samples/sec: 1790.00 - lr: 0.000007 - momentum: 0.000000
2024-03-26 10:31:33,856 epoch 9 - iter 81/95 - loss 0.03043791 - time (sec): 17.05 - samples/sec: 1741.38 - lr: 0.000007 - momentum: 0.000000
2024-03-26 10:31:35,497 epoch 9 - iter 90/95 - loss 0.02846557 - time (sec): 18.69 - samples/sec: 1762.25 - lr: 0.000006 - momentum: 0.000000
2024-03-26 10:31:36,157 ----------------------------------------------------------------------------------------------------
2024-03-26 10:31:36,158 EPOCH 9 done: loss 0.0293 - lr: 0.000006
2024-03-26 10:31:37,075 DEV : loss 0.17136143147945404 - f1-score (micro avg)  0.9443
2024-03-26 10:31:37,076 saving best model
2024-03-26 10:31:37,540 ----------------------------------------------------------------------------------------------------
2024-03-26 10:31:39,384 epoch 10 - iter 9/95 - loss 0.03543689 - time (sec): 1.84 - samples/sec: 1682.13 - lr: 0.000005 - momentum: 0.000000
2024-03-26 10:31:41,567 epoch 10 - iter 18/95 - loss 0.03398521 - time (sec): 4.03 - samples/sec: 1655.13 - lr: 0.000005 - momentum: 0.000000
2024-03-26 10:31:42,989 epoch 10 - iter 27/95 - loss 0.02864523 - time (sec): 5.45 - samples/sec: 1800.34 - lr: 0.000004 - momentum: 0.000000
2024-03-26 10:31:44,708 epoch 10 - iter 36/95 - loss 0.02526881 - time (sec): 7.17 - samples/sec: 1841.01 - lr: 0.000004 - momentum: 0.000000
2024-03-26 10:31:46,128 epoch 10 - iter 45/95 - loss 0.02433110 - time (sec): 8.59 - samples/sec: 1870.07 - lr: 0.000003 - momentum: 0.000000
2024-03-26 10:31:47,144 epoch 10 - iter 54/95 - loss 0.02218444 - time (sec): 9.60 - samples/sec: 1943.11 - lr: 0.000003 - momentum: 0.000000
2024-03-26 10:31:48,949 epoch 10 - iter 63/95 - loss 0.02021689 - time (sec): 11.41 - samples/sec: 1915.96 - lr: 0.000002 - momentum: 0.000000
2024-03-26 10:31:51,203 epoch 10 - iter 72/95 - loss 0.02517433 - time (sec): 13.66 - samples/sec: 1866.01 - lr: 0.000002 - momentum: 0.000000
2024-03-26 10:31:52,834 epoch 10 - iter 81/95 - loss 0.02584878 - time (sec): 15.29 - samples/sec: 1857.82 - lr: 0.000001 - momentum: 0.000000
2024-03-26 10:31:55,141 epoch 10 - iter 90/95 - loss 0.02442898 - time (sec): 17.60 - samples/sec: 1846.49 - lr: 0.000001 - momentum: 0.000000
2024-03-26 10:31:56,381 ----------------------------------------------------------------------------------------------------
2024-03-26 10:31:56,381 EPOCH 10 done: loss 0.0242 - lr: 0.000001
2024-03-26 10:31:57,328 DEV : loss 0.18190717697143555 - f1-score (micro avg)  0.9364
2024-03-26 10:31:57,605 ----------------------------------------------------------------------------------------------------
2024-03-26 10:31:57,605 Loading model from best epoch ...
2024-03-26 10:31:58,541 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 10:31:59,301 
Results:
- F-score (micro) 0.9155
- F-score (macro) 0.6953
- Accuracy 0.8466

By class:
              precision    recall  f1-score   support

 Unternehmen     0.9325    0.8835    0.9073       266
 Auslagerung     0.8702    0.9157    0.8924       249
         Ort     0.9708    0.9925    0.9815       134
    Software     0.0000    0.0000    0.0000         0

   micro avg     0.9127    0.9183    0.9155       649
   macro avg     0.6934    0.6979    0.6953       649
weighted avg     0.9165    0.9183    0.9169       649

2024-03-26 10:31:59,302 ----------------------------------------------------------------------------------------------------