Hugging Face's logo

stefan-it
/
flair-co-funer-german_bert_base-bs16-e10-lr3e-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-bs16-e10-lr3e-05-5
File size: 26,662 Bytes 
710c26b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
267
 
2024-03-26 12:02:56,881 ----------------------------------------------------------------------------------------------------
2024-03-26 12:02:56,881 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 12:02:56,881 ----------------------------------------------------------------------------------------------------
2024-03-26 12:02:56,881 Corpus: 758 train + 94 dev + 96 test sentences
2024-03-26 12:02:56,881 ----------------------------------------------------------------------------------------------------
2024-03-26 12:02:56,881 Train:  758 sentences
2024-03-26 12:02:56,881         (train_with_dev=False, train_with_test=False)
2024-03-26 12:02:56,881 ----------------------------------------------------------------------------------------------------
2024-03-26 12:02:56,881 Training Params:
2024-03-26 12:02:56,881  - learning_rate: "3e-05" 
2024-03-26 12:02:56,881  - mini_batch_size: "16"
2024-03-26 12:02:56,881  - max_epochs: "10"
2024-03-26 12:02:56,881  - shuffle: "True"
2024-03-26 12:02:56,881 ----------------------------------------------------------------------------------------------------
2024-03-26 12:02:56,881 Plugins:
2024-03-26 12:02:56,881  - TensorboardLogger
2024-03-26 12:02:56,881  - LinearScheduler | warmup_fraction: '0.1'
2024-03-26 12:02:56,881 ----------------------------------------------------------------------------------------------------
2024-03-26 12:02:56,881 Final evaluation on model from best epoch (best-model.pt)
2024-03-26 12:02:56,881  - metric: "('micro avg', 'f1-score')"
2024-03-26 12:02:56,881 ----------------------------------------------------------------------------------------------------
2024-03-26 12:02:56,881 Computation:
2024-03-26 12:02:56,881  - compute on device: cuda:0
2024-03-26 12:02:56,881  - embedding storage: none
2024-03-26 12:02:56,881 ----------------------------------------------------------------------------------------------------
2024-03-26 12:02:56,881 Model training base path: "flair-co-funer-german_bert_base-bs16-e10-lr3e-05-5"
2024-03-26 12:02:56,881 ----------------------------------------------------------------------------------------------------
2024-03-26 12:02:56,881 ----------------------------------------------------------------------------------------------------
2024-03-26 12:02:56,881 Logging anything other than scalars to TensorBoard is currently not supported.
2024-03-26 12:02:58,395 epoch 1 - iter 4/48 - loss 3.12129509 - time (sec): 1.51 - samples/sec: 1732.52 - lr: 0.000002 - momentum: 0.000000
2024-03-26 12:03:01,185 epoch 1 - iter 8/48 - loss 3.09794484 - time (sec): 4.30 - samples/sec: 1414.35 - lr: 0.000004 - momentum: 0.000000
2024-03-26 12:03:03,091 epoch 1 - iter 12/48 - loss 3.10138069 - time (sec): 6.21 - samples/sec: 1434.38 - lr: 0.000007 - momentum: 0.000000
2024-03-26 12:03:04,753 epoch 1 - iter 16/48 - loss 3.01723077 - time (sec): 7.87 - samples/sec: 1528.46 - lr: 0.000009 - momentum: 0.000000
2024-03-26 12:03:06,948 epoch 1 - iter 20/48 - loss 2.88339031 - time (sec): 10.07 - samples/sec: 1497.84 - lr: 0.000012 - momentum: 0.000000
2024-03-26 12:03:09,697 epoch 1 - iter 24/48 - loss 2.73555277 - time (sec): 12.82 - samples/sec: 1439.57 - lr: 0.000014 - momentum: 0.000000
2024-03-26 12:03:11,383 epoch 1 - iter 28/48 - loss 2.62931161 - time (sec): 14.50 - samples/sec: 1447.29 - lr: 0.000017 - momentum: 0.000000
2024-03-26 12:03:13,547 epoch 1 - iter 32/48 - loss 2.51120225 - time (sec): 16.67 - samples/sec: 1442.48 - lr: 0.000019 - momentum: 0.000000
2024-03-26 12:03:15,134 epoch 1 - iter 36/48 - loss 2.42560844 - time (sec): 18.25 - samples/sec: 1462.79 - lr: 0.000022 - momentum: 0.000000
2024-03-26 12:03:17,965 epoch 1 - iter 40/48 - loss 2.31439325 - time (sec): 21.08 - samples/sec: 1418.37 - lr: 0.000024 - momentum: 0.000000
2024-03-26 12:03:19,175 epoch 1 - iter 44/48 - loss 2.22836577 - time (sec): 22.29 - samples/sec: 1441.81 - lr: 0.000027 - momentum: 0.000000
2024-03-26 12:03:21,051 epoch 1 - iter 48/48 - loss 2.15578626 - time (sec): 24.17 - samples/sec: 1426.28 - lr: 0.000029 - momentum: 0.000000
2024-03-26 12:03:21,051 ----------------------------------------------------------------------------------------------------
2024-03-26 12:03:21,051 EPOCH 1 done: loss 2.1558 - lr: 0.000029
2024-03-26 12:03:21,894 DEV : loss 0.7538434863090515 - f1-score (micro avg)  0.5155
2024-03-26 12:03:21,895 saving best model
2024-03-26 12:03:22,152 ----------------------------------------------------------------------------------------------------
2024-03-26 12:03:24,936 epoch 2 - iter 4/48 - loss 0.89958872 - time (sec): 2.78 - samples/sec: 1240.73 - lr: 0.000030 - momentum: 0.000000
2024-03-26 12:03:26,870 epoch 2 - iter 8/48 - loss 0.83054446 - time (sec): 4.72 - samples/sec: 1299.11 - lr: 0.000030 - momentum: 0.000000
2024-03-26 12:03:28,843 epoch 2 - iter 12/48 - loss 0.78361854 - time (sec): 6.69 - samples/sec: 1333.57 - lr: 0.000029 - momentum: 0.000000
2024-03-26 12:03:31,588 epoch 2 - iter 16/48 - loss 0.70998971 - time (sec): 9.44 - samples/sec: 1341.21 - lr: 0.000029 - momentum: 0.000000
2024-03-26 12:03:33,023 epoch 2 - iter 20/48 - loss 0.67967096 - time (sec): 10.87 - samples/sec: 1378.25 - lr: 0.000029 - momentum: 0.000000
2024-03-26 12:03:35,936 epoch 2 - iter 24/48 - loss 0.63814992 - time (sec): 13.78 - samples/sec: 1298.51 - lr: 0.000028 - momentum: 0.000000
2024-03-26 12:03:37,606 epoch 2 - iter 28/48 - loss 0.62425146 - time (sec): 15.45 - samples/sec: 1330.52 - lr: 0.000028 - momentum: 0.000000
2024-03-26 12:03:39,660 epoch 2 - iter 32/48 - loss 0.59545641 - time (sec): 17.51 - samples/sec: 1326.21 - lr: 0.000028 - momentum: 0.000000
2024-03-26 12:03:41,485 epoch 2 - iter 36/48 - loss 0.58132540 - time (sec): 19.33 - samples/sec: 1355.74 - lr: 0.000028 - momentum: 0.000000
2024-03-26 12:03:43,881 epoch 2 - iter 40/48 - loss 0.57243950 - time (sec): 21.73 - samples/sec: 1347.17 - lr: 0.000027 - momentum: 0.000000
2024-03-26 12:03:46,129 epoch 2 - iter 44/48 - loss 0.55195339 - time (sec): 23.98 - samples/sec: 1352.77 - lr: 0.000027 - momentum: 0.000000
2024-03-26 12:03:47,408 epoch 2 - iter 48/48 - loss 0.54352570 - time (sec): 25.26 - samples/sec: 1364.94 - lr: 0.000027 - momentum: 0.000000
2024-03-26 12:03:47,408 ----------------------------------------------------------------------------------------------------
2024-03-26 12:03:47,408 EPOCH 2 done: loss 0.5435 - lr: 0.000027
2024-03-26 12:03:48,330 DEV : loss 0.33131173253059387 - f1-score (micro avg)  0.785
2024-03-26 12:03:48,331 saving best model
2024-03-26 12:03:48,763 ----------------------------------------------------------------------------------------------------
2024-03-26 12:03:49,867 epoch 3 - iter 4/48 - loss 0.42482105 - time (sec): 1.10 - samples/sec: 2024.56 - lr: 0.000026 - momentum: 0.000000
2024-03-26 12:03:51,807 epoch 3 - iter 8/48 - loss 0.40870155 - time (sec): 3.04 - samples/sec: 1619.65 - lr: 0.000026 - momentum: 0.000000
2024-03-26 12:03:54,064 epoch 3 - iter 12/48 - loss 0.35237639 - time (sec): 5.30 - samples/sec: 1618.21 - lr: 0.000026 - momentum: 0.000000
2024-03-26 12:03:56,038 epoch 3 - iter 16/48 - loss 0.34486620 - time (sec): 7.27 - samples/sec: 1563.96 - lr: 0.000026 - momentum: 0.000000
2024-03-26 12:03:57,966 epoch 3 - iter 20/48 - loss 0.33509474 - time (sec): 9.20 - samples/sec: 1540.83 - lr: 0.000025 - momentum: 0.000000
2024-03-26 12:03:59,958 epoch 3 - iter 24/48 - loss 0.31801538 - time (sec): 11.19 - samples/sec: 1497.60 - lr: 0.000025 - momentum: 0.000000
2024-03-26 12:04:03,262 epoch 3 - iter 28/48 - loss 0.30314619 - time (sec): 14.50 - samples/sec: 1380.74 - lr: 0.000025 - momentum: 0.000000
2024-03-26 12:04:04,816 epoch 3 - iter 32/48 - loss 0.30246473 - time (sec): 16.05 - samples/sec: 1403.32 - lr: 0.000025 - momentum: 0.000000
2024-03-26 12:04:08,198 epoch 3 - iter 36/48 - loss 0.29146006 - time (sec): 19.43 - samples/sec: 1334.78 - lr: 0.000024 - momentum: 0.000000
2024-03-26 12:04:10,657 epoch 3 - iter 40/48 - loss 0.28968058 - time (sec): 21.89 - samples/sec: 1336.63 - lr: 0.000024 - momentum: 0.000000
2024-03-26 12:04:12,775 epoch 3 - iter 44/48 - loss 0.28024638 - time (sec): 24.01 - samples/sec: 1336.44 - lr: 0.000024 - momentum: 0.000000
2024-03-26 12:04:14,398 epoch 3 - iter 48/48 - loss 0.27767314 - time (sec): 25.63 - samples/sec: 1344.83 - lr: 0.000023 - momentum: 0.000000
2024-03-26 12:04:14,398 ----------------------------------------------------------------------------------------------------
2024-03-26 12:04:14,398 EPOCH 3 done: loss 0.2777 - lr: 0.000023
2024-03-26 12:04:15,322 DEV : loss 0.2567653954029083 - f1-score (micro avg)  0.8532
2024-03-26 12:04:15,323 saving best model
2024-03-26 12:04:15,739 ----------------------------------------------------------------------------------------------------
2024-03-26 12:04:18,888 epoch 4 - iter 4/48 - loss 0.15456626 - time (sec): 3.15 - samples/sec: 1184.99 - lr: 0.000023 - momentum: 0.000000
2024-03-26 12:04:20,372 epoch 4 - iter 8/48 - loss 0.19317901 - time (sec): 4.63 - samples/sec: 1343.11 - lr: 0.000023 - momentum: 0.000000
2024-03-26 12:04:22,985 epoch 4 - iter 12/48 - loss 0.18252728 - time (sec): 7.24 - samples/sec: 1283.90 - lr: 0.000023 - momentum: 0.000000
2024-03-26 12:04:25,682 epoch 4 - iter 16/48 - loss 0.17118891 - time (sec): 9.94 - samples/sec: 1277.14 - lr: 0.000022 - momentum: 0.000000
2024-03-26 12:04:27,947 epoch 4 - iter 20/48 - loss 0.16696842 - time (sec): 12.21 - samples/sec: 1292.89 - lr: 0.000022 - momentum: 0.000000
2024-03-26 12:04:29,478 epoch 4 - iter 24/48 - loss 0.16439449 - time (sec): 13.74 - samples/sec: 1327.14 - lr: 0.000022 - momentum: 0.000000
2024-03-26 12:04:31,901 epoch 4 - iter 28/48 - loss 0.16802806 - time (sec): 16.16 - samples/sec: 1314.54 - lr: 0.000022 - momentum: 0.000000
2024-03-26 12:04:34,935 epoch 4 - iter 32/48 - loss 0.16616945 - time (sec): 19.19 - samples/sec: 1305.61 - lr: 0.000021 - momentum: 0.000000
2024-03-26 12:04:36,640 epoch 4 - iter 36/48 - loss 0.17149344 - time (sec): 20.90 - samples/sec: 1328.15 - lr: 0.000021 - momentum: 0.000000
2024-03-26 12:04:37,626 epoch 4 - iter 40/48 - loss 0.17449164 - time (sec): 21.88 - samples/sec: 1372.17 - lr: 0.000021 - momentum: 0.000000
2024-03-26 12:04:39,157 epoch 4 - iter 44/48 - loss 0.17147288 - time (sec): 23.42 - samples/sec: 1389.35 - lr: 0.000020 - momentum: 0.000000
2024-03-26 12:04:40,044 epoch 4 - iter 48/48 - loss 0.17402778 - time (sec): 24.30 - samples/sec: 1418.47 - lr: 0.000020 - momentum: 0.000000
2024-03-26 12:04:40,044 ----------------------------------------------------------------------------------------------------
2024-03-26 12:04:40,044 EPOCH 4 done: loss 0.1740 - lr: 0.000020
2024-03-26 12:04:40,972 DEV : loss 0.23388732969760895 - f1-score (micro avg)  0.8688
2024-03-26 12:04:40,973 saving best model
2024-03-26 12:04:41,384 ----------------------------------------------------------------------------------------------------
2024-03-26 12:04:43,222 epoch 5 - iter 4/48 - loss 0.16130966 - time (sec): 1.84 - samples/sec: 1564.18 - lr: 0.000020 - momentum: 0.000000
2024-03-26 12:04:45,120 epoch 5 - iter 8/48 - loss 0.13106811 - time (sec): 3.73 - samples/sec: 1661.67 - lr: 0.000020 - momentum: 0.000000
2024-03-26 12:04:48,241 epoch 5 - iter 12/48 - loss 0.12117448 - time (sec): 6.86 - samples/sec: 1403.40 - lr: 0.000019 - momentum: 0.000000
2024-03-26 12:04:49,616 epoch 5 - iter 16/48 - loss 0.11428761 - time (sec): 8.23 - samples/sec: 1444.70 - lr: 0.000019 - momentum: 0.000000
2024-03-26 12:04:52,012 epoch 5 - iter 20/48 - loss 0.12659685 - time (sec): 10.63 - samples/sec: 1418.01 - lr: 0.000019 - momentum: 0.000000
2024-03-26 12:04:54,173 epoch 5 - iter 24/48 - loss 0.12691480 - time (sec): 12.79 - samples/sec: 1390.07 - lr: 0.000018 - momentum: 0.000000
2024-03-26 12:04:55,597 epoch 5 - iter 28/48 - loss 0.13411938 - time (sec): 14.21 - samples/sec: 1427.90 - lr: 0.000018 - momentum: 0.000000
2024-03-26 12:04:57,007 epoch 5 - iter 32/48 - loss 0.13634792 - time (sec): 15.62 - samples/sec: 1459.36 - lr: 0.000018 - momentum: 0.000000
2024-03-26 12:04:59,293 epoch 5 - iter 36/48 - loss 0.13690522 - time (sec): 17.91 - samples/sec: 1442.07 - lr: 0.000018 - momentum: 0.000000
2024-03-26 12:05:01,169 epoch 5 - iter 40/48 - loss 0.13384191 - time (sec): 19.78 - samples/sec: 1441.02 - lr: 0.000017 - momentum: 0.000000
2024-03-26 12:05:03,269 epoch 5 - iter 44/48 - loss 0.13274461 - time (sec): 21.88 - samples/sec: 1450.66 - lr: 0.000017 - momentum: 0.000000
2024-03-26 12:05:05,441 epoch 5 - iter 48/48 - loss 0.12821407 - time (sec): 24.06 - samples/sec: 1433.05 - lr: 0.000017 - momentum: 0.000000
2024-03-26 12:05:05,441 ----------------------------------------------------------------------------------------------------
2024-03-26 12:05:05,441 EPOCH 5 done: loss 0.1282 - lr: 0.000017
2024-03-26 12:05:06,374 DEV : loss 0.2178356647491455 - f1-score (micro avg)  0.893
2024-03-26 12:05:06,375 saving best model
2024-03-26 12:05:06,798 ----------------------------------------------------------------------------------------------------
2024-03-26 12:05:08,730 epoch 6 - iter 4/48 - loss 0.09708101 - time (sec): 1.93 - samples/sec: 1421.89 - lr: 0.000017 - momentum: 0.000000
2024-03-26 12:05:11,610 epoch 6 - iter 8/48 - loss 0.11059477 - time (sec): 4.81 - samples/sec: 1321.14 - lr: 0.000016 - momentum: 0.000000
2024-03-26 12:05:13,501 epoch 6 - iter 12/48 - loss 0.11422747 - time (sec): 6.70 - samples/sec: 1347.27 - lr: 0.000016 - momentum: 0.000000
2024-03-26 12:05:15,088 epoch 6 - iter 16/48 - loss 0.11763326 - time (sec): 8.29 - samples/sec: 1395.80 - lr: 0.000016 - momentum: 0.000000
2024-03-26 12:05:17,827 epoch 6 - iter 20/48 - loss 0.11242158 - time (sec): 11.03 - samples/sec: 1322.13 - lr: 0.000015 - momentum: 0.000000
2024-03-26 12:05:20,563 epoch 6 - iter 24/48 - loss 0.10277800 - time (sec): 13.76 - samples/sec: 1298.04 - lr: 0.000015 - momentum: 0.000000
2024-03-26 12:05:23,079 epoch 6 - iter 28/48 - loss 0.10001828 - time (sec): 16.28 - samples/sec: 1274.53 - lr: 0.000015 - momentum: 0.000000
2024-03-26 12:05:24,485 epoch 6 - iter 32/48 - loss 0.10667881 - time (sec): 17.69 - samples/sec: 1318.22 - lr: 0.000015 - momentum: 0.000000
2024-03-26 12:05:26,389 epoch 6 - iter 36/48 - loss 0.10414212 - time (sec): 19.59 - samples/sec: 1329.32 - lr: 0.000014 - momentum: 0.000000
2024-03-26 12:05:27,410 epoch 6 - iter 40/48 - loss 0.10365325 - time (sec): 20.61 - samples/sec: 1368.68 - lr: 0.000014 - momentum: 0.000000
2024-03-26 12:05:29,930 epoch 6 - iter 44/48 - loss 0.10035451 - time (sec): 23.13 - samples/sec: 1345.14 - lr: 0.000014 - momentum: 0.000000
2024-03-26 12:05:32,791 epoch 6 - iter 48/48 - loss 0.09524719 - time (sec): 25.99 - samples/sec: 1326.28 - lr: 0.000014 - momentum: 0.000000
2024-03-26 12:05:32,791 ----------------------------------------------------------------------------------------------------
2024-03-26 12:05:32,791 EPOCH 6 done: loss 0.0952 - lr: 0.000014
2024-03-26 12:05:33,740 DEV : loss 0.19964341819286346 - f1-score (micro avg)  0.8942
2024-03-26 12:05:33,741 saving best model
2024-03-26 12:05:34,172 ----------------------------------------------------------------------------------------------------
2024-03-26 12:05:36,357 epoch 7 - iter 4/48 - loss 0.05253045 - time (sec): 2.18 - samples/sec: 1333.19 - lr: 0.000013 - momentum: 0.000000
2024-03-26 12:05:38,086 epoch 7 - iter 8/48 - loss 0.05348429 - time (sec): 3.91 - samples/sec: 1362.53 - lr: 0.000013 - momentum: 0.000000
2024-03-26 12:05:39,541 epoch 7 - iter 12/48 - loss 0.08415717 - time (sec): 5.37 - samples/sec: 1414.39 - lr: 0.000013 - momentum: 0.000000
2024-03-26 12:05:41,498 epoch 7 - iter 16/48 - loss 0.07694373 - time (sec): 7.32 - samples/sec: 1449.66 - lr: 0.000012 - momentum: 0.000000
2024-03-26 12:05:43,817 epoch 7 - iter 20/48 - loss 0.09023838 - time (sec): 9.64 - samples/sec: 1502.91 - lr: 0.000012 - momentum: 0.000000
2024-03-26 12:05:45,166 epoch 7 - iter 24/48 - loss 0.08683075 - time (sec): 10.99 - samples/sec: 1550.09 - lr: 0.000012 - momentum: 0.000000
2024-03-26 12:05:47,411 epoch 7 - iter 28/48 - loss 0.08520615 - time (sec): 13.24 - samples/sec: 1506.85 - lr: 0.000012 - momentum: 0.000000
2024-03-26 12:05:49,331 epoch 7 - iter 32/48 - loss 0.08645376 - time (sec): 15.16 - samples/sec: 1502.21 - lr: 0.000011 - momentum: 0.000000
2024-03-26 12:05:51,355 epoch 7 - iter 36/48 - loss 0.08423135 - time (sec): 17.18 - samples/sec: 1471.25 - lr: 0.000011 - momentum: 0.000000
2024-03-26 12:05:54,206 epoch 7 - iter 40/48 - loss 0.08031771 - time (sec): 20.03 - samples/sec: 1453.99 - lr: 0.000011 - momentum: 0.000000
2024-03-26 12:05:55,736 epoch 7 - iter 44/48 - loss 0.08069773 - time (sec): 21.56 - samples/sec: 1468.81 - lr: 0.000010 - momentum: 0.000000
2024-03-26 12:05:57,912 epoch 7 - iter 48/48 - loss 0.07961125 - time (sec): 23.74 - samples/sec: 1452.24 - lr: 0.000010 - momentum: 0.000000
2024-03-26 12:05:57,912 ----------------------------------------------------------------------------------------------------
2024-03-26 12:05:57,912 EPOCH 7 done: loss 0.0796 - lr: 0.000010
2024-03-26 12:05:58,929 DEV : loss 0.2036362737417221 - f1-score (micro avg)  0.9066
2024-03-26 12:05:58,930 saving best model
2024-03-26 12:05:59,354 ----------------------------------------------------------------------------------------------------
2024-03-26 12:06:01,640 epoch 8 - iter 4/48 - loss 0.10170390 - time (sec): 2.28 - samples/sec: 1221.01 - lr: 0.000010 - momentum: 0.000000
2024-03-26 12:06:03,208 epoch 8 - iter 8/48 - loss 0.06787866 - time (sec): 3.85 - samples/sec: 1411.19 - lr: 0.000010 - momentum: 0.000000
2024-03-26 12:06:06,239 epoch 8 - iter 12/48 - loss 0.05875894 - time (sec): 6.88 - samples/sec: 1307.39 - lr: 0.000009 - momentum: 0.000000
2024-03-26 12:06:08,665 epoch 8 - iter 16/48 - loss 0.05794682 - time (sec): 9.31 - samples/sec: 1319.75 - lr: 0.000009 - momentum: 0.000000
2024-03-26 12:06:10,144 epoch 8 - iter 20/48 - loss 0.05562347 - time (sec): 10.79 - samples/sec: 1377.00 - lr: 0.000009 - momentum: 0.000000
2024-03-26 12:06:11,562 epoch 8 - iter 24/48 - loss 0.05757691 - time (sec): 12.21 - samples/sec: 1449.59 - lr: 0.000009 - momentum: 0.000000
2024-03-26 12:06:12,869 epoch 8 - iter 28/48 - loss 0.05949732 - time (sec): 13.51 - samples/sec: 1513.12 - lr: 0.000008 - momentum: 0.000000
2024-03-26 12:06:15,170 epoch 8 - iter 32/48 - loss 0.06077559 - time (sec): 15.81 - samples/sec: 1464.99 - lr: 0.000008 - momentum: 0.000000
2024-03-26 12:06:17,741 epoch 8 - iter 36/48 - loss 0.05962745 - time (sec): 18.38 - samples/sec: 1424.12 - lr: 0.000008 - momentum: 0.000000
2024-03-26 12:06:19,780 epoch 8 - iter 40/48 - loss 0.06048251 - time (sec): 20.42 - samples/sec: 1428.96 - lr: 0.000007 - momentum: 0.000000
2024-03-26 12:06:21,921 epoch 8 - iter 44/48 - loss 0.06214127 - time (sec): 22.56 - samples/sec: 1415.58 - lr: 0.000007 - momentum: 0.000000
2024-03-26 12:06:23,545 epoch 8 - iter 48/48 - loss 0.06199253 - time (sec): 24.19 - samples/sec: 1425.12 - lr: 0.000007 - momentum: 0.000000
2024-03-26 12:06:23,545 ----------------------------------------------------------------------------------------------------
2024-03-26 12:06:23,545 EPOCH 8 done: loss 0.0620 - lr: 0.000007
2024-03-26 12:06:24,474 DEV : loss 0.18240593373775482 - f1-score (micro avg)  0.9167
2024-03-26 12:06:24,475 saving best model
2024-03-26 12:06:24,909 ----------------------------------------------------------------------------------------------------
2024-03-26 12:06:27,685 epoch 9 - iter 4/48 - loss 0.05441554 - time (sec): 2.77 - samples/sec: 1262.13 - lr: 0.000007 - momentum: 0.000000
2024-03-26 12:06:29,758 epoch 9 - iter 8/48 - loss 0.04400094 - time (sec): 4.85 - samples/sec: 1317.94 - lr: 0.000006 - momentum: 0.000000
2024-03-26 12:06:32,758 epoch 9 - iter 12/48 - loss 0.04559467 - time (sec): 7.85 - samples/sec: 1239.88 - lr: 0.000006 - momentum: 0.000000
2024-03-26 12:06:35,906 epoch 9 - iter 16/48 - loss 0.05720080 - time (sec): 10.99 - samples/sec: 1222.91 - lr: 0.000006 - momentum: 0.000000
2024-03-26 12:06:36,775 epoch 9 - iter 20/48 - loss 0.05506132 - time (sec): 11.86 - samples/sec: 1313.05 - lr: 0.000006 - momentum: 0.000000
2024-03-26 12:06:38,705 epoch 9 - iter 24/48 - loss 0.05366432 - time (sec): 13.79 - samples/sec: 1306.91 - lr: 0.000005 - momentum: 0.000000
2024-03-26 12:06:40,743 epoch 9 - iter 28/48 - loss 0.05227759 - time (sec): 15.83 - samples/sec: 1322.70 - lr: 0.000005 - momentum: 0.000000
2024-03-26 12:06:41,781 epoch 9 - iter 32/48 - loss 0.05472457 - time (sec): 16.87 - samples/sec: 1384.85 - lr: 0.000005 - momentum: 0.000000
2024-03-26 12:06:42,962 epoch 9 - iter 36/48 - loss 0.05406804 - time (sec): 18.05 - samples/sec: 1435.35 - lr: 0.000004 - momentum: 0.000000
2024-03-26 12:06:44,289 epoch 9 - iter 40/48 - loss 0.05402247 - time (sec): 19.38 - samples/sec: 1463.70 - lr: 0.000004 - momentum: 0.000000
2024-03-26 12:06:47,367 epoch 9 - iter 44/48 - loss 0.05348785 - time (sec): 22.46 - samples/sec: 1436.47 - lr: 0.000004 - momentum: 0.000000
2024-03-26 12:06:48,928 epoch 9 - iter 48/48 - loss 0.05219946 - time (sec): 24.02 - samples/sec: 1435.34 - lr: 0.000004 - momentum: 0.000000
2024-03-26 12:06:48,928 ----------------------------------------------------------------------------------------------------
2024-03-26 12:06:48,928 EPOCH 9 done: loss 0.0522 - lr: 0.000004
2024-03-26 12:06:49,863 DEV : loss 0.18798069655895233 - f1-score (micro avg)  0.9247
2024-03-26 12:06:49,864 saving best model
2024-03-26 12:06:50,308 ----------------------------------------------------------------------------------------------------
2024-03-26 12:06:53,223 epoch 10 - iter 4/48 - loss 0.04554635 - time (sec): 2.91 - samples/sec: 1274.34 - lr: 0.000003 - momentum: 0.000000
2024-03-26 12:06:55,200 epoch 10 - iter 8/48 - loss 0.04254462 - time (sec): 4.89 - samples/sec: 1321.40 - lr: 0.000003 - momentum: 0.000000
2024-03-26 12:06:57,403 epoch 10 - iter 12/48 - loss 0.04412043 - time (sec): 7.09 - samples/sec: 1282.08 - lr: 0.000003 - momentum: 0.000000
2024-03-26 12:06:59,990 epoch 10 - iter 16/48 - loss 0.04292351 - time (sec): 9.68 - samples/sec: 1233.56 - lr: 0.000002 - momentum: 0.000000
2024-03-26 12:07:02,611 epoch 10 - iter 20/48 - loss 0.04280469 - time (sec): 12.30 - samples/sec: 1237.58 - lr: 0.000002 - momentum: 0.000000
2024-03-26 12:07:04,033 epoch 10 - iter 24/48 - loss 0.04266044 - time (sec): 13.72 - samples/sec: 1299.71 - lr: 0.000002 - momentum: 0.000000
2024-03-26 12:07:04,956 epoch 10 - iter 28/48 - loss 0.04409514 - time (sec): 14.65 - samples/sec: 1367.63 - lr: 0.000002 - momentum: 0.000000
2024-03-26 12:07:06,939 epoch 10 - iter 32/48 - loss 0.04788591 - time (sec): 16.63 - samples/sec: 1386.99 - lr: 0.000001 - momentum: 0.000000
2024-03-26 12:07:09,238 epoch 10 - iter 36/48 - loss 0.04807467 - time (sec): 18.93 - samples/sec: 1366.69 - lr: 0.000001 - momentum: 0.000000
2024-03-26 12:07:10,954 epoch 10 - iter 40/48 - loss 0.04909381 - time (sec): 20.64 - samples/sec: 1391.12 - lr: 0.000001 - momentum: 0.000000
2024-03-26 12:07:14,181 epoch 10 - iter 44/48 - loss 0.04792634 - time (sec): 23.87 - samples/sec: 1372.93 - lr: 0.000001 - momentum: 0.000000
2024-03-26 12:07:14,899 epoch 10 - iter 48/48 - loss 0.04830699 - time (sec): 24.59 - samples/sec: 1401.87 - lr: 0.000000 - momentum: 0.000000
2024-03-26 12:07:14,900 ----------------------------------------------------------------------------------------------------
2024-03-26 12:07:14,900 EPOCH 10 done: loss 0.0483 - lr: 0.000000
2024-03-26 12:07:15,835 DEV : loss 0.18821127712726593 - f1-score (micro avg)  0.9204
2024-03-26 12:07:16,096 ----------------------------------------------------------------------------------------------------
2024-03-26 12:07:16,097 Loading model from best epoch ...
2024-03-26 12:07:16,803 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 12:07:17,653 
Results:
- F-score (micro) 0.9094
- F-score (macro) 0.6907
- Accuracy 0.8362

By class:
              precision    recall  f1-score   support

 Unternehmen     0.8923    0.8722    0.8821       266
 Auslagerung     0.8976    0.9157    0.9066       249
         Ort     0.9635    0.9851    0.9742       134
    Software     0.0000    0.0000    0.0000         0

   micro avg     0.9066    0.9122    0.9094       649
   macro avg     0.6884    0.6932    0.6907       649
weighted avg     0.9091    0.9122    0.9105       649

2024-03-26 12:07:17,653 ----------------------------------------------------------------------------------------------------