Hugging Face's logo

stefan-it
/
flair-co-funer-german_bert_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
2
stefan-it commited on
Commit
d000a19
·
verified ·
1 Parent(s): 197f0e7

Upload ./training.log with huggingface_hub

Browse files
Files changed (1) hide show
  1. training.log +244 -0
training.log ADDED
@@ -0,0 +1,244 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2024-03-26 11:59:09,057 ----------------------------------------------------------------------------------------------------
2
+ 2024-03-26 11:59:09,057 Model: "SequenceTagger(
3
+ (embeddings): TransformerWordEmbeddings(
4
+ (model): BertModel(
5
+ (embeddings): BertEmbeddings(
6
+ (word_embeddings): Embedding(30001, 768)
7
+ (position_embeddings): Embedding(512, 768)
8
+ (token_type_embeddings): Embedding(2, 768)
9
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
10
+ (dropout): Dropout(p=0.1, inplace=False)
11
+ )
12
+ (encoder): BertEncoder(
13
+ (layer): ModuleList(
14
+ (0-11): 12 x BertLayer(
15
+ (attention): BertAttention(
16
+ (self): BertSelfAttention(
17
+ (query): Linear(in_features=768, out_features=768, bias=True)
18
+ (key): Linear(in_features=768, out_features=768, bias=True)
19
+ (value): Linear(in_features=768, out_features=768, bias=True)
20
+ (dropout): Dropout(p=0.1, inplace=False)
21
+ )
22
+ (output): BertSelfOutput(
23
+ (dense): Linear(in_features=768, out_features=768, bias=True)
24
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
25
+ (dropout): Dropout(p=0.1, inplace=False)
26
+ )
27
+ )
28
+ (intermediate): BertIntermediate(
29
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
30
+ (intermediate_act_fn): GELUActivation()
31
+ )
32
+ (output): BertOutput(
33
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
34
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
35
+ (dropout): Dropout(p=0.1, inplace=False)
36
+ )
37
+ )
38
+ )
39
+ )
40
+ (pooler): BertPooler(
41
+ (dense): Linear(in_features=768, out_features=768, bias=True)
42
+ (activation): Tanh()
43
+ )
44
+ )
45
+ )
46
+ (locked_dropout): LockedDropout(p=0.5)
47
+ (linear): Linear(in_features=768, out_features=17, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2024-03-26 11:59:09,057 ----------------------------------------------------------------------------------------------------
51
+ 2024-03-26 11:59:09,057 Corpus: 758 train + 94 dev + 96 test sentences
52
+ 2024-03-26 11:59:09,057 ----------------------------------------------------------------------------------------------------
53
+ 2024-03-26 11:59:09,057 Train: 758 sentences
54
+ 2024-03-26 11:59:09,057 (train_with_dev=False, train_with_test=False)
55
+ 2024-03-26 11:59:09,057 ----------------------------------------------------------------------------------------------------
56
+ 2024-03-26 11:59:09,057 Training Params:
57
+ 2024-03-26 11:59:09,057 - learning_rate: "5e-05"
58
+ 2024-03-26 11:59:09,057 - mini_batch_size: "8"
59
+ 2024-03-26 11:59:09,057 - max_epochs: "10"
60
+ 2024-03-26 11:59:09,057 - shuffle: "True"
61
+ 2024-03-26 11:59:09,057 ----------------------------------------------------------------------------------------------------
62
+ 2024-03-26 11:59:09,057 Plugins:
63
+ 2024-03-26 11:59:09,057 - TensorboardLogger
64
+ 2024-03-26 11:59:09,057 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2024-03-26 11:59:09,057 ----------------------------------------------------------------------------------------------------
66
+ 2024-03-26 11:59:09,057 Final evaluation on model from best epoch (best-model.pt)
67
+ 2024-03-26 11:59:09,057 - metric: "('micro avg', 'f1-score')"
68
+ 2024-03-26 11:59:09,057 ----------------------------------------------------------------------------------------------------
69
+ 2024-03-26 11:59:09,057 Computation:
70
+ 2024-03-26 11:59:09,057 - compute on device: cuda:0
71
+ 2024-03-26 11:59:09,057 - embedding storage: none
72
+ 2024-03-26 11:59:09,057 ----------------------------------------------------------------------------------------------------
73
+ 2024-03-26 11:59:09,057 Model training base path: "flair-co-funer-german_bert_base-bs8-e10-lr5e-05-4"
74
+ 2024-03-26 11:59:09,057 ----------------------------------------------------------------------------------------------------
75
+ 2024-03-26 11:59:09,057 ----------------------------------------------------------------------------------------------------
76
+ 2024-03-26 11:59:09,058 Logging anything other than scalars to TensorBoard is currently not supported.
77
+ 2024-03-26 11:59:10,418 epoch 1 - iter 9/95 - loss 3.11551508 - time (sec): 1.36 - samples/sec: 2132.11 - lr: 0.000004 - momentum: 0.000000
78
+ 2024-03-26 11:59:11,850 epoch 1 - iter 18/95 - loss 2.91592715 - time (sec): 2.79 - samples/sec: 1975.14 - lr: 0.000009 - momentum: 0.000000
79
+ 2024-03-26 11:59:13,523 epoch 1 - iter 27/95 - loss 2.63491605 - time (sec): 4.46 - samples/sec: 1920.85 - lr: 0.000014 - momentum: 0.000000
80
+ 2024-03-26 11:59:15,532 epoch 1 - iter 36/95 - loss 2.39303542 - time (sec): 6.47 - samples/sec: 1843.82 - lr: 0.000018 - momentum: 0.000000
81
+ 2024-03-26 11:59:17,444 epoch 1 - iter 45/95 - loss 2.19445986 - time (sec): 8.39 - samples/sec: 1859.55 - lr: 0.000023 - momentum: 0.000000
82
+ 2024-03-26 11:59:19,715 epoch 1 - iter 54/95 - loss 2.01807178 - time (sec): 10.66 - samples/sec: 1796.53 - lr: 0.000028 - momentum: 0.000000
83
+ 2024-03-26 11:59:21,763 epoch 1 - iter 63/95 - loss 1.85928116 - time (sec): 12.71 - samples/sec: 1778.94 - lr: 0.000033 - momentum: 0.000000
84
+ 2024-03-26 11:59:22,764 epoch 1 - iter 72/95 - loss 1.76634872 - time (sec): 13.71 - samples/sec: 1822.57 - lr: 0.000037 - momentum: 0.000000
85
+ 2024-03-26 11:59:25,096 epoch 1 - iter 81/95 - loss 1.62019393 - time (sec): 16.04 - samples/sec: 1771.09 - lr: 0.000042 - momentum: 0.000000
86
+ 2024-03-26 11:59:26,515 epoch 1 - iter 90/95 - loss 1.49289861 - time (sec): 17.46 - samples/sec: 1830.77 - lr: 0.000047 - momentum: 0.000000
87
+ 2024-03-26 11:59:27,817 ----------------------------------------------------------------------------------------------------
88
+ 2024-03-26 11:59:27,817 EPOCH 1 done: loss 1.4311 - lr: 0.000047
89
+ 2024-03-26 11:59:28,673 DEV : loss 0.3974745273590088 - f1-score (micro avg) 0.7207
90
+ 2024-03-26 11:59:28,674 saving best model
91
+ 2024-03-26 11:59:28,954 ----------------------------------------------------------------------------------------------------
92
+ 2024-03-26 11:59:30,567 epoch 2 - iter 9/95 - loss 0.51132915 - time (sec): 1.61 - samples/sec: 1787.31 - lr: 0.000050 - momentum: 0.000000
93
+ 2024-03-26 11:59:32,312 epoch 2 - iter 18/95 - loss 0.41287581 - time (sec): 3.36 - samples/sec: 1840.74 - lr: 0.000049 - momentum: 0.000000
94
+ 2024-03-26 11:59:34,170 epoch 2 - iter 27/95 - loss 0.38177177 - time (sec): 5.22 - samples/sec: 1813.09 - lr: 0.000048 - momentum: 0.000000
95
+ 2024-03-26 11:59:36,623 epoch 2 - iter 36/95 - loss 0.33454736 - time (sec): 7.67 - samples/sec: 1701.24 - lr: 0.000048 - momentum: 0.000000
96
+ 2024-03-26 11:59:38,634 epoch 2 - iter 45/95 - loss 0.32462192 - time (sec): 9.68 - samples/sec: 1704.50 - lr: 0.000047 - momentum: 0.000000
97
+ 2024-03-26 11:59:40,436 epoch 2 - iter 54/95 - loss 0.33859423 - time (sec): 11.48 - samples/sec: 1727.29 - lr: 0.000047 - momentum: 0.000000
98
+ 2024-03-26 11:59:42,905 epoch 2 - iter 63/95 - loss 0.32027748 - time (sec): 13.95 - samples/sec: 1713.90 - lr: 0.000046 - momentum: 0.000000
99
+ 2024-03-26 11:59:44,796 epoch 2 - iter 72/95 - loss 0.31809533 - time (sec): 15.84 - samples/sec: 1707.05 - lr: 0.000046 - momentum: 0.000000
100
+ 2024-03-26 11:59:47,076 epoch 2 - iter 81/95 - loss 0.30946687 - time (sec): 18.12 - samples/sec: 1688.25 - lr: 0.000045 - momentum: 0.000000
101
+ 2024-03-26 11:59:48,400 epoch 2 - iter 90/95 - loss 0.30584654 - time (sec): 19.45 - samples/sec: 1713.26 - lr: 0.000045 - momentum: 0.000000
102
+ 2024-03-26 11:59:48,844 ----------------------------------------------------------------------------------------------------
103
+ 2024-03-26 11:59:48,845 EPOCH 2 done: loss 0.3017 - lr: 0.000045
104
+ 2024-03-26 11:59:49,798 DEV : loss 0.24508343636989594 - f1-score (micro avg) 0.8649
105
+ 2024-03-26 11:59:49,800 saving best model
106
+ 2024-03-26 11:59:50,246 ----------------------------------------------------------------------------------------------------
107
+ 2024-03-26 11:59:51,810 epoch 3 - iter 9/95 - loss 0.17662022 - time (sec): 1.56 - samples/sec: 1694.57 - lr: 0.000044 - momentum: 0.000000
108
+ 2024-03-26 11:59:53,538 epoch 3 - iter 18/95 - loss 0.14946269 - time (sec): 3.29 - samples/sec: 1697.74 - lr: 0.000043 - momentum: 0.000000
109
+ 2024-03-26 11:59:55,440 epoch 3 - iter 27/95 - loss 0.17150757 - time (sec): 5.19 - samples/sec: 1714.83 - lr: 0.000043 - momentum: 0.000000
110
+ 2024-03-26 11:59:57,385 epoch 3 - iter 36/95 - loss 0.17605043 - time (sec): 7.14 - samples/sec: 1690.97 - lr: 0.000042 - momentum: 0.000000
111
+ 2024-03-26 11:59:59,448 epoch 3 - iter 45/95 - loss 0.17196591 - time (sec): 9.20 - samples/sec: 1715.95 - lr: 0.000042 - momentum: 0.000000
112
+ 2024-03-26 12:00:01,784 epoch 3 - iter 54/95 - loss 0.16879250 - time (sec): 11.54 - samples/sec: 1678.78 - lr: 0.000041 - momentum: 0.000000
113
+ 2024-03-26 12:00:03,576 epoch 3 - iter 63/95 - loss 0.16109343 - time (sec): 13.33 - samples/sec: 1678.31 - lr: 0.000041 - momentum: 0.000000
114
+ 2024-03-26 12:00:05,589 epoch 3 - iter 72/95 - loss 0.15752743 - time (sec): 15.34 - samples/sec: 1687.28 - lr: 0.000040 - momentum: 0.000000
115
+ 2024-03-26 12:00:07,623 epoch 3 - iter 81/95 - loss 0.16724057 - time (sec): 17.38 - samples/sec: 1704.42 - lr: 0.000040 - momentum: 0.000000
116
+ 2024-03-26 12:00:09,915 epoch 3 - iter 90/95 - loss 0.16469999 - time (sec): 19.67 - samples/sec: 1686.10 - lr: 0.000039 - momentum: 0.000000
117
+ 2024-03-26 12:00:10,534 ----------------------------------------------------------------------------------------------------
118
+ 2024-03-26 12:00:10,534 EPOCH 3 done: loss 0.1671 - lr: 0.000039
119
+ 2024-03-26 12:00:11,475 DEV : loss 0.20654602348804474 - f1-score (micro avg) 0.866
120
+ 2024-03-26 12:00:11,476 saving best model
121
+ 2024-03-26 12:00:11,925 ----------------------------------------------------------------------------------------------------
122
+ 2024-03-26 12:00:14,379 epoch 4 - iter 9/95 - loss 0.06959102 - time (sec): 2.45 - samples/sec: 1612.63 - lr: 0.000039 - momentum: 0.000000
123
+ 2024-03-26 12:00:15,556 epoch 4 - iter 18/95 - loss 0.07924386 - time (sec): 3.63 - samples/sec: 1763.18 - lr: 0.000038 - momentum: 0.000000
124
+ 2024-03-26 12:00:17,785 epoch 4 - iter 27/95 - loss 0.09207853 - time (sec): 5.86 - samples/sec: 1774.88 - lr: 0.000037 - momentum: 0.000000
125
+ 2024-03-26 12:00:19,293 epoch 4 - iter 36/95 - loss 0.09786047 - time (sec): 7.37 - samples/sec: 1810.88 - lr: 0.000037 - momentum: 0.000000
126
+ 2024-03-26 12:00:20,634 epoch 4 - iter 45/95 - loss 0.09845056 - time (sec): 8.71 - samples/sec: 1842.74 - lr: 0.000036 - momentum: 0.000000
127
+ 2024-03-26 12:00:22,742 epoch 4 - iter 54/95 - loss 0.09663956 - time (sec): 10.82 - samples/sec: 1789.42 - lr: 0.000036 - momentum: 0.000000
128
+ 2024-03-26 12:00:25,103 epoch 4 - iter 63/95 - loss 0.10785088 - time (sec): 13.18 - samples/sec: 1757.00 - lr: 0.000035 - momentum: 0.000000
129
+ 2024-03-26 12:00:26,614 epoch 4 - iter 72/95 - loss 0.10490869 - time (sec): 14.69 - samples/sec: 1789.10 - lr: 0.000035 - momentum: 0.000000
130
+ 2024-03-26 12:00:28,246 epoch 4 - iter 81/95 - loss 0.10385446 - time (sec): 16.32 - samples/sec: 1819.91 - lr: 0.000034 - momentum: 0.000000
131
+ 2024-03-26 12:00:29,896 epoch 4 - iter 90/95 - loss 0.10468697 - time (sec): 17.97 - samples/sec: 1847.82 - lr: 0.000034 - momentum: 0.000000
132
+ 2024-03-26 12:00:30,524 ----------------------------------------------------------------------------------------------------
133
+ 2024-03-26 12:00:30,524 EPOCH 4 done: loss 0.1058 - lr: 0.000034
134
+ 2024-03-26 12:00:31,464 DEV : loss 0.20639045536518097 - f1-score (micro avg) 0.9061
135
+ 2024-03-26 12:00:31,465 saving best model
136
+ 2024-03-26 12:00:31,899 ----------------------------------------------------------------------------------------------------
137
+ 2024-03-26 12:00:33,151 epoch 5 - iter 9/95 - loss 0.11849622 - time (sec): 1.25 - samples/sec: 2367.15 - lr: 0.000033 - momentum: 0.000000
138
+ 2024-03-26 12:00:34,644 epoch 5 - iter 18/95 - loss 0.09990788 - time (sec): 2.74 - samples/sec: 2125.68 - lr: 0.000032 - momentum: 0.000000
139
+ 2024-03-26 12:00:36,707 epoch 5 - iter 27/95 - loss 0.08403115 - time (sec): 4.80 - samples/sec: 1913.94 - lr: 0.000032 - momentum: 0.000000
140
+ 2024-03-26 12:00:39,162 epoch 5 - iter 36/95 - loss 0.08445462 - time (sec): 7.26 - samples/sec: 1749.59 - lr: 0.000031 - momentum: 0.000000
141
+ 2024-03-26 12:00:40,406 epoch 5 - iter 45/95 - loss 0.09122409 - time (sec): 8.50 - samples/sec: 1796.12 - lr: 0.000031 - momentum: 0.000000
142
+ 2024-03-26 12:00:42,378 epoch 5 - iter 54/95 - loss 0.08649690 - time (sec): 10.48 - samples/sec: 1827.16 - lr: 0.000030 - momentum: 0.000000
143
+ 2024-03-26 12:00:44,512 epoch 5 - iter 63/95 - loss 0.08178669 - time (sec): 12.61 - samples/sec: 1812.39 - lr: 0.000030 - momentum: 0.000000
144
+ 2024-03-26 12:00:45,811 epoch 5 - iter 72/95 - loss 0.08147182 - time (sec): 13.91 - samples/sec: 1840.43 - lr: 0.000029 - momentum: 0.000000
145
+ 2024-03-26 12:00:48,448 epoch 5 - iter 81/95 - loss 0.07532693 - time (sec): 16.55 - samples/sec: 1775.15 - lr: 0.000029 - momentum: 0.000000
146
+ 2024-03-26 12:00:50,528 epoch 5 - iter 90/95 - loss 0.07517831 - time (sec): 18.63 - samples/sec: 1757.85 - lr: 0.000028 - momentum: 0.000000
147
+ 2024-03-26 12:00:51,418 ----------------------------------------------------------------------------------------------------
148
+ 2024-03-26 12:00:51,418 EPOCH 5 done: loss 0.0778 - lr: 0.000028
149
+ 2024-03-26 12:00:52,372 DEV : loss 0.1956116110086441 - f1-score (micro avg) 0.9178
150
+ 2024-03-26 12:00:52,374 saving best model
151
+ 2024-03-26 12:00:52,849 ----------------------------------------------------------------------------------------------------
152
+ 2024-03-26 12:00:54,564 epoch 6 - iter 9/95 - loss 0.09104755 - time (sec): 1.71 - samples/sec: 1934.88 - lr: 0.000027 - momentum: 0.000000
153
+ 2024-03-26 12:00:56,671 epoch 6 - iter 18/95 - loss 0.06055749 - time (sec): 3.82 - samples/sec: 1773.75 - lr: 0.000027 - momentum: 0.000000
154
+ 2024-03-26 12:00:58,142 epoch 6 - iter 27/95 - loss 0.06306007 - time (sec): 5.29 - samples/sec: 1805.78 - lr: 0.000026 - momentum: 0.000000
155
+ 2024-03-26 12:01:00,604 epoch 6 - iter 36/95 - loss 0.05149544 - time (sec): 7.75 - samples/sec: 1656.27 - lr: 0.000026 - momentum: 0.000000
156
+ 2024-03-26 12:01:02,418 epoch 6 - iter 45/95 - loss 0.04808451 - time (sec): 9.57 - samples/sec: 1681.48 - lr: 0.000025 - momentum: 0.000000
157
+ 2024-03-26 12:01:04,989 epoch 6 - iter 54/95 - loss 0.05677027 - time (sec): 12.14 - samples/sec: 1658.67 - lr: 0.000025 - momentum: 0.000000
158
+ 2024-03-26 12:01:06,572 epoch 6 - iter 63/95 - loss 0.05658119 - time (sec): 13.72 - samples/sec: 1672.98 - lr: 0.000024 - momentum: 0.000000
159
+ 2024-03-26 12:01:08,132 epoch 6 - iter 72/95 - loss 0.05676127 - time (sec): 15.28 - samples/sec: 1697.53 - lr: 0.000024 - momentum: 0.000000
160
+ 2024-03-26 12:01:10,274 epoch 6 - iter 81/95 - loss 0.05615297 - time (sec): 17.42 - samples/sec: 1692.00 - lr: 0.000023 - momentum: 0.000000
161
+ 2024-03-26 12:01:11,487 epoch 6 - iter 90/95 - loss 0.05920230 - time (sec): 18.64 - samples/sec: 1735.54 - lr: 0.000023 - momentum: 0.000000
162
+ 2024-03-26 12:01:12,892 ----------------------------------------------------------------------------------------------------
163
+ 2024-03-26 12:01:12,892 EPOCH 6 done: loss 0.0576 - lr: 0.000023
164
+ 2024-03-26 12:01:13,859 DEV : loss 0.19573496282100677 - f1-score (micro avg) 0.9042
165
+ 2024-03-26 12:01:13,861 ----------------------------------------------------------------------------------------------------
166
+ 2024-03-26 12:01:15,247 epoch 7 - iter 9/95 - loss 0.03915603 - time (sec): 1.39 - samples/sec: 2290.04 - lr: 0.000022 - momentum: 0.000000
167
+ 2024-03-26 12:01:17,434 epoch 7 - iter 18/95 - loss 0.03185381 - time (sec): 3.57 - samples/sec: 1887.66 - lr: 0.000021 - momentum: 0.000000
168
+ 2024-03-26 12:01:19,422 epoch 7 - iter 27/95 - loss 0.03946797 - time (sec): 5.56 - samples/sec: 1757.96 - lr: 0.000021 - momentum: 0.000000
169
+ 2024-03-26 12:01:20,775 epoch 7 - iter 36/95 - loss 0.03799302 - time (sec): 6.91 - samples/sec: 1809.47 - lr: 0.000020 - momentum: 0.000000
170
+ 2024-03-26 12:01:22,488 epoch 7 - iter 45/95 - loss 0.03831134 - time (sec): 8.63 - samples/sec: 1824.07 - lr: 0.000020 - momentum: 0.000000
171
+ 2024-03-26 12:01:24,868 epoch 7 - iter 54/95 - loss 0.03448802 - time (sec): 11.01 - samples/sec: 1780.19 - lr: 0.000019 - momentum: 0.000000
172
+ 2024-03-26 12:01:26,952 epoch 7 - iter 63/95 - loss 0.03711925 - time (sec): 13.09 - samples/sec: 1739.36 - lr: 0.000019 - momentum: 0.000000
173
+ 2024-03-26 12:01:29,235 epoch 7 - iter 72/95 - loss 0.03678211 - time (sec): 15.37 - samples/sec: 1707.00 - lr: 0.000018 - momentum: 0.000000
174
+ 2024-03-26 12:01:30,770 epoch 7 - iter 81/95 - loss 0.04143827 - time (sec): 16.91 - samples/sec: 1717.57 - lr: 0.000018 - momentum: 0.000000
175
+ 2024-03-26 12:01:32,742 epoch 7 - iter 90/95 - loss 0.04455468 - time (sec): 18.88 - samples/sec: 1742.11 - lr: 0.000017 - momentum: 0.000000
176
+ 2024-03-26 12:01:33,439 ----------------------------------------------------------------------------------------------------
177
+ 2024-03-26 12:01:33,439 EPOCH 7 done: loss 0.0439 - lr: 0.000017
178
+ 2024-03-26 12:01:34,398 DEV : loss 0.20444811880588531 - f1-score (micro avg) 0.9106
179
+ 2024-03-26 12:01:34,400 ----------------------------------------------------------------------------------------------------
180
+ 2024-03-26 12:01:36,096 epoch 8 - iter 9/95 - loss 0.01074191 - time (sec): 1.70 - samples/sec: 1735.19 - lr: 0.000016 - momentum: 0.000000
181
+ 2024-03-26 12:01:38,281 epoch 8 - iter 18/95 - loss 0.01421139 - time (sec): 3.88 - samples/sec: 1708.18 - lr: 0.000016 - momentum: 0.000000
182
+ 2024-03-26 12:01:40,155 epoch 8 - iter 27/95 - loss 0.02248889 - time (sec): 5.75 - samples/sec: 1686.55 - lr: 0.000015 - momentum: 0.000000
183
+ 2024-03-26 12:01:42,191 epoch 8 - iter 36/95 - loss 0.02397763 - time (sec): 7.79 - samples/sec: 1690.53 - lr: 0.000015 - momentum: 0.000000
184
+ 2024-03-26 12:01:43,228 epoch 8 - iter 45/95 - loss 0.03170376 - time (sec): 8.83 - samples/sec: 1776.94 - lr: 0.000014 - momentum: 0.000000
185
+ 2024-03-26 12:01:45,207 epoch 8 - iter 54/95 - loss 0.03599503 - time (sec): 10.81 - samples/sec: 1767.59 - lr: 0.000014 - momentum: 0.000000
186
+ 2024-03-26 12:01:47,502 epoch 8 - iter 63/95 - loss 0.03927120 - time (sec): 13.10 - samples/sec: 1749.17 - lr: 0.000013 - momentum: 0.000000
187
+ 2024-03-26 12:01:49,770 epoch 8 - iter 72/95 - loss 0.03887791 - time (sec): 15.37 - samples/sec: 1739.28 - lr: 0.000013 - momentum: 0.000000
188
+ 2024-03-26 12:01:51,507 epoch 8 - iter 81/95 - loss 0.03775282 - time (sec): 17.11 - samples/sec: 1744.68 - lr: 0.000012 - momentum: 0.000000
189
+ 2024-03-26 12:01:53,554 epoch 8 - iter 90/95 - loss 0.03465067 - time (sec): 19.15 - samples/sec: 1733.30 - lr: 0.000012 - momentum: 0.000000
190
+ 2024-03-26 12:01:54,156 ----------------------------------------------------------------------------------------------------
191
+ 2024-03-26 12:01:54,156 EPOCH 8 done: loss 0.0348 - lr: 0.000012
192
+ 2024-03-26 12:01:55,092 DEV : loss 0.22374196350574493 - f1-score (micro avg) 0.9219
193
+ 2024-03-26 12:01:55,093 saving best model
194
+ 2024-03-26 12:01:55,540 ----------------------------------------------------------------------------------------------------
195
+ 2024-03-26 12:01:57,115 epoch 9 - iter 9/95 - loss 0.02255766 - time (sec): 1.57 - samples/sec: 2020.25 - lr: 0.000011 - momentum: 0.000000
196
+ 2024-03-26 12:01:59,514 epoch 9 - iter 18/95 - loss 0.02441884 - time (sec): 3.97 - samples/sec: 1714.10 - lr: 0.000010 - momentum: 0.000000
197
+ 2024-03-26 12:02:01,166 epoch 9 - iter 27/95 - loss 0.02100220 - time (sec): 5.62 - samples/sec: 1732.76 - lr: 0.000010 - momentum: 0.000000
198
+ 2024-03-26 12:02:03,532 epoch 9 - iter 36/95 - loss 0.02224059 - time (sec): 7.99 - samples/sec: 1697.76 - lr: 0.000009 - momentum: 0.000000
199
+ 2024-03-26 12:02:05,456 epoch 9 - iter 45/95 - loss 0.02327845 - time (sec): 9.91 - samples/sec: 1678.15 - lr: 0.000009 - momentum: 0.000000
200
+ 2024-03-26 12:02:06,885 epoch 9 - iter 54/95 - loss 0.02693223 - time (sec): 11.34 - samples/sec: 1720.22 - lr: 0.000008 - momentum: 0.000000
201
+ 2024-03-26 12:02:09,040 epoch 9 - iter 63/95 - loss 0.02451513 - time (sec): 13.50 - samples/sec: 1698.43 - lr: 0.000008 - momentum: 0.000000
202
+ 2024-03-26 12:02:10,272 epoch 9 - iter 72/95 - loss 0.02812439 - time (sec): 14.73 - samples/sec: 1735.32 - lr: 0.000007 - momentum: 0.000000
203
+ 2024-03-26 12:02:13,095 epoch 9 - iter 81/95 - loss 0.02633959 - time (sec): 17.55 - samples/sec: 1691.71 - lr: 0.000007 - momentum: 0.000000
204
+ 2024-03-26 12:02:14,755 epoch 9 - iter 90/95 - loss 0.02469000 - time (sec): 19.21 - samples/sec: 1714.60 - lr: 0.000006 - momentum: 0.000000
205
+ 2024-03-26 12:02:15,429 ----------------------------------------------------------------------------------------------------
206
+ 2024-03-26 12:02:15,429 EPOCH 9 done: loss 0.0262 - lr: 0.000006
207
+ 2024-03-26 12:02:16,387 DEV : loss 0.22455130517482758 - f1-score (micro avg) 0.9291
208
+ 2024-03-26 12:02:16,388 saving best model
209
+ 2024-03-26 12:02:16,842 ----------------------------------------------------------------------------------------------------
210
+ 2024-03-26 12:02:18,756 epoch 10 - iter 9/95 - loss 0.02413491 - time (sec): 1.91 - samples/sec: 1621.19 - lr: 0.000005 - momentum: 0.000000
211
+ 2024-03-26 12:02:20,996 epoch 10 - iter 18/95 - loss 0.02195416 - time (sec): 4.15 - samples/sec: 1605.14 - lr: 0.000005 - momentum: 0.000000
212
+ 2024-03-26 12:02:22,415 epoch 10 - iter 27/95 - loss 0.02161648 - time (sec): 5.57 - samples/sec: 1760.59 - lr: 0.000004 - momentum: 0.000000
213
+ 2024-03-26 12:02:24,160 epoch 10 - iter 36/95 - loss 0.01921928 - time (sec): 7.32 - samples/sec: 1803.60 - lr: 0.000004 - momentum: 0.000000
214
+ 2024-03-26 12:02:25,592 epoch 10 - iter 45/95 - loss 0.01857663 - time (sec): 8.75 - samples/sec: 1835.65 - lr: 0.000003 - momentum: 0.000000
215
+ 2024-03-26 12:02:26,637 epoch 10 - iter 54/95 - loss 0.01707115 - time (sec): 9.79 - samples/sec: 1905.49 - lr: 0.000003 - momentum: 0.000000
216
+ 2024-03-26 12:02:28,484 epoch 10 - iter 63/95 - loss 0.01548965 - time (sec): 11.64 - samples/sec: 1877.87 - lr: 0.000002 - momentum: 0.000000
217
+ 2024-03-26 12:02:30,794 epoch 10 - iter 72/95 - loss 0.02073655 - time (sec): 13.95 - samples/sec: 1827.57 - lr: 0.000002 - momentum: 0.000000
218
+ 2024-03-26 12:02:32,482 epoch 10 - iter 81/95 - loss 0.02161353 - time (sec): 15.64 - samples/sec: 1816.89 - lr: 0.000001 - momentum: 0.000000
219
+ 2024-03-26 12:02:34,943 epoch 10 - iter 90/95 - loss 0.02064998 - time (sec): 18.10 - samples/sec: 1795.54 - lr: 0.000001 - momentum: 0.000000
220
+ 2024-03-26 12:02:36,200 ----------------------------------------------------------------------------------------------------
221
+ 2024-03-26 12:02:36,200 EPOCH 10 done: loss 0.0209 - lr: 0.000001
222
+ 2024-03-26 12:02:37,134 DEV : loss 0.24257376790046692 - f1-score (micro avg) 0.9269
223
+ 2024-03-26 12:02:37,396 ----------------------------------------------------------------------------------------------------
224
+ 2024-03-26 12:02:37,397 Loading model from best epoch ...
225
+ 2024-03-26 12:02:38,241 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
226
+ 2024-03-26 12:02:39,079
227
+ Results:
228
+ - F-score (micro) 0.9062
229
+ - F-score (macro) 0.6887
230
+ - Accuracy 0.8319
231
+
232
+ By class:
233
+ precision recall f1-score support
234
+
235
+ Unternehmen 0.8902 0.8835 0.8868 266
236
+ Auslagerung 0.8764 0.9116 0.8937 249
237
+ Ort 0.9635 0.9851 0.9742 134
238
+ Software 0.0000 0.0000 0.0000 0
239
+
240
+ micro avg 0.8973 0.9153 0.9062 649
241
+ macro avg 0.6825 0.6950 0.6887 649
242
+ weighted avg 0.9000 0.9153 0.9075 649
243
+
244
+ 2024-03-26 12:02:39,079 ----------------------------------------------------------------------------------------------------