Hugging Face's logo

stefan-it
/
flair-co-funer-german_bert_base-bs16-e10-lr5e-05-5

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
5f1b20f
•
1 Parent(s): 22e7afd

Upload ./training.log with huggingface_hub

Browse files
Files changed (1) hide show
  1. training.log +265 -0
training.log ADDED
@@ -0,0 +1,265 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2024-03-26 12:07:35,602 ----------------------------------------------------------------------------------------------------
2
+ 2024-03-26 12:07:35,603 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 12:07:35,603 ----------------------------------------------------------------------------------------------------
51
+ 2024-03-26 12:07:35,603 Corpus: 758 train + 94 dev + 96 test sentences
52
+ 2024-03-26 12:07:35,603 ----------------------------------------------------------------------------------------------------
53
+ 2024-03-26 12:07:35,603 Train: 758 sentences
54
+ 2024-03-26 12:07:35,603 (train_with_dev=False, train_with_test=False)
55
+ 2024-03-26 12:07:35,603 ----------------------------------------------------------------------------------------------------
56
+ 2024-03-26 12:07:35,603 Training Params:
57
+ 2024-03-26 12:07:35,603 - learning_rate: "5e-05"
58
+ 2024-03-26 12:07:35,603 - mini_batch_size: "16"
59
+ 2024-03-26 12:07:35,603 - max_epochs: "10"
60
+ 2024-03-26 12:07:35,603 - shuffle: "True"
61
+ 2024-03-26 12:07:35,603 ----------------------------------------------------------------------------------------------------
62
+ 2024-03-26 12:07:35,603 Plugins:
63
+ 2024-03-26 12:07:35,603 - TensorboardLogger
64
+ 2024-03-26 12:07:35,603 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2024-03-26 12:07:35,603 ----------------------------------------------------------------------------------------------------
66
+ 2024-03-26 12:07:35,603 Final evaluation on model from best epoch (best-model.pt)
67
+ 2024-03-26 12:07:35,603 - metric: "('micro avg', 'f1-score')"
68
+ 2024-03-26 12:07:35,603 ----------------------------------------------------------------------------------------------------
69
+ 2024-03-26 12:07:35,603 Computation:
70
+ 2024-03-26 12:07:35,603 - compute on device: cuda:0
71
+ 2024-03-26 12:07:35,603 - embedding storage: none
72
+ 2024-03-26 12:07:35,603 ----------------------------------------------------------------------------------------------------
73
+ 2024-03-26 12:07:35,603 Model training base path: "flair-co-funer-german_bert_base-bs16-e10-lr5e-05-5"
74
+ 2024-03-26 12:07:35,603 ----------------------------------------------------------------------------------------------------
75
+ 2024-03-26 12:07:35,603 ----------------------------------------------------------------------------------------------------
76
+ 2024-03-26 12:07:35,603 Logging anything other than scalars to TensorBoard is currently not supported.
77
+ 2024-03-26 12:07:37,116 epoch 1 - iter 4/48 - loss 3.11778714 - time (sec): 1.51 - samples/sec: 1733.05 - lr: 0.000003 - momentum: 0.000000
78
+ 2024-03-26 12:07:40,030 epoch 1 - iter 8/48 - loss 3.06180791 - time (sec): 4.43 - samples/sec: 1374.97 - lr: 0.000007 - momentum: 0.000000
79
+ 2024-03-26 12:07:41,942 epoch 1 - iter 12/48 - loss 3.01711290 - time (sec): 6.34 - samples/sec: 1405.22 - lr: 0.000011 - momentum: 0.000000
80
+ 2024-03-26 12:07:43,616 epoch 1 - iter 16/48 - loss 2.87606922 - time (sec): 8.01 - samples/sec: 1501.64 - lr: 0.000016 - momentum: 0.000000
81
+ 2024-03-26 12:07:45,827 epoch 1 - iter 20/48 - loss 2.72415315 - time (sec): 10.22 - samples/sec: 1474.93 - lr: 0.000020 - momentum: 0.000000
82
+ 2024-03-26 12:07:48,596 epoch 1 - iter 24/48 - loss 2.55773822 - time (sec): 12.99 - samples/sec: 1420.04 - lr: 0.000024 - momentum: 0.000000
83
+ 2024-03-26 12:07:50,292 epoch 1 - iter 28/48 - loss 2.44445691 - time (sec): 14.69 - samples/sec: 1428.78 - lr: 0.000028 - momentum: 0.000000
84
+ 2024-03-26 12:07:52,479 epoch 1 - iter 32/48 - loss 2.31125320 - time (sec): 16.88 - samples/sec: 1424.49 - lr: 0.000032 - momentum: 0.000000
85
+ 2024-03-26 12:07:54,080 epoch 1 - iter 36/48 - loss 2.21174392 - time (sec): 18.48 - samples/sec: 1445.00 - lr: 0.000036 - momentum: 0.000000
86
+ 2024-03-26 12:07:56,933 epoch 1 - iter 40/48 - loss 2.08390632 - time (sec): 21.33 - samples/sec: 1402.01 - lr: 0.000041 - momentum: 0.000000
87
+ 2024-03-26 12:07:58,146 epoch 1 - iter 44/48 - loss 1.99032800 - time (sec): 22.54 - samples/sec: 1425.92 - lr: 0.000045 - momentum: 0.000000
88
+ 2024-03-26 12:08:00,046 epoch 1 - iter 48/48 - loss 1.90868499 - time (sec): 24.44 - samples/sec: 1410.32 - lr: 0.000049 - momentum: 0.000000
89
+ 2024-03-26 12:08:00,047 ----------------------------------------------------------------------------------------------------
90
+ 2024-03-26 12:08:00,047 EPOCH 1 done: loss 1.9087 - lr: 0.000049
91
+ 2024-03-26 12:08:00,907 DEV : loss 0.543185293674469 - f1-score (micro avg) 0.5736
92
+ 2024-03-26 12:08:00,908 saving best model
93
+ 2024-03-26 12:08:01,166 ----------------------------------------------------------------------------------------------------
94
+ 2024-03-26 12:08:03,961 epoch 2 - iter 4/48 - loss 0.63153066 - time (sec): 2.79 - samples/sec: 1235.52 - lr: 0.000050 - momentum: 0.000000
95
+ 2024-03-26 12:08:05,900 epoch 2 - iter 8/48 - loss 0.57681578 - time (sec): 4.73 - samples/sec: 1294.70 - lr: 0.000049 - momentum: 0.000000
96
+ 2024-03-26 12:08:07,882 epoch 2 - iter 12/48 - loss 0.55300388 - time (sec): 6.72 - samples/sec: 1328.57 - lr: 0.000049 - momentum: 0.000000
97
+ 2024-03-26 12:08:10,632 epoch 2 - iter 16/48 - loss 0.50764240 - time (sec): 9.46 - samples/sec: 1337.04 - lr: 0.000048 - momentum: 0.000000
98
+ 2024-03-26 12:08:12,057 epoch 2 - iter 20/48 - loss 0.48930833 - time (sec): 10.89 - samples/sec: 1375.75 - lr: 0.000048 - momentum: 0.000000
99
+ 2024-03-26 12:08:14,974 epoch 2 - iter 24/48 - loss 0.45960951 - time (sec): 13.81 - samples/sec: 1296.23 - lr: 0.000047 - momentum: 0.000000
100
+ 2024-03-26 12:08:16,643 epoch 2 - iter 28/48 - loss 0.45452998 - time (sec): 15.48 - samples/sec: 1328.57 - lr: 0.000047 - momentum: 0.000000
101
+ 2024-03-26 12:08:18,696 epoch 2 - iter 32/48 - loss 0.43339098 - time (sec): 17.53 - samples/sec: 1324.50 - lr: 0.000046 - momentum: 0.000000
102
+ 2024-03-26 12:08:20,523 epoch 2 - iter 36/48 - loss 0.42883053 - time (sec): 19.36 - samples/sec: 1354.11 - lr: 0.000046 - momentum: 0.000000
103
+ 2024-03-26 12:08:22,919 epoch 2 - iter 40/48 - loss 0.42732543 - time (sec): 21.75 - samples/sec: 1345.70 - lr: 0.000046 - momentum: 0.000000
104
+ 2024-03-26 12:08:25,174 epoch 2 - iter 44/48 - loss 0.41218116 - time (sec): 24.01 - samples/sec: 1351.01 - lr: 0.000045 - momentum: 0.000000
105
+ 2024-03-26 12:08:26,460 epoch 2 - iter 48/48 - loss 0.40616007 - time (sec): 25.29 - samples/sec: 1362.89 - lr: 0.000045 - momentum: 0.000000
106
+ 2024-03-26 12:08:26,460 ----------------------------------------------------------------------------------------------------
107
+ 2024-03-26 12:08:26,460 EPOCH 2 done: loss 0.4062 - lr: 0.000045
108
+ 2024-03-26 12:08:27,390 DEV : loss 0.27681225538253784 - f1-score (micro avg) 0.8318
109
+ 2024-03-26 12:08:27,391 saving best model
110
+ 2024-03-26 12:08:27,828 ----------------------------------------------------------------------------------------------------
111
+ 2024-03-26 12:08:28,926 epoch 3 - iter 4/48 - loss 0.32628564 - time (sec): 1.10 - samples/sec: 2035.62 - lr: 0.000044 - momentum: 0.000000
112
+ 2024-03-26 12:08:30,864 epoch 3 - iter 8/48 - loss 0.33108745 - time (sec): 3.03 - samples/sec: 1623.66 - lr: 0.000044 - momentum: 0.000000
113
+ 2024-03-26 12:08:33,122 epoch 3 - iter 12/48 - loss 0.27482260 - time (sec): 5.29 - samples/sec: 1620.20 - lr: 0.000043 - momentum: 0.000000
114
+ 2024-03-26 12:08:35,099 epoch 3 - iter 16/48 - loss 0.26946915 - time (sec): 7.27 - samples/sec: 1564.76 - lr: 0.000043 - momentum: 0.000000
115
+ 2024-03-26 12:08:37,028 epoch 3 - iter 20/48 - loss 0.26156635 - time (sec): 9.20 - samples/sec: 1541.20 - lr: 0.000042 - momentum: 0.000000
116
+ 2024-03-26 12:08:39,027 epoch 3 - iter 24/48 - loss 0.24592552 - time (sec): 11.20 - samples/sec: 1497.12 - lr: 0.000042 - momentum: 0.000000
117
+ 2024-03-26 12:08:42,330 epoch 3 - iter 28/48 - loss 0.23213405 - time (sec): 14.50 - samples/sec: 1380.43 - lr: 0.000041 - momentum: 0.000000
118
+ 2024-03-26 12:08:43,890 epoch 3 - iter 32/48 - loss 0.23144593 - time (sec): 16.06 - samples/sec: 1402.51 - lr: 0.000041 - momentum: 0.000000
119
+ 2024-03-26 12:08:47,297 epoch 3 - iter 36/48 - loss 0.22295835 - time (sec): 19.47 - samples/sec: 1332.44 - lr: 0.000040 - momentum: 0.000000
120
+ 2024-03-26 12:08:49,782 epoch 3 - iter 40/48 - loss 0.22193128 - time (sec): 21.95 - samples/sec: 1332.98 - lr: 0.000040 - momentum: 0.000000
121
+ 2024-03-26 12:08:51,917 epoch 3 - iter 44/48 - loss 0.21408438 - time (sec): 24.09 - samples/sec: 1332.17 - lr: 0.000040 - momentum: 0.000000
122
+ 2024-03-26 12:08:53,552 epoch 3 - iter 48/48 - loss 0.21266740 - time (sec): 25.72 - samples/sec: 1340.16 - lr: 0.000039 - momentum: 0.000000
123
+ 2024-03-26 12:08:53,552 ----------------------------------------------------------------------------------------------------
124
+ 2024-03-26 12:08:53,553 EPOCH 3 done: loss 0.2127 - lr: 0.000039
125
+ 2024-03-26 12:08:54,484 DEV : loss 0.21906423568725586 - f1-score (micro avg) 0.8593
126
+ 2024-03-26 12:08:54,484 saving best model
127
+ 2024-03-26 12:08:54,900 ----------------------------------------------------------------------------------------------------
128
+ 2024-03-26 12:08:57,975 epoch 4 - iter 4/48 - loss 0.11740549 - time (sec): 3.07 - samples/sec: 1213.56 - lr: 0.000039 - momentum: 0.000000
129
+ 2024-03-26 12:08:59,460 epoch 4 - iter 8/48 - loss 0.14421906 - time (sec): 4.56 - samples/sec: 1364.34 - lr: 0.000038 - momentum: 0.000000
130
+ 2024-03-26 12:09:02,077 epoch 4 - iter 12/48 - loss 0.12918348 - time (sec): 7.17 - samples/sec: 1296.27 - lr: 0.000038 - momentum: 0.000000
131
+ 2024-03-26 12:09:04,779 epoch 4 - iter 16/48 - loss 0.11996799 - time (sec): 9.88 - samples/sec: 1285.46 - lr: 0.000037 - momentum: 0.000000
132
+ 2024-03-26 12:09:07,047 epoch 4 - iter 20/48 - loss 0.11541074 - time (sec): 12.14 - samples/sec: 1299.45 - lr: 0.000037 - momentum: 0.000000
133
+ 2024-03-26 12:09:08,575 epoch 4 - iter 24/48 - loss 0.11274160 - time (sec): 13.67 - samples/sec: 1333.36 - lr: 0.000036 - momentum: 0.000000
134
+ 2024-03-26 12:09:11,000 epoch 4 - iter 28/48 - loss 0.11571890 - time (sec): 16.10 - samples/sec: 1319.57 - lr: 0.000036 - momentum: 0.000000
135
+ 2024-03-26 12:09:14,037 epoch 4 - iter 32/48 - loss 0.11758783 - time (sec): 19.13 - samples/sec: 1309.67 - lr: 0.000035 - momentum: 0.000000
136
+ 2024-03-26 12:09:15,742 epoch 4 - iter 36/48 - loss 0.12156915 - time (sec): 20.84 - samples/sec: 1331.89 - lr: 0.000035 - momentum: 0.000000
137
+ 2024-03-26 12:09:16,734 epoch 4 - iter 40/48 - loss 0.12483162 - time (sec): 21.83 - samples/sec: 1375.49 - lr: 0.000034 - momentum: 0.000000
138
+ 2024-03-26 12:09:18,269 epoch 4 - iter 44/48 - loss 0.12429368 - time (sec): 23.37 - samples/sec: 1392.27 - lr: 0.000034 - momentum: 0.000000
139
+ 2024-03-26 12:09:19,156 epoch 4 - iter 48/48 - loss 0.12572472 - time (sec): 24.25 - samples/sec: 1421.29 - lr: 0.000034 - momentum: 0.000000
140
+ 2024-03-26 12:09:19,156 ----------------------------------------------------------------------------------------------------
141
+ 2024-03-26 12:09:19,156 EPOCH 4 done: loss 0.1257 - lr: 0.000034
142
+ 2024-03-26 12:09:20,185 DEV : loss 0.20740923285484314 - f1-score (micro avg) 0.8866
143
+ 2024-03-26 12:09:20,186 saving best model
144
+ 2024-03-26 12:09:20,621 ----------------------------------------------------------------------------------------------------
145
+ 2024-03-26 12:09:22,459 epoch 5 - iter 4/48 - loss 0.12090102 - time (sec): 1.84 - samples/sec: 1564.49 - lr: 0.000033 - momentum: 0.000000
146
+ 2024-03-26 12:09:24,361 epoch 5 - iter 8/48 - loss 0.08690955 - time (sec): 3.74 - samples/sec: 1659.97 - lr: 0.000033 - momentum: 0.000000
147
+ 2024-03-26 12:09:27,482 epoch 5 - iter 12/48 - loss 0.08537459 - time (sec): 6.86 - samples/sec: 1402.63 - lr: 0.000032 - momentum: 0.000000
148
+ 2024-03-26 12:09:28,855 epoch 5 - iter 16/48 - loss 0.08004747 - time (sec): 8.23 - samples/sec: 1444.41 - lr: 0.000032 - momentum: 0.000000
149
+ 2024-03-26 12:09:31,229 epoch 5 - iter 20/48 - loss 0.09057165 - time (sec): 10.61 - samples/sec: 1420.83 - lr: 0.000031 - momentum: 0.000000
150
+ 2024-03-26 12:09:33,382 epoch 5 - iter 24/48 - loss 0.09118019 - time (sec): 12.76 - samples/sec: 1393.20 - lr: 0.000031 - momentum: 0.000000
151
+ 2024-03-26 12:09:34,792 epoch 5 - iter 28/48 - loss 0.09617832 - time (sec): 14.17 - samples/sec: 1432.13 - lr: 0.000030 - momentum: 0.000000
152
+ 2024-03-26 12:09:36,185 epoch 5 - iter 32/48 - loss 0.09826878 - time (sec): 15.56 - samples/sec: 1464.92 - lr: 0.000030 - momentum: 0.000000
153
+ 2024-03-26 12:09:38,457 epoch 5 - iter 36/48 - loss 0.09846321 - time (sec): 17.83 - samples/sec: 1448.02 - lr: 0.000029 - momentum: 0.000000
154
+ 2024-03-26 12:09:40,324 epoch 5 - iter 40/48 - loss 0.09674286 - time (sec): 19.70 - samples/sec: 1447.10 - lr: 0.000029 - momentum: 0.000000
155
+ 2024-03-26 12:09:42,409 epoch 5 - iter 44/48 - loss 0.09638321 - time (sec): 21.79 - samples/sec: 1457.16 - lr: 0.000029 - momentum: 0.000000
156
+ 2024-03-26 12:09:44,567 epoch 5 - iter 48/48 - loss 0.09397791 - time (sec): 23.94 - samples/sec: 1439.72 - lr: 0.000028 - momentum: 0.000000
157
+ 2024-03-26 12:09:44,567 ----------------------------------------------------------------------------------------------------
158
+ 2024-03-26 12:09:44,567 EPOCH 5 done: loss 0.0940 - lr: 0.000028
159
+ 2024-03-26 12:09:45,505 DEV : loss 0.20168490707874298 - f1-score (micro avg) 0.9022
160
+ 2024-03-26 12:09:45,506 saving best model
161
+ 2024-03-26 12:09:45,935 ----------------------------------------------------------------------------------------------------
162
+ 2024-03-26 12:09:47,883 epoch 6 - iter 4/48 - loss 0.07053445 - time (sec): 1.95 - samples/sec: 1410.89 - lr: 0.000028 - momentum: 0.000000
163
+ 2024-03-26 12:09:50,791 epoch 6 - iter 8/48 - loss 0.08016896 - time (sec): 4.85 - samples/sec: 1309.39 - lr: 0.000027 - momentum: 0.000000
164
+ 2024-03-26 12:09:52,691 epoch 6 - iter 12/48 - loss 0.08291448 - time (sec): 6.75 - samples/sec: 1336.93 - lr: 0.000027 - momentum: 0.000000
165
+ 2024-03-26 12:09:54,289 epoch 6 - iter 16/48 - loss 0.08691650 - time (sec): 8.35 - samples/sec: 1385.28 - lr: 0.000026 - momentum: 0.000000
166
+ 2024-03-26 12:09:57,045 epoch 6 - iter 20/48 - loss 0.08104279 - time (sec): 11.11 - samples/sec: 1312.62 - lr: 0.000026 - momentum: 0.000000
167
+ 2024-03-26 12:09:59,801 epoch 6 - iter 24/48 - loss 0.07372277 - time (sec): 13.86 - samples/sec: 1288.63 - lr: 0.000025 - momentum: 0.000000
168
+ 2024-03-26 12:10:02,342 epoch 6 - iter 28/48 - loss 0.07189241 - time (sec): 16.40 - samples/sec: 1264.89 - lr: 0.000025 - momentum: 0.000000
169
+ 2024-03-26 12:10:03,750 epoch 6 - iter 32/48 - loss 0.07946674 - time (sec): 17.81 - samples/sec: 1308.82 - lr: 0.000024 - momentum: 0.000000
170
+ 2024-03-26 12:10:05,666 epoch 6 - iter 36/48 - loss 0.07790067 - time (sec): 19.73 - samples/sec: 1320.00 - lr: 0.000024 - momentum: 0.000000
171
+ 2024-03-26 12:10:06,692 epoch 6 - iter 40/48 - loss 0.07657290 - time (sec): 20.75 - samples/sec: 1359.22 - lr: 0.000023 - momentum: 0.000000
172
+ 2024-03-26 12:10:09,222 epoch 6 - iter 44/48 - loss 0.07393162 - time (sec): 23.28 - samples/sec: 1336.25 - lr: 0.000023 - momentum: 0.000000
173
+ 2024-03-26 12:10:12,106 epoch 6 - iter 48/48 - loss 0.06893913 - time (sec): 26.17 - samples/sec: 1317.26 - lr: 0.000023 - momentum: 0.000000
174
+ 2024-03-26 12:10:12,107 ----------------------------------------------------------------------------------------------------
175
+ 2024-03-26 12:10:12,107 EPOCH 6 done: loss 0.0689 - lr: 0.000023
176
+ 2024-03-26 12:10:13,053 DEV : loss 0.19028829038143158 - f1-score (micro avg) 0.9094
177
+ 2024-03-26 12:10:13,054 saving best model
178
+ 2024-03-26 12:10:13,523 ----------------------------------------------------------------------------------------------------
179
+ 2024-03-26 12:10:15,702 epoch 7 - iter 4/48 - loss 0.03780100 - time (sec): 2.18 - samples/sec: 1335.71 - lr: 0.000022 - momentum: 0.000000
180
+ 2024-03-26 12:10:17,432 epoch 7 - iter 8/48 - loss 0.03164946 - time (sec): 3.91 - samples/sec: 1363.75 - lr: 0.000022 - momentum: 0.000000
181
+ 2024-03-26 12:10:18,885 epoch 7 - iter 12/48 - loss 0.05940567 - time (sec): 5.36 - samples/sec: 1415.95 - lr: 0.000021 - momentum: 0.000000
182
+ 2024-03-26 12:10:20,851 epoch 7 - iter 16/48 - loss 0.05605664 - time (sec): 7.33 - samples/sec: 1448.89 - lr: 0.000021 - momentum: 0.000000
183
+ 2024-03-26 12:10:23,171 epoch 7 - iter 20/48 - loss 0.06559647 - time (sec): 9.65 - samples/sec: 1502.25 - lr: 0.000020 - momentum: 0.000000
184
+ 2024-03-26 12:10:24,522 epoch 7 - iter 24/48 - loss 0.06256993 - time (sec): 11.00 - samples/sec: 1549.19 - lr: 0.000020 - momentum: 0.000000
185
+ 2024-03-26 12:10:26,774 epoch 7 - iter 28/48 - loss 0.06026949 - time (sec): 13.25 - samples/sec: 1505.41 - lr: 0.000019 - momentum: 0.000000
186
+ 2024-03-26 12:10:28,696 epoch 7 - iter 32/48 - loss 0.06041284 - time (sec): 15.17 - samples/sec: 1500.68 - lr: 0.000019 - momentum: 0.000000
187
+ 2024-03-26 12:10:30,717 epoch 7 - iter 36/48 - loss 0.05927659 - time (sec): 17.19 - samples/sec: 1470.21 - lr: 0.000018 - momentum: 0.000000
188
+ 2024-03-26 12:10:33,586 epoch 7 - iter 40/48 - loss 0.05610429 - time (sec): 20.06 - samples/sec: 1451.74 - lr: 0.000018 - momentum: 0.000000
189
+ 2024-03-26 12:10:35,124 epoch 7 - iter 44/48 - loss 0.05626883 - time (sec): 21.60 - samples/sec: 1466.20 - lr: 0.000017 - momentum: 0.000000
190
+ 2024-03-26 12:10:37,316 epoch 7 - iter 48/48 - loss 0.05544531 - time (sec): 23.79 - samples/sec: 1448.93 - lr: 0.000017 - momentum: 0.000000
191
+ 2024-03-26 12:10:37,316 ----------------------------------------------------------------------------------------------------
192
+ 2024-03-26 12:10:37,316 EPOCH 7 done: loss 0.0554 - lr: 0.000017
193
+ 2024-03-26 12:10:38,258 DEV : loss 0.1981896311044693 - f1-score (micro avg) 0.9053
194
+ 2024-03-26 12:10:38,259 ----------------------------------------------------------------------------------------------------
195
+ 2024-03-26 12:10:40,567 epoch 8 - iter 4/48 - loss 0.04534416 - time (sec): 2.31 - samples/sec: 1208.18 - lr: 0.000017 - momentum: 0.000000
196
+ 2024-03-26 12:10:42,143 epoch 8 - iter 8/48 - loss 0.02965056 - time (sec): 3.88 - samples/sec: 1399.44 - lr: 0.000016 - momentum: 0.000000
197
+ 2024-03-26 12:10:45,204 epoch 8 - iter 12/48 - loss 0.03215632 - time (sec): 6.95 - samples/sec: 1295.70 - lr: 0.000016 - momentum: 0.000000
198
+ 2024-03-26 12:10:47,642 epoch 8 - iter 16/48 - loss 0.03129876 - time (sec): 9.38 - samples/sec: 1309.25 - lr: 0.000015 - momentum: 0.000000
199
+ 2024-03-26 12:10:49,132 epoch 8 - iter 20/48 - loss 0.03091358 - time (sec): 10.87 - samples/sec: 1366.26 - lr: 0.000015 - momentum: 0.000000
200
+ 2024-03-26 12:10:50,563 epoch 8 - iter 24/48 - loss 0.03131696 - time (sec): 12.30 - samples/sec: 1438.11 - lr: 0.000014 - momentum: 0.000000
201
+ 2024-03-26 12:10:51,878 epoch 8 - iter 28/48 - loss 0.03425023 - time (sec): 13.62 - samples/sec: 1501.33 - lr: 0.000014 - momentum: 0.000000
202
+ 2024-03-26 12:10:54,281 epoch 8 - iter 32/48 - loss 0.03651616 - time (sec): 16.02 - samples/sec: 1446.00 - lr: 0.000013 - momentum: 0.000000
203
+ 2024-03-26 12:10:56,877 epoch 8 - iter 36/48 - loss 0.03602537 - time (sec): 18.62 - samples/sec: 1406.31 - lr: 0.000013 - momentum: 0.000000
204
+ 2024-03-26 12:10:58,925 epoch 8 - iter 40/48 - loss 0.03873332 - time (sec): 20.67 - samples/sec: 1412.22 - lr: 0.000012 - momentum: 0.000000
205
+ 2024-03-26 12:11:01,074 epoch 8 - iter 44/48 - loss 0.04052895 - time (sec): 22.81 - samples/sec: 1400.07 - lr: 0.000012 - momentum: 0.000000
206
+ 2024-03-26 12:11:02,700 epoch 8 - iter 48/48 - loss 0.04041399 - time (sec): 24.44 - samples/sec: 1410.43 - lr: 0.000011 - momentum: 0.000000
207
+ 2024-03-26 12:11:02,700 ----------------------------------------------------------------------------------------------------
208
+ 2024-03-26 12:11:02,700 EPOCH 8 done: loss 0.0404 - lr: 0.000011
209
+ 2024-03-26 12:11:03,635 DEV : loss 0.18661607801914215 - f1-score (micro avg) 0.9188
210
+ 2024-03-26 12:11:03,636 saving best model
211
+ 2024-03-26 12:11:04,065 ----------------------------------------------------------------------------------------------------
212
+ 2024-03-26 12:11:06,839 epoch 9 - iter 4/48 - loss 0.02788980 - time (sec): 2.77 - samples/sec: 1262.60 - lr: 0.000011 - momentum: 0.000000
213
+ 2024-03-26 12:11:08,912 epoch 9 - iter 8/48 - loss 0.02223975 - time (sec): 4.85 - samples/sec: 1318.02 - lr: 0.000011 - momentum: 0.000000
214
+ 2024-03-26 12:11:11,916 epoch 9 - iter 12/48 - loss 0.02509804 - time (sec): 7.85 - samples/sec: 1239.30 - lr: 0.000010 - momentum: 0.000000
215
+ 2024-03-26 12:11:15,050 epoch 9 - iter 16/48 - loss 0.03662374 - time (sec): 10.98 - samples/sec: 1224.07 - lr: 0.000010 - momentum: 0.000000
216
+ 2024-03-26 12:11:15,918 epoch 9 - iter 20/48 - loss 0.03356456 - time (sec): 11.85 - samples/sec: 1314.29 - lr: 0.000009 - momentum: 0.000000
217
+ 2024-03-26 12:11:17,836 epoch 9 - iter 24/48 - loss 0.03206316 - time (sec): 13.77 - samples/sec: 1309.12 - lr: 0.000009 - momentum: 0.000000
218
+ 2024-03-26 12:11:19,863 epoch 9 - iter 28/48 - loss 0.03079525 - time (sec): 15.80 - samples/sec: 1325.63 - lr: 0.000008 - momentum: 0.000000
219
+ 2024-03-26 12:11:20,896 epoch 9 - iter 32/48 - loss 0.03134437 - time (sec): 16.83 - samples/sec: 1388.15 - lr: 0.000008 - momentum: 0.000000
220
+ 2024-03-26 12:11:22,071 epoch 9 - iter 36/48 - loss 0.02974533 - time (sec): 18.00 - samples/sec: 1438.98 - lr: 0.000007 - momentum: 0.000000
221
+ 2024-03-26 12:11:23,388 epoch 9 - iter 40/48 - loss 0.02947689 - time (sec): 19.32 - samples/sec: 1467.99 - lr: 0.000007 - momentum: 0.000000
222
+ 2024-03-26 12:11:26,444 epoch 9 - iter 44/48 - loss 0.03086982 - time (sec): 22.38 - samples/sec: 1441.46 - lr: 0.000006 - momentum: 0.000000
223
+ 2024-03-26 12:11:27,995 epoch 9 - iter 48/48 - loss 0.03045941 - time (sec): 23.93 - samples/sec: 1440.61 - lr: 0.000006 - momentum: 0.000000
224
+ 2024-03-26 12:11:27,995 ----------------------------------------------------------------------------------------------------
225
+ 2024-03-26 12:11:27,996 EPOCH 9 done: loss 0.0305 - lr: 0.000006
226
+ 2024-03-26 12:11:28,933 DEV : loss 0.20570887625217438 - f1-score (micro avg) 0.9248
227
+ 2024-03-26 12:11:28,936 saving best model
228
+ 2024-03-26 12:11:29,364 ----------------------------------------------------------------------------------------------------
229
+ 2024-03-26 12:11:32,286 epoch 10 - iter 4/48 - loss 0.02076248 - time (sec): 2.92 - samples/sec: 1271.59 - lr: 0.000006 - momentum: 0.000000
230
+ 2024-03-26 12:11:34,276 epoch 10 - iter 8/48 - loss 0.01986618 - time (sec): 4.91 - samples/sec: 1316.11 - lr: 0.000005 - momentum: 0.000000
231
+ 2024-03-26 12:11:36,485 epoch 10 - iter 12/48 - loss 0.02131288 - time (sec): 7.12 - samples/sec: 1277.40 - lr: 0.000005 - momentum: 0.000000
232
+ 2024-03-26 12:11:39,078 epoch 10 - iter 16/48 - loss 0.02075093 - time (sec): 9.71 - samples/sec: 1229.58 - lr: 0.000004 - momentum: 0.000000
233
+ 2024-03-26 12:11:41,706 epoch 10 - iter 20/48 - loss 0.02169695 - time (sec): 12.34 - samples/sec: 1233.69 - lr: 0.000004 - momentum: 0.000000
234
+ 2024-03-26 12:11:43,127 epoch 10 - iter 24/48 - loss 0.02159281 - time (sec): 13.76 - samples/sec: 1296.16 - lr: 0.000003 - momentum: 0.000000
235
+ 2024-03-26 12:11:44,051 epoch 10 - iter 28/48 - loss 0.02282303 - time (sec): 14.69 - samples/sec: 1364.03 - lr: 0.000003 - momentum: 0.000000
236
+ 2024-03-26 12:11:46,036 epoch 10 - iter 32/48 - loss 0.02684181 - time (sec): 16.67 - samples/sec: 1383.65 - lr: 0.000002 - momentum: 0.000000
237
+ 2024-03-26 12:11:48,346 epoch 10 - iter 36/48 - loss 0.02575977 - time (sec): 18.98 - samples/sec: 1363.04 - lr: 0.000002 - momentum: 0.000000
238
+ 2024-03-26 12:11:50,063 epoch 10 - iter 40/48 - loss 0.02573594 - time (sec): 20.70 - samples/sec: 1387.55 - lr: 0.000001 - momentum: 0.000000
239
+ 2024-03-26 12:11:53,301 epoch 10 - iter 44/48 - loss 0.02535055 - time (sec): 23.94 - samples/sec: 1369.28 - lr: 0.000001 - momentum: 0.000000
240
+ 2024-03-26 12:11:54,015 epoch 10 - iter 48/48 - loss 0.02542471 - time (sec): 24.65 - samples/sec: 1398.49 - lr: 0.000000 - momentum: 0.000000
241
+ 2024-03-26 12:11:54,015 ----------------------------------------------------------------------------------------------------
242
+ 2024-03-26 12:11:54,016 EPOCH 10 done: loss 0.0254 - lr: 0.000000
243
+ 2024-03-26 12:11:54,951 DEV : loss 0.2053409367799759 - f1-score (micro avg) 0.9242
244
+ 2024-03-26 12:11:55,211 ----------------------------------------------------------------------------------------------------
245
+ 2024-03-26 12:11:55,211 Loading model from best epoch ...
246
+ 2024-03-26 12:11:56,053 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
247
+ 2024-03-26 12:11:56,904
248
+ Results:
249
+ - F-score (micro) 0.9106
250
+ - F-score (macro) 0.6927
251
+ - Accuracy 0.8383
252
+
253
+ By class:
254
+ precision recall f1-score support
255
+
256
+ Unternehmen 0.9046 0.8910 0.8977 266
257
+ Auslagerung 0.8692 0.9076 0.8880 249
258
+ Ort 0.9779 0.9925 0.9852 134
259
+ Software 0.0000 0.0000 0.0000 0
260
+
261
+ micro avg 0.9030 0.9183 0.9106 649
262
+ macro avg 0.6879 0.6978 0.6927 649
263
+ weighted avg 0.9062 0.9183 0.9121 649
264
+
265
+ 2024-03-26 12:11:56,905 ----------------------------------------------------------------------------------------------------