Hugging Face's logo

stefan-it
/
flair-co-funer-german_bert_base-bs16-e10-lr3e-05-1

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
4ce6349
•
1 Parent(s): 7ce66b4

Upload ./training.log with huggingface_hub

Browse files
Files changed (1) hide show
  1. training.log +267 -0
training.log ADDED
@@ -0,0 +1,267 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2024-03-26 10:55:09,597 ----------------------------------------------------------------------------------------------------
2
+ 2024-03-26 10:55:09,597 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 10:55:09,597 ----------------------------------------------------------------------------------------------------
51
+ 2024-03-26 10:55:09,597 Corpus: 758 train + 94 dev + 96 test sentences
52
+ 2024-03-26 10:55:09,597 ----------------------------------------------------------------------------------------------------
53
+ 2024-03-26 10:55:09,597 Train: 758 sentences
54
+ 2024-03-26 10:55:09,597 (train_with_dev=False, train_with_test=False)
55
+ 2024-03-26 10:55:09,597 ----------------------------------------------------------------------------------------------------
56
+ 2024-03-26 10:55:09,598 Training Params:
57
+ 2024-03-26 10:55:09,598 - learning_rate: "3e-05"
58
+ 2024-03-26 10:55:09,598 - mini_batch_size: "16"
59
+ 2024-03-26 10:55:09,598 - max_epochs: "10"
60
+ 2024-03-26 10:55:09,598 - shuffle: "True"
61
+ 2024-03-26 10:55:09,598 ----------------------------------------------------------------------------------------------------
62
+ 2024-03-26 10:55:09,598 Plugins:
63
+ 2024-03-26 10:55:09,598 - TensorboardLogger
64
+ 2024-03-26 10:55:09,598 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2024-03-26 10:55:09,598 ----------------------------------------------------------------------------------------------------
66
+ 2024-03-26 10:55:09,598 Final evaluation on model from best epoch (best-model.pt)
67
+ 2024-03-26 10:55:09,598 - metric: "('micro avg', 'f1-score')"
68
+ 2024-03-26 10:55:09,598 ----------------------------------------------------------------------------------------------------
69
+ 2024-03-26 10:55:09,598 Computation:
70
+ 2024-03-26 10:55:09,598 - compute on device: cuda:0
71
+ 2024-03-26 10:55:09,598 - embedding storage: none
72
+ 2024-03-26 10:55:09,598 ----------------------------------------------------------------------------------------------------
73
+ 2024-03-26 10:55:09,598 Model training base path: "flair-co-funer-german_bert_base-bs16-e10-lr3e-05-1"
74
+ 2024-03-26 10:55:09,598 ----------------------------------------------------------------------------------------------------
75
+ 2024-03-26 10:55:09,598 ----------------------------------------------------------------------------------------------------
76
+ 2024-03-26 10:55:09,598 Logging anything other than scalars to TensorBoard is currently not supported.
77
+ 2024-03-26 10:55:11,752 epoch 1 - iter 4/48 - loss 3.18136204 - time (sec): 2.15 - samples/sec: 1260.53 - lr: 0.000002 - momentum: 0.000000
78
+ 2024-03-26 10:55:13,064 epoch 1 - iter 8/48 - loss 3.26104187 - time (sec): 3.47 - samples/sec: 1555.09 - lr: 0.000004 - momentum: 0.000000
79
+ 2024-03-26 10:55:16,152 epoch 1 - iter 12/48 - loss 3.20211529 - time (sec): 6.55 - samples/sec: 1327.87 - lr: 0.000007 - momentum: 0.000000
80
+ 2024-03-26 10:55:19,396 epoch 1 - iter 16/48 - loss 3.11692024 - time (sec): 9.80 - samples/sec: 1244.50 - lr: 0.000009 - momentum: 0.000000
81
+ 2024-03-26 10:55:21,894 epoch 1 - iter 20/48 - loss 2.98646881 - time (sec): 12.30 - samples/sec: 1251.03 - lr: 0.000012 - momentum: 0.000000
82
+ 2024-03-26 10:55:23,619 epoch 1 - iter 24/48 - loss 2.85768210 - time (sec): 14.02 - samples/sec: 1300.72 - lr: 0.000014 - momentum: 0.000000
83
+ 2024-03-26 10:55:25,219 epoch 1 - iter 28/48 - loss 2.74291622 - time (sec): 15.62 - samples/sec: 1325.19 - lr: 0.000017 - momentum: 0.000000
84
+ 2024-03-26 10:55:27,326 epoch 1 - iter 32/48 - loss 2.63361563 - time (sec): 17.73 - samples/sec: 1333.24 - lr: 0.000019 - momentum: 0.000000
85
+ 2024-03-26 10:55:28,289 epoch 1 - iter 36/48 - loss 2.55063314 - time (sec): 18.69 - samples/sec: 1393.65 - lr: 0.000022 - momentum: 0.000000
86
+ 2024-03-26 10:55:30,274 epoch 1 - iter 40/48 - loss 2.46376363 - time (sec): 20.68 - samples/sec: 1408.22 - lr: 0.000024 - momentum: 0.000000
87
+ 2024-03-26 10:55:32,292 epoch 1 - iter 44/48 - loss 2.37711797 - time (sec): 22.69 - samples/sec: 1396.02 - lr: 0.000027 - momentum: 0.000000
88
+ 2024-03-26 10:55:33,804 epoch 1 - iter 48/48 - loss 2.27943829 - time (sec): 24.21 - samples/sec: 1424.14 - lr: 0.000029 - momentum: 0.000000
89
+ 2024-03-26 10:55:33,804 ----------------------------------------------------------------------------------------------------
90
+ 2024-03-26 10:55:33,804 EPOCH 1 done: loss 2.2794 - lr: 0.000029
91
+ 2024-03-26 10:55:34,638 DEV : loss 0.8841983079910278 - f1-score (micro avg) 0.3292
92
+ 2024-03-26 10:55:34,639 saving best model
93
+ 2024-03-26 10:55:34,944 ----------------------------------------------------------------------------------------------------
94
+ 2024-03-26 10:55:37,559 epoch 2 - iter 4/48 - loss 1.06963551 - time (sec): 2.61 - samples/sec: 1186.48 - lr: 0.000030 - momentum: 0.000000
95
+ 2024-03-26 10:55:39,674 epoch 2 - iter 8/48 - loss 1.02472331 - time (sec): 4.73 - samples/sec: 1397.79 - lr: 0.000030 - momentum: 0.000000
96
+ 2024-03-26 10:55:41,963 epoch 2 - iter 12/48 - loss 0.95761277 - time (sec): 7.02 - samples/sec: 1319.21 - lr: 0.000029 - momentum: 0.000000
97
+ 2024-03-26 10:55:44,022 epoch 2 - iter 16/48 - loss 0.90790118 - time (sec): 9.08 - samples/sec: 1312.57 - lr: 0.000029 - momentum: 0.000000
98
+ 2024-03-26 10:55:46,126 epoch 2 - iter 20/48 - loss 0.85036938 - time (sec): 11.18 - samples/sec: 1341.05 - lr: 0.000029 - momentum: 0.000000
99
+ 2024-03-26 10:55:49,236 epoch 2 - iter 24/48 - loss 0.77726093 - time (sec): 14.29 - samples/sec: 1294.67 - lr: 0.000028 - momentum: 0.000000
100
+ 2024-03-26 10:55:51,622 epoch 2 - iter 28/48 - loss 0.75313006 - time (sec): 16.68 - samples/sec: 1291.59 - lr: 0.000028 - momentum: 0.000000
101
+ 2024-03-26 10:55:53,372 epoch 2 - iter 32/48 - loss 0.72383530 - time (sec): 18.43 - samples/sec: 1309.23 - lr: 0.000028 - momentum: 0.000000
102
+ 2024-03-26 10:55:54,428 epoch 2 - iter 36/48 - loss 0.70149258 - time (sec): 19.48 - samples/sec: 1357.69 - lr: 0.000028 - momentum: 0.000000
103
+ 2024-03-26 10:55:56,284 epoch 2 - iter 40/48 - loss 0.67432397 - time (sec): 21.34 - samples/sec: 1377.62 - lr: 0.000027 - momentum: 0.000000
104
+ 2024-03-26 10:55:58,311 epoch 2 - iter 44/48 - loss 0.65523371 - time (sec): 23.37 - samples/sec: 1374.06 - lr: 0.000027 - momentum: 0.000000
105
+ 2024-03-26 10:55:59,735 epoch 2 - iter 48/48 - loss 0.63761461 - time (sec): 24.79 - samples/sec: 1390.52 - lr: 0.000027 - momentum: 0.000000
106
+ 2024-03-26 10:55:59,735 ----------------------------------------------------------------------------------------------------
107
+ 2024-03-26 10:55:59,735 EPOCH 2 done: loss 0.6376 - lr: 0.000027
108
+ 2024-03-26 10:56:00,749 DEV : loss 0.3372988700866699 - f1-score (micro avg) 0.7573
109
+ 2024-03-26 10:56:00,750 saving best model
110
+ 2024-03-26 10:56:01,223 ----------------------------------------------------------------------------------------------------
111
+ 2024-03-26 10:56:03,751 epoch 3 - iter 4/48 - loss 0.38597010 - time (sec): 2.53 - samples/sec: 1208.26 - lr: 0.000026 - momentum: 0.000000
112
+ 2024-03-26 10:56:05,605 epoch 3 - iter 8/48 - loss 0.34811982 - time (sec): 4.38 - samples/sec: 1339.46 - lr: 0.000026 - momentum: 0.000000
113
+ 2024-03-26 10:56:07,422 epoch 3 - iter 12/48 - loss 0.35445530 - time (sec): 6.20 - samples/sec: 1417.83 - lr: 0.000026 - momentum: 0.000000
114
+ 2024-03-26 10:56:09,791 epoch 3 - iter 16/48 - loss 0.33399667 - time (sec): 8.57 - samples/sec: 1425.65 - lr: 0.000026 - momentum: 0.000000
115
+ 2024-03-26 10:56:11,255 epoch 3 - iter 20/48 - loss 0.34563570 - time (sec): 10.03 - samples/sec: 1475.18 - lr: 0.000025 - momentum: 0.000000
116
+ 2024-03-26 10:56:14,190 epoch 3 - iter 24/48 - loss 0.32843371 - time (sec): 12.96 - samples/sec: 1458.63 - lr: 0.000025 - momentum: 0.000000
117
+ 2024-03-26 10:56:14,942 epoch 3 - iter 28/48 - loss 0.31446138 - time (sec): 13.72 - samples/sec: 1533.97 - lr: 0.000025 - momentum: 0.000000
118
+ 2024-03-26 10:56:17,608 epoch 3 - iter 32/48 - loss 0.29864600 - time (sec): 16.38 - samples/sec: 1466.40 - lr: 0.000025 - momentum: 0.000000
119
+ 2024-03-26 10:56:19,631 epoch 3 - iter 36/48 - loss 0.28700422 - time (sec): 18.41 - samples/sec: 1459.20 - lr: 0.000024 - momentum: 0.000000
120
+ 2024-03-26 10:56:21,564 epoch 3 - iter 40/48 - loss 0.28584502 - time (sec): 20.34 - samples/sec: 1446.69 - lr: 0.000024 - momentum: 0.000000
121
+ 2024-03-26 10:56:23,776 epoch 3 - iter 44/48 - loss 0.27720640 - time (sec): 22.55 - samples/sec: 1446.34 - lr: 0.000024 - momentum: 0.000000
122
+ 2024-03-26 10:56:25,049 epoch 3 - iter 48/48 - loss 0.27449823 - time (sec): 23.82 - samples/sec: 1446.95 - lr: 0.000023 - momentum: 0.000000
123
+ 2024-03-26 10:56:25,049 ----------------------------------------------------------------------------------------------------
124
+ 2024-03-26 10:56:25,049 EPOCH 3 done: loss 0.2745 - lr: 0.000023
125
+ 2024-03-26 10:56:25,977 DEV : loss 0.26073935627937317 - f1-score (micro avg) 0.8484
126
+ 2024-03-26 10:56:25,979 saving best model
127
+ 2024-03-26 10:56:26,464 ----------------------------------------------------------------------------------------------------
128
+ 2024-03-26 10:56:27,937 epoch 4 - iter 4/48 - loss 0.19093305 - time (sec): 1.47 - samples/sec: 1852.02 - lr: 0.000023 - momentum: 0.000000
129
+ 2024-03-26 10:56:30,418 epoch 4 - iter 8/48 - loss 0.18898170 - time (sec): 3.95 - samples/sec: 1451.37 - lr: 0.000023 - momentum: 0.000000
130
+ 2024-03-26 10:56:32,511 epoch 4 - iter 12/48 - loss 0.20110181 - time (sec): 6.05 - samples/sec: 1445.00 - lr: 0.000023 - momentum: 0.000000
131
+ 2024-03-26 10:56:34,730 epoch 4 - iter 16/48 - loss 0.18320119 - time (sec): 8.27 - samples/sec: 1448.64 - lr: 0.000022 - momentum: 0.000000
132
+ 2024-03-26 10:56:37,704 epoch 4 - iter 20/48 - loss 0.17234055 - time (sec): 11.24 - samples/sec: 1378.13 - lr: 0.000022 - momentum: 0.000000
133
+ 2024-03-26 10:56:39,167 epoch 4 - iter 24/48 - loss 0.17634685 - time (sec): 12.70 - samples/sec: 1418.82 - lr: 0.000022 - momentum: 0.000000
134
+ 2024-03-26 10:56:40,684 epoch 4 - iter 28/48 - loss 0.17459195 - time (sec): 14.22 - samples/sec: 1458.49 - lr: 0.000022 - momentum: 0.000000
135
+ 2024-03-26 10:56:43,284 epoch 4 - iter 32/48 - loss 0.17854116 - time (sec): 16.82 - samples/sec: 1441.08 - lr: 0.000021 - momentum: 0.000000
136
+ 2024-03-26 10:56:44,303 epoch 4 - iter 36/48 - loss 0.18073101 - time (sec): 17.84 - samples/sec: 1489.43 - lr: 0.000021 - momentum: 0.000000
137
+ 2024-03-26 10:56:46,702 epoch 4 - iter 40/48 - loss 0.17591819 - time (sec): 20.24 - samples/sec: 1443.13 - lr: 0.000021 - momentum: 0.000000
138
+ 2024-03-26 10:56:48,542 epoch 4 - iter 44/48 - loss 0.17572825 - time (sec): 22.08 - samples/sec: 1461.47 - lr: 0.000020 - momentum: 0.000000
139
+ 2024-03-26 10:56:49,925 epoch 4 - iter 48/48 - loss 0.17431063 - time (sec): 23.46 - samples/sec: 1469.37 - lr: 0.000020 - momentum: 0.000000
140
+ 2024-03-26 10:56:49,925 ----------------------------------------------------------------------------------------------------
141
+ 2024-03-26 10:56:49,925 EPOCH 4 done: loss 0.1743 - lr: 0.000020
142
+ 2024-03-26 10:56:50,845 DEV : loss 0.22804546356201172 - f1-score (micro avg) 0.8739
143
+ 2024-03-26 10:56:50,846 saving best model
144
+ 2024-03-26 10:56:51,332 ----------------------------------------------------------------------------------------------------
145
+ 2024-03-26 10:56:53,239 epoch 5 - iter 4/48 - loss 0.14175185 - time (sec): 1.91 - samples/sec: 1463.66 - lr: 0.000020 - momentum: 0.000000
146
+ 2024-03-26 10:56:55,735 epoch 5 - iter 8/48 - loss 0.13481281 - time (sec): 4.40 - samples/sec: 1348.74 - lr: 0.000020 - momentum: 0.000000
147
+ 2024-03-26 10:56:57,772 epoch 5 - iter 12/48 - loss 0.13837671 - time (sec): 6.44 - samples/sec: 1330.44 - lr: 0.000019 - momentum: 0.000000
148
+ 2024-03-26 10:56:59,801 epoch 5 - iter 16/48 - loss 0.13341679 - time (sec): 8.47 - samples/sec: 1364.24 - lr: 0.000019 - momentum: 0.000000
149
+ 2024-03-26 10:57:01,739 epoch 5 - iter 20/48 - loss 0.13770967 - time (sec): 10.41 - samples/sec: 1375.23 - lr: 0.000019 - momentum: 0.000000
150
+ 2024-03-26 10:57:03,284 epoch 5 - iter 24/48 - loss 0.14293971 - time (sec): 11.95 - samples/sec: 1423.20 - lr: 0.000018 - momentum: 0.000000
151
+ 2024-03-26 10:57:05,585 epoch 5 - iter 28/48 - loss 0.14527389 - time (sec): 14.25 - samples/sec: 1413.39 - lr: 0.000018 - momentum: 0.000000
152
+ 2024-03-26 10:57:08,217 epoch 5 - iter 32/48 - loss 0.14233404 - time (sec): 16.88 - samples/sec: 1402.10 - lr: 0.000018 - momentum: 0.000000
153
+ 2024-03-26 10:57:10,561 epoch 5 - iter 36/48 - loss 0.13484154 - time (sec): 19.23 - samples/sec: 1411.34 - lr: 0.000018 - momentum: 0.000000
154
+ 2024-03-26 10:57:11,473 epoch 5 - iter 40/48 - loss 0.13458303 - time (sec): 20.14 - samples/sec: 1452.79 - lr: 0.000017 - momentum: 0.000000
155
+ 2024-03-26 10:57:14,103 epoch 5 - iter 44/48 - loss 0.12839504 - time (sec): 22.77 - samples/sec: 1422.22 - lr: 0.000017 - momentum: 0.000000
156
+ 2024-03-26 10:57:15,569 epoch 5 - iter 48/48 - loss 0.12741136 - time (sec): 24.24 - samples/sec: 1422.38 - lr: 0.000017 - momentum: 0.000000
157
+ 2024-03-26 10:57:15,569 ----------------------------------------------------------------------------------------------------
158
+ 2024-03-26 10:57:15,569 EPOCH 5 done: loss 0.1274 - lr: 0.000017
159
+ 2024-03-26 10:57:16,515 DEV : loss 0.20272430777549744 - f1-score (micro avg) 0.8804
160
+ 2024-03-26 10:57:16,516 saving best model
161
+ 2024-03-26 10:57:16,984 ----------------------------------------------------------------------------------------------------
162
+ 2024-03-26 10:57:18,992 epoch 6 - iter 4/48 - loss 0.07592408 - time (sec): 2.01 - samples/sec: 1317.35 - lr: 0.000017 - momentum: 0.000000
163
+ 2024-03-26 10:57:21,186 epoch 6 - iter 8/48 - loss 0.10267334 - time (sec): 4.20 - samples/sec: 1316.74 - lr: 0.000016 - momentum: 0.000000
164
+ 2024-03-26 10:57:23,020 epoch 6 - iter 12/48 - loss 0.10704529 - time (sec): 6.04 - samples/sec: 1431.77 - lr: 0.000016 - momentum: 0.000000
165
+ 2024-03-26 10:57:25,277 epoch 6 - iter 16/48 - loss 0.10464018 - time (sec): 8.29 - samples/sec: 1385.41 - lr: 0.000016 - momentum: 0.000000
166
+ 2024-03-26 10:57:27,081 epoch 6 - iter 20/48 - loss 0.11108015 - time (sec): 10.10 - samples/sec: 1389.85 - lr: 0.000015 - momentum: 0.000000
167
+ 2024-03-26 10:57:29,574 epoch 6 - iter 24/48 - loss 0.10630629 - time (sec): 12.59 - samples/sec: 1367.78 - lr: 0.000015 - momentum: 0.000000
168
+ 2024-03-26 10:57:31,519 epoch 6 - iter 28/48 - loss 0.10378445 - time (sec): 14.53 - samples/sec: 1360.83 - lr: 0.000015 - momentum: 0.000000
169
+ 2024-03-26 10:57:34,025 epoch 6 - iter 32/48 - loss 0.10357046 - time (sec): 17.04 - samples/sec: 1341.33 - lr: 0.000015 - momentum: 0.000000
170
+ 2024-03-26 10:57:37,488 epoch 6 - iter 36/48 - loss 0.09818676 - time (sec): 20.50 - samples/sec: 1300.63 - lr: 0.000014 - momentum: 0.000000
171
+ 2024-03-26 10:57:39,188 epoch 6 - iter 40/48 - loss 0.09582576 - time (sec): 22.20 - samples/sec: 1330.90 - lr: 0.000014 - momentum: 0.000000
172
+ 2024-03-26 10:57:41,073 epoch 6 - iter 44/48 - loss 0.09400591 - time (sec): 24.09 - samples/sec: 1333.22 - lr: 0.000014 - momentum: 0.000000
173
+ 2024-03-26 10:57:42,362 epoch 6 - iter 48/48 - loss 0.09775463 - time (sec): 25.38 - samples/sec: 1358.38 - lr: 0.000014 - momentum: 0.000000
174
+ 2024-03-26 10:57:42,362 ----------------------------------------------------------------------------------------------------
175
+ 2024-03-26 10:57:42,362 EPOCH 6 done: loss 0.0978 - lr: 0.000014
176
+ 2024-03-26 10:57:43,416 DEV : loss 0.20098459720611572 - f1-score (micro avg) 0.8917
177
+ 2024-03-26 10:57:43,417 saving best model
178
+ 2024-03-26 10:57:43,882 ----------------------------------------------------------------------------------------------------
179
+ 2024-03-26 10:57:45,548 epoch 7 - iter 4/48 - loss 0.09646501 - time (sec): 1.67 - samples/sec: 1650.04 - lr: 0.000013 - momentum: 0.000000
180
+ 2024-03-26 10:57:47,647 epoch 7 - iter 8/48 - loss 0.08084940 - time (sec): 3.76 - samples/sec: 1428.49 - lr: 0.000013 - momentum: 0.000000
181
+ 2024-03-26 10:57:49,930 epoch 7 - iter 12/48 - loss 0.08590358 - time (sec): 6.05 - samples/sec: 1372.76 - lr: 0.000013 - momentum: 0.000000
182
+ 2024-03-26 10:57:52,559 epoch 7 - iter 16/48 - loss 0.08260125 - time (sec): 8.68 - samples/sec: 1330.22 - lr: 0.000012 - momentum: 0.000000
183
+ 2024-03-26 10:57:54,936 epoch 7 - iter 20/48 - loss 0.08060122 - time (sec): 11.05 - samples/sec: 1322.25 - lr: 0.000012 - momentum: 0.000000
184
+ 2024-03-26 10:57:56,303 epoch 7 - iter 24/48 - loss 0.07855381 - time (sec): 12.42 - samples/sec: 1377.33 - lr: 0.000012 - momentum: 0.000000
185
+ 2024-03-26 10:57:57,723 epoch 7 - iter 28/48 - loss 0.07878124 - time (sec): 13.84 - samples/sec: 1441.14 - lr: 0.000012 - momentum: 0.000000
186
+ 2024-03-26 10:57:59,728 epoch 7 - iter 32/48 - loss 0.07627449 - time (sec): 15.85 - samples/sec: 1431.35 - lr: 0.000011 - momentum: 0.000000
187
+ 2024-03-26 10:58:01,913 epoch 7 - iter 36/48 - loss 0.07425843 - time (sec): 18.03 - samples/sec: 1420.51 - lr: 0.000011 - momentum: 0.000000
188
+ 2024-03-26 10:58:04,470 epoch 7 - iter 40/48 - loss 0.07509344 - time (sec): 20.59 - samples/sec: 1396.05 - lr: 0.000011 - momentum: 0.000000
189
+ 2024-03-26 10:58:06,346 epoch 7 - iter 44/48 - loss 0.07606956 - time (sec): 22.46 - samples/sec: 1411.21 - lr: 0.000010 - momentum: 0.000000
190
+ 2024-03-26 10:58:08,338 epoch 7 - iter 48/48 - loss 0.07594269 - time (sec): 24.46 - samples/sec: 1409.57 - lr: 0.000010 - momentum: 0.000000
191
+ 2024-03-26 10:58:08,338 ----------------------------------------------------------------------------------------------------
192
+ 2024-03-26 10:58:08,338 EPOCH 7 done: loss 0.0759 - lr: 0.000010
193
+ 2024-03-26 10:58:09,277 DEV : loss 0.18951921164989471 - f1-score (micro avg) 0.8989
194
+ 2024-03-26 10:58:09,278 saving best model
195
+ 2024-03-26 10:58:09,759 ----------------------------------------------------------------------------------------------------
196
+ 2024-03-26 10:58:11,825 epoch 8 - iter 4/48 - loss 0.06179609 - time (sec): 2.07 - samples/sec: 1308.88 - lr: 0.000010 - momentum: 0.000000
197
+ 2024-03-26 10:58:14,623 epoch 8 - iter 8/48 - loss 0.05406802 - time (sec): 4.86 - samples/sec: 1142.42 - lr: 0.000010 - momentum: 0.000000
198
+ 2024-03-26 10:58:15,947 epoch 8 - iter 12/48 - loss 0.05651994 - time (sec): 6.19 - samples/sec: 1290.20 - lr: 0.000009 - momentum: 0.000000
199
+ 2024-03-26 10:58:18,398 epoch 8 - iter 16/48 - loss 0.06729804 - time (sec): 8.64 - samples/sec: 1303.19 - lr: 0.000009 - momentum: 0.000000
200
+ 2024-03-26 10:58:20,988 epoch 8 - iter 20/48 - loss 0.06041800 - time (sec): 11.23 - samples/sec: 1341.99 - lr: 0.000009 - momentum: 0.000000
201
+ 2024-03-26 10:58:22,321 epoch 8 - iter 24/48 - loss 0.06144765 - time (sec): 12.56 - samples/sec: 1416.83 - lr: 0.000009 - momentum: 0.000000
202
+ 2024-03-26 10:58:25,656 epoch 8 - iter 28/48 - loss 0.06053862 - time (sec): 15.90 - samples/sec: 1372.70 - lr: 0.000008 - momentum: 0.000000
203
+ 2024-03-26 10:58:27,673 epoch 8 - iter 32/48 - loss 0.06204024 - time (sec): 17.91 - samples/sec: 1376.29 - lr: 0.000008 - momentum: 0.000000
204
+ 2024-03-26 10:58:28,730 epoch 8 - iter 36/48 - loss 0.06119637 - time (sec): 18.97 - samples/sec: 1415.54 - lr: 0.000008 - momentum: 0.000000
205
+ 2024-03-26 10:58:30,430 epoch 8 - iter 40/48 - loss 0.06095629 - time (sec): 20.67 - samples/sec: 1413.97 - lr: 0.000007 - momentum: 0.000000
206
+ 2024-03-26 10:58:32,073 epoch 8 - iter 44/48 - loss 0.06097632 - time (sec): 22.31 - samples/sec: 1431.98 - lr: 0.000007 - momentum: 0.000000
207
+ 2024-03-26 10:58:34,072 epoch 8 - iter 48/48 - loss 0.06172598 - time (sec): 24.31 - samples/sec: 1417.89 - lr: 0.000007 - momentum: 0.000000
208
+ 2024-03-26 10:58:34,073 ----------------------------------------------------------------------------------------------------
209
+ 2024-03-26 10:58:34,073 EPOCH 8 done: loss 0.0617 - lr: 0.000007
210
+ 2024-03-26 10:58:35,009 DEV : loss 0.19442327320575714 - f1-score (micro avg) 0.9037
211
+ 2024-03-26 10:58:35,010 saving best model
212
+ 2024-03-26 10:58:35,492 ----------------------------------------------------------------------------------------------------
213
+ 2024-03-26 10:58:37,420 epoch 9 - iter 4/48 - loss 0.03787709 - time (sec): 1.93 - samples/sec: 1389.85 - lr: 0.000007 - momentum: 0.000000
214
+ 2024-03-26 10:58:40,657 epoch 9 - iter 8/48 - loss 0.02721893 - time (sec): 5.16 - samples/sec: 1209.21 - lr: 0.000006 - momentum: 0.000000
215
+ 2024-03-26 10:58:42,375 epoch 9 - iter 12/48 - loss 0.03793841 - time (sec): 6.88 - samples/sec: 1262.76 - lr: 0.000006 - momentum: 0.000000
216
+ 2024-03-26 10:58:44,628 epoch 9 - iter 16/48 - loss 0.04341471 - time (sec): 9.13 - samples/sec: 1261.32 - lr: 0.000006 - momentum: 0.000000
217
+ 2024-03-26 10:58:46,982 epoch 9 - iter 20/48 - loss 0.05014430 - time (sec): 11.49 - samples/sec: 1287.78 - lr: 0.000006 - momentum: 0.000000
218
+ 2024-03-26 10:58:49,212 epoch 9 - iter 24/48 - loss 0.05183345 - time (sec): 13.72 - samples/sec: 1303.60 - lr: 0.000005 - momentum: 0.000000
219
+ 2024-03-26 10:58:51,578 epoch 9 - iter 28/48 - loss 0.05030911 - time (sec): 16.09 - samples/sec: 1302.97 - lr: 0.000005 - momentum: 0.000000
220
+ 2024-03-26 10:58:54,009 epoch 9 - iter 32/48 - loss 0.05050612 - time (sec): 18.52 - samples/sec: 1296.84 - lr: 0.000005 - momentum: 0.000000
221
+ 2024-03-26 10:58:55,907 epoch 9 - iter 36/48 - loss 0.05285281 - time (sec): 20.41 - samples/sec: 1311.41 - lr: 0.000004 - momentum: 0.000000
222
+ 2024-03-26 10:58:58,155 epoch 9 - iter 40/48 - loss 0.05425255 - time (sec): 22.66 - samples/sec: 1301.12 - lr: 0.000004 - momentum: 0.000000
223
+ 2024-03-26 10:59:00,300 epoch 9 - iter 44/48 - loss 0.05319438 - time (sec): 24.81 - samples/sec: 1314.23 - lr: 0.000004 - momentum: 0.000000
224
+ 2024-03-26 10:59:01,098 epoch 9 - iter 48/48 - loss 0.05376715 - time (sec): 25.61 - samples/sec: 1346.29 - lr: 0.000004 - momentum: 0.000000
225
+ 2024-03-26 10:59:01,098 ----------------------------------------------------------------------------------------------------
226
+ 2024-03-26 10:59:01,098 EPOCH 9 done: loss 0.0538 - lr: 0.000004
227
+ 2024-03-26 10:59:02,047 DEV : loss 0.18321390450000763 - f1-score (micro avg) 0.9084
228
+ 2024-03-26 10:59:02,048 saving best model
229
+ 2024-03-26 10:59:02,515 ----------------------------------------------------------------------------------------------------
230
+ 2024-03-26 10:59:04,383 epoch 10 - iter 4/48 - loss 0.03045480 - time (sec): 1.87 - samples/sec: 1407.80 - lr: 0.000003 - momentum: 0.000000
231
+ 2024-03-26 10:59:06,446 epoch 10 - iter 8/48 - loss 0.03749981 - time (sec): 3.93 - samples/sec: 1409.71 - lr: 0.000003 - momentum: 0.000000
232
+ 2024-03-26 10:59:09,113 epoch 10 - iter 12/48 - loss 0.03813115 - time (sec): 6.60 - samples/sec: 1322.73 - lr: 0.000003 - momentum: 0.000000
233
+ 2024-03-26 10:59:11,061 epoch 10 - iter 16/48 - loss 0.04613621 - time (sec): 8.55 - samples/sec: 1342.43 - lr: 0.000002 - momentum: 0.000000
234
+ 2024-03-26 10:59:12,999 epoch 10 - iter 20/48 - loss 0.04639203 - time (sec): 10.48 - samples/sec: 1379.82 - lr: 0.000002 - momentum: 0.000000
235
+ 2024-03-26 10:59:14,686 epoch 10 - iter 24/48 - loss 0.05611524 - time (sec): 12.17 - samples/sec: 1393.39 - lr: 0.000002 - momentum: 0.000000
236
+ 2024-03-26 10:59:16,506 epoch 10 - iter 28/48 - loss 0.05319972 - time (sec): 13.99 - samples/sec: 1414.20 - lr: 0.000002 - momentum: 0.000000
237
+ 2024-03-26 10:59:17,731 epoch 10 - iter 32/48 - loss 0.05157465 - time (sec): 15.22 - samples/sec: 1447.46 - lr: 0.000001 - momentum: 0.000000
238
+ 2024-03-26 10:59:20,788 epoch 10 - iter 36/48 - loss 0.04699753 - time (sec): 18.27 - samples/sec: 1401.78 - lr: 0.000001 - momentum: 0.000000
239
+ 2024-03-26 10:59:23,662 epoch 10 - iter 40/48 - loss 0.05009458 - time (sec): 21.15 - samples/sec: 1375.18 - lr: 0.000001 - momentum: 0.000000
240
+ 2024-03-26 10:59:26,468 epoch 10 - iter 44/48 - loss 0.04782682 - time (sec): 23.95 - samples/sec: 1347.81 - lr: 0.000001 - momentum: 0.000000
241
+ 2024-03-26 10:59:28,113 epoch 10 - iter 48/48 - loss 0.04668607 - time (sec): 25.60 - samples/sec: 1346.68 - lr: 0.000000 - momentum: 0.000000
242
+ 2024-03-26 10:59:28,114 ----------------------------------------------------------------------------------------------------
243
+ 2024-03-26 10:59:28,114 EPOCH 10 done: loss 0.0467 - lr: 0.000000
244
+ 2024-03-26 10:59:29,068 DEV : loss 0.18393239378929138 - f1-score (micro avg) 0.9136
245
+ 2024-03-26 10:59:29,069 saving best model
246
+ 2024-03-26 10:59:29,855 ----------------------------------------------------------------------------------------------------
247
+ 2024-03-26 10:59:29,855 Loading model from best epoch ...
248
+ 2024-03-26 10:59:30,803 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
249
+ 2024-03-26 10:59:31,572
250
+ Results:
251
+ - F-score (micro) 0.9003
252
+ - F-score (macro) 0.6853
253
+ - Accuracy 0.821
254
+
255
+ By class:
256
+ precision recall f1-score support
257
+
258
+ Unternehmen 0.8923 0.8722 0.8821 266
259
+ Auslagerung 0.8677 0.8956 0.8814 249
260
+ Ort 0.9706 0.9851 0.9778 134
261
+ Software 0.0000 0.0000 0.0000 0
262
+
263
+ micro avg 0.8962 0.9045 0.9003 649
264
+ macro avg 0.6827 0.6882 0.6853 649
265
+ weighted avg 0.8990 0.9045 0.9016 649
266
+
267
+ 2024-03-26 10:59:31,572 ----------------------------------------------------------------------------------------------------