Hugging Face's logo

stefan-it
/
hmbench-ajmc-de-hmbert_64k-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-5

Token Classification
Flair
PyTorch
TensorBoard
German
sequence-tagger-model
Model card Files Metrics Community
1
stefan-it commited on
Commit
3e21217
•
1 Parent(s): e2b8de6

Upload ./training.log with huggingface_hub

Browse files
Files changed (1) hide show
  1. training.log +246 -0
training.log ADDED
@@ -0,0 +1,246 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-23 15:54:49,408 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-23 15:54:49,409 Model: "SequenceTagger(
3
+ (embeddings): TransformerWordEmbeddings(
4
+ (model): BertModel(
5
+ (embeddings): BertEmbeddings(
6
+ (word_embeddings): Embedding(64001, 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=25, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-23 15:54:49,409 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-23 15:54:49,409 MultiCorpus: 1100 train + 206 dev + 240 test sentences
52
+ - NER_HIPE_2022 Corpus: 1100 train + 206 dev + 240 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/de/with_doc_seperator
53
+ 2023-10-23 15:54:49,409 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-23 15:54:49,409 Train: 1100 sentences
55
+ 2023-10-23 15:54:49,409 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-23 15:54:49,409 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-23 15:54:49,409 Training Params:
58
+ 2023-10-23 15:54:49,409 - learning_rate: "3e-05"
59
+ 2023-10-23 15:54:49,409 - mini_batch_size: "8"
60
+ 2023-10-23 15:54:49,409 - max_epochs: "10"
61
+ 2023-10-23 15:54:49,410 - shuffle: "True"
62
+ 2023-10-23 15:54:49,410 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-23 15:54:49,410 Plugins:
64
+ 2023-10-23 15:54:49,410 - TensorboardLogger
65
+ 2023-10-23 15:54:49,410 - LinearScheduler | warmup_fraction: '0.1'
66
+ 2023-10-23 15:54:49,410 ----------------------------------------------------------------------------------------------------
67
+ 2023-10-23 15:54:49,410 Final evaluation on model from best epoch (best-model.pt)
68
+ 2023-10-23 15:54:49,410 - metric: "('micro avg', 'f1-score')"
69
+ 2023-10-23 15:54:49,410 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-23 15:54:49,410 Computation:
71
+ 2023-10-23 15:54:49,410 - compute on device: cuda:0
72
+ 2023-10-23 15:54:49,410 - embedding storage: none
73
+ 2023-10-23 15:54:49,410 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-23 15:54:49,410 Model training base path: "hmbench-ajmc/de-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-5"
75
+ 2023-10-23 15:54:49,410 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-23 15:54:49,410 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-23 15:54:49,410 Logging anything other than scalars to TensorBoard is currently not supported.
78
+ 2023-10-23 15:54:50,141 epoch 1 - iter 13/138 - loss 3.00405917 - time (sec): 0.73 - samples/sec: 2373.99 - lr: 0.000003 - momentum: 0.000000
79
+ 2023-10-23 15:54:50,875 epoch 1 - iter 26/138 - loss 2.56287802 - time (sec): 1.46 - samples/sec: 2693.60 - lr: 0.000005 - momentum: 0.000000
80
+ 2023-10-23 15:54:51,628 epoch 1 - iter 39/138 - loss 2.05445276 - time (sec): 2.22 - samples/sec: 2813.21 - lr: 0.000008 - momentum: 0.000000
81
+ 2023-10-23 15:54:52,397 epoch 1 - iter 52/138 - loss 1.76142833 - time (sec): 2.99 - samples/sec: 2810.56 - lr: 0.000011 - momentum: 0.000000
82
+ 2023-10-23 15:54:53,182 epoch 1 - iter 65/138 - loss 1.54424671 - time (sec): 3.77 - samples/sec: 2837.08 - lr: 0.000014 - momentum: 0.000000
83
+ 2023-10-23 15:54:53,971 epoch 1 - iter 78/138 - loss 1.35513027 - time (sec): 4.56 - samples/sec: 2841.14 - lr: 0.000017 - momentum: 0.000000
84
+ 2023-10-23 15:54:54,733 epoch 1 - iter 91/138 - loss 1.22250128 - time (sec): 5.32 - samples/sec: 2819.82 - lr: 0.000020 - momentum: 0.000000
85
+ 2023-10-23 15:54:55,485 epoch 1 - iter 104/138 - loss 1.11598652 - time (sec): 6.07 - samples/sec: 2841.53 - lr: 0.000022 - momentum: 0.000000
86
+ 2023-10-23 15:54:56,239 epoch 1 - iter 117/138 - loss 1.02403952 - time (sec): 6.83 - samples/sec: 2874.76 - lr: 0.000025 - momentum: 0.000000
87
+ 2023-10-23 15:54:56,997 epoch 1 - iter 130/138 - loss 0.95915259 - time (sec): 7.59 - samples/sec: 2853.35 - lr: 0.000028 - momentum: 0.000000
88
+ 2023-10-23 15:54:57,457 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-23 15:54:57,457 EPOCH 1 done: loss 0.9234 - lr: 0.000028
90
+ 2023-10-23 15:54:58,036 DEV : loss 0.19589945673942566 - f1-score (micro avg) 0.7295
91
+ 2023-10-23 15:54:58,042 saving best model
92
+ 2023-10-23 15:54:58,436 ----------------------------------------------------------------------------------------------------
93
+ 2023-10-23 15:54:59,187 epoch 2 - iter 13/138 - loss 0.24513018 - time (sec): 0.75 - samples/sec: 3157.13 - lr: 0.000030 - momentum: 0.000000
94
+ 2023-10-23 15:54:59,947 epoch 2 - iter 26/138 - loss 0.23709938 - time (sec): 1.51 - samples/sec: 2974.25 - lr: 0.000029 - momentum: 0.000000
95
+ 2023-10-23 15:55:00,669 epoch 2 - iter 39/138 - loss 0.20795711 - time (sec): 2.23 - samples/sec: 2891.60 - lr: 0.000029 - momentum: 0.000000
96
+ 2023-10-23 15:55:01,392 epoch 2 - iter 52/138 - loss 0.20218652 - time (sec): 2.95 - samples/sec: 2802.42 - lr: 0.000029 - momentum: 0.000000
97
+ 2023-10-23 15:55:02,109 epoch 2 - iter 65/138 - loss 0.19026323 - time (sec): 3.67 - samples/sec: 2809.17 - lr: 0.000028 - momentum: 0.000000
98
+ 2023-10-23 15:55:02,835 epoch 2 - iter 78/138 - loss 0.18132874 - time (sec): 4.40 - samples/sec: 2844.07 - lr: 0.000028 - momentum: 0.000000
99
+ 2023-10-23 15:55:03,562 epoch 2 - iter 91/138 - loss 0.17220623 - time (sec): 5.12 - samples/sec: 2839.33 - lr: 0.000028 - momentum: 0.000000
100
+ 2023-10-23 15:55:04,321 epoch 2 - iter 104/138 - loss 0.16856554 - time (sec): 5.88 - samples/sec: 2842.82 - lr: 0.000028 - momentum: 0.000000
101
+ 2023-10-23 15:55:05,080 epoch 2 - iter 117/138 - loss 0.17136679 - time (sec): 6.64 - samples/sec: 2867.72 - lr: 0.000027 - momentum: 0.000000
102
+ 2023-10-23 15:55:05,831 epoch 2 - iter 130/138 - loss 0.16835947 - time (sec): 7.39 - samples/sec: 2904.80 - lr: 0.000027 - momentum: 0.000000
103
+ 2023-10-23 15:55:06,309 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-23 15:55:06,309 EPOCH 2 done: loss 0.1688 - lr: 0.000027
105
+ 2023-10-23 15:55:06,848 DEV : loss 0.1276298463344574 - f1-score (micro avg) 0.8109
106
+ 2023-10-23 15:55:06,854 saving best model
107
+ 2023-10-23 15:55:07,392 ----------------------------------------------------------------------------------------------------
108
+ 2023-10-23 15:55:08,123 epoch 3 - iter 13/138 - loss 0.11073512 - time (sec): 0.73 - samples/sec: 3098.87 - lr: 0.000026 - momentum: 0.000000
109
+ 2023-10-23 15:55:08,869 epoch 3 - iter 26/138 - loss 0.09498375 - time (sec): 1.47 - samples/sec: 3028.40 - lr: 0.000026 - momentum: 0.000000
110
+ 2023-10-23 15:55:09,584 epoch 3 - iter 39/138 - loss 0.09514575 - time (sec): 2.19 - samples/sec: 2991.48 - lr: 0.000026 - momentum: 0.000000
111
+ 2023-10-23 15:55:10,302 epoch 3 - iter 52/138 - loss 0.10425731 - time (sec): 2.91 - samples/sec: 3042.18 - lr: 0.000025 - momentum: 0.000000
112
+ 2023-10-23 15:55:11,014 epoch 3 - iter 65/138 - loss 0.09912849 - time (sec): 3.62 - samples/sec: 3030.06 - lr: 0.000025 - momentum: 0.000000
113
+ 2023-10-23 15:55:11,744 epoch 3 - iter 78/138 - loss 0.09280214 - time (sec): 4.35 - samples/sec: 3000.12 - lr: 0.000025 - momentum: 0.000000
114
+ 2023-10-23 15:55:12,486 epoch 3 - iter 91/138 - loss 0.09344571 - time (sec): 5.09 - samples/sec: 2982.33 - lr: 0.000025 - momentum: 0.000000
115
+ 2023-10-23 15:55:13,228 epoch 3 - iter 104/138 - loss 0.09045760 - time (sec): 5.83 - samples/sec: 2967.07 - lr: 0.000024 - momentum: 0.000000
116
+ 2023-10-23 15:55:13,972 epoch 3 - iter 117/138 - loss 0.09303884 - time (sec): 6.58 - samples/sec: 2971.02 - lr: 0.000024 - momentum: 0.000000
117
+ 2023-10-23 15:55:14,712 epoch 3 - iter 130/138 - loss 0.09075384 - time (sec): 7.32 - samples/sec: 2950.98 - lr: 0.000024 - momentum: 0.000000
118
+ 2023-10-23 15:55:15,164 ----------------------------------------------------------------------------------------------------
119
+ 2023-10-23 15:55:15,164 EPOCH 3 done: loss 0.0923 - lr: 0.000024
120
+ 2023-10-23 15:55:15,702 DEV : loss 0.12066850066184998 - f1-score (micro avg) 0.8379
121
+ 2023-10-23 15:55:15,708 saving best model
122
+ 2023-10-23 15:55:16,258 ----------------------------------------------------------------------------------------------------
123
+ 2023-10-23 15:55:16,991 epoch 4 - iter 13/138 - loss 0.11350050 - time (sec): 0.73 - samples/sec: 3101.74 - lr: 0.000023 - momentum: 0.000000
124
+ 2023-10-23 15:55:17,700 epoch 4 - iter 26/138 - loss 0.08075514 - time (sec): 1.44 - samples/sec: 3158.96 - lr: 0.000023 - momentum: 0.000000
125
+ 2023-10-23 15:55:18,428 epoch 4 - iter 39/138 - loss 0.07346125 - time (sec): 2.17 - samples/sec: 3008.78 - lr: 0.000022 - momentum: 0.000000
126
+ 2023-10-23 15:55:19,126 epoch 4 - iter 52/138 - loss 0.06656305 - time (sec): 2.87 - samples/sec: 2975.49 - lr: 0.000022 - momentum: 0.000000
127
+ 2023-10-23 15:55:19,869 epoch 4 - iter 65/138 - loss 0.06313594 - time (sec): 3.61 - samples/sec: 2992.76 - lr: 0.000022 - momentum: 0.000000
128
+ 2023-10-23 15:55:20,597 epoch 4 - iter 78/138 - loss 0.06522584 - time (sec): 4.34 - samples/sec: 3016.50 - lr: 0.000021 - momentum: 0.000000
129
+ 2023-10-23 15:55:21,310 epoch 4 - iter 91/138 - loss 0.06281046 - time (sec): 5.05 - samples/sec: 3010.97 - lr: 0.000021 - momentum: 0.000000
130
+ 2023-10-23 15:55:22,054 epoch 4 - iter 104/138 - loss 0.06475346 - time (sec): 5.79 - samples/sec: 3005.96 - lr: 0.000021 - momentum: 0.000000
131
+ 2023-10-23 15:55:22,749 epoch 4 - iter 117/138 - loss 0.06295527 - time (sec): 6.49 - samples/sec: 2979.84 - lr: 0.000021 - momentum: 0.000000
132
+ 2023-10-23 15:55:23,481 epoch 4 - iter 130/138 - loss 0.06220209 - time (sec): 7.22 - samples/sec: 2944.77 - lr: 0.000020 - momentum: 0.000000
133
+ 2023-10-23 15:55:23,918 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-23 15:55:23,918 EPOCH 4 done: loss 0.0608 - lr: 0.000020
135
+ 2023-10-23 15:55:24,446 DEV : loss 0.1310967206954956 - f1-score (micro avg) 0.8609
136
+ 2023-10-23 15:55:24,452 saving best model
137
+ 2023-10-23 15:55:24,998 ----------------------------------------------------------------------------------------------------
138
+ 2023-10-23 15:55:25,720 epoch 5 - iter 13/138 - loss 0.05355793 - time (sec): 0.72 - samples/sec: 2992.96 - lr: 0.000020 - momentum: 0.000000
139
+ 2023-10-23 15:55:26,423 epoch 5 - iter 26/138 - loss 0.06962420 - time (sec): 1.42 - samples/sec: 2987.17 - lr: 0.000019 - momentum: 0.000000
140
+ 2023-10-23 15:55:27,155 epoch 5 - iter 39/138 - loss 0.05818477 - time (sec): 2.16 - samples/sec: 3006.77 - lr: 0.000019 - momentum: 0.000000
141
+ 2023-10-23 15:55:27,903 epoch 5 - iter 52/138 - loss 0.04671793 - time (sec): 2.90 - samples/sec: 2959.51 - lr: 0.000019 - momentum: 0.000000
142
+ 2023-10-23 15:55:28,647 epoch 5 - iter 65/138 - loss 0.04604634 - time (sec): 3.65 - samples/sec: 2997.02 - lr: 0.000018 - momentum: 0.000000
143
+ 2023-10-23 15:55:29,391 epoch 5 - iter 78/138 - loss 0.04370062 - time (sec): 4.39 - samples/sec: 2928.05 - lr: 0.000018 - momentum: 0.000000
144
+ 2023-10-23 15:55:30,130 epoch 5 - iter 91/138 - loss 0.04395983 - time (sec): 5.13 - samples/sec: 2938.58 - lr: 0.000018 - momentum: 0.000000
145
+ 2023-10-23 15:55:30,889 epoch 5 - iter 104/138 - loss 0.04812229 - time (sec): 5.89 - samples/sec: 2925.69 - lr: 0.000018 - momentum: 0.000000
146
+ 2023-10-23 15:55:31,639 epoch 5 - iter 117/138 - loss 0.04842003 - time (sec): 6.64 - samples/sec: 2891.87 - lr: 0.000017 - momentum: 0.000000
147
+ 2023-10-23 15:55:32,383 epoch 5 - iter 130/138 - loss 0.04784414 - time (sec): 7.38 - samples/sec: 2887.52 - lr: 0.000017 - momentum: 0.000000
148
+ 2023-10-23 15:55:32,838 ----------------------------------------------------------------------------------------------------
149
+ 2023-10-23 15:55:32,838 EPOCH 5 done: loss 0.0472 - lr: 0.000017
150
+ 2023-10-23 15:55:33,377 DEV : loss 0.13556700944900513 - f1-score (micro avg) 0.881
151
+ 2023-10-23 15:55:33,383 saving best model
152
+ 2023-10-23 15:55:33,924 ----------------------------------------------------------------------------------------------------
153
+ 2023-10-23 15:55:34,699 epoch 6 - iter 13/138 - loss 0.00447347 - time (sec): 0.77 - samples/sec: 3105.62 - lr: 0.000016 - momentum: 0.000000
154
+ 2023-10-23 15:55:35,433 epoch 6 - iter 26/138 - loss 0.03663128 - time (sec): 1.51 - samples/sec: 2857.73 - lr: 0.000016 - momentum: 0.000000
155
+ 2023-10-23 15:55:36,190 epoch 6 - iter 39/138 - loss 0.02970269 - time (sec): 2.26 - samples/sec: 2836.41 - lr: 0.000016 - momentum: 0.000000
156
+ 2023-10-23 15:55:36,929 epoch 6 - iter 52/138 - loss 0.03038221 - time (sec): 3.00 - samples/sec: 2898.43 - lr: 0.000015 - momentum: 0.000000
157
+ 2023-10-23 15:55:37,666 epoch 6 - iter 65/138 - loss 0.02843939 - time (sec): 3.74 - samples/sec: 2877.33 - lr: 0.000015 - momentum: 0.000000
158
+ 2023-10-23 15:55:38,403 epoch 6 - iter 78/138 - loss 0.03163609 - time (sec): 4.48 - samples/sec: 2867.41 - lr: 0.000015 - momentum: 0.000000
159
+ 2023-10-23 15:55:39,145 epoch 6 - iter 91/138 - loss 0.03089174 - time (sec): 5.22 - samples/sec: 2838.67 - lr: 0.000015 - momentum: 0.000000
160
+ 2023-10-23 15:55:39,889 epoch 6 - iter 104/138 - loss 0.03317619 - time (sec): 5.96 - samples/sec: 2839.11 - lr: 0.000014 - momentum: 0.000000
161
+ 2023-10-23 15:55:40,628 epoch 6 - iter 117/138 - loss 0.03642630 - time (sec): 6.70 - samples/sec: 2865.20 - lr: 0.000014 - momentum: 0.000000
162
+ 2023-10-23 15:55:41,372 epoch 6 - iter 130/138 - loss 0.03570597 - time (sec): 7.44 - samples/sec: 2869.61 - lr: 0.000014 - momentum: 0.000000
163
+ 2023-10-23 15:55:41,829 ----------------------------------------------------------------------------------------------------
164
+ 2023-10-23 15:55:41,829 EPOCH 6 done: loss 0.0388 - lr: 0.000014
165
+ 2023-10-23 15:55:42,364 DEV : loss 0.12986908853054047 - f1-score (micro avg) 0.8857
166
+ 2023-10-23 15:55:42,370 saving best model
167
+ 2023-10-23 15:55:42,912 ----------------------------------------------------------------------------------------------------
168
+ 2023-10-23 15:55:43,641 epoch 7 - iter 13/138 - loss 0.04560714 - time (sec): 0.72 - samples/sec: 3046.17 - lr: 0.000013 - momentum: 0.000000
169
+ 2023-10-23 15:55:44,362 epoch 7 - iter 26/138 - loss 0.04097470 - time (sec): 1.45 - samples/sec: 2967.32 - lr: 0.000013 - momentum: 0.000000
170
+ 2023-10-23 15:55:45,104 epoch 7 - iter 39/138 - loss 0.03509241 - time (sec): 2.19 - samples/sec: 2861.35 - lr: 0.000012 - momentum: 0.000000
171
+ 2023-10-23 15:55:45,854 epoch 7 - iter 52/138 - loss 0.03079550 - time (sec): 2.94 - samples/sec: 2872.75 - lr: 0.000012 - momentum: 0.000000
172
+ 2023-10-23 15:55:46,591 epoch 7 - iter 65/138 - loss 0.02734683 - time (sec): 3.67 - samples/sec: 2913.33 - lr: 0.000012 - momentum: 0.000000
173
+ 2023-10-23 15:55:47,325 epoch 7 - iter 78/138 - loss 0.02674504 - time (sec): 4.41 - samples/sec: 2904.82 - lr: 0.000012 - momentum: 0.000000
174
+ 2023-10-23 15:55:48,052 epoch 7 - iter 91/138 - loss 0.02597626 - time (sec): 5.14 - samples/sec: 2898.08 - lr: 0.000011 - momentum: 0.000000
175
+ 2023-10-23 15:55:48,786 epoch 7 - iter 104/138 - loss 0.02279908 - time (sec): 5.87 - samples/sec: 2938.68 - lr: 0.000011 - momentum: 0.000000
176
+ 2023-10-23 15:55:49,527 epoch 7 - iter 117/138 - loss 0.02768589 - time (sec): 6.61 - samples/sec: 2959.75 - lr: 0.000011 - momentum: 0.000000
177
+ 2023-10-23 15:55:50,264 epoch 7 - iter 130/138 - loss 0.02689635 - time (sec): 7.35 - samples/sec: 2936.27 - lr: 0.000010 - momentum: 0.000000
178
+ 2023-10-23 15:55:50,716 ----------------------------------------------------------------------------------------------------
179
+ 2023-10-23 15:55:50,716 EPOCH 7 done: loss 0.0261 - lr: 0.000010
180
+ 2023-10-23 15:55:51,252 DEV : loss 0.14399342238903046 - f1-score (micro avg) 0.8766
181
+ 2023-10-23 15:55:51,258 ----------------------------------------------------------------------------------------------------
182
+ 2023-10-23 15:55:52,007 epoch 8 - iter 13/138 - loss 0.03629714 - time (sec): 0.75 - samples/sec: 2708.89 - lr: 0.000010 - momentum: 0.000000
183
+ 2023-10-23 15:55:52,767 epoch 8 - iter 26/138 - loss 0.02677088 - time (sec): 1.51 - samples/sec: 2954.84 - lr: 0.000009 - momentum: 0.000000
184
+ 2023-10-23 15:55:53,495 epoch 8 - iter 39/138 - loss 0.03630954 - time (sec): 2.24 - samples/sec: 2889.29 - lr: 0.000009 - momentum: 0.000000
185
+ 2023-10-23 15:55:54,276 epoch 8 - iter 52/138 - loss 0.03402449 - time (sec): 3.02 - samples/sec: 2934.37 - lr: 0.000009 - momentum: 0.000000
186
+ 2023-10-23 15:55:55,016 epoch 8 - iter 65/138 - loss 0.03031141 - time (sec): 3.76 - samples/sec: 2962.17 - lr: 0.000009 - momentum: 0.000000
187
+ 2023-10-23 15:55:55,781 epoch 8 - iter 78/138 - loss 0.02709042 - time (sec): 4.52 - samples/sec: 2964.19 - lr: 0.000008 - momentum: 0.000000
188
+ 2023-10-23 15:55:56,513 epoch 8 - iter 91/138 - loss 0.02482129 - time (sec): 5.25 - samples/sec: 2943.89 - lr: 0.000008 - momentum: 0.000000
189
+ 2023-10-23 15:55:57,244 epoch 8 - iter 104/138 - loss 0.02348331 - time (sec): 5.98 - samples/sec: 2909.14 - lr: 0.000008 - momentum: 0.000000
190
+ 2023-10-23 15:55:58,010 epoch 8 - iter 117/138 - loss 0.02455497 - time (sec): 6.75 - samples/sec: 2883.47 - lr: 0.000007 - momentum: 0.000000
191
+ 2023-10-23 15:55:58,754 epoch 8 - iter 130/138 - loss 0.02442744 - time (sec): 7.49 - samples/sec: 2877.71 - lr: 0.000007 - momentum: 0.000000
192
+ 2023-10-23 15:55:59,224 ----------------------------------------------------------------------------------------------------
193
+ 2023-10-23 15:55:59,224 EPOCH 8 done: loss 0.0243 - lr: 0.000007
194
+ 2023-10-23 15:55:59,776 DEV : loss 0.14874805510044098 - f1-score (micro avg) 0.8948
195
+ 2023-10-23 15:55:59,782 saving best model
196
+ 2023-10-23 15:56:00,334 ----------------------------------------------------------------------------------------------------
197
+ 2023-10-23 15:56:01,062 epoch 9 - iter 13/138 - loss 0.01544187 - time (sec): 0.72 - samples/sec: 3068.12 - lr: 0.000006 - momentum: 0.000000
198
+ 2023-10-23 15:56:01,793 epoch 9 - iter 26/138 - loss 0.01943197 - time (sec): 1.45 - samples/sec: 2962.08 - lr: 0.000006 - momentum: 0.000000
199
+ 2023-10-23 15:56:02,529 epoch 9 - iter 39/138 - loss 0.02702448 - time (sec): 2.19 - samples/sec: 3028.97 - lr: 0.000006 - momentum: 0.000000
200
+ 2023-10-23 15:56:03,253 epoch 9 - iter 52/138 - loss 0.02759916 - time (sec): 2.91 - samples/sec: 2961.00 - lr: 0.000005 - momentum: 0.000000
201
+ 2023-10-23 15:56:03,988 epoch 9 - iter 65/138 - loss 0.02404425 - time (sec): 3.65 - samples/sec: 2930.40 - lr: 0.000005 - momentum: 0.000000
202
+ 2023-10-23 15:56:04,729 epoch 9 - iter 78/138 - loss 0.01987204 - time (sec): 4.39 - samples/sec: 3004.47 - lr: 0.000005 - momentum: 0.000000
203
+ 2023-10-23 15:56:05,471 epoch 9 - iter 91/138 - loss 0.01814794 - time (sec): 5.13 - samples/sec: 2981.58 - lr: 0.000005 - momentum: 0.000000
204
+ 2023-10-23 15:56:06,204 epoch 9 - iter 104/138 - loss 0.01732655 - time (sec): 5.86 - samples/sec: 2973.69 - lr: 0.000004 - momentum: 0.000000
205
+ 2023-10-23 15:56:06,931 epoch 9 - iter 117/138 - loss 0.01633533 - time (sec): 6.59 - samples/sec: 2991.47 - lr: 0.000004 - momentum: 0.000000
206
+ 2023-10-23 15:56:07,675 epoch 9 - iter 130/138 - loss 0.01746274 - time (sec): 7.33 - samples/sec: 2956.10 - lr: 0.000004 - momentum: 0.000000
207
+ 2023-10-23 15:56:08,128 ----------------------------------------------------------------------------------------------------
208
+ 2023-10-23 15:56:08,129 EPOCH 9 done: loss 0.0179 - lr: 0.000004
209
+ 2023-10-23 15:56:08,663 DEV : loss 0.1484120488166809 - f1-score (micro avg) 0.8852
210
+ 2023-10-23 15:56:08,669 ----------------------------------------------------------------------------------------------------
211
+ 2023-10-23 15:56:09,391 epoch 10 - iter 13/138 - loss 0.02130054 - time (sec): 0.72 - samples/sec: 2830.96 - lr: 0.000003 - momentum: 0.000000
212
+ 2023-10-23 15:56:10,108 epoch 10 - iter 26/138 - loss 0.01334521 - time (sec): 1.44 - samples/sec: 2911.14 - lr: 0.000003 - momentum: 0.000000
213
+ 2023-10-23 15:56:10,826 epoch 10 - iter 39/138 - loss 0.01106197 - time (sec): 2.16 - samples/sec: 2984.20 - lr: 0.000002 - momentum: 0.000000
214
+ 2023-10-23 15:56:11,576 epoch 10 - iter 52/138 - loss 0.00961286 - time (sec): 2.91 - samples/sec: 3018.09 - lr: 0.000002 - momentum: 0.000000
215
+ 2023-10-23 15:56:12,281 epoch 10 - iter 65/138 - loss 0.00864498 - time (sec): 3.61 - samples/sec: 2954.18 - lr: 0.000002 - momentum: 0.000000
216
+ 2023-10-23 15:56:13,017 epoch 10 - iter 78/138 - loss 0.01211412 - time (sec): 4.35 - samples/sec: 2938.79 - lr: 0.000002 - momentum: 0.000000
217
+ 2023-10-23 15:56:13,741 epoch 10 - iter 91/138 - loss 0.01154617 - time (sec): 5.07 - samples/sec: 2914.28 - lr: 0.000001 - momentum: 0.000000
218
+ 2023-10-23 15:56:14,451 epoch 10 - iter 104/138 - loss 0.01059741 - time (sec): 5.78 - samples/sec: 2936.14 - lr: 0.000001 - momentum: 0.000000
219
+ 2023-10-23 15:56:15,177 epoch 10 - iter 117/138 - loss 0.01141697 - time (sec): 6.51 - samples/sec: 2935.76 - lr: 0.000001 - momentum: 0.000000
220
+ 2023-10-23 15:56:15,924 epoch 10 - iter 130/138 - loss 0.01581859 - time (sec): 7.25 - samples/sec: 2959.38 - lr: 0.000000 - momentum: 0.000000
221
+ 2023-10-23 15:56:16,375 ----------------------------------------------------------------------------------------------------
222
+ 2023-10-23 15:56:16,375 EPOCH 10 done: loss 0.0156 - lr: 0.000000
223
+ 2023-10-23 15:56:16,904 DEV : loss 0.14890803396701813 - f1-score (micro avg) 0.8918
224
+ 2023-10-23 15:56:17,307 ----------------------------------------------------------------------------------------------------
225
+ 2023-10-23 15:56:17,308 Loading model from best epoch ...
226
+ 2023-10-23 15:56:18,949 SequenceTagger predicts: Dictionary with 25 tags: O, S-scope, B-scope, E-scope, I-scope, S-pers, B-pers, E-pers, I-pers, S-work, B-work, E-work, I-work, S-loc, B-loc, E-loc, I-loc, S-object, B-object, E-object, I-object, S-date, B-date, E-date, I-date
227
+ 2023-10-23 15:56:19,626
228
+ Results:
229
+ - F-score (micro) 0.9026
230
+ - F-score (macro) 0.8039
231
+ - Accuracy 0.8265
232
+
233
+ By class:
234
+ precision recall f1-score support
235
+
236
+ scope 0.8827 0.8977 0.8901 176
237
+ pers 0.9612 0.9688 0.9650 128
238
+ work 0.8676 0.7973 0.8310 74
239
+ object 1.0000 0.5000 0.6667 2
240
+ loc 1.0000 0.5000 0.6667 2
241
+
242
+ micro avg 0.9074 0.8979 0.9026 382
243
+ macro avg 0.9423 0.7328 0.8039 382
244
+ weighted avg 0.9073 0.8979 0.9014 382
245
+
246
+ 2023-10-23 15:56:19,626 ----------------------------------------------------------------------------------------------------