Hugging Face's logo

stefan-it
/
hmbench-newseye-fi-hmbert_64k-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4

Token Classification
Flair
PyTorch
TensorBoard
Finnish
sequence-tagger-model
Model card Files Metrics Community
stefan-it commited on
Commit
b8ee441
•
1 Parent(s): fdad6c1

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
+ 2023-10-25 21:17:46,995 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-25 21:17:46,996 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=17, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-25 21:17:46,996 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-25 21:17:46,996 MultiCorpus: 1166 train + 165 dev + 415 test sentences
52
+ - NER_HIPE_2022 Corpus: 1166 train + 165 dev + 415 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fi/with_doc_seperator
53
+ 2023-10-25 21:17:46,996 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-25 21:17:46,996 Train: 1166 sentences
55
+ 2023-10-25 21:17:46,996 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-25 21:17:46,996 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-25 21:17:46,996 Training Params:
58
+ 2023-10-25 21:17:46,996 - learning_rate: "5e-05"
59
+ 2023-10-25 21:17:46,996 - mini_batch_size: "8"
60
+ 2023-10-25 21:17:46,996 - max_epochs: "10"
61
+ 2023-10-25 21:17:46,996 - shuffle: "True"
62
+ 2023-10-25 21:17:46,996 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-25 21:17:46,996 Plugins:
64
+ 2023-10-25 21:17:46,997 - TensorboardLogger
65
+ 2023-10-25 21:17:46,997 - LinearScheduler | warmup_fraction: '0.1'
66
+ 2023-10-25 21:17:46,997 ----------------------------------------------------------------------------------------------------
67
+ 2023-10-25 21:17:46,997 Final evaluation on model from best epoch (best-model.pt)
68
+ 2023-10-25 21:17:46,997 - metric: "('micro avg', 'f1-score')"
69
+ 2023-10-25 21:17:46,997 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-25 21:17:46,997 Computation:
71
+ 2023-10-25 21:17:46,997 - compute on device: cuda:0
72
+ 2023-10-25 21:17:46,997 - embedding storage: none
73
+ 2023-10-25 21:17:46,997 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-25 21:17:46,997 Model training base path: "hmbench-newseye/fi-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4"
75
+ 2023-10-25 21:17:46,997 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-25 21:17:46,997 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-25 21:17:46,997 Logging anything other than scalars to TensorBoard is currently not supported.
78
+ 2023-10-25 21:17:47,929 epoch 1 - iter 14/146 - loss 3.25226463 - time (sec): 0.93 - samples/sec: 4885.87 - lr: 0.000004 - momentum: 0.000000
79
+ 2023-10-25 21:17:48,770 epoch 1 - iter 28/146 - loss 2.55317170 - time (sec): 1.77 - samples/sec: 4857.98 - lr: 0.000009 - momentum: 0.000000
80
+ 2023-10-25 21:17:49,542 epoch 1 - iter 42/146 - loss 2.05595893 - time (sec): 2.54 - samples/sec: 4866.50 - lr: 0.000014 - momentum: 0.000000
81
+ 2023-10-25 21:17:50,368 epoch 1 - iter 56/146 - loss 1.67247460 - time (sec): 3.37 - samples/sec: 4975.54 - lr: 0.000019 - momentum: 0.000000
82
+ 2023-10-25 21:17:51,219 epoch 1 - iter 70/146 - loss 1.44306959 - time (sec): 4.22 - samples/sec: 4875.31 - lr: 0.000024 - momentum: 0.000000
83
+ 2023-10-25 21:17:52,277 epoch 1 - iter 84/146 - loss 1.24458071 - time (sec): 5.28 - samples/sec: 4795.51 - lr: 0.000028 - momentum: 0.000000
84
+ 2023-10-25 21:17:53,133 epoch 1 - iter 98/146 - loss 1.10964271 - time (sec): 6.14 - samples/sec: 4834.68 - lr: 0.000033 - momentum: 0.000000
85
+ 2023-10-25 21:17:53,951 epoch 1 - iter 112/146 - loss 1.01272313 - time (sec): 6.95 - samples/sec: 4803.87 - lr: 0.000038 - momentum: 0.000000
86
+ 2023-10-25 21:17:54,877 epoch 1 - iter 126/146 - loss 0.90550822 - time (sec): 7.88 - samples/sec: 4861.78 - lr: 0.000043 - momentum: 0.000000
87
+ 2023-10-25 21:17:55,745 epoch 1 - iter 140/146 - loss 0.83647696 - time (sec): 8.75 - samples/sec: 4825.74 - lr: 0.000048 - momentum: 0.000000
88
+ 2023-10-25 21:17:56,215 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-25 21:17:56,215 EPOCH 1 done: loss 0.8028 - lr: 0.000048
90
+ 2023-10-25 21:17:56,870 DEV : loss 0.15297161042690277 - f1-score (micro avg) 0.5579
91
+ 2023-10-25 21:17:56,874 saving best model
92
+ 2023-10-25 21:17:57,259 ----------------------------------------------------------------------------------------------------
93
+ 2023-10-25 21:17:58,168 epoch 2 - iter 14/146 - loss 0.22538995 - time (sec): 0.91 - samples/sec: 4654.15 - lr: 0.000050 - momentum: 0.000000
94
+ 2023-10-25 21:17:59,136 epoch 2 - iter 28/146 - loss 0.17893266 - time (sec): 1.88 - samples/sec: 4749.32 - lr: 0.000049 - momentum: 0.000000
95
+ 2023-10-25 21:17:59,957 epoch 2 - iter 42/146 - loss 0.17000545 - time (sec): 2.70 - samples/sec: 4841.19 - lr: 0.000048 - momentum: 0.000000
96
+ 2023-10-25 21:18:00,877 epoch 2 - iter 56/146 - loss 0.17256647 - time (sec): 3.62 - samples/sec: 4838.73 - lr: 0.000048 - momentum: 0.000000
97
+ 2023-10-25 21:18:01,792 epoch 2 - iter 70/146 - loss 0.16856484 - time (sec): 4.53 - samples/sec: 4708.65 - lr: 0.000047 - momentum: 0.000000
98
+ 2023-10-25 21:18:02,569 epoch 2 - iter 84/146 - loss 0.16763435 - time (sec): 5.31 - samples/sec: 4689.76 - lr: 0.000047 - momentum: 0.000000
99
+ 2023-10-25 21:18:03,413 epoch 2 - iter 98/146 - loss 0.16570481 - time (sec): 6.15 - samples/sec: 4716.09 - lr: 0.000046 - momentum: 0.000000
100
+ 2023-10-25 21:18:04,405 epoch 2 - iter 112/146 - loss 0.16574954 - time (sec): 7.14 - samples/sec: 4690.41 - lr: 0.000046 - momentum: 0.000000
101
+ 2023-10-25 21:18:05,376 epoch 2 - iter 126/146 - loss 0.16070878 - time (sec): 8.12 - samples/sec: 4703.69 - lr: 0.000045 - momentum: 0.000000
102
+ 2023-10-25 21:18:06,290 epoch 2 - iter 140/146 - loss 0.15828404 - time (sec): 9.03 - samples/sec: 4735.04 - lr: 0.000045 - momentum: 0.000000
103
+ 2023-10-25 21:18:06,674 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-25 21:18:06,674 EPOCH 2 done: loss 0.1575 - lr: 0.000045
105
+ 2023-10-25 21:18:07,586 DEV : loss 0.10719096660614014 - f1-score (micro avg) 0.689
106
+ 2023-10-25 21:18:07,590 saving best model
107
+ 2023-10-25 21:18:08,264 ----------------------------------------------------------------------------------------------------
108
+ 2023-10-25 21:18:09,307 epoch 3 - iter 14/146 - loss 0.08892807 - time (sec): 1.04 - samples/sec: 4531.22 - lr: 0.000044 - momentum: 0.000000
109
+ 2023-10-25 21:18:10,135 epoch 3 - iter 28/146 - loss 0.09012832 - time (sec): 1.87 - samples/sec: 4706.43 - lr: 0.000043 - momentum: 0.000000
110
+ 2023-10-25 21:18:10,916 epoch 3 - iter 42/146 - loss 0.09342662 - time (sec): 2.65 - samples/sec: 4776.18 - lr: 0.000043 - momentum: 0.000000
111
+ 2023-10-25 21:18:11,924 epoch 3 - iter 56/146 - loss 0.10492414 - time (sec): 3.66 - samples/sec: 4678.82 - lr: 0.000042 - momentum: 0.000000
112
+ 2023-10-25 21:18:12,834 epoch 3 - iter 70/146 - loss 0.11480081 - time (sec): 4.57 - samples/sec: 4716.90 - lr: 0.000042 - momentum: 0.000000
113
+ 2023-10-25 21:18:13,629 epoch 3 - iter 84/146 - loss 0.11121453 - time (sec): 5.36 - samples/sec: 4662.18 - lr: 0.000041 - momentum: 0.000000
114
+ 2023-10-25 21:18:14,461 epoch 3 - iter 98/146 - loss 0.10452791 - time (sec): 6.19 - samples/sec: 4690.07 - lr: 0.000041 - momentum: 0.000000
115
+ 2023-10-25 21:18:15,365 epoch 3 - iter 112/146 - loss 0.10131230 - time (sec): 7.10 - samples/sec: 4669.60 - lr: 0.000040 - momentum: 0.000000
116
+ 2023-10-25 21:18:16,388 epoch 3 - iter 126/146 - loss 0.10122777 - time (sec): 8.12 - samples/sec: 4643.65 - lr: 0.000040 - momentum: 0.000000
117
+ 2023-10-25 21:18:17,337 epoch 3 - iter 140/146 - loss 0.09817162 - time (sec): 9.07 - samples/sec: 4660.96 - lr: 0.000039 - momentum: 0.000000
118
+ 2023-10-25 21:18:17,762 ----------------------------------------------------------------------------------------------------
119
+ 2023-10-25 21:18:17,762 EPOCH 3 done: loss 0.0959 - lr: 0.000039
120
+ 2023-10-25 21:18:18,668 DEV : loss 0.1064475029706955 - f1-score (micro avg) 0.7389
121
+ 2023-10-25 21:18:18,673 saving best model
122
+ 2023-10-25 21:18:19,350 ----------------------------------------------------------------------------------------------------
123
+ 2023-10-25 21:18:20,237 epoch 4 - iter 14/146 - loss 0.04861052 - time (sec): 0.89 - samples/sec: 4719.25 - lr: 0.000038 - momentum: 0.000000
124
+ 2023-10-25 21:18:21,215 epoch 4 - iter 28/146 - loss 0.05191836 - time (sec): 1.86 - samples/sec: 4877.09 - lr: 0.000038 - momentum: 0.000000
125
+ 2023-10-25 21:18:22,236 epoch 4 - iter 42/146 - loss 0.06395996 - time (sec): 2.88 - samples/sec: 4809.19 - lr: 0.000037 - momentum: 0.000000
126
+ 2023-10-25 21:18:23,076 epoch 4 - iter 56/146 - loss 0.06048888 - time (sec): 3.72 - samples/sec: 4874.77 - lr: 0.000037 - momentum: 0.000000
127
+ 2023-10-25 21:18:23,950 epoch 4 - iter 70/146 - loss 0.06286319 - time (sec): 4.60 - samples/sec: 4835.86 - lr: 0.000036 - momentum: 0.000000
128
+ 2023-10-25 21:18:24,738 epoch 4 - iter 84/146 - loss 0.06582380 - time (sec): 5.39 - samples/sec: 4807.83 - lr: 0.000036 - momentum: 0.000000
129
+ 2023-10-25 21:18:25,554 epoch 4 - iter 98/146 - loss 0.06500042 - time (sec): 6.20 - samples/sec: 4816.03 - lr: 0.000035 - momentum: 0.000000
130
+ 2023-10-25 21:18:26,501 epoch 4 - iter 112/146 - loss 0.06090963 - time (sec): 7.15 - samples/sec: 4799.24 - lr: 0.000035 - momentum: 0.000000
131
+ 2023-10-25 21:18:27,383 epoch 4 - iter 126/146 - loss 0.05983610 - time (sec): 8.03 - samples/sec: 4786.06 - lr: 0.000034 - momentum: 0.000000
132
+ 2023-10-25 21:18:28,318 epoch 4 - iter 140/146 - loss 0.05759203 - time (sec): 8.97 - samples/sec: 4790.33 - lr: 0.000034 - momentum: 0.000000
133
+ 2023-10-25 21:18:28,721 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-25 21:18:28,721 EPOCH 4 done: loss 0.0581 - lr: 0.000034
135
+ 2023-10-25 21:18:29,633 DEV : loss 0.11335166543722153 - f1-score (micro avg) 0.7289
136
+ 2023-10-25 21:18:29,637 ----------------------------------------------------------------------------------------------------
137
+ 2023-10-25 21:18:30,635 epoch 5 - iter 14/146 - loss 0.02310730 - time (sec): 1.00 - samples/sec: 4565.19 - lr: 0.000033 - momentum: 0.000000
138
+ 2023-10-25 21:18:31,564 epoch 5 - iter 28/146 - loss 0.03667454 - time (sec): 1.93 - samples/sec: 4622.32 - lr: 0.000032 - momentum: 0.000000
139
+ 2023-10-25 21:18:32,482 epoch 5 - iter 42/146 - loss 0.03684771 - time (sec): 2.84 - samples/sec: 4751.29 - lr: 0.000032 - momentum: 0.000000
140
+ 2023-10-25 21:18:33,524 epoch 5 - iter 56/146 - loss 0.03238162 - time (sec): 3.89 - samples/sec: 4664.15 - lr: 0.000031 - momentum: 0.000000
141
+ 2023-10-25 21:18:34,483 epoch 5 - iter 70/146 - loss 0.03312718 - time (sec): 4.84 - samples/sec: 4705.98 - lr: 0.000031 - momentum: 0.000000
142
+ 2023-10-25 21:18:35,374 epoch 5 - iter 84/146 - loss 0.03488909 - time (sec): 5.74 - samples/sec: 4658.50 - lr: 0.000030 - momentum: 0.000000
143
+ 2023-10-25 21:18:36,254 epoch 5 - iter 98/146 - loss 0.03511040 - time (sec): 6.62 - samples/sec: 4588.03 - lr: 0.000030 - momentum: 0.000000
144
+ 2023-10-25 21:18:37,094 epoch 5 - iter 112/146 - loss 0.03459069 - time (sec): 7.46 - samples/sec: 4634.93 - lr: 0.000029 - momentum: 0.000000
145
+ 2023-10-25 21:18:37,946 epoch 5 - iter 126/146 - loss 0.03501141 - time (sec): 8.31 - samples/sec: 4628.01 - lr: 0.000029 - momentum: 0.000000
146
+ 2023-10-25 21:18:38,791 epoch 5 - iter 140/146 - loss 0.03582364 - time (sec): 9.15 - samples/sec: 4661.80 - lr: 0.000028 - momentum: 0.000000
147
+ 2023-10-25 21:18:39,129 ----------------------------------------------------------------------------------------------------
148
+ 2023-10-25 21:18:39,129 EPOCH 5 done: loss 0.0355 - lr: 0.000028
149
+ 2023-10-25 21:18:40,050 DEV : loss 0.1298971325159073 - f1-score (micro avg) 0.7406
150
+ 2023-10-25 21:18:40,054 saving best model
151
+ 2023-10-25 21:18:40,729 ----------------------------------------------------------------------------------------------------
152
+ 2023-10-25 21:18:41,777 epoch 6 - iter 14/146 - loss 0.03101011 - time (sec): 1.04 - samples/sec: 3989.79 - lr: 0.000027 - momentum: 0.000000
153
+ 2023-10-25 21:18:42,674 epoch 6 - iter 28/146 - loss 0.02855869 - time (sec): 1.94 - samples/sec: 4176.16 - lr: 0.000027 - momentum: 0.000000
154
+ 2023-10-25 21:18:43,620 epoch 6 - iter 42/146 - loss 0.02671193 - time (sec): 2.88 - samples/sec: 4207.24 - lr: 0.000026 - momentum: 0.000000
155
+ 2023-10-25 21:18:44,432 epoch 6 - iter 56/146 - loss 0.02864686 - time (sec): 3.70 - samples/sec: 4361.20 - lr: 0.000026 - momentum: 0.000000
156
+ 2023-10-25 21:18:45,347 epoch 6 - iter 70/146 - loss 0.02700276 - time (sec): 4.61 - samples/sec: 4449.25 - lr: 0.000025 - momentum: 0.000000
157
+ 2023-10-25 21:18:46,316 epoch 6 - iter 84/146 - loss 0.02771092 - time (sec): 5.58 - samples/sec: 4453.29 - lr: 0.000025 - momentum: 0.000000
158
+ 2023-10-25 21:18:47,196 epoch 6 - iter 98/146 - loss 0.02716645 - time (sec): 6.46 - samples/sec: 4495.58 - lr: 0.000024 - momentum: 0.000000
159
+ 2023-10-25 21:18:48,272 epoch 6 - iter 112/146 - loss 0.02790357 - time (sec): 7.54 - samples/sec: 4587.95 - lr: 0.000024 - momentum: 0.000000
160
+ 2023-10-25 21:18:49,070 epoch 6 - iter 126/146 - loss 0.02763179 - time (sec): 8.33 - samples/sec: 4624.59 - lr: 0.000023 - momentum: 0.000000
161
+ 2023-10-25 21:18:50,011 epoch 6 - iter 140/146 - loss 0.02596328 - time (sec): 9.28 - samples/sec: 4628.42 - lr: 0.000023 - momentum: 0.000000
162
+ 2023-10-25 21:18:50,348 ----------------------------------------------------------------------------------------------------
163
+ 2023-10-25 21:18:50,348 EPOCH 6 done: loss 0.0260 - lr: 0.000023
164
+ 2023-10-25 21:18:51,258 DEV : loss 0.13975274562835693 - f1-score (micro avg) 0.7446
165
+ 2023-10-25 21:18:51,263 saving best model
166
+ 2023-10-25 21:18:51,943 ----------------------------------------------------------------------------------------------------
167
+ 2023-10-25 21:18:52,792 epoch 7 - iter 14/146 - loss 0.01511146 - time (sec): 0.84 - samples/sec: 4339.48 - lr: 0.000022 - momentum: 0.000000
168
+ 2023-10-25 21:18:53,708 epoch 7 - iter 28/146 - loss 0.03038031 - time (sec): 1.76 - samples/sec: 4499.40 - lr: 0.000021 - momentum: 0.000000
169
+ 2023-10-25 21:18:54,773 epoch 7 - iter 42/146 - loss 0.02451599 - time (sec): 2.82 - samples/sec: 4488.20 - lr: 0.000021 - momentum: 0.000000
170
+ 2023-10-25 21:18:55,594 epoch 7 - iter 56/146 - loss 0.02420964 - time (sec): 3.65 - samples/sec: 4517.24 - lr: 0.000020 - momentum: 0.000000
171
+ 2023-10-25 21:18:56,404 epoch 7 - iter 70/146 - loss 0.02264035 - time (sec): 4.46 - samples/sec: 4517.72 - lr: 0.000020 - momentum: 0.000000
172
+ 2023-10-25 21:18:57,429 epoch 7 - iter 84/146 - loss 0.01956782 - time (sec): 5.48 - samples/sec: 4525.12 - lr: 0.000019 - momentum: 0.000000
173
+ 2023-10-25 21:18:58,408 epoch 7 - iter 98/146 - loss 0.01927721 - time (sec): 6.46 - samples/sec: 4651.79 - lr: 0.000019 - momentum: 0.000000
174
+ 2023-10-25 21:18:59,240 epoch 7 - iter 112/146 - loss 0.01849363 - time (sec): 7.29 - samples/sec: 4668.93 - lr: 0.000018 - momentum: 0.000000
175
+ 2023-10-25 21:19:00,152 epoch 7 - iter 126/146 - loss 0.01922696 - time (sec): 8.20 - samples/sec: 4654.81 - lr: 0.000018 - momentum: 0.000000
176
+ 2023-10-25 21:19:01,102 epoch 7 - iter 140/146 - loss 0.01984644 - time (sec): 9.15 - samples/sec: 4657.12 - lr: 0.000017 - momentum: 0.000000
177
+ 2023-10-25 21:19:01,430 ----------------------------------------------------------------------------------------------------
178
+ 2023-10-25 21:19:01,431 EPOCH 7 done: loss 0.0197 - lr: 0.000017
179
+ 2023-10-25 21:19:02,383 DEV : loss 0.1919669657945633 - f1-score (micro avg) 0.7032
180
+ 2023-10-25 21:19:02,388 ----------------------------------------------------------------------------------------------------
181
+ 2023-10-25 21:19:03,290 epoch 8 - iter 14/146 - loss 0.00645562 - time (sec): 0.90 - samples/sec: 4461.30 - lr: 0.000016 - momentum: 0.000000
182
+ 2023-10-25 21:19:04,288 epoch 8 - iter 28/146 - loss 0.01024029 - time (sec): 1.90 - samples/sec: 4851.43 - lr: 0.000016 - momentum: 0.000000
183
+ 2023-10-25 21:19:05,111 epoch 8 - iter 42/146 - loss 0.01291880 - time (sec): 2.72 - samples/sec: 4749.49 - lr: 0.000015 - momentum: 0.000000
184
+ 2023-10-25 21:19:05,943 epoch 8 - iter 56/146 - loss 0.01125943 - time (sec): 3.55 - samples/sec: 4866.42 - lr: 0.000015 - momentum: 0.000000
185
+ 2023-10-25 21:19:06,777 epoch 8 - iter 70/146 - loss 0.01175429 - time (sec): 4.39 - samples/sec: 4870.43 - lr: 0.000014 - momentum: 0.000000
186
+ 2023-10-25 21:19:07,788 epoch 8 - iter 84/146 - loss 0.01352109 - time (sec): 5.40 - samples/sec: 4832.43 - lr: 0.000014 - momentum: 0.000000
187
+ 2023-10-25 21:19:08,711 epoch 8 - iter 98/146 - loss 0.01341075 - time (sec): 6.32 - samples/sec: 4766.57 - lr: 0.000013 - momentum: 0.000000
188
+ 2023-10-25 21:19:09,567 epoch 8 - iter 112/146 - loss 0.01412573 - time (sec): 7.18 - samples/sec: 4730.64 - lr: 0.000013 - momentum: 0.000000
189
+ 2023-10-25 21:19:10,473 epoch 8 - iter 126/146 - loss 0.01398333 - time (sec): 8.08 - samples/sec: 4746.89 - lr: 0.000012 - momentum: 0.000000
190
+ 2023-10-25 21:19:11,543 epoch 8 - iter 140/146 - loss 0.01416295 - time (sec): 9.15 - samples/sec: 4722.43 - lr: 0.000012 - momentum: 0.000000
191
+ 2023-10-25 21:19:11,859 ----------------------------------------------------------------------------------------------------
192
+ 2023-10-25 21:19:11,859 EPOCH 8 done: loss 0.0141 - lr: 0.000012
193
+ 2023-10-25 21:19:12,769 DEV : loss 0.1696723997592926 - f1-score (micro avg) 0.7516
194
+ 2023-10-25 21:19:12,773 saving best model
195
+ 2023-10-25 21:19:13,447 ----------------------------------------------------------------------------------------------------
196
+ 2023-10-25 21:19:14,627 epoch 9 - iter 14/146 - loss 0.00262122 - time (sec): 1.18 - samples/sec: 3820.50 - lr: 0.000011 - momentum: 0.000000
197
+ 2023-10-25 21:19:15,556 epoch 9 - iter 28/146 - loss 0.00845440 - time (sec): 2.11 - samples/sec: 4034.86 - lr: 0.000010 - momentum: 0.000000
198
+ 2023-10-25 21:19:16,437 epoch 9 - iter 42/146 - loss 0.00754201 - time (sec): 2.99 - samples/sec: 4232.15 - lr: 0.000010 - momentum: 0.000000
199
+ 2023-10-25 21:19:17,421 epoch 9 - iter 56/146 - loss 0.00744900 - time (sec): 3.97 - samples/sec: 4383.29 - lr: 0.000009 - momentum: 0.000000
200
+ 2023-10-25 21:19:18,283 epoch 9 - iter 70/146 - loss 0.00890328 - time (sec): 4.83 - samples/sec: 4481.93 - lr: 0.000009 - momentum: 0.000000
201
+ 2023-10-25 21:19:19,192 epoch 9 - iter 84/146 - loss 0.00998982 - time (sec): 5.74 - samples/sec: 4490.71 - lr: 0.000008 - momentum: 0.000000
202
+ 2023-10-25 21:19:20,017 epoch 9 - iter 98/146 - loss 0.00904694 - time (sec): 6.57 - samples/sec: 4555.33 - lr: 0.000008 - momentum: 0.000000
203
+ 2023-10-25 21:19:20,783 epoch 9 - iter 112/146 - loss 0.00888017 - time (sec): 7.33 - samples/sec: 4505.62 - lr: 0.000007 - momentum: 0.000000
204
+ 2023-10-25 21:19:21,773 epoch 9 - iter 126/146 - loss 0.00832595 - time (sec): 8.32 - samples/sec: 4546.73 - lr: 0.000007 - momentum: 0.000000
205
+ 2023-10-25 21:19:22,838 epoch 9 - iter 140/146 - loss 0.00937880 - time (sec): 9.39 - samples/sec: 4534.27 - lr: 0.000006 - momentum: 0.000000
206
+ 2023-10-25 21:19:23,210 ----------------------------------------------------------------------------------------------------
207
+ 2023-10-25 21:19:23,210 EPOCH 9 done: loss 0.0091 - lr: 0.000006
208
+ 2023-10-25 21:19:24,121 DEV : loss 0.19342197477817535 - f1-score (micro avg) 0.7395
209
+ 2023-10-25 21:19:24,126 ----------------------------------------------------------------------------------------------------
210
+ 2023-10-25 21:19:25,053 epoch 10 - iter 14/146 - loss 0.00885036 - time (sec): 0.93 - samples/sec: 4744.47 - lr: 0.000005 - momentum: 0.000000
211
+ 2023-10-25 21:19:25,916 epoch 10 - iter 28/146 - loss 0.00830142 - time (sec): 1.79 - samples/sec: 4598.25 - lr: 0.000005 - momentum: 0.000000
212
+ 2023-10-25 21:19:26,734 epoch 10 - iter 42/146 - loss 0.00659488 - time (sec): 2.61 - samples/sec: 4611.18 - lr: 0.000004 - momentum: 0.000000
213
+ 2023-10-25 21:19:27,676 epoch 10 - iter 56/146 - loss 0.00754351 - time (sec): 3.55 - samples/sec: 4657.30 - lr: 0.000004 - momentum: 0.000000
214
+ 2023-10-25 21:19:28,550 epoch 10 - iter 70/146 - loss 0.00629635 - time (sec): 4.42 - samples/sec: 4767.11 - lr: 0.000003 - momentum: 0.000000
215
+ 2023-10-25 21:19:29,452 epoch 10 - iter 84/146 - loss 0.00537279 - time (sec): 5.33 - samples/sec: 4796.56 - lr: 0.000003 - momentum: 0.000000
216
+ 2023-10-25 21:19:30,589 epoch 10 - iter 98/146 - loss 0.00525309 - time (sec): 6.46 - samples/sec: 4767.88 - lr: 0.000002 - momentum: 0.000000
217
+ 2023-10-25 21:19:31,437 epoch 10 - iter 112/146 - loss 0.00525225 - time (sec): 7.31 - samples/sec: 4708.13 - lr: 0.000002 - momentum: 0.000000
218
+ 2023-10-25 21:19:32,326 epoch 10 - iter 126/146 - loss 0.00605529 - time (sec): 8.20 - samples/sec: 4673.20 - lr: 0.000001 - momentum: 0.000000
219
+ 2023-10-25 21:19:33,253 epoch 10 - iter 140/146 - loss 0.00601103 - time (sec): 9.13 - samples/sec: 4665.94 - lr: 0.000000 - momentum: 0.000000
220
+ 2023-10-25 21:19:33,591 ----------------------------------------------------------------------------------------------------
221
+ 2023-10-25 21:19:33,591 EPOCH 10 done: loss 0.0058 - lr: 0.000000
222
+ 2023-10-25 21:19:34,501 DEV : loss 0.19659662246704102 - f1-score (micro avg) 0.742
223
+ 2023-10-25 21:19:34,900 ----------------------------------------------------------------------------------------------------
224
+ 2023-10-25 21:19:34,901 Loading model from best epoch ...
225
+ 2023-10-25 21:19:36,575 SequenceTagger predicts: Dictionary with 17 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-PER, B-PER, E-PER, I-PER, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
226
+ 2023-10-25 21:19:38,113
227
+ Results:
228
+ - F-score (micro) 0.7569
229
+ - F-score (macro) 0.6826
230
+ - Accuracy 0.6335
231
+
232
+ By class:
233
+ precision recall f1-score support
234
+
235
+ PER 0.7879 0.8218 0.8045 348
236
+ LOC 0.6769 0.8429 0.7509 261
237
+ ORG 0.5102 0.4808 0.4950 52
238
+ HumanProd 0.6071 0.7727 0.6800 22
239
+
240
+ micro avg 0.7163 0.8023 0.7569 683
241
+ macro avg 0.6455 0.7296 0.6826 683
242
+ weighted avg 0.7185 0.8023 0.7564 683
243
+
244
+ 2023-10-25 21:19:38,113 ----------------------------------------------------------------------------------------------------