Hugging Face's logo

stefan-it
/
hmbench-newseye-fi-hmbert_64k-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-1

Token Classification
Flair
PyTorch
TensorBoard
Finnish
sequence-tagger-model
Model card Files Metrics Community
stefan-it commited on
Commit
3db2b3f
•
1 Parent(s): 64fb5cc

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 20:49:55,434 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-25 20:49:55,434 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 20:49:55,435 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-25 20:49:55,435 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 20:49:55,435 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-25 20:49:55,435 Train: 1166 sentences
55
+ 2023-10-25 20:49:55,435 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-25 20:49:55,435 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-25 20:49:55,435 Training Params:
58
+ 2023-10-25 20:49:55,435 - learning_rate: "3e-05"
59
+ 2023-10-25 20:49:55,435 - mini_batch_size: "4"
60
+ 2023-10-25 20:49:55,435 - max_epochs: "10"
61
+ 2023-10-25 20:49:55,435 - shuffle: "True"
62
+ 2023-10-25 20:49:55,435 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-25 20:49:55,435 Plugins:
64
+ 2023-10-25 20:49:55,435 - TensorboardLogger
65
+ 2023-10-25 20:49:55,435 - LinearScheduler | warmup_fraction: '0.1'
66
+ 2023-10-25 20:49:55,435 ----------------------------------------------------------------------------------------------------
67
+ 2023-10-25 20:49:55,435 Final evaluation on model from best epoch (best-model.pt)
68
+ 2023-10-25 20:49:55,435 - metric: "('micro avg', 'f1-score')"
69
+ 2023-10-25 20:49:55,435 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-25 20:49:55,435 Computation:
71
+ 2023-10-25 20:49:55,435 - compute on device: cuda:0
72
+ 2023-10-25 20:49:55,435 - embedding storage: none
73
+ 2023-10-25 20:49:55,435 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-25 20:49:55,435 Model training base path: "hmbench-newseye/fi-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-1"
75
+ 2023-10-25 20:49:55,435 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-25 20:49:55,435 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-25 20:49:55,436 Logging anything other than scalars to TensorBoard is currently not supported.
78
+ 2023-10-25 20:49:56,721 epoch 1 - iter 29/292 - loss 3.39043164 - time (sec): 1.28 - samples/sec: 3706.84 - lr: 0.000003 - momentum: 0.000000
79
+ 2023-10-25 20:49:57,994 epoch 1 - iter 58/292 - loss 2.75211348 - time (sec): 2.56 - samples/sec: 3400.18 - lr: 0.000006 - momentum: 0.000000
80
+ 2023-10-25 20:49:59,292 epoch 1 - iter 87/292 - loss 2.03241543 - time (sec): 3.86 - samples/sec: 3423.52 - lr: 0.000009 - momentum: 0.000000
81
+ 2023-10-25 20:50:00,587 epoch 1 - iter 116/292 - loss 1.65465457 - time (sec): 5.15 - samples/sec: 3448.70 - lr: 0.000012 - momentum: 0.000000
82
+ 2023-10-25 20:50:01,918 epoch 1 - iter 145/292 - loss 1.40479820 - time (sec): 6.48 - samples/sec: 3481.90 - lr: 0.000015 - momentum: 0.000000
83
+ 2023-10-25 20:50:03,215 epoch 1 - iter 174/292 - loss 1.22215741 - time (sec): 7.78 - samples/sec: 3521.21 - lr: 0.000018 - momentum: 0.000000
84
+ 2023-10-25 20:50:04,516 epoch 1 - iter 203/292 - loss 1.09287149 - time (sec): 9.08 - samples/sec: 3492.27 - lr: 0.000021 - momentum: 0.000000
85
+ 2023-10-25 20:50:05,787 epoch 1 - iter 232/292 - loss 0.99311006 - time (sec): 10.35 - samples/sec: 3465.97 - lr: 0.000024 - momentum: 0.000000
86
+ 2023-10-25 20:50:07,109 epoch 1 - iter 261/292 - loss 0.90839928 - time (sec): 11.67 - samples/sec: 3451.01 - lr: 0.000027 - momentum: 0.000000
87
+ 2023-10-25 20:50:08,352 epoch 1 - iter 290/292 - loss 0.84739090 - time (sec): 12.92 - samples/sec: 3425.59 - lr: 0.000030 - momentum: 0.000000
88
+ 2023-10-25 20:50:08,431 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-25 20:50:08,431 EPOCH 1 done: loss 0.8460 - lr: 0.000030
90
+ 2023-10-25 20:50:08,940 DEV : loss 0.17541182041168213 - f1-score (micro avg) 0.5276
91
+ 2023-10-25 20:50:08,944 saving best model
92
+ 2023-10-25 20:50:09,418 ----------------------------------------------------------------------------------------------------
93
+ 2023-10-25 20:50:10,662 epoch 2 - iter 29/292 - loss 0.18632175 - time (sec): 1.24 - samples/sec: 3417.03 - lr: 0.000030 - momentum: 0.000000
94
+ 2023-10-25 20:50:11,914 epoch 2 - iter 58/292 - loss 0.19639063 - time (sec): 2.49 - samples/sec: 3432.14 - lr: 0.000029 - momentum: 0.000000
95
+ 2023-10-25 20:50:13,238 epoch 2 - iter 87/292 - loss 0.19518792 - time (sec): 3.82 - samples/sec: 3532.20 - lr: 0.000029 - momentum: 0.000000
96
+ 2023-10-25 20:50:14,521 epoch 2 - iter 116/292 - loss 0.18816835 - time (sec): 5.10 - samples/sec: 3473.17 - lr: 0.000029 - momentum: 0.000000
97
+ 2023-10-25 20:50:15,787 epoch 2 - iter 145/292 - loss 0.18505163 - time (sec): 6.37 - samples/sec: 3337.12 - lr: 0.000028 - momentum: 0.000000
98
+ 2023-10-25 20:50:17,089 epoch 2 - iter 174/292 - loss 0.17260294 - time (sec): 7.67 - samples/sec: 3408.94 - lr: 0.000028 - momentum: 0.000000
99
+ 2023-10-25 20:50:18,400 epoch 2 - iter 203/292 - loss 0.16289497 - time (sec): 8.98 - samples/sec: 3520.34 - lr: 0.000028 - momentum: 0.000000
100
+ 2023-10-25 20:50:19,622 epoch 2 - iter 232/292 - loss 0.15958523 - time (sec): 10.20 - samples/sec: 3494.66 - lr: 0.000027 - momentum: 0.000000
101
+ 2023-10-25 20:50:20,816 epoch 2 - iter 261/292 - loss 0.16168193 - time (sec): 11.40 - samples/sec: 3445.66 - lr: 0.000027 - momentum: 0.000000
102
+ 2023-10-25 20:50:22,195 epoch 2 - iter 290/292 - loss 0.16115890 - time (sec): 12.78 - samples/sec: 3460.71 - lr: 0.000027 - momentum: 0.000000
103
+ 2023-10-25 20:50:22,276 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-25 20:50:22,276 EPOCH 2 done: loss 0.1610 - lr: 0.000027
105
+ 2023-10-25 20:50:23,351 DEV : loss 0.09794748574495316 - f1-score (micro avg) 0.7061
106
+ 2023-10-25 20:50:23,355 saving best model
107
+ 2023-10-25 20:50:23,977 ----------------------------------------------------------------------------------------------------
108
+ 2023-10-25 20:50:25,325 epoch 3 - iter 29/292 - loss 0.08203820 - time (sec): 1.35 - samples/sec: 3769.10 - lr: 0.000026 - momentum: 0.000000
109
+ 2023-10-25 20:50:26,684 epoch 3 - iter 58/292 - loss 0.07682316 - time (sec): 2.70 - samples/sec: 3779.79 - lr: 0.000026 - momentum: 0.000000
110
+ 2023-10-25 20:50:27,948 epoch 3 - iter 87/292 - loss 0.08420685 - time (sec): 3.97 - samples/sec: 3502.65 - lr: 0.000026 - momentum: 0.000000
111
+ 2023-10-25 20:50:29,252 epoch 3 - iter 116/292 - loss 0.08287543 - time (sec): 5.27 - samples/sec: 3436.51 - lr: 0.000025 - momentum: 0.000000
112
+ 2023-10-25 20:50:30,559 epoch 3 - iter 145/292 - loss 0.09196539 - time (sec): 6.58 - samples/sec: 3446.27 - lr: 0.000025 - momentum: 0.000000
113
+ 2023-10-25 20:50:31,793 epoch 3 - iter 174/292 - loss 0.08691728 - time (sec): 7.81 - samples/sec: 3325.74 - lr: 0.000025 - momentum: 0.000000
114
+ 2023-10-25 20:50:33,140 epoch 3 - iter 203/292 - loss 0.08593496 - time (sec): 9.16 - samples/sec: 3392.81 - lr: 0.000024 - momentum: 0.000000
115
+ 2023-10-25 20:50:34,459 epoch 3 - iter 232/292 - loss 0.08704569 - time (sec): 10.48 - samples/sec: 3379.82 - lr: 0.000024 - momentum: 0.000000
116
+ 2023-10-25 20:50:35,769 epoch 3 - iter 261/292 - loss 0.08983051 - time (sec): 11.79 - samples/sec: 3401.24 - lr: 0.000024 - momentum: 0.000000
117
+ 2023-10-25 20:50:37,036 epoch 3 - iter 290/292 - loss 0.08782705 - time (sec): 13.06 - samples/sec: 3396.89 - lr: 0.000023 - momentum: 0.000000
118
+ 2023-10-25 20:50:37,113 ----------------------------------------------------------------------------------------------------
119
+ 2023-10-25 20:50:37,113 EPOCH 3 done: loss 0.0878 - lr: 0.000023
120
+ 2023-10-25 20:50:38,028 DEV : loss 0.12430848926305771 - f1-score (micro avg) 0.6871
121
+ 2023-10-25 20:50:38,032 ----------------------------------------------------------------------------------------------------
122
+ 2023-10-25 20:50:39,272 epoch 4 - iter 29/292 - loss 0.03595201 - time (sec): 1.24 - samples/sec: 2975.54 - lr: 0.000023 - momentum: 0.000000
123
+ 2023-10-25 20:50:40,532 epoch 4 - iter 58/292 - loss 0.03957711 - time (sec): 2.50 - samples/sec: 3106.93 - lr: 0.000023 - momentum: 0.000000
124
+ 2023-10-25 20:50:41,907 epoch 4 - iter 87/292 - loss 0.04087163 - time (sec): 3.87 - samples/sec: 3227.91 - lr: 0.000022 - momentum: 0.000000
125
+ 2023-10-25 20:50:43,212 epoch 4 - iter 116/292 - loss 0.04261847 - time (sec): 5.18 - samples/sec: 3324.13 - lr: 0.000022 - momentum: 0.000000
126
+ 2023-10-25 20:50:44,576 epoch 4 - iter 145/292 - loss 0.04049488 - time (sec): 6.54 - samples/sec: 3511.55 - lr: 0.000022 - momentum: 0.000000
127
+ 2023-10-25 20:50:45,851 epoch 4 - iter 174/292 - loss 0.04321196 - time (sec): 7.82 - samples/sec: 3405.50 - lr: 0.000021 - momentum: 0.000000
128
+ 2023-10-25 20:50:47,245 epoch 4 - iter 203/292 - loss 0.04658455 - time (sec): 9.21 - samples/sec: 3423.75 - lr: 0.000021 - momentum: 0.000000
129
+ 2023-10-25 20:50:48,489 epoch 4 - iter 232/292 - loss 0.05107851 - time (sec): 10.46 - samples/sec: 3402.58 - lr: 0.000021 - momentum: 0.000000
130
+ 2023-10-25 20:50:49,760 epoch 4 - iter 261/292 - loss 0.05070514 - time (sec): 11.73 - samples/sec: 3372.72 - lr: 0.000020 - momentum: 0.000000
131
+ 2023-10-25 20:50:51,027 epoch 4 - iter 290/292 - loss 0.05233506 - time (sec): 12.99 - samples/sec: 3405.20 - lr: 0.000020 - momentum: 0.000000
132
+ 2023-10-25 20:50:51,109 ----------------------------------------------------------------------------------------------------
133
+ 2023-10-25 20:50:51,109 EPOCH 4 done: loss 0.0529 - lr: 0.000020
134
+ 2023-10-25 20:50:52,018 DEV : loss 0.10681485384702682 - f1-score (micro avg) 0.7445
135
+ 2023-10-25 20:50:52,022 saving best model
136
+ 2023-10-25 20:50:52,627 ----------------------------------------------------------------------------------------------------
137
+ 2023-10-25 20:50:53,899 epoch 5 - iter 29/292 - loss 0.03458033 - time (sec): 1.27 - samples/sec: 3108.24 - lr: 0.000020 - momentum: 0.000000
138
+ 2023-10-25 20:50:55,212 epoch 5 - iter 58/292 - loss 0.02994252 - time (sec): 2.58 - samples/sec: 3329.96 - lr: 0.000019 - momentum: 0.000000
139
+ 2023-10-25 20:50:56,544 epoch 5 - iter 87/292 - loss 0.04035193 - time (sec): 3.92 - samples/sec: 3380.24 - lr: 0.000019 - momentum: 0.000000
140
+ 2023-10-25 20:50:57,849 epoch 5 - iter 116/292 - loss 0.03752758 - time (sec): 5.22 - samples/sec: 3384.76 - lr: 0.000019 - momentum: 0.000000
141
+ 2023-10-25 20:50:59,099 epoch 5 - iter 145/292 - loss 0.03514150 - time (sec): 6.47 - samples/sec: 3385.83 - lr: 0.000018 - momentum: 0.000000
142
+ 2023-10-25 20:51:00,425 epoch 5 - iter 174/292 - loss 0.03515028 - time (sec): 7.80 - samples/sec: 3397.01 - lr: 0.000018 - momentum: 0.000000
143
+ 2023-10-25 20:51:01,703 epoch 5 - iter 203/292 - loss 0.03610470 - time (sec): 9.07 - samples/sec: 3396.17 - lr: 0.000018 - momentum: 0.000000
144
+ 2023-10-25 20:51:03,021 epoch 5 - iter 232/292 - loss 0.03513941 - time (sec): 10.39 - samples/sec: 3412.76 - lr: 0.000017 - momentum: 0.000000
145
+ 2023-10-25 20:51:04,335 epoch 5 - iter 261/292 - loss 0.03355999 - time (sec): 11.71 - samples/sec: 3368.72 - lr: 0.000017 - momentum: 0.000000
146
+ 2023-10-25 20:51:05,650 epoch 5 - iter 290/292 - loss 0.03482035 - time (sec): 13.02 - samples/sec: 3399.33 - lr: 0.000017 - momentum: 0.000000
147
+ 2023-10-25 20:51:05,732 ----------------------------------------------------------------------------------------------------
148
+ 2023-10-25 20:51:05,732 EPOCH 5 done: loss 0.0347 - lr: 0.000017
149
+ 2023-10-25 20:51:06,653 DEV : loss 0.1422182321548462 - f1-score (micro avg) 0.7585
150
+ 2023-10-25 20:51:06,657 saving best model
151
+ 2023-10-25 20:51:07,266 ----------------------------------------------------------------------------------------------------
152
+ 2023-10-25 20:51:08,561 epoch 6 - iter 29/292 - loss 0.02969331 - time (sec): 1.29 - samples/sec: 3244.31 - lr: 0.000016 - momentum: 0.000000
153
+ 2023-10-25 20:51:09,836 epoch 6 - iter 58/292 - loss 0.01985153 - time (sec): 2.57 - samples/sec: 3337.81 - lr: 0.000016 - momentum: 0.000000
154
+ 2023-10-25 20:51:11,291 epoch 6 - iter 87/292 - loss 0.01900497 - time (sec): 4.02 - samples/sec: 3468.04 - lr: 0.000016 - momentum: 0.000000
155
+ 2023-10-25 20:51:12,604 epoch 6 - iter 116/292 - loss 0.02209573 - time (sec): 5.34 - samples/sec: 3480.15 - lr: 0.000015 - momentum: 0.000000
156
+ 2023-10-25 20:51:13,984 epoch 6 - iter 145/292 - loss 0.02181314 - time (sec): 6.72 - samples/sec: 3326.83 - lr: 0.000015 - momentum: 0.000000
157
+ 2023-10-25 20:51:15,286 epoch 6 - iter 174/292 - loss 0.02134735 - time (sec): 8.02 - samples/sec: 3352.70 - lr: 0.000015 - momentum: 0.000000
158
+ 2023-10-25 20:51:16,565 epoch 6 - iter 203/292 - loss 0.02078533 - time (sec): 9.30 - samples/sec: 3371.65 - lr: 0.000014 - momentum: 0.000000
159
+ 2023-10-25 20:51:17,855 epoch 6 - iter 232/292 - loss 0.02097319 - time (sec): 10.59 - samples/sec: 3391.07 - lr: 0.000014 - momentum: 0.000000
160
+ 2023-10-25 20:51:19,118 epoch 6 - iter 261/292 - loss 0.02180842 - time (sec): 11.85 - samples/sec: 3359.49 - lr: 0.000014 - momentum: 0.000000
161
+ 2023-10-25 20:51:20,424 epoch 6 - iter 290/292 - loss 0.02475954 - time (sec): 13.16 - samples/sec: 3366.18 - lr: 0.000013 - momentum: 0.000000
162
+ 2023-10-25 20:51:20,514 ----------------------------------------------------------------------------------------------------
163
+ 2023-10-25 20:51:20,515 EPOCH 6 done: loss 0.0247 - lr: 0.000013
164
+ 2023-10-25 20:51:21,576 DEV : loss 0.16126354038715363 - f1-score (micro avg) 0.7652
165
+ 2023-10-25 20:51:21,580 saving best model
166
+ 2023-10-25 20:51:22,185 ----------------------------------------------------------------------------------------------------
167
+ 2023-10-25 20:51:23,507 epoch 7 - iter 29/292 - loss 0.01483357 - time (sec): 1.32 - samples/sec: 3168.20 - lr: 0.000013 - momentum: 0.000000
168
+ 2023-10-25 20:51:24,768 epoch 7 - iter 58/292 - loss 0.01300097 - time (sec): 2.58 - samples/sec: 3239.79 - lr: 0.000013 - momentum: 0.000000
169
+ 2023-10-25 20:51:26,062 epoch 7 - iter 87/292 - loss 0.01498788 - time (sec): 3.88 - samples/sec: 3265.72 - lr: 0.000012 - momentum: 0.000000
170
+ 2023-10-25 20:51:27,306 epoch 7 - iter 116/292 - loss 0.01630833 - time (sec): 5.12 - samples/sec: 3270.30 - lr: 0.000012 - momentum: 0.000000
171
+ 2023-10-25 20:51:28,706 epoch 7 - iter 145/292 - loss 0.01682322 - time (sec): 6.52 - samples/sec: 3374.65 - lr: 0.000012 - momentum: 0.000000
172
+ 2023-10-25 20:51:30,087 epoch 7 - iter 174/292 - loss 0.01645327 - time (sec): 7.90 - samples/sec: 3390.80 - lr: 0.000011 - momentum: 0.000000
173
+ 2023-10-25 20:51:31,501 epoch 7 - iter 203/292 - loss 0.01898085 - time (sec): 9.32 - samples/sec: 3370.72 - lr: 0.000011 - momentum: 0.000000
174
+ 2023-10-25 20:51:32,860 epoch 7 - iter 232/292 - loss 0.01845577 - time (sec): 10.67 - samples/sec: 3335.24 - lr: 0.000011 - momentum: 0.000000
175
+ 2023-10-25 20:51:34,225 epoch 7 - iter 261/292 - loss 0.01880091 - time (sec): 12.04 - samples/sec: 3319.52 - lr: 0.000010 - momentum: 0.000000
176
+ 2023-10-25 20:51:35,514 epoch 7 - iter 290/292 - loss 0.01778838 - time (sec): 13.33 - samples/sec: 3313.71 - lr: 0.000010 - momentum: 0.000000
177
+ 2023-10-25 20:51:35,601 ----------------------------------------------------------------------------------------------------
178
+ 2023-10-25 20:51:35,602 EPOCH 7 done: loss 0.0177 - lr: 0.000010
179
+ 2023-10-25 20:51:36,522 DEV : loss 0.16213594377040863 - f1-score (micro avg) 0.7632
180
+ 2023-10-25 20:51:36,527 ----------------------------------------------------------------------------------------------------
181
+ 2023-10-25 20:51:37,830 epoch 8 - iter 29/292 - loss 0.00618455 - time (sec): 1.30 - samples/sec: 3400.26 - lr: 0.000010 - momentum: 0.000000
182
+ 2023-10-25 20:51:39,173 epoch 8 - iter 58/292 - loss 0.01018591 - time (sec): 2.65 - samples/sec: 3765.66 - lr: 0.000009 - momentum: 0.000000
183
+ 2023-10-25 20:51:40,427 epoch 8 - iter 87/292 - loss 0.01208414 - time (sec): 3.90 - samples/sec: 3546.09 - lr: 0.000009 - momentum: 0.000000
184
+ 2023-10-25 20:51:41,708 epoch 8 - iter 116/292 - loss 0.01178172 - time (sec): 5.18 - samples/sec: 3512.34 - lr: 0.000009 - momentum: 0.000000
185
+ 2023-10-25 20:51:42,987 epoch 8 - iter 145/292 - loss 0.01121372 - time (sec): 6.46 - samples/sec: 3484.78 - lr: 0.000008 - momentum: 0.000000
186
+ 2023-10-25 20:51:44,330 epoch 8 - iter 174/292 - loss 0.01231181 - time (sec): 7.80 - samples/sec: 3475.09 - lr: 0.000008 - momentum: 0.000000
187
+ 2023-10-25 20:51:45,677 epoch 8 - iter 203/292 - loss 0.01430244 - time (sec): 9.15 - samples/sec: 3486.84 - lr: 0.000008 - momentum: 0.000000
188
+ 2023-10-25 20:51:47,011 epoch 8 - iter 232/292 - loss 0.01575816 - time (sec): 10.48 - samples/sec: 3458.53 - lr: 0.000007 - momentum: 0.000000
189
+ 2023-10-25 20:51:48,271 epoch 8 - iter 261/292 - loss 0.01464034 - time (sec): 11.74 - samples/sec: 3409.17 - lr: 0.000007 - momentum: 0.000000
190
+ 2023-10-25 20:51:49,562 epoch 8 - iter 290/292 - loss 0.01437000 - time (sec): 13.03 - samples/sec: 3388.14 - lr: 0.000007 - momentum: 0.000000
191
+ 2023-10-25 20:51:49,647 ----------------------------------------------------------------------------------------------------
192
+ 2023-10-25 20:51:49,647 EPOCH 8 done: loss 0.0143 - lr: 0.000007
193
+ 2023-10-25 20:51:50,562 DEV : loss 0.16556185483932495 - f1-score (micro avg) 0.7716
194
+ 2023-10-25 20:51:50,567 saving best model
195
+ 2023-10-25 20:51:51,186 ----------------------------------------------------------------------------------------------------
196
+ 2023-10-25 20:51:52,526 epoch 9 - iter 29/292 - loss 0.01512189 - time (sec): 1.34 - samples/sec: 3442.17 - lr: 0.000006 - momentum: 0.000000
197
+ 2023-10-25 20:51:53,782 epoch 9 - iter 58/292 - loss 0.00904168 - time (sec): 2.59 - samples/sec: 3320.38 - lr: 0.000006 - momentum: 0.000000
198
+ 2023-10-25 20:51:55,011 epoch 9 - iter 87/292 - loss 0.00954412 - time (sec): 3.82 - samples/sec: 3176.88 - lr: 0.000006 - momentum: 0.000000
199
+ 2023-10-25 20:51:56,327 epoch 9 - iter 116/292 - loss 0.00857609 - time (sec): 5.14 - samples/sec: 3297.91 - lr: 0.000005 - momentum: 0.000000
200
+ 2023-10-25 20:51:57,569 epoch 9 - iter 145/292 - loss 0.00870679 - time (sec): 6.38 - samples/sec: 3273.04 - lr: 0.000005 - momentum: 0.000000
201
+ 2023-10-25 20:51:58,850 epoch 9 - iter 174/292 - loss 0.00944735 - time (sec): 7.66 - samples/sec: 3267.62 - lr: 0.000005 - momentum: 0.000000
202
+ 2023-10-25 20:52:00,165 epoch 9 - iter 203/292 - loss 0.00895757 - time (sec): 8.98 - samples/sec: 3343.36 - lr: 0.000004 - momentum: 0.000000
203
+ 2023-10-25 20:52:01,442 epoch 9 - iter 232/292 - loss 0.00789994 - time (sec): 10.25 - samples/sec: 3365.25 - lr: 0.000004 - momentum: 0.000000
204
+ 2023-10-25 20:52:02,796 epoch 9 - iter 261/292 - loss 0.00780787 - time (sec): 11.61 - samples/sec: 3440.83 - lr: 0.000004 - momentum: 0.000000
205
+ 2023-10-25 20:52:04,106 epoch 9 - iter 290/292 - loss 0.00830024 - time (sec): 12.92 - samples/sec: 3422.83 - lr: 0.000003 - momentum: 0.000000
206
+ 2023-10-25 20:52:04,189 ----------------------------------------------------------------------------------------------------
207
+ 2023-10-25 20:52:04,189 EPOCH 9 done: loss 0.0083 - lr: 0.000003
208
+ 2023-10-25 20:52:05,105 DEV : loss 0.17480386793613434 - f1-score (micro avg) 0.7709
209
+ 2023-10-25 20:52:05,110 ----------------------------------------------------------------------------------------------------
210
+ 2023-10-25 20:52:06,344 epoch 10 - iter 29/292 - loss 0.00779031 - time (sec): 1.23 - samples/sec: 3541.08 - lr: 0.000003 - momentum: 0.000000
211
+ 2023-10-25 20:52:07,662 epoch 10 - iter 58/292 - loss 0.01153876 - time (sec): 2.55 - samples/sec: 3328.64 - lr: 0.000003 - momentum: 0.000000
212
+ 2023-10-25 20:52:08,928 epoch 10 - iter 87/292 - loss 0.00889995 - time (sec): 3.82 - samples/sec: 3422.10 - lr: 0.000002 - momentum: 0.000000
213
+ 2023-10-25 20:52:10,215 epoch 10 - iter 116/292 - loss 0.00719468 - time (sec): 5.10 - samples/sec: 3393.53 - lr: 0.000002 - momentum: 0.000000
214
+ 2023-10-25 20:52:11,523 epoch 10 - iter 145/292 - loss 0.00702750 - time (sec): 6.41 - samples/sec: 3391.15 - lr: 0.000002 - momentum: 0.000000
215
+ 2023-10-25 20:52:12,870 epoch 10 - iter 174/292 - loss 0.00688014 - time (sec): 7.76 - samples/sec: 3317.01 - lr: 0.000001 - momentum: 0.000000
216
+ 2023-10-25 20:52:14,180 epoch 10 - iter 203/292 - loss 0.00644392 - time (sec): 9.07 - samples/sec: 3404.29 - lr: 0.000001 - momentum: 0.000000
217
+ 2023-10-25 20:52:15,503 epoch 10 - iter 232/292 - loss 0.00741474 - time (sec): 10.39 - samples/sec: 3428.84 - lr: 0.000001 - momentum: 0.000000
218
+ 2023-10-25 20:52:16,780 epoch 10 - iter 261/292 - loss 0.00729106 - time (sec): 11.67 - samples/sec: 3395.49 - lr: 0.000000 - momentum: 0.000000
219
+ 2023-10-25 20:52:18,049 epoch 10 - iter 290/292 - loss 0.00662543 - time (sec): 12.94 - samples/sec: 3409.87 - lr: 0.000000 - momentum: 0.000000
220
+ 2023-10-25 20:52:18,137 ----------------------------------------------------------------------------------------------------
221
+ 2023-10-25 20:52:18,137 EPOCH 10 done: loss 0.0066 - lr: 0.000000
222
+ 2023-10-25 20:52:19,049 DEV : loss 0.17905840277671814 - f1-score (micro avg) 0.7643
223
+ 2023-10-25 20:52:19,529 ----------------------------------------------------------------------------------------------------
224
+ 2023-10-25 20:52:19,530 Loading model from best epoch ...
225
+ 2023-10-25 20:52:21,304 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 20:52:22,880
227
+ Results:
228
+ - F-score (micro) 0.7843
229
+ - F-score (macro) 0.6944
230
+ - Accuracy 0.6691
231
+
232
+ By class:
233
+ precision recall f1-score support
234
+
235
+ PER 0.7899 0.8534 0.8204 348
236
+ LOC 0.7611 0.8544 0.8051 261
237
+ ORG 0.4565 0.4038 0.4286 52
238
+ HumanProd 0.6800 0.7727 0.7234 22
239
+
240
+ micro avg 0.7541 0.8170 0.7843 683
241
+ macro avg 0.6719 0.7211 0.6944 683
242
+ weighted avg 0.7500 0.8170 0.7816 683
243
+
244
+ 2023-10-25 20:52:22,880 ----------------------------------------------------------------------------------------------------