Hugging Face's logo

stefan-it
/
hmbench-icdar-fr-hmteams-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3

Token Classification
Flair
PyTorch
TensorBoard
French
sequence-tagger-model
Model card Files Metrics Community
stefan-it commited on
Commit
f2580d4
·
1 Parent(s): 1b8d6e5

Upload folder using huggingface_hub

Browse files
Files changed (6) hide show
  1. best-model.pt +3 -0
  2. dev.tsv +0 -0
  3. loss.tsv +11 -0
  4. runs/events.out.tfevents.1697546819.bce904bcef33.2023.9 +3 -0
  5. test.tsv +0 -0
  6. training.log +236 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:63a0f7e2211202e1ae482d9de47ce29f804bd284e322cfabd2a8ea86b174e639
3
+ size 440941957
dev.tsv ADDED
The diff for this file is too large to render. See raw diff
 
loss.tsv ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ EPOCH TIMESTAMP LEARNING_RATE TRAIN_LOSS DEV_LOSS DEV_PRECISION DEV_RECALL DEV_F1 DEV_ACCURACY
2
+ 1 12:48:34 0.0000 0.3670 0.0871 0.7015 0.7070 0.7042 0.5575
3
+ 2 12:50:13 0.0000 0.1217 0.1113 0.7453 0.7647 0.7549 0.6225
4
+ 3 12:51:49 0.0000 0.0921 0.1251 0.7588 0.7330 0.7457 0.6067
5
+ 4 12:53:26 0.0000 0.0719 0.1501 0.7213 0.7670 0.7434 0.6130
6
+ 5 12:55:02 0.0000 0.0543 0.2085 0.7376 0.7568 0.7471 0.6115
7
+ 6 12:56:39 0.0000 0.0396 0.2126 0.7338 0.7704 0.7517 0.6197
8
+ 7 12:58:16 0.0000 0.0299 0.2321 0.7158 0.7805 0.7468 0.6139
9
+ 8 12:59:51 0.0000 0.0203 0.2372 0.7465 0.7794 0.7626 0.6321
10
+ 9 13:01:29 0.0000 0.0138 0.2417 0.7179 0.7919 0.7531 0.6217
11
+ 10 13:03:07 0.0000 0.0092 0.2555 0.7310 0.7839 0.7566 0.6249
runs/events.out.tfevents.1697546819.bce904bcef33.2023.9 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ca298b7e6acd6ab95016ae4dea501ac9c37d980776f82ef15c1f4fe79d2a4e55
3
+ size 1108164
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,236 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-17 12:46:59,129 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-17 12:46:59,130 Model: "SequenceTagger(
3
+ (embeddings): TransformerWordEmbeddings(
4
+ (model): ElectraModel(
5
+ (embeddings): ElectraEmbeddings(
6
+ (word_embeddings): Embedding(32001, 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): ElectraEncoder(
13
+ (layer): ModuleList(
14
+ (0-11): 12 x ElectraLayer(
15
+ (attention): ElectraAttention(
16
+ (self): ElectraSelfAttention(
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): ElectraSelfOutput(
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): ElectraIntermediate(
29
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
30
+ (intermediate_act_fn): GELUActivation()
31
+ )
32
+ (output): ElectraOutput(
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
+ )
41
+ )
42
+ (locked_dropout): LockedDropout(p=0.5)
43
+ (linear): Linear(in_features=768, out_features=13, bias=True)
44
+ (loss_function): CrossEntropyLoss()
45
+ )"
46
+ 2023-10-17 12:46:59,130 ----------------------------------------------------------------------------------------------------
47
+ 2023-10-17 12:46:59,130 MultiCorpus: 7936 train + 992 dev + 992 test sentences
48
+ - NER_ICDAR_EUROPEANA Corpus: 7936 train + 992 dev + 992 test sentences - /root/.flair/datasets/ner_icdar_europeana/fr
49
+ 2023-10-17 12:46:59,130 ----------------------------------------------------------------------------------------------------
50
+ 2023-10-17 12:46:59,130 Train: 7936 sentences
51
+ 2023-10-17 12:46:59,130 (train_with_dev=False, train_with_test=False)
52
+ 2023-10-17 12:46:59,130 ----------------------------------------------------------------------------------------------------
53
+ 2023-10-17 12:46:59,130 Training Params:
54
+ 2023-10-17 12:46:59,130 - learning_rate: "5e-05"
55
+ 2023-10-17 12:46:59,130 - mini_batch_size: "4"
56
+ 2023-10-17 12:46:59,130 - max_epochs: "10"
57
+ 2023-10-17 12:46:59,130 - shuffle: "True"
58
+ 2023-10-17 12:46:59,130 ----------------------------------------------------------------------------------------------------
59
+ 2023-10-17 12:46:59,130 Plugins:
60
+ 2023-10-17 12:46:59,130 - TensorboardLogger
61
+ 2023-10-17 12:46:59,130 - LinearScheduler | warmup_fraction: '0.1'
62
+ 2023-10-17 12:46:59,130 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-17 12:46:59,130 Final evaluation on model from best epoch (best-model.pt)
64
+ 2023-10-17 12:46:59,130 - metric: "('micro avg', 'f1-score')"
65
+ 2023-10-17 12:46:59,130 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-17 12:46:59,130 Computation:
67
+ 2023-10-17 12:46:59,130 - compute on device: cuda:0
68
+ 2023-10-17 12:46:59,131 - embedding storage: none
69
+ 2023-10-17 12:46:59,131 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-17 12:46:59,131 Model training base path: "hmbench-icdar/fr-hmteams/teams-base-historic-multilingual-discriminator-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3"
71
+ 2023-10-17 12:46:59,131 ----------------------------------------------------------------------------------------------------
72
+ 2023-10-17 12:46:59,131 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-17 12:46:59,131 Logging anything other than scalars to TensorBoard is currently not supported.
74
+ 2023-10-17 12:47:08,412 epoch 1 - iter 198/1984 - loss 1.95138784 - time (sec): 9.28 - samples/sec: 1840.03 - lr: 0.000005 - momentum: 0.000000
75
+ 2023-10-17 12:47:17,762 epoch 1 - iter 396/1984 - loss 1.16755511 - time (sec): 18.63 - samples/sec: 1773.29 - lr: 0.000010 - momentum: 0.000000
76
+ 2023-10-17 12:47:27,276 epoch 1 - iter 594/1984 - loss 0.85631491 - time (sec): 28.14 - samples/sec: 1778.06 - lr: 0.000015 - momentum: 0.000000
77
+ 2023-10-17 12:47:36,514 epoch 1 - iter 792/1984 - loss 0.69613444 - time (sec): 37.38 - samples/sec: 1762.40 - lr: 0.000020 - momentum: 0.000000
78
+ 2023-10-17 12:47:45,647 epoch 1 - iter 990/1984 - loss 0.58337391 - time (sec): 46.52 - samples/sec: 1784.74 - lr: 0.000025 - momentum: 0.000000
79
+ 2023-10-17 12:47:54,968 epoch 1 - iter 1188/1984 - loss 0.50718607 - time (sec): 55.84 - samples/sec: 1799.80 - lr: 0.000030 - momentum: 0.000000
80
+ 2023-10-17 12:48:04,334 epoch 1 - iter 1386/1984 - loss 0.45379013 - time (sec): 65.20 - samples/sec: 1804.34 - lr: 0.000035 - momentum: 0.000000
81
+ 2023-10-17 12:48:13,644 epoch 1 - iter 1584/1984 - loss 0.42012684 - time (sec): 74.51 - samples/sec: 1783.88 - lr: 0.000040 - momentum: 0.000000
82
+ 2023-10-17 12:48:22,679 epoch 1 - iter 1782/1984 - loss 0.39242579 - time (sec): 83.55 - samples/sec: 1771.63 - lr: 0.000045 - momentum: 0.000000
83
+ 2023-10-17 12:48:31,375 epoch 1 - iter 1980/1984 - loss 0.36710615 - time (sec): 92.24 - samples/sec: 1774.74 - lr: 0.000050 - momentum: 0.000000
84
+ 2023-10-17 12:48:31,545 ----------------------------------------------------------------------------------------------------
85
+ 2023-10-17 12:48:31,545 EPOCH 1 done: loss 0.3670 - lr: 0.000050
86
+ 2023-10-17 12:48:34,796 DEV : loss 0.08705931901931763 - f1-score (micro avg) 0.7042
87
+ 2023-10-17 12:48:34,817 saving best model
88
+ 2023-10-17 12:48:35,263 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-17 12:48:44,219 epoch 2 - iter 198/1984 - loss 0.13447725 - time (sec): 8.95 - samples/sec: 1684.39 - lr: 0.000049 - momentum: 0.000000
90
+ 2023-10-17 12:48:53,342 epoch 2 - iter 396/1984 - loss 0.12519486 - time (sec): 18.08 - samples/sec: 1746.79 - lr: 0.000049 - momentum: 0.000000
91
+ 2023-10-17 12:49:02,617 epoch 2 - iter 594/1984 - loss 0.12768946 - time (sec): 27.35 - samples/sec: 1716.69 - lr: 0.000048 - momentum: 0.000000
92
+ 2023-10-17 12:49:12,897 epoch 2 - iter 792/1984 - loss 0.12434640 - time (sec): 37.63 - samples/sec: 1686.98 - lr: 0.000048 - momentum: 0.000000
93
+ 2023-10-17 12:49:22,703 epoch 2 - iter 990/1984 - loss 0.12438353 - time (sec): 47.44 - samples/sec: 1692.29 - lr: 0.000047 - momentum: 0.000000
94
+ 2023-10-17 12:49:31,761 epoch 2 - iter 1188/1984 - loss 0.12321391 - time (sec): 56.50 - samples/sec: 1711.71 - lr: 0.000047 - momentum: 0.000000
95
+ 2023-10-17 12:49:40,930 epoch 2 - iter 1386/1984 - loss 0.12223100 - time (sec): 65.67 - samples/sec: 1715.45 - lr: 0.000046 - momentum: 0.000000
96
+ 2023-10-17 12:49:50,575 epoch 2 - iter 1584/1984 - loss 0.12393554 - time (sec): 75.31 - samples/sec: 1718.04 - lr: 0.000046 - momentum: 0.000000
97
+ 2023-10-17 12:49:59,914 epoch 2 - iter 1782/1984 - loss 0.12338924 - time (sec): 84.65 - samples/sec: 1734.21 - lr: 0.000045 - momentum: 0.000000
98
+ 2023-10-17 12:50:09,035 epoch 2 - iter 1980/1984 - loss 0.12187740 - time (sec): 93.77 - samples/sec: 1745.61 - lr: 0.000044 - momentum: 0.000000
99
+ 2023-10-17 12:50:09,216 ----------------------------------------------------------------------------------------------------
100
+ 2023-10-17 12:50:09,217 EPOCH 2 done: loss 0.1217 - lr: 0.000044
101
+ 2023-10-17 12:50:13,258 DEV : loss 0.11133266240358353 - f1-score (micro avg) 0.7549
102
+ 2023-10-17 12:50:13,279 saving best model
103
+ 2023-10-17 12:50:13,803 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-17 12:50:23,285 epoch 3 - iter 198/1984 - loss 0.10677996 - time (sec): 9.48 - samples/sec: 1681.00 - lr: 0.000044 - momentum: 0.000000
105
+ 2023-10-17 12:50:32,630 epoch 3 - iter 396/1984 - loss 0.10024302 - time (sec): 18.82 - samples/sec: 1736.37 - lr: 0.000043 - momentum: 0.000000
106
+ 2023-10-17 12:50:41,612 epoch 3 - iter 594/1984 - loss 0.09584361 - time (sec): 27.80 - samples/sec: 1762.32 - lr: 0.000043 - momentum: 0.000000
107
+ 2023-10-17 12:50:50,953 epoch 3 - iter 792/1984 - loss 0.09534706 - time (sec): 37.15 - samples/sec: 1757.15 - lr: 0.000042 - momentum: 0.000000
108
+ 2023-10-17 12:51:00,042 epoch 3 - iter 990/1984 - loss 0.09438992 - time (sec): 46.23 - samples/sec: 1747.78 - lr: 0.000042 - momentum: 0.000000
109
+ 2023-10-17 12:51:09,296 epoch 3 - iter 1188/1984 - loss 0.09441971 - time (sec): 55.49 - samples/sec: 1746.25 - lr: 0.000041 - momentum: 0.000000
110
+ 2023-10-17 12:51:18,660 epoch 3 - iter 1386/1984 - loss 0.09258276 - time (sec): 64.85 - samples/sec: 1752.18 - lr: 0.000041 - momentum: 0.000000
111
+ 2023-10-17 12:51:27,819 epoch 3 - iter 1584/1984 - loss 0.09280077 - time (sec): 74.01 - samples/sec: 1778.28 - lr: 0.000040 - momentum: 0.000000
112
+ 2023-10-17 12:51:36,996 epoch 3 - iter 1782/1984 - loss 0.09287512 - time (sec): 83.19 - samples/sec: 1776.15 - lr: 0.000039 - momentum: 0.000000
113
+ 2023-10-17 12:51:46,023 epoch 3 - iter 1980/1984 - loss 0.09227744 - time (sec): 92.22 - samples/sec: 1774.99 - lr: 0.000039 - momentum: 0.000000
114
+ 2023-10-17 12:51:46,205 ----------------------------------------------------------------------------------------------------
115
+ 2023-10-17 12:51:46,205 EPOCH 3 done: loss 0.0921 - lr: 0.000039
116
+ 2023-10-17 12:51:49,815 DEV : loss 0.12508752942085266 - f1-score (micro avg) 0.7457
117
+ 2023-10-17 12:51:49,839 ----------------------------------------------------------------------------------------------------
118
+ 2023-10-17 12:51:59,362 epoch 4 - iter 198/1984 - loss 0.07168588 - time (sec): 9.52 - samples/sec: 1798.75 - lr: 0.000038 - momentum: 0.000000
119
+ 2023-10-17 12:52:08,583 epoch 4 - iter 396/1984 - loss 0.06670880 - time (sec): 18.74 - samples/sec: 1756.76 - lr: 0.000038 - momentum: 0.000000
120
+ 2023-10-17 12:52:17,944 epoch 4 - iter 594/1984 - loss 0.06287904 - time (sec): 28.10 - samples/sec: 1742.76 - lr: 0.000037 - momentum: 0.000000
121
+ 2023-10-17 12:52:27,311 epoch 4 - iter 792/1984 - loss 0.06538020 - time (sec): 37.47 - samples/sec: 1735.71 - lr: 0.000037 - momentum: 0.000000
122
+ 2023-10-17 12:52:36,780 epoch 4 - iter 990/1984 - loss 0.06605263 - time (sec): 46.94 - samples/sec: 1743.88 - lr: 0.000036 - momentum: 0.000000
123
+ 2023-10-17 12:52:46,097 epoch 4 - iter 1188/1984 - loss 0.06749132 - time (sec): 56.26 - samples/sec: 1734.01 - lr: 0.000036 - momentum: 0.000000
124
+ 2023-10-17 12:52:55,393 epoch 4 - iter 1386/1984 - loss 0.07064539 - time (sec): 65.55 - samples/sec: 1746.22 - lr: 0.000035 - momentum: 0.000000
125
+ 2023-10-17 12:53:04,548 epoch 4 - iter 1584/1984 - loss 0.07083371 - time (sec): 74.71 - samples/sec: 1753.43 - lr: 0.000034 - momentum: 0.000000
126
+ 2023-10-17 12:53:13,831 epoch 4 - iter 1782/1984 - loss 0.07085589 - time (sec): 83.99 - samples/sec: 1755.38 - lr: 0.000034 - momentum: 0.000000
127
+ 2023-10-17 12:53:23,039 epoch 4 - iter 1980/1984 - loss 0.07179751 - time (sec): 93.20 - samples/sec: 1755.99 - lr: 0.000033 - momentum: 0.000000
128
+ 2023-10-17 12:53:23,235 ----------------------------------------------------------------------------------------------------
129
+ 2023-10-17 12:53:23,235 EPOCH 4 done: loss 0.0719 - lr: 0.000033
130
+ 2023-10-17 12:53:26,874 DEV : loss 0.15007272362709045 - f1-score (micro avg) 0.7434
131
+ 2023-10-17 12:53:26,898 ----------------------------------------------------------------------------------------------------
132
+ 2023-10-17 12:53:36,403 epoch 5 - iter 198/1984 - loss 0.05191204 - time (sec): 9.50 - samples/sec: 1735.42 - lr: 0.000033 - momentum: 0.000000
133
+ 2023-10-17 12:53:45,521 epoch 5 - iter 396/1984 - loss 0.04935558 - time (sec): 18.62 - samples/sec: 1774.10 - lr: 0.000032 - momentum: 0.000000
134
+ 2023-10-17 12:53:54,441 epoch 5 - iter 594/1984 - loss 0.05488619 - time (sec): 27.54 - samples/sec: 1768.34 - lr: 0.000032 - momentum: 0.000000
135
+ 2023-10-17 12:54:03,624 epoch 5 - iter 792/1984 - loss 0.05643838 - time (sec): 36.72 - samples/sec: 1766.61 - lr: 0.000031 - momentum: 0.000000
136
+ 2023-10-17 12:54:12,819 epoch 5 - iter 990/1984 - loss 0.05590674 - time (sec): 45.92 - samples/sec: 1773.21 - lr: 0.000031 - momentum: 0.000000
137
+ 2023-10-17 12:54:21,976 epoch 5 - iter 1188/1984 - loss 0.05482986 - time (sec): 55.08 - samples/sec: 1755.80 - lr: 0.000030 - momentum: 0.000000
138
+ 2023-10-17 12:54:31,108 epoch 5 - iter 1386/1984 - loss 0.05386415 - time (sec): 64.21 - samples/sec: 1772.47 - lr: 0.000029 - momentum: 0.000000
139
+ 2023-10-17 12:54:40,403 epoch 5 - iter 1584/1984 - loss 0.05340092 - time (sec): 73.50 - samples/sec: 1770.69 - lr: 0.000029 - momentum: 0.000000
140
+ 2023-10-17 12:54:49,745 epoch 5 - iter 1782/1984 - loss 0.05379444 - time (sec): 82.85 - samples/sec: 1771.56 - lr: 0.000028 - momentum: 0.000000
141
+ 2023-10-17 12:54:59,027 epoch 5 - iter 1980/1984 - loss 0.05432464 - time (sec): 92.13 - samples/sec: 1776.36 - lr: 0.000028 - momentum: 0.000000
142
+ 2023-10-17 12:54:59,213 ----------------------------------------------------------------------------------------------------
143
+ 2023-10-17 12:54:59,214 EPOCH 5 done: loss 0.0543 - lr: 0.000028
144
+ 2023-10-17 12:55:02,841 DEV : loss 0.20846031606197357 - f1-score (micro avg) 0.7471
145
+ 2023-10-17 12:55:02,866 ----------------------------------------------------------------------------------------------------
146
+ 2023-10-17 12:55:12,114 epoch 6 - iter 198/1984 - loss 0.03953401 - time (sec): 9.25 - samples/sec: 1769.03 - lr: 0.000027 - momentum: 0.000000
147
+ 2023-10-17 12:55:21,364 epoch 6 - iter 396/1984 - loss 0.03812773 - time (sec): 18.50 - samples/sec: 1765.39 - lr: 0.000027 - momentum: 0.000000
148
+ 2023-10-17 12:55:30,551 epoch 6 - iter 594/1984 - loss 0.04268890 - time (sec): 27.68 - samples/sec: 1778.53 - lr: 0.000026 - momentum: 0.000000
149
+ 2023-10-17 12:55:40,019 epoch 6 - iter 792/1984 - loss 0.04138806 - time (sec): 37.15 - samples/sec: 1769.41 - lr: 0.000026 - momentum: 0.000000
150
+ 2023-10-17 12:55:49,073 epoch 6 - iter 990/1984 - loss 0.04025963 - time (sec): 46.21 - samples/sec: 1763.61 - lr: 0.000025 - momentum: 0.000000
151
+ 2023-10-17 12:55:58,111 epoch 6 - iter 1188/1984 - loss 0.03971711 - time (sec): 55.24 - samples/sec: 1746.68 - lr: 0.000024 - momentum: 0.000000
152
+ 2023-10-17 12:56:07,294 epoch 6 - iter 1386/1984 - loss 0.04025272 - time (sec): 64.43 - samples/sec: 1752.53 - lr: 0.000024 - momentum: 0.000000
153
+ 2023-10-17 12:56:16,467 epoch 6 - iter 1584/1984 - loss 0.03959126 - time (sec): 73.60 - samples/sec: 1771.43 - lr: 0.000023 - momentum: 0.000000
154
+ 2023-10-17 12:56:25,663 epoch 6 - iter 1782/1984 - loss 0.03963302 - time (sec): 82.80 - samples/sec: 1766.72 - lr: 0.000023 - momentum: 0.000000
155
+ 2023-10-17 12:56:35,200 epoch 6 - iter 1980/1984 - loss 0.03953747 - time (sec): 92.33 - samples/sec: 1772.28 - lr: 0.000022 - momentum: 0.000000
156
+ 2023-10-17 12:56:35,381 ----------------------------------------------------------------------------------------------------
157
+ 2023-10-17 12:56:35,381 EPOCH 6 done: loss 0.0396 - lr: 0.000022
158
+ 2023-10-17 12:56:39,041 DEV : loss 0.21264401078224182 - f1-score (micro avg) 0.7517
159
+ 2023-10-17 12:56:39,065 ----------------------------------------------------------------------------------------------------
160
+ 2023-10-17 12:56:48,296 epoch 7 - iter 198/1984 - loss 0.02510474 - time (sec): 9.23 - samples/sec: 1692.24 - lr: 0.000022 - momentum: 0.000000
161
+ 2023-10-17 12:56:57,912 epoch 7 - iter 396/1984 - loss 0.02487195 - time (sec): 18.85 - samples/sec: 1760.14 - lr: 0.000021 - momentum: 0.000000
162
+ 2023-10-17 12:57:07,100 epoch 7 - iter 594/1984 - loss 0.02627481 - time (sec): 28.03 - samples/sec: 1757.59 - lr: 0.000021 - momentum: 0.000000
163
+ 2023-10-17 12:57:16,372 epoch 7 - iter 792/1984 - loss 0.03015138 - time (sec): 37.31 - samples/sec: 1746.05 - lr: 0.000020 - momentum: 0.000000
164
+ 2023-10-17 12:57:25,611 epoch 7 - iter 990/1984 - loss 0.02986521 - time (sec): 46.54 - samples/sec: 1749.85 - lr: 0.000019 - momentum: 0.000000
165
+ 2023-10-17 12:57:34,638 epoch 7 - iter 1188/1984 - loss 0.02921708 - time (sec): 55.57 - samples/sec: 1768.09 - lr: 0.000019 - momentum: 0.000000
166
+ 2023-10-17 12:57:44,038 epoch 7 - iter 1386/1984 - loss 0.02876629 - time (sec): 64.97 - samples/sec: 1773.71 - lr: 0.000018 - momentum: 0.000000
167
+ 2023-10-17 12:57:53,266 epoch 7 - iter 1584/1984 - loss 0.02894997 - time (sec): 74.20 - samples/sec: 1770.84 - lr: 0.000018 - momentum: 0.000000
168
+ 2023-10-17 12:58:02,591 epoch 7 - iter 1782/1984 - loss 0.03019572 - time (sec): 83.52 - samples/sec: 1765.81 - lr: 0.000017 - momentum: 0.000000
169
+ 2023-10-17 12:58:11,832 epoch 7 - iter 1980/1984 - loss 0.02983351 - time (sec): 92.77 - samples/sec: 1764.81 - lr: 0.000017 - momentum: 0.000000
170
+ 2023-10-17 12:58:12,018 ----------------------------------------------------------------------------------------------------
171
+ 2023-10-17 12:58:12,018 EPOCH 7 done: loss 0.0299 - lr: 0.000017
172
+ 2023-10-17 12:58:16,123 DEV : loss 0.23210309445858002 - f1-score (micro avg) 0.7468
173
+ 2023-10-17 12:58:16,144 ----------------------------------------------------------------------------------------------------
174
+ 2023-10-17 12:58:25,261 epoch 8 - iter 198/1984 - loss 0.02662398 - time (sec): 9.12 - samples/sec: 1763.88 - lr: 0.000016 - momentum: 0.000000
175
+ 2023-10-17 12:58:34,458 epoch 8 - iter 396/1984 - loss 0.02320812 - time (sec): 18.31 - samples/sec: 1753.52 - lr: 0.000016 - momentum: 0.000000
176
+ 2023-10-17 12:58:43,623 epoch 8 - iter 594/1984 - loss 0.02078829 - time (sec): 27.48 - samples/sec: 1758.96 - lr: 0.000015 - momentum: 0.000000
177
+ 2023-10-17 12:58:53,133 epoch 8 - iter 792/1984 - loss 0.01981068 - time (sec): 36.99 - samples/sec: 1754.70 - lr: 0.000014 - momentum: 0.000000
178
+ 2023-10-17 12:59:02,118 epoch 8 - iter 990/1984 - loss 0.02018388 - time (sec): 45.97 - samples/sec: 1780.30 - lr: 0.000014 - momentum: 0.000000
179
+ 2023-10-17 12:59:11,160 epoch 8 - iter 1188/1984 - loss 0.02143587 - time (sec): 55.02 - samples/sec: 1768.00 - lr: 0.000013 - momentum: 0.000000
180
+ 2023-10-17 12:59:20,403 epoch 8 - iter 1386/1984 - loss 0.02108215 - time (sec): 64.26 - samples/sec: 1779.78 - lr: 0.000013 - momentum: 0.000000
181
+ 2023-10-17 12:59:29,551 epoch 8 - iter 1584/1984 - loss 0.02050008 - time (sec): 73.41 - samples/sec: 1790.79 - lr: 0.000012 - momentum: 0.000000
182
+ 2023-10-17 12:59:38,685 epoch 8 - iter 1782/1984 - loss 0.02034437 - time (sec): 82.54 - samples/sec: 1790.41 - lr: 0.000012 - momentum: 0.000000
183
+ 2023-10-17 12:59:47,933 epoch 8 - iter 1980/1984 - loss 0.02018234 - time (sec): 91.79 - samples/sec: 1782.77 - lr: 0.000011 - momentum: 0.000000
184
+ 2023-10-17 12:59:48,133 ----------------------------------------------------------------------------------------------------
185
+ 2023-10-17 12:59:48,133 EPOCH 8 done: loss 0.0203 - lr: 0.000011
186
+ 2023-10-17 12:59:51,750 DEV : loss 0.23719041049480438 - f1-score (micro avg) 0.7626
187
+ 2023-10-17 12:59:51,774 saving best model
188
+ 2023-10-17 12:59:52,342 ----------------------------------------------------------------------------------------------------
189
+ 2023-10-17 13:00:01,612 epoch 9 - iter 198/1984 - loss 0.01128630 - time (sec): 9.27 - samples/sec: 1795.34 - lr: 0.000011 - momentum: 0.000000
190
+ 2023-10-17 13:00:10,749 epoch 9 - iter 396/1984 - loss 0.00943051 - time (sec): 18.41 - samples/sec: 1820.75 - lr: 0.000010 - momentum: 0.000000
191
+ 2023-10-17 13:00:19,875 epoch 9 - iter 594/1984 - loss 0.01123883 - time (sec): 27.53 - samples/sec: 1777.99 - lr: 0.000009 - momentum: 0.000000
192
+ 2023-10-17 13:00:29,376 epoch 9 - iter 792/1984 - loss 0.01104369 - time (sec): 37.03 - samples/sec: 1767.64 - lr: 0.000009 - momentum: 0.000000
193
+ 2023-10-17 13:00:38,816 epoch 9 - iter 990/1984 - loss 0.01110156 - time (sec): 46.47 - samples/sec: 1763.83 - lr: 0.000008 - momentum: 0.000000
194
+ 2023-10-17 13:00:48,150 epoch 9 - iter 1188/1984 - loss 0.01314728 - time (sec): 55.81 - samples/sec: 1775.48 - lr: 0.000008 - momentum: 0.000000
195
+ 2023-10-17 13:00:57,267 epoch 9 - iter 1386/1984 - loss 0.01317730 - time (sec): 64.92 - samples/sec: 1764.93 - lr: 0.000007 - momentum: 0.000000
196
+ 2023-10-17 13:01:06,731 epoch 9 - iter 1584/1984 - loss 0.01382017 - time (sec): 74.39 - samples/sec: 1760.10 - lr: 0.000007 - momentum: 0.000000
197
+ 2023-10-17 13:01:15,926 epoch 9 - iter 1782/1984 - loss 0.01358330 - time (sec): 83.58 - samples/sec: 1759.37 - lr: 0.000006 - momentum: 0.000000
198
+ 2023-10-17 13:01:25,451 epoch 9 - iter 1980/1984 - loss 0.01373408 - time (sec): 93.11 - samples/sec: 1756.06 - lr: 0.000006 - momentum: 0.000000
199
+ 2023-10-17 13:01:25,664 ----------------------------------------------------------------------------------------------------
200
+ 2023-10-17 13:01:25,665 EPOCH 9 done: loss 0.0138 - lr: 0.000006
201
+ 2023-10-17 13:01:29,408 DEV : loss 0.2416975200176239 - f1-score (micro avg) 0.7531
202
+ 2023-10-17 13:01:29,434 ----------------------------------------------------------------------------------------------------
203
+ 2023-10-17 13:01:38,722 epoch 10 - iter 198/1984 - loss 0.00724275 - time (sec): 9.29 - samples/sec: 1731.53 - lr: 0.000005 - momentum: 0.000000
204
+ 2023-10-17 13:01:48,062 epoch 10 - iter 396/1984 - loss 0.00878620 - time (sec): 18.63 - samples/sec: 1741.75 - lr: 0.000004 - momentum: 0.000000
205
+ 2023-10-17 13:01:57,341 epoch 10 - iter 594/1984 - loss 0.00792020 - time (sec): 27.91 - samples/sec: 1788.05 - lr: 0.000004 - momentum: 0.000000
206
+ 2023-10-17 13:02:06,885 epoch 10 - iter 792/1984 - loss 0.00864729 - time (sec): 37.45 - samples/sec: 1766.05 - lr: 0.000003 - momentum: 0.000000
207
+ 2023-10-17 13:02:16,926 epoch 10 - iter 990/1984 - loss 0.00875296 - time (sec): 47.49 - samples/sec: 1730.47 - lr: 0.000003 - momentum: 0.000000
208
+ 2023-10-17 13:02:26,316 epoch 10 - iter 1188/1984 - loss 0.00957040 - time (sec): 56.88 - samples/sec: 1731.11 - lr: 0.000002 - momentum: 0.000000
209
+ 2023-10-17 13:02:35,536 epoch 10 - iter 1386/1984 - loss 0.00980326 - time (sec): 66.10 - samples/sec: 1733.47 - lr: 0.000002 - momentum: 0.000000
210
+ 2023-10-17 13:02:44,848 epoch 10 - iter 1584/1984 - loss 0.00931466 - time (sec): 75.41 - samples/sec: 1747.44 - lr: 0.000001 - momentum: 0.000000
211
+ 2023-10-17 13:02:53,940 epoch 10 - iter 1782/1984 - loss 0.00906989 - time (sec): 84.50 - samples/sec: 1750.28 - lr: 0.000001 - momentum: 0.000000
212
+ 2023-10-17 13:03:03,073 epoch 10 - iter 1980/1984 - loss 0.00906925 - time (sec): 93.64 - samples/sec: 1748.58 - lr: 0.000000 - momentum: 0.000000
213
+ 2023-10-17 13:03:03,259 ----------------------------------------------------------------------------------------------------
214
+ 2023-10-17 13:03:03,259 EPOCH 10 done: loss 0.0092 - lr: 0.000000
215
+ 2023-10-17 13:03:06,977 DEV : loss 0.25551512837409973 - f1-score (micro avg) 0.7566
216
+ 2023-10-17 13:03:07,490 ----------------------------------------------------------------------------------------------------
217
+ 2023-10-17 13:03:07,492 Loading model from best epoch ...
218
+ 2023-10-17 13:03:09,090 SequenceTagger predicts: Dictionary with 13 tags: O, S-PER, B-PER, E-PER, I-PER, S-LOC, B-LOC, E-LOC, I-LOC, S-ORG, B-ORG, E-ORG, I-ORG
219
+ 2023-10-17 13:03:12,571
220
+ Results:
221
+ - F-score (micro) 0.7774
222
+ - F-score (macro) 0.6822
223
+ - Accuracy 0.6592
224
+
225
+ By class:
226
+ precision recall f1-score support
227
+
228
+ LOC 0.8343 0.8611 0.8475 655
229
+ PER 0.6920 0.7758 0.7315 223
230
+ ORG 0.5192 0.4252 0.4675 127
231
+
232
+ micro avg 0.7680 0.7871 0.7774 1005
233
+ macro avg 0.6819 0.6874 0.6822 1005
234
+ weighted avg 0.7629 0.7871 0.7737 1005
235
+
236
+ 2023-10-17 13:03:12,571 ----------------------------------------------------------------------------------------------------