Hugging Face's logo

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

Token Classification
Flair
PyTorch
TensorBoard
French
sequence-tagger-model
Model card Files Metrics Community
stefan-it commited on
Commit
f549f2f
•
1 Parent(s): 750847e

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.1697540060.bce904bcef33.2023.1 +3 -0
  5. test.tsv +0 -0
  6. training.log +237 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:680446c99fc3ed5ae28e1ca18d3d30019b8ce12d3e2590d5201b6fcaf9e59806
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 10:55:53 0.0000 0.3305 0.3863 0.0169 0.0023 0.0040 0.0020
3
+ 2 10:57:27 0.0000 0.1296 0.1098 0.7564 0.6991 0.7266 0.5830
4
+ 3 10:59:03 0.0000 0.0927 0.1182 0.7643 0.7704 0.7673 0.6418
5
+ 4 11:00:37 0.0000 0.0722 0.1527 0.7703 0.7624 0.7663 0.6358
6
+ 5 11:02:11 0.0000 0.0539 0.1992 0.7436 0.7579 0.7507 0.6198
7
+ 6 11:03:47 0.0000 0.0429 0.1953 0.7357 0.7839 0.7590 0.6317
8
+ 7 11:05:23 0.0000 0.0292 0.2106 0.7495 0.7817 0.7652 0.6386
9
+ 8 11:06:57 0.0000 0.0189 0.2515 0.7465 0.7828 0.7642 0.6337
10
+ 9 11:08:31 0.0000 0.0136 0.2389 0.7415 0.7885 0.7643 0.6354
11
+ 10 11:10:03 0.0000 0.0076 0.2602 0.7566 0.7805 0.7684 0.6389
runs/events.out.tfevents.1697540060.bce904bcef33.2023.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:52ea632b95cf1b2e6d58dd1a39bf0ad051a86a0a6cd194ee7ef8f9c7c23a1d95
3
+ size 1108164
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,237 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-17 10:54:20,438 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-17 10:54:20,439 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 10:54:20,439 ----------------------------------------------------------------------------------------------------
47
+ 2023-10-17 10:54:20,440 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 10:54:20,440 ----------------------------------------------------------------------------------------------------
50
+ 2023-10-17 10:54:20,440 Train: 7936 sentences
51
+ 2023-10-17 10:54:20,440 (train_with_dev=False, train_with_test=False)
52
+ 2023-10-17 10:54:20,440 ----------------------------------------------------------------------------------------------------
53
+ 2023-10-17 10:54:20,440 Training Params:
54
+ 2023-10-17 10:54:20,440 - learning_rate: "5e-05"
55
+ 2023-10-17 10:54:20,440 - mini_batch_size: "4"
56
+ 2023-10-17 10:54:20,440 - max_epochs: "10"
57
+ 2023-10-17 10:54:20,440 - shuffle: "True"
58
+ 2023-10-17 10:54:20,440 ----------------------------------------------------------------------------------------------------
59
+ 2023-10-17 10:54:20,440 Plugins:
60
+ 2023-10-17 10:54:20,440 - TensorboardLogger
61
+ 2023-10-17 10:54:20,440 - LinearScheduler | warmup_fraction: '0.1'
62
+ 2023-10-17 10:54:20,440 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-17 10:54:20,440 Final evaluation on model from best epoch (best-model.pt)
64
+ 2023-10-17 10:54:20,440 - metric: "('micro avg', 'f1-score')"
65
+ 2023-10-17 10:54:20,440 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-17 10:54:20,440 Computation:
67
+ 2023-10-17 10:54:20,440 - compute on device: cuda:0
68
+ 2023-10-17 10:54:20,440 - embedding storage: none
69
+ 2023-10-17 10:54:20,440 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-17 10:54:20,440 Model training base path: "hmbench-icdar/fr-hmteams/teams-base-historic-multilingual-discriminator-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1"
71
+ 2023-10-17 10:54:20,440 ----------------------------------------------------------------------------------------------------
72
+ 2023-10-17 10:54:20,440 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-17 10:54:20,440 Logging anything other than scalars to TensorBoard is currently not supported.
74
+ 2023-10-17 10:54:28,907 epoch 1 - iter 198/1984 - loss 1.62211158 - time (sec): 8.47 - samples/sec: 1911.31 - lr: 0.000005 - momentum: 0.000000
75
+ 2023-10-17 10:54:37,933 epoch 1 - iter 396/1984 - loss 0.94619822 - time (sec): 17.49 - samples/sec: 1900.28 - lr: 0.000010 - momentum: 0.000000
76
+ 2023-10-17 10:54:47,160 epoch 1 - iter 594/1984 - loss 0.70711697 - time (sec): 26.72 - samples/sec: 1857.08 - lr: 0.000015 - momentum: 0.000000
77
+ 2023-10-17 10:54:56,036 epoch 1 - iter 792/1984 - loss 0.58185081 - time (sec): 35.59 - samples/sec: 1837.14 - lr: 0.000020 - momentum: 0.000000
78
+ 2023-10-17 10:55:04,916 epoch 1 - iter 990/1984 - loss 0.49792196 - time (sec): 44.47 - samples/sec: 1840.07 - lr: 0.000025 - momentum: 0.000000
79
+ 2023-10-17 10:55:14,026 epoch 1 - iter 1188/1984 - loss 0.43844628 - time (sec): 53.58 - samples/sec: 1840.99 - lr: 0.000030 - momentum: 0.000000
80
+ 2023-10-17 10:55:22,978 epoch 1 - iter 1386/1984 - loss 0.39736182 - time (sec): 62.54 - samples/sec: 1835.63 - lr: 0.000035 - momentum: 0.000000
81
+ 2023-10-17 10:55:32,299 epoch 1 - iter 1584/1984 - loss 0.36824360 - time (sec): 71.86 - samples/sec: 1827.19 - lr: 0.000040 - momentum: 0.000000
82
+ 2023-10-17 10:55:41,543 epoch 1 - iter 1782/1984 - loss 0.34485394 - time (sec): 81.10 - samples/sec: 1815.57 - lr: 0.000045 - momentum: 0.000000
83
+ 2023-10-17 10:55:50,632 epoch 1 - iter 1980/1984 - loss 0.33011193 - time (sec): 90.19 - samples/sec: 1814.56 - lr: 0.000050 - momentum: 0.000000
84
+ 2023-10-17 10:55:50,815 ----------------------------------------------------------------------------------------------------
85
+ 2023-10-17 10:55:50,815 EPOCH 1 done: loss 0.3305 - lr: 0.000050
86
+ 2023-10-17 10:55:53,856 DEV : loss 0.3863277733325958 - f1-score (micro avg) 0.004
87
+ 2023-10-17 10:55:53,876 saving best model
88
+ 2023-10-17 10:55:54,244 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-17 10:56:03,257 epoch 2 - iter 198/1984 - loss 0.16096443 - time (sec): 9.01 - samples/sec: 1876.10 - lr: 0.000049 - momentum: 0.000000
90
+ 2023-10-17 10:56:11,979 epoch 2 - iter 396/1984 - loss 0.14525930 - time (sec): 17.73 - samples/sec: 1867.53 - lr: 0.000049 - momentum: 0.000000
91
+ 2023-10-17 10:56:20,750 epoch 2 - iter 594/1984 - loss 0.14335165 - time (sec): 26.50 - samples/sec: 1872.28 - lr: 0.000048 - momentum: 0.000000
92
+ 2023-10-17 10:56:29,643 epoch 2 - iter 792/1984 - loss 0.13927613 - time (sec): 35.40 - samples/sec: 1877.08 - lr: 0.000048 - momentum: 0.000000
93
+ 2023-10-17 10:56:38,576 epoch 2 - iter 990/1984 - loss 0.13757993 - time (sec): 44.33 - samples/sec: 1862.58 - lr: 0.000047 - momentum: 0.000000
94
+ 2023-10-17 10:56:47,747 epoch 2 - iter 1188/1984 - loss 0.13290183 - time (sec): 53.50 - samples/sec: 1856.82 - lr: 0.000047 - momentum: 0.000000
95
+ 2023-10-17 10:56:56,765 epoch 2 - iter 1386/1984 - loss 0.13087811 - time (sec): 62.52 - samples/sec: 1854.94 - lr: 0.000046 - momentum: 0.000000
96
+ 2023-10-17 10:57:05,839 epoch 2 - iter 1584/1984 - loss 0.13141431 - time (sec): 71.59 - samples/sec: 1839.27 - lr: 0.000046 - momentum: 0.000000
97
+ 2023-10-17 10:57:14,783 epoch 2 - iter 1782/1984 - loss 0.13018538 - time (sec): 80.54 - samples/sec: 1828.57 - lr: 0.000045 - momentum: 0.000000
98
+ 2023-10-17 10:57:23,835 epoch 2 - iter 1980/1984 - loss 0.12957544 - time (sec): 89.59 - samples/sec: 1826.70 - lr: 0.000044 - momentum: 0.000000
99
+ 2023-10-17 10:57:24,015 ----------------------------------------------------------------------------------------------------
100
+ 2023-10-17 10:57:24,015 EPOCH 2 done: loss 0.1296 - lr: 0.000044
101
+ 2023-10-17 10:57:27,871 DEV : loss 0.10983909666538239 - f1-score (micro avg) 0.7266
102
+ 2023-10-17 10:57:27,892 saving best model
103
+ 2023-10-17 10:57:28,372 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-17 10:57:37,508 epoch 3 - iter 198/1984 - loss 0.09127543 - time (sec): 9.13 - samples/sec: 1814.20 - lr: 0.000044 - momentum: 0.000000
105
+ 2023-10-17 10:57:46,635 epoch 3 - iter 396/1984 - loss 0.09356135 - time (sec): 18.26 - samples/sec: 1799.61 - lr: 0.000043 - momentum: 0.000000
106
+ 2023-10-17 10:57:55,783 epoch 3 - iter 594/1984 - loss 0.09391553 - time (sec): 27.41 - samples/sec: 1810.02 - lr: 0.000043 - momentum: 0.000000
107
+ 2023-10-17 10:58:04,831 epoch 3 - iter 792/1984 - loss 0.09087245 - time (sec): 36.46 - samples/sec: 1821.93 - lr: 0.000042 - momentum: 0.000000
108
+ 2023-10-17 10:58:13,828 epoch 3 - iter 990/1984 - loss 0.09252505 - time (sec): 45.45 - samples/sec: 1819.54 - lr: 0.000042 - momentum: 0.000000
109
+ 2023-10-17 10:58:22,910 epoch 3 - iter 1188/1984 - loss 0.09198370 - time (sec): 54.54 - samples/sec: 1805.55 - lr: 0.000041 - momentum: 0.000000
110
+ 2023-10-17 10:58:31,977 epoch 3 - iter 1386/1984 - loss 0.09210720 - time (sec): 63.60 - samples/sec: 1804.93 - lr: 0.000041 - momentum: 0.000000
111
+ 2023-10-17 10:58:41,054 epoch 3 - iter 1584/1984 - loss 0.09115675 - time (sec): 72.68 - samples/sec: 1804.76 - lr: 0.000040 - momentum: 0.000000
112
+ 2023-10-17 10:58:50,159 epoch 3 - iter 1782/1984 - loss 0.09117253 - time (sec): 81.79 - samples/sec: 1808.36 - lr: 0.000039 - momentum: 0.000000
113
+ 2023-10-17 10:58:59,410 epoch 3 - iter 1980/1984 - loss 0.09250784 - time (sec): 91.04 - samples/sec: 1797.18 - lr: 0.000039 - momentum: 0.000000
114
+ 2023-10-17 10:58:59,604 ----------------------------------------------------------------------------------------------------
115
+ 2023-10-17 10:58:59,604 EPOCH 3 done: loss 0.0927 - lr: 0.000039
116
+ 2023-10-17 10:59:03,046 DEV : loss 0.11815514415502548 - f1-score (micro avg) 0.7673
117
+ 2023-10-17 10:59:03,068 saving best model
118
+ 2023-10-17 10:59:03,547 ----------------------------------------------------------------------------------------------------
119
+ 2023-10-17 10:59:12,752 epoch 4 - iter 198/1984 - loss 0.06981980 - time (sec): 9.20 - samples/sec: 1783.67 - lr: 0.000038 - momentum: 0.000000
120
+ 2023-10-17 10:59:22,063 epoch 4 - iter 396/1984 - loss 0.07072403 - time (sec): 18.51 - samples/sec: 1844.43 - lr: 0.000038 - momentum: 0.000000
121
+ 2023-10-17 10:59:31,256 epoch 4 - iter 594/1984 - loss 0.07503217 - time (sec): 27.71 - samples/sec: 1838.25 - lr: 0.000037 - momentum: 0.000000
122
+ 2023-10-17 10:59:40,298 epoch 4 - iter 792/1984 - loss 0.07356156 - time (sec): 36.75 - samples/sec: 1835.12 - lr: 0.000037 - momentum: 0.000000
123
+ 2023-10-17 10:59:49,285 epoch 4 - iter 990/1984 - loss 0.07317536 - time (sec): 45.74 - samples/sec: 1829.88 - lr: 0.000036 - momentum: 0.000000
124
+ 2023-10-17 10:59:58,054 epoch 4 - iter 1188/1984 - loss 0.07519900 - time (sec): 54.51 - samples/sec: 1822.01 - lr: 0.000036 - momentum: 0.000000
125
+ 2023-10-17 11:00:06,621 epoch 4 - iter 1386/1984 - loss 0.07351265 - time (sec): 63.07 - samples/sec: 1822.06 - lr: 0.000035 - momentum: 0.000000
126
+ 2023-10-17 11:00:15,739 epoch 4 - iter 1584/1984 - loss 0.07234519 - time (sec): 72.19 - samples/sec: 1814.86 - lr: 0.000034 - momentum: 0.000000
127
+ 2023-10-17 11:00:24,737 epoch 4 - iter 1782/1984 - loss 0.07317391 - time (sec): 81.19 - samples/sec: 1819.38 - lr: 0.000034 - momentum: 0.000000
128
+ 2023-10-17 11:00:33,691 epoch 4 - iter 1980/1984 - loss 0.07222348 - time (sec): 90.14 - samples/sec: 1816.69 - lr: 0.000033 - momentum: 0.000000
129
+ 2023-10-17 11:00:33,875 ----------------------------------------------------------------------------------------------------
130
+ 2023-10-17 11:00:33,875 EPOCH 4 done: loss 0.0722 - lr: 0.000033
131
+ 2023-10-17 11:00:37,278 DEV : loss 0.15266923606395721 - f1-score (micro avg) 0.7663
132
+ 2023-10-17 11:00:37,300 ----------------------------------------------------------------------------------------------------
133
+ 2023-10-17 11:00:46,355 epoch 5 - iter 198/1984 - loss 0.05046205 - time (sec): 9.05 - samples/sec: 1795.87 - lr: 0.000033 - momentum: 0.000000
134
+ 2023-10-17 11:00:55,385 epoch 5 - iter 396/1984 - loss 0.05247464 - time (sec): 18.08 - samples/sec: 1852.85 - lr: 0.000032 - momentum: 0.000000
135
+ 2023-10-17 11:01:04,492 epoch 5 - iter 594/1984 - loss 0.05489266 - time (sec): 27.19 - samples/sec: 1846.51 - lr: 0.000032 - momentum: 0.000000
136
+ 2023-10-17 11:01:13,843 epoch 5 - iter 792/1984 - loss 0.05563933 - time (sec): 36.54 - samples/sec: 1853.29 - lr: 0.000031 - momentum: 0.000000
137
+ 2023-10-17 11:01:23,006 epoch 5 - iter 990/1984 - loss 0.05335869 - time (sec): 45.70 - samples/sec: 1841.01 - lr: 0.000031 - momentum: 0.000000
138
+ 2023-10-17 11:01:32,134 epoch 5 - iter 1188/1984 - loss 0.05269990 - time (sec): 54.83 - samples/sec: 1823.48 - lr: 0.000030 - momentum: 0.000000
139
+ 2023-10-17 11:01:41,271 epoch 5 - iter 1386/1984 - loss 0.05371472 - time (sec): 63.97 - samples/sec: 1819.47 - lr: 0.000029 - momentum: 0.000000
140
+ 2023-10-17 11:01:50,241 epoch 5 - iter 1584/1984 - loss 0.05415103 - time (sec): 72.94 - samples/sec: 1812.74 - lr: 0.000029 - momentum: 0.000000
141
+ 2023-10-17 11:01:59,308 epoch 5 - iter 1782/1984 - loss 0.05436597 - time (sec): 82.01 - samples/sec: 1807.24 - lr: 0.000028 - momentum: 0.000000
142
+ 2023-10-17 11:02:08,369 epoch 5 - iter 1980/1984 - loss 0.05389026 - time (sec): 91.07 - samples/sec: 1796.79 - lr: 0.000028 - momentum: 0.000000
143
+ 2023-10-17 11:02:08,550 ----------------------------------------------------------------------------------------------------
144
+ 2023-10-17 11:02:08,551 EPOCH 5 done: loss 0.0539 - lr: 0.000028
145
+ 2023-10-17 11:02:11,912 DEV : loss 0.19919680058956146 - f1-score (micro avg) 0.7507
146
+ 2023-10-17 11:02:11,932 ----------------------------------------------------------------------------------------------------
147
+ 2023-10-17 11:02:21,160 epoch 6 - iter 198/1984 - loss 0.04620617 - time (sec): 9.23 - samples/sec: 1782.45 - lr: 0.000027 - momentum: 0.000000
148
+ 2023-10-17 11:02:30,334 epoch 6 - iter 396/1984 - loss 0.04843778 - time (sec): 18.40 - samples/sec: 1774.97 - lr: 0.000027 - momentum: 0.000000
149
+ 2023-10-17 11:02:39,513 epoch 6 - iter 594/1984 - loss 0.04484302 - time (sec): 27.58 - samples/sec: 1822.04 - lr: 0.000026 - momentum: 0.000000
150
+ 2023-10-17 11:02:48,566 epoch 6 - iter 792/1984 - loss 0.04250884 - time (sec): 36.63 - samples/sec: 1818.99 - lr: 0.000026 - momentum: 0.000000
151
+ 2023-10-17 11:02:57,657 epoch 6 - iter 990/1984 - loss 0.04152118 - time (sec): 45.72 - samples/sec: 1824.08 - lr: 0.000025 - momentum: 0.000000
152
+ 2023-10-17 11:03:06,709 epoch 6 - iter 1188/1984 - loss 0.04219659 - time (sec): 54.78 - samples/sec: 1820.23 - lr: 0.000024 - momentum: 0.000000
153
+ 2023-10-17 11:03:15,840 epoch 6 - iter 1386/1984 - loss 0.04406527 - time (sec): 63.91 - samples/sec: 1804.54 - lr: 0.000024 - momentum: 0.000000
154
+ 2023-10-17 11:03:25,050 epoch 6 - iter 1584/1984 - loss 0.04345490 - time (sec): 73.12 - samples/sec: 1793.93 - lr: 0.000023 - momentum: 0.000000
155
+ 2023-10-17 11:03:34,090 epoch 6 - iter 1782/1984 - loss 0.04365630 - time (sec): 82.16 - samples/sec: 1792.95 - lr: 0.000023 - momentum: 0.000000
156
+ 2023-10-17 11:03:43,312 epoch 6 - iter 1980/1984 - loss 0.04303766 - time (sec): 91.38 - samples/sec: 1790.06 - lr: 0.000022 - momentum: 0.000000
157
+ 2023-10-17 11:03:43,494 ----------------------------------------------------------------------------------------------------
158
+ 2023-10-17 11:03:43,494 EPOCH 6 done: loss 0.0429 - lr: 0.000022
159
+ 2023-10-17 11:03:46,982 DEV : loss 0.19526612758636475 - f1-score (micro avg) 0.759
160
+ 2023-10-17 11:03:47,005 ----------------------------------------------------------------------------------------------------
161
+ 2023-10-17 11:03:56,272 epoch 7 - iter 198/1984 - loss 0.02150771 - time (sec): 9.27 - samples/sec: 1758.81 - lr: 0.000022 - momentum: 0.000000
162
+ 2023-10-17 11:04:05,509 epoch 7 - iter 396/1984 - loss 0.02251021 - time (sec): 18.50 - samples/sec: 1773.66 - lr: 0.000021 - momentum: 0.000000
163
+ 2023-10-17 11:04:14,821 epoch 7 - iter 594/1984 - loss 0.02431276 - time (sec): 27.82 - samples/sec: 1781.84 - lr: 0.000021 - momentum: 0.000000
164
+ 2023-10-17 11:04:23,989 epoch 7 - iter 792/1984 - loss 0.02510360 - time (sec): 36.98 - samples/sec: 1775.37 - lr: 0.000020 - momentum: 0.000000
165
+ 2023-10-17 11:04:33,324 epoch 7 - iter 990/1984 - loss 0.02564549 - time (sec): 46.32 - samples/sec: 1769.45 - lr: 0.000019 - momentum: 0.000000
166
+ 2023-10-17 11:04:42,517 epoch 7 - iter 1188/1984 - loss 0.02559000 - time (sec): 55.51 - samples/sec: 1773.24 - lr: 0.000019 - momentum: 0.000000
167
+ 2023-10-17 11:04:51,705 epoch 7 - iter 1386/1984 - loss 0.02670608 - time (sec): 64.70 - samples/sec: 1773.77 - lr: 0.000018 - momentum: 0.000000
168
+ 2023-10-17 11:05:00,894 epoch 7 - iter 1584/1984 - loss 0.02761748 - time (sec): 73.89 - samples/sec: 1763.66 - lr: 0.000018 - momentum: 0.000000
169
+ 2023-10-17 11:05:10,013 epoch 7 - iter 1782/1984 - loss 0.02914139 - time (sec): 83.01 - samples/sec: 1776.01 - lr: 0.000017 - momentum: 0.000000
170
+ 2023-10-17 11:05:19,040 epoch 7 - iter 1980/1984 - loss 0.02930128 - time (sec): 92.03 - samples/sec: 1778.63 - lr: 0.000017 - momentum: 0.000000
171
+ 2023-10-17 11:05:19,223 ----------------------------------------------------------------------------------------------------
172
+ 2023-10-17 11:05:19,223 EPOCH 7 done: loss 0.0292 - lr: 0.000017
173
+ 2023-10-17 11:05:23,023 DEV : loss 0.2106274515390396 - f1-score (micro avg) 0.7652
174
+ 2023-10-17 11:05:23,044 ----------------------------------------------------------------------------------------------------
175
+ 2023-10-17 11:05:32,176 epoch 8 - iter 198/1984 - loss 0.01532135 - time (sec): 9.13 - samples/sec: 1792.67 - lr: 0.000016 - momentum: 0.000000
176
+ 2023-10-17 11:05:41,290 epoch 8 - iter 396/1984 - loss 0.01844700 - time (sec): 18.24 - samples/sec: 1775.71 - lr: 0.000016 - momentum: 0.000000
177
+ 2023-10-17 11:05:50,631 epoch 8 - iter 594/1984 - loss 0.01680914 - time (sec): 27.59 - samples/sec: 1813.29 - lr: 0.000015 - momentum: 0.000000
178
+ 2023-10-17 11:05:59,572 epoch 8 - iter 792/1984 - loss 0.01532722 - time (sec): 36.53 - samples/sec: 1809.50 - lr: 0.000014 - momentum: 0.000000
179
+ 2023-10-17 11:06:08,828 epoch 8 - iter 990/1984 - loss 0.01651685 - time (sec): 45.78 - samples/sec: 1817.17 - lr: 0.000014 - momentum: 0.000000
180
+ 2023-10-17 11:06:17,849 epoch 8 - iter 1188/1984 - loss 0.01693879 - time (sec): 54.80 - samples/sec: 1823.06 - lr: 0.000013 - momentum: 0.000000
181
+ 2023-10-17 11:06:26,896 epoch 8 - iter 1386/1984 - loss 0.01791079 - time (sec): 63.85 - samples/sec: 1810.51 - lr: 0.000013 - momentum: 0.000000
182
+ 2023-10-17 11:06:35,832 epoch 8 - iter 1584/1984 - loss 0.01862913 - time (sec): 72.79 - samples/sec: 1794.24 - lr: 0.000012 - momentum: 0.000000
183
+ 2023-10-17 11:06:44,860 epoch 8 - iter 1782/1984 - loss 0.01920153 - time (sec): 81.81 - samples/sec: 1799.31 - lr: 0.000012 - momentum: 0.000000
184
+ 2023-10-17 11:06:53,979 epoch 8 - iter 1980/1984 - loss 0.01890437 - time (sec): 90.93 - samples/sec: 1799.51 - lr: 0.000011 - momentum: 0.000000
185
+ 2023-10-17 11:06:54,163 ----------------------------------------------------------------------------------------------------
186
+ 2023-10-17 11:06:54,163 EPOCH 8 done: loss 0.0189 - lr: 0.000011
187
+ 2023-10-17 11:06:57,535 DEV : loss 0.25147587060928345 - f1-score (micro avg) 0.7642
188
+ 2023-10-17 11:06:57,556 ----------------------------------------------------------------------------------------------------
189
+ 2023-10-17 11:07:06,501 epoch 9 - iter 198/1984 - loss 0.01800551 - time (sec): 8.94 - samples/sec: 1771.15 - lr: 0.000011 - momentum: 0.000000
190
+ 2023-10-17 11:07:15,377 epoch 9 - iter 396/1984 - loss 0.01378590 - time (sec): 17.82 - samples/sec: 1863.42 - lr: 0.000010 - momentum: 0.000000
191
+ 2023-10-17 11:07:24,608 epoch 9 - iter 594/1984 - loss 0.01314436 - time (sec): 27.05 - samples/sec: 1863.71 - lr: 0.000009 - momentum: 0.000000
192
+ 2023-10-17 11:07:33,694 epoch 9 - iter 792/1984 - loss 0.01248684 - time (sec): 36.14 - samples/sec: 1845.31 - lr: 0.000009 - momentum: 0.000000
193
+ 2023-10-17 11:07:42,786 epoch 9 - iter 990/1984 - loss 0.01186486 - time (sec): 45.23 - samples/sec: 1825.45 - lr: 0.000008 - momentum: 0.000000
194
+ 2023-10-17 11:07:51,933 epoch 9 - iter 1188/1984 - loss 0.01172450 - time (sec): 54.38 - samples/sec: 1815.57 - lr: 0.000008 - momentum: 0.000000
195
+ 2023-10-17 11:08:01,131 epoch 9 - iter 1386/1984 - loss 0.01171225 - time (sec): 63.57 - samples/sec: 1805.34 - lr: 0.000007 - momentum: 0.000000
196
+ 2023-10-17 11:08:10,295 epoch 9 - iter 1584/1984 - loss 0.01234481 - time (sec): 72.74 - samples/sec: 1808.45 - lr: 0.000007 - momentum: 0.000000
197
+ 2023-10-17 11:08:19,267 epoch 9 - iter 1782/1984 - loss 0.01340571 - time (sec): 81.71 - samples/sec: 1807.81 - lr: 0.000006 - momentum: 0.000000
198
+ 2023-10-17 11:08:28,340 epoch 9 - iter 1980/1984 - loss 0.01367155 - time (sec): 90.78 - samples/sec: 1803.06 - lr: 0.000006 - momentum: 0.000000
199
+ 2023-10-17 11:08:28,512 ----------------------------------------------------------------------------------------------------
200
+ 2023-10-17 11:08:28,512 EPOCH 9 done: loss 0.0136 - lr: 0.000006
201
+ 2023-10-17 11:08:31,940 DEV : loss 0.23887786269187927 - f1-score (micro avg) 0.7643
202
+ 2023-10-17 11:08:31,962 ----------------------------------------------------------------------------------------------------
203
+ 2023-10-17 11:08:40,850 epoch 10 - iter 198/1984 - loss 0.00470077 - time (sec): 8.89 - samples/sec: 1858.08 - lr: 0.000005 - momentum: 0.000000
204
+ 2023-10-17 11:08:49,392 epoch 10 - iter 396/1984 - loss 0.00422202 - time (sec): 17.43 - samples/sec: 1844.72 - lr: 0.000004 - momentum: 0.000000
205
+ 2023-10-17 11:08:58,007 epoch 10 - iter 594/1984 - loss 0.00543694 - time (sec): 26.04 - samples/sec: 1866.92 - lr: 0.000004 - momentum: 0.000000
206
+ 2023-10-17 11:09:06,736 epoch 10 - iter 792/1984 - loss 0.00646607 - time (sec): 34.77 - samples/sec: 1863.80 - lr: 0.000003 - momentum: 0.000000
207
+ 2023-10-17 11:09:15,340 epoch 10 - iter 990/1984 - loss 0.00650601 - time (sec): 43.38 - samples/sec: 1857.52 - lr: 0.000003 - momentum: 0.000000
208
+ 2023-10-17 11:09:23,989 epoch 10 - iter 1188/1984 - loss 0.00670870 - time (sec): 52.03 - samples/sec: 1873.09 - lr: 0.000002 - momentum: 0.000000
209
+ 2023-10-17 11:09:32,785 epoch 10 - iter 1386/1984 - loss 0.00698514 - time (sec): 60.82 - samples/sec: 1884.69 - lr: 0.000002 - momentum: 0.000000
210
+ 2023-10-17 11:09:41,466 epoch 10 - iter 1584/1984 - loss 0.00711906 - time (sec): 69.50 - samples/sec: 1886.06 - lr: 0.000001 - momentum: 0.000000
211
+ 2023-10-17 11:09:50,532 epoch 10 - iter 1782/1984 - loss 0.00702130 - time (sec): 78.57 - samples/sec: 1870.40 - lr: 0.000001 - momentum: 0.000000
212
+ 2023-10-17 11:09:59,828 epoch 10 - iter 1980/1984 - loss 0.00764656 - time (sec): 87.86 - samples/sec: 1863.19 - lr: 0.000000 - momentum: 0.000000
213
+ 2023-10-17 11:10:00,016 ----------------------------------------------------------------------------------------------------
214
+ 2023-10-17 11:10:00,016 EPOCH 10 done: loss 0.0076 - lr: 0.000000
215
+ 2023-10-17 11:10:03,438 DEV : loss 0.2601605951786041 - f1-score (micro avg) 0.7684
216
+ 2023-10-17 11:10:03,459 saving best model
217
+ 2023-10-17 11:10:04,382 ----------------------------------------------------------------------------------------------------
218
+ 2023-10-17 11:10:04,383 Loading model from best epoch ...
219
+ 2023-10-17 11:10:05,772 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
220
+ 2023-10-17 11:10:08,979
221
+ Results:
222
+ - F-score (micro) 0.7792
223
+ - F-score (macro) 0.6974
224
+ - Accuracy 0.6658
225
+
226
+ By class:
227
+ precision recall f1-score support
228
+
229
+ LOC 0.8303 0.8443 0.8372 655
230
+ PER 0.7028 0.7848 0.7415 223
231
+ ORG 0.6000 0.4488 0.5135 127
232
+
233
+ micro avg 0.7772 0.7811 0.7792 1005
234
+ macro avg 0.7110 0.6926 0.6974 1005
235
+ weighted avg 0.7729 0.7811 0.7751 1005
236
+
237
+ 2023-10-17 11:10:08,979 ----------------------------------------------------------------------------------------------------