Hugging Face's logo

stefan-it
/
hmbench-icdar-fr-hmbert-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3

Token Classification
Flair
PyTorch
French
sequence-tagger-model
Model card Files Community
1
stefan-it commited on
Commit
58a38b8
·
1 Parent(s): c7cf99a

Upload folder using huggingface_hub

Browse files
Files changed (5) hide show
  1. best-model.pt +3 -0
  2. dev.tsv +0 -0
  3. loss.tsv +11 -0
  4. test.tsv +0 -0
  5. training.log +242 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c67017bd3ec2300cd184067f1a28f8fbc7306d13ab24880755d9caffcf671885
3
+ size 443311111
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 23:28:56 0.0000 0.3509 0.0934 0.6601 0.7602 0.7066 0.5690
3
+ 2 23:29:58 0.0000 0.1053 0.0894 0.7198 0.7557 0.7373 0.5996
4
+ 3 23:31:02 0.0000 0.0730 0.1066 0.7214 0.7557 0.7381 0.6034
5
+ 4 23:32:05 0.0000 0.0530 0.1528 0.7147 0.7681 0.7405 0.6128
6
+ 5 23:33:08 0.0000 0.0424 0.1442 0.7086 0.7704 0.7382 0.6021
7
+ 6 23:34:11 0.0000 0.0315 0.1808 0.7371 0.7579 0.7474 0.6204
8
+ 7 23:35:15 0.0000 0.0236 0.1939 0.7370 0.7704 0.7533 0.6259
9
+ 8 23:36:18 0.0000 0.0179 0.2065 0.7327 0.7658 0.7489 0.6194
10
+ 9 23:37:20 0.0000 0.0122 0.2183 0.7348 0.7681 0.7511 0.6224
11
+ 10 23:38:23 0.0000 0.0086 0.2139 0.7417 0.7794 0.7601 0.6310
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,242 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-13 23:27:53,960 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-13 23:27:53,961 Model: "SequenceTagger(
3
+ (embeddings): TransformerWordEmbeddings(
4
+ (model): BertModel(
5
+ (embeddings): BertEmbeddings(
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): 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=13, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-13 23:27:53,961 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-13 23:27:53,961 MultiCorpus: 7936 train + 992 dev + 992 test sentences
52
+ - NER_ICDAR_EUROPEANA Corpus: 7936 train + 992 dev + 992 test sentences - /root/.flair/datasets/ner_icdar_europeana/fr
53
+ 2023-10-13 23:27:53,961 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-13 23:27:53,961 Train: 7936 sentences
55
+ 2023-10-13 23:27:53,961 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-13 23:27:53,961 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-13 23:27:53,961 Training Params:
58
+ 2023-10-13 23:27:53,961 - learning_rate: "5e-05"
59
+ 2023-10-13 23:27:53,961 - mini_batch_size: "8"
60
+ 2023-10-13 23:27:53,961 - max_epochs: "10"
61
+ 2023-10-13 23:27:53,961 - shuffle: "True"
62
+ 2023-10-13 23:27:53,961 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-13 23:27:53,961 Plugins:
64
+ 2023-10-13 23:27:53,961 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-13 23:27:53,961 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-13 23:27:53,961 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-13 23:27:53,961 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-13 23:27:53,961 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-13 23:27:53,961 Computation:
70
+ 2023-10-13 23:27:53,961 - compute on device: cuda:0
71
+ 2023-10-13 23:27:53,962 - embedding storage: none
72
+ 2023-10-13 23:27:53,962 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-13 23:27:53,962 Model training base path: "hmbench-icdar/fr-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3"
74
+ 2023-10-13 23:27:53,962 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-13 23:27:53,962 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-13 23:28:00,314 epoch 1 - iter 99/992 - loss 1.84420524 - time (sec): 6.35 - samples/sec: 2623.00 - lr: 0.000005 - momentum: 0.000000
77
+ 2023-10-13 23:28:06,201 epoch 1 - iter 198/992 - loss 1.10827375 - time (sec): 12.24 - samples/sec: 2677.78 - lr: 0.000010 - momentum: 0.000000
78
+ 2023-10-13 23:28:12,435 epoch 1 - iter 297/992 - loss 0.80875373 - time (sec): 18.47 - samples/sec: 2680.34 - lr: 0.000015 - momentum: 0.000000
79
+ 2023-10-13 23:28:18,266 epoch 1 - iter 396/992 - loss 0.66063021 - time (sec): 24.30 - samples/sec: 2704.15 - lr: 0.000020 - momentum: 0.000000
80
+ 2023-10-13 23:28:24,336 epoch 1 - iter 495/992 - loss 0.56248912 - time (sec): 30.37 - samples/sec: 2713.70 - lr: 0.000025 - momentum: 0.000000
81
+ 2023-10-13 23:28:29,967 epoch 1 - iter 594/992 - loss 0.49232335 - time (sec): 36.00 - samples/sec: 2749.86 - lr: 0.000030 - momentum: 0.000000
82
+ 2023-10-13 23:28:35,712 epoch 1 - iter 693/992 - loss 0.44224564 - time (sec): 41.75 - samples/sec: 2760.11 - lr: 0.000035 - momentum: 0.000000
83
+ 2023-10-13 23:28:41,632 epoch 1 - iter 792/992 - loss 0.40489892 - time (sec): 47.67 - samples/sec: 2764.55 - lr: 0.000040 - momentum: 0.000000
84
+ 2023-10-13 23:28:47,351 epoch 1 - iter 891/992 - loss 0.37416440 - time (sec): 53.39 - samples/sec: 2768.49 - lr: 0.000045 - momentum: 0.000000
85
+ 2023-10-13 23:28:53,006 epoch 1 - iter 990/992 - loss 0.35114103 - time (sec): 59.04 - samples/sec: 2773.46 - lr: 0.000050 - momentum: 0.000000
86
+ 2023-10-13 23:28:53,112 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-13 23:28:53,112 EPOCH 1 done: loss 0.3509 - lr: 0.000050
88
+ 2023-10-13 23:28:56,206 DEV : loss 0.0934109166264534 - f1-score (micro avg) 0.7066
89
+ 2023-10-13 23:28:56,226 saving best model
90
+ 2023-10-13 23:28:56,661 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-13 23:29:02,883 epoch 2 - iter 99/992 - loss 0.12408130 - time (sec): 6.22 - samples/sec: 2830.88 - lr: 0.000049 - momentum: 0.000000
92
+ 2023-10-13 23:29:09,014 epoch 2 - iter 198/992 - loss 0.11801383 - time (sec): 12.35 - samples/sec: 2720.96 - lr: 0.000049 - momentum: 0.000000
93
+ 2023-10-13 23:29:14,757 epoch 2 - iter 297/992 - loss 0.11422384 - time (sec): 18.09 - samples/sec: 2769.68 - lr: 0.000048 - momentum: 0.000000
94
+ 2023-10-13 23:29:20,479 epoch 2 - iter 396/992 - loss 0.11197551 - time (sec): 23.82 - samples/sec: 2753.22 - lr: 0.000048 - momentum: 0.000000
95
+ 2023-10-13 23:29:26,220 epoch 2 - iter 495/992 - loss 0.11065679 - time (sec): 29.56 - samples/sec: 2763.91 - lr: 0.000047 - momentum: 0.000000
96
+ 2023-10-13 23:29:32,096 epoch 2 - iter 594/992 - loss 0.10893379 - time (sec): 35.43 - samples/sec: 2759.56 - lr: 0.000047 - momentum: 0.000000
97
+ 2023-10-13 23:29:37,819 epoch 2 - iter 693/992 - loss 0.10716282 - time (sec): 41.16 - samples/sec: 2765.86 - lr: 0.000046 - momentum: 0.000000
98
+ 2023-10-13 23:29:43,750 epoch 2 - iter 792/992 - loss 0.10687508 - time (sec): 47.09 - samples/sec: 2774.79 - lr: 0.000046 - momentum: 0.000000
99
+ 2023-10-13 23:29:49,391 epoch 2 - iter 891/992 - loss 0.10650512 - time (sec): 52.73 - samples/sec: 2772.57 - lr: 0.000045 - momentum: 0.000000
100
+ 2023-10-13 23:29:55,474 epoch 2 - iter 990/992 - loss 0.10524813 - time (sec): 58.81 - samples/sec: 2784.28 - lr: 0.000044 - momentum: 0.000000
101
+ 2023-10-13 23:29:55,580 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-13 23:29:55,580 EPOCH 2 done: loss 0.1053 - lr: 0.000044
103
+ 2023-10-13 23:29:58,941 DEV : loss 0.0894036665558815 - f1-score (micro avg) 0.7373
104
+ 2023-10-13 23:29:58,961 saving best model
105
+ 2023-10-13 23:29:59,934 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-13 23:30:05,864 epoch 3 - iter 99/992 - loss 0.06449001 - time (sec): 5.93 - samples/sec: 2757.86 - lr: 0.000044 - momentum: 0.000000
107
+ 2023-10-13 23:30:11,666 epoch 3 - iter 198/992 - loss 0.06269725 - time (sec): 11.73 - samples/sec: 2753.00 - lr: 0.000043 - momentum: 0.000000
108
+ 2023-10-13 23:30:17,362 epoch 3 - iter 297/992 - loss 0.06703869 - time (sec): 17.42 - samples/sec: 2775.83 - lr: 0.000043 - momentum: 0.000000
109
+ 2023-10-13 23:30:23,235 epoch 3 - iter 396/992 - loss 0.06854271 - time (sec): 23.30 - samples/sec: 2803.73 - lr: 0.000042 - momentum: 0.000000
110
+ 2023-10-13 23:30:29,364 epoch 3 - iter 495/992 - loss 0.07038135 - time (sec): 29.43 - samples/sec: 2792.79 - lr: 0.000042 - momentum: 0.000000
111
+ 2023-10-13 23:30:35,163 epoch 3 - iter 594/992 - loss 0.07075691 - time (sec): 35.22 - samples/sec: 2788.96 - lr: 0.000041 - momentum: 0.000000
112
+ 2023-10-13 23:30:41,272 epoch 3 - iter 693/992 - loss 0.07173481 - time (sec): 41.33 - samples/sec: 2773.31 - lr: 0.000041 - momentum: 0.000000
113
+ 2023-10-13 23:30:47,040 epoch 3 - iter 792/992 - loss 0.07257445 - time (sec): 47.10 - samples/sec: 2771.06 - lr: 0.000040 - momentum: 0.000000
114
+ 2023-10-13 23:30:52,852 epoch 3 - iter 891/992 - loss 0.07399709 - time (sec): 52.91 - samples/sec: 2775.77 - lr: 0.000039 - momentum: 0.000000
115
+ 2023-10-13 23:30:58,651 epoch 3 - iter 990/992 - loss 0.07304166 - time (sec): 58.71 - samples/sec: 2787.44 - lr: 0.000039 - momentum: 0.000000
116
+ 2023-10-13 23:30:58,762 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-13 23:30:58,762 EPOCH 3 done: loss 0.0730 - lr: 0.000039
118
+ 2023-10-13 23:31:02,203 DEV : loss 0.10655300319194794 - f1-score (micro avg) 0.7381
119
+ 2023-10-13 23:31:02,224 saving best model
120
+ 2023-10-13 23:31:02,734 ----------------------------------------------------------------------------------------------------
121
+ 2023-10-13 23:31:08,758 epoch 4 - iter 99/992 - loss 0.04574350 - time (sec): 6.02 - samples/sec: 2832.57 - lr: 0.000038 - momentum: 0.000000
122
+ 2023-10-13 23:31:14,898 epoch 4 - iter 198/992 - loss 0.04764950 - time (sec): 12.16 - samples/sec: 2726.71 - lr: 0.000038 - momentum: 0.000000
123
+ 2023-10-13 23:31:20,551 epoch 4 - iter 297/992 - loss 0.04943979 - time (sec): 17.81 - samples/sec: 2747.51 - lr: 0.000037 - momentum: 0.000000
124
+ 2023-10-13 23:31:26,202 epoch 4 - iter 396/992 - loss 0.05038291 - time (sec): 23.46 - samples/sec: 2783.86 - lr: 0.000037 - momentum: 0.000000
125
+ 2023-10-13 23:31:32,240 epoch 4 - iter 495/992 - loss 0.05076880 - time (sec): 29.50 - samples/sec: 2764.50 - lr: 0.000036 - momentum: 0.000000
126
+ 2023-10-13 23:31:38,242 epoch 4 - iter 594/992 - loss 0.05156694 - time (sec): 35.50 - samples/sec: 2763.16 - lr: 0.000036 - momentum: 0.000000
127
+ 2023-10-13 23:31:43,936 epoch 4 - iter 693/992 - loss 0.05246850 - time (sec): 41.19 - samples/sec: 2782.77 - lr: 0.000035 - momentum: 0.000000
128
+ 2023-10-13 23:31:49,696 epoch 4 - iter 792/992 - loss 0.05316471 - time (sec): 46.95 - samples/sec: 2783.75 - lr: 0.000034 - momentum: 0.000000
129
+ 2023-10-13 23:31:55,681 epoch 4 - iter 891/992 - loss 0.05363958 - time (sec): 52.94 - samples/sec: 2787.31 - lr: 0.000034 - momentum: 0.000000
130
+ 2023-10-13 23:32:01,604 epoch 4 - iter 990/992 - loss 0.05300719 - time (sec): 58.86 - samples/sec: 2783.74 - lr: 0.000033 - momentum: 0.000000
131
+ 2023-10-13 23:32:01,704 ----------------------------------------------------------------------------------------------------
132
+ 2023-10-13 23:32:01,705 EPOCH 4 done: loss 0.0530 - lr: 0.000033
133
+ 2023-10-13 23:32:05,121 DEV : loss 0.15282906591892242 - f1-score (micro avg) 0.7405
134
+ 2023-10-13 23:32:05,142 saving best model
135
+ 2023-10-13 23:32:05,677 ----------------------------------------------------------------------------------------------------
136
+ 2023-10-13 23:32:11,818 epoch 5 - iter 99/992 - loss 0.03701588 - time (sec): 6.14 - samples/sec: 2763.56 - lr: 0.000033 - momentum: 0.000000
137
+ 2023-10-13 23:32:17,742 epoch 5 - iter 198/992 - loss 0.03627021 - time (sec): 12.06 - samples/sec: 2740.35 - lr: 0.000032 - momentum: 0.000000
138
+ 2023-10-13 23:32:24,414 epoch 5 - iter 297/992 - loss 0.03801493 - time (sec): 18.73 - samples/sec: 2692.51 - lr: 0.000032 - momentum: 0.000000
139
+ 2023-10-13 23:32:30,313 epoch 5 - iter 396/992 - loss 0.04018940 - time (sec): 24.63 - samples/sec: 2697.06 - lr: 0.000031 - momentum: 0.000000
140
+ 2023-10-13 23:32:36,272 epoch 5 - iter 495/992 - loss 0.03982197 - time (sec): 30.59 - samples/sec: 2710.43 - lr: 0.000031 - momentum: 0.000000
141
+ 2023-10-13 23:32:42,145 epoch 5 - iter 594/992 - loss 0.04135012 - time (sec): 36.46 - samples/sec: 2712.40 - lr: 0.000030 - momentum: 0.000000
142
+ 2023-10-13 23:32:47,697 epoch 5 - iter 693/992 - loss 0.04198723 - time (sec): 42.02 - samples/sec: 2723.82 - lr: 0.000029 - momentum: 0.000000
143
+ 2023-10-13 23:32:53,463 epoch 5 - iter 792/992 - loss 0.04154554 - time (sec): 47.78 - samples/sec: 2733.14 - lr: 0.000029 - momentum: 0.000000
144
+ 2023-10-13 23:32:59,465 epoch 5 - iter 891/992 - loss 0.04180375 - time (sec): 53.78 - samples/sec: 2751.31 - lr: 0.000028 - momentum: 0.000000
145
+ 2023-10-13 23:33:05,084 epoch 5 - iter 990/992 - loss 0.04232641 - time (sec): 59.40 - samples/sec: 2752.10 - lr: 0.000028 - momentum: 0.000000
146
+ 2023-10-13 23:33:05,307 ----------------------------------------------------------------------------------------------------
147
+ 2023-10-13 23:33:05,307 EPOCH 5 done: loss 0.0424 - lr: 0.000028
148
+ 2023-10-13 23:33:08,786 DEV : loss 0.14422008395195007 - f1-score (micro avg) 0.7382
149
+ 2023-10-13 23:33:08,807 ----------------------------------------------------------------------------------------------------
150
+ 2023-10-13 23:33:14,515 epoch 6 - iter 99/992 - loss 0.03270406 - time (sec): 5.71 - samples/sec: 2753.34 - lr: 0.000027 - momentum: 0.000000
151
+ 2023-10-13 23:33:20,341 epoch 6 - iter 198/992 - loss 0.03440711 - time (sec): 11.53 - samples/sec: 2762.06 - lr: 0.000027 - momentum: 0.000000
152
+ 2023-10-13 23:33:26,151 epoch 6 - iter 297/992 - loss 0.03164332 - time (sec): 17.34 - samples/sec: 2786.75 - lr: 0.000026 - momentum: 0.000000
153
+ 2023-10-13 23:33:32,289 epoch 6 - iter 396/992 - loss 0.03087269 - time (sec): 23.48 - samples/sec: 2774.22 - lr: 0.000026 - momentum: 0.000000
154
+ 2023-10-13 23:33:38,140 epoch 6 - iter 495/992 - loss 0.03029381 - time (sec): 29.33 - samples/sec: 2779.29 - lr: 0.000025 - momentum: 0.000000
155
+ 2023-10-13 23:33:44,152 epoch 6 - iter 594/992 - loss 0.03194290 - time (sec): 35.34 - samples/sec: 2771.66 - lr: 0.000024 - momentum: 0.000000
156
+ 2023-10-13 23:33:50,524 epoch 6 - iter 693/992 - loss 0.03148955 - time (sec): 41.72 - samples/sec: 2751.98 - lr: 0.000024 - momentum: 0.000000
157
+ 2023-10-13 23:33:56,555 epoch 6 - iter 792/992 - loss 0.03151560 - time (sec): 47.75 - samples/sec: 2745.14 - lr: 0.000023 - momentum: 0.000000
158
+ 2023-10-13 23:34:02,346 epoch 6 - iter 891/992 - loss 0.03159422 - time (sec): 53.54 - samples/sec: 2737.07 - lr: 0.000023 - momentum: 0.000000
159
+ 2023-10-13 23:34:08,209 epoch 6 - iter 990/992 - loss 0.03160206 - time (sec): 59.40 - samples/sec: 2752.33 - lr: 0.000022 - momentum: 0.000000
160
+ 2023-10-13 23:34:08,344 ----------------------------------------------------------------------------------------------------
161
+ 2023-10-13 23:34:08,344 EPOCH 6 done: loss 0.0315 - lr: 0.000022
162
+ 2023-10-13 23:34:11,797 DEV : loss 0.18077921867370605 - f1-score (micro avg) 0.7474
163
+ 2023-10-13 23:34:11,818 saving best model
164
+ 2023-10-13 23:34:12,359 ----------------------------------------------------------------------------------------------------
165
+ 2023-10-13 23:34:18,498 epoch 7 - iter 99/992 - loss 0.02270269 - time (sec): 6.14 - samples/sec: 2795.67 - lr: 0.000022 - momentum: 0.000000
166
+ 2023-10-13 23:34:24,910 epoch 7 - iter 198/992 - loss 0.02677230 - time (sec): 12.55 - samples/sec: 2687.84 - lr: 0.000021 - momentum: 0.000000
167
+ 2023-10-13 23:34:30,704 epoch 7 - iter 297/992 - loss 0.02627154 - time (sec): 18.34 - samples/sec: 2745.71 - lr: 0.000021 - momentum: 0.000000
168
+ 2023-10-13 23:34:36,691 epoch 7 - iter 396/992 - loss 0.02486597 - time (sec): 24.33 - samples/sec: 2749.78 - lr: 0.000020 - momentum: 0.000000
169
+ 2023-10-13 23:34:42,564 epoch 7 - iter 495/992 - loss 0.02427333 - time (sec): 30.20 - samples/sec: 2755.90 - lr: 0.000019 - momentum: 0.000000
170
+ 2023-10-13 23:34:48,585 epoch 7 - iter 594/992 - loss 0.02499253 - time (sec): 36.22 - samples/sec: 2771.56 - lr: 0.000019 - momentum: 0.000000
171
+ 2023-10-13 23:34:54,382 epoch 7 - iter 693/992 - loss 0.02475078 - time (sec): 42.02 - samples/sec: 2755.90 - lr: 0.000018 - momentum: 0.000000
172
+ 2023-10-13 23:35:00,122 epoch 7 - iter 792/992 - loss 0.02346216 - time (sec): 47.76 - samples/sec: 2761.23 - lr: 0.000018 - momentum: 0.000000
173
+ 2023-10-13 23:35:05,982 epoch 7 - iter 891/992 - loss 0.02381140 - time (sec): 53.62 - samples/sec: 2764.08 - lr: 0.000017 - momentum: 0.000000
174
+ 2023-10-13 23:35:11,552 epoch 7 - iter 990/992 - loss 0.02363118 - time (sec): 59.19 - samples/sec: 2764.57 - lr: 0.000017 - momentum: 0.000000
175
+ 2023-10-13 23:35:11,659 ----------------------------------------------------------------------------------------------------
176
+ 2023-10-13 23:35:11,659 EPOCH 7 done: loss 0.0236 - lr: 0.000017
177
+ 2023-10-13 23:35:15,132 DEV : loss 0.19388847053050995 - f1-score (micro avg) 0.7533
178
+ 2023-10-13 23:35:15,154 saving best model
179
+ 2023-10-13 23:35:15,701 ----------------------------------------------------------------------------------------------------
180
+ 2023-10-13 23:35:21,741 epoch 8 - iter 99/992 - loss 0.02606675 - time (sec): 6.03 - samples/sec: 2751.12 - lr: 0.000016 - momentum: 0.000000
181
+ 2023-10-13 23:35:27,536 epoch 8 - iter 198/992 - loss 0.02087024 - time (sec): 11.82 - samples/sec: 2773.87 - lr: 0.000016 - momentum: 0.000000
182
+ 2023-10-13 23:35:33,595 epoch 8 - iter 297/992 - loss 0.02083409 - time (sec): 17.88 - samples/sec: 2756.05 - lr: 0.000015 - momentum: 0.000000
183
+ 2023-10-13 23:35:39,370 epoch 8 - iter 396/992 - loss 0.01959564 - time (sec): 23.66 - samples/sec: 2759.52 - lr: 0.000014 - momentum: 0.000000
184
+ 2023-10-13 23:35:45,228 epoch 8 - iter 495/992 - loss 0.01772775 - time (sec): 29.51 - samples/sec: 2753.18 - lr: 0.000014 - momentum: 0.000000
185
+ 2023-10-13 23:35:51,290 epoch 8 - iter 594/992 - loss 0.01876618 - time (sec): 35.58 - samples/sec: 2755.48 - lr: 0.000013 - momentum: 0.000000
186
+ 2023-10-13 23:35:57,441 epoch 8 - iter 693/992 - loss 0.01844572 - time (sec): 41.73 - samples/sec: 2754.59 - lr: 0.000013 - momentum: 0.000000
187
+ 2023-10-13 23:36:03,063 epoch 8 - iter 792/992 - loss 0.01837206 - time (sec): 47.35 - samples/sec: 2766.17 - lr: 0.000012 - momentum: 0.000000
188
+ 2023-10-13 23:36:08,882 epoch 8 - iter 891/992 - loss 0.01808332 - time (sec): 53.17 - samples/sec: 2764.84 - lr: 0.000012 - momentum: 0.000000
189
+ 2023-10-13 23:36:14,706 epoch 8 - iter 990/992 - loss 0.01781156 - time (sec): 58.99 - samples/sec: 2774.06 - lr: 0.000011 - momentum: 0.000000
190
+ 2023-10-13 23:36:14,815 ----------------------------------------------------------------------------------------------------
191
+ 2023-10-13 23:36:14,815 EPOCH 8 done: loss 0.0179 - lr: 0.000011
192
+ 2023-10-13 23:36:18,231 DEV : loss 0.2065057009458542 - f1-score (micro avg) 0.7489
193
+ 2023-10-13 23:36:18,252 ----------------------------------------------------------------------------------------------------
194
+ 2023-10-13 23:36:23,837 epoch 9 - iter 99/992 - loss 0.00810002 - time (sec): 5.58 - samples/sec: 2755.74 - lr: 0.000011 - momentum: 0.000000
195
+ 2023-10-13 23:36:29,544 epoch 9 - iter 198/992 - loss 0.01228895 - time (sec): 11.29 - samples/sec: 2762.22 - lr: 0.000010 - momentum: 0.000000
196
+ 2023-10-13 23:36:35,476 epoch 9 - iter 297/992 - loss 0.01142656 - time (sec): 17.22 - samples/sec: 2805.13 - lr: 0.000009 - momentum: 0.000000
197
+ 2023-10-13 23:36:41,470 epoch 9 - iter 396/992 - loss 0.01157152 - time (sec): 23.22 - samples/sec: 2800.36 - lr: 0.000009 - momentum: 0.000000
198
+ 2023-10-13 23:36:48,284 epoch 9 - iter 495/992 - loss 0.01250551 - time (sec): 30.03 - samples/sec: 2754.58 - lr: 0.000008 - momentum: 0.000000
199
+ 2023-10-13 23:36:54,144 epoch 9 - iter 594/992 - loss 0.01181795 - time (sec): 35.89 - samples/sec: 2761.00 - lr: 0.000008 - momentum: 0.000000
200
+ 2023-10-13 23:36:59,911 epoch 9 - iter 693/992 - loss 0.01134550 - time (sec): 41.66 - samples/sec: 2767.20 - lr: 0.000007 - momentum: 0.000000
201
+ 2023-10-13 23:37:05,693 epoch 9 - iter 792/992 - loss 0.01157164 - time (sec): 47.44 - samples/sec: 2773.18 - lr: 0.000007 - momentum: 0.000000
202
+ 2023-10-13 23:37:11,344 epoch 9 - iter 891/992 - loss 0.01261003 - time (sec): 53.09 - samples/sec: 2773.26 - lr: 0.000006 - momentum: 0.000000
203
+ 2023-10-13 23:37:17,361 epoch 9 - iter 990/992 - loss 0.01216585 - time (sec): 59.11 - samples/sec: 2769.37 - lr: 0.000006 - momentum: 0.000000
204
+ 2023-10-13 23:37:17,483 ----------------------------------------------------------------------------------------------------
205
+ 2023-10-13 23:37:17,483 EPOCH 9 done: loss 0.0122 - lr: 0.000006
206
+ 2023-10-13 23:37:20,901 DEV : loss 0.21828265488147736 - f1-score (micro avg) 0.7511
207
+ 2023-10-13 23:37:20,922 ----------------------------------------------------------------------------------------------------
208
+ 2023-10-13 23:37:26,797 epoch 10 - iter 99/992 - loss 0.00487076 - time (sec): 5.87 - samples/sec: 2621.15 - lr: 0.000005 - momentum: 0.000000
209
+ 2023-10-13 23:37:32,458 epoch 10 - iter 198/992 - loss 0.00499497 - time (sec): 11.53 - samples/sec: 2735.44 - lr: 0.000004 - momentum: 0.000000
210
+ 2023-10-13 23:37:38,187 epoch 10 - iter 297/992 - loss 0.00669737 - time (sec): 17.26 - samples/sec: 2734.38 - lr: 0.000004 - momentum: 0.000000
211
+ 2023-10-13 23:37:44,180 epoch 10 - iter 396/992 - loss 0.00712915 - time (sec): 23.26 - samples/sec: 2726.30 - lr: 0.000003 - momentum: 0.000000
212
+ 2023-10-13 23:37:49,921 epoch 10 - iter 495/992 - loss 0.00806367 - time (sec): 29.00 - samples/sec: 2749.29 - lr: 0.000003 - momentum: 0.000000
213
+ 2023-10-13 23:37:55,956 epoch 10 - iter 594/992 - loss 0.00780873 - time (sec): 35.03 - samples/sec: 2762.96 - lr: 0.000002 - momentum: 0.000000
214
+ 2023-10-13 23:38:01,974 epoch 10 - iter 693/992 - loss 0.00776724 - time (sec): 41.05 - samples/sec: 2789.12 - lr: 0.000002 - momentum: 0.000000
215
+ 2023-10-13 23:38:07,834 epoch 10 - iter 792/992 - loss 0.00771472 - time (sec): 46.91 - samples/sec: 2797.30 - lr: 0.000001 - momentum: 0.000000
216
+ 2023-10-13 23:38:13,576 epoch 10 - iter 891/992 - loss 0.00813066 - time (sec): 52.65 - samples/sec: 2798.27 - lr: 0.000001 - momentum: 0.000000
217
+ 2023-10-13 23:38:19,408 epoch 10 - iter 990/992 - loss 0.00863417 - time (sec): 58.48 - samples/sec: 2798.97 - lr: 0.000000 - momentum: 0.000000
218
+ 2023-10-13 23:38:19,516 ----------------------------------------------------------------------------------------------------
219
+ 2023-10-13 23:38:19,516 EPOCH 10 done: loss 0.0086 - lr: 0.000000
220
+ 2023-10-13 23:38:22,986 DEV : loss 0.2138698697090149 - f1-score (micro avg) 0.7601
221
+ 2023-10-13 23:38:23,006 saving best model
222
+ 2023-10-13 23:38:23,968 ----------------------------------------------------------------------------------------------------
223
+ 2023-10-13 23:38:23,969 Loading model from best epoch ...
224
+ 2023-10-13 23:38:25,336 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
225
+ 2023-10-13 23:38:28,578
226
+ Results:
227
+ - F-score (micro) 0.7934
228
+ - F-score (macro) 0.7137
229
+ - Accuracy 0.6822
230
+
231
+ By class:
232
+ precision recall f1-score support
233
+
234
+ LOC 0.8217 0.8794 0.8496 655
235
+ PER 0.7188 0.8251 0.7683 223
236
+ ORG 0.5636 0.4882 0.5232 127
237
+
238
+ micro avg 0.7704 0.8179 0.7934 1005
239
+ macro avg 0.7014 0.7309 0.7137 1005
240
+ weighted avg 0.7662 0.8179 0.7903 1005
241
+
242
+ 2023-10-13 23:38:28,579 ----------------------------------------------------------------------------------------------------