Hugging Face's logo

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

Token Classification
Flair
PyTorch
TensorBoard
French
sequence-tagger-model
Model card Files Metrics Community
1
stefan-it commited on
Commit
04b2386
·
1 Parent(s): c34f0c4

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.1697660802.46dc0c540dd0.3341.1 +3 -0
  5. test.tsv +0 -0
  6. training.log +245 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:282d14078571a9f7115a0264073e7926494675e4de75333f64c18caaaf1136fb
3
+ size 19045922
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 20:27:14 0.0000 0.8373 0.2124 0.3097 0.3020 0.3058 0.2027
3
+ 2 20:27:46 0.0000 0.2517 0.1658 0.3969 0.4310 0.4132 0.2880
4
+ 3 20:28:19 0.0000 0.2043 0.1485 0.5016 0.5249 0.5130 0.3769
5
+ 4 20:28:51 0.0000 0.1776 0.1476 0.5699 0.5441 0.5567 0.4172
6
+ 5 20:29:23 0.0000 0.1631 0.1393 0.5625 0.6109 0.5857 0.4485
7
+ 6 20:29:55 0.0000 0.1508 0.1444 0.5618 0.6063 0.5832 0.4459
8
+ 7 20:30:28 0.0000 0.1405 0.1465 0.5876 0.5803 0.5839 0.4457
9
+ 8 20:31:00 0.0000 0.1331 0.1468 0.5842 0.6120 0.5978 0.4604
10
+ 9 20:31:33 0.0000 0.1285 0.1488 0.5888 0.6075 0.5980 0.4602
11
+ 10 20:32:05 0.0000 0.1234 0.1480 0.5889 0.6143 0.6013 0.4629
runs/events.out.tfevents.1697660802.46dc0c540dd0.3341.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:089523743dd4143afc128e50f638f7e6d034ad0e98f7998b48b72b803b06080f
3
+ size 1108164
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,245 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-18 20:26:42,123 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-18 20:26:42,123 Model: "SequenceTagger(
3
+ (embeddings): TransformerWordEmbeddings(
4
+ (model): BertModel(
5
+ (embeddings): BertEmbeddings(
6
+ (word_embeddings): Embedding(32001, 128)
7
+ (position_embeddings): Embedding(512, 128)
8
+ (token_type_embeddings): Embedding(2, 128)
9
+ (LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
10
+ (dropout): Dropout(p=0.1, inplace=False)
11
+ )
12
+ (encoder): BertEncoder(
13
+ (layer): ModuleList(
14
+ (0-1): 2 x BertLayer(
15
+ (attention): BertAttention(
16
+ (self): BertSelfAttention(
17
+ (query): Linear(in_features=128, out_features=128, bias=True)
18
+ (key): Linear(in_features=128, out_features=128, bias=True)
19
+ (value): Linear(in_features=128, out_features=128, bias=True)
20
+ (dropout): Dropout(p=0.1, inplace=False)
21
+ )
22
+ (output): BertSelfOutput(
23
+ (dense): Linear(in_features=128, out_features=128, bias=True)
24
+ (LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
25
+ (dropout): Dropout(p=0.1, inplace=False)
26
+ )
27
+ )
28
+ (intermediate): BertIntermediate(
29
+ (dense): Linear(in_features=128, out_features=512, bias=True)
30
+ (intermediate_act_fn): GELUActivation()
31
+ )
32
+ (output): BertOutput(
33
+ (dense): Linear(in_features=512, out_features=128, bias=True)
34
+ (LayerNorm): LayerNorm((128,), 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=128, out_features=128, bias=True)
42
+ (activation): Tanh()
43
+ )
44
+ )
45
+ )
46
+ (locked_dropout): LockedDropout(p=0.5)
47
+ (linear): Linear(in_features=128, out_features=13, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-18 20:26:42,123 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-18 20:26:42,123 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
+ 2023-10-18 20:26:42,123 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-18 20:26:42,123 Train: 7936 sentences
55
+ 2023-10-18 20:26:42,123 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-18 20:26:42,123 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-18 20:26:42,123 Training Params:
58
+ 2023-10-18 20:26:42,123 - learning_rate: "5e-05"
59
+ 2023-10-18 20:26:42,123 - mini_batch_size: "4"
60
+ 2023-10-18 20:26:42,124 - max_epochs: "10"
61
+ 2023-10-18 20:26:42,124 - shuffle: "True"
62
+ 2023-10-18 20:26:42,124 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-18 20:26:42,124 Plugins:
64
+ 2023-10-18 20:26:42,124 - TensorboardLogger
65
+ 2023-10-18 20:26:42,124 - LinearScheduler | warmup_fraction: '0.1'
66
+ 2023-10-18 20:26:42,124 ----------------------------------------------------------------------------------------------------
67
+ 2023-10-18 20:26:42,124 Final evaluation on model from best epoch (best-model.pt)
68
+ 2023-10-18 20:26:42,124 - metric: "('micro avg', 'f1-score')"
69
+ 2023-10-18 20:26:42,124 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-18 20:26:42,124 Computation:
71
+ 2023-10-18 20:26:42,124 - compute on device: cuda:0
72
+ 2023-10-18 20:26:42,124 - embedding storage: none
73
+ 2023-10-18 20:26:42,124 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-18 20:26:42,124 Model training base path: "hmbench-icdar/fr-dbmdz/bert-tiny-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1"
75
+ 2023-10-18 20:26:42,124 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-18 20:26:42,124 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-18 20:26:42,124 Logging anything other than scalars to TensorBoard is currently not supported.
78
+ 2023-10-18 20:26:45,221 epoch 1 - iter 198/1984 - loss 3.10528833 - time (sec): 3.10 - samples/sec: 5261.08 - lr: 0.000005 - momentum: 0.000000
79
+ 2023-10-18 20:26:48,238 epoch 1 - iter 396/1984 - loss 2.51418141 - time (sec): 6.11 - samples/sec: 5312.12 - lr: 0.000010 - momentum: 0.000000
80
+ 2023-10-18 20:26:51,300 epoch 1 - iter 594/1984 - loss 1.89531636 - time (sec): 9.18 - samples/sec: 5362.40 - lr: 0.000015 - momentum: 0.000000
81
+ 2023-10-18 20:26:54,272 epoch 1 - iter 792/1984 - loss 1.54862823 - time (sec): 12.15 - samples/sec: 5412.02 - lr: 0.000020 - momentum: 0.000000
82
+ 2023-10-18 20:26:57,272 epoch 1 - iter 990/1984 - loss 1.33496354 - time (sec): 15.15 - samples/sec: 5376.01 - lr: 0.000025 - momentum: 0.000000
83
+ 2023-10-18 20:27:00,328 epoch 1 - iter 1188/1984 - loss 1.17925593 - time (sec): 18.20 - samples/sec: 5378.00 - lr: 0.000030 - momentum: 0.000000
84
+ 2023-10-18 20:27:03,304 epoch 1 - iter 1386/1984 - loss 1.06364809 - time (sec): 21.18 - samples/sec: 5382.30 - lr: 0.000035 - momentum: 0.000000
85
+ 2023-10-18 20:27:06,316 epoch 1 - iter 1584/1984 - loss 0.97153194 - time (sec): 24.19 - samples/sec: 5413.52 - lr: 0.000040 - momentum: 0.000000
86
+ 2023-10-18 20:27:09,339 epoch 1 - iter 1782/1984 - loss 0.89943309 - time (sec): 27.21 - samples/sec: 5406.86 - lr: 0.000045 - momentum: 0.000000
87
+ 2023-10-18 20:27:12,376 epoch 1 - iter 1980/1984 - loss 0.83845518 - time (sec): 30.25 - samples/sec: 5408.71 - lr: 0.000050 - momentum: 0.000000
88
+ 2023-10-18 20:27:12,437 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-18 20:27:12,437 EPOCH 1 done: loss 0.8373 - lr: 0.000050
90
+ 2023-10-18 20:27:14,301 DEV : loss 0.2123669981956482 - f1-score (micro avg) 0.3058
91
+ 2023-10-18 20:27:14,319 saving best model
92
+ 2023-10-18 20:27:14,350 ----------------------------------------------------------------------------------------------------
93
+ 2023-10-18 20:27:17,415 epoch 2 - iter 198/1984 - loss 0.30751315 - time (sec): 3.06 - samples/sec: 5547.01 - lr: 0.000049 - momentum: 0.000000
94
+ 2023-10-18 20:27:20,409 epoch 2 - iter 396/1984 - loss 0.28880847 - time (sec): 6.06 - samples/sec: 5584.38 - lr: 0.000049 - momentum: 0.000000
95
+ 2023-10-18 20:27:23,444 epoch 2 - iter 594/1984 - loss 0.28273759 - time (sec): 9.09 - samples/sec: 5507.73 - lr: 0.000048 - momentum: 0.000000
96
+ 2023-10-18 20:27:26,439 epoch 2 - iter 792/1984 - loss 0.27088719 - time (sec): 12.09 - samples/sec: 5458.52 - lr: 0.000048 - momentum: 0.000000
97
+ 2023-10-18 20:27:29,491 epoch 2 - iter 990/1984 - loss 0.26504813 - time (sec): 15.14 - samples/sec: 5485.12 - lr: 0.000047 - momentum: 0.000000
98
+ 2023-10-18 20:27:32,572 epoch 2 - iter 1188/1984 - loss 0.26339405 - time (sec): 18.22 - samples/sec: 5459.96 - lr: 0.000047 - momentum: 0.000000
99
+ 2023-10-18 20:27:35,603 epoch 2 - iter 1386/1984 - loss 0.25689801 - time (sec): 21.25 - samples/sec: 5470.87 - lr: 0.000046 - momentum: 0.000000
100
+ 2023-10-18 20:27:38,702 epoch 2 - iter 1584/1984 - loss 0.25609983 - time (sec): 24.35 - samples/sec: 5445.66 - lr: 0.000046 - momentum: 0.000000
101
+ 2023-10-18 20:27:41,727 epoch 2 - iter 1782/1984 - loss 0.25426883 - time (sec): 27.38 - samples/sec: 5413.62 - lr: 0.000045 - momentum: 0.000000
102
+ 2023-10-18 20:27:44,747 epoch 2 - iter 1980/1984 - loss 0.25189498 - time (sec): 30.40 - samples/sec: 5383.25 - lr: 0.000044 - momentum: 0.000000
103
+ 2023-10-18 20:27:44,809 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-18 20:27:44,809 EPOCH 2 done: loss 0.2517 - lr: 0.000044
105
+ 2023-10-18 20:27:46,652 DEV : loss 0.16580356657505035 - f1-score (micro avg) 0.4132
106
+ 2023-10-18 20:27:46,671 saving best model
107
+ 2023-10-18 20:27:46,706 ----------------------------------------------------------------------------------------------------
108
+ 2023-10-18 20:27:49,900 epoch 3 - iter 198/1984 - loss 0.19382973 - time (sec): 3.19 - samples/sec: 5103.32 - lr: 0.000044 - momentum: 0.000000
109
+ 2023-10-18 20:27:52,946 epoch 3 - iter 396/1984 - loss 0.18909038 - time (sec): 6.24 - samples/sec: 5257.64 - lr: 0.000043 - momentum: 0.000000
110
+ 2023-10-18 20:27:55,955 epoch 3 - iter 594/1984 - loss 0.20993255 - time (sec): 9.25 - samples/sec: 5268.06 - lr: 0.000043 - momentum: 0.000000
111
+ 2023-10-18 20:27:59,021 epoch 3 - iter 792/1984 - loss 0.20714776 - time (sec): 12.31 - samples/sec: 5332.54 - lr: 0.000042 - momentum: 0.000000
112
+ 2023-10-18 20:28:02,011 epoch 3 - iter 990/1984 - loss 0.20713127 - time (sec): 15.30 - samples/sec: 5309.82 - lr: 0.000042 - momentum: 0.000000
113
+ 2023-10-18 20:28:05,286 epoch 3 - iter 1188/1984 - loss 0.20727047 - time (sec): 18.58 - samples/sec: 5280.75 - lr: 0.000041 - momentum: 0.000000
114
+ 2023-10-18 20:28:08,347 epoch 3 - iter 1386/1984 - loss 0.20873667 - time (sec): 21.64 - samples/sec: 5271.59 - lr: 0.000041 - momentum: 0.000000
115
+ 2023-10-18 20:28:11,390 epoch 3 - iter 1584/1984 - loss 0.20804768 - time (sec): 24.68 - samples/sec: 5300.39 - lr: 0.000040 - momentum: 0.000000
116
+ 2023-10-18 20:28:14,388 epoch 3 - iter 1782/1984 - loss 0.20568264 - time (sec): 27.68 - samples/sec: 5321.57 - lr: 0.000039 - momentum: 0.000000
117
+ 2023-10-18 20:28:17,407 epoch 3 - iter 1980/1984 - loss 0.20417751 - time (sec): 30.70 - samples/sec: 5326.77 - lr: 0.000039 - momentum: 0.000000
118
+ 2023-10-18 20:28:17,476 ----------------------------------------------------------------------------------------------------
119
+ 2023-10-18 20:28:17,476 EPOCH 3 done: loss 0.2043 - lr: 0.000039
120
+ 2023-10-18 20:28:19,319 DEV : loss 0.14851002395153046 - f1-score (micro avg) 0.513
121
+ 2023-10-18 20:28:19,337 saving best model
122
+ 2023-10-18 20:28:19,377 ----------------------------------------------------------------------------------------------------
123
+ 2023-10-18 20:28:22,449 epoch 4 - iter 198/1984 - loss 0.18877601 - time (sec): 3.07 - samples/sec: 5327.50 - lr: 0.000038 - momentum: 0.000000
124
+ 2023-10-18 20:28:25,530 epoch 4 - iter 396/1984 - loss 0.19191756 - time (sec): 6.15 - samples/sec: 5277.12 - lr: 0.000038 - momentum: 0.000000
125
+ 2023-10-18 20:28:28,538 epoch 4 - iter 594/1984 - loss 0.18721169 - time (sec): 9.16 - samples/sec: 5235.98 - lr: 0.000037 - momentum: 0.000000
126
+ 2023-10-18 20:28:31,607 epoch 4 - iter 792/1984 - loss 0.19120631 - time (sec): 12.23 - samples/sec: 5184.55 - lr: 0.000037 - momentum: 0.000000
127
+ 2023-10-18 20:28:34,601 epoch 4 - iter 990/1984 - loss 0.18849200 - time (sec): 15.22 - samples/sec: 5215.47 - lr: 0.000036 - momentum: 0.000000
128
+ 2023-10-18 20:28:37,624 epoch 4 - iter 1188/1984 - loss 0.18373488 - time (sec): 18.25 - samples/sec: 5250.54 - lr: 0.000036 - momentum: 0.000000
129
+ 2023-10-18 20:28:40,678 epoch 4 - iter 1386/1984 - loss 0.18383898 - time (sec): 21.30 - samples/sec: 5325.35 - lr: 0.000035 - momentum: 0.000000
130
+ 2023-10-18 20:28:43,747 epoch 4 - iter 1584/1984 - loss 0.18100800 - time (sec): 24.37 - samples/sec: 5341.76 - lr: 0.000034 - momentum: 0.000000
131
+ 2023-10-18 20:28:46,758 epoch 4 - iter 1782/1984 - loss 0.18083878 - time (sec): 27.38 - samples/sec: 5332.30 - lr: 0.000034 - momentum: 0.000000
132
+ 2023-10-18 20:28:49,823 epoch 4 - iter 1980/1984 - loss 0.17764091 - time (sec): 30.44 - samples/sec: 5375.24 - lr: 0.000033 - momentum: 0.000000
133
+ 2023-10-18 20:28:49,882 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-18 20:28:49,883 EPOCH 4 done: loss 0.1776 - lr: 0.000033
135
+ 2023-10-18 20:28:51,692 DEV : loss 0.14755003154277802 - f1-score (micro avg) 0.5567
136
+ 2023-10-18 20:28:51,711 saving best model
137
+ 2023-10-18 20:28:51,746 ----------------------------------------------------------------------------------------------------
138
+ 2023-10-18 20:28:54,714 epoch 5 - iter 198/1984 - loss 0.19822197 - time (sec): 2.97 - samples/sec: 5025.42 - lr: 0.000033 - momentum: 0.000000
139
+ 2023-10-18 20:28:57,801 epoch 5 - iter 396/1984 - loss 0.16894933 - time (sec): 6.05 - samples/sec: 5373.09 - lr: 0.000032 - momentum: 0.000000
140
+ 2023-10-18 20:29:00,850 epoch 5 - iter 594/1984 - loss 0.16781950 - time (sec): 9.10 - samples/sec: 5386.35 - lr: 0.000032 - momentum: 0.000000
141
+ 2023-10-18 20:29:03,958 epoch 5 - iter 792/1984 - loss 0.16415652 - time (sec): 12.21 - samples/sec: 5406.11 - lr: 0.000031 - momentum: 0.000000
142
+ 2023-10-18 20:29:06,959 epoch 5 - iter 990/1984 - loss 0.16205996 - time (sec): 15.21 - samples/sec: 5362.44 - lr: 0.000031 - momentum: 0.000000
143
+ 2023-10-18 20:29:09,981 epoch 5 - iter 1188/1984 - loss 0.16241670 - time (sec): 18.23 - samples/sec: 5393.56 - lr: 0.000030 - momentum: 0.000000
144
+ 2023-10-18 20:29:13,014 epoch 5 - iter 1386/1984 - loss 0.16432711 - time (sec): 21.27 - samples/sec: 5406.25 - lr: 0.000029 - momentum: 0.000000
145
+ 2023-10-18 20:29:16,025 epoch 5 - iter 1584/1984 - loss 0.16348040 - time (sec): 24.28 - samples/sec: 5420.92 - lr: 0.000029 - momentum: 0.000000
146
+ 2023-10-18 20:29:19,029 epoch 5 - iter 1782/1984 - loss 0.16364024 - time (sec): 27.28 - samples/sec: 5416.91 - lr: 0.000028 - momentum: 0.000000
147
+ 2023-10-18 20:29:22,045 epoch 5 - iter 1980/1984 - loss 0.16319243 - time (sec): 30.30 - samples/sec: 5401.10 - lr: 0.000028 - momentum: 0.000000
148
+ 2023-10-18 20:29:22,104 ----------------------------------------------------------------------------------------------------
149
+ 2023-10-18 20:29:22,104 EPOCH 5 done: loss 0.1631 - lr: 0.000028
150
+ 2023-10-18 20:29:23,921 DEV : loss 0.13933779299259186 - f1-score (micro avg) 0.5857
151
+ 2023-10-18 20:29:23,939 saving best model
152
+ 2023-10-18 20:29:23,973 ----------------------------------------------------------------------------------------------------
153
+ 2023-10-18 20:29:27,070 epoch 6 - iter 198/1984 - loss 0.17797563 - time (sec): 3.10 - samples/sec: 4977.75 - lr: 0.000027 - momentum: 0.000000
154
+ 2023-10-18 20:29:30,099 epoch 6 - iter 396/1984 - loss 0.16344503 - time (sec): 6.12 - samples/sec: 5175.17 - lr: 0.000027 - momentum: 0.000000
155
+ 2023-10-18 20:29:33,245 epoch 6 - iter 594/1984 - loss 0.16069556 - time (sec): 9.27 - samples/sec: 5117.65 - lr: 0.000026 - momentum: 0.000000
156
+ 2023-10-18 20:29:36,261 epoch 6 - iter 792/1984 - loss 0.16257303 - time (sec): 12.29 - samples/sec: 5175.21 - lr: 0.000026 - momentum: 0.000000
157
+ 2023-10-18 20:29:39,313 epoch 6 - iter 990/1984 - loss 0.16151177 - time (sec): 15.34 - samples/sec: 5240.88 - lr: 0.000025 - momentum: 0.000000
158
+ 2023-10-18 20:29:42,401 epoch 6 - iter 1188/1984 - loss 0.15627333 - time (sec): 18.43 - samples/sec: 5283.96 - lr: 0.000024 - momentum: 0.000000
159
+ 2023-10-18 20:29:45,429 epoch 6 - iter 1386/1984 - loss 0.15261274 - time (sec): 21.46 - samples/sec: 5332.64 - lr: 0.000024 - momentum: 0.000000
160
+ 2023-10-18 20:29:48,133 epoch 6 - iter 1584/1984 - loss 0.15177407 - time (sec): 24.16 - samples/sec: 5415.88 - lr: 0.000023 - momentum: 0.000000
161
+ 2023-10-18 20:29:50,811 epoch 6 - iter 1782/1984 - loss 0.15322561 - time (sec): 26.84 - samples/sec: 5487.27 - lr: 0.000023 - momentum: 0.000000
162
+ 2023-10-18 20:29:53,582 epoch 6 - iter 1980/1984 - loss 0.15084002 - time (sec): 29.61 - samples/sec: 5527.18 - lr: 0.000022 - momentum: 0.000000
163
+ 2023-10-18 20:29:53,643 ----------------------------------------------------------------------------------------------------
164
+ 2023-10-18 20:29:53,644 EPOCH 6 done: loss 0.1508 - lr: 0.000022
165
+ 2023-10-18 20:29:55,863 DEV : loss 0.14441362023353577 - f1-score (micro avg) 0.5832
166
+ 2023-10-18 20:29:55,882 ----------------------------------------------------------------------------------------------------
167
+ 2023-10-18 20:29:58,975 epoch 7 - iter 198/1984 - loss 0.17696709 - time (sec): 3.09 - samples/sec: 5179.02 - lr: 0.000022 - momentum: 0.000000
168
+ 2023-10-18 20:30:01,974 epoch 7 - iter 396/1984 - loss 0.15547142 - time (sec): 6.09 - samples/sec: 5311.56 - lr: 0.000021 - momentum: 0.000000
169
+ 2023-10-18 20:30:05,067 epoch 7 - iter 594/1984 - loss 0.15230504 - time (sec): 9.18 - samples/sec: 5276.61 - lr: 0.000021 - momentum: 0.000000
170
+ 2023-10-18 20:30:08,135 epoch 7 - iter 792/1984 - loss 0.14464304 - time (sec): 12.25 - samples/sec: 5360.99 - lr: 0.000020 - momentum: 0.000000
171
+ 2023-10-18 20:30:11,196 epoch 7 - iter 990/1984 - loss 0.14375802 - time (sec): 15.31 - samples/sec: 5387.28 - lr: 0.000019 - momentum: 0.000000
172
+ 2023-10-18 20:30:14,217 epoch 7 - iter 1188/1984 - loss 0.14330999 - time (sec): 18.33 - samples/sec: 5362.74 - lr: 0.000019 - momentum: 0.000000
173
+ 2023-10-18 20:30:17,322 epoch 7 - iter 1386/1984 - loss 0.14105304 - time (sec): 21.44 - samples/sec: 5365.32 - lr: 0.000018 - momentum: 0.000000
174
+ 2023-10-18 20:30:20,389 epoch 7 - iter 1584/1984 - loss 0.14079064 - time (sec): 24.51 - samples/sec: 5340.24 - lr: 0.000018 - momentum: 0.000000
175
+ 2023-10-18 20:30:23,438 epoch 7 - iter 1782/1984 - loss 0.14054323 - time (sec): 27.56 - samples/sec: 5332.07 - lr: 0.000017 - momentum: 0.000000
176
+ 2023-10-18 20:30:26,642 epoch 7 - iter 1980/1984 - loss 0.14054386 - time (sec): 30.76 - samples/sec: 5324.44 - lr: 0.000017 - momentum: 0.000000
177
+ 2023-10-18 20:30:26,701 ----------------------------------------------------------------------------------------------------
178
+ 2023-10-18 20:30:26,701 EPOCH 7 done: loss 0.1405 - lr: 0.000017
179
+ 2023-10-18 20:30:28,540 DEV : loss 0.14645995199680328 - f1-score (micro avg) 0.5839
180
+ 2023-10-18 20:30:28,559 ----------------------------------------------------------------------------------------------------
181
+ 2023-10-18 20:30:31,574 epoch 8 - iter 198/1984 - loss 0.13505995 - time (sec): 3.01 - samples/sec: 5338.26 - lr: 0.000016 - momentum: 0.000000
182
+ 2023-10-18 20:30:34,585 epoch 8 - iter 396/1984 - loss 0.13300452 - time (sec): 6.03 - samples/sec: 5274.07 - lr: 0.000016 - momentum: 0.000000
183
+ 2023-10-18 20:30:37,605 epoch 8 - iter 594/1984 - loss 0.13513264 - time (sec): 9.05 - samples/sec: 5218.46 - lr: 0.000015 - momentum: 0.000000
184
+ 2023-10-18 20:30:40,619 epoch 8 - iter 792/1984 - loss 0.13631868 - time (sec): 12.06 - samples/sec: 5324.96 - lr: 0.000014 - momentum: 0.000000
185
+ 2023-10-18 20:30:43,473 epoch 8 - iter 990/1984 - loss 0.13573158 - time (sec): 14.91 - samples/sec: 5368.03 - lr: 0.000014 - momentum: 0.000000
186
+ 2023-10-18 20:30:46,628 epoch 8 - iter 1188/1984 - loss 0.13343905 - time (sec): 18.07 - samples/sec: 5410.44 - lr: 0.000013 - momentum: 0.000000
187
+ 2023-10-18 20:30:49,733 epoch 8 - iter 1386/1984 - loss 0.13185324 - time (sec): 21.17 - samples/sec: 5358.71 - lr: 0.000013 - momentum: 0.000000
188
+ 2023-10-18 20:30:52,837 epoch 8 - iter 1584/1984 - loss 0.13284754 - time (sec): 24.28 - samples/sec: 5373.28 - lr: 0.000012 - momentum: 0.000000
189
+ 2023-10-18 20:30:55,870 epoch 8 - iter 1782/1984 - loss 0.13371484 - time (sec): 27.31 - samples/sec: 5380.92 - lr: 0.000012 - momentum: 0.000000
190
+ 2023-10-18 20:30:58,973 epoch 8 - iter 1980/1984 - loss 0.13321577 - time (sec): 30.41 - samples/sec: 5383.22 - lr: 0.000011 - momentum: 0.000000
191
+ 2023-10-18 20:30:59,035 ----------------------------------------------------------------------------------------------------
192
+ 2023-10-18 20:30:59,035 EPOCH 8 done: loss 0.1331 - lr: 0.000011
193
+ 2023-10-18 20:31:00,874 DEV : loss 0.1467994898557663 - f1-score (micro avg) 0.5978
194
+ 2023-10-18 20:31:00,893 saving best model
195
+ 2023-10-18 20:31:00,928 ----------------------------------------------------------------------------------------------------
196
+ 2023-10-18 20:31:03,991 epoch 9 - iter 198/1984 - loss 0.11768644 - time (sec): 3.06 - samples/sec: 5303.58 - lr: 0.000011 - momentum: 0.000000
197
+ 2023-10-18 20:31:07,031 epoch 9 - iter 396/1984 - loss 0.13017349 - time (sec): 6.10 - samples/sec: 5376.12 - lr: 0.000010 - momentum: 0.000000
198
+ 2023-10-18 20:31:10,059 epoch 9 - iter 594/1984 - loss 0.13044117 - time (sec): 9.13 - samples/sec: 5287.11 - lr: 0.000009 - momentum: 0.000000
199
+ 2023-10-18 20:31:13,053 epoch 9 - iter 792/1984 - loss 0.12921804 - time (sec): 12.12 - samples/sec: 5270.42 - lr: 0.000009 - momentum: 0.000000
200
+ 2023-10-18 20:31:16,142 epoch 9 - iter 990/1984 - loss 0.12884571 - time (sec): 15.21 - samples/sec: 5310.97 - lr: 0.000008 - momentum: 0.000000
201
+ 2023-10-18 20:31:19,206 epoch 9 - iter 1188/1984 - loss 0.12866639 - time (sec): 18.28 - samples/sec: 5297.82 - lr: 0.000008 - momentum: 0.000000
202
+ 2023-10-18 20:31:22,278 epoch 9 - iter 1386/1984 - loss 0.12832734 - time (sec): 21.35 - samples/sec: 5353.47 - lr: 0.000007 - momentum: 0.000000
203
+ 2023-10-18 20:31:25,302 epoch 9 - iter 1584/1984 - loss 0.12983739 - time (sec): 24.37 - samples/sec: 5340.26 - lr: 0.000007 - momentum: 0.000000
204
+ 2023-10-18 20:31:28,367 epoch 9 - iter 1782/1984 - loss 0.12868462 - time (sec): 27.44 - samples/sec: 5343.64 - lr: 0.000006 - momentum: 0.000000
205
+ 2023-10-18 20:31:31,449 epoch 9 - iter 1980/1984 - loss 0.12845863 - time (sec): 30.52 - samples/sec: 5363.40 - lr: 0.000006 - momentum: 0.000000
206
+ 2023-10-18 20:31:31,511 ----------------------------------------------------------------------------------------------------
207
+ 2023-10-18 20:31:31,511 EPOCH 9 done: loss 0.1285 - lr: 0.000006
208
+ 2023-10-18 20:31:33,363 DEV : loss 0.1488402932882309 - f1-score (micro avg) 0.598
209
+ 2023-10-18 20:31:33,382 saving best model
210
+ 2023-10-18 20:31:33,419 ----------------------------------------------------------------------------------------------------
211
+ 2023-10-18 20:31:36,322 epoch 10 - iter 198/1984 - loss 0.12152017 - time (sec): 2.90 - samples/sec: 5410.22 - lr: 0.000005 - momentum: 0.000000
212
+ 2023-10-18 20:31:39,369 epoch 10 - iter 396/1984 - loss 0.12775156 - time (sec): 5.95 - samples/sec: 5394.26 - lr: 0.000004 - momentum: 0.000000
213
+ 2023-10-18 20:31:42,403 epoch 10 - iter 594/1984 - loss 0.12825070 - time (sec): 8.98 - samples/sec: 5336.97 - lr: 0.000004 - momentum: 0.000000
214
+ 2023-10-18 20:31:45,472 epoch 10 - iter 792/1984 - loss 0.12234506 - time (sec): 12.05 - samples/sec: 5362.92 - lr: 0.000003 - momentum: 0.000000
215
+ 2023-10-18 20:31:48,524 epoch 10 - iter 990/1984 - loss 0.12321278 - time (sec): 15.11 - samples/sec: 5364.67 - lr: 0.000003 - momentum: 0.000000
216
+ 2023-10-18 20:31:51,571 epoch 10 - iter 1188/1984 - loss 0.12468032 - time (sec): 18.15 - samples/sec: 5333.23 - lr: 0.000002 - momentum: 0.000000
217
+ 2023-10-18 20:31:54,673 epoch 10 - iter 1386/1984 - loss 0.12362803 - time (sec): 21.25 - samples/sec: 5378.68 - lr: 0.000002 - momentum: 0.000000
218
+ 2023-10-18 20:31:57,703 epoch 10 - iter 1584/1984 - loss 0.12226510 - time (sec): 24.28 - samples/sec: 5359.19 - lr: 0.000001 - momentum: 0.000000
219
+ 2023-10-18 20:32:00,719 epoch 10 - iter 1782/1984 - loss 0.12224833 - time (sec): 27.30 - samples/sec: 5379.10 - lr: 0.000001 - momentum: 0.000000
220
+ 2023-10-18 20:32:03,781 epoch 10 - iter 1980/1984 - loss 0.12319689 - time (sec): 30.36 - samples/sec: 5388.44 - lr: 0.000000 - momentum: 0.000000
221
+ 2023-10-18 20:32:03,854 ----------------------------------------------------------------------------------------------------
222
+ 2023-10-18 20:32:03,854 EPOCH 10 done: loss 0.1234 - lr: 0.000000
223
+ 2023-10-18 20:32:05,695 DEV : loss 0.14801497757434845 - f1-score (micro avg) 0.6013
224
+ 2023-10-18 20:32:05,713 saving best model
225
+ 2023-10-18 20:32:05,773 ----------------------------------------------------------------------------------------------------
226
+ 2023-10-18 20:32:05,774 Loading model from best epoch ...
227
+ 2023-10-18 20:32:05,849 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
228
+ 2023-10-18 20:32:07,351
229
+ Results:
230
+ - F-score (micro) 0.619
231
+ - F-score (macro) 0.4877
232
+ - Accuracy 0.4937
233
+
234
+ By class:
235
+ precision recall f1-score support
236
+
237
+ LOC 0.7309 0.7008 0.7155 655
238
+ PER 0.4337 0.6457 0.5189 223
239
+ ORG 0.4167 0.1575 0.2286 127
240
+
241
+ micro avg 0.6181 0.6199 0.6190 1005
242
+ macro avg 0.5271 0.5013 0.4877 1005
243
+ weighted avg 0.6252 0.6199 0.6104 1005
244
+
245
+ 2023-10-18 20:32:07,351 ----------------------------------------------------------------------------------------------------