Hugging Face's logo

stefan-it
/
hmbench-icdar-fr-hmbert_tiny-bs8-wsFalse-e10-lr3e-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
ac144df
•
1 Parent(s): 75fcb6a

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.1697661137.46dc0c540dd0.3341.2 +3 -0
  5. test.tsv +0 -0
  6. training.log +246 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f7a4b91e310e8d3f741cf6b5fecad29a0d1e33bff6d75e2f3ceed291aaf91f87
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 20:32:41 0.0000 1.1686 0.2587 0.3529 0.0611 0.1041 0.0567
3
+ 2 20:33:05 0.0000 0.3001 0.1854 0.3742 0.3247 0.3477 0.2289
4
+ 3 20:33:30 0.0000 0.2428 0.1634 0.4297 0.4186 0.4241 0.2927
5
+ 4 20:33:54 0.0000 0.2173 0.1565 0.4913 0.4785 0.4848 0.3459
6
+ 5 20:34:19 0.0000 0.2012 0.1485 0.5058 0.5407 0.5227 0.3855
7
+ 6 20:34:43 0.0000 0.1906 0.1433 0.5397 0.5611 0.5502 0.4092
8
+ 7 20:35:08 0.0000 0.1826 0.1391 0.5600 0.5645 0.5623 0.4200
9
+ 8 20:35:32 0.0000 0.1782 0.1373 0.5745 0.5713 0.5729 0.4313
10
+ 9 20:35:57 0.0000 0.1719 0.1383 0.5737 0.5679 0.5708 0.4287
11
+ 10 20:36:21 0.0000 0.1715 0.1362 0.5717 0.5769 0.5743 0.4333
runs/events.out.tfevents.1697661137.46dc0c540dd0.3341.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7f682027a69221628a8622f851db4aa53314b01cc8797656048c31d59e464b6a
3
+ size 556612
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,246 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-18 20:32:17,044 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-18 20:32:17,044 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 20:32:17,044 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-18 20:32:17,044 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 20:32:17,044 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-18 20:32:17,044 Train: 7936 sentences
55
+ 2023-10-18 20:32:17,044 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-18 20:32:17,044 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-18 20:32:17,045 Training Params:
58
+ 2023-10-18 20:32:17,045 - learning_rate: "3e-05"
59
+ 2023-10-18 20:32:17,045 - mini_batch_size: "8"
60
+ 2023-10-18 20:32:17,045 - max_epochs: "10"
61
+ 2023-10-18 20:32:17,045 - shuffle: "True"
62
+ 2023-10-18 20:32:17,045 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-18 20:32:17,045 Plugins:
64
+ 2023-10-18 20:32:17,045 - TensorboardLogger
65
+ 2023-10-18 20:32:17,045 - LinearScheduler | warmup_fraction: '0.1'
66
+ 2023-10-18 20:32:17,045 ----------------------------------------------------------------------------------------------------
67
+ 2023-10-18 20:32:17,045 Final evaluation on model from best epoch (best-model.pt)
68
+ 2023-10-18 20:32:17,045 - metric: "('micro avg', 'f1-score')"
69
+ 2023-10-18 20:32:17,045 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-18 20:32:17,045 Computation:
71
+ 2023-10-18 20:32:17,045 - compute on device: cuda:0
72
+ 2023-10-18 20:32:17,045 - embedding storage: none
73
+ 2023-10-18 20:32:17,045 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-18 20:32:17,045 Model training base path: "hmbench-icdar/fr-dbmdz/bert-tiny-historic-multilingual-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-1"
75
+ 2023-10-18 20:32:17,045 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-18 20:32:17,045 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-18 20:32:17,045 Logging anything other than scalars to TensorBoard is currently not supported.
78
+ 2023-10-18 20:32:19,153 epoch 1 - iter 99/992 - loss 3.26900601 - time (sec): 2.11 - samples/sec: 7728.61 - lr: 0.000003 - momentum: 0.000000
79
+ 2023-10-18 20:32:21,668 epoch 1 - iter 198/992 - loss 3.01174688 - time (sec): 4.62 - samples/sec: 7025.90 - lr: 0.000006 - momentum: 0.000000
80
+ 2023-10-18 20:32:23,881 epoch 1 - iter 297/992 - loss 2.64131737 - time (sec): 6.84 - samples/sec: 7198.35 - lr: 0.000009 - momentum: 0.000000
81
+ 2023-10-18 20:32:26,462 epoch 1 - iter 396/992 - loss 2.23150630 - time (sec): 9.42 - samples/sec: 6982.05 - lr: 0.000012 - momentum: 0.000000
82
+ 2023-10-18 20:32:28,635 epoch 1 - iter 495/992 - loss 1.90904405 - time (sec): 11.59 - samples/sec: 7026.98 - lr: 0.000015 - momentum: 0.000000
83
+ 2023-10-18 20:32:30,871 epoch 1 - iter 594/992 - loss 1.67059385 - time (sec): 13.83 - samples/sec: 7081.35 - lr: 0.000018 - momentum: 0.000000
84
+ 2023-10-18 20:32:33,068 epoch 1 - iter 693/992 - loss 1.49967758 - time (sec): 16.02 - samples/sec: 7114.89 - lr: 0.000021 - momentum: 0.000000
85
+ 2023-10-18 20:32:35,281 epoch 1 - iter 792/992 - loss 1.36387729 - time (sec): 18.24 - samples/sec: 7181.71 - lr: 0.000024 - momentum: 0.000000
86
+ 2023-10-18 20:32:37,528 epoch 1 - iter 891/992 - loss 1.25925333 - time (sec): 20.48 - samples/sec: 7183.73 - lr: 0.000027 - momentum: 0.000000
87
+ 2023-10-18 20:32:39,760 epoch 1 - iter 990/992 - loss 1.17027992 - time (sec): 22.71 - samples/sec: 7203.37 - lr: 0.000030 - momentum: 0.000000
88
+ 2023-10-18 20:32:39,805 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-18 20:32:39,805 EPOCH 1 done: loss 1.1686 - lr: 0.000030
90
+ 2023-10-18 20:32:41,275 DEV : loss 0.25871363282203674 - f1-score (micro avg) 0.1041
91
+ 2023-10-18 20:32:41,293 saving best model
92
+ 2023-10-18 20:32:41,328 ----------------------------------------------------------------------------------------------------
93
+ 2023-10-18 20:32:43,608 epoch 2 - iter 99/992 - loss 0.38667417 - time (sec): 2.28 - samples/sec: 7457.05 - lr: 0.000030 - momentum: 0.000000
94
+ 2023-10-18 20:32:45,809 epoch 2 - iter 198/992 - loss 0.35532658 - time (sec): 4.48 - samples/sec: 7551.86 - lr: 0.000029 - momentum: 0.000000
95
+ 2023-10-18 20:32:48,030 epoch 2 - iter 297/992 - loss 0.34540051 - time (sec): 6.70 - samples/sec: 7473.14 - lr: 0.000029 - momentum: 0.000000
96
+ 2023-10-18 20:32:50,309 epoch 2 - iter 396/992 - loss 0.33013814 - time (sec): 8.98 - samples/sec: 7347.87 - lr: 0.000029 - momentum: 0.000000
97
+ 2023-10-18 20:32:52,637 epoch 2 - iter 495/992 - loss 0.32197045 - time (sec): 11.31 - samples/sec: 7343.57 - lr: 0.000028 - momentum: 0.000000
98
+ 2023-10-18 20:32:54,913 epoch 2 - iter 594/992 - loss 0.31801653 - time (sec): 13.59 - samples/sec: 7323.60 - lr: 0.000028 - momentum: 0.000000
99
+ 2023-10-18 20:32:57,144 epoch 2 - iter 693/992 - loss 0.30876076 - time (sec): 15.82 - samples/sec: 7351.65 - lr: 0.000028 - momentum: 0.000000
100
+ 2023-10-18 20:32:59,360 epoch 2 - iter 792/992 - loss 0.30673219 - time (sec): 18.03 - samples/sec: 7354.43 - lr: 0.000027 - momentum: 0.000000
101
+ 2023-10-18 20:33:01,529 epoch 2 - iter 891/992 - loss 0.30357404 - time (sec): 20.20 - samples/sec: 7336.57 - lr: 0.000027 - momentum: 0.000000
102
+ 2023-10-18 20:33:03,726 epoch 2 - iter 990/992 - loss 0.30024812 - time (sec): 22.40 - samples/sec: 7305.76 - lr: 0.000027 - momentum: 0.000000
103
+ 2023-10-18 20:33:03,769 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-18 20:33:03,769 EPOCH 2 done: loss 0.3001 - lr: 0.000027
105
+ 2023-10-18 20:33:05,604 DEV : loss 0.1853763908147812 - f1-score (micro avg) 0.3477
106
+ 2023-10-18 20:33:05,622 saving best model
107
+ 2023-10-18 20:33:05,658 ----------------------------------------------------------------------------------------------------
108
+ 2023-10-18 20:33:07,896 epoch 3 - iter 99/992 - loss 0.23413942 - time (sec): 2.24 - samples/sec: 7285.51 - lr: 0.000026 - momentum: 0.000000
109
+ 2023-10-18 20:33:10,145 epoch 3 - iter 198/992 - loss 0.22991950 - time (sec): 4.49 - samples/sec: 7312.34 - lr: 0.000026 - momentum: 0.000000
110
+ 2023-10-18 20:33:12,378 epoch 3 - iter 297/992 - loss 0.25254134 - time (sec): 6.72 - samples/sec: 7251.19 - lr: 0.000026 - momentum: 0.000000
111
+ 2023-10-18 20:33:14,643 epoch 3 - iter 396/992 - loss 0.24877043 - time (sec): 8.98 - samples/sec: 7308.99 - lr: 0.000025 - momentum: 0.000000
112
+ 2023-10-18 20:33:16,809 epoch 3 - iter 495/992 - loss 0.24911355 - time (sec): 11.15 - samples/sec: 7288.56 - lr: 0.000025 - momentum: 0.000000
113
+ 2023-10-18 20:33:19,031 epoch 3 - iter 594/992 - loss 0.24658049 - time (sec): 13.37 - samples/sec: 7337.19 - lr: 0.000025 - momentum: 0.000000
114
+ 2023-10-18 20:33:21,280 epoch 3 - iter 693/992 - loss 0.24936620 - time (sec): 15.62 - samples/sec: 7303.17 - lr: 0.000024 - momentum: 0.000000
115
+ 2023-10-18 20:33:23,521 epoch 3 - iter 792/992 - loss 0.24772256 - time (sec): 17.86 - samples/sec: 7324.51 - lr: 0.000024 - momentum: 0.000000
116
+ 2023-10-18 20:33:25,756 epoch 3 - iter 891/992 - loss 0.24405967 - time (sec): 20.10 - samples/sec: 7330.03 - lr: 0.000024 - momentum: 0.000000
117
+ 2023-10-18 20:33:27,957 epoch 3 - iter 990/992 - loss 0.24272779 - time (sec): 22.30 - samples/sec: 7334.09 - lr: 0.000023 - momentum: 0.000000
118
+ 2023-10-18 20:33:28,002 ----------------------------------------------------------------------------------------------------
119
+ 2023-10-18 20:33:28,002 EPOCH 3 done: loss 0.2428 - lr: 0.000023
120
+ 2023-10-18 20:33:30,215 DEV : loss 0.16335979104042053 - f1-score (micro avg) 0.4241
121
+ 2023-10-18 20:33:30,233 saving best model
122
+ 2023-10-18 20:33:30,266 ----------------------------------------------------------------------------------------------------
123
+ 2023-10-18 20:33:32,469 epoch 4 - iter 99/992 - loss 0.22443699 - time (sec): 2.20 - samples/sec: 7430.86 - lr: 0.000023 - momentum: 0.000000
124
+ 2023-10-18 20:33:34,801 epoch 4 - iter 198/992 - loss 0.23701210 - time (sec): 4.53 - samples/sec: 7160.62 - lr: 0.000023 - momentum: 0.000000
125
+ 2023-10-18 20:33:37,037 epoch 4 - iter 297/992 - loss 0.23242108 - time (sec): 6.77 - samples/sec: 7084.78 - lr: 0.000022 - momentum: 0.000000
126
+ 2023-10-18 20:33:39,238 epoch 4 - iter 396/992 - loss 0.23657149 - time (sec): 8.97 - samples/sec: 7067.25 - lr: 0.000022 - momentum: 0.000000
127
+ 2023-10-18 20:33:41,445 epoch 4 - iter 495/992 - loss 0.23101340 - time (sec): 11.18 - samples/sec: 7103.06 - lr: 0.000022 - momentum: 0.000000
128
+ 2023-10-18 20:33:43,690 epoch 4 - iter 594/992 - loss 0.22285698 - time (sec): 13.42 - samples/sec: 7137.36 - lr: 0.000021 - momentum: 0.000000
129
+ 2023-10-18 20:33:45,950 epoch 4 - iter 693/992 - loss 0.22245975 - time (sec): 15.68 - samples/sec: 7232.76 - lr: 0.000021 - momentum: 0.000000
130
+ 2023-10-18 20:33:48,271 epoch 4 - iter 792/992 - loss 0.21974109 - time (sec): 18.00 - samples/sec: 7230.28 - lr: 0.000021 - momentum: 0.000000
131
+ 2023-10-18 20:33:50,429 epoch 4 - iter 891/992 - loss 0.22008264 - time (sec): 20.16 - samples/sec: 7241.30 - lr: 0.000020 - momentum: 0.000000
132
+ 2023-10-18 20:33:52,672 epoch 4 - iter 990/992 - loss 0.21727144 - time (sec): 22.41 - samples/sec: 7304.02 - lr: 0.000020 - momentum: 0.000000
133
+ 2023-10-18 20:33:52,723 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-18 20:33:52,723 EPOCH 4 done: loss 0.2173 - lr: 0.000020
135
+ 2023-10-18 20:33:54,524 DEV : loss 0.15653517842292786 - f1-score (micro avg) 0.4848
136
+ 2023-10-18 20:33:54,542 saving best model
137
+ 2023-10-18 20:33:54,578 ----------------------------------------------------------------------------------------------------
138
+ 2023-10-18 20:33:56,710 epoch 5 - iter 99/992 - loss 0.24537712 - time (sec): 2.13 - samples/sec: 6996.16 - lr: 0.000020 - momentum: 0.000000
139
+ 2023-10-18 20:33:58,940 epoch 5 - iter 198/992 - loss 0.21134377 - time (sec): 4.36 - samples/sec: 7459.07 - lr: 0.000019 - momentum: 0.000000
140
+ 2023-10-18 20:34:01,175 epoch 5 - iter 297/992 - loss 0.20640811 - time (sec): 6.60 - samples/sec: 7433.04 - lr: 0.000019 - momentum: 0.000000
141
+ 2023-10-18 20:34:03,545 epoch 5 - iter 396/992 - loss 0.20197952 - time (sec): 8.97 - samples/sec: 7362.38 - lr: 0.000019 - momentum: 0.000000
142
+ 2023-10-18 20:34:05,772 epoch 5 - iter 495/992 - loss 0.20092622 - time (sec): 11.19 - samples/sec: 7287.58 - lr: 0.000018 - momentum: 0.000000
143
+ 2023-10-18 20:34:08,064 epoch 5 - iter 594/992 - loss 0.20292860 - time (sec): 13.49 - samples/sec: 7292.89 - lr: 0.000018 - momentum: 0.000000
144
+ 2023-10-18 20:34:10,381 epoch 5 - iter 693/992 - loss 0.20258902 - time (sec): 15.80 - samples/sec: 7276.17 - lr: 0.000018 - momentum: 0.000000
145
+ 2023-10-18 20:34:12,621 epoch 5 - iter 792/992 - loss 0.20192145 - time (sec): 18.04 - samples/sec: 7294.30 - lr: 0.000017 - momentum: 0.000000
146
+ 2023-10-18 20:34:14,841 epoch 5 - iter 891/992 - loss 0.20140955 - time (sec): 20.26 - samples/sec: 7293.58 - lr: 0.000017 - momentum: 0.000000
147
+ 2023-10-18 20:34:17,121 epoch 5 - iter 990/992 - loss 0.20126076 - time (sec): 22.54 - samples/sec: 7259.31 - lr: 0.000017 - momentum: 0.000000
148
+ 2023-10-18 20:34:17,169 ----------------------------------------------------------------------------------------------------
149
+ 2023-10-18 20:34:17,169 EPOCH 5 done: loss 0.2012 - lr: 0.000017
150
+ 2023-10-18 20:34:18,982 DEV : loss 0.1485081911087036 - f1-score (micro avg) 0.5227
151
+ 2023-10-18 20:34:19,001 saving best model
152
+ 2023-10-18 20:34:19,036 ----------------------------------------------------------------------------------------------------
153
+ 2023-10-18 20:34:21,233 epoch 6 - iter 99/992 - loss 0.22430179 - time (sec): 2.20 - samples/sec: 7017.32 - lr: 0.000016 - momentum: 0.000000
154
+ 2023-10-18 20:34:23,529 epoch 6 - iter 198/992 - loss 0.20436870 - time (sec): 4.49 - samples/sec: 7055.19 - lr: 0.000016 - momentum: 0.000000
155
+ 2023-10-18 20:34:25,787 epoch 6 - iter 297/992 - loss 0.20029647 - time (sec): 6.75 - samples/sec: 7028.76 - lr: 0.000016 - momentum: 0.000000
156
+ 2023-10-18 20:34:28,088 epoch 6 - iter 396/992 - loss 0.20256720 - time (sec): 9.05 - samples/sec: 7025.13 - lr: 0.000015 - momentum: 0.000000
157
+ 2023-10-18 20:34:30,365 epoch 6 - iter 495/992 - loss 0.20317747 - time (sec): 11.33 - samples/sec: 7096.80 - lr: 0.000015 - momentum: 0.000000
158
+ 2023-10-18 20:34:32,649 epoch 6 - iter 594/992 - loss 0.19651111 - time (sec): 13.61 - samples/sec: 7152.66 - lr: 0.000015 - momentum: 0.000000
159
+ 2023-10-18 20:34:34,851 epoch 6 - iter 693/992 - loss 0.19268692 - time (sec): 15.81 - samples/sec: 7234.86 - lr: 0.000014 - momentum: 0.000000
160
+ 2023-10-18 20:34:37,113 epoch 6 - iter 792/992 - loss 0.19172945 - time (sec): 18.08 - samples/sec: 7238.22 - lr: 0.000014 - momentum: 0.000000
161
+ 2023-10-18 20:34:39,321 epoch 6 - iter 891/992 - loss 0.19360741 - time (sec): 20.28 - samples/sec: 7259.92 - lr: 0.000014 - momentum: 0.000000
162
+ 2023-10-18 20:34:41,590 epoch 6 - iter 990/992 - loss 0.19064340 - time (sec): 22.55 - samples/sec: 7256.27 - lr: 0.000013 - momentum: 0.000000
163
+ 2023-10-18 20:34:41,642 ----------------------------------------------------------------------------------------------------
164
+ 2023-10-18 20:34:41,642 EPOCH 6 done: loss 0.1906 - lr: 0.000013
165
+ 2023-10-18 20:34:43,465 DEV : loss 0.1433064043521881 - f1-score (micro avg) 0.5502
166
+ 2023-10-18 20:34:43,483 saving best model
167
+ 2023-10-18 20:34:43,516 ----------------------------------------------------------------------------------------------------
168
+ 2023-10-18 20:34:45,840 epoch 7 - iter 99/992 - loss 0.21206962 - time (sec): 2.32 - samples/sec: 6893.32 - lr: 0.000013 - momentum: 0.000000
169
+ 2023-10-18 20:34:48,076 epoch 7 - iter 198/992 - loss 0.19780535 - time (sec): 4.56 - samples/sec: 7095.06 - lr: 0.000013 - momentum: 0.000000
170
+ 2023-10-18 20:34:50,388 epoch 7 - iter 297/992 - loss 0.19634963 - time (sec): 6.87 - samples/sec: 7052.69 - lr: 0.000012 - momentum: 0.000000
171
+ 2023-10-18 20:34:52,686 epoch 7 - iter 396/992 - loss 0.18709449 - time (sec): 9.17 - samples/sec: 7163.54 - lr: 0.000012 - momentum: 0.000000
172
+ 2023-10-18 20:34:54,903 epoch 7 - iter 495/992 - loss 0.18615423 - time (sec): 11.39 - samples/sec: 7245.35 - lr: 0.000012 - momentum: 0.000000
173
+ 2023-10-18 20:34:57,221 epoch 7 - iter 594/992 - loss 0.18575283 - time (sec): 13.70 - samples/sec: 7174.74 - lr: 0.000011 - momentum: 0.000000
174
+ 2023-10-18 20:34:59,457 epoch 7 - iter 693/992 - loss 0.18299791 - time (sec): 15.94 - samples/sec: 7216.08 - lr: 0.000011 - momentum: 0.000000
175
+ 2023-10-18 20:35:01,674 epoch 7 - iter 792/992 - loss 0.18308143 - time (sec): 18.16 - samples/sec: 7207.35 - lr: 0.000011 - momentum: 0.000000
176
+ 2023-10-18 20:35:03,934 epoch 7 - iter 891/992 - loss 0.18248990 - time (sec): 20.42 - samples/sec: 7196.20 - lr: 0.000010 - momentum: 0.000000
177
+ 2023-10-18 20:35:06,151 epoch 7 - iter 990/992 - loss 0.18260769 - time (sec): 22.63 - samples/sec: 7235.83 - lr: 0.000010 - momentum: 0.000000
178
+ 2023-10-18 20:35:06,190 ----------------------------------------------------------------------------------------------------
179
+ 2023-10-18 20:35:06,190 EPOCH 7 done: loss 0.1826 - lr: 0.000010
180
+ 2023-10-18 20:35:08,035 DEV : loss 0.13907533884048462 - f1-score (micro avg) 0.5623
181
+ 2023-10-18 20:35:08,053 saving best model
182
+ 2023-10-18 20:35:08,086 ----------------------------------------------------------------------------------------------------
183
+ 2023-10-18 20:35:10,249 epoch 8 - iter 99/992 - loss 0.18323978 - time (sec): 2.16 - samples/sec: 7443.73 - lr: 0.000010 - momentum: 0.000000
184
+ 2023-10-18 20:35:12,450 epoch 8 - iter 198/992 - loss 0.17828271 - time (sec): 4.36 - samples/sec: 7283.96 - lr: 0.000009 - momentum: 0.000000
185
+ 2023-10-18 20:35:14,693 epoch 8 - iter 297/992 - loss 0.17876581 - time (sec): 6.61 - samples/sec: 7145.60 - lr: 0.000009 - momentum: 0.000000
186
+ 2023-10-18 20:35:17,019 epoch 8 - iter 396/992 - loss 0.18045852 - time (sec): 8.93 - samples/sec: 7189.34 - lr: 0.000009 - momentum: 0.000000
187
+ 2023-10-18 20:35:19,211 epoch 8 - iter 495/992 - loss 0.18102038 - time (sec): 11.12 - samples/sec: 7197.08 - lr: 0.000008 - momentum: 0.000000
188
+ 2023-10-18 20:35:21,483 epoch 8 - iter 594/992 - loss 0.17757856 - time (sec): 13.40 - samples/sec: 7297.71 - lr: 0.000008 - momentum: 0.000000
189
+ 2023-10-18 20:35:23,754 epoch 8 - iter 693/992 - loss 0.17685881 - time (sec): 15.67 - samples/sec: 7241.95 - lr: 0.000008 - momentum: 0.000000
190
+ 2023-10-18 20:35:25,987 epoch 8 - iter 792/992 - loss 0.17740407 - time (sec): 17.90 - samples/sec: 7287.46 - lr: 0.000007 - momentum: 0.000000
191
+ 2023-10-18 20:35:28,216 epoch 8 - iter 891/992 - loss 0.17815940 - time (sec): 20.13 - samples/sec: 7300.64 - lr: 0.000007 - momentum: 0.000000
192
+ 2023-10-18 20:35:30,534 epoch 8 - iter 990/992 - loss 0.17824616 - time (sec): 22.45 - samples/sec: 7293.89 - lr: 0.000007 - momentum: 0.000000
193
+ 2023-10-18 20:35:30,576 ----------------------------------------------------------------------------------------------------
194
+ 2023-10-18 20:35:30,576 EPOCH 8 done: loss 0.1782 - lr: 0.000007
195
+ 2023-10-18 20:35:32,760 DEV : loss 0.1372818648815155 - f1-score (micro avg) 0.5729
196
+ 2023-10-18 20:35:32,778 saving best model
197
+ 2023-10-18 20:35:32,810 ----------------------------------------------------------------------------------------------------
198
+ 2023-10-18 20:35:35,075 epoch 9 - iter 99/992 - loss 0.15969196 - time (sec): 2.26 - samples/sec: 7173.67 - lr: 0.000006 - momentum: 0.000000
199
+ 2023-10-18 20:35:37,180 epoch 9 - iter 198/992 - loss 0.17448311 - time (sec): 4.37 - samples/sec: 7508.59 - lr: 0.000006 - momentum: 0.000000
200
+ 2023-10-18 20:35:39,362 epoch 9 - iter 297/992 - loss 0.17623452 - time (sec): 6.55 - samples/sec: 7368.61 - lr: 0.000006 - momentum: 0.000000
201
+ 2023-10-18 20:35:41,713 epoch 9 - iter 396/992 - loss 0.17504733 - time (sec): 8.90 - samples/sec: 7178.40 - lr: 0.000005 - momentum: 0.000000
202
+ 2023-10-18 20:35:43,965 epoch 9 - iter 495/992 - loss 0.17349129 - time (sec): 11.15 - samples/sec: 7243.56 - lr: 0.000005 - momentum: 0.000000
203
+ 2023-10-18 20:35:46,194 epoch 9 - iter 594/992 - loss 0.17420767 - time (sec): 13.38 - samples/sec: 7235.24 - lr: 0.000005 - momentum: 0.000000
204
+ 2023-10-18 20:35:48,470 epoch 9 - iter 693/992 - loss 0.17251140 - time (sec): 15.66 - samples/sec: 7298.69 - lr: 0.000004 - momentum: 0.000000
205
+ 2023-10-18 20:35:50,625 epoch 9 - iter 792/992 - loss 0.17353050 - time (sec): 17.81 - samples/sec: 7306.77 - lr: 0.000004 - momentum: 0.000000
206
+ 2023-10-18 20:35:52,833 epoch 9 - iter 891/992 - loss 0.17204365 - time (sec): 20.02 - samples/sec: 7322.75 - lr: 0.000004 - momentum: 0.000000
207
+ 2023-10-18 20:35:55,115 epoch 9 - iter 990/992 - loss 0.17187736 - time (sec): 22.30 - samples/sec: 7339.19 - lr: 0.000003 - momentum: 0.000000
208
+ 2023-10-18 20:35:55,161 ----------------------------------------------------------------------------------------------------
209
+ 2023-10-18 20:35:55,161 EPOCH 9 done: loss 0.1719 - lr: 0.000003
210
+ 2023-10-18 20:35:56,988 DEV : loss 0.13832369446754456 - f1-score (micro avg) 0.5708
211
+ 2023-10-18 20:35:57,009 ----------------------------------------------------------------------------------------------------
212
+ 2023-10-18 20:35:59,237 epoch 10 - iter 99/992 - loss 0.17701827 - time (sec): 2.23 - samples/sec: 7049.38 - lr: 0.000003 - momentum: 0.000000
213
+ 2023-10-18 20:36:01,471 epoch 10 - iter 198/992 - loss 0.17977925 - time (sec): 4.46 - samples/sec: 7194.13 - lr: 0.000003 - momentum: 0.000000
214
+ 2023-10-18 20:36:03,693 epoch 10 - iter 297/992 - loss 0.17530848 - time (sec): 6.68 - samples/sec: 7174.37 - lr: 0.000002 - momentum: 0.000000
215
+ 2023-10-18 20:36:05,955 epoch 10 - iter 396/992 - loss 0.16854571 - time (sec): 8.95 - samples/sec: 7225.38 - lr: 0.000002 - momentum: 0.000000
216
+ 2023-10-18 20:36:08,161 epoch 10 - iter 495/992 - loss 0.17158421 - time (sec): 11.15 - samples/sec: 7266.51 - lr: 0.000002 - momentum: 0.000000
217
+ 2023-10-18 20:36:10,360 epoch 10 - iter 594/992 - loss 0.17275765 - time (sec): 13.35 - samples/sec: 7251.27 - lr: 0.000001 - momentum: 0.000000
218
+ 2023-10-18 20:36:12,695 epoch 10 - iter 693/992 - loss 0.17130222 - time (sec): 15.69 - samples/sec: 7288.10 - lr: 0.000001 - momentum: 0.000000
219
+ 2023-10-18 20:36:14,930 epoch 10 - iter 792/992 - loss 0.17016111 - time (sec): 17.92 - samples/sec: 7261.87 - lr: 0.000001 - momentum: 0.000000
220
+ 2023-10-18 20:36:17,124 epoch 10 - iter 891/992 - loss 0.17025831 - time (sec): 20.11 - samples/sec: 7300.56 - lr: 0.000000 - momentum: 0.000000
221
+ 2023-10-18 20:36:19,367 epoch 10 - iter 990/992 - loss 0.17132536 - time (sec): 22.36 - samples/sec: 7317.61 - lr: 0.000000 - momentum: 0.000000
222
+ 2023-10-18 20:36:19,412 ----------------------------------------------------------------------------------------------------
223
+ 2023-10-18 20:36:19,412 EPOCH 10 done: loss 0.1715 - lr: 0.000000
224
+ 2023-10-18 20:36:21,225 DEV : loss 0.13621357083320618 - f1-score (micro avg) 0.5743
225
+ 2023-10-18 20:36:21,244 saving best model
226
+ 2023-10-18 20:36:21,307 ----------------------------------------------------------------------------------------------------
227
+ 2023-10-18 20:36:21,307 Loading model from best epoch ...
228
+ 2023-10-18 20:36:21,389 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
229
+ 2023-10-18 20:36:22,927
230
+ Results:
231
+ - F-score (micro) 0.5768
232
+ - F-score (macro) 0.3939
233
+ - Accuracy 0.4531
234
+
235
+ By class:
236
+ precision recall f1-score support
237
+
238
+ LOC 0.7394 0.6672 0.7014 655
239
+ PER 0.3516 0.6323 0.4519 223
240
+ ORG 0.1429 0.0157 0.0284 127
241
+
242
+ micro avg 0.5765 0.5771 0.5768 1005
243
+ macro avg 0.4113 0.4384 0.3939 1005
244
+ weighted avg 0.5780 0.5771 0.5610 1005
245
+
246
+ 2023-10-18 20:36:22,927 ----------------------------------------------------------------------------------------------------