Hugging Face's logo

stefan-it
/
hmbench-icdar-fr-hmbert_tiny-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2

Token Classification
Flair
PyTorch
TensorBoard
French
sequence-tagger-model
Model card Files Metrics Community
1
stefan-it commited on
Commit
4bdf4fb
•
1 Parent(s): 9314c88

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.1697662310.46dc0c540dd0.3341.6 +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:d3fb90b24348371823627063f2d3f7298221363ccfd442546b88d1434c58d9c8
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 20:52:15 0.0000 0.9089 0.2375 0.2994 0.1731 0.2194 0.1354
3
+ 2 20:52:39 0.0000 0.2981 0.2020 0.3380 0.3281 0.3330 0.2224
4
+ 3 20:53:04 0.0000 0.2605 0.1907 0.3456 0.3835 0.3635 0.2482
5
+ 4 20:53:28 0.0000 0.2376 0.1791 0.3418 0.4106 0.3731 0.2560
6
+ 5 20:53:53 0.0000 0.2252 0.1707 0.3582 0.4231 0.3880 0.2664
7
+ 6 20:54:17 0.0000 0.2121 0.1651 0.3584 0.4367 0.3937 0.2724
8
+ 7 20:54:42 0.0000 0.2040 0.1611 0.3857 0.4333 0.4081 0.2812
9
+ 8 20:55:06 0.0000 0.1978 0.1614 0.3761 0.4378 0.4046 0.2792
10
+ 9 20:55:30 0.0000 0.1941 0.1605 0.3735 0.4491 0.4078 0.2816
11
+ 10 20:55:54 0.0000 0.1919 0.1607 0.3829 0.4457 0.4119 0.2849
runs/events.out.tfevents.1697662310.46dc0c540dd0.3341.6 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:85f46398a308a8944f7df69f45c183266454f63180aff83991fd68482ac34a72
3
+ size 556612
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,245 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-18 20:51:50,862 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-18 20:51:50,862 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 20:51:50,862 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-18 20:51:50,863 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 20:51:50,863 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-18 20:51:50,863 Train: 7936 sentences
55
+ 2023-10-18 20:51:50,863 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-18 20:51:50,863 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-18 20:51:50,863 Training Params:
58
+ 2023-10-18 20:51:50,863 - learning_rate: "3e-05"
59
+ 2023-10-18 20:51:50,863 - mini_batch_size: "8"
60
+ 2023-10-18 20:51:50,863 - max_epochs: "10"
61
+ 2023-10-18 20:51:50,863 - shuffle: "True"
62
+ 2023-10-18 20:51:50,863 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-18 20:51:50,863 Plugins:
64
+ 2023-10-18 20:51:50,863 - TensorboardLogger
65
+ 2023-10-18 20:51:50,863 - LinearScheduler | warmup_fraction: '0.1'
66
+ 2023-10-18 20:51:50,863 ----------------------------------------------------------------------------------------------------
67
+ 2023-10-18 20:51:50,863 Final evaluation on model from best epoch (best-model.pt)
68
+ 2023-10-18 20:51:50,863 - metric: "('micro avg', 'f1-score')"
69
+ 2023-10-18 20:51:50,863 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-18 20:51:50,863 Computation:
71
+ 2023-10-18 20:51:50,863 - compute on device: cuda:0
72
+ 2023-10-18 20:51:50,863 - embedding storage: none
73
+ 2023-10-18 20:51:50,863 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-18 20:51:50,863 Model training base path: "hmbench-icdar/fr-dbmdz/bert-tiny-historic-multilingual-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2"
75
+ 2023-10-18 20:51:50,863 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-18 20:51:50,863 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-18 20:51:50,863 Logging anything other than scalars to TensorBoard is currently not supported.
78
+ 2023-10-18 20:51:53,281 epoch 1 - iter 99/992 - loss 2.42306355 - time (sec): 2.42 - samples/sec: 6817.15 - lr: 0.000003 - momentum: 0.000000
79
+ 2023-10-18 20:51:55,609 epoch 1 - iter 198/992 - loss 2.21764434 - time (sec): 4.75 - samples/sec: 6893.42 - lr: 0.000006 - momentum: 0.000000
80
+ 2023-10-18 20:51:57,874 epoch 1 - iter 297/992 - loss 1.90812069 - time (sec): 7.01 - samples/sec: 7104.88 - lr: 0.000009 - momentum: 0.000000
81
+ 2023-10-18 20:52:00,102 epoch 1 - iter 396/992 - loss 1.60298911 - time (sec): 9.24 - samples/sec: 7162.65 - lr: 0.000012 - momentum: 0.000000
82
+ 2023-10-18 20:52:02,336 epoch 1 - iter 495/992 - loss 1.39460148 - time (sec): 11.47 - samples/sec: 7253.67 - lr: 0.000015 - momentum: 0.000000
83
+ 2023-10-18 20:52:04,570 epoch 1 - iter 594/992 - loss 1.24571099 - time (sec): 13.71 - samples/sec: 7251.89 - lr: 0.000018 - momentum: 0.000000
84
+ 2023-10-18 20:52:06,806 epoch 1 - iter 693/992 - loss 1.13088356 - time (sec): 15.94 - samples/sec: 7261.58 - lr: 0.000021 - momentum: 0.000000
85
+ 2023-10-18 20:52:09,007 epoch 1 - iter 792/992 - loss 1.03882783 - time (sec): 18.14 - samples/sec: 7255.30 - lr: 0.000024 - momentum: 0.000000
86
+ 2023-10-18 20:52:11,267 epoch 1 - iter 891/992 - loss 0.97090243 - time (sec): 20.40 - samples/sec: 7218.69 - lr: 0.000027 - momentum: 0.000000
87
+ 2023-10-18 20:52:13,572 epoch 1 - iter 990/992 - loss 0.91052903 - time (sec): 22.71 - samples/sec: 7208.11 - lr: 0.000030 - momentum: 0.000000
88
+ 2023-10-18 20:52:13,616 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-18 20:52:13,616 EPOCH 1 done: loss 0.9089 - lr: 0.000030
90
+ 2023-10-18 20:52:15,124 DEV : loss 0.23747262358665466 - f1-score (micro avg) 0.2194
91
+ 2023-10-18 20:52:15,143 saving best model
92
+ 2023-10-18 20:52:15,175 ----------------------------------------------------------------------------------------------------
93
+ 2023-10-18 20:52:17,403 epoch 2 - iter 99/992 - loss 0.34838492 - time (sec): 2.23 - samples/sec: 7243.16 - lr: 0.000030 - momentum: 0.000000
94
+ 2023-10-18 20:52:20,106 epoch 2 - iter 198/992 - loss 0.31688513 - time (sec): 4.93 - samples/sec: 6753.96 - lr: 0.000029 - momentum: 0.000000
95
+ 2023-10-18 20:52:22,343 epoch 2 - iter 297/992 - loss 0.31470515 - time (sec): 7.17 - samples/sec: 6839.20 - lr: 0.000029 - momentum: 0.000000
96
+ 2023-10-18 20:52:24,563 epoch 2 - iter 396/992 - loss 0.31410203 - time (sec): 9.39 - samples/sec: 6947.41 - lr: 0.000029 - momentum: 0.000000
97
+ 2023-10-18 20:52:26,781 epoch 2 - iter 495/992 - loss 0.30497184 - time (sec): 11.61 - samples/sec: 7005.44 - lr: 0.000028 - momentum: 0.000000
98
+ 2023-10-18 20:52:29,033 epoch 2 - iter 594/992 - loss 0.30271100 - time (sec): 13.86 - samples/sec: 7014.07 - lr: 0.000028 - momentum: 0.000000
99
+ 2023-10-18 20:52:31,321 epoch 2 - iter 693/992 - loss 0.30336907 - time (sec): 16.15 - samples/sec: 7001.83 - lr: 0.000028 - momentum: 0.000000
100
+ 2023-10-18 20:52:33,566 epoch 2 - iter 792/992 - loss 0.30106027 - time (sec): 18.39 - samples/sec: 7067.54 - lr: 0.000027 - momentum: 0.000000
101
+ 2023-10-18 20:52:35,840 epoch 2 - iter 891/992 - loss 0.30102108 - time (sec): 20.66 - samples/sec: 7119.36 - lr: 0.000027 - momentum: 0.000000
102
+ 2023-10-18 20:52:38,098 epoch 2 - iter 990/992 - loss 0.29802060 - time (sec): 22.92 - samples/sec: 7139.84 - lr: 0.000027 - momentum: 0.000000
103
+ 2023-10-18 20:52:38,146 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-18 20:52:38,146 EPOCH 2 done: loss 0.2981 - lr: 0.000027
105
+ 2023-10-18 20:52:39,977 DEV : loss 0.2019718885421753 - f1-score (micro avg) 0.333
106
+ 2023-10-18 20:52:39,995 saving best model
107
+ 2023-10-18 20:52:40,030 ----------------------------------------------------------------------------------------------------
108
+ 2023-10-18 20:52:42,204 epoch 3 - iter 99/992 - loss 0.25674067 - time (sec): 2.17 - samples/sec: 7699.56 - lr: 0.000026 - momentum: 0.000000
109
+ 2023-10-18 20:52:44,405 epoch 3 - iter 198/992 - loss 0.26870989 - time (sec): 4.37 - samples/sec: 7471.38 - lr: 0.000026 - momentum: 0.000000
110
+ 2023-10-18 20:52:46,670 epoch 3 - iter 297/992 - loss 0.27125039 - time (sec): 6.64 - samples/sec: 7366.32 - lr: 0.000026 - momentum: 0.000000
111
+ 2023-10-18 20:52:48,945 epoch 3 - iter 396/992 - loss 0.26816170 - time (sec): 8.91 - samples/sec: 7272.18 - lr: 0.000025 - momentum: 0.000000
112
+ 2023-10-18 20:52:51,223 epoch 3 - iter 495/992 - loss 0.26031992 - time (sec): 11.19 - samples/sec: 7295.87 - lr: 0.000025 - momentum: 0.000000
113
+ 2023-10-18 20:52:53,401 epoch 3 - iter 594/992 - loss 0.26294179 - time (sec): 13.37 - samples/sec: 7260.85 - lr: 0.000025 - momentum: 0.000000
114
+ 2023-10-18 20:52:55,672 epoch 3 - iter 693/992 - loss 0.27016265 - time (sec): 15.64 - samples/sec: 7260.38 - lr: 0.000024 - momentum: 0.000000
115
+ 2023-10-18 20:52:57,880 epoch 3 - iter 792/992 - loss 0.26656315 - time (sec): 17.85 - samples/sec: 7303.47 - lr: 0.000024 - momentum: 0.000000
116
+ 2023-10-18 20:53:00,090 epoch 3 - iter 891/992 - loss 0.26240928 - time (sec): 20.06 - samples/sec: 7327.88 - lr: 0.000024 - momentum: 0.000000
117
+ 2023-10-18 20:53:02,420 epoch 3 - iter 990/992 - loss 0.26065555 - time (sec): 22.39 - samples/sec: 7304.54 - lr: 0.000023 - momentum: 0.000000
118
+ 2023-10-18 20:53:02,472 ----------------------------------------------------------------------------------------------------
119
+ 2023-10-18 20:53:02,472 EPOCH 3 done: loss 0.2605 - lr: 0.000023
120
+ 2023-10-18 20:53:04,287 DEV : loss 0.19067135453224182 - f1-score (micro avg) 0.3635
121
+ 2023-10-18 20:53:04,306 saving best model
122
+ 2023-10-18 20:53:04,341 ----------------------------------------------------------------------------------------------------
123
+ 2023-10-18 20:53:06,510 epoch 4 - iter 99/992 - loss 0.25239806 - time (sec): 2.17 - samples/sec: 7264.69 - lr: 0.000023 - momentum: 0.000000
124
+ 2023-10-18 20:53:08,706 epoch 4 - iter 198/992 - loss 0.24991484 - time (sec): 4.36 - samples/sec: 7480.62 - lr: 0.000023 - momentum: 0.000000
125
+ 2023-10-18 20:53:10,917 epoch 4 - iter 297/992 - loss 0.24563286 - time (sec): 6.58 - samples/sec: 7280.98 - lr: 0.000022 - momentum: 0.000000
126
+ 2023-10-18 20:53:13,138 epoch 4 - iter 396/992 - loss 0.24305426 - time (sec): 8.80 - samples/sec: 7222.81 - lr: 0.000022 - momentum: 0.000000
127
+ 2023-10-18 20:53:15,385 epoch 4 - iter 495/992 - loss 0.24520370 - time (sec): 11.04 - samples/sec: 7298.48 - lr: 0.000022 - momentum: 0.000000
128
+ 2023-10-18 20:53:17,680 epoch 4 - iter 594/992 - loss 0.24264463 - time (sec): 13.34 - samples/sec: 7270.56 - lr: 0.000021 - momentum: 0.000000
129
+ 2023-10-18 20:53:19,961 epoch 4 - iter 693/992 - loss 0.24046413 - time (sec): 15.62 - samples/sec: 7253.51 - lr: 0.000021 - momentum: 0.000000
130
+ 2023-10-18 20:53:22,226 epoch 4 - iter 792/992 - loss 0.24173072 - time (sec): 17.88 - samples/sec: 7263.53 - lr: 0.000021 - momentum: 0.000000
131
+ 2023-10-18 20:53:24,493 epoch 4 - iter 891/992 - loss 0.23789589 - time (sec): 20.15 - samples/sec: 7283.37 - lr: 0.000020 - momentum: 0.000000
132
+ 2023-10-18 20:53:26,770 epoch 4 - iter 990/992 - loss 0.23782954 - time (sec): 22.43 - samples/sec: 7294.84 - lr: 0.000020 - momentum: 0.000000
133
+ 2023-10-18 20:53:26,816 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-18 20:53:26,816 EPOCH 4 done: loss 0.2376 - lr: 0.000020
135
+ 2023-10-18 20:53:28,648 DEV : loss 0.17912989854812622 - f1-score (micro avg) 0.3731
136
+ 2023-10-18 20:53:28,667 saving best model
137
+ 2023-10-18 20:53:28,703 ----------------------------------------------------------------------------------------------------
138
+ 2023-10-18 20:53:31,192 epoch 5 - iter 99/992 - loss 0.20190982 - time (sec): 2.49 - samples/sec: 6724.85 - lr: 0.000020 - momentum: 0.000000
139
+ 2023-10-18 20:53:33,466 epoch 5 - iter 198/992 - loss 0.21079062 - time (sec): 4.76 - samples/sec: 6812.55 - lr: 0.000019 - momentum: 0.000000
140
+ 2023-10-18 20:53:35,707 epoch 5 - iter 297/992 - loss 0.21616584 - time (sec): 7.00 - samples/sec: 6809.75 - lr: 0.000019 - momentum: 0.000000
141
+ 2023-10-18 20:53:37,906 epoch 5 - iter 396/992 - loss 0.21632794 - time (sec): 9.20 - samples/sec: 7007.42 - lr: 0.000019 - momentum: 0.000000
142
+ 2023-10-18 20:53:40,108 epoch 5 - iter 495/992 - loss 0.21820585 - time (sec): 11.40 - samples/sec: 7057.11 - lr: 0.000018 - momentum: 0.000000
143
+ 2023-10-18 20:53:42,289 epoch 5 - iter 594/992 - loss 0.21997889 - time (sec): 13.59 - samples/sec: 7110.83 - lr: 0.000018 - momentum: 0.000000
144
+ 2023-10-18 20:53:44,510 epoch 5 - iter 693/992 - loss 0.22087534 - time (sec): 15.81 - samples/sec: 7163.55 - lr: 0.000018 - momentum: 0.000000
145
+ 2023-10-18 20:53:46,779 epoch 5 - iter 792/992 - loss 0.21945109 - time (sec): 18.08 - samples/sec: 7190.02 - lr: 0.000017 - momentum: 0.000000
146
+ 2023-10-18 20:53:48,989 epoch 5 - iter 891/992 - loss 0.22199106 - time (sec): 20.29 - samples/sec: 7244.25 - lr: 0.000017 - momentum: 0.000000
147
+ 2023-10-18 20:53:51,198 epoch 5 - iter 990/992 - loss 0.22507917 - time (sec): 22.49 - samples/sec: 7273.09 - lr: 0.000017 - momentum: 0.000000
148
+ 2023-10-18 20:53:51,244 ----------------------------------------------------------------------------------------------------
149
+ 2023-10-18 20:53:51,244 EPOCH 5 done: loss 0.2252 - lr: 0.000017
150
+ 2023-10-18 20:53:53,099 DEV : loss 0.17070569097995758 - f1-score (micro avg) 0.388
151
+ 2023-10-18 20:53:53,117 saving best model
152
+ 2023-10-18 20:53:53,152 ----------------------------------------------------------------------------------------------------
153
+ 2023-10-18 20:53:55,176 epoch 6 - iter 99/992 - loss 0.21435041 - time (sec): 2.02 - samples/sec: 8064.65 - lr: 0.000016 - momentum: 0.000000
154
+ 2023-10-18 20:53:57,335 epoch 6 - iter 198/992 - loss 0.20832594 - time (sec): 4.18 - samples/sec: 7647.91 - lr: 0.000016 - momentum: 0.000000
155
+ 2023-10-18 20:53:59,580 epoch 6 - iter 297/992 - loss 0.20727939 - time (sec): 6.43 - samples/sec: 7625.66 - lr: 0.000016 - momentum: 0.000000
156
+ 2023-10-18 20:54:01,825 epoch 6 - iter 396/992 - loss 0.20671655 - time (sec): 8.67 - samples/sec: 7502.11 - lr: 0.000015 - momentum: 0.000000
157
+ 2023-10-18 20:54:04,029 epoch 6 - iter 495/992 - loss 0.20859057 - time (sec): 10.88 - samples/sec: 7448.22 - lr: 0.000015 - momentum: 0.000000
158
+ 2023-10-18 20:54:06,262 epoch 6 - iter 594/992 - loss 0.20604312 - time (sec): 13.11 - samples/sec: 7436.72 - lr: 0.000015 - momentum: 0.000000
159
+ 2023-10-18 20:54:08,487 epoch 6 - iter 693/992 - loss 0.20717609 - time (sec): 15.33 - samples/sec: 7394.52 - lr: 0.000014 - momentum: 0.000000
160
+ 2023-10-18 20:54:10,832 epoch 6 - iter 792/992 - loss 0.20730451 - time (sec): 17.68 - samples/sec: 7321.27 - lr: 0.000014 - momentum: 0.000000
161
+ 2023-10-18 20:54:13,157 epoch 6 - iter 891/992 - loss 0.20936951 - time (sec): 20.00 - samples/sec: 7280.12 - lr: 0.000014 - momentum: 0.000000
162
+ 2023-10-18 20:54:15,481 epoch 6 - iter 990/992 - loss 0.21222474 - time (sec): 22.33 - samples/sec: 7330.00 - lr: 0.000013 - momentum: 0.000000
163
+ 2023-10-18 20:54:15,533 ----------------------------------------------------------------------------------------------------
164
+ 2023-10-18 20:54:15,533 EPOCH 6 done: loss 0.2121 - lr: 0.000013
165
+ 2023-10-18 20:54:17,751 DEV : loss 0.16514191031455994 - f1-score (micro avg) 0.3937
166
+ 2023-10-18 20:54:17,770 saving best model
167
+ 2023-10-18 20:54:17,806 ----------------------------------------------------------------------------------------------------
168
+ 2023-10-18 20:54:20,052 epoch 7 - iter 99/992 - loss 0.21964137 - time (sec): 2.25 - samples/sec: 7179.31 - lr: 0.000013 - momentum: 0.000000
169
+ 2023-10-18 20:54:22,341 epoch 7 - iter 198/992 - loss 0.21098183 - time (sec): 4.53 - samples/sec: 7076.71 - lr: 0.000013 - momentum: 0.000000
170
+ 2023-10-18 20:54:24,694 epoch 7 - iter 297/992 - loss 0.21018550 - time (sec): 6.89 - samples/sec: 7109.21 - lr: 0.000012 - momentum: 0.000000
171
+ 2023-10-18 20:54:26,927 epoch 7 - iter 396/992 - loss 0.21566329 - time (sec): 9.12 - samples/sec: 7173.22 - lr: 0.000012 - momentum: 0.000000
172
+ 2023-10-18 20:54:29,180 epoch 7 - iter 495/992 - loss 0.20941511 - time (sec): 11.37 - samples/sec: 7197.51 - lr: 0.000012 - momentum: 0.000000
173
+ 2023-10-18 20:54:31,434 epoch 7 - iter 594/992 - loss 0.20773689 - time (sec): 13.63 - samples/sec: 7200.12 - lr: 0.000011 - momentum: 0.000000
174
+ 2023-10-18 20:54:33,652 epoch 7 - iter 693/992 - loss 0.20830501 - time (sec): 15.85 - samples/sec: 7200.02 - lr: 0.000011 - momentum: 0.000000
175
+ 2023-10-18 20:54:35,933 epoch 7 - iter 792/992 - loss 0.20710786 - time (sec): 18.13 - samples/sec: 7175.38 - lr: 0.000011 - momentum: 0.000000
176
+ 2023-10-18 20:54:38,175 epoch 7 - iter 891/992 - loss 0.20292449 - time (sec): 20.37 - samples/sec: 7254.20 - lr: 0.000010 - momentum: 0.000000
177
+ 2023-10-18 20:54:40,445 epoch 7 - iter 990/992 - loss 0.20430713 - time (sec): 22.64 - samples/sec: 7229.07 - lr: 0.000010 - momentum: 0.000000
178
+ 2023-10-18 20:54:40,490 ----------------------------------------------------------------------------------------------------
179
+ 2023-10-18 20:54:40,490 EPOCH 7 done: loss 0.2040 - lr: 0.000010
180
+ 2023-10-18 20:54:42,328 DEV : loss 0.16114094853401184 - f1-score (micro avg) 0.4081
181
+ 2023-10-18 20:54:42,346 saving best model
182
+ 2023-10-18 20:54:42,381 ----------------------------------------------------------------------------------------------------
183
+ 2023-10-18 20:54:44,630 epoch 8 - iter 99/992 - loss 0.19783164 - time (sec): 2.25 - samples/sec: 7440.34 - lr: 0.000010 - momentum: 0.000000
184
+ 2023-10-18 20:54:46,732 epoch 8 - iter 198/992 - loss 0.19042092 - time (sec): 4.35 - samples/sec: 7605.07 - lr: 0.000009 - momentum: 0.000000
185
+ 2023-10-18 20:54:48,937 epoch 8 - iter 297/992 - loss 0.19412905 - time (sec): 6.56 - samples/sec: 7746.45 - lr: 0.000009 - momentum: 0.000000
186
+ 2023-10-18 20:54:50,961 epoch 8 - iter 396/992 - loss 0.19712824 - time (sec): 8.58 - samples/sec: 7750.93 - lr: 0.000009 - momentum: 0.000000
187
+ 2023-10-18 20:54:53,123 epoch 8 - iter 495/992 - loss 0.19440607 - time (sec): 10.74 - samples/sec: 7628.27 - lr: 0.000008 - momentum: 0.000000
188
+ 2023-10-18 20:54:55,313 epoch 8 - iter 594/992 - loss 0.20059272 - time (sec): 12.93 - samples/sec: 7589.21 - lr: 0.000008 - momentum: 0.000000
189
+ 2023-10-18 20:54:57,485 epoch 8 - iter 693/992 - loss 0.19929992 - time (sec): 15.10 - samples/sec: 7557.80 - lr: 0.000008 - momentum: 0.000000
190
+ 2023-10-18 20:54:59,693 epoch 8 - iter 792/992 - loss 0.19752093 - time (sec): 17.31 - samples/sec: 7525.17 - lr: 0.000007 - momentum: 0.000000
191
+ 2023-10-18 20:55:01,938 epoch 8 - iter 891/992 - loss 0.19794149 - time (sec): 19.56 - samples/sec: 7506.92 - lr: 0.000007 - momentum: 0.000000
192
+ 2023-10-18 20:55:04,171 epoch 8 - iter 990/992 - loss 0.19798238 - time (sec): 21.79 - samples/sec: 7508.41 - lr: 0.000007 - momentum: 0.000000
193
+ 2023-10-18 20:55:04,220 ----------------------------------------------------------------------------------------------------
194
+ 2023-10-18 20:55:04,220 EPOCH 8 done: loss 0.1978 - lr: 0.000007
195
+ 2023-10-18 20:55:06,037 DEV : loss 0.16143764555454254 - f1-score (micro avg) 0.4046
196
+ 2023-10-18 20:55:06,056 ----------------------------------------------------------------------------------------------------
197
+ 2023-10-18 20:55:08,164 epoch 9 - iter 99/992 - loss 0.19886128 - time (sec): 2.11 - samples/sec: 7962.82 - lr: 0.000006 - momentum: 0.000000
198
+ 2023-10-18 20:55:10,429 epoch 9 - iter 198/992 - loss 0.20268705 - time (sec): 4.37 - samples/sec: 7900.46 - lr: 0.000006 - momentum: 0.000000
199
+ 2023-10-18 20:55:12,615 epoch 9 - iter 297/992 - loss 0.19583829 - time (sec): 6.56 - samples/sec: 7683.31 - lr: 0.000006 - momentum: 0.000000
200
+ 2023-10-18 20:55:14,782 epoch 9 - iter 396/992 - loss 0.19318088 - time (sec): 8.73 - samples/sec: 7552.63 - lr: 0.000005 - momentum: 0.000000
201
+ 2023-10-18 20:55:16,985 epoch 9 - iter 495/992 - loss 0.19300654 - time (sec): 10.93 - samples/sec: 7540.55 - lr: 0.000005 - momentum: 0.000000
202
+ 2023-10-18 20:55:19,255 epoch 9 - iter 594/992 - loss 0.19864381 - time (sec): 13.20 - samples/sec: 7477.14 - lr: 0.000005 - momentum: 0.000000
203
+ 2023-10-18 20:55:21,468 epoch 9 - iter 693/992 - loss 0.19900744 - time (sec): 15.41 - samples/sec: 7405.54 - lr: 0.000004 - momentum: 0.000000
204
+ 2023-10-18 20:55:23,682 epoch 9 - iter 792/992 - loss 0.19612859 - time (sec): 17.63 - samples/sec: 7398.32 - lr: 0.000004 - momentum: 0.000000
205
+ 2023-10-18 20:55:25,923 epoch 9 - iter 891/992 - loss 0.19581107 - time (sec): 19.87 - samples/sec: 7408.49 - lr: 0.000004 - momentum: 0.000000
206
+ 2023-10-18 20:55:28,250 epoch 9 - iter 990/992 - loss 0.19433849 - time (sec): 22.19 - samples/sec: 7373.99 - lr: 0.000003 - momentum: 0.000000
207
+ 2023-10-18 20:55:28,298 ----------------------------------------------------------------------------------------------------
208
+ 2023-10-18 20:55:28,298 EPOCH 9 done: loss 0.1941 - lr: 0.000003
209
+ 2023-10-18 20:55:30,137 DEV : loss 0.16051384806632996 - f1-score (micro avg) 0.4078
210
+ 2023-10-18 20:55:30,156 ----------------------------------------------------------------------------------------------------
211
+ 2023-10-18 20:55:32,394 epoch 10 - iter 99/992 - loss 0.20330896 - time (sec): 2.24 - samples/sec: 7007.16 - lr: 0.000003 - momentum: 0.000000
212
+ 2023-10-18 20:55:34,656 epoch 10 - iter 198/992 - loss 0.18894617 - time (sec): 4.50 - samples/sec: 7227.09 - lr: 0.000003 - momentum: 0.000000
213
+ 2023-10-18 20:55:36,948 epoch 10 - iter 297/992 - loss 0.19045374 - time (sec): 6.79 - samples/sec: 7212.88 - lr: 0.000002 - momentum: 0.000000
214
+ 2023-10-18 20:55:39,134 epoch 10 - iter 396/992 - loss 0.19170091 - time (sec): 8.98 - samples/sec: 7256.78 - lr: 0.000002 - momentum: 0.000000
215
+ 2023-10-18 20:55:41,390 epoch 10 - iter 495/992 - loss 0.18876878 - time (sec): 11.23 - samples/sec: 7289.13 - lr: 0.000002 - momentum: 0.000000
216
+ 2023-10-18 20:55:43,609 epoch 10 - iter 594/992 - loss 0.19227055 - time (sec): 13.45 - samples/sec: 7327.60 - lr: 0.000001 - momentum: 0.000000
217
+ 2023-10-18 20:55:45,880 epoch 10 - iter 693/992 - loss 0.19144431 - time (sec): 15.72 - samples/sec: 7322.12 - lr: 0.000001 - momentum: 0.000000
218
+ 2023-10-18 20:55:48,086 epoch 10 - iter 792/992 - loss 0.19069240 - time (sec): 17.93 - samples/sec: 7344.98 - lr: 0.000001 - momentum: 0.000000
219
+ 2023-10-18 20:55:50,261 epoch 10 - iter 891/992 - loss 0.19135351 - time (sec): 20.11 - samples/sec: 7328.13 - lr: 0.000000 - momentum: 0.000000
220
+ 2023-10-18 20:55:52,556 epoch 10 - iter 990/992 - loss 0.19175148 - time (sec): 22.40 - samples/sec: 7310.16 - lr: 0.000000 - momentum: 0.000000
221
+ 2023-10-18 20:55:52,602 ----------------------------------------------------------------------------------------------------
222
+ 2023-10-18 20:55:52,602 EPOCH 10 done: loss 0.1919 - lr: 0.000000
223
+ 2023-10-18 20:55:54,456 DEV : loss 0.16066327691078186 - f1-score (micro avg) 0.4119
224
+ 2023-10-18 20:55:54,475 saving best model
225
+ 2023-10-18 20:55:54,537 ----------------------------------------------------------------------------------------------------
226
+ 2023-10-18 20:55:54,538 Loading model from best epoch ...
227
+ 2023-10-18 20:55:54,607 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 20:55:56,064
229
+ Results:
230
+ - F-score (micro) 0.4631
231
+ - F-score (macro) 0.3022
232
+ - Accuracy 0.3478
233
+
234
+ By class:
235
+ precision recall f1-score support
236
+
237
+ LOC 0.6305 0.5679 0.5976 655
238
+ PER 0.2269 0.4843 0.3090 223
239
+ ORG 0.0000 0.0000 0.0000 127
240
+
241
+ micro avg 0.4494 0.4776 0.4631 1005
242
+ macro avg 0.2858 0.3507 0.3022 1005
243
+ weighted avg 0.4613 0.4776 0.4580 1005
244
+
245
+ 2023-10-18 20:55:56,064 ----------------------------------------------------------------------------------------------------