Hugging Face's logo

stefan-it
/
hmbench-newseye-fr-hmbert-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-5

Token Classification
Flair
PyTorch
French
sequence-tagger-model
Model card Files Community
stefan-it commited on
Commit
d4b11ff
•
1 Parent(s): 3a670fd

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 +244 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:77df3601c90b40c4b79c69b5c786f0dd16abec4cca1919d7289cb84fc1760dd7
3
+ size 443323527
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 13:44:58 0.0000 0.5208 0.1208 0.6651 0.7537 0.7066 0.5711
3
+ 2 13:46:10 0.0000 0.1055 0.1181 0.6479 0.7837 0.7094 0.5720
4
+ 3 13:47:24 0.0000 0.0644 0.1145 0.7673 0.7986 0.7827 0.6716
5
+ 4 13:48:38 0.0000 0.0467 0.1236 0.7663 0.7986 0.7821 0.6633
6
+ 5 13:49:50 0.0000 0.0358 0.1685 0.7762 0.8259 0.8003 0.6859
7
+ 6 13:51:04 0.0000 0.0290 0.1786 0.7882 0.8150 0.8013 0.6861
8
+ 7 13:52:17 0.0000 0.0239 0.1898 0.7955 0.8150 0.8051 0.6957
9
+ 8 13:53:31 0.0000 0.0175 0.1964 0.7899 0.8082 0.7989 0.6891
10
+ 9 13:54:45 0.0000 0.0132 0.1960 0.8128 0.8272 0.8200 0.7178
11
+ 10 13:55:58 0.0000 0.0115 0.1951 0.8112 0.8299 0.8204 0.7185
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,244 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-16 13:43:46,897 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-16 13:43:46,898 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=17, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-16 13:43:46,898 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-16 13:43:46,898 MultiCorpus: 7142 train + 698 dev + 2570 test sentences
52
+ - NER_HIPE_2022 Corpus: 7142 train + 698 dev + 2570 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fr/with_doc_seperator
53
+ 2023-10-16 13:43:46,898 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-16 13:43:46,898 Train: 7142 sentences
55
+ 2023-10-16 13:43:46,898 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-16 13:43:46,899 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-16 13:43:46,899 Training Params:
58
+ 2023-10-16 13:43:46,899 - learning_rate: "3e-05"
59
+ 2023-10-16 13:43:46,899 - mini_batch_size: "8"
60
+ 2023-10-16 13:43:46,899 - max_epochs: "10"
61
+ 2023-10-16 13:43:46,899 - shuffle: "True"
62
+ 2023-10-16 13:43:46,899 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-16 13:43:46,899 Plugins:
64
+ 2023-10-16 13:43:46,899 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-16 13:43:46,899 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-16 13:43:46,899 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-16 13:43:46,899 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-16 13:43:46,899 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-16 13:43:46,899 Computation:
70
+ 2023-10-16 13:43:46,899 - compute on device: cuda:0
71
+ 2023-10-16 13:43:46,899 - embedding storage: none
72
+ 2023-10-16 13:43:46,899 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-16 13:43:46,899 Model training base path: "hmbench-newseye/fr-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-5"
74
+ 2023-10-16 13:43:46,899 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-16 13:43:46,899 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-16 13:43:53,652 epoch 1 - iter 89/893 - loss 2.49154230 - time (sec): 6.75 - samples/sec: 3658.17 - lr: 0.000003 - momentum: 0.000000
77
+ 2023-10-16 13:44:00,646 epoch 1 - iter 178/893 - loss 1.62126567 - time (sec): 13.75 - samples/sec: 3614.22 - lr: 0.000006 - momentum: 0.000000
78
+ 2023-10-16 13:44:07,488 epoch 1 - iter 267/893 - loss 1.22582402 - time (sec): 20.59 - samples/sec: 3653.03 - lr: 0.000009 - momentum: 0.000000
79
+ 2023-10-16 13:44:14,523 epoch 1 - iter 356/893 - loss 0.99528059 - time (sec): 27.62 - samples/sec: 3661.62 - lr: 0.000012 - momentum: 0.000000
80
+ 2023-10-16 13:44:21,361 epoch 1 - iter 445/893 - loss 0.85563832 - time (sec): 34.46 - samples/sec: 3639.09 - lr: 0.000015 - momentum: 0.000000
81
+ 2023-10-16 13:44:27,857 epoch 1 - iter 534/893 - loss 0.75454568 - time (sec): 40.96 - samples/sec: 3635.00 - lr: 0.000018 - momentum: 0.000000
82
+ 2023-10-16 13:44:34,891 epoch 1 - iter 623/893 - loss 0.67109967 - time (sec): 47.99 - samples/sec: 3647.33 - lr: 0.000021 - momentum: 0.000000
83
+ 2023-10-16 13:44:41,532 epoch 1 - iter 712/893 - loss 0.61218176 - time (sec): 54.63 - samples/sec: 3661.46 - lr: 0.000024 - momentum: 0.000000
84
+ 2023-10-16 13:44:48,148 epoch 1 - iter 801/893 - loss 0.56314889 - time (sec): 61.25 - samples/sec: 3652.60 - lr: 0.000027 - momentum: 0.000000
85
+ 2023-10-16 13:44:55,012 epoch 1 - iter 890/893 - loss 0.52200864 - time (sec): 68.11 - samples/sec: 3640.49 - lr: 0.000030 - momentum: 0.000000
86
+ 2023-10-16 13:44:55,226 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-16 13:44:55,226 EPOCH 1 done: loss 0.5208 - lr: 0.000030
88
+ 2023-10-16 13:44:58,272 DEV : loss 0.12075067311525345 - f1-score (micro avg) 0.7066
89
+ 2023-10-16 13:44:58,288 saving best model
90
+ 2023-10-16 13:44:58,731 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-16 13:45:05,320 epoch 2 - iter 89/893 - loss 0.13241194 - time (sec): 6.59 - samples/sec: 3790.57 - lr: 0.000030 - momentum: 0.000000
92
+ 2023-10-16 13:45:12,294 epoch 2 - iter 178/893 - loss 0.12021519 - time (sec): 13.56 - samples/sec: 3795.99 - lr: 0.000029 - momentum: 0.000000
93
+ 2023-10-16 13:45:19,078 epoch 2 - iter 267/893 - loss 0.11750837 - time (sec): 20.35 - samples/sec: 3792.31 - lr: 0.000029 - momentum: 0.000000
94
+ 2023-10-16 13:45:25,904 epoch 2 - iter 356/893 - loss 0.11841414 - time (sec): 27.17 - samples/sec: 3728.66 - lr: 0.000029 - momentum: 0.000000
95
+ 2023-10-16 13:45:32,684 epoch 2 - iter 445/893 - loss 0.11544678 - time (sec): 33.95 - samples/sec: 3718.33 - lr: 0.000028 - momentum: 0.000000
96
+ 2023-10-16 13:45:39,485 epoch 2 - iter 534/893 - loss 0.11398255 - time (sec): 40.75 - samples/sec: 3711.64 - lr: 0.000028 - momentum: 0.000000
97
+ 2023-10-16 13:45:46,276 epoch 2 - iter 623/893 - loss 0.11078897 - time (sec): 47.54 - samples/sec: 3690.56 - lr: 0.000028 - momentum: 0.000000
98
+ 2023-10-16 13:45:52,795 epoch 2 - iter 712/893 - loss 0.10867892 - time (sec): 54.06 - samples/sec: 3704.94 - lr: 0.000027 - momentum: 0.000000
99
+ 2023-10-16 13:45:59,583 epoch 2 - iter 801/893 - loss 0.10557874 - time (sec): 60.85 - samples/sec: 3706.50 - lr: 0.000027 - momentum: 0.000000
100
+ 2023-10-16 13:46:05,978 epoch 2 - iter 890/893 - loss 0.10564723 - time (sec): 67.25 - samples/sec: 3683.34 - lr: 0.000027 - momentum: 0.000000
101
+ 2023-10-16 13:46:06,201 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-16 13:46:06,201 EPOCH 2 done: loss 0.1055 - lr: 0.000027
103
+ 2023-10-16 13:46:10,199 DEV : loss 0.11806550621986389 - f1-score (micro avg) 0.7094
104
+ 2023-10-16 13:46:10,215 saving best model
105
+ 2023-10-16 13:46:10,792 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-16 13:46:17,508 epoch 3 - iter 89/893 - loss 0.06756403 - time (sec): 6.71 - samples/sec: 3546.43 - lr: 0.000026 - momentum: 0.000000
107
+ 2023-10-16 13:46:24,311 epoch 3 - iter 178/893 - loss 0.06905887 - time (sec): 13.52 - samples/sec: 3559.14 - lr: 0.000026 - momentum: 0.000000
108
+ 2023-10-16 13:46:31,239 epoch 3 - iter 267/893 - loss 0.06890371 - time (sec): 20.44 - samples/sec: 3556.52 - lr: 0.000026 - momentum: 0.000000
109
+ 2023-10-16 13:46:38,289 epoch 3 - iter 356/893 - loss 0.06587791 - time (sec): 27.49 - samples/sec: 3546.44 - lr: 0.000025 - momentum: 0.000000
110
+ 2023-10-16 13:46:45,243 epoch 3 - iter 445/893 - loss 0.06381783 - time (sec): 34.45 - samples/sec: 3560.17 - lr: 0.000025 - momentum: 0.000000
111
+ 2023-10-16 13:46:51,806 epoch 3 - iter 534/893 - loss 0.06384950 - time (sec): 41.01 - samples/sec: 3583.39 - lr: 0.000025 - momentum: 0.000000
112
+ 2023-10-16 13:46:59,175 epoch 3 - iter 623/893 - loss 0.06292702 - time (sec): 48.38 - samples/sec: 3566.95 - lr: 0.000024 - momentum: 0.000000
113
+ 2023-10-16 13:47:06,251 epoch 3 - iter 712/893 - loss 0.06408425 - time (sec): 55.45 - samples/sec: 3577.80 - lr: 0.000024 - momentum: 0.000000
114
+ 2023-10-16 13:47:13,444 epoch 3 - iter 801/893 - loss 0.06538790 - time (sec): 62.65 - samples/sec: 3554.54 - lr: 0.000024 - momentum: 0.000000
115
+ 2023-10-16 13:47:20,279 epoch 3 - iter 890/893 - loss 0.06456476 - time (sec): 69.48 - samples/sec: 3566.45 - lr: 0.000023 - momentum: 0.000000
116
+ 2023-10-16 13:47:20,513 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-16 13:47:20,514 EPOCH 3 done: loss 0.0644 - lr: 0.000023
118
+ 2023-10-16 13:47:24,535 DEV : loss 0.11452696472406387 - f1-score (micro avg) 0.7827
119
+ 2023-10-16 13:47:24,551 saving best model
120
+ 2023-10-16 13:47:25,042 ----------------------------------------------------------------------------------------------------
121
+ 2023-10-16 13:47:32,102 epoch 4 - iter 89/893 - loss 0.04287257 - time (sec): 7.06 - samples/sec: 3684.25 - lr: 0.000023 - momentum: 0.000000
122
+ 2023-10-16 13:47:39,186 epoch 4 - iter 178/893 - loss 0.04063102 - time (sec): 14.14 - samples/sec: 3584.36 - lr: 0.000023 - momentum: 0.000000
123
+ 2023-10-16 13:47:45,973 epoch 4 - iter 267/893 - loss 0.04147794 - time (sec): 20.93 - samples/sec: 3584.08 - lr: 0.000022 - momentum: 0.000000
124
+ 2023-10-16 13:47:53,237 epoch 4 - iter 356/893 - loss 0.04278598 - time (sec): 28.19 - samples/sec: 3559.01 - lr: 0.000022 - momentum: 0.000000
125
+ 2023-10-16 13:47:59,557 epoch 4 - iter 445/893 - loss 0.04283638 - time (sec): 34.51 - samples/sec: 3604.20 - lr: 0.000022 - momentum: 0.000000
126
+ 2023-10-16 13:48:06,337 epoch 4 - iter 534/893 - loss 0.04347932 - time (sec): 41.29 - samples/sec: 3619.66 - lr: 0.000021 - momentum: 0.000000
127
+ 2023-10-16 13:48:12,846 epoch 4 - iter 623/893 - loss 0.04461082 - time (sec): 47.80 - samples/sec: 3609.56 - lr: 0.000021 - momentum: 0.000000
128
+ 2023-10-16 13:48:19,452 epoch 4 - iter 712/893 - loss 0.04492176 - time (sec): 54.41 - samples/sec: 3624.66 - lr: 0.000021 - momentum: 0.000000
129
+ 2023-10-16 13:48:26,362 epoch 4 - iter 801/893 - loss 0.04597372 - time (sec): 61.32 - samples/sec: 3619.82 - lr: 0.000020 - momentum: 0.000000
130
+ 2023-10-16 13:48:33,542 epoch 4 - iter 890/893 - loss 0.04676504 - time (sec): 68.50 - samples/sec: 3619.95 - lr: 0.000020 - momentum: 0.000000
131
+ 2023-10-16 13:48:33,772 ----------------------------------------------------------------------------------------------------
132
+ 2023-10-16 13:48:33,772 EPOCH 4 done: loss 0.0467 - lr: 0.000020
133
+ 2023-10-16 13:48:38,330 DEV : loss 0.1235646903514862 - f1-score (micro avg) 0.7821
134
+ 2023-10-16 13:48:38,345 ----------------------------------------------------------------------------------------------------
135
+ 2023-10-16 13:48:45,005 epoch 5 - iter 89/893 - loss 0.02387243 - time (sec): 6.66 - samples/sec: 3482.39 - lr: 0.000020 - momentum: 0.000000
136
+ 2023-10-16 13:48:52,056 epoch 5 - iter 178/893 - loss 0.03032860 - time (sec): 13.71 - samples/sec: 3679.13 - lr: 0.000019 - momentum: 0.000000
137
+ 2023-10-16 13:48:58,977 epoch 5 - iter 267/893 - loss 0.03259763 - time (sec): 20.63 - samples/sec: 3668.08 - lr: 0.000019 - momentum: 0.000000
138
+ 2023-10-16 13:49:05,752 epoch 5 - iter 356/893 - loss 0.03570155 - time (sec): 27.41 - samples/sec: 3680.40 - lr: 0.000019 - momentum: 0.000000
139
+ 2023-10-16 13:49:12,035 epoch 5 - iter 445/893 - loss 0.03481697 - time (sec): 33.69 - samples/sec: 3660.13 - lr: 0.000018 - momentum: 0.000000
140
+ 2023-10-16 13:49:19,355 epoch 5 - iter 534/893 - loss 0.03425923 - time (sec): 41.01 - samples/sec: 3658.78 - lr: 0.000018 - momentum: 0.000000
141
+ 2023-10-16 13:49:26,083 epoch 5 - iter 623/893 - loss 0.03439068 - time (sec): 47.74 - samples/sec: 3652.14 - lr: 0.000018 - momentum: 0.000000
142
+ 2023-10-16 13:49:32,693 epoch 5 - iter 712/893 - loss 0.03544740 - time (sec): 54.35 - samples/sec: 3649.21 - lr: 0.000017 - momentum: 0.000000
143
+ 2023-10-16 13:49:39,657 epoch 5 - iter 801/893 - loss 0.03607589 - time (sec): 61.31 - samples/sec: 3633.76 - lr: 0.000017 - momentum: 0.000000
144
+ 2023-10-16 13:49:46,685 epoch 5 - iter 890/893 - loss 0.03580004 - time (sec): 68.34 - samples/sec: 3632.42 - lr: 0.000017 - momentum: 0.000000
145
+ 2023-10-16 13:49:46,906 ----------------------------------------------------------------------------------------------------
146
+ 2023-10-16 13:49:46,906 EPOCH 5 done: loss 0.0358 - lr: 0.000017
147
+ 2023-10-16 13:49:50,900 DEV : loss 0.16850945353507996 - f1-score (micro avg) 0.8003
148
+ 2023-10-16 13:49:50,916 saving best model
149
+ 2023-10-16 13:49:52,046 ----------------------------------------------------------------------------------------------------
150
+ 2023-10-16 13:49:58,748 epoch 6 - iter 89/893 - loss 0.02777395 - time (sec): 6.70 - samples/sec: 3551.20 - lr: 0.000016 - momentum: 0.000000
151
+ 2023-10-16 13:50:05,732 epoch 6 - iter 178/893 - loss 0.02629674 - time (sec): 13.68 - samples/sec: 3643.60 - lr: 0.000016 - momentum: 0.000000
152
+ 2023-10-16 13:50:12,637 epoch 6 - iter 267/893 - loss 0.02593427 - time (sec): 20.59 - samples/sec: 3624.05 - lr: 0.000016 - momentum: 0.000000
153
+ 2023-10-16 13:50:19,772 epoch 6 - iter 356/893 - loss 0.02773488 - time (sec): 27.72 - samples/sec: 3605.78 - lr: 0.000015 - momentum: 0.000000
154
+ 2023-10-16 13:50:26,258 epoch 6 - iter 445/893 - loss 0.02870583 - time (sec): 34.21 - samples/sec: 3627.65 - lr: 0.000015 - momentum: 0.000000
155
+ 2023-10-16 13:50:33,263 epoch 6 - iter 534/893 - loss 0.02775510 - time (sec): 41.21 - samples/sec: 3667.80 - lr: 0.000015 - momentum: 0.000000
156
+ 2023-10-16 13:50:39,881 epoch 6 - iter 623/893 - loss 0.02870404 - time (sec): 47.83 - samples/sec: 3678.92 - lr: 0.000014 - momentum: 0.000000
157
+ 2023-10-16 13:50:46,806 epoch 6 - iter 712/893 - loss 0.02927233 - time (sec): 54.76 - samples/sec: 3670.61 - lr: 0.000014 - momentum: 0.000000
158
+ 2023-10-16 13:50:53,440 epoch 6 - iter 801/893 - loss 0.02905059 - time (sec): 61.39 - samples/sec: 3651.61 - lr: 0.000014 - momentum: 0.000000
159
+ 2023-10-16 13:51:00,351 epoch 6 - iter 890/893 - loss 0.02892722 - time (sec): 68.30 - samples/sec: 3634.00 - lr: 0.000013 - momentum: 0.000000
160
+ 2023-10-16 13:51:00,543 ----------------------------------------------------------------------------------------------------
161
+ 2023-10-16 13:51:00,543 EPOCH 6 done: loss 0.0290 - lr: 0.000013
162
+ 2023-10-16 13:51:04,562 DEV : loss 0.17858995497226715 - f1-score (micro avg) 0.8013
163
+ 2023-10-16 13:51:04,578 saving best model
164
+ 2023-10-16 13:51:05,103 ----------------------------------------------------------------------------------------------------
165
+ 2023-10-16 13:51:11,331 epoch 7 - iter 89/893 - loss 0.01538452 - time (sec): 6.22 - samples/sec: 3772.50 - lr: 0.000013 - momentum: 0.000000
166
+ 2023-10-16 13:51:17,722 epoch 7 - iter 178/893 - loss 0.02131337 - time (sec): 12.62 - samples/sec: 3726.95 - lr: 0.000013 - momentum: 0.000000
167
+ 2023-10-16 13:51:24,347 epoch 7 - iter 267/893 - loss 0.02097900 - time (sec): 19.24 - samples/sec: 3757.18 - lr: 0.000012 - momentum: 0.000000
168
+ 2023-10-16 13:51:31,181 epoch 7 - iter 356/893 - loss 0.02190829 - time (sec): 26.07 - samples/sec: 3745.44 - lr: 0.000012 - momentum: 0.000000
169
+ 2023-10-16 13:51:37,865 epoch 7 - iter 445/893 - loss 0.02094727 - time (sec): 32.76 - samples/sec: 3702.99 - lr: 0.000012 - momentum: 0.000000
170
+ 2023-10-16 13:51:44,624 epoch 7 - iter 534/893 - loss 0.02163093 - time (sec): 39.52 - samples/sec: 3697.98 - lr: 0.000011 - momentum: 0.000000
171
+ 2023-10-16 13:51:51,853 epoch 7 - iter 623/893 - loss 0.02291209 - time (sec): 46.75 - samples/sec: 3689.77 - lr: 0.000011 - momentum: 0.000000
172
+ 2023-10-16 13:51:59,294 epoch 7 - iter 712/893 - loss 0.02293232 - time (sec): 54.19 - samples/sec: 3678.83 - lr: 0.000011 - momentum: 0.000000
173
+ 2023-10-16 13:52:06,044 epoch 7 - iter 801/893 - loss 0.02333221 - time (sec): 60.94 - samples/sec: 3662.26 - lr: 0.000010 - momentum: 0.000000
174
+ 2023-10-16 13:52:12,889 epoch 7 - iter 890/893 - loss 0.02382874 - time (sec): 67.78 - samples/sec: 3659.34 - lr: 0.000010 - momentum: 0.000000
175
+ 2023-10-16 13:52:13,099 ----------------------------------------------------------------------------------------------------
176
+ 2023-10-16 13:52:13,099 EPOCH 7 done: loss 0.0239 - lr: 0.000010
177
+ 2023-10-16 13:52:17,661 DEV : loss 0.18977805972099304 - f1-score (micro avg) 0.8051
178
+ 2023-10-16 13:52:17,677 saving best model
179
+ 2023-10-16 13:52:18,230 ----------------------------------------------------------------------------------------------------
180
+ 2023-10-16 13:52:25,001 epoch 8 - iter 89/893 - loss 0.01760355 - time (sec): 6.77 - samples/sec: 3525.99 - lr: 0.000010 - momentum: 0.000000
181
+ 2023-10-16 13:52:31,737 epoch 8 - iter 178/893 - loss 0.01600479 - time (sec): 13.51 - samples/sec: 3609.06 - lr: 0.000009 - momentum: 0.000000
182
+ 2023-10-16 13:52:38,483 epoch 8 - iter 267/893 - loss 0.01545751 - time (sec): 20.25 - samples/sec: 3645.30 - lr: 0.000009 - momentum: 0.000000
183
+ 2023-10-16 13:52:45,395 epoch 8 - iter 356/893 - loss 0.01612712 - time (sec): 27.16 - samples/sec: 3651.36 - lr: 0.000009 - momentum: 0.000000
184
+ 2023-10-16 13:52:51,691 epoch 8 - iter 445/893 - loss 0.01694443 - time (sec): 33.46 - samples/sec: 3648.63 - lr: 0.000008 - momentum: 0.000000
185
+ 2023-10-16 13:52:58,579 epoch 8 - iter 534/893 - loss 0.01655607 - time (sec): 40.35 - samples/sec: 3624.80 - lr: 0.000008 - momentum: 0.000000
186
+ 2023-10-16 13:53:05,827 epoch 8 - iter 623/893 - loss 0.01701668 - time (sec): 47.60 - samples/sec: 3641.53 - lr: 0.000008 - momentum: 0.000000
187
+ 2023-10-16 13:53:13,155 epoch 8 - iter 712/893 - loss 0.01685952 - time (sec): 54.92 - samples/sec: 3628.71 - lr: 0.000007 - momentum: 0.000000
188
+ 2023-10-16 13:53:19,500 epoch 8 - iter 801/893 - loss 0.01672613 - time (sec): 61.27 - samples/sec: 3625.67 - lr: 0.000007 - momentum: 0.000000
189
+ 2023-10-16 13:53:26,651 epoch 8 - iter 890/893 - loss 0.01754531 - time (sec): 68.42 - samples/sec: 3618.81 - lr: 0.000007 - momentum: 0.000000
190
+ 2023-10-16 13:53:26,999 ----------------------------------------------------------------------------------------------------
191
+ 2023-10-16 13:53:26,999 EPOCH 8 done: loss 0.0175 - lr: 0.000007
192
+ 2023-10-16 13:53:31,625 DEV : loss 0.19642271101474762 - f1-score (micro avg) 0.7989
193
+ 2023-10-16 13:53:31,642 ----------------------------------------------------------------------------------------------------
194
+ 2023-10-16 13:53:38,444 epoch 9 - iter 89/893 - loss 0.01210769 - time (sec): 6.80 - samples/sec: 3501.99 - lr: 0.000006 - momentum: 0.000000
195
+ 2023-10-16 13:53:45,076 epoch 9 - iter 178/893 - loss 0.01089980 - time (sec): 13.43 - samples/sec: 3607.10 - lr: 0.000006 - momentum: 0.000000
196
+ 2023-10-16 13:53:51,952 epoch 9 - iter 267/893 - loss 0.01206946 - time (sec): 20.31 - samples/sec: 3558.60 - lr: 0.000006 - momentum: 0.000000
197
+ 2023-10-16 13:53:58,996 epoch 9 - iter 356/893 - loss 0.01230578 - time (sec): 27.35 - samples/sec: 3545.34 - lr: 0.000005 - momentum: 0.000000
198
+ 2023-10-16 13:54:05,513 epoch 9 - iter 445/893 - loss 0.01310413 - time (sec): 33.87 - samples/sec: 3557.03 - lr: 0.000005 - momentum: 0.000000
199
+ 2023-10-16 13:54:12,606 epoch 9 - iter 534/893 - loss 0.01304358 - time (sec): 40.96 - samples/sec: 3563.63 - lr: 0.000005 - momentum: 0.000000
200
+ 2023-10-16 13:54:19,791 epoch 9 - iter 623/893 - loss 0.01306085 - time (sec): 48.15 - samples/sec: 3546.44 - lr: 0.000004 - momentum: 0.000000
201
+ 2023-10-16 13:54:26,979 epoch 9 - iter 712/893 - loss 0.01371970 - time (sec): 55.34 - samples/sec: 3565.24 - lr: 0.000004 - momentum: 0.000000
202
+ 2023-10-16 13:54:33,781 epoch 9 - iter 801/893 - loss 0.01284582 - time (sec): 62.14 - samples/sec: 3588.43 - lr: 0.000004 - momentum: 0.000000
203
+ 2023-10-16 13:54:40,711 epoch 9 - iter 890/893 - loss 0.01318723 - time (sec): 69.07 - samples/sec: 3589.78 - lr: 0.000003 - momentum: 0.000000
204
+ 2023-10-16 13:54:40,933 ----------------------------------------------------------------------------------------------------
205
+ 2023-10-16 13:54:40,933 EPOCH 9 done: loss 0.0132 - lr: 0.000003
206
+ 2023-10-16 13:54:44,996 DEV : loss 0.19601404666900635 - f1-score (micro avg) 0.82
207
+ 2023-10-16 13:54:45,012 saving best model
208
+ 2023-10-16 13:54:45,569 ----------------------------------------------------------------------------------------------------
209
+ 2023-10-16 13:54:52,458 epoch 10 - iter 89/893 - loss 0.01232102 - time (sec): 6.89 - samples/sec: 3580.68 - lr: 0.000003 - momentum: 0.000000
210
+ 2023-10-16 13:54:59,061 epoch 10 - iter 178/893 - loss 0.01172762 - time (sec): 13.49 - samples/sec: 3553.37 - lr: 0.000003 - momentum: 0.000000
211
+ 2023-10-16 13:55:05,897 epoch 10 - iter 267/893 - loss 0.01214811 - time (sec): 20.33 - samples/sec: 3580.72 - lr: 0.000002 - momentum: 0.000000
212
+ 2023-10-16 13:55:12,611 epoch 10 - iter 356/893 - loss 0.01159662 - time (sec): 27.04 - samples/sec: 3624.28 - lr: 0.000002 - momentum: 0.000000
213
+ 2023-10-16 13:55:19,351 epoch 10 - iter 445/893 - loss 0.01125182 - time (sec): 33.78 - samples/sec: 3627.33 - lr: 0.000002 - momentum: 0.000000
214
+ 2023-10-16 13:55:26,203 epoch 10 - iter 534/893 - loss 0.01148185 - time (sec): 40.63 - samples/sec: 3623.21 - lr: 0.000001 - momentum: 0.000000
215
+ 2023-10-16 13:55:32,925 epoch 10 - iter 623/893 - loss 0.01082743 - time (sec): 47.35 - samples/sec: 3628.25 - lr: 0.000001 - momentum: 0.000000
216
+ 2023-10-16 13:55:39,767 epoch 10 - iter 712/893 - loss 0.01097241 - time (sec): 54.20 - samples/sec: 3633.23 - lr: 0.000001 - momentum: 0.000000
217
+ 2023-10-16 13:55:46,432 epoch 10 - iter 801/893 - loss 0.01107939 - time (sec): 60.86 - samples/sec: 3640.46 - lr: 0.000000 - momentum: 0.000000
218
+ 2023-10-16 13:55:53,393 epoch 10 - iter 890/893 - loss 0.01153753 - time (sec): 67.82 - samples/sec: 3653.55 - lr: 0.000000 - momentum: 0.000000
219
+ 2023-10-16 13:55:53,634 ----------------------------------------------------------------------------------------------------
220
+ 2023-10-16 13:55:53,634 EPOCH 10 done: loss 0.0115 - lr: 0.000000
221
+ 2023-10-16 13:55:58,213 DEV : loss 0.19514599442481995 - f1-score (micro avg) 0.8204
222
+ 2023-10-16 13:55:58,230 saving best model
223
+ 2023-10-16 13:55:59,218 ----------------------------------------------------------------------------------------------------
224
+ 2023-10-16 13:55:59,219 Loading model from best epoch ...
225
+ 2023-10-16 13:56:00,770 SequenceTagger predicts: Dictionary with 17 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, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
226
+ 2023-10-16 13:56:10,110
227
+ Results:
228
+ - F-score (micro) 0.6969
229
+ - F-score (macro) 0.6147
230
+ - Accuracy 0.559
231
+
232
+ By class:
233
+ precision recall f1-score support
234
+
235
+ LOC 0.7240 0.6995 0.7116 1095
236
+ PER 0.7760 0.7806 0.7783 1012
237
+ ORG 0.4223 0.5938 0.4936 357
238
+ HumanProd 0.3529 0.7273 0.4752 33
239
+
240
+ micro avg 0.6772 0.7177 0.6969 2497
241
+ macro avg 0.5688 0.7003 0.6147 2497
242
+ weighted avg 0.6971 0.7177 0.7043 2497
243
+
244
+ 2023-10-16 13:56:10,110 ----------------------------------------------------------------------------------------------------