Hugging Face's logo

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

Token Classification
Flair
PyTorch
French
sequence-tagger-model
Model card Files Community
stefan-it commited on
Commit
dbc6bd5
·
1 Parent(s): b68465f

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 +241 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0202e7c2a50531707315d754c5265a5f92d2cfd5ddd08612494fff3f8ee70dd4
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 08:31:35 0.0000 0.4535 0.1111 0.7035 0.7619 0.7315 0.5964
3
+ 2 08:32:47 0.0000 0.1078 0.1104 0.7268 0.7565 0.7413 0.6157
4
+ 3 08:34:02 0.0000 0.0691 0.1074 0.7372 0.7823 0.7591 0.6340
5
+ 4 08:35:17 0.0000 0.0492 0.1580 0.7345 0.7905 0.7615 0.6413
6
+ 5 08:36:31 0.0000 0.0386 0.1329 0.8105 0.7973 0.8038 0.6910
7
+ 6 08:37:44 0.0000 0.0294 0.1681 0.7992 0.8068 0.8030 0.6895
8
+ 7 08:38:58 0.0000 0.0223 0.1885 0.7739 0.8150 0.7939 0.6799
9
+ 8 08:40:11 0.0000 0.0168 0.1790 0.7843 0.8163 0.8000 0.6912
10
+ 9 08:41:24 0.0000 0.0121 0.1874 0.7957 0.8109 0.8032 0.6938
11
+ 10 08:42:37 0.0000 0.0081 0.2023 0.7923 0.8095 0.8008 0.6927
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,241 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-16 08:30:24,478 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-16 08:30:24,479 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 08:30:24,479 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-16 08:30:24,479 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 08:30:24,479 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-16 08:30:24,479 Train: 7142 sentences
55
+ 2023-10-16 08:30:24,479 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-16 08:30:24,479 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-16 08:30:24,479 Training Params:
58
+ 2023-10-16 08:30:24,480 - learning_rate: "5e-05"
59
+ 2023-10-16 08:30:24,480 - mini_batch_size: "8"
60
+ 2023-10-16 08:30:24,480 - max_epochs: "10"
61
+ 2023-10-16 08:30:24,480 - shuffle: "True"
62
+ 2023-10-16 08:30:24,480 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-16 08:30:24,480 Plugins:
64
+ 2023-10-16 08:30:24,480 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-16 08:30:24,480 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-16 08:30:24,480 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-16 08:30:24,480 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-16 08:30:24,480 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-16 08:30:24,480 Computation:
70
+ 2023-10-16 08:30:24,480 - compute on device: cuda:0
71
+ 2023-10-16 08:30:24,480 - embedding storage: none
72
+ 2023-10-16 08:30:24,480 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-16 08:30:24,480 Model training base path: "hmbench-newseye/fr-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1"
74
+ 2023-10-16 08:30:24,480 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-16 08:30:24,480 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-16 08:30:31,442 epoch 1 - iter 89/893 - loss 2.29929295 - time (sec): 6.96 - samples/sec: 3536.66 - lr: 0.000005 - momentum: 0.000000
77
+ 2023-10-16 08:30:38,127 epoch 1 - iter 178/893 - loss 1.42154576 - time (sec): 13.65 - samples/sec: 3658.06 - lr: 0.000010 - momentum: 0.000000
78
+ 2023-10-16 08:30:44,918 epoch 1 - iter 267/893 - loss 1.07378330 - time (sec): 20.44 - samples/sec: 3647.03 - lr: 0.000015 - momentum: 0.000000
79
+ 2023-10-16 08:30:52,159 epoch 1 - iter 356/893 - loss 0.86632795 - time (sec): 27.68 - samples/sec: 3637.11 - lr: 0.000020 - momentum: 0.000000
80
+ 2023-10-16 08:30:58,762 epoch 1 - iter 445/893 - loss 0.74218889 - time (sec): 34.28 - samples/sec: 3634.12 - lr: 0.000025 - momentum: 0.000000
81
+ 2023-10-16 08:31:05,491 epoch 1 - iter 534/893 - loss 0.65263176 - time (sec): 41.01 - samples/sec: 3622.44 - lr: 0.000030 - momentum: 0.000000
82
+ 2023-10-16 08:31:12,213 epoch 1 - iter 623/893 - loss 0.58561512 - time (sec): 47.73 - samples/sec: 3632.46 - lr: 0.000035 - momentum: 0.000000
83
+ 2023-10-16 08:31:19,397 epoch 1 - iter 712/893 - loss 0.52966299 - time (sec): 54.92 - samples/sec: 3606.30 - lr: 0.000040 - momentum: 0.000000
84
+ 2023-10-16 08:31:26,275 epoch 1 - iter 801/893 - loss 0.48632560 - time (sec): 61.79 - samples/sec: 3630.22 - lr: 0.000045 - momentum: 0.000000
85
+ 2023-10-16 08:31:32,715 epoch 1 - iter 890/893 - loss 0.45469344 - time (sec): 68.23 - samples/sec: 3633.38 - lr: 0.000050 - momentum: 0.000000
86
+ 2023-10-16 08:31:32,923 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-16 08:31:32,923 EPOCH 1 done: loss 0.4535 - lr: 0.000050
88
+ 2023-10-16 08:31:35,862 DEV : loss 0.1111108735203743 - f1-score (micro avg) 0.7315
89
+ 2023-10-16 08:31:35,877 saving best model
90
+ 2023-10-16 08:31:36,283 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-16 08:31:42,955 epoch 2 - iter 89/893 - loss 0.10314043 - time (sec): 6.67 - samples/sec: 3630.28 - lr: 0.000049 - momentum: 0.000000
92
+ 2023-10-16 08:31:49,778 epoch 2 - iter 178/893 - loss 0.10488832 - time (sec): 13.49 - samples/sec: 3679.44 - lr: 0.000049 - momentum: 0.000000
93
+ 2023-10-16 08:31:56,312 epoch 2 - iter 267/893 - loss 0.10593916 - time (sec): 20.03 - samples/sec: 3676.30 - lr: 0.000048 - momentum: 0.000000
94
+ 2023-10-16 08:32:02,734 epoch 2 - iter 356/893 - loss 0.10528509 - time (sec): 26.45 - samples/sec: 3705.74 - lr: 0.000048 - momentum: 0.000000
95
+ 2023-10-16 08:32:09,647 epoch 2 - iter 445/893 - loss 0.10726148 - time (sec): 33.36 - samples/sec: 3668.93 - lr: 0.000047 - momentum: 0.000000
96
+ 2023-10-16 08:32:16,082 epoch 2 - iter 534/893 - loss 0.10633760 - time (sec): 39.80 - samples/sec: 3700.04 - lr: 0.000047 - momentum: 0.000000
97
+ 2023-10-16 08:32:22,837 epoch 2 - iter 623/893 - loss 0.10924451 - time (sec): 46.55 - samples/sec: 3708.20 - lr: 0.000046 - momentum: 0.000000
98
+ 2023-10-16 08:32:29,534 epoch 2 - iter 712/893 - loss 0.10912244 - time (sec): 53.25 - samples/sec: 3718.64 - lr: 0.000046 - momentum: 0.000000
99
+ 2023-10-16 08:32:36,384 epoch 2 - iter 801/893 - loss 0.10852736 - time (sec): 60.10 - samples/sec: 3711.26 - lr: 0.000045 - momentum: 0.000000
100
+ 2023-10-16 08:32:42,972 epoch 2 - iter 890/893 - loss 0.10756996 - time (sec): 66.69 - samples/sec: 3716.30 - lr: 0.000044 - momentum: 0.000000
101
+ 2023-10-16 08:32:43,168 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-16 08:32:43,168 EPOCH 2 done: loss 0.1078 - lr: 0.000044
103
+ 2023-10-16 08:32:47,663 DEV : loss 0.11037413775920868 - f1-score (micro avg) 0.7413
104
+ 2023-10-16 08:32:47,677 saving best model
105
+ 2023-10-16 08:32:48,276 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-16 08:32:54,910 epoch 3 - iter 89/893 - loss 0.06571927 - time (sec): 6.63 - samples/sec: 3555.83 - lr: 0.000044 - momentum: 0.000000
107
+ 2023-10-16 08:33:01,822 epoch 3 - iter 178/893 - loss 0.06442290 - time (sec): 13.54 - samples/sec: 3510.35 - lr: 0.000043 - momentum: 0.000000
108
+ 2023-10-16 08:33:08,524 epoch 3 - iter 267/893 - loss 0.06524693 - time (sec): 20.24 - samples/sec: 3522.78 - lr: 0.000043 - momentum: 0.000000
109
+ 2023-10-16 08:33:15,999 epoch 3 - iter 356/893 - loss 0.06535829 - time (sec): 27.72 - samples/sec: 3522.68 - lr: 0.000042 - momentum: 0.000000
110
+ 2023-10-16 08:33:22,881 epoch 3 - iter 445/893 - loss 0.06584378 - time (sec): 34.60 - samples/sec: 3530.86 - lr: 0.000042 - momentum: 0.000000
111
+ 2023-10-16 08:33:29,970 epoch 3 - iter 534/893 - loss 0.06636128 - time (sec): 41.69 - samples/sec: 3539.01 - lr: 0.000041 - momentum: 0.000000
112
+ 2023-10-16 08:33:36,636 epoch 3 - iter 623/893 - loss 0.06671742 - time (sec): 48.36 - samples/sec: 3557.43 - lr: 0.000041 - momentum: 0.000000
113
+ 2023-10-16 08:33:44,150 epoch 3 - iter 712/893 - loss 0.06656220 - time (sec): 55.87 - samples/sec: 3550.81 - lr: 0.000040 - momentum: 0.000000
114
+ 2023-10-16 08:33:50,912 epoch 3 - iter 801/893 - loss 0.06974675 - time (sec): 62.63 - samples/sec: 3532.98 - lr: 0.000039 - momentum: 0.000000
115
+ 2023-10-16 08:33:58,301 epoch 3 - iter 890/893 - loss 0.06917257 - time (sec): 70.02 - samples/sec: 3544.19 - lr: 0.000039 - momentum: 0.000000
116
+ 2023-10-16 08:33:58,506 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-16 08:33:58,506 EPOCH 3 done: loss 0.0691 - lr: 0.000039
118
+ 2023-10-16 08:34:02,598 DEV : loss 0.10738271474838257 - f1-score (micro avg) 0.7591
119
+ 2023-10-16 08:34:02,615 saving best model
120
+ 2023-10-16 08:34:03,145 ----------------------------------------------------------------------------------------------------
121
+ 2023-10-16 08:34:10,375 epoch 4 - iter 89/893 - loss 0.04971979 - time (sec): 7.23 - samples/sec: 3448.16 - lr: 0.000038 - momentum: 0.000000
122
+ 2023-10-16 08:34:17,473 epoch 4 - iter 178/893 - loss 0.05173471 - time (sec): 14.33 - samples/sec: 3533.67 - lr: 0.000038 - momentum: 0.000000
123
+ 2023-10-16 08:34:24,210 epoch 4 - iter 267/893 - loss 0.04932522 - time (sec): 21.06 - samples/sec: 3535.37 - lr: 0.000037 - momentum: 0.000000
124
+ 2023-10-16 08:34:31,209 epoch 4 - iter 356/893 - loss 0.04733186 - time (sec): 28.06 - samples/sec: 3595.03 - lr: 0.000037 - momentum: 0.000000
125
+ 2023-10-16 08:34:38,106 epoch 4 - iter 445/893 - loss 0.04625692 - time (sec): 34.96 - samples/sec: 3580.71 - lr: 0.000036 - momentum: 0.000000
126
+ 2023-10-16 08:34:45,075 epoch 4 - iter 534/893 - loss 0.04657375 - time (sec): 41.93 - samples/sec: 3594.96 - lr: 0.000036 - momentum: 0.000000
127
+ 2023-10-16 08:34:51,136 epoch 4 - iter 623/893 - loss 0.04778283 - time (sec): 47.99 - samples/sec: 3604.56 - lr: 0.000035 - momentum: 0.000000
128
+ 2023-10-16 08:34:58,087 epoch 4 - iter 712/893 - loss 0.04869638 - time (sec): 54.94 - samples/sec: 3605.56 - lr: 0.000034 - momentum: 0.000000
129
+ 2023-10-16 08:35:05,077 epoch 4 - iter 801/893 - loss 0.04824241 - time (sec): 61.93 - samples/sec: 3597.71 - lr: 0.000034 - momentum: 0.000000
130
+ 2023-10-16 08:35:12,187 epoch 4 - iter 890/893 - loss 0.04911274 - time (sec): 69.04 - samples/sec: 3589.56 - lr: 0.000033 - momentum: 0.000000
131
+ 2023-10-16 08:35:12,443 ----------------------------------------------------------------------------------------------------
132
+ 2023-10-16 08:35:12,443 EPOCH 4 done: loss 0.0492 - lr: 0.000033
133
+ 2023-10-16 08:35:17,390 DEV : loss 0.15796300768852234 - f1-score (micro avg) 0.7615
134
+ 2023-10-16 08:35:17,408 saving best model
135
+ 2023-10-16 08:35:17,922 ----------------------------------------------------------------------------------------------------
136
+ 2023-10-16 08:35:25,304 epoch 5 - iter 89/893 - loss 0.03757668 - time (sec): 7.38 - samples/sec: 3595.51 - lr: 0.000033 - momentum: 0.000000
137
+ 2023-10-16 08:35:32,363 epoch 5 - iter 178/893 - loss 0.03834790 - time (sec): 14.44 - samples/sec: 3577.11 - lr: 0.000032 - momentum: 0.000000
138
+ 2023-10-16 08:35:39,882 epoch 5 - iter 267/893 - loss 0.03835783 - time (sec): 21.96 - samples/sec: 3507.06 - lr: 0.000032 - momentum: 0.000000
139
+ 2023-10-16 08:35:47,072 epoch 5 - iter 356/893 - loss 0.03539896 - time (sec): 29.15 - samples/sec: 3548.29 - lr: 0.000031 - momentum: 0.000000
140
+ 2023-10-16 08:35:53,849 epoch 5 - iter 445/893 - loss 0.03572467 - time (sec): 35.93 - samples/sec: 3579.89 - lr: 0.000031 - momentum: 0.000000
141
+ 2023-10-16 08:36:00,409 epoch 5 - iter 534/893 - loss 0.03593015 - time (sec): 42.49 - samples/sec: 3594.42 - lr: 0.000030 - momentum: 0.000000
142
+ 2023-10-16 08:36:07,226 epoch 5 - iter 623/893 - loss 0.03603536 - time (sec): 49.30 - samples/sec: 3598.46 - lr: 0.000029 - momentum: 0.000000
143
+ 2023-10-16 08:36:13,538 epoch 5 - iter 712/893 - loss 0.03658109 - time (sec): 55.61 - samples/sec: 3600.77 - lr: 0.000029 - momentum: 0.000000
144
+ 2023-10-16 08:36:19,945 epoch 5 - iter 801/893 - loss 0.03767615 - time (sec): 62.02 - samples/sec: 3599.50 - lr: 0.000028 - momentum: 0.000000
145
+ 2023-10-16 08:36:26,516 epoch 5 - iter 890/893 - loss 0.03849175 - time (sec): 68.59 - samples/sec: 3607.72 - lr: 0.000028 - momentum: 0.000000
146
+ 2023-10-16 08:36:26,836 ----------------------------------------------------------------------------------------------------
147
+ 2023-10-16 08:36:26,836 EPOCH 5 done: loss 0.0386 - lr: 0.000028
148
+ 2023-10-16 08:36:31,509 DEV : loss 0.1328819841146469 - f1-score (micro avg) 0.8038
149
+ 2023-10-16 08:36:31,524 saving best model
150
+ 2023-10-16 08:36:32,006 ----------------------------------------------------------------------------------------------------
151
+ 2023-10-16 08:36:38,905 epoch 6 - iter 89/893 - loss 0.02758864 - time (sec): 6.90 - samples/sec: 3569.96 - lr: 0.000027 - momentum: 0.000000
152
+ 2023-10-16 08:36:45,504 epoch 6 - iter 178/893 - loss 0.02542332 - time (sec): 13.50 - samples/sec: 3655.39 - lr: 0.000027 - momentum: 0.000000
153
+ 2023-10-16 08:36:52,704 epoch 6 - iter 267/893 - loss 0.02811667 - time (sec): 20.70 - samples/sec: 3610.59 - lr: 0.000026 - momentum: 0.000000
154
+ 2023-10-16 08:36:59,729 epoch 6 - iter 356/893 - loss 0.03024642 - time (sec): 27.72 - samples/sec: 3648.47 - lr: 0.000026 - momentum: 0.000000
155
+ 2023-10-16 08:37:06,662 epoch 6 - iter 445/893 - loss 0.02860895 - time (sec): 34.65 - samples/sec: 3636.20 - lr: 0.000025 - momentum: 0.000000
156
+ 2023-10-16 08:37:13,334 epoch 6 - iter 534/893 - loss 0.02935119 - time (sec): 41.33 - samples/sec: 3648.62 - lr: 0.000024 - momentum: 0.000000
157
+ 2023-10-16 08:37:20,091 epoch 6 - iter 623/893 - loss 0.02876917 - time (sec): 48.08 - samples/sec: 3673.57 - lr: 0.000024 - momentum: 0.000000
158
+ 2023-10-16 08:37:27,048 epoch 6 - iter 712/893 - loss 0.02875323 - time (sec): 55.04 - samples/sec: 3665.73 - lr: 0.000023 - momentum: 0.000000
159
+ 2023-10-16 08:37:33,705 epoch 6 - iter 801/893 - loss 0.02975422 - time (sec): 61.70 - samples/sec: 3631.98 - lr: 0.000023 - momentum: 0.000000
160
+ 2023-10-16 08:37:40,582 epoch 6 - iter 890/893 - loss 0.02924042 - time (sec): 68.57 - samples/sec: 3618.99 - lr: 0.000022 - momentum: 0.000000
161
+ 2023-10-16 08:37:40,783 ----------------------------------------------------------------------------------------------------
162
+ 2023-10-16 08:37:40,784 EPOCH 6 done: loss 0.0294 - lr: 0.000022
163
+ 2023-10-16 08:37:44,862 DEV : loss 0.16811993718147278 - f1-score (micro avg) 0.803
164
+ 2023-10-16 08:37:44,878 ----------------------------------------------------------------------------------------------------
165
+ 2023-10-16 08:37:51,877 epoch 7 - iter 89/893 - loss 0.01923575 - time (sec): 7.00 - samples/sec: 3720.25 - lr: 0.000022 - momentum: 0.000000
166
+ 2023-10-16 08:37:59,090 epoch 7 - iter 178/893 - loss 0.01984026 - time (sec): 14.21 - samples/sec: 3574.00 - lr: 0.000021 - momentum: 0.000000
167
+ 2023-10-16 08:38:06,084 epoch 7 - iter 267/893 - loss 0.02090345 - time (sec): 21.20 - samples/sec: 3617.39 - lr: 0.000021 - momentum: 0.000000
168
+ 2023-10-16 08:38:12,677 epoch 7 - iter 356/893 - loss 0.02107347 - time (sec): 27.80 - samples/sec: 3634.44 - lr: 0.000020 - momentum: 0.000000
169
+ 2023-10-16 08:38:19,274 epoch 7 - iter 445/893 - loss 0.01912080 - time (sec): 34.39 - samples/sec: 3595.85 - lr: 0.000019 - momentum: 0.000000
170
+ 2023-10-16 08:38:25,861 epoch 7 - iter 534/893 - loss 0.02166335 - time (sec): 40.98 - samples/sec: 3631.27 - lr: 0.000019 - momentum: 0.000000
171
+ 2023-10-16 08:38:32,527 epoch 7 - iter 623/893 - loss 0.02231858 - time (sec): 47.65 - samples/sec: 3642.54 - lr: 0.000018 - momentum: 0.000000
172
+ 2023-10-16 08:38:39,347 epoch 7 - iter 712/893 - loss 0.02297249 - time (sec): 54.47 - samples/sec: 3639.54 - lr: 0.000018 - momentum: 0.000000
173
+ 2023-10-16 08:38:46,464 epoch 7 - iter 801/893 - loss 0.02302119 - time (sec): 61.58 - samples/sec: 3629.65 - lr: 0.000017 - momentum: 0.000000
174
+ 2023-10-16 08:38:53,241 epoch 7 - iter 890/893 - loss 0.02234366 - time (sec): 68.36 - samples/sec: 3629.66 - lr: 0.000017 - momentum: 0.000000
175
+ 2023-10-16 08:38:53,453 ----------------------------------------------------------------------------------------------------
176
+ 2023-10-16 08:38:53,453 EPOCH 7 done: loss 0.0223 - lr: 0.000017
177
+ 2023-10-16 08:38:58,034 DEV : loss 0.18849003314971924 - f1-score (micro avg) 0.7939
178
+ 2023-10-16 08:38:58,050 ----------------------------------------------------------------------------------------------------
179
+ 2023-10-16 08:39:04,891 epoch 8 - iter 89/893 - loss 0.01614628 - time (sec): 6.84 - samples/sec: 3609.86 - lr: 0.000016 - momentum: 0.000000
180
+ 2023-10-16 08:39:12,208 epoch 8 - iter 178/893 - loss 0.01529648 - time (sec): 14.16 - samples/sec: 3638.21 - lr: 0.000016 - momentum: 0.000000
181
+ 2023-10-16 08:39:18,795 epoch 8 - iter 267/893 - loss 0.01647957 - time (sec): 20.74 - samples/sec: 3621.92 - lr: 0.000015 - momentum: 0.000000
182
+ 2023-10-16 08:39:25,204 epoch 8 - iter 356/893 - loss 0.01719578 - time (sec): 27.15 - samples/sec: 3630.10 - lr: 0.000014 - momentum: 0.000000
183
+ 2023-10-16 08:39:31,952 epoch 8 - iter 445/893 - loss 0.01781583 - time (sec): 33.90 - samples/sec: 3576.50 - lr: 0.000014 - momentum: 0.000000
184
+ 2023-10-16 08:39:39,376 epoch 8 - iter 534/893 - loss 0.01686429 - time (sec): 41.32 - samples/sec: 3573.14 - lr: 0.000013 - momentum: 0.000000
185
+ 2023-10-16 08:39:45,998 epoch 8 - iter 623/893 - loss 0.01704050 - time (sec): 47.95 - samples/sec: 3598.61 - lr: 0.000013 - momentum: 0.000000
186
+ 2023-10-16 08:39:53,311 epoch 8 - iter 712/893 - loss 0.01722286 - time (sec): 55.26 - samples/sec: 3607.80 - lr: 0.000012 - momentum: 0.000000
187
+ 2023-10-16 08:40:00,416 epoch 8 - iter 801/893 - loss 0.01694712 - time (sec): 62.36 - samples/sec: 3603.89 - lr: 0.000012 - momentum: 0.000000
188
+ 2023-10-16 08:40:07,089 epoch 8 - iter 890/893 - loss 0.01671342 - time (sec): 69.04 - samples/sec: 3595.86 - lr: 0.000011 - momentum: 0.000000
189
+ 2023-10-16 08:40:07,249 ----------------------------------------------------------------------------------------------------
190
+ 2023-10-16 08:40:07,249 EPOCH 8 done: loss 0.0168 - lr: 0.000011
191
+ 2023-10-16 08:40:11,861 DEV : loss 0.17899882793426514 - f1-score (micro avg) 0.8
192
+ 2023-10-16 08:40:11,877 ----------------------------------------------------------------------------------------------------
193
+ 2023-10-16 08:40:19,154 epoch 9 - iter 89/893 - loss 0.02097746 - time (sec): 7.28 - samples/sec: 3410.06 - lr: 0.000011 - momentum: 0.000000
194
+ 2023-10-16 08:40:25,833 epoch 9 - iter 178/893 - loss 0.01641454 - time (sec): 13.95 - samples/sec: 3554.96 - lr: 0.000010 - momentum: 0.000000
195
+ 2023-10-16 08:40:32,789 epoch 9 - iter 267/893 - loss 0.01440771 - time (sec): 20.91 - samples/sec: 3613.57 - lr: 0.000009 - momentum: 0.000000
196
+ 2023-10-16 08:40:39,245 epoch 9 - iter 356/893 - loss 0.01250180 - time (sec): 27.37 - samples/sec: 3645.03 - lr: 0.000009 - momentum: 0.000000
197
+ 2023-10-16 08:40:46,116 epoch 9 - iter 445/893 - loss 0.01202762 - time (sec): 34.24 - samples/sec: 3625.94 - lr: 0.000008 - momentum: 0.000000
198
+ 2023-10-16 08:40:53,435 epoch 9 - iter 534/893 - loss 0.01191690 - time (sec): 41.56 - samples/sec: 3609.26 - lr: 0.000008 - momentum: 0.000000
199
+ 2023-10-16 08:41:00,371 epoch 9 - iter 623/893 - loss 0.01186562 - time (sec): 48.49 - samples/sec: 3622.78 - lr: 0.000007 - momentum: 0.000000
200
+ 2023-10-16 08:41:07,388 epoch 9 - iter 712/893 - loss 0.01142829 - time (sec): 55.51 - samples/sec: 3617.83 - lr: 0.000007 - momentum: 0.000000
201
+ 2023-10-16 08:41:13,708 epoch 9 - iter 801/893 - loss 0.01159964 - time (sec): 61.83 - samples/sec: 3623.46 - lr: 0.000006 - momentum: 0.000000
202
+ 2023-10-16 08:41:20,376 epoch 9 - iter 890/893 - loss 0.01208136 - time (sec): 68.50 - samples/sec: 3620.63 - lr: 0.000006 - momentum: 0.000000
203
+ 2023-10-16 08:41:20,635 ----------------------------------------------------------------------------------------------------
204
+ 2023-10-16 08:41:20,635 EPOCH 9 done: loss 0.0121 - lr: 0.000006
205
+ 2023-10-16 08:41:24,716 DEV : loss 0.1874045431613922 - f1-score (micro avg) 0.8032
206
+ 2023-10-16 08:41:24,732 ----------------------------------------------------------------------------------------------------
207
+ 2023-10-16 08:41:31,855 epoch 10 - iter 89/893 - loss 0.00193443 - time (sec): 7.12 - samples/sec: 3639.40 - lr: 0.000005 - momentum: 0.000000
208
+ 2023-10-16 08:41:39,260 epoch 10 - iter 178/893 - loss 0.00618707 - time (sec): 14.53 - samples/sec: 3615.48 - lr: 0.000004 - momentum: 0.000000
209
+ 2023-10-16 08:41:46,242 epoch 10 - iter 267/893 - loss 0.00721601 - time (sec): 21.51 - samples/sec: 3602.66 - lr: 0.000004 - momentum: 0.000000
210
+ 2023-10-16 08:41:52,700 epoch 10 - iter 356/893 - loss 0.00741060 - time (sec): 27.97 - samples/sec: 3659.77 - lr: 0.000003 - momentum: 0.000000
211
+ 2023-10-16 08:41:59,344 epoch 10 - iter 445/893 - loss 0.00736920 - time (sec): 34.61 - samples/sec: 3657.22 - lr: 0.000003 - momentum: 0.000000
212
+ 2023-10-16 08:42:05,827 epoch 10 - iter 534/893 - loss 0.00689678 - time (sec): 41.09 - samples/sec: 3651.95 - lr: 0.000002 - momentum: 0.000000
213
+ 2023-10-16 08:42:12,520 epoch 10 - iter 623/893 - loss 0.00774315 - time (sec): 47.79 - samples/sec: 3628.93 - lr: 0.000002 - momentum: 0.000000
214
+ 2023-10-16 08:42:19,505 epoch 10 - iter 712/893 - loss 0.00762537 - time (sec): 54.77 - samples/sec: 3623.92 - lr: 0.000001 - momentum: 0.000000
215
+ 2023-10-16 08:42:26,433 epoch 10 - iter 801/893 - loss 0.00794685 - time (sec): 61.70 - samples/sec: 3623.69 - lr: 0.000001 - momentum: 0.000000
216
+ 2023-10-16 08:42:33,079 epoch 10 - iter 890/893 - loss 0.00809313 - time (sec): 68.35 - samples/sec: 3628.42 - lr: 0.000000 - momentum: 0.000000
217
+ 2023-10-16 08:42:33,254 ----------------------------------------------------------------------------------------------------
218
+ 2023-10-16 08:42:33,254 EPOCH 10 done: loss 0.0081 - lr: 0.000000
219
+ 2023-10-16 08:42:37,798 DEV : loss 0.20228144526481628 - f1-score (micro avg) 0.8008
220
+ 2023-10-16 08:42:38,197 ----------------------------------------------------------------------------------------------------
221
+ 2023-10-16 08:42:38,198 Loading model from best epoch ...
222
+ 2023-10-16 08:42:39,883 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
223
+ 2023-10-16 08:42:49,235
224
+ Results:
225
+ - F-score (micro) 0.6844
226
+ - F-score (macro) 0.6123
227
+ - Accuracy 0.5361
228
+
229
+ By class:
230
+ precision recall f1-score support
231
+
232
+ LOC 0.6996 0.6785 0.6889 1095
233
+ PER 0.7892 0.7105 0.7478 1012
234
+ ORG 0.5038 0.5518 0.5267 357
235
+ HumanProd 0.4595 0.5152 0.4857 33
236
+
237
+ micro avg 0.6980 0.6712 0.6844 2497
238
+ macro avg 0.6130 0.6140 0.6123 2497
239
+ weighted avg 0.7048 0.6712 0.6869 2497
240
+
241
+ 2023-10-16 08:42:49,235 ----------------------------------------------------------------------------------------------------