Hugging Face's logo

stefan-it
/
hmbench-icdar-fr-hmbert-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2

Token Classification
Flair
PyTorch
French
sequence-tagger-model
Model card Files Community
1
stefan-it commited on
Commit
415103f
·
1 Parent(s): 39aad70

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 +242 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:91edca03741f5c09ac936e8b3af9426a58e4e92faf6a6d7acef8186b984e0396
3
+ size 443311111
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 22:33:38 0.0000 0.3735 0.1084 0.7053 0.6606 0.6822 0.5448
3
+ 2 22:34:43 0.0000 0.1015 0.0859 0.6950 0.7964 0.7422 0.6059
4
+ 3 22:35:46 0.0000 0.0737 0.1023 0.7576 0.7602 0.7589 0.6298
5
+ 4 22:36:49 0.0000 0.0510 0.1315 0.7689 0.7489 0.7587 0.6281
6
+ 5 22:37:52 0.0000 0.0421 0.1539 0.7345 0.7794 0.7563 0.6286
7
+ 6 22:38:55 0.0000 0.0296 0.1960 0.7508 0.7636 0.7572 0.6314
8
+ 7 22:39:57 0.0000 0.0217 0.1889 0.7322 0.8009 0.7650 0.6407
9
+ 8 22:41:00 0.0000 0.0159 0.1949 0.7455 0.7885 0.7664 0.6400
10
+ 9 22:42:04 0.0000 0.0110 0.2040 0.7508 0.7839 0.7670 0.6435
11
+ 10 22:43:08 0.0000 0.0075 0.2077 0.7552 0.7851 0.7698 0.6456
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,242 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-13 22:32:35,467 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-13 22:32:35,468 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=13, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-13 22:32:35,468 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-13 22:32:35,468 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
+ 2023-10-13 22:32:35,469 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-13 22:32:35,469 Train: 7936 sentences
55
+ 2023-10-13 22:32:35,469 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-13 22:32:35,469 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-13 22:32:35,469 Training Params:
58
+ 2023-10-13 22:32:35,469 - learning_rate: "5e-05"
59
+ 2023-10-13 22:32:35,469 - mini_batch_size: "8"
60
+ 2023-10-13 22:32:35,469 - max_epochs: "10"
61
+ 2023-10-13 22:32:35,469 - shuffle: "True"
62
+ 2023-10-13 22:32:35,469 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-13 22:32:35,469 Plugins:
64
+ 2023-10-13 22:32:35,469 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-13 22:32:35,469 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-13 22:32:35,469 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-13 22:32:35,469 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-13 22:32:35,469 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-13 22:32:35,469 Computation:
70
+ 2023-10-13 22:32:35,469 - compute on device: cuda:0
71
+ 2023-10-13 22:32:35,469 - embedding storage: none
72
+ 2023-10-13 22:32:35,469 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-13 22:32:35,469 Model training base path: "hmbench-icdar/fr-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2"
74
+ 2023-10-13 22:32:35,469 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-13 22:32:35,469 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-13 22:32:42,105 epoch 1 - iter 99/992 - loss 1.97625335 - time (sec): 6.63 - samples/sec: 2619.78 - lr: 0.000005 - momentum: 0.000000
77
+ 2023-10-13 22:32:48,119 epoch 1 - iter 198/992 - loss 1.20445619 - time (sec): 12.65 - samples/sec: 2622.63 - lr: 0.000010 - momentum: 0.000000
78
+ 2023-10-13 22:32:53,990 epoch 1 - iter 297/992 - loss 0.90127542 - time (sec): 18.52 - samples/sec: 2645.29 - lr: 0.000015 - momentum: 0.000000
79
+ 2023-10-13 22:32:59,842 epoch 1 - iter 396/992 - loss 0.73031528 - time (sec): 24.37 - samples/sec: 2664.09 - lr: 0.000020 - momentum: 0.000000
80
+ 2023-10-13 22:33:05,553 epoch 1 - iter 495/992 - loss 0.61732090 - time (sec): 30.08 - samples/sec: 2695.40 - lr: 0.000025 - momentum: 0.000000
81
+ 2023-10-13 22:33:11,400 epoch 1 - iter 594/992 - loss 0.53727279 - time (sec): 35.93 - samples/sec: 2719.61 - lr: 0.000030 - momentum: 0.000000
82
+ 2023-10-13 22:33:17,199 epoch 1 - iter 693/992 - loss 0.48213515 - time (sec): 41.73 - samples/sec: 2724.53 - lr: 0.000035 - momentum: 0.000000
83
+ 2023-10-13 22:33:23,140 epoch 1 - iter 792/992 - loss 0.43768861 - time (sec): 47.67 - samples/sec: 2731.72 - lr: 0.000040 - momentum: 0.000000
84
+ 2023-10-13 22:33:29,370 epoch 1 - iter 891/992 - loss 0.40269124 - time (sec): 53.90 - samples/sec: 2727.47 - lr: 0.000045 - momentum: 0.000000
85
+ 2023-10-13 22:33:35,471 epoch 1 - iter 990/992 - loss 0.37457303 - time (sec): 60.00 - samples/sec: 2724.88 - lr: 0.000050 - momentum: 0.000000
86
+ 2023-10-13 22:33:35,623 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-13 22:33:35,624 EPOCH 1 done: loss 0.3735 - lr: 0.000050
88
+ 2023-10-13 22:33:38,899 DEV : loss 0.10837739706039429 - f1-score (micro avg) 0.6822
89
+ 2023-10-13 22:33:38,924 saving best model
90
+ 2023-10-13 22:33:39,350 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-13 22:33:45,003 epoch 2 - iter 99/992 - loss 0.10295012 - time (sec): 5.65 - samples/sec: 2750.21 - lr: 0.000049 - momentum: 0.000000
92
+ 2023-10-13 22:33:50,901 epoch 2 - iter 198/992 - loss 0.10973798 - time (sec): 11.55 - samples/sec: 2718.58 - lr: 0.000049 - momentum: 0.000000
93
+ 2023-10-13 22:33:56,904 epoch 2 - iter 297/992 - loss 0.10543489 - time (sec): 17.55 - samples/sec: 2761.93 - lr: 0.000048 - momentum: 0.000000
94
+ 2023-10-13 22:34:02,862 epoch 2 - iter 396/992 - loss 0.10381763 - time (sec): 23.51 - samples/sec: 2769.50 - lr: 0.000048 - momentum: 0.000000
95
+ 2023-10-13 22:34:08,919 epoch 2 - iter 495/992 - loss 0.10311459 - time (sec): 29.57 - samples/sec: 2751.80 - lr: 0.000047 - momentum: 0.000000
96
+ 2023-10-13 22:34:14,729 epoch 2 - iter 594/992 - loss 0.10171878 - time (sec): 35.38 - samples/sec: 2751.64 - lr: 0.000047 - momentum: 0.000000
97
+ 2023-10-13 22:34:20,492 epoch 2 - iter 693/992 - loss 0.09979812 - time (sec): 41.14 - samples/sec: 2754.11 - lr: 0.000046 - momentum: 0.000000
98
+ 2023-10-13 22:34:26,580 epoch 2 - iter 792/992 - loss 0.10186744 - time (sec): 47.23 - samples/sec: 2753.33 - lr: 0.000046 - momentum: 0.000000
99
+ 2023-10-13 22:34:32,670 epoch 2 - iter 891/992 - loss 0.10186196 - time (sec): 53.32 - samples/sec: 2741.16 - lr: 0.000045 - momentum: 0.000000
100
+ 2023-10-13 22:34:38,791 epoch 2 - iter 990/992 - loss 0.10164595 - time (sec): 59.44 - samples/sec: 2752.43 - lr: 0.000044 - momentum: 0.000000
101
+ 2023-10-13 22:34:38,920 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-13 22:34:38,920 EPOCH 2 done: loss 0.1015 - lr: 0.000044
103
+ 2023-10-13 22:34:43,263 DEV : loss 0.08593912422657013 - f1-score (micro avg) 0.7422
104
+ 2023-10-13 22:34:43,287 saving best model
105
+ 2023-10-13 22:34:43,770 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-13 22:34:49,664 epoch 3 - iter 99/992 - loss 0.08176039 - time (sec): 5.89 - samples/sec: 2717.35 - lr: 0.000044 - momentum: 0.000000
107
+ 2023-10-13 22:34:55,620 epoch 3 - iter 198/992 - loss 0.07620139 - time (sec): 11.85 - samples/sec: 2720.74 - lr: 0.000043 - momentum: 0.000000
108
+ 2023-10-13 22:35:01,359 epoch 3 - iter 297/992 - loss 0.07242783 - time (sec): 17.59 - samples/sec: 2743.33 - lr: 0.000043 - momentum: 0.000000
109
+ 2023-10-13 22:35:07,179 epoch 3 - iter 396/992 - loss 0.07362963 - time (sec): 23.41 - samples/sec: 2759.74 - lr: 0.000042 - momentum: 0.000000
110
+ 2023-10-13 22:35:13,060 epoch 3 - iter 495/992 - loss 0.07299823 - time (sec): 29.29 - samples/sec: 2761.91 - lr: 0.000042 - momentum: 0.000000
111
+ 2023-10-13 22:35:19,048 epoch 3 - iter 594/992 - loss 0.07330654 - time (sec): 35.27 - samples/sec: 2761.44 - lr: 0.000041 - momentum: 0.000000
112
+ 2023-10-13 22:35:25,090 epoch 3 - iter 693/992 - loss 0.07489880 - time (sec): 41.32 - samples/sec: 2747.94 - lr: 0.000041 - momentum: 0.000000
113
+ 2023-10-13 22:35:31,036 epoch 3 - iter 792/992 - loss 0.07562313 - time (sec): 47.26 - samples/sec: 2760.10 - lr: 0.000040 - momentum: 0.000000
114
+ 2023-10-13 22:35:37,046 epoch 3 - iter 891/992 - loss 0.07372307 - time (sec): 53.27 - samples/sec: 2766.49 - lr: 0.000039 - momentum: 0.000000
115
+ 2023-10-13 22:35:42,895 epoch 3 - iter 990/992 - loss 0.07373060 - time (sec): 59.12 - samples/sec: 2769.61 - lr: 0.000039 - momentum: 0.000000
116
+ 2023-10-13 22:35:42,996 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-13 22:35:42,996 EPOCH 3 done: loss 0.0737 - lr: 0.000039
118
+ 2023-10-13 22:35:46,388 DEV : loss 0.10231217741966248 - f1-score (micro avg) 0.7589
119
+ 2023-10-13 22:35:46,412 saving best model
120
+ 2023-10-13 22:35:46,925 ----------------------------------------------------------------------------------------------------
121
+ 2023-10-13 22:35:52,809 epoch 4 - iter 99/992 - loss 0.05134981 - time (sec): 5.88 - samples/sec: 2723.80 - lr: 0.000038 - momentum: 0.000000
122
+ 2023-10-13 22:35:59,141 epoch 4 - iter 198/992 - loss 0.05017639 - time (sec): 12.21 - samples/sec: 2716.64 - lr: 0.000038 - momentum: 0.000000
123
+ 2023-10-13 22:36:05,090 epoch 4 - iter 297/992 - loss 0.04727321 - time (sec): 18.16 - samples/sec: 2718.31 - lr: 0.000037 - momentum: 0.000000
124
+ 2023-10-13 22:36:10,869 epoch 4 - iter 396/992 - loss 0.04808307 - time (sec): 23.94 - samples/sec: 2749.13 - lr: 0.000037 - momentum: 0.000000
125
+ 2023-10-13 22:36:16,706 epoch 4 - iter 495/992 - loss 0.04713806 - time (sec): 29.78 - samples/sec: 2768.47 - lr: 0.000036 - momentum: 0.000000
126
+ 2023-10-13 22:36:22,555 epoch 4 - iter 594/992 - loss 0.04822839 - time (sec): 35.63 - samples/sec: 2769.58 - lr: 0.000036 - momentum: 0.000000
127
+ 2023-10-13 22:36:28,395 epoch 4 - iter 693/992 - loss 0.04982517 - time (sec): 41.47 - samples/sec: 2764.17 - lr: 0.000035 - momentum: 0.000000
128
+ 2023-10-13 22:36:34,340 epoch 4 - iter 792/992 - loss 0.05129636 - time (sec): 47.41 - samples/sec: 2765.07 - lr: 0.000034 - momentum: 0.000000
129
+ 2023-10-13 22:36:39,871 epoch 4 - iter 891/992 - loss 0.05132015 - time (sec): 52.94 - samples/sec: 2782.54 - lr: 0.000034 - momentum: 0.000000
130
+ 2023-10-13 22:36:46,060 epoch 4 - iter 990/992 - loss 0.05099074 - time (sec): 59.13 - samples/sec: 2765.99 - lr: 0.000033 - momentum: 0.000000
131
+ 2023-10-13 22:36:46,190 ----------------------------------------------------------------------------------------------------
132
+ 2023-10-13 22:36:46,190 EPOCH 4 done: loss 0.0510 - lr: 0.000033
133
+ 2023-10-13 22:36:49,798 DEV : loss 0.13146057724952698 - f1-score (micro avg) 0.7587
134
+ 2023-10-13 22:36:49,822 ----------------------------------------------------------------------------------------------------
135
+ 2023-10-13 22:36:55,595 epoch 5 - iter 99/992 - loss 0.03914982 - time (sec): 5.77 - samples/sec: 2761.13 - lr: 0.000033 - momentum: 0.000000
136
+ 2023-10-13 22:37:01,451 epoch 5 - iter 198/992 - loss 0.03993438 - time (sec): 11.63 - samples/sec: 2786.06 - lr: 0.000032 - momentum: 0.000000
137
+ 2023-10-13 22:37:07,215 epoch 5 - iter 297/992 - loss 0.03848388 - time (sec): 17.39 - samples/sec: 2811.92 - lr: 0.000032 - momentum: 0.000000
138
+ 2023-10-13 22:37:13,113 epoch 5 - iter 396/992 - loss 0.03928207 - time (sec): 23.29 - samples/sec: 2821.97 - lr: 0.000031 - momentum: 0.000000
139
+ 2023-10-13 22:37:19,287 epoch 5 - iter 495/992 - loss 0.04014864 - time (sec): 29.46 - samples/sec: 2803.85 - lr: 0.000031 - momentum: 0.000000
140
+ 2023-10-13 22:37:25,177 epoch 5 - iter 594/992 - loss 0.04251426 - time (sec): 35.35 - samples/sec: 2787.71 - lr: 0.000030 - momentum: 0.000000
141
+ 2023-10-13 22:37:30,988 epoch 5 - iter 693/992 - loss 0.04162004 - time (sec): 41.17 - samples/sec: 2781.73 - lr: 0.000029 - momentum: 0.000000
142
+ 2023-10-13 22:37:36,743 epoch 5 - iter 792/992 - loss 0.04073969 - time (sec): 46.92 - samples/sec: 2780.60 - lr: 0.000029 - momentum: 0.000000
143
+ 2023-10-13 22:37:43,244 epoch 5 - iter 891/992 - loss 0.04252912 - time (sec): 53.42 - samples/sec: 2758.70 - lr: 0.000028 - momentum: 0.000000
144
+ 2023-10-13 22:37:48,877 epoch 5 - iter 990/992 - loss 0.04211590 - time (sec): 59.05 - samples/sec: 2768.82 - lr: 0.000028 - momentum: 0.000000
145
+ 2023-10-13 22:37:49,038 ----------------------------------------------------------------------------------------------------
146
+ 2023-10-13 22:37:49,038 EPOCH 5 done: loss 0.0421 - lr: 0.000028
147
+ 2023-10-13 22:37:52,595 DEV : loss 0.15390822291374207 - f1-score (micro avg) 0.7563
148
+ 2023-10-13 22:37:52,618 ----------------------------------------------------------------------------------------------------
149
+ 2023-10-13 22:37:58,397 epoch 6 - iter 99/992 - loss 0.02447859 - time (sec): 5.78 - samples/sec: 2857.09 - lr: 0.000027 - momentum: 0.000000
150
+ 2023-10-13 22:38:04,483 epoch 6 - iter 198/992 - loss 0.02281979 - time (sec): 11.86 - samples/sec: 2814.99 - lr: 0.000027 - momentum: 0.000000
151
+ 2023-10-13 22:38:10,375 epoch 6 - iter 297/992 - loss 0.02449871 - time (sec): 17.76 - samples/sec: 2836.72 - lr: 0.000026 - momentum: 0.000000
152
+ 2023-10-13 22:38:16,330 epoch 6 - iter 396/992 - loss 0.02562274 - time (sec): 23.71 - samples/sec: 2816.53 - lr: 0.000026 - momentum: 0.000000
153
+ 2023-10-13 22:38:22,242 epoch 6 - iter 495/992 - loss 0.02613520 - time (sec): 29.62 - samples/sec: 2780.83 - lr: 0.000025 - momentum: 0.000000
154
+ 2023-10-13 22:38:28,192 epoch 6 - iter 594/992 - loss 0.02619509 - time (sec): 35.57 - samples/sec: 2768.90 - lr: 0.000024 - momentum: 0.000000
155
+ 2023-10-13 22:38:33,861 epoch 6 - iter 693/992 - loss 0.02665766 - time (sec): 41.24 - samples/sec: 2754.02 - lr: 0.000024 - momentum: 0.000000
156
+ 2023-10-13 22:38:39,952 epoch 6 - iter 792/992 - loss 0.02787193 - time (sec): 47.33 - samples/sec: 2770.66 - lr: 0.000023 - momentum: 0.000000
157
+ 2023-10-13 22:38:45,783 epoch 6 - iter 891/992 - loss 0.02982103 - time (sec): 53.16 - samples/sec: 2766.54 - lr: 0.000023 - momentum: 0.000000
158
+ 2023-10-13 22:38:51,599 epoch 6 - iter 990/992 - loss 0.02967950 - time (sec): 58.98 - samples/sec: 2772.64 - lr: 0.000022 - momentum: 0.000000
159
+ 2023-10-13 22:38:51,724 ----------------------------------------------------------------------------------------------------
160
+ 2023-10-13 22:38:51,724 EPOCH 6 done: loss 0.0296 - lr: 0.000022
161
+ 2023-10-13 22:38:55,149 DEV : loss 0.19599080085754395 - f1-score (micro avg) 0.7572
162
+ 2023-10-13 22:38:55,170 ----------------------------------------------------------------------------------------------------
163
+ 2023-10-13 22:39:00,912 epoch 7 - iter 99/992 - loss 0.02376152 - time (sec): 5.74 - samples/sec: 2826.61 - lr: 0.000022 - momentum: 0.000000
164
+ 2023-10-13 22:39:06,655 epoch 7 - iter 198/992 - loss 0.01958953 - time (sec): 11.48 - samples/sec: 2810.22 - lr: 0.000021 - momentum: 0.000000
165
+ 2023-10-13 22:39:12,687 epoch 7 - iter 297/992 - loss 0.01962767 - time (sec): 17.52 - samples/sec: 2831.96 - lr: 0.000021 - momentum: 0.000000
166
+ 2023-10-13 22:39:18,688 epoch 7 - iter 396/992 - loss 0.01948507 - time (sec): 23.52 - samples/sec: 2822.98 - lr: 0.000020 - momentum: 0.000000
167
+ 2023-10-13 22:39:24,654 epoch 7 - iter 495/992 - loss 0.01985791 - time (sec): 29.48 - samples/sec: 2810.77 - lr: 0.000019 - momentum: 0.000000
168
+ 2023-10-13 22:39:30,455 epoch 7 - iter 594/992 - loss 0.02068738 - time (sec): 35.28 - samples/sec: 2793.89 - lr: 0.000019 - momentum: 0.000000
169
+ 2023-10-13 22:39:36,232 epoch 7 - iter 693/992 - loss 0.02241871 - time (sec): 41.06 - samples/sec: 2796.58 - lr: 0.000018 - momentum: 0.000000
170
+ 2023-10-13 22:39:41,978 epoch 7 - iter 792/992 - loss 0.02181208 - time (sec): 46.81 - samples/sec: 2804.38 - lr: 0.000018 - momentum: 0.000000
171
+ 2023-10-13 22:39:47,996 epoch 7 - iter 891/992 - loss 0.02217841 - time (sec): 52.83 - samples/sec: 2783.90 - lr: 0.000017 - momentum: 0.000000
172
+ 2023-10-13 22:39:54,016 epoch 7 - iter 990/992 - loss 0.02172978 - time (sec): 58.84 - samples/sec: 2783.19 - lr: 0.000017 - momentum: 0.000000
173
+ 2023-10-13 22:39:54,115 ----------------------------------------------------------------------------------------------------
174
+ 2023-10-13 22:39:54,115 EPOCH 7 done: loss 0.0217 - lr: 0.000017
175
+ 2023-10-13 22:39:57,955 DEV : loss 0.18885478377342224 - f1-score (micro avg) 0.765
176
+ 2023-10-13 22:39:57,976 saving best model
177
+ 2023-10-13 22:39:58,428 ----------------------------------------------------------------------------------------------------
178
+ 2023-10-13 22:40:04,318 epoch 8 - iter 99/992 - loss 0.02286923 - time (sec): 5.88 - samples/sec: 2805.77 - lr: 0.000016 - momentum: 0.000000
179
+ 2023-10-13 22:40:10,386 epoch 8 - iter 198/992 - loss 0.01729121 - time (sec): 11.95 - samples/sec: 2800.86 - lr: 0.000016 - momentum: 0.000000
180
+ 2023-10-13 22:40:16,196 epoch 8 - iter 297/992 - loss 0.01585346 - time (sec): 17.76 - samples/sec: 2818.05 - lr: 0.000015 - momentum: 0.000000
181
+ 2023-10-13 22:40:22,131 epoch 8 - iter 396/992 - loss 0.01479878 - time (sec): 23.70 - samples/sec: 2838.92 - lr: 0.000014 - momentum: 0.000000
182
+ 2023-10-13 22:40:27,927 epoch 8 - iter 495/992 - loss 0.01559774 - time (sec): 29.49 - samples/sec: 2832.36 - lr: 0.000014 - momentum: 0.000000
183
+ 2023-10-13 22:40:33,638 epoch 8 - iter 594/992 - loss 0.01479709 - time (sec): 35.20 - samples/sec: 2821.11 - lr: 0.000013 - momentum: 0.000000
184
+ 2023-10-13 22:40:39,365 epoch 8 - iter 693/992 - loss 0.01447454 - time (sec): 40.93 - samples/sec: 2808.13 - lr: 0.000013 - momentum: 0.000000
185
+ 2023-10-13 22:40:45,325 epoch 8 - iter 792/992 - loss 0.01516714 - time (sec): 46.89 - samples/sec: 2803.50 - lr: 0.000012 - momentum: 0.000000
186
+ 2023-10-13 22:40:51,152 epoch 8 - iter 891/992 - loss 0.01551720 - time (sec): 52.72 - samples/sec: 2801.05 - lr: 0.000012 - momentum: 0.000000
187
+ 2023-10-13 22:40:56,972 epoch 8 - iter 990/992 - loss 0.01557435 - time (sec): 58.54 - samples/sec: 2798.83 - lr: 0.000011 - momentum: 0.000000
188
+ 2023-10-13 22:40:57,071 ----------------------------------------------------------------------------------------------------
189
+ 2023-10-13 22:40:57,072 EPOCH 8 done: loss 0.0159 - lr: 0.000011
190
+ 2023-10-13 22:41:00,599 DEV : loss 0.19485069811344147 - f1-score (micro avg) 0.7664
191
+ 2023-10-13 22:41:00,622 saving best model
192
+ 2023-10-13 22:41:01,122 ----------------------------------------------------------------------------------------------------
193
+ 2023-10-13 22:41:06,817 epoch 9 - iter 99/992 - loss 0.00725184 - time (sec): 5.69 - samples/sec: 2598.12 - lr: 0.000011 - momentum: 0.000000
194
+ 2023-10-13 22:41:12,615 epoch 9 - iter 198/992 - loss 0.01324789 - time (sec): 11.49 - samples/sec: 2671.94 - lr: 0.000010 - momentum: 0.000000
195
+ 2023-10-13 22:41:18,618 epoch 9 - iter 297/992 - loss 0.01262387 - time (sec): 17.49 - samples/sec: 2676.85 - lr: 0.000009 - momentum: 0.000000
196
+ 2023-10-13 22:41:24,527 epoch 9 - iter 396/992 - loss 0.01078576 - time (sec): 23.40 - samples/sec: 2700.15 - lr: 0.000009 - momentum: 0.000000
197
+ 2023-10-13 22:41:30,687 epoch 9 - iter 495/992 - loss 0.01086785 - time (sec): 29.56 - samples/sec: 2708.13 - lr: 0.000008 - momentum: 0.000000
198
+ 2023-10-13 22:41:36,631 epoch 9 - iter 594/992 - loss 0.01119434 - time (sec): 35.50 - samples/sec: 2722.13 - lr: 0.000008 - momentum: 0.000000
199
+ 2023-10-13 22:41:42,383 epoch 9 - iter 693/992 - loss 0.01095970 - time (sec): 41.26 - samples/sec: 2736.96 - lr: 0.000007 - momentum: 0.000000
200
+ 2023-10-13 22:41:48,299 epoch 9 - iter 792/992 - loss 0.01070482 - time (sec): 47.17 - samples/sec: 2753.00 - lr: 0.000007 - momentum: 0.000000
201
+ 2023-10-13 22:41:54,240 epoch 9 - iter 891/992 - loss 0.01101838 - time (sec): 53.11 - samples/sec: 2759.58 - lr: 0.000006 - momentum: 0.000000
202
+ 2023-10-13 22:42:00,495 epoch 9 - iter 990/992 - loss 0.01099219 - time (sec): 59.37 - samples/sec: 2755.51 - lr: 0.000006 - momentum: 0.000000
203
+ 2023-10-13 22:42:00,600 ----------------------------------------------------------------------------------------------------
204
+ 2023-10-13 22:42:00,600 EPOCH 9 done: loss 0.0110 - lr: 0.000006
205
+ 2023-10-13 22:42:04,597 DEV : loss 0.20399770140647888 - f1-score (micro avg) 0.767
206
+ 2023-10-13 22:42:04,619 saving best model
207
+ 2023-10-13 22:42:05,118 ----------------------------------------------------------------------------------------------------
208
+ 2023-10-13 22:42:11,036 epoch 10 - iter 99/992 - loss 0.00694056 - time (sec): 5.92 - samples/sec: 2777.84 - lr: 0.000005 - momentum: 0.000000
209
+ 2023-10-13 22:42:17,345 epoch 10 - iter 198/992 - loss 0.00836868 - time (sec): 12.23 - samples/sec: 2802.73 - lr: 0.000004 - momentum: 0.000000
210
+ 2023-10-13 22:42:22,986 epoch 10 - iter 297/992 - loss 0.00886989 - time (sec): 17.87 - samples/sec: 2785.15 - lr: 0.000004 - momentum: 0.000000
211
+ 2023-10-13 22:42:28,741 epoch 10 - iter 396/992 - loss 0.00862895 - time (sec): 23.62 - samples/sec: 2804.19 - lr: 0.000003 - momentum: 0.000000
212
+ 2023-10-13 22:42:35,030 epoch 10 - iter 495/992 - loss 0.00902308 - time (sec): 29.91 - samples/sec: 2785.89 - lr: 0.000003 - momentum: 0.000000
213
+ 2023-10-13 22:42:40,907 epoch 10 - iter 594/992 - loss 0.00870118 - time (sec): 35.79 - samples/sec: 2776.25 - lr: 0.000002 - momentum: 0.000000
214
+ 2023-10-13 22:42:46,703 epoch 10 - iter 693/992 - loss 0.00845402 - time (sec): 41.58 - samples/sec: 2756.78 - lr: 0.000002 - momentum: 0.000000
215
+ 2023-10-13 22:42:52,747 epoch 10 - iter 792/992 - loss 0.00817124 - time (sec): 47.63 - samples/sec: 2751.49 - lr: 0.000001 - momentum: 0.000000
216
+ 2023-10-13 22:42:58,806 epoch 10 - iter 891/992 - loss 0.00784789 - time (sec): 53.69 - samples/sec: 2751.57 - lr: 0.000001 - momentum: 0.000000
217
+ 2023-10-13 22:43:04,469 epoch 10 - iter 990/992 - loss 0.00753226 - time (sec): 59.35 - samples/sec: 2757.97 - lr: 0.000000 - momentum: 0.000000
218
+ 2023-10-13 22:43:04,580 ----------------------------------------------------------------------------------------------------
219
+ 2023-10-13 22:43:04,580 EPOCH 10 done: loss 0.0075 - lr: 0.000000
220
+ 2023-10-13 22:43:07,991 DEV : loss 0.20772945880889893 - f1-score (micro avg) 0.7698
221
+ 2023-10-13 22:43:08,013 saving best model
222
+ 2023-10-13 22:43:08,849 ----------------------------------------------------------------------------------------------------
223
+ 2023-10-13 22:43:08,850 Loading model from best epoch ...
224
+ 2023-10-13 22:43:10,240 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
225
+ 2023-10-13 22:43:13,645
226
+ Results:
227
+ - F-score (micro) 0.7805
228
+ - F-score (macro) 0.6917
229
+ - Accuracy 0.6599
230
+
231
+ By class:
232
+ precision recall f1-score support
233
+
234
+ LOC 0.8191 0.8641 0.8410 655
235
+ PER 0.7171 0.8296 0.7692 223
236
+ ORG 0.4912 0.4409 0.4647 127
237
+
238
+ micro avg 0.7592 0.8030 0.7805 1005
239
+ macro avg 0.6758 0.7116 0.6917 1005
240
+ weighted avg 0.7550 0.8030 0.7775 1005
241
+
242
+ 2023-10-13 22:43:13,646 ----------------------------------------------------------------------------------------------------