Hugging Face's logo

stefan-it
/
hmbench-letemps-fr-hmbert-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-1

Token Classification
Flair
PyTorch
French
sequence-tagger-model
Model card Files Community
1
stefan-it commited on
Commit
eb83e71
·
1 Parent(s): d375000

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 +240 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8ddfa3258042b97a8a64c1405ad19b75912023185c4a618b42615e61e3bf9c90
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 18:35:45 0.0000 0.2746 0.1470 0.5080 0.6865 0.5839 0.4237
3
+ 2 18:39:02 0.0000 0.0970 0.1427 0.5730 0.6739 0.6193 0.4555
4
+ 3 18:42:17 0.0000 0.0730 0.1966 0.5240 0.7632 0.6213 0.4626
5
+ 4 18:45:18 0.0000 0.0514 0.2727 0.5171 0.7803 0.6220 0.4630
6
+ 5 18:48:33 0.0000 0.0386 0.3290 0.5429 0.8043 0.6482 0.4899
7
+ 6 18:51:50 0.0000 0.0272 0.3021 0.5614 0.7220 0.6316 0.4720
8
+ 7 18:55:05 0.0000 0.0203 0.3369 0.5313 0.7574 0.6245 0.4642
9
+ 8 18:58:20 0.0000 0.0134 0.3728 0.5353 0.7632 0.6292 0.4704
10
+ 9 19:01:36 0.0000 0.0079 0.3810 0.5484 0.7838 0.6453 0.4858
11
+ 10 19:04:30 0.0000 0.0053 0.3955 0.5550 0.7620 0.6422 0.4830
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,240 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-14 18:32:53,099 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-14 18:32:53,100 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
+ 2023-10-14 18:32:53,100 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-14 18:32:53,100 MultiCorpus: 14465 train + 1392 dev + 2432 test sentences
52
+ - NER_HIPE_2022 Corpus: 14465 train + 1392 dev + 2432 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/letemps/fr/with_doc_seperator
53
+ 2023-10-14 18:32:53,100 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-14 18:32:53,100 Train: 14465 sentences
55
+ 2023-10-14 18:32:53,100 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-14 18:32:53,100 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-14 18:32:53,100 Training Params:
58
+ 2023-10-14 18:32:53,101 - learning_rate: "3e-05"
59
+ 2023-10-14 18:32:53,101 - mini_batch_size: "4"
60
+ 2023-10-14 18:32:53,101 - max_epochs: "10"
61
+ 2023-10-14 18:32:53,101 - shuffle: "True"
62
+ 2023-10-14 18:32:53,101 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-14 18:32:53,101 Plugins:
64
+ 2023-10-14 18:32:53,101 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-14 18:32:53,101 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-14 18:32:53,101 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-14 18:32:53,101 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-14 18:32:53,101 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-14 18:32:53,101 Computation:
70
+ 2023-10-14 18:32:53,101 - compute on device: cuda:0
71
+ 2023-10-14 18:32:53,101 - embedding storage: none
72
+ 2023-10-14 18:32:53,101 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-14 18:32:53,101 Model training base path: "hmbench-letemps/fr-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-1"
74
+ 2023-10-14 18:32:53,101 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-14 18:32:53,101 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-14 18:33:11,011 epoch 1 - iter 361/3617 - loss 1.49793342 - time (sec): 17.91 - samples/sec: 2144.90 - lr: 0.000003 - momentum: 0.000000
77
+ 2023-10-14 18:33:28,000 epoch 1 - iter 722/3617 - loss 0.85914597 - time (sec): 34.90 - samples/sec: 2185.64 - lr: 0.000006 - momentum: 0.000000
78
+ 2023-10-14 18:33:44,513 epoch 1 - iter 1083/3617 - loss 0.63406158 - time (sec): 51.41 - samples/sec: 2213.73 - lr: 0.000009 - momentum: 0.000000
79
+ 2023-10-14 18:34:01,032 epoch 1 - iter 1444/3617 - loss 0.50536750 - time (sec): 67.93 - samples/sec: 2252.43 - lr: 0.000012 - momentum: 0.000000
80
+ 2023-10-14 18:34:17,641 epoch 1 - iter 1805/3617 - loss 0.43148914 - time (sec): 84.54 - samples/sec: 2252.23 - lr: 0.000015 - momentum: 0.000000
81
+ 2023-10-14 18:34:33,662 epoch 1 - iter 2166/3617 - loss 0.38143642 - time (sec): 100.56 - samples/sec: 2265.58 - lr: 0.000018 - momentum: 0.000000
82
+ 2023-10-14 18:34:50,400 epoch 1 - iter 2527/3617 - loss 0.34330335 - time (sec): 117.30 - samples/sec: 2260.05 - lr: 0.000021 - momentum: 0.000000
83
+ 2023-10-14 18:35:06,830 epoch 1 - iter 2888/3617 - loss 0.31397767 - time (sec): 133.73 - samples/sec: 2279.40 - lr: 0.000024 - momentum: 0.000000
84
+ 2023-10-14 18:35:23,147 epoch 1 - iter 3249/3617 - loss 0.29300894 - time (sec): 150.04 - samples/sec: 2278.06 - lr: 0.000027 - momentum: 0.000000
85
+ 2023-10-14 18:35:39,897 epoch 1 - iter 3610/3617 - loss 0.27489902 - time (sec): 166.80 - samples/sec: 2273.67 - lr: 0.000030 - momentum: 0.000000
86
+ 2023-10-14 18:35:40,216 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-14 18:35:40,217 EPOCH 1 done: loss 0.2746 - lr: 0.000030
88
+ 2023-10-14 18:35:45,572 DEV : loss 0.1469813734292984 - f1-score (micro avg) 0.5839
89
+ 2023-10-14 18:35:45,603 saving best model
90
+ 2023-10-14 18:35:46,156 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-14 18:36:05,275 epoch 2 - iter 361/3617 - loss 0.09657648 - time (sec): 19.12 - samples/sec: 1994.77 - lr: 0.000030 - momentum: 0.000000
92
+ 2023-10-14 18:36:24,315 epoch 2 - iter 722/3617 - loss 0.09526196 - time (sec): 38.16 - samples/sec: 2002.34 - lr: 0.000029 - momentum: 0.000000
93
+ 2023-10-14 18:36:43,401 epoch 2 - iter 1083/3617 - loss 0.09794662 - time (sec): 57.24 - samples/sec: 2007.27 - lr: 0.000029 - momentum: 0.000000
94
+ 2023-10-14 18:37:02,359 epoch 2 - iter 1444/3617 - loss 0.10000382 - time (sec): 76.20 - samples/sec: 1996.00 - lr: 0.000029 - momentum: 0.000000
95
+ 2023-10-14 18:37:21,435 epoch 2 - iter 1805/3617 - loss 0.09984349 - time (sec): 95.28 - samples/sec: 1984.72 - lr: 0.000028 - momentum: 0.000000
96
+ 2023-10-14 18:37:40,211 epoch 2 - iter 2166/3617 - loss 0.09842580 - time (sec): 114.05 - samples/sec: 1997.07 - lr: 0.000028 - momentum: 0.000000
97
+ 2023-10-14 18:37:59,219 epoch 2 - iter 2527/3617 - loss 0.09886238 - time (sec): 133.06 - samples/sec: 1999.42 - lr: 0.000028 - momentum: 0.000000
98
+ 2023-10-14 18:38:18,114 epoch 2 - iter 2888/3617 - loss 0.09750237 - time (sec): 151.96 - samples/sec: 1997.68 - lr: 0.000027 - momentum: 0.000000
99
+ 2023-10-14 18:38:36,932 epoch 2 - iter 3249/3617 - loss 0.09684901 - time (sec): 170.77 - samples/sec: 2000.95 - lr: 0.000027 - momentum: 0.000000
100
+ 2023-10-14 18:38:55,778 epoch 2 - iter 3610/3617 - loss 0.09693424 - time (sec): 189.62 - samples/sec: 2000.05 - lr: 0.000027 - momentum: 0.000000
101
+ 2023-10-14 18:38:56,131 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-14 18:38:56,131 EPOCH 2 done: loss 0.0970 - lr: 0.000027
103
+ 2023-10-14 18:39:02,429 DEV : loss 0.14265793561935425 - f1-score (micro avg) 0.6193
104
+ 2023-10-14 18:39:02,457 saving best model
105
+ 2023-10-14 18:39:02,982 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-14 18:39:21,781 epoch 3 - iter 361/3617 - loss 0.06955254 - time (sec): 18.79 - samples/sec: 1884.93 - lr: 0.000026 - momentum: 0.000000
107
+ 2023-10-14 18:39:40,836 epoch 3 - iter 722/3617 - loss 0.06922338 - time (sec): 37.85 - samples/sec: 1943.55 - lr: 0.000026 - momentum: 0.000000
108
+ 2023-10-14 18:39:59,620 epoch 3 - iter 1083/3617 - loss 0.07232891 - time (sec): 56.63 - samples/sec: 1967.42 - lr: 0.000026 - momentum: 0.000000
109
+ 2023-10-14 18:40:18,296 epoch 3 - iter 1444/3617 - loss 0.07277401 - time (sec): 75.31 - samples/sec: 2002.36 - lr: 0.000025 - momentum: 0.000000
110
+ 2023-10-14 18:40:36,983 epoch 3 - iter 1805/3617 - loss 0.07138983 - time (sec): 94.00 - samples/sec: 2006.08 - lr: 0.000025 - momentum: 0.000000
111
+ 2023-10-14 18:40:55,720 epoch 3 - iter 2166/3617 - loss 0.07379404 - time (sec): 112.73 - samples/sec: 2016.27 - lr: 0.000025 - momentum: 0.000000
112
+ 2023-10-14 18:41:14,768 epoch 3 - iter 2527/3617 - loss 0.07306798 - time (sec): 131.78 - samples/sec: 2017.47 - lr: 0.000024 - momentum: 0.000000
113
+ 2023-10-14 18:41:33,623 epoch 3 - iter 2888/3617 - loss 0.07400133 - time (sec): 150.64 - samples/sec: 2014.28 - lr: 0.000024 - momentum: 0.000000
114
+ 2023-10-14 18:41:52,426 epoch 3 - iter 3249/3617 - loss 0.07385904 - time (sec): 169.44 - samples/sec: 2012.00 - lr: 0.000024 - momentum: 0.000000
115
+ 2023-10-14 18:42:11,275 epoch 3 - iter 3610/3617 - loss 0.07300804 - time (sec): 188.29 - samples/sec: 2014.74 - lr: 0.000023 - momentum: 0.000000
116
+ 2023-10-14 18:42:11,633 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-14 18:42:11,633 EPOCH 3 done: loss 0.0730 - lr: 0.000023
118
+ 2023-10-14 18:42:17,786 DEV : loss 0.19656439125537872 - f1-score (micro avg) 0.6213
119
+ 2023-10-14 18:42:17,815 saving best model
120
+ 2023-10-14 18:42:18,413 ----------------------------------------------------------------------------------------------------
121
+ 2023-10-14 18:42:37,058 epoch 4 - iter 361/3617 - loss 0.04567558 - time (sec): 18.64 - samples/sec: 2104.23 - lr: 0.000023 - momentum: 0.000000
122
+ 2023-10-14 18:42:55,665 epoch 4 - iter 722/3617 - loss 0.04907222 - time (sec): 37.25 - samples/sec: 2057.31 - lr: 0.000023 - momentum: 0.000000
123
+ 2023-10-14 18:43:14,466 epoch 4 - iter 1083/3617 - loss 0.04822914 - time (sec): 56.05 - samples/sec: 2037.22 - lr: 0.000022 - momentum: 0.000000
124
+ 2023-10-14 18:43:31,749 epoch 4 - iter 1444/3617 - loss 0.04834016 - time (sec): 73.33 - samples/sec: 2053.38 - lr: 0.000022 - momentum: 0.000000
125
+ 2023-10-14 18:43:48,091 epoch 4 - iter 1805/3617 - loss 0.04851364 - time (sec): 89.68 - samples/sec: 2110.20 - lr: 0.000022 - momentum: 0.000000
126
+ 2023-10-14 18:44:04,310 epoch 4 - iter 2166/3617 - loss 0.04902986 - time (sec): 105.90 - samples/sec: 2140.07 - lr: 0.000021 - momentum: 0.000000
127
+ 2023-10-14 18:44:20,694 epoch 4 - iter 2527/3617 - loss 0.04948457 - time (sec): 122.28 - samples/sec: 2166.31 - lr: 0.000021 - momentum: 0.000000
128
+ 2023-10-14 18:44:36,979 epoch 4 - iter 2888/3617 - loss 0.05023602 - time (sec): 138.57 - samples/sec: 2189.47 - lr: 0.000021 - momentum: 0.000000
129
+ 2023-10-14 18:44:54,061 epoch 4 - iter 3249/3617 - loss 0.05065579 - time (sec): 155.65 - samples/sec: 2194.24 - lr: 0.000020 - momentum: 0.000000
130
+ 2023-10-14 18:45:12,956 epoch 4 - iter 3610/3617 - loss 0.05150400 - time (sec): 174.54 - samples/sec: 2172.39 - lr: 0.000020 - momentum: 0.000000
131
+ 2023-10-14 18:45:13,320 ----------------------------------------------------------------------------------------------------
132
+ 2023-10-14 18:45:13,320 EPOCH 4 done: loss 0.0514 - lr: 0.000020
133
+ 2023-10-14 18:45:18,927 DEV : loss 0.27270472049713135 - f1-score (micro avg) 0.622
134
+ 2023-10-14 18:45:18,960 saving best model
135
+ 2023-10-14 18:45:19,676 ----------------------------------------------------------------------------------------------------
136
+ 2023-10-14 18:45:36,826 epoch 5 - iter 361/3617 - loss 0.03804518 - time (sec): 17.15 - samples/sec: 2296.61 - lr: 0.000020 - momentum: 0.000000
137
+ 2023-10-14 18:45:55,785 epoch 5 - iter 722/3617 - loss 0.03841622 - time (sec): 36.11 - samples/sec: 2135.41 - lr: 0.000019 - momentum: 0.000000
138
+ 2023-10-14 18:46:14,740 epoch 5 - iter 1083/3617 - loss 0.03924782 - time (sec): 55.06 - samples/sec: 2090.46 - lr: 0.000019 - momentum: 0.000000
139
+ 2023-10-14 18:46:33,750 epoch 5 - iter 1444/3617 - loss 0.04097206 - time (sec): 74.07 - samples/sec: 2051.11 - lr: 0.000019 - momentum: 0.000000
140
+ 2023-10-14 18:46:53,472 epoch 5 - iter 1805/3617 - loss 0.03960546 - time (sec): 93.79 - samples/sec: 2036.03 - lr: 0.000018 - momentum: 0.000000
141
+ 2023-10-14 18:47:12,427 epoch 5 - iter 2166/3617 - loss 0.04004957 - time (sec): 112.75 - samples/sec: 2034.50 - lr: 0.000018 - momentum: 0.000000
142
+ 2023-10-14 18:47:31,472 epoch 5 - iter 2527/3617 - loss 0.03876434 - time (sec): 131.79 - samples/sec: 2032.00 - lr: 0.000018 - momentum: 0.000000
143
+ 2023-10-14 18:47:50,424 epoch 5 - iter 2888/3617 - loss 0.03792790 - time (sec): 150.75 - samples/sec: 2021.21 - lr: 0.000017 - momentum: 0.000000
144
+ 2023-10-14 18:48:09,228 epoch 5 - iter 3249/3617 - loss 0.03873215 - time (sec): 169.55 - samples/sec: 2022.58 - lr: 0.000017 - momentum: 0.000000
145
+ 2023-10-14 18:48:28,017 epoch 5 - iter 3610/3617 - loss 0.03865296 - time (sec): 188.34 - samples/sec: 2013.85 - lr: 0.000017 - momentum: 0.000000
146
+ 2023-10-14 18:48:28,376 ----------------------------------------------------------------------------------------------------
147
+ 2023-10-14 18:48:28,377 EPOCH 5 done: loss 0.0386 - lr: 0.000017
148
+ 2023-10-14 18:48:33,939 DEV : loss 0.32902124524116516 - f1-score (micro avg) 0.6482
149
+ 2023-10-14 18:48:33,971 saving best model
150
+ 2023-10-14 18:48:34,499 ----------------------------------------------------------------------------------------------------
151
+ 2023-10-14 18:48:53,479 epoch 6 - iter 361/3617 - loss 0.02506183 - time (sec): 18.98 - samples/sec: 2000.72 - lr: 0.000016 - momentum: 0.000000
152
+ 2023-10-14 18:49:12,488 epoch 6 - iter 722/3617 - loss 0.02531762 - time (sec): 37.98 - samples/sec: 1980.58 - lr: 0.000016 - momentum: 0.000000
153
+ 2023-10-14 18:49:31,432 epoch 6 - iter 1083/3617 - loss 0.02644327 - time (sec): 56.93 - samples/sec: 1964.83 - lr: 0.000016 - momentum: 0.000000
154
+ 2023-10-14 18:49:50,306 epoch 6 - iter 1444/3617 - loss 0.02682723 - time (sec): 75.80 - samples/sec: 1972.03 - lr: 0.000015 - momentum: 0.000000
155
+ 2023-10-14 18:50:09,249 epoch 6 - iter 1805/3617 - loss 0.02642346 - time (sec): 94.75 - samples/sec: 1987.90 - lr: 0.000015 - momentum: 0.000000
156
+ 2023-10-14 18:50:28,143 epoch 6 - iter 2166/3617 - loss 0.02688605 - time (sec): 113.64 - samples/sec: 1995.11 - lr: 0.000015 - momentum: 0.000000
157
+ 2023-10-14 18:50:47,094 epoch 6 - iter 2527/3617 - loss 0.02597740 - time (sec): 132.59 - samples/sec: 1991.28 - lr: 0.000014 - momentum: 0.000000
158
+ 2023-10-14 18:51:06,110 epoch 6 - iter 2888/3617 - loss 0.02657743 - time (sec): 151.61 - samples/sec: 2004.75 - lr: 0.000014 - momentum: 0.000000
159
+ 2023-10-14 18:51:24,765 epoch 6 - iter 3249/3617 - loss 0.02701549 - time (sec): 170.26 - samples/sec: 2001.72 - lr: 0.000014 - momentum: 0.000000
160
+ 2023-10-14 18:51:43,525 epoch 6 - iter 3610/3617 - loss 0.02721336 - time (sec): 189.02 - samples/sec: 2006.30 - lr: 0.000013 - momentum: 0.000000
161
+ 2023-10-14 18:51:43,876 ----------------------------------------------------------------------------------------------------
162
+ 2023-10-14 18:51:43,877 EPOCH 6 done: loss 0.0272 - lr: 0.000013
163
+ 2023-10-14 18:51:50,080 DEV : loss 0.3020953834056854 - f1-score (micro avg) 0.6316
164
+ 2023-10-14 18:51:50,109 ----------------------------------------------------------------------------------------------------
165
+ 2023-10-14 18:52:09,031 epoch 7 - iter 361/3617 - loss 0.01566413 - time (sec): 18.92 - samples/sec: 2050.15 - lr: 0.000013 - momentum: 0.000000
166
+ 2023-10-14 18:52:28,180 epoch 7 - iter 722/3617 - loss 0.01602060 - time (sec): 38.07 - samples/sec: 2024.60 - lr: 0.000013 - momentum: 0.000000
167
+ 2023-10-14 18:52:47,111 epoch 7 - iter 1083/3617 - loss 0.01755375 - time (sec): 57.00 - samples/sec: 2017.47 - lr: 0.000012 - momentum: 0.000000
168
+ 2023-10-14 18:53:05,964 epoch 7 - iter 1444/3617 - loss 0.01898152 - time (sec): 75.85 - samples/sec: 2014.71 - lr: 0.000012 - momentum: 0.000000
169
+ 2023-10-14 18:53:24,801 epoch 7 - iter 1805/3617 - loss 0.02029282 - time (sec): 94.69 - samples/sec: 2013.35 - lr: 0.000012 - momentum: 0.000000
170
+ 2023-10-14 18:53:43,723 epoch 7 - iter 2166/3617 - loss 0.01945143 - time (sec): 113.61 - samples/sec: 2011.18 - lr: 0.000011 - momentum: 0.000000
171
+ 2023-10-14 18:54:02,484 epoch 7 - iter 2527/3617 - loss 0.02008824 - time (sec): 132.37 - samples/sec: 2009.24 - lr: 0.000011 - momentum: 0.000000
172
+ 2023-10-14 18:54:21,323 epoch 7 - iter 2888/3617 - loss 0.02026028 - time (sec): 151.21 - samples/sec: 2011.04 - lr: 0.000011 - momentum: 0.000000
173
+ 2023-10-14 18:54:40,125 epoch 7 - iter 3249/3617 - loss 0.02016038 - time (sec): 170.01 - samples/sec: 2007.85 - lr: 0.000010 - momentum: 0.000000
174
+ 2023-10-14 18:54:58,847 epoch 7 - iter 3610/3617 - loss 0.02025047 - time (sec): 188.74 - samples/sec: 2009.28 - lr: 0.000010 - momentum: 0.000000
175
+ 2023-10-14 18:54:59,208 ----------------------------------------------------------------------------------------------------
176
+ 2023-10-14 18:54:59,209 EPOCH 7 done: loss 0.0203 - lr: 0.000010
177
+ 2023-10-14 18:55:05,497 DEV : loss 0.33693569898605347 - f1-score (micro avg) 0.6245
178
+ 2023-10-14 18:55:05,525 ----------------------------------------------------------------------------------------------------
179
+ 2023-10-14 18:55:24,336 epoch 8 - iter 361/3617 - loss 0.01327463 - time (sec): 18.81 - samples/sec: 1964.88 - lr: 0.000010 - momentum: 0.000000
180
+ 2023-10-14 18:55:43,223 epoch 8 - iter 722/3617 - loss 0.01473725 - time (sec): 37.70 - samples/sec: 1989.05 - lr: 0.000009 - momentum: 0.000000
181
+ 2023-10-14 18:56:02,086 epoch 8 - iter 1083/3617 - loss 0.01272106 - time (sec): 56.56 - samples/sec: 2011.45 - lr: 0.000009 - momentum: 0.000000
182
+ 2023-10-14 18:56:20,916 epoch 8 - iter 1444/3617 - loss 0.01306734 - time (sec): 75.39 - samples/sec: 2011.73 - lr: 0.000009 - momentum: 0.000000
183
+ 2023-10-14 18:56:39,693 epoch 8 - iter 1805/3617 - loss 0.01336643 - time (sec): 94.17 - samples/sec: 2006.28 - lr: 0.000008 - momentum: 0.000000
184
+ 2023-10-14 18:56:58,759 epoch 8 - iter 2166/3617 - loss 0.01331363 - time (sec): 113.23 - samples/sec: 1982.43 - lr: 0.000008 - momentum: 0.000000
185
+ 2023-10-14 18:57:17,808 epoch 8 - iter 2527/3617 - loss 0.01360786 - time (sec): 132.28 - samples/sec: 1995.37 - lr: 0.000008 - momentum: 0.000000
186
+ 2023-10-14 18:57:36,722 epoch 8 - iter 2888/3617 - loss 0.01309021 - time (sec): 151.20 - samples/sec: 1993.35 - lr: 0.000007 - momentum: 0.000000
187
+ 2023-10-14 18:57:55,357 epoch 8 - iter 3249/3617 - loss 0.01286633 - time (sec): 169.83 - samples/sec: 2004.16 - lr: 0.000007 - momentum: 0.000000
188
+ 2023-10-14 18:58:14,255 epoch 8 - iter 3610/3617 - loss 0.01337274 - time (sec): 188.73 - samples/sec: 2008.80 - lr: 0.000007 - momentum: 0.000000
189
+ 2023-10-14 18:58:14,611 ----------------------------------------------------------------------------------------------------
190
+ 2023-10-14 18:58:14,611 EPOCH 8 done: loss 0.0134 - lr: 0.000007
191
+ 2023-10-14 18:58:20,871 DEV : loss 0.3728400766849518 - f1-score (micro avg) 0.6292
192
+ 2023-10-14 18:58:20,900 ----------------------------------------------------------------------------------------------------
193
+ 2023-10-14 18:58:39,836 epoch 9 - iter 361/3617 - loss 0.00636871 - time (sec): 18.93 - samples/sec: 2004.28 - lr: 0.000006 - momentum: 0.000000
194
+ 2023-10-14 18:58:58,882 epoch 9 - iter 722/3617 - loss 0.00907563 - time (sec): 37.98 - samples/sec: 2028.07 - lr: 0.000006 - momentum: 0.000000
195
+ 2023-10-14 18:59:17,616 epoch 9 - iter 1083/3617 - loss 0.00828252 - time (sec): 56.72 - samples/sec: 2036.27 - lr: 0.000006 - momentum: 0.000000
196
+ 2023-10-14 18:59:36,465 epoch 9 - iter 1444/3617 - loss 0.00898628 - time (sec): 75.56 - samples/sec: 2012.65 - lr: 0.000005 - momentum: 0.000000
197
+ 2023-10-14 18:59:55,560 epoch 9 - iter 1805/3617 - loss 0.00817358 - time (sec): 94.66 - samples/sec: 2003.84 - lr: 0.000005 - momentum: 0.000000
198
+ 2023-10-14 19:00:14,343 epoch 9 - iter 2166/3617 - loss 0.00817291 - time (sec): 113.44 - samples/sec: 2002.35 - lr: 0.000005 - momentum: 0.000000
199
+ 2023-10-14 19:00:33,183 epoch 9 - iter 2527/3617 - loss 0.00832167 - time (sec): 132.28 - samples/sec: 1994.85 - lr: 0.000004 - momentum: 0.000000
200
+ 2023-10-14 19:00:52,035 epoch 9 - iter 2888/3617 - loss 0.00797829 - time (sec): 151.13 - samples/sec: 1996.91 - lr: 0.000004 - momentum: 0.000000
201
+ 2023-10-14 19:01:10,800 epoch 9 - iter 3249/3617 - loss 0.00799275 - time (sec): 169.90 - samples/sec: 2003.98 - lr: 0.000004 - momentum: 0.000000
202
+ 2023-10-14 19:01:30,189 epoch 9 - iter 3610/3617 - loss 0.00795879 - time (sec): 189.29 - samples/sec: 2003.69 - lr: 0.000003 - momentum: 0.000000
203
+ 2023-10-14 19:01:30,546 ----------------------------------------------------------------------------------------------------
204
+ 2023-10-14 19:01:30,546 EPOCH 9 done: loss 0.0079 - lr: 0.000003
205
+ 2023-10-14 19:01:36,746 DEV : loss 0.3810341954231262 - f1-score (micro avg) 0.6453
206
+ 2023-10-14 19:01:36,775 ----------------------------------------------------------------------------------------------------
207
+ 2023-10-14 19:01:55,572 epoch 10 - iter 361/3617 - loss 0.00233359 - time (sec): 18.80 - samples/sec: 2041.41 - lr: 0.000003 - momentum: 0.000000
208
+ 2023-10-14 19:02:12,702 epoch 10 - iter 722/3617 - loss 0.00511219 - time (sec): 35.93 - samples/sec: 2151.13 - lr: 0.000003 - momentum: 0.000000
209
+ 2023-10-14 19:02:29,091 epoch 10 - iter 1083/3617 - loss 0.00547615 - time (sec): 52.32 - samples/sec: 2185.00 - lr: 0.000002 - momentum: 0.000000
210
+ 2023-10-14 19:02:45,447 epoch 10 - iter 1444/3617 - loss 0.00486842 - time (sec): 68.67 - samples/sec: 2216.76 - lr: 0.000002 - momentum: 0.000000
211
+ 2023-10-14 19:03:02,177 epoch 10 - iter 1805/3617 - loss 0.00509004 - time (sec): 85.40 - samples/sec: 2231.64 - lr: 0.000002 - momentum: 0.000000
212
+ 2023-10-14 19:03:19,019 epoch 10 - iter 2166/3617 - loss 0.00541113 - time (sec): 102.24 - samples/sec: 2238.08 - lr: 0.000001 - momentum: 0.000000
213
+ 2023-10-14 19:03:35,397 epoch 10 - iter 2527/3617 - loss 0.00522420 - time (sec): 118.62 - samples/sec: 2236.37 - lr: 0.000001 - momentum: 0.000000
214
+ 2023-10-14 19:03:51,652 epoch 10 - iter 2888/3617 - loss 0.00553422 - time (sec): 134.88 - samples/sec: 2230.99 - lr: 0.000001 - momentum: 0.000000
215
+ 2023-10-14 19:04:08,143 epoch 10 - iter 3249/3617 - loss 0.00538172 - time (sec): 151.37 - samples/sec: 2243.70 - lr: 0.000000 - momentum: 0.000000
216
+ 2023-10-14 19:04:24,734 epoch 10 - iter 3610/3617 - loss 0.00535691 - time (sec): 167.96 - samples/sec: 2258.53 - lr: 0.000000 - momentum: 0.000000
217
+ 2023-10-14 19:04:25,047 ----------------------------------------------------------------------------------------------------
218
+ 2023-10-14 19:04:25,047 EPOCH 10 done: loss 0.0053 - lr: 0.000000
219
+ 2023-10-14 19:04:30,571 DEV : loss 0.395516961812973 - f1-score (micro avg) 0.6422
220
+ 2023-10-14 19:04:31,008 ----------------------------------------------------------------------------------------------------
221
+ 2023-10-14 19:04:31,009 Loading model from best epoch ...
222
+ 2023-10-14 19:04:33,509 SequenceTagger predicts: Dictionary with 13 tags: O, S-loc, B-loc, E-loc, I-loc, S-pers, B-pers, E-pers, I-pers, S-org, B-org, E-org, I-org
223
+ 2023-10-14 19:04:40,299
224
+ Results:
225
+ - F-score (micro) 0.6431
226
+ - F-score (macro) 0.5165
227
+ - Accuracy 0.4908
228
+
229
+ By class:
230
+ precision recall f1-score support
231
+
232
+ loc 0.6159 0.7868 0.6909 591
233
+ pers 0.5281 0.8151 0.6410 357
234
+ org 0.2353 0.2025 0.2177 79
235
+
236
+ micro avg 0.5619 0.7517 0.6431 1027
237
+ macro avg 0.4598 0.6015 0.5165 1027
238
+ weighted avg 0.5561 0.7517 0.6372 1027
239
+
240
+ 2023-10-14 19:04:40,299 ----------------------------------------------------------------------------------------------------