Hugging Face's logo

stefan-it
/
hmbench-hipe2020-de-hmbert-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-3

Token Classification
Flair
PyTorch
German
sequence-tagger-model
Model card Files Community
1
stefan-it commited on
Commit
786a0c5
•
1 Parent(s): a1adb2c

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 +243 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:07ac1f902f4a3067ab4c392df3f1a7e4158591ffe0866ecc0c7ce3b5071acab2
3
+ size 443335879
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 12:56:01 0.0000 0.5935 0.1924 0.6141 0.5512 0.5810 0.4196
3
+ 2 12:56:52 0.0000 0.1618 0.1460 0.6879 0.7271 0.7070 0.5636
4
+ 3 12:57:42 0.0000 0.0919 0.1695 0.7033 0.7412 0.7217 0.5848
5
+ 4 12:58:33 0.0000 0.0594 0.2061 0.7695 0.7568 0.7631 0.6314
6
+ 5 12:59:26 0.0000 0.0435 0.2056 0.7655 0.7912 0.7782 0.6516
7
+ 6 13:00:18 0.0000 0.0277 0.2001 0.7429 0.7795 0.7608 0.6326
8
+ 7 13:01:09 0.0000 0.0183 0.2389 0.7820 0.7740 0.7780 0.6517
9
+ 8 13:02:01 0.0000 0.0131 0.2317 0.7614 0.7858 0.7734 0.6459
10
+ 9 13:02:51 0.0000 0.0084 0.2373 0.7682 0.7928 0.7803 0.6559
11
+ 10 13:03:43 0.0000 0.0054 0.2383 0.7747 0.7850 0.7798 0.6545
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,243 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-13 12:55:14,652 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-13 12:55:14,653 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=21, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-13 12:55:14,653 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-13 12:55:14,653 MultiCorpus: 3575 train + 1235 dev + 1266 test sentences
52
+ - NER_HIPE_2022 Corpus: 3575 train + 1235 dev + 1266 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/hipe2020/de/with_doc_seperator
53
+ 2023-10-13 12:55:14,653 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-13 12:55:14,653 Train: 3575 sentences
55
+ 2023-10-13 12:55:14,653 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-13 12:55:14,653 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-13 12:55:14,653 Training Params:
58
+ 2023-10-13 12:55:14,653 - learning_rate: "3e-05"
59
+ 2023-10-13 12:55:14,653 - mini_batch_size: "4"
60
+ 2023-10-13 12:55:14,653 - max_epochs: "10"
61
+ 2023-10-13 12:55:14,653 - shuffle: "True"
62
+ 2023-10-13 12:55:14,653 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-13 12:55:14,653 Plugins:
64
+ 2023-10-13 12:55:14,654 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-13 12:55:14,654 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-13 12:55:14,654 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-13 12:55:14,654 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-13 12:55:14,654 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-13 12:55:14,654 Computation:
70
+ 2023-10-13 12:55:14,654 - compute on device: cuda:0
71
+ 2023-10-13 12:55:14,654 - embedding storage: none
72
+ 2023-10-13 12:55:14,654 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-13 12:55:14,654 Model training base path: "hmbench-hipe2020/de-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-3"
74
+ 2023-10-13 12:55:14,654 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-13 12:55:14,654 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-13 12:55:18,848 epoch 1 - iter 89/894 - loss 2.69593370 - time (sec): 4.19 - samples/sec: 2000.73 - lr: 0.000003 - momentum: 0.000000
77
+ 2023-10-13 12:55:23,146 epoch 1 - iter 178/894 - loss 1.71958865 - time (sec): 8.49 - samples/sec: 2028.76 - lr: 0.000006 - momentum: 0.000000
78
+ 2023-10-13 12:55:27,200 epoch 1 - iter 267/894 - loss 1.32701386 - time (sec): 12.54 - samples/sec: 2010.55 - lr: 0.000009 - momentum: 0.000000
79
+ 2023-10-13 12:55:31,451 epoch 1 - iter 356/894 - loss 1.05861260 - time (sec): 16.80 - samples/sec: 2078.11 - lr: 0.000012 - momentum: 0.000000
80
+ 2023-10-13 12:55:35,742 epoch 1 - iter 445/894 - loss 0.90380884 - time (sec): 21.09 - samples/sec: 2070.33 - lr: 0.000015 - momentum: 0.000000
81
+ 2023-10-13 12:55:39,944 epoch 1 - iter 534/894 - loss 0.79977481 - time (sec): 25.29 - samples/sec: 2079.85 - lr: 0.000018 - momentum: 0.000000
82
+ 2023-10-13 12:55:44,011 epoch 1 - iter 623/894 - loss 0.73143982 - time (sec): 29.36 - samples/sec: 2066.38 - lr: 0.000021 - momentum: 0.000000
83
+ 2023-10-13 12:55:48,084 epoch 1 - iter 712/894 - loss 0.68005300 - time (sec): 33.43 - samples/sec: 2056.86 - lr: 0.000024 - momentum: 0.000000
84
+ 2023-10-13 12:55:52,305 epoch 1 - iter 801/894 - loss 0.63023691 - time (sec): 37.65 - samples/sec: 2070.97 - lr: 0.000027 - momentum: 0.000000
85
+ 2023-10-13 12:55:56,579 epoch 1 - iter 890/894 - loss 0.59407354 - time (sec): 41.92 - samples/sec: 2057.15 - lr: 0.000030 - momentum: 0.000000
86
+ 2023-10-13 12:55:56,770 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-13 12:55:56,770 EPOCH 1 done: loss 0.5935 - lr: 0.000030
88
+ 2023-10-13 12:56:01,600 DEV : loss 0.19242174923419952 - f1-score (micro avg) 0.581
89
+ 2023-10-13 12:56:01,629 saving best model
90
+ 2023-10-13 12:56:01,951 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-13 12:56:05,928 epoch 2 - iter 89/894 - loss 0.20351765 - time (sec): 3.98 - samples/sec: 2070.12 - lr: 0.000030 - momentum: 0.000000
92
+ 2023-10-13 12:56:09,978 epoch 2 - iter 178/894 - loss 0.19371181 - time (sec): 8.03 - samples/sec: 2100.43 - lr: 0.000029 - momentum: 0.000000
93
+ 2023-10-13 12:56:14,065 epoch 2 - iter 267/894 - loss 0.17774358 - time (sec): 12.11 - samples/sec: 2072.98 - lr: 0.000029 - momentum: 0.000000
94
+ 2023-10-13 12:56:18,280 epoch 2 - iter 356/894 - loss 0.16603566 - time (sec): 16.33 - samples/sec: 2063.76 - lr: 0.000029 - momentum: 0.000000
95
+ 2023-10-13 12:56:22,329 epoch 2 - iter 445/894 - loss 0.17216562 - time (sec): 20.38 - samples/sec: 2057.61 - lr: 0.000028 - momentum: 0.000000
96
+ 2023-10-13 12:56:26,707 epoch 2 - iter 534/894 - loss 0.16847025 - time (sec): 24.75 - samples/sec: 2027.62 - lr: 0.000028 - momentum: 0.000000
97
+ 2023-10-13 12:56:31,007 epoch 2 - iter 623/894 - loss 0.16459087 - time (sec): 29.06 - samples/sec: 2036.86 - lr: 0.000028 - momentum: 0.000000
98
+ 2023-10-13 12:56:35,095 epoch 2 - iter 712/894 - loss 0.16278123 - time (sec): 33.14 - samples/sec: 2049.87 - lr: 0.000027 - momentum: 0.000000
99
+ 2023-10-13 12:56:39,067 epoch 2 - iter 801/894 - loss 0.16215338 - time (sec): 37.11 - samples/sec: 2088.95 - lr: 0.000027 - momentum: 0.000000
100
+ 2023-10-13 12:56:43,379 epoch 2 - iter 890/894 - loss 0.16148233 - time (sec): 41.43 - samples/sec: 2081.25 - lr: 0.000027 - momentum: 0.000000
101
+ 2023-10-13 12:56:43,566 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-13 12:56:43,566 EPOCH 2 done: loss 0.1618 - lr: 0.000027
103
+ 2023-10-13 12:56:52,106 DEV : loss 0.14601190388202667 - f1-score (micro avg) 0.707
104
+ 2023-10-13 12:56:52,135 saving best model
105
+ 2023-10-13 12:56:52,581 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-13 12:56:56,686 epoch 3 - iter 89/894 - loss 0.09364191 - time (sec): 4.10 - samples/sec: 1998.31 - lr: 0.000026 - momentum: 0.000000
107
+ 2023-10-13 12:57:00,659 epoch 3 - iter 178/894 - loss 0.08926448 - time (sec): 8.07 - samples/sec: 1991.65 - lr: 0.000026 - momentum: 0.000000
108
+ 2023-10-13 12:57:04,820 epoch 3 - iter 267/894 - loss 0.08910757 - time (sec): 12.23 - samples/sec: 2046.42 - lr: 0.000026 - momentum: 0.000000
109
+ 2023-10-13 12:57:08,871 epoch 3 - iter 356/894 - loss 0.09229143 - time (sec): 16.29 - samples/sec: 2054.75 - lr: 0.000025 - momentum: 0.000000
110
+ 2023-10-13 12:57:13,084 epoch 3 - iter 445/894 - loss 0.09330711 - time (sec): 20.50 - samples/sec: 2026.30 - lr: 0.000025 - momentum: 0.000000
111
+ 2023-10-13 12:57:17,109 epoch 3 - iter 534/894 - loss 0.08872648 - time (sec): 24.52 - samples/sec: 2052.27 - lr: 0.000025 - momentum: 0.000000
112
+ 2023-10-13 12:57:21,286 epoch 3 - iter 623/894 - loss 0.08943539 - time (sec): 28.70 - samples/sec: 2046.90 - lr: 0.000024 - momentum: 0.000000
113
+ 2023-10-13 12:57:25,411 epoch 3 - iter 712/894 - loss 0.09181165 - time (sec): 32.83 - samples/sec: 2051.35 - lr: 0.000024 - momentum: 0.000000
114
+ 2023-10-13 12:57:29,642 epoch 3 - iter 801/894 - loss 0.09100852 - time (sec): 37.06 - samples/sec: 2060.24 - lr: 0.000024 - momentum: 0.000000
115
+ 2023-10-13 12:57:34,154 epoch 3 - iter 890/894 - loss 0.09219008 - time (sec): 41.57 - samples/sec: 2072.83 - lr: 0.000023 - momentum: 0.000000
116
+ 2023-10-13 12:57:34,340 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-13 12:57:34,340 EPOCH 3 done: loss 0.0919 - lr: 0.000023
118
+ 2023-10-13 12:57:42,891 DEV : loss 0.169508159160614 - f1-score (micro avg) 0.7217
119
+ 2023-10-13 12:57:42,921 saving best model
120
+ 2023-10-13 12:57:43,338 ----------------------------------------------------------------------------------------------------
121
+ 2023-10-13 12:57:47,481 epoch 4 - iter 89/894 - loss 0.07066538 - time (sec): 4.14 - samples/sec: 2006.71 - lr: 0.000023 - momentum: 0.000000
122
+ 2023-10-13 12:57:51,485 epoch 4 - iter 178/894 - loss 0.06521536 - time (sec): 8.15 - samples/sec: 2053.11 - lr: 0.000023 - momentum: 0.000000
123
+ 2023-10-13 12:57:55,505 epoch 4 - iter 267/894 - loss 0.06095332 - time (sec): 12.17 - samples/sec: 2061.80 - lr: 0.000022 - momentum: 0.000000
124
+ 2023-10-13 12:57:59,865 epoch 4 - iter 356/894 - loss 0.05999315 - time (sec): 16.53 - samples/sec: 2133.54 - lr: 0.000022 - momentum: 0.000000
125
+ 2023-10-13 12:58:03,915 epoch 4 - iter 445/894 - loss 0.05842531 - time (sec): 20.58 - samples/sec: 2114.71 - lr: 0.000022 - momentum: 0.000000
126
+ 2023-10-13 12:58:08,237 epoch 4 - iter 534/894 - loss 0.05992046 - time (sec): 24.90 - samples/sec: 2095.16 - lr: 0.000021 - momentum: 0.000000
127
+ 2023-10-13 12:58:12,418 epoch 4 - iter 623/894 - loss 0.05852169 - time (sec): 29.08 - samples/sec: 2093.88 - lr: 0.000021 - momentum: 0.000000
128
+ 2023-10-13 12:58:16,496 epoch 4 - iter 712/894 - loss 0.05864176 - time (sec): 33.16 - samples/sec: 2086.38 - lr: 0.000021 - momentum: 0.000000
129
+ 2023-10-13 12:58:20,933 epoch 4 - iter 801/894 - loss 0.05904040 - time (sec): 37.59 - samples/sec: 2083.56 - lr: 0.000020 - momentum: 0.000000
130
+ 2023-10-13 12:58:24,881 epoch 4 - iter 890/894 - loss 0.05946598 - time (sec): 41.54 - samples/sec: 2076.25 - lr: 0.000020 - momentum: 0.000000
131
+ 2023-10-13 12:58:25,052 ----------------------------------------------------------------------------------------------------
132
+ 2023-10-13 12:58:25,052 EPOCH 4 done: loss 0.0594 - lr: 0.000020
133
+ 2023-10-13 12:58:33,488 DEV : loss 0.20608599483966827 - f1-score (micro avg) 0.7631
134
+ 2023-10-13 12:58:33,518 saving best model
135
+ 2023-10-13 12:58:34,001 ----------------------------------------------------------------------------------------------------
136
+ 2023-10-13 12:58:38,060 epoch 5 - iter 89/894 - loss 0.05023316 - time (sec): 4.06 - samples/sec: 2106.72 - lr: 0.000020 - momentum: 0.000000
137
+ 2023-10-13 12:58:42,103 epoch 5 - iter 178/894 - loss 0.04072573 - time (sec): 8.10 - samples/sec: 2066.73 - lr: 0.000019 - momentum: 0.000000
138
+ 2023-10-13 12:58:46,306 epoch 5 - iter 267/894 - loss 0.04046872 - time (sec): 12.30 - samples/sec: 2056.71 - lr: 0.000019 - momentum: 0.000000
139
+ 2023-10-13 12:58:50,544 epoch 5 - iter 356/894 - loss 0.04285187 - time (sec): 16.54 - samples/sec: 2012.79 - lr: 0.000019 - momentum: 0.000000
140
+ 2023-10-13 12:58:54,937 epoch 5 - iter 445/894 - loss 0.04118917 - time (sec): 20.93 - samples/sec: 2034.33 - lr: 0.000018 - momentum: 0.000000
141
+ 2023-10-13 12:58:59,181 epoch 5 - iter 534/894 - loss 0.04312736 - time (sec): 25.18 - samples/sec: 2018.89 - lr: 0.000018 - momentum: 0.000000
142
+ 2023-10-13 12:59:03,441 epoch 5 - iter 623/894 - loss 0.04081111 - time (sec): 29.44 - samples/sec: 1999.35 - lr: 0.000018 - momentum: 0.000000
143
+ 2023-10-13 12:59:07,980 epoch 5 - iter 712/894 - loss 0.04391013 - time (sec): 33.98 - samples/sec: 2030.64 - lr: 0.000017 - momentum: 0.000000
144
+ 2023-10-13 12:59:12,804 epoch 5 - iter 801/894 - loss 0.04351538 - time (sec): 38.80 - samples/sec: 2012.42 - lr: 0.000017 - momentum: 0.000000
145
+ 2023-10-13 12:59:17,151 epoch 5 - iter 890/894 - loss 0.04355119 - time (sec): 43.15 - samples/sec: 1999.22 - lr: 0.000017 - momentum: 0.000000
146
+ 2023-10-13 12:59:17,355 ----------------------------------------------------------------------------------------------------
147
+ 2023-10-13 12:59:17,355 EPOCH 5 done: loss 0.0435 - lr: 0.000017
148
+ 2023-10-13 12:59:26,229 DEV : loss 0.20556075870990753 - f1-score (micro avg) 0.7782
149
+ 2023-10-13 12:59:26,260 saving best model
150
+ 2023-10-13 12:59:26,730 ----------------------------------------------------------------------------------------------------
151
+ 2023-10-13 12:59:31,174 epoch 6 - iter 89/894 - loss 0.02131180 - time (sec): 4.44 - samples/sec: 2185.91 - lr: 0.000016 - momentum: 0.000000
152
+ 2023-10-13 12:59:35,181 epoch 6 - iter 178/894 - loss 0.01740524 - time (sec): 8.45 - samples/sec: 2134.49 - lr: 0.000016 - momentum: 0.000000
153
+ 2023-10-13 12:59:39,711 epoch 6 - iter 267/894 - loss 0.01734140 - time (sec): 12.98 - samples/sec: 2072.83 - lr: 0.000016 - momentum: 0.000000
154
+ 2023-10-13 12:59:44,017 epoch 6 - iter 356/894 - loss 0.02350570 - time (sec): 17.28 - samples/sec: 2074.33 - lr: 0.000015 - momentum: 0.000000
155
+ 2023-10-13 12:59:48,252 epoch 6 - iter 445/894 - loss 0.02543387 - time (sec): 21.52 - samples/sec: 2080.84 - lr: 0.000015 - momentum: 0.000000
156
+ 2023-10-13 12:59:52,367 epoch 6 - iter 534/894 - loss 0.02567710 - time (sec): 25.63 - samples/sec: 2060.05 - lr: 0.000015 - momentum: 0.000000
157
+ 2023-10-13 12:59:56,761 epoch 6 - iter 623/894 - loss 0.02518657 - time (sec): 30.03 - samples/sec: 2023.69 - lr: 0.000014 - momentum: 0.000000
158
+ 2023-10-13 13:00:00,899 epoch 6 - iter 712/894 - loss 0.02743067 - time (sec): 34.17 - samples/sec: 2015.73 - lr: 0.000014 - momentum: 0.000000
159
+ 2023-10-13 13:00:05,151 epoch 6 - iter 801/894 - loss 0.02804290 - time (sec): 38.42 - samples/sec: 2022.12 - lr: 0.000014 - momentum: 0.000000
160
+ 2023-10-13 13:00:09,498 epoch 6 - iter 890/894 - loss 0.02756516 - time (sec): 42.77 - samples/sec: 2017.73 - lr: 0.000013 - momentum: 0.000000
161
+ 2023-10-13 13:00:09,683 ----------------------------------------------------------------------------------------------------
162
+ 2023-10-13 13:00:09,684 EPOCH 6 done: loss 0.0277 - lr: 0.000013
163
+ 2023-10-13 13:00:18,535 DEV : loss 0.20008665323257446 - f1-score (micro avg) 0.7608
164
+ 2023-10-13 13:00:18,570 ----------------------------------------------------------------------------------------------------
165
+ 2023-10-13 13:00:22,657 epoch 7 - iter 89/894 - loss 0.02473568 - time (sec): 4.09 - samples/sec: 2283.61 - lr: 0.000013 - momentum: 0.000000
166
+ 2023-10-13 13:00:27,033 epoch 7 - iter 178/894 - loss 0.02214234 - time (sec): 8.46 - samples/sec: 2258.02 - lr: 0.000013 - momentum: 0.000000
167
+ 2023-10-13 13:00:31,190 epoch 7 - iter 267/894 - loss 0.01933197 - time (sec): 12.62 - samples/sec: 2215.83 - lr: 0.000012 - momentum: 0.000000
168
+ 2023-10-13 13:00:35,256 epoch 7 - iter 356/894 - loss 0.01840027 - time (sec): 16.68 - samples/sec: 2236.09 - lr: 0.000012 - momentum: 0.000000
169
+ 2023-10-13 13:00:39,376 epoch 7 - iter 445/894 - loss 0.01736996 - time (sec): 20.81 - samples/sec: 2212.78 - lr: 0.000012 - momentum: 0.000000
170
+ 2023-10-13 13:00:43,341 epoch 7 - iter 534/894 - loss 0.01710256 - time (sec): 24.77 - samples/sec: 2178.14 - lr: 0.000011 - momentum: 0.000000
171
+ 2023-10-13 13:00:47,660 epoch 7 - iter 623/894 - loss 0.01847066 - time (sec): 29.09 - samples/sec: 2116.68 - lr: 0.000011 - momentum: 0.000000
172
+ 2023-10-13 13:00:51,880 epoch 7 - iter 712/894 - loss 0.01768895 - time (sec): 33.31 - samples/sec: 2101.35 - lr: 0.000011 - momentum: 0.000000
173
+ 2023-10-13 13:00:55,976 epoch 7 - iter 801/894 - loss 0.01830601 - time (sec): 37.40 - samples/sec: 2086.22 - lr: 0.000010 - momentum: 0.000000
174
+ 2023-10-13 13:01:00,075 epoch 7 - iter 890/894 - loss 0.01838476 - time (sec): 41.50 - samples/sec: 2074.90 - lr: 0.000010 - momentum: 0.000000
175
+ 2023-10-13 13:01:00,255 ----------------------------------------------------------------------------------------------------
176
+ 2023-10-13 13:01:00,255 EPOCH 7 done: loss 0.0183 - lr: 0.000010
177
+ 2023-10-13 13:01:09,025 DEV : loss 0.23889903724193573 - f1-score (micro avg) 0.778
178
+ 2023-10-13 13:01:09,055 ----------------------------------------------------------------------------------------------------
179
+ 2023-10-13 13:01:13,200 epoch 8 - iter 89/894 - loss 0.01264123 - time (sec): 4.14 - samples/sec: 2109.44 - lr: 0.000010 - momentum: 0.000000
180
+ 2023-10-13 13:01:17,329 epoch 8 - iter 178/894 - loss 0.01441008 - time (sec): 8.27 - samples/sec: 2044.81 - lr: 0.000009 - momentum: 0.000000
181
+ 2023-10-13 13:01:21,833 epoch 8 - iter 267/894 - loss 0.01359192 - time (sec): 12.78 - samples/sec: 2117.06 - lr: 0.000009 - momentum: 0.000000
182
+ 2023-10-13 13:01:26,166 epoch 8 - iter 356/894 - loss 0.01220036 - time (sec): 17.11 - samples/sec: 2075.45 - lr: 0.000009 - momentum: 0.000000
183
+ 2023-10-13 13:01:30,406 epoch 8 - iter 445/894 - loss 0.01217291 - time (sec): 21.35 - samples/sec: 2063.59 - lr: 0.000008 - momentum: 0.000000
184
+ 2023-10-13 13:01:34,751 epoch 8 - iter 534/894 - loss 0.01395930 - time (sec): 25.69 - samples/sec: 2061.54 - lr: 0.000008 - momentum: 0.000000
185
+ 2023-10-13 13:01:38,837 epoch 8 - iter 623/894 - loss 0.01363106 - time (sec): 29.78 - samples/sec: 2061.57 - lr: 0.000008 - momentum: 0.000000
186
+ 2023-10-13 13:01:43,036 epoch 8 - iter 712/894 - loss 0.01331376 - time (sec): 33.98 - samples/sec: 2049.69 - lr: 0.000007 - momentum: 0.000000
187
+ 2023-10-13 13:01:47,676 epoch 8 - iter 801/894 - loss 0.01313221 - time (sec): 38.62 - samples/sec: 2032.02 - lr: 0.000007 - momentum: 0.000000
188
+ 2023-10-13 13:01:52,010 epoch 8 - iter 890/894 - loss 0.01300296 - time (sec): 42.95 - samples/sec: 2008.08 - lr: 0.000007 - momentum: 0.000000
189
+ 2023-10-13 13:01:52,187 ----------------------------------------------------------------------------------------------------
190
+ 2023-10-13 13:01:52,188 EPOCH 8 done: loss 0.0131 - lr: 0.000007
191
+ 2023-10-13 13:02:01,000 DEV : loss 0.23165372014045715 - f1-score (micro avg) 0.7734
192
+ 2023-10-13 13:02:01,032 ----------------------------------------------------------------------------------------------------
193
+ 2023-10-13 13:02:05,174 epoch 9 - iter 89/894 - loss 0.00510865 - time (sec): 4.14 - samples/sec: 1989.00 - lr: 0.000006 - momentum: 0.000000
194
+ 2023-10-13 13:02:09,509 epoch 9 - iter 178/894 - loss 0.00451444 - time (sec): 8.48 - samples/sec: 2090.14 - lr: 0.000006 - momentum: 0.000000
195
+ 2023-10-13 13:02:13,747 epoch 9 - iter 267/894 - loss 0.00647433 - time (sec): 12.71 - samples/sec: 2067.70 - lr: 0.000006 - momentum: 0.000000
196
+ 2023-10-13 13:02:18,039 epoch 9 - iter 356/894 - loss 0.00857438 - time (sec): 17.01 - samples/sec: 2043.58 - lr: 0.000005 - momentum: 0.000000
197
+ 2023-10-13 13:02:22,150 epoch 9 - iter 445/894 - loss 0.01024290 - time (sec): 21.12 - samples/sec: 2049.28 - lr: 0.000005 - momentum: 0.000000
198
+ 2023-10-13 13:02:26,250 epoch 9 - iter 534/894 - loss 0.00899714 - time (sec): 25.22 - samples/sec: 2043.77 - lr: 0.000005 - momentum: 0.000000
199
+ 2023-10-13 13:02:30,329 epoch 9 - iter 623/894 - loss 0.00792510 - time (sec): 29.30 - samples/sec: 2035.48 - lr: 0.000004 - momentum: 0.000000
200
+ 2023-10-13 13:02:34,640 epoch 9 - iter 712/894 - loss 0.00742665 - time (sec): 33.61 - samples/sec: 2032.31 - lr: 0.000004 - momentum: 0.000000
201
+ 2023-10-13 13:02:38,893 epoch 9 - iter 801/894 - loss 0.00799065 - time (sec): 37.86 - samples/sec: 2058.77 - lr: 0.000004 - momentum: 0.000000
202
+ 2023-10-13 13:02:42,941 epoch 9 - iter 890/894 - loss 0.00818853 - time (sec): 41.91 - samples/sec: 2056.11 - lr: 0.000003 - momentum: 0.000000
203
+ 2023-10-13 13:02:43,123 ----------------------------------------------------------------------------------------------------
204
+ 2023-10-13 13:02:43,123 EPOCH 9 done: loss 0.0084 - lr: 0.000003
205
+ 2023-10-13 13:02:51,867 DEV : loss 0.2373068630695343 - f1-score (micro avg) 0.7803
206
+ 2023-10-13 13:02:51,898 saving best model
207
+ 2023-10-13 13:02:52,380 ----------------------------------------------------------------------------------------------------
208
+ 2023-10-13 13:02:56,667 epoch 10 - iter 89/894 - loss 0.00299024 - time (sec): 4.28 - samples/sec: 2137.65 - lr: 0.000003 - momentum: 0.000000
209
+ 2023-10-13 13:03:00,728 epoch 10 - iter 178/894 - loss 0.00773541 - time (sec): 8.34 - samples/sec: 2091.33 - lr: 0.000003 - momentum: 0.000000
210
+ 2023-10-13 13:03:05,105 epoch 10 - iter 267/894 - loss 0.00678755 - time (sec): 12.72 - samples/sec: 2063.17 - lr: 0.000002 - momentum: 0.000000
211
+ 2023-10-13 13:03:09,449 epoch 10 - iter 356/894 - loss 0.00753006 - time (sec): 17.06 - samples/sec: 2048.86 - lr: 0.000002 - momentum: 0.000000
212
+ 2023-10-13 13:03:13,636 epoch 10 - iter 445/894 - loss 0.00720676 - time (sec): 21.25 - samples/sec: 2025.78 - lr: 0.000002 - momentum: 0.000000
213
+ 2023-10-13 13:03:17,696 epoch 10 - iter 534/894 - loss 0.00687943 - time (sec): 25.31 - samples/sec: 2026.65 - lr: 0.000001 - momentum: 0.000000
214
+ 2023-10-13 13:03:22,040 epoch 10 - iter 623/894 - loss 0.00586593 - time (sec): 29.65 - samples/sec: 2033.97 - lr: 0.000001 - momentum: 0.000000
215
+ 2023-10-13 13:03:26,626 epoch 10 - iter 712/894 - loss 0.00544570 - time (sec): 34.24 - samples/sec: 2055.88 - lr: 0.000001 - momentum: 0.000000
216
+ 2023-10-13 13:03:30,764 epoch 10 - iter 801/894 - loss 0.00536573 - time (sec): 38.38 - samples/sec: 2038.80 - lr: 0.000000 - momentum: 0.000000
217
+ 2023-10-13 13:03:34,874 epoch 10 - iter 890/894 - loss 0.00540931 - time (sec): 42.49 - samples/sec: 2027.55 - lr: 0.000000 - momentum: 0.000000
218
+ 2023-10-13 13:03:35,057 ----------------------------------------------------------------------------------------------------
219
+ 2023-10-13 13:03:35,058 EPOCH 10 done: loss 0.0054 - lr: 0.000000
220
+ 2023-10-13 13:03:43,593 DEV : loss 0.23833510279655457 - f1-score (micro avg) 0.7798
221
+ 2023-10-13 13:03:43,975 ----------------------------------------------------------------------------------------------------
222
+ 2023-10-13 13:03:43,976 Loading model from best epoch ...
223
+ 2023-10-13 13:03:45,339 SequenceTagger predicts: Dictionary with 21 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, S-prod, B-prod, E-prod, I-prod, S-time, B-time, E-time, I-time
224
+ 2023-10-13 13:03:49,655
225
+ Results:
226
+ - F-score (micro) 0.7373
227
+ - F-score (macro) 0.662
228
+ - Accuracy 0.6036
229
+
230
+ By class:
231
+ precision recall f1-score support
232
+
233
+ loc 0.8468 0.8255 0.8360 596
234
+ pers 0.6525 0.7387 0.6930 333
235
+ org 0.4965 0.5303 0.5128 132
236
+ prod 0.6531 0.4848 0.5565 66
237
+ time 0.6727 0.7551 0.7115 49
238
+
239
+ micro avg 0.7290 0.7457 0.7373 1176
240
+ macro avg 0.6643 0.6669 0.6620 1176
241
+ weighted avg 0.7343 0.7457 0.7384 1176
242
+
243
+ 2023-10-13 13:03:49,655 ----------------------------------------------------------------------------------------------------