Hugging Face's logo

stefan-it
/
hmbench-hipe2020-de-hmbert-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-5

Token Classification
Flair
PyTorch
German
sequence-tagger-model
Model card Files Community
1
stefan-it commited on
Commit
54365e7
·
1 Parent(s): 9e1e18c

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 +247 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bb97d07f977e13664541f5fa722090786be50068b377f8fa9c8771bef07c2cd8
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 14:08:58 0.0000 0.5727 0.1898 0.5619 0.6247 0.5916 0.4357
3
+ 2 14:09:51 0.0000 0.1577 0.1675 0.6934 0.7021 0.6977 0.5543
4
+ 3 14:10:40 0.0000 0.0939 0.2000 0.7538 0.7060 0.7291 0.5921
5
+ 4 14:11:32 0.0000 0.0638 0.2237 0.7581 0.7326 0.7451 0.6080
6
+ 5 14:12:22 0.0000 0.0441 0.2385 0.7788 0.7185 0.7475 0.6143
7
+ 6 14:13:13 0.0000 0.0293 0.2394 0.7629 0.7522 0.7575 0.6255
8
+ 7 14:14:07 0.0000 0.0209 0.2525 0.7654 0.7576 0.7615 0.6321
9
+ 8 14:15:00 0.0000 0.0094 0.2569 0.7765 0.7553 0.7658 0.6368
10
+ 9 14:15:52 0.0000 0.0069 0.2671 0.7773 0.7694 0.7733 0.6444
11
+ 10 14:16:43 0.0000 0.0051 0.2596 0.7819 0.7654 0.7736 0.6445
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,247 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-13 14:08:11,374 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-13 14:08:11,375 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 14:08:11,375 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-13 14:08:11,375 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 14:08:11,375 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-13 14:08:11,375 Train: 3575 sentences
55
+ 2023-10-13 14:08:11,375 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-13 14:08:11,375 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-13 14:08:11,375 Training Params:
58
+ 2023-10-13 14:08:11,375 - learning_rate: "5e-05"
59
+ 2023-10-13 14:08:11,375 - mini_batch_size: "4"
60
+ 2023-10-13 14:08:11,375 - max_epochs: "10"
61
+ 2023-10-13 14:08:11,375 - shuffle: "True"
62
+ 2023-10-13 14:08:11,375 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-13 14:08:11,375 Plugins:
64
+ 2023-10-13 14:08:11,375 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-13 14:08:11,375 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-13 14:08:11,375 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-13 14:08:11,375 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-13 14:08:11,375 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-13 14:08:11,375 Computation:
70
+ 2023-10-13 14:08:11,376 - compute on device: cuda:0
71
+ 2023-10-13 14:08:11,376 - embedding storage: none
72
+ 2023-10-13 14:08:11,376 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-13 14:08:11,376 Model training base path: "hmbench-hipe2020/de-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-5"
74
+ 2023-10-13 14:08:11,376 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-13 14:08:11,376 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-13 14:08:15,898 epoch 1 - iter 89/894 - loss 2.52923575 - time (sec): 4.52 - samples/sec: 2182.09 - lr: 0.000005 - momentum: 0.000000
77
+ 2023-10-13 14:08:20,012 epoch 1 - iter 178/894 - loss 1.58342524 - time (sec): 8.64 - samples/sec: 2207.18 - lr: 0.000010 - momentum: 0.000000
78
+ 2023-10-13 14:08:24,203 epoch 1 - iter 267/894 - loss 1.23639423 - time (sec): 12.83 - samples/sec: 2136.52 - lr: 0.000015 - momentum: 0.000000
79
+ 2023-10-13 14:08:28,550 epoch 1 - iter 356/894 - loss 1.01715624 - time (sec): 17.17 - samples/sec: 2108.54 - lr: 0.000020 - momentum: 0.000000
80
+ 2023-10-13 14:08:32,844 epoch 1 - iter 445/894 - loss 0.87794477 - time (sec): 21.47 - samples/sec: 2090.89 - lr: 0.000025 - momentum: 0.000000
81
+ 2023-10-13 14:08:36,910 epoch 1 - iter 534/894 - loss 0.77886231 - time (sec): 25.53 - samples/sec: 2086.30 - lr: 0.000030 - momentum: 0.000000
82
+ 2023-10-13 14:08:40,939 epoch 1 - iter 623/894 - loss 0.70820496 - time (sec): 29.56 - samples/sec: 2091.56 - lr: 0.000035 - momentum: 0.000000
83
+ 2023-10-13 14:08:45,129 epoch 1 - iter 712/894 - loss 0.65158695 - time (sec): 33.75 - samples/sec: 2073.39 - lr: 0.000040 - momentum: 0.000000
84
+ 2023-10-13 14:08:49,139 epoch 1 - iter 801/894 - loss 0.60985499 - time (sec): 37.76 - samples/sec: 2064.58 - lr: 0.000045 - momentum: 0.000000
85
+ 2023-10-13 14:08:53,278 epoch 1 - iter 890/894 - loss 0.57386967 - time (sec): 41.90 - samples/sec: 2057.76 - lr: 0.000050 - momentum: 0.000000
86
+ 2023-10-13 14:08:53,459 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-13 14:08:53,459 EPOCH 1 done: loss 0.5727 - lr: 0.000050
88
+ 2023-10-13 14:08:58,500 DEV : loss 0.18978139758110046 - f1-score (micro avg) 0.5916
89
+ 2023-10-13 14:08:58,525 saving best model
90
+ 2023-10-13 14:08:58,878 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-13 14:09:03,117 epoch 2 - iter 89/894 - loss 0.21194614 - time (sec): 4.24 - samples/sec: 2181.41 - lr: 0.000049 - momentum: 0.000000
92
+ 2023-10-13 14:09:07,553 epoch 2 - iter 178/894 - loss 0.19054416 - time (sec): 8.67 - samples/sec: 2045.48 - lr: 0.000049 - momentum: 0.000000
93
+ 2023-10-13 14:09:12,119 epoch 2 - iter 267/894 - loss 0.17595858 - time (sec): 13.24 - samples/sec: 1960.01 - lr: 0.000048 - momentum: 0.000000
94
+ 2023-10-13 14:09:16,730 epoch 2 - iter 356/894 - loss 0.17594605 - time (sec): 17.85 - samples/sec: 1879.92 - lr: 0.000048 - momentum: 0.000000
95
+ 2023-10-13 14:09:21,206 epoch 2 - iter 445/894 - loss 0.17142673 - time (sec): 22.33 - samples/sec: 1886.32 - lr: 0.000047 - momentum: 0.000000
96
+ 2023-10-13 14:09:25,650 epoch 2 - iter 534/894 - loss 0.16657106 - time (sec): 26.77 - samples/sec: 1893.29 - lr: 0.000047 - momentum: 0.000000
97
+ 2023-10-13 14:09:29,793 epoch 2 - iter 623/894 - loss 0.16819979 - time (sec): 30.91 - samples/sec: 1920.54 - lr: 0.000046 - momentum: 0.000000
98
+ 2023-10-13 14:09:33,961 epoch 2 - iter 712/894 - loss 0.16248898 - time (sec): 35.08 - samples/sec: 1942.69 - lr: 0.000046 - momentum: 0.000000
99
+ 2023-10-13 14:09:38,270 epoch 2 - iter 801/894 - loss 0.16082101 - time (sec): 39.39 - samples/sec: 1957.79 - lr: 0.000045 - momentum: 0.000000
100
+ 2023-10-13 14:09:42,495 epoch 2 - iter 890/894 - loss 0.15753470 - time (sec): 43.62 - samples/sec: 1972.33 - lr: 0.000044 - momentum: 0.000000
101
+ 2023-10-13 14:09:42,682 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-13 14:09:42,682 EPOCH 2 done: loss 0.1577 - lr: 0.000044
103
+ 2023-10-13 14:09:51,308 DEV : loss 0.16749663650989532 - f1-score (micro avg) 0.6977
104
+ 2023-10-13 14:09:51,338 saving best model
105
+ 2023-10-13 14:09:51,838 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-13 14:09:55,733 epoch 3 - iter 89/894 - loss 0.08007067 - time (sec): 3.89 - samples/sec: 2360.54 - lr: 0.000044 - momentum: 0.000000
107
+ 2023-10-13 14:09:59,650 epoch 3 - iter 178/894 - loss 0.07640027 - time (sec): 7.81 - samples/sec: 2295.76 - lr: 0.000043 - momentum: 0.000000
108
+ 2023-10-13 14:10:03,873 epoch 3 - iter 267/894 - loss 0.08683829 - time (sec): 12.03 - samples/sec: 2266.26 - lr: 0.000043 - momentum: 0.000000
109
+ 2023-10-13 14:10:07,876 epoch 3 - iter 356/894 - loss 0.09438784 - time (sec): 16.04 - samples/sec: 2263.33 - lr: 0.000042 - momentum: 0.000000
110
+ 2023-10-13 14:10:11,886 epoch 3 - iter 445/894 - loss 0.09349886 - time (sec): 20.05 - samples/sec: 2216.06 - lr: 0.000042 - momentum: 0.000000
111
+ 2023-10-13 14:10:15,754 epoch 3 - iter 534/894 - loss 0.09086137 - time (sec): 23.91 - samples/sec: 2207.82 - lr: 0.000041 - momentum: 0.000000
112
+ 2023-10-13 14:10:19,673 epoch 3 - iter 623/894 - loss 0.09013375 - time (sec): 27.83 - samples/sec: 2181.57 - lr: 0.000041 - momentum: 0.000000
113
+ 2023-10-13 14:10:23,738 epoch 3 - iter 712/894 - loss 0.09024790 - time (sec): 31.90 - samples/sec: 2171.31 - lr: 0.000040 - momentum: 0.000000
114
+ 2023-10-13 14:10:27,643 epoch 3 - iter 801/894 - loss 0.09257071 - time (sec): 35.80 - samples/sec: 2178.38 - lr: 0.000039 - momentum: 0.000000
115
+ 2023-10-13 14:10:31,533 epoch 3 - iter 890/894 - loss 0.09331365 - time (sec): 39.69 - samples/sec: 2172.02 - lr: 0.000039 - momentum: 0.000000
116
+ 2023-10-13 14:10:31,707 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-13 14:10:31,707 EPOCH 3 done: loss 0.0939 - lr: 0.000039
118
+ 2023-10-13 14:10:40,166 DEV : loss 0.1999683529138565 - f1-score (micro avg) 0.7291
119
+ 2023-10-13 14:10:40,193 saving best model
120
+ 2023-10-13 14:10:40,669 ----------------------------------------------------------------------------------------------------
121
+ 2023-10-13 14:10:44,976 epoch 4 - iter 89/894 - loss 0.07161300 - time (sec): 4.30 - samples/sec: 2161.37 - lr: 0.000038 - momentum: 0.000000
122
+ 2023-10-13 14:10:49,749 epoch 4 - iter 178/894 - loss 0.06237632 - time (sec): 9.07 - samples/sec: 1927.98 - lr: 0.000038 - momentum: 0.000000
123
+ 2023-10-13 14:10:54,687 epoch 4 - iter 267/894 - loss 0.06571349 - time (sec): 14.01 - samples/sec: 1972.83 - lr: 0.000037 - momentum: 0.000000
124
+ 2023-10-13 14:10:58,893 epoch 4 - iter 356/894 - loss 0.06326807 - time (sec): 18.22 - samples/sec: 1959.55 - lr: 0.000037 - momentum: 0.000000
125
+ 2023-10-13 14:11:03,057 epoch 4 - iter 445/894 - loss 0.06462152 - time (sec): 22.38 - samples/sec: 1972.59 - lr: 0.000036 - momentum: 0.000000
126
+ 2023-10-13 14:11:07,375 epoch 4 - iter 534/894 - loss 0.06356342 - time (sec): 26.70 - samples/sec: 2004.24 - lr: 0.000036 - momentum: 0.000000
127
+ 2023-10-13 14:11:11,407 epoch 4 - iter 623/894 - loss 0.06359237 - time (sec): 30.73 - samples/sec: 2007.14 - lr: 0.000035 - momentum: 0.000000
128
+ 2023-10-13 14:11:15,432 epoch 4 - iter 712/894 - loss 0.06394484 - time (sec): 34.76 - samples/sec: 2010.72 - lr: 0.000034 - momentum: 0.000000
129
+ 2023-10-13 14:11:19,477 epoch 4 - iter 801/894 - loss 0.06383650 - time (sec): 38.80 - samples/sec: 2010.42 - lr: 0.000034 - momentum: 0.000000
130
+ 2023-10-13 14:11:23,578 epoch 4 - iter 890/894 - loss 0.06368701 - time (sec): 42.90 - samples/sec: 2009.83 - lr: 0.000033 - momentum: 0.000000
131
+ 2023-10-13 14:11:23,755 ----------------------------------------------------------------------------------------------------
132
+ 2023-10-13 14:11:23,755 EPOCH 4 done: loss 0.0638 - lr: 0.000033
133
+ 2023-10-13 14:11:32,038 DEV : loss 0.22368095815181732 - f1-score (micro avg) 0.7451
134
+ 2023-10-13 14:11:32,065 saving best model
135
+ 2023-10-13 14:11:32,473 ----------------------------------------------------------------------------------------------------
136
+ 2023-10-13 14:11:36,581 epoch 5 - iter 89/894 - loss 0.05326395 - time (sec): 4.10 - samples/sec: 2126.63 - lr: 0.000033 - momentum: 0.000000
137
+ 2023-10-13 14:11:40,582 epoch 5 - iter 178/894 - loss 0.04466412 - time (sec): 8.10 - samples/sec: 2102.13 - lr: 0.000032 - momentum: 0.000000
138
+ 2023-10-13 14:11:44,899 epoch 5 - iter 267/894 - loss 0.04574289 - time (sec): 12.42 - samples/sec: 2112.80 - lr: 0.000032 - momentum: 0.000000
139
+ 2023-10-13 14:11:48,919 epoch 5 - iter 356/894 - loss 0.04504252 - time (sec): 16.44 - samples/sec: 2116.26 - lr: 0.000031 - momentum: 0.000000
140
+ 2023-10-13 14:11:53,162 epoch 5 - iter 445/894 - loss 0.04436721 - time (sec): 20.68 - samples/sec: 2130.98 - lr: 0.000031 - momentum: 0.000000
141
+ 2023-10-13 14:11:57,502 epoch 5 - iter 534/894 - loss 0.04372206 - time (sec): 25.02 - samples/sec: 2107.92 - lr: 0.000030 - momentum: 0.000000
142
+ 2023-10-13 14:12:01,606 epoch 5 - iter 623/894 - loss 0.04442546 - time (sec): 29.13 - samples/sec: 2096.61 - lr: 0.000029 - momentum: 0.000000
143
+ 2023-10-13 14:12:05,829 epoch 5 - iter 712/894 - loss 0.04469383 - time (sec): 33.35 - samples/sec: 2095.05 - lr: 0.000029 - momentum: 0.000000
144
+ 2023-10-13 14:12:09,867 epoch 5 - iter 801/894 - loss 0.04557369 - time (sec): 37.39 - samples/sec: 2080.06 - lr: 0.000028 - momentum: 0.000000
145
+ 2023-10-13 14:12:14,004 epoch 5 - iter 890/894 - loss 0.04426329 - time (sec): 41.52 - samples/sec: 2074.21 - lr: 0.000028 - momentum: 0.000000
146
+ 2023-10-13 14:12:14,193 ----------------------------------------------------------------------------------------------------
147
+ 2023-10-13 14:12:14,193 EPOCH 5 done: loss 0.0441 - lr: 0.000028
148
+ 2023-10-13 14:12:22,702 DEV : loss 0.23852457106113434 - f1-score (micro avg) 0.7475
149
+ 2023-10-13 14:12:22,729 saving best model
150
+ 2023-10-13 14:12:23,193 ----------------------------------------------------------------------------------------------------
151
+ 2023-10-13 14:12:27,238 epoch 6 - iter 89/894 - loss 0.01889556 - time (sec): 4.04 - samples/sec: 2069.29 - lr: 0.000027 - momentum: 0.000000
152
+ 2023-10-13 14:12:31,500 epoch 6 - iter 178/894 - loss 0.02626898 - time (sec): 8.30 - samples/sec: 2019.45 - lr: 0.000027 - momentum: 0.000000
153
+ 2023-10-13 14:12:36,124 epoch 6 - iter 267/894 - loss 0.02606331 - time (sec): 12.93 - samples/sec: 2097.55 - lr: 0.000026 - momentum: 0.000000
154
+ 2023-10-13 14:12:40,260 epoch 6 - iter 356/894 - loss 0.02937353 - time (sec): 17.06 - samples/sec: 2099.91 - lr: 0.000026 - momentum: 0.000000
155
+ 2023-10-13 14:12:44,576 epoch 6 - iter 445/894 - loss 0.02750190 - time (sec): 21.38 - samples/sec: 2134.02 - lr: 0.000025 - momentum: 0.000000
156
+ 2023-10-13 14:12:48,628 epoch 6 - iter 534/894 - loss 0.02780421 - time (sec): 25.43 - samples/sec: 2105.98 - lr: 0.000024 - momentum: 0.000000
157
+ 2023-10-13 14:12:52,732 epoch 6 - iter 623/894 - loss 0.02856228 - time (sec): 29.54 - samples/sec: 2085.01 - lr: 0.000024 - momentum: 0.000000
158
+ 2023-10-13 14:12:56,924 epoch 6 - iter 712/894 - loss 0.02984563 - time (sec): 33.73 - samples/sec: 2073.64 - lr: 0.000023 - momentum: 0.000000
159
+ 2023-10-13 14:13:00,987 epoch 6 - iter 801/894 - loss 0.02950493 - time (sec): 37.79 - samples/sec: 2076.16 - lr: 0.000023 - momentum: 0.000000
160
+ 2023-10-13 14:13:05,151 epoch 6 - iter 890/894 - loss 0.02941361 - time (sec): 41.96 - samples/sec: 2053.84 - lr: 0.000022 - momentum: 0.000000
161
+ 2023-10-13 14:13:05,332 ----------------------------------------------------------------------------------------------------
162
+ 2023-10-13 14:13:05,333 EPOCH 6 done: loss 0.0293 - lr: 0.000022
163
+ 2023-10-13 14:13:13,835 DEV : loss 0.23937579989433289 - f1-score (micro avg) 0.7575
164
+ 2023-10-13 14:13:13,863 saving best model
165
+ 2023-10-13 14:13:14,366 ----------------------------------------------------------------------------------------------------
166
+ 2023-10-13 14:13:18,512 epoch 7 - iter 89/894 - loss 0.01705401 - time (sec): 4.14 - samples/sec: 1884.74 - lr: 0.000022 - momentum: 0.000000
167
+ 2023-10-13 14:13:22,533 epoch 7 - iter 178/894 - loss 0.01759034 - time (sec): 8.17 - samples/sec: 1917.37 - lr: 0.000021 - momentum: 0.000000
168
+ 2023-10-13 14:13:27,065 epoch 7 - iter 267/894 - loss 0.01849552 - time (sec): 12.70 - samples/sec: 1997.91 - lr: 0.000021 - momentum: 0.000000
169
+ 2023-10-13 14:13:31,220 epoch 7 - iter 356/894 - loss 0.02024248 - time (sec): 16.85 - samples/sec: 2024.97 - lr: 0.000020 - momentum: 0.000000
170
+ 2023-10-13 14:13:35,408 epoch 7 - iter 445/894 - loss 0.02149739 - time (sec): 21.04 - samples/sec: 2048.75 - lr: 0.000019 - momentum: 0.000000
171
+ 2023-10-13 14:13:39,801 epoch 7 - iter 534/894 - loss 0.01928417 - time (sec): 25.43 - samples/sec: 2018.47 - lr: 0.000019 - momentum: 0.000000
172
+ 2023-10-13 14:13:44,296 epoch 7 - iter 623/894 - loss 0.02198540 - time (sec): 29.93 - samples/sec: 1990.90 - lr: 0.000018 - momentum: 0.000000
173
+ 2023-10-13 14:13:49,168 epoch 7 - iter 712/894 - loss 0.02082455 - time (sec): 34.80 - samples/sec: 1978.14 - lr: 0.000018 - momentum: 0.000000
174
+ 2023-10-13 14:13:53,734 epoch 7 - iter 801/894 - loss 0.02160685 - time (sec): 39.37 - samples/sec: 1966.85 - lr: 0.000017 - momentum: 0.000000
175
+ 2023-10-13 14:13:58,424 epoch 7 - iter 890/894 - loss 0.02101591 - time (sec): 44.06 - samples/sec: 1955.83 - lr: 0.000017 - momentum: 0.000000
176
+ 2023-10-13 14:13:58,627 ----------------------------------------------------------------------------------------------------
177
+ 2023-10-13 14:13:58,627 EPOCH 7 done: loss 0.0209 - lr: 0.000017
178
+ 2023-10-13 14:14:07,516 DEV : loss 0.2525382936000824 - f1-score (micro avg) 0.7615
179
+ 2023-10-13 14:14:07,560 saving best model
180
+ 2023-10-13 14:14:08,054 ----------------------------------------------------------------------------------------------------
181
+ 2023-10-13 14:14:12,693 epoch 8 - iter 89/894 - loss 0.01225659 - time (sec): 4.64 - samples/sec: 1823.64 - lr: 0.000016 - momentum: 0.000000
182
+ 2023-10-13 14:14:17,331 epoch 8 - iter 178/894 - loss 0.00698362 - time (sec): 9.28 - samples/sec: 1794.94 - lr: 0.000016 - momentum: 0.000000
183
+ 2023-10-13 14:14:22,187 epoch 8 - iter 267/894 - loss 0.00689475 - time (sec): 14.13 - samples/sec: 1845.24 - lr: 0.000015 - momentum: 0.000000
184
+ 2023-10-13 14:14:26,650 epoch 8 - iter 356/894 - loss 0.00651321 - time (sec): 18.59 - samples/sec: 1906.91 - lr: 0.000014 - momentum: 0.000000
185
+ 2023-10-13 14:14:30,820 epoch 8 - iter 445/894 - loss 0.00846668 - time (sec): 22.76 - samples/sec: 1928.85 - lr: 0.000014 - momentum: 0.000000
186
+ 2023-10-13 14:14:35,003 epoch 8 - iter 534/894 - loss 0.00904011 - time (sec): 26.95 - samples/sec: 1942.81 - lr: 0.000013 - momentum: 0.000000
187
+ 2023-10-13 14:14:39,112 epoch 8 - iter 623/894 - loss 0.00884653 - time (sec): 31.06 - samples/sec: 1959.93 - lr: 0.000013 - momentum: 0.000000
188
+ 2023-10-13 14:14:43,204 epoch 8 - iter 712/894 - loss 0.00918400 - time (sec): 35.15 - samples/sec: 1971.65 - lr: 0.000012 - momentum: 0.000000
189
+ 2023-10-13 14:14:47,448 epoch 8 - iter 801/894 - loss 0.00950962 - time (sec): 39.39 - samples/sec: 1969.14 - lr: 0.000012 - momentum: 0.000000
190
+ 2023-10-13 14:14:51,628 epoch 8 - iter 890/894 - loss 0.00947519 - time (sec): 43.57 - samples/sec: 1978.33 - lr: 0.000011 - momentum: 0.000000
191
+ 2023-10-13 14:14:51,804 ----------------------------------------------------------------------------------------------------
192
+ 2023-10-13 14:14:51,804 EPOCH 8 done: loss 0.0094 - lr: 0.000011
193
+ 2023-10-13 14:15:00,785 DEV : loss 0.2568976879119873 - f1-score (micro avg) 0.7658
194
+ 2023-10-13 14:15:00,815 saving best model
195
+ 2023-10-13 14:15:01,340 ----------------------------------------------------------------------------------------------------
196
+ 2023-10-13 14:15:05,956 epoch 9 - iter 89/894 - loss 0.00391659 - time (sec): 4.61 - samples/sec: 1787.79 - lr: 0.000011 - momentum: 0.000000
197
+ 2023-10-13 14:15:10,412 epoch 9 - iter 178/894 - loss 0.00453941 - time (sec): 9.07 - samples/sec: 1958.05 - lr: 0.000010 - momentum: 0.000000
198
+ 2023-10-13 14:15:14,661 epoch 9 - iter 267/894 - loss 0.00419557 - time (sec): 13.32 - samples/sec: 2014.76 - lr: 0.000009 - momentum: 0.000000
199
+ 2023-10-13 14:15:18,739 epoch 9 - iter 356/894 - loss 0.00626206 - time (sec): 17.40 - samples/sec: 2030.77 - lr: 0.000009 - momentum: 0.000000
200
+ 2023-10-13 14:15:22,908 epoch 9 - iter 445/894 - loss 0.00766227 - time (sec): 21.57 - samples/sec: 2017.53 - lr: 0.000008 - momentum: 0.000000
201
+ 2023-10-13 14:15:26,928 epoch 9 - iter 534/894 - loss 0.00757263 - time (sec): 25.59 - samples/sec: 2035.91 - lr: 0.000008 - momentum: 0.000000
202
+ 2023-10-13 14:15:31,064 epoch 9 - iter 623/894 - loss 0.00774753 - time (sec): 29.72 - samples/sec: 2044.80 - lr: 0.000007 - momentum: 0.000000
203
+ 2023-10-13 14:15:35,087 epoch 9 - iter 712/894 - loss 0.00696736 - time (sec): 33.74 - samples/sec: 2055.60 - lr: 0.000007 - momentum: 0.000000
204
+ 2023-10-13 14:15:39,286 epoch 9 - iter 801/894 - loss 0.00720776 - time (sec): 37.94 - samples/sec: 2045.59 - lr: 0.000006 - momentum: 0.000000
205
+ 2023-10-13 14:15:43,335 epoch 9 - iter 890/894 - loss 0.00695611 - time (sec): 41.99 - samples/sec: 2052.94 - lr: 0.000006 - momentum: 0.000000
206
+ 2023-10-13 14:15:43,509 ----------------------------------------------------------------------------------------------------
207
+ 2023-10-13 14:15:43,509 EPOCH 9 done: loss 0.0069 - lr: 0.000006
208
+ 2023-10-13 14:15:52,300 DEV : loss 0.26707446575164795 - f1-score (micro avg) 0.7733
209
+ 2023-10-13 14:15:52,332 saving best model
210
+ 2023-10-13 14:15:52,842 ----------------------------------------------------------------------------------------------------
211
+ 2023-10-13 14:15:56,987 epoch 10 - iter 89/894 - loss 0.01316612 - time (sec): 4.14 - samples/sec: 2068.39 - lr: 0.000005 - momentum: 0.000000
212
+ 2023-10-13 14:16:01,198 epoch 10 - iter 178/894 - loss 0.00979750 - time (sec): 8.35 - samples/sec: 1970.51 - lr: 0.000004 - momentum: 0.000000
213
+ 2023-10-13 14:16:05,313 epoch 10 - iter 267/894 - loss 0.00661329 - time (sec): 12.47 - samples/sec: 2005.21 - lr: 0.000004 - momentum: 0.000000
214
+ 2023-10-13 14:16:09,316 epoch 10 - iter 356/894 - loss 0.00704547 - time (sec): 16.47 - samples/sec: 2011.59 - lr: 0.000003 - momentum: 0.000000
215
+ 2023-10-13 14:16:13,661 epoch 10 - iter 445/894 - loss 0.00624802 - time (sec): 20.82 - samples/sec: 2058.36 - lr: 0.000003 - momentum: 0.000000
216
+ 2023-10-13 14:16:18,017 epoch 10 - iter 534/894 - loss 0.00619476 - time (sec): 25.17 - samples/sec: 2071.72 - lr: 0.000002 - momentum: 0.000000
217
+ 2023-10-13 14:16:22,076 epoch 10 - iter 623/894 - loss 0.00569995 - time (sec): 29.23 - samples/sec: 2070.04 - lr: 0.000002 - momentum: 0.000000
218
+ 2023-10-13 14:16:26,128 epoch 10 - iter 712/894 - loss 0.00579249 - time (sec): 33.28 - samples/sec: 2059.38 - lr: 0.000001 - momentum: 0.000000
219
+ 2023-10-13 14:16:30,235 epoch 10 - iter 801/894 - loss 0.00526208 - time (sec): 37.39 - samples/sec: 2083.28 - lr: 0.000001 - momentum: 0.000000
220
+ 2023-10-13 14:16:34,265 epoch 10 - iter 890/894 - loss 0.00512287 - time (sec): 41.42 - samples/sec: 2080.80 - lr: 0.000000 - momentum: 0.000000
221
+ 2023-10-13 14:16:34,441 ----------------------------------------------------------------------------------------------------
222
+ 2023-10-13 14:16:34,441 EPOCH 10 done: loss 0.0051 - lr: 0.000000
223
+ 2023-10-13 14:16:43,026 DEV : loss 0.2595662772655487 - f1-score (micro avg) 0.7736
224
+ 2023-10-13 14:16:43,057 saving best model
225
+ 2023-10-13 14:16:44,098 ----------------------------------------------------------------------------------------------------
226
+ 2023-10-13 14:16:44,100 Loading model from best epoch ...
227
+ 2023-10-13 14:16:45,623 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
228
+ 2023-10-13 14:16:50,883
229
+ Results:
230
+ - F-score (micro) 0.7347
231
+ - F-score (macro) 0.6609
232
+ - Accuracy 0.5974
233
+
234
+ By class:
235
+ precision recall f1-score support
236
+
237
+ loc 0.8176 0.8423 0.8298 596
238
+ pers 0.6575 0.7147 0.6849 333
239
+ org 0.5225 0.4394 0.4774 132
240
+ prod 0.6522 0.4545 0.5357 66
241
+ time 0.7407 0.8163 0.7767 49
242
+
243
+ micro avg 0.7313 0.7381 0.7347 1176
244
+ macro avg 0.6781 0.6535 0.6609 1176
245
+ weighted avg 0.7266 0.7381 0.7305 1176
246
+
247
+ 2023-10-13 14:16:50,883 ----------------------------------------------------------------------------------------------------