Hugging Face's logo

stefan-it
/
hmbench-hipe2020-de-hmbert-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3

Token Classification
Flair
PyTorch
German
sequence-tagger-model
Model card Files Community
1
stefan-it commited on
Commit
b357a9c
·
1 Parent(s): 916a535

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 +244 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eebd18dc0d746ba85f6d36babd2ee198a4b5a1e11c662e6760f2c3b68d5dee46
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 13:05:04 0.0000 0.5152 0.1790 0.6490 0.5653 0.6043 0.4468
3
+ 2 13:05:54 0.0000 0.1631 0.1371 0.7075 0.6677 0.6870 0.5408
4
+ 3 13:06:46 0.0000 0.0973 0.1664 0.7937 0.7068 0.7477 0.6121
5
+ 4 13:07:36 0.0000 0.0661 0.2019 0.7584 0.7561 0.7572 0.6239
6
+ 5 13:08:28 0.0000 0.0421 0.2171 0.7029 0.7842 0.7413 0.6079
7
+ 6 13:09:20 0.0000 0.0306 0.2127 0.7316 0.7780 0.7541 0.6238
8
+ 7 13:10:11 0.0000 0.0199 0.2425 0.7679 0.7811 0.7744 0.6487
9
+ 8 13:11:02 0.0000 0.0112 0.2334 0.8035 0.7639 0.7832 0.6570
10
+ 9 13:11:54 0.0000 0.0061 0.2255 0.7649 0.7936 0.7790 0.6565
11
+ 10 13:12:44 0.0000 0.0043 0.2382 0.7834 0.7834 0.7834 0.6618
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,244 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-13 13:04:16,455 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-13 13:04:16,455 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 13:04:16,456 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-13 13:04:16,456 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 13:04:16,456 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-13 13:04:16,456 Train: 3575 sentences
55
+ 2023-10-13 13:04:16,456 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-13 13:04:16,456 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-13 13:04:16,456 Training Params:
58
+ 2023-10-13 13:04:16,456 - learning_rate: "5e-05"
59
+ 2023-10-13 13:04:16,456 - mini_batch_size: "4"
60
+ 2023-10-13 13:04:16,456 - max_epochs: "10"
61
+ 2023-10-13 13:04:16,456 - shuffle: "True"
62
+ 2023-10-13 13:04:16,456 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-13 13:04:16,456 Plugins:
64
+ 2023-10-13 13:04:16,456 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-13 13:04:16,456 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-13 13:04:16,456 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-13 13:04:16,456 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-13 13:04:16,456 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-13 13:04:16,456 Computation:
70
+ 2023-10-13 13:04:16,456 - compute on device: cuda:0
71
+ 2023-10-13 13:04:16,456 - embedding storage: none
72
+ 2023-10-13 13:04:16,456 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-13 13:04:16,456 Model training base path: "hmbench-hipe2020/de-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3"
74
+ 2023-10-13 13:04:16,456 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-13 13:04:16,456 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-13 13:04:20,590 epoch 1 - iter 89/894 - loss 2.36634984 - time (sec): 4.13 - samples/sec: 2030.01 - lr: 0.000005 - momentum: 0.000000
77
+ 2023-10-13 13:04:24,739 epoch 1 - iter 178/894 - loss 1.43239919 - time (sec): 8.28 - samples/sec: 2080.21 - lr: 0.000010 - momentum: 0.000000
78
+ 2023-10-13 13:04:28,865 epoch 1 - iter 267/894 - loss 1.11148343 - time (sec): 12.41 - samples/sec: 2032.80 - lr: 0.000015 - momentum: 0.000000
79
+ 2023-10-13 13:04:33,154 epoch 1 - iter 356/894 - loss 0.89126147 - time (sec): 16.70 - samples/sec: 2090.54 - lr: 0.000020 - momentum: 0.000000
80
+ 2023-10-13 13:04:37,385 epoch 1 - iter 445/894 - loss 0.76537662 - time (sec): 20.93 - samples/sec: 2086.15 - lr: 0.000025 - momentum: 0.000000
81
+ 2023-10-13 13:04:41,659 epoch 1 - iter 534/894 - loss 0.68100211 - time (sec): 25.20 - samples/sec: 2087.08 - lr: 0.000030 - momentum: 0.000000
82
+ 2023-10-13 13:04:45,922 epoch 1 - iter 623/894 - loss 0.62588240 - time (sec): 29.46 - samples/sec: 2058.75 - lr: 0.000035 - momentum: 0.000000
83
+ 2023-10-13 13:04:50,658 epoch 1 - iter 712/894 - loss 0.58637517 - time (sec): 34.20 - samples/sec: 2010.46 - lr: 0.000040 - momentum: 0.000000
84
+ 2023-10-13 13:04:55,024 epoch 1 - iter 801/894 - loss 0.54465662 - time (sec): 38.57 - samples/sec: 2021.76 - lr: 0.000045 - momentum: 0.000000
85
+ 2023-10-13 13:04:59,165 epoch 1 - iter 890/894 - loss 0.51580101 - time (sec): 42.71 - samples/sec: 2019.42 - lr: 0.000050 - momentum: 0.000000
86
+ 2023-10-13 13:04:59,339 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-13 13:04:59,339 EPOCH 1 done: loss 0.5152 - lr: 0.000050
88
+ 2023-10-13 13:05:04,379 DEV : loss 0.1789691001176834 - f1-score (micro avg) 0.6043
89
+ 2023-10-13 13:05:04,404 saving best model
90
+ 2023-10-13 13:05:04,708 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-13 13:05:08,595 epoch 2 - iter 89/894 - loss 0.19123209 - time (sec): 3.89 - samples/sec: 2118.22 - lr: 0.000049 - momentum: 0.000000
92
+ 2023-10-13 13:05:12,714 epoch 2 - iter 178/894 - loss 0.18653465 - time (sec): 8.00 - samples/sec: 2105.89 - lr: 0.000049 - momentum: 0.000000
93
+ 2023-10-13 13:05:16,707 epoch 2 - iter 267/894 - loss 0.17320392 - time (sec): 12.00 - samples/sec: 2092.69 - lr: 0.000048 - momentum: 0.000000
94
+ 2023-10-13 13:05:20,941 epoch 2 - iter 356/894 - loss 0.16508210 - time (sec): 16.23 - samples/sec: 2076.00 - lr: 0.000048 - momentum: 0.000000
95
+ 2023-10-13 13:05:25,000 epoch 2 - iter 445/894 - loss 0.17103083 - time (sec): 20.29 - samples/sec: 2066.33 - lr: 0.000047 - momentum: 0.000000
96
+ 2023-10-13 13:05:29,170 epoch 2 - iter 534/894 - loss 0.16503055 - time (sec): 24.46 - samples/sec: 2052.01 - lr: 0.000047 - momentum: 0.000000
97
+ 2023-10-13 13:05:33,290 epoch 2 - iter 623/894 - loss 0.16223494 - time (sec): 28.58 - samples/sec: 2070.65 - lr: 0.000046 - momentum: 0.000000
98
+ 2023-10-13 13:05:37,456 epoch 2 - iter 712/894 - loss 0.16356294 - time (sec): 32.75 - samples/sec: 2074.67 - lr: 0.000046 - momentum: 0.000000
99
+ 2023-10-13 13:05:41,548 epoch 2 - iter 801/894 - loss 0.16281990 - time (sec): 36.84 - samples/sec: 2104.60 - lr: 0.000045 - momentum: 0.000000
100
+ 2023-10-13 13:05:45,629 epoch 2 - iter 890/894 - loss 0.16303453 - time (sec): 40.92 - samples/sec: 2107.04 - lr: 0.000044 - momentum: 0.000000
101
+ 2023-10-13 13:05:45,808 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-13 13:05:45,808 EPOCH 2 done: loss 0.1631 - lr: 0.000044
103
+ 2023-10-13 13:05:54,534 DEV : loss 0.137100949883461 - f1-score (micro avg) 0.687
104
+ 2023-10-13 13:05:54,562 saving best model
105
+ 2023-10-13 13:05:55,032 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-13 13:05:59,180 epoch 3 - iter 89/894 - loss 0.09699773 - time (sec): 4.15 - samples/sec: 1976.35 - lr: 0.000044 - momentum: 0.000000
107
+ 2023-10-13 13:06:03,141 epoch 3 - iter 178/894 - loss 0.09307940 - time (sec): 8.11 - samples/sec: 1983.41 - lr: 0.000043 - momentum: 0.000000
108
+ 2023-10-13 13:06:07,398 epoch 3 - iter 267/894 - loss 0.09537427 - time (sec): 12.36 - samples/sec: 2024.87 - lr: 0.000043 - momentum: 0.000000
109
+ 2023-10-13 13:06:11,519 epoch 3 - iter 356/894 - loss 0.10190669 - time (sec): 16.49 - samples/sec: 2029.87 - lr: 0.000042 - momentum: 0.000000
110
+ 2023-10-13 13:06:15,669 epoch 3 - iter 445/894 - loss 0.10548613 - time (sec): 20.64 - samples/sec: 2012.81 - lr: 0.000042 - momentum: 0.000000
111
+ 2023-10-13 13:06:19,846 epoch 3 - iter 534/894 - loss 0.09892079 - time (sec): 24.81 - samples/sec: 2028.38 - lr: 0.000041 - momentum: 0.000000
112
+ 2023-10-13 13:06:24,075 epoch 3 - iter 623/894 - loss 0.09652969 - time (sec): 29.04 - samples/sec: 2022.93 - lr: 0.000041 - momentum: 0.000000
113
+ 2023-10-13 13:06:28,473 epoch 3 - iter 712/894 - loss 0.09893057 - time (sec): 33.44 - samples/sec: 2013.66 - lr: 0.000040 - momentum: 0.000000
114
+ 2023-10-13 13:06:32,779 epoch 3 - iter 801/894 - loss 0.09689469 - time (sec): 37.74 - samples/sec: 2022.65 - lr: 0.000039 - momentum: 0.000000
115
+ 2023-10-13 13:06:37,176 epoch 3 - iter 890/894 - loss 0.09758921 - time (sec): 42.14 - samples/sec: 2044.61 - lr: 0.000039 - momentum: 0.000000
116
+ 2023-10-13 13:06:37,358 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-13 13:06:37,359 EPOCH 3 done: loss 0.0973 - lr: 0.000039
118
+ 2023-10-13 13:06:46,038 DEV : loss 0.1663622260093689 - f1-score (micro avg) 0.7477
119
+ 2023-10-13 13:06:46,074 saving best model
120
+ 2023-10-13 13:06:46,554 ----------------------------------------------------------------------------------------------------
121
+ 2023-10-13 13:06:50,999 epoch 4 - iter 89/894 - loss 0.06861891 - time (sec): 4.44 - samples/sec: 1871.87 - lr: 0.000038 - momentum: 0.000000
122
+ 2023-10-13 13:06:55,159 epoch 4 - iter 178/894 - loss 0.06191697 - time (sec): 8.60 - samples/sec: 1944.64 - lr: 0.000038 - momentum: 0.000000
123
+ 2023-10-13 13:06:59,111 epoch 4 - iter 267/894 - loss 0.05888640 - time (sec): 12.55 - samples/sec: 1998.40 - lr: 0.000037 - momentum: 0.000000
124
+ 2023-10-13 13:07:03,348 epoch 4 - iter 356/894 - loss 0.06116699 - time (sec): 16.79 - samples/sec: 2100.08 - lr: 0.000037 - momentum: 0.000000
125
+ 2023-10-13 13:07:07,307 epoch 4 - iter 445/894 - loss 0.05715439 - time (sec): 20.75 - samples/sec: 2097.09 - lr: 0.000036 - momentum: 0.000000
126
+ 2023-10-13 13:07:11,513 epoch 4 - iter 534/894 - loss 0.06420373 - time (sec): 24.95 - samples/sec: 2090.38 - lr: 0.000036 - momentum: 0.000000
127
+ 2023-10-13 13:07:15,549 epoch 4 - iter 623/894 - loss 0.06272790 - time (sec): 28.99 - samples/sec: 2100.24 - lr: 0.000035 - momentum: 0.000000
128
+ 2023-10-13 13:07:19,635 epoch 4 - iter 712/894 - loss 0.06336972 - time (sec): 33.08 - samples/sec: 2091.46 - lr: 0.000034 - momentum: 0.000000
129
+ 2023-10-13 13:07:23,854 epoch 4 - iter 801/894 - loss 0.06486021 - time (sec): 37.30 - samples/sec: 2100.20 - lr: 0.000034 - momentum: 0.000000
130
+ 2023-10-13 13:07:27,851 epoch 4 - iter 890/894 - loss 0.06614027 - time (sec): 41.29 - samples/sec: 2088.79 - lr: 0.000033 - momentum: 0.000000
131
+ 2023-10-13 13:07:28,035 ----------------------------------------------------------------------------------------------------
132
+ 2023-10-13 13:07:28,035 EPOCH 4 done: loss 0.0661 - lr: 0.000033
133
+ 2023-10-13 13:07:36,744 DEV : loss 0.20190493762493134 - f1-score (micro avg) 0.7572
134
+ 2023-10-13 13:07:36,774 saving best model
135
+ 2023-10-13 13:07:37,188 ----------------------------------------------------------------------------------------------------
136
+ 2023-10-13 13:07:41,561 epoch 5 - iter 89/894 - loss 0.04288380 - time (sec): 4.37 - samples/sec: 1956.39 - lr: 0.000033 - momentum: 0.000000
137
+ 2023-10-13 13:07:45,694 epoch 5 - iter 178/894 - loss 0.03764053 - time (sec): 8.50 - samples/sec: 1968.92 - lr: 0.000032 - momentum: 0.000000
138
+ 2023-10-13 13:07:49,851 epoch 5 - iter 267/894 - loss 0.03485234 - time (sec): 12.66 - samples/sec: 1998.91 - lr: 0.000032 - momentum: 0.000000
139
+ 2023-10-13 13:07:53,898 epoch 5 - iter 356/894 - loss 0.03891244 - time (sec): 16.71 - samples/sec: 1992.95 - lr: 0.000031 - momentum: 0.000000
140
+ 2023-10-13 13:07:58,344 epoch 5 - iter 445/894 - loss 0.04283138 - time (sec): 21.15 - samples/sec: 2013.45 - lr: 0.000031 - momentum: 0.000000
141
+ 2023-10-13 13:08:02,706 epoch 5 - iter 534/894 - loss 0.04395209 - time (sec): 25.51 - samples/sec: 1992.27 - lr: 0.000030 - momentum: 0.000000
142
+ 2023-10-13 13:08:07,000 epoch 5 - iter 623/894 - loss 0.04187519 - time (sec): 29.81 - samples/sec: 1974.53 - lr: 0.000029 - momentum: 0.000000
143
+ 2023-10-13 13:08:11,451 epoch 5 - iter 712/894 - loss 0.04371575 - time (sec): 34.26 - samples/sec: 2013.92 - lr: 0.000029 - momentum: 0.000000
144
+ 2023-10-13 13:08:15,665 epoch 5 - iter 801/894 - loss 0.04232226 - time (sec): 38.47 - samples/sec: 2029.58 - lr: 0.000028 - momentum: 0.000000
145
+ 2023-10-13 13:08:19,803 epoch 5 - iter 890/894 - loss 0.04227247 - time (sec): 42.61 - samples/sec: 2024.39 - lr: 0.000028 - momentum: 0.000000
146
+ 2023-10-13 13:08:19,982 ----------------------------------------------------------------------------------------------------
147
+ 2023-10-13 13:08:19,982 EPOCH 5 done: loss 0.0421 - lr: 0.000028
148
+ 2023-10-13 13:08:28,483 DEV : loss 0.2170535773038864 - f1-score (micro avg) 0.7413
149
+ 2023-10-13 13:08:28,510 ----------------------------------------------------------------------------------------------------
150
+ 2023-10-13 13:08:33,040 epoch 6 - iter 89/894 - loss 0.03105688 - time (sec): 4.53 - samples/sec: 2143.95 - lr: 0.000027 - momentum: 0.000000
151
+ 2023-10-13 13:08:37,152 epoch 6 - iter 178/894 - loss 0.02598800 - time (sec): 8.64 - samples/sec: 2086.86 - lr: 0.000027 - momentum: 0.000000
152
+ 2023-10-13 13:08:41,573 epoch 6 - iter 267/894 - loss 0.02511682 - time (sec): 13.06 - samples/sec: 2059.81 - lr: 0.000026 - momentum: 0.000000
153
+ 2023-10-13 13:08:45,626 epoch 6 - iter 356/894 - loss 0.02507832 - time (sec): 17.11 - samples/sec: 2094.96 - lr: 0.000026 - momentum: 0.000000
154
+ 2023-10-13 13:08:49,625 epoch 6 - iter 445/894 - loss 0.02412952 - time (sec): 21.11 - samples/sec: 2120.88 - lr: 0.000025 - momentum: 0.000000
155
+ 2023-10-13 13:08:54,136 epoch 6 - iter 534/894 - loss 0.02552282 - time (sec): 25.63 - samples/sec: 2060.82 - lr: 0.000024 - momentum: 0.000000
156
+ 2023-10-13 13:08:58,762 epoch 6 - iter 623/894 - loss 0.02567017 - time (sec): 30.25 - samples/sec: 2008.90 - lr: 0.000024 - momentum: 0.000000
157
+ 2023-10-13 13:09:03,242 epoch 6 - iter 712/894 - loss 0.02907428 - time (sec): 34.73 - samples/sec: 1982.97 - lr: 0.000023 - momentum: 0.000000
158
+ 2023-10-13 13:09:07,685 epoch 6 - iter 801/894 - loss 0.03070030 - time (sec): 39.17 - samples/sec: 1983.15 - lr: 0.000023 - momentum: 0.000000
159
+ 2023-10-13 13:09:12,113 epoch 6 - iter 890/894 - loss 0.03031654 - time (sec): 43.60 - samples/sec: 1979.02 - lr: 0.000022 - momentum: 0.000000
160
+ 2023-10-13 13:09:12,288 ----------------------------------------------------------------------------------------------------
161
+ 2023-10-13 13:09:12,289 EPOCH 6 done: loss 0.0306 - lr: 0.000022
162
+ 2023-10-13 13:09:20,941 DEV : loss 0.21267157793045044 - f1-score (micro avg) 0.7541
163
+ 2023-10-13 13:09:20,968 ----------------------------------------------------------------------------------------------------
164
+ 2023-10-13 13:09:25,137 epoch 7 - iter 89/894 - loss 0.02116415 - time (sec): 4.17 - samples/sec: 2238.75 - lr: 0.000022 - momentum: 0.000000
165
+ 2023-10-13 13:09:29,616 epoch 7 - iter 178/894 - loss 0.02415449 - time (sec): 8.65 - samples/sec: 2209.79 - lr: 0.000021 - momentum: 0.000000
166
+ 2023-10-13 13:09:34,148 epoch 7 - iter 267/894 - loss 0.02137786 - time (sec): 13.18 - samples/sec: 2121.64 - lr: 0.000021 - momentum: 0.000000
167
+ 2023-10-13 13:09:38,226 epoch 7 - iter 356/894 - loss 0.02169246 - time (sec): 17.26 - samples/sec: 2162.02 - lr: 0.000020 - momentum: 0.000000
168
+ 2023-10-13 13:09:42,374 epoch 7 - iter 445/894 - loss 0.01995436 - time (sec): 21.40 - samples/sec: 2150.81 - lr: 0.000019 - momentum: 0.000000
169
+ 2023-10-13 13:09:46,490 epoch 7 - iter 534/894 - loss 0.01839450 - time (sec): 25.52 - samples/sec: 2114.02 - lr: 0.000019 - momentum: 0.000000
170
+ 2023-10-13 13:09:50,578 epoch 7 - iter 623/894 - loss 0.01849719 - time (sec): 29.61 - samples/sec: 2079.51 - lr: 0.000018 - momentum: 0.000000
171
+ 2023-10-13 13:09:54,790 epoch 7 - iter 712/894 - loss 0.01855825 - time (sec): 33.82 - samples/sec: 2069.57 - lr: 0.000018 - momentum: 0.000000
172
+ 2023-10-13 13:09:58,936 epoch 7 - iter 801/894 - loss 0.01877532 - time (sec): 37.97 - samples/sec: 2055.33 - lr: 0.000017 - momentum: 0.000000
173
+ 2023-10-13 13:10:02,992 epoch 7 - iter 890/894 - loss 0.02001437 - time (sec): 42.02 - samples/sec: 2049.29 - lr: 0.000017 - momentum: 0.000000
174
+ 2023-10-13 13:10:03,172 ----------------------------------------------------------------------------------------------------
175
+ 2023-10-13 13:10:03,172 EPOCH 7 done: loss 0.0199 - lr: 0.000017
176
+ 2023-10-13 13:10:11,819 DEV : loss 0.24245555698871613 - f1-score (micro avg) 0.7744
177
+ 2023-10-13 13:10:11,846 saving best model
178
+ 2023-10-13 13:10:12,251 ----------------------------------------------------------------------------------------------------
179
+ 2023-10-13 13:10:16,359 epoch 8 - iter 89/894 - loss 0.01603205 - time (sec): 4.10 - samples/sec: 2129.71 - lr: 0.000016 - momentum: 0.000000
180
+ 2023-10-13 13:10:20,346 epoch 8 - iter 178/894 - loss 0.01420650 - time (sec): 8.09 - samples/sec: 2090.79 - lr: 0.000016 - momentum: 0.000000
181
+ 2023-10-13 13:10:24,729 epoch 8 - iter 267/894 - loss 0.01506146 - time (sec): 12.47 - samples/sec: 2168.47 - lr: 0.000015 - momentum: 0.000000
182
+ 2023-10-13 13:10:28,889 epoch 8 - iter 356/894 - loss 0.01227759 - time (sec): 16.63 - samples/sec: 2134.89 - lr: 0.000014 - momentum: 0.000000
183
+ 2023-10-13 13:10:33,378 epoch 8 - iter 445/894 - loss 0.01221774 - time (sec): 21.12 - samples/sec: 2085.76 - lr: 0.000014 - momentum: 0.000000
184
+ 2023-10-13 13:10:37,636 epoch 8 - iter 534/894 - loss 0.01151621 - time (sec): 25.38 - samples/sec: 2087.00 - lr: 0.000013 - momentum: 0.000000
185
+ 2023-10-13 13:10:41,814 epoch 8 - iter 623/894 - loss 0.01137032 - time (sec): 29.56 - samples/sec: 2077.04 - lr: 0.000013 - momentum: 0.000000
186
+ 2023-10-13 13:10:45,868 epoch 8 - iter 712/894 - loss 0.01073686 - time (sec): 33.61 - samples/sec: 2072.07 - lr: 0.000012 - momentum: 0.000000
187
+ 2023-10-13 13:10:50,055 epoch 8 - iter 801/894 - loss 0.01085969 - time (sec): 37.80 - samples/sec: 2076.10 - lr: 0.000012 - momentum: 0.000000
188
+ 2023-10-13 13:10:54,024 epoch 8 - iter 890/894 - loss 0.01108987 - time (sec): 41.77 - samples/sec: 2065.03 - lr: 0.000011 - momentum: 0.000000
189
+ 2023-10-13 13:10:54,198 ----------------------------------------------------------------------------------------------------
190
+ 2023-10-13 13:10:54,198 EPOCH 8 done: loss 0.0112 - lr: 0.000011
191
+ 2023-10-13 13:11:02,846 DEV : loss 0.233389750123024 - f1-score (micro avg) 0.7832
192
+ 2023-10-13 13:11:02,873 saving best model
193
+ 2023-10-13 13:11:03,298 ----------------------------------------------------------------------------------------------------
194
+ 2023-10-13 13:11:07,662 epoch 9 - iter 89/894 - loss 0.00502968 - time (sec): 4.36 - samples/sec: 1888.24 - lr: 0.000011 - momentum: 0.000000
195
+ 2023-10-13 13:11:12,220 epoch 9 - iter 178/894 - loss 0.00345860 - time (sec): 8.92 - samples/sec: 1986.19 - lr: 0.000010 - momentum: 0.000000
196
+ 2023-10-13 13:11:16,359 epoch 9 - iter 267/894 - loss 0.00469124 - time (sec): 13.06 - samples/sec: 2013.22 - lr: 0.000009 - momentum: 0.000000
197
+ 2023-10-13 13:11:20,410 epoch 9 - iter 356/894 - loss 0.00482172 - time (sec): 17.11 - samples/sec: 2031.12 - lr: 0.000009 - momentum: 0.000000
198
+ 2023-10-13 13:11:24,331 epoch 9 - iter 445/894 - loss 0.00561556 - time (sec): 21.03 - samples/sec: 2057.73 - lr: 0.000008 - momentum: 0.000000
199
+ 2023-10-13 13:11:28,315 epoch 9 - iter 534/894 - loss 0.00515886 - time (sec): 25.01 - samples/sec: 2060.32 - lr: 0.000008 - momentum: 0.000000
200
+ 2023-10-13 13:11:32,522 epoch 9 - iter 623/894 - loss 0.00534634 - time (sec): 29.22 - samples/sec: 2040.68 - lr: 0.000007 - momentum: 0.000000
201
+ 2023-10-13 13:11:36,786 epoch 9 - iter 712/894 - loss 0.00559174 - time (sec): 33.49 - samples/sec: 2039.71 - lr: 0.000007 - momentum: 0.000000
202
+ 2023-10-13 13:11:41,058 epoch 9 - iter 801/894 - loss 0.00552963 - time (sec): 37.76 - samples/sec: 2064.38 - lr: 0.000006 - momentum: 0.000000
203
+ 2023-10-13 13:11:45,102 epoch 9 - iter 890/894 - loss 0.00582293 - time (sec): 41.80 - samples/sec: 2061.36 - lr: 0.000006 - momentum: 0.000000
204
+ 2023-10-13 13:11:45,305 ----------------------------------------------------------------------------------------------------
205
+ 2023-10-13 13:11:45,306 EPOCH 9 done: loss 0.0061 - lr: 0.000006
206
+ 2023-10-13 13:11:54,109 DEV : loss 0.2255241870880127 - f1-score (micro avg) 0.779
207
+ 2023-10-13 13:11:54,139 ----------------------------------------------------------------------------------------------------
208
+ 2023-10-13 13:11:58,538 epoch 10 - iter 89/894 - loss 0.00368147 - time (sec): 4.40 - samples/sec: 2080.65 - lr: 0.000005 - momentum: 0.000000
209
+ 2023-10-13 13:12:02,657 epoch 10 - iter 178/894 - loss 0.00436238 - time (sec): 8.52 - samples/sec: 2048.21 - lr: 0.000004 - momentum: 0.000000
210
+ 2023-10-13 13:12:06,746 epoch 10 - iter 267/894 - loss 0.00379776 - time (sec): 12.61 - samples/sec: 2081.50 - lr: 0.000004 - momentum: 0.000000
211
+ 2023-10-13 13:12:10,863 epoch 10 - iter 356/894 - loss 0.00400577 - time (sec): 16.72 - samples/sec: 2090.41 - lr: 0.000003 - momentum: 0.000000
212
+ 2023-10-13 13:12:15,238 epoch 10 - iter 445/894 - loss 0.00371452 - time (sec): 21.10 - samples/sec: 2040.31 - lr: 0.000003 - momentum: 0.000000
213
+ 2023-10-13 13:12:19,338 epoch 10 - iter 534/894 - loss 0.00453871 - time (sec): 25.20 - samples/sec: 2035.60 - lr: 0.000002 - momentum: 0.000000
214
+ 2023-10-13 13:12:23,456 epoch 10 - iter 623/894 - loss 0.00446673 - time (sec): 29.32 - samples/sec: 2057.39 - lr: 0.000002 - momentum: 0.000000
215
+ 2023-10-13 13:12:27,853 epoch 10 - iter 712/894 - loss 0.00466561 - time (sec): 33.71 - samples/sec: 2088.01 - lr: 0.000001 - momentum: 0.000000
216
+ 2023-10-13 13:12:31,952 epoch 10 - iter 801/894 - loss 0.00442535 - time (sec): 37.81 - samples/sec: 2069.29 - lr: 0.000001 - momentum: 0.000000
217
+ 2023-10-13 13:12:36,079 epoch 10 - iter 890/894 - loss 0.00430193 - time (sec): 41.94 - samples/sec: 2054.07 - lr: 0.000000 - momentum: 0.000000
218
+ 2023-10-13 13:12:36,255 ----------------------------------------------------------------------------------------------------
219
+ 2023-10-13 13:12:36,256 EPOCH 10 done: loss 0.0043 - lr: 0.000000
220
+ 2023-10-13 13:12:44,962 DEV : loss 0.23823896050453186 - f1-score (micro avg) 0.7834
221
+ 2023-10-13 13:12:44,989 saving best model
222
+ 2023-10-13 13:12:45,734 ----------------------------------------------------------------------------------------------------
223
+ 2023-10-13 13:12:45,736 Loading model from best epoch ...
224
+ 2023-10-13 13:12:47,180 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
225
+ 2023-10-13 13:12:51,907
226
+ Results:
227
+ - F-score (micro) 0.7438
228
+ - F-score (macro) 0.6448
229
+ - Accuracy 0.6112
230
+
231
+ By class:
232
+ precision recall f1-score support
233
+
234
+ loc 0.8333 0.8473 0.8403 596
235
+ pers 0.6614 0.7568 0.7059 333
236
+ org 0.6118 0.3939 0.4793 132
237
+ prod 0.6222 0.4242 0.5045 66
238
+ time 0.6939 0.6939 0.6939 49
239
+
240
+ micro avg 0.7470 0.7406 0.7438 1176
241
+ macro avg 0.6845 0.6232 0.6448 1176
242
+ weighted avg 0.7421 0.7406 0.7367 1176
243
+
244
+ 2023-10-13 13:12:51,908 ----------------------------------------------------------------------------------------------------