Hugging Face's logo

stefan-it
/
hmbench-hipe2020-de-hmbert-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4

Token Classification
Flair
PyTorch
German
sequence-tagger-model
Model card Files Community
1
stefan-it commited on
Commit
0ad93eb
•
1 Parent(s): 168bca2

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 +245 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:be5a4c6411a563e22b335d3b2e52b8959375ab1a58416fd031bbe7584238ddc4
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 13:27:42 0.0000 0.6313 0.1942 0.6323 0.5903 0.6106 0.4489
3
+ 2 13:28:35 0.0000 0.1574 0.1417 0.6334 0.7349 0.6804 0.5387
4
+ 3 13:29:28 0.0000 0.0872 0.1514 0.6909 0.7639 0.7256 0.5857
5
+ 4 13:30:23 0.0000 0.0589 0.1777 0.7235 0.7694 0.7457 0.6154
6
+ 5 13:31:15 0.0000 0.0397 0.2007 0.7675 0.7615 0.7645 0.6358
7
+ 6 13:32:06 0.0000 0.0264 0.2041 0.7447 0.7936 0.7684 0.6396
8
+ 7 13:32:58 0.0000 0.0174 0.2201 0.7688 0.7905 0.7795 0.6544
9
+ 8 13:33:50 0.0000 0.0122 0.2316 0.7654 0.7756 0.7705 0.6421
10
+ 9 13:34:41 0.0000 0.0076 0.2367 0.7784 0.7936 0.7859 0.6625
11
+ 10 13:35:33 0.0000 0.0053 0.2382 0.7719 0.7991 0.7852 0.6619
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,245 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-13 13:26:55,352 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-13 13:26:55,353 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 13:26:55,353 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-13 13:26:55,353 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 13:26:55,353 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-13 13:26:55,353 Train: 3575 sentences
55
+ 2023-10-13 13:26:55,353 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-13 13:26:55,353 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-13 13:26:55,353 Training Params:
58
+ 2023-10-13 13:26:55,353 - learning_rate: "3e-05"
59
+ 2023-10-13 13:26:55,353 - mini_batch_size: "4"
60
+ 2023-10-13 13:26:55,353 - max_epochs: "10"
61
+ 2023-10-13 13:26:55,353 - shuffle: "True"
62
+ 2023-10-13 13:26:55,353 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-13 13:26:55,353 Plugins:
64
+ 2023-10-13 13:26:55,353 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-13 13:26:55,353 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-13 13:26:55,353 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-13 13:26:55,354 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-13 13:26:55,354 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-13 13:26:55,354 Computation:
70
+ 2023-10-13 13:26:55,354 - compute on device: cuda:0
71
+ 2023-10-13 13:26:55,354 - embedding storage: none
72
+ 2023-10-13 13:26:55,354 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-13 13:26:55,354 Model training base path: "hmbench-hipe2020/de-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4"
74
+ 2023-10-13 13:26:55,354 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-13 13:26:55,354 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-13 13:26:59,604 epoch 1 - iter 89/894 - loss 2.95558386 - time (sec): 4.25 - samples/sec: 2080.59 - lr: 0.000003 - momentum: 0.000000
77
+ 2023-10-13 13:27:03,798 epoch 1 - iter 178/894 - loss 1.86962681 - time (sec): 8.44 - samples/sec: 2140.01 - lr: 0.000006 - momentum: 0.000000
78
+ 2023-10-13 13:27:07,907 epoch 1 - iter 267/894 - loss 1.43922230 - time (sec): 12.55 - samples/sec: 2070.79 - lr: 0.000009 - momentum: 0.000000
79
+ 2023-10-13 13:27:12,189 epoch 1 - iter 356/894 - loss 1.16332138 - time (sec): 16.83 - samples/sec: 2074.45 - lr: 0.000012 - momentum: 0.000000
80
+ 2023-10-13 13:27:16,305 epoch 1 - iter 445/894 - loss 0.99639681 - time (sec): 20.95 - samples/sec: 2063.30 - lr: 0.000015 - momentum: 0.000000
81
+ 2023-10-13 13:27:20,409 epoch 1 - iter 534/894 - loss 0.88501649 - time (sec): 25.05 - samples/sec: 2062.28 - lr: 0.000018 - momentum: 0.000000
82
+ 2023-10-13 13:27:24,535 epoch 1 - iter 623/894 - loss 0.80459753 - time (sec): 29.18 - samples/sec: 2051.80 - lr: 0.000021 - momentum: 0.000000
83
+ 2023-10-13 13:27:28,784 epoch 1 - iter 712/894 - loss 0.73830399 - time (sec): 33.43 - samples/sec: 2048.10 - lr: 0.000024 - momentum: 0.000000
84
+ 2023-10-13 13:27:32,761 epoch 1 - iter 801/894 - loss 0.68135095 - time (sec): 37.41 - samples/sec: 2044.69 - lr: 0.000027 - momentum: 0.000000
85
+ 2023-10-13 13:27:37,145 epoch 1 - iter 890/894 - loss 0.63356747 - time (sec): 41.79 - samples/sec: 2062.28 - lr: 0.000030 - momentum: 0.000000
86
+ 2023-10-13 13:27:37,323 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-13 13:27:37,324 EPOCH 1 done: loss 0.6313 - lr: 0.000030
88
+ 2023-10-13 13:27:42,761 DEV : loss 0.1941666156053543 - f1-score (micro avg) 0.6106
89
+ 2023-10-13 13:27:42,790 saving best model
90
+ 2023-10-13 13:27:43,110 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-13 13:27:47,503 epoch 2 - iter 89/894 - loss 0.18794288 - time (sec): 4.39 - samples/sec: 2054.48 - lr: 0.000030 - momentum: 0.000000
92
+ 2023-10-13 13:27:52,262 epoch 2 - iter 178/894 - loss 0.17923060 - time (sec): 9.15 - samples/sec: 2032.58 - lr: 0.000029 - momentum: 0.000000
93
+ 2023-10-13 13:27:56,703 epoch 2 - iter 267/894 - loss 0.16858546 - time (sec): 13.59 - samples/sec: 1950.55 - lr: 0.000029 - momentum: 0.000000
94
+ 2023-10-13 13:28:00,953 epoch 2 - iter 356/894 - loss 0.17427875 - time (sec): 17.84 - samples/sec: 1952.36 - lr: 0.000029 - momentum: 0.000000
95
+ 2023-10-13 13:28:05,270 epoch 2 - iter 445/894 - loss 0.17144978 - time (sec): 22.16 - samples/sec: 1968.14 - lr: 0.000028 - momentum: 0.000000
96
+ 2023-10-13 13:28:09,579 epoch 2 - iter 534/894 - loss 0.16359629 - time (sec): 26.47 - samples/sec: 1976.51 - lr: 0.000028 - momentum: 0.000000
97
+ 2023-10-13 13:28:13,700 epoch 2 - iter 623/894 - loss 0.16276876 - time (sec): 30.59 - samples/sec: 1976.29 - lr: 0.000028 - momentum: 0.000000
98
+ 2023-10-13 13:28:17,828 epoch 2 - iter 712/894 - loss 0.16121921 - time (sec): 34.72 - samples/sec: 1977.28 - lr: 0.000027 - momentum: 0.000000
99
+ 2023-10-13 13:28:22,050 epoch 2 - iter 801/894 - loss 0.15844776 - time (sec): 38.94 - samples/sec: 1990.54 - lr: 0.000027 - momentum: 0.000000
100
+ 2023-10-13 13:28:26,052 epoch 2 - iter 890/894 - loss 0.15768333 - time (sec): 42.94 - samples/sec: 2007.70 - lr: 0.000027 - momentum: 0.000000
101
+ 2023-10-13 13:28:26,230 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-13 13:28:26,230 EPOCH 2 done: loss 0.1574 - lr: 0.000027
103
+ 2023-10-13 13:28:35,343 DEV : loss 0.1417360156774521 - f1-score (micro avg) 0.6804
104
+ 2023-10-13 13:28:35,382 saving best model
105
+ 2023-10-13 13:28:35,903 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-13 13:28:40,362 epoch 3 - iter 89/894 - loss 0.10870491 - time (sec): 4.46 - samples/sec: 1945.07 - lr: 0.000026 - momentum: 0.000000
107
+ 2023-10-13 13:28:44,659 epoch 3 - iter 178/894 - loss 0.09511837 - time (sec): 8.75 - samples/sec: 2060.06 - lr: 0.000026 - momentum: 0.000000
108
+ 2023-10-13 13:28:48,835 epoch 3 - iter 267/894 - loss 0.08476480 - time (sec): 12.93 - samples/sec: 2045.85 - lr: 0.000026 - momentum: 0.000000
109
+ 2023-10-13 13:28:52,984 epoch 3 - iter 356/894 - loss 0.08905987 - time (sec): 17.08 - samples/sec: 2036.20 - lr: 0.000025 - momentum: 0.000000
110
+ 2023-10-13 13:28:57,234 epoch 3 - iter 445/894 - loss 0.08587345 - time (sec): 21.33 - samples/sec: 1995.76 - lr: 0.000025 - momentum: 0.000000
111
+ 2023-10-13 13:29:01,699 epoch 3 - iter 534/894 - loss 0.08587494 - time (sec): 25.79 - samples/sec: 1983.40 - lr: 0.000025 - momentum: 0.000000
112
+ 2023-10-13 13:29:06,110 epoch 3 - iter 623/894 - loss 0.08738808 - time (sec): 30.20 - samples/sec: 1965.13 - lr: 0.000024 - momentum: 0.000000
113
+ 2023-10-13 13:29:10,497 epoch 3 - iter 712/894 - loss 0.08364026 - time (sec): 34.59 - samples/sec: 1980.18 - lr: 0.000024 - momentum: 0.000000
114
+ 2023-10-13 13:29:14,555 epoch 3 - iter 801/894 - loss 0.08812239 - time (sec): 38.65 - samples/sec: 1985.64 - lr: 0.000024 - momentum: 0.000000
115
+ 2023-10-13 13:29:18,956 epoch 3 - iter 890/894 - loss 0.08729426 - time (sec): 43.05 - samples/sec: 2001.73 - lr: 0.000023 - momentum: 0.000000
116
+ 2023-10-13 13:29:19,149 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-13 13:29:19,149 EPOCH 3 done: loss 0.0872 - lr: 0.000023
118
+ 2023-10-13 13:29:28,104 DEV : loss 0.1514357179403305 - f1-score (micro avg) 0.7256
119
+ 2023-10-13 13:29:28,139 saving best model
120
+ 2023-10-13 13:29:28,599 ----------------------------------------------------------------------------------------------------
121
+ 2023-10-13 13:29:33,101 epoch 4 - iter 89/894 - loss 0.05364078 - time (sec): 4.50 - samples/sec: 1918.44 - lr: 0.000023 - momentum: 0.000000
122
+ 2023-10-13 13:29:37,515 epoch 4 - iter 178/894 - loss 0.05117238 - time (sec): 8.91 - samples/sec: 1863.71 - lr: 0.000023 - momentum: 0.000000
123
+ 2023-10-13 13:29:42,069 epoch 4 - iter 267/894 - loss 0.05442001 - time (sec): 13.46 - samples/sec: 1876.18 - lr: 0.000022 - momentum: 0.000000
124
+ 2023-10-13 13:29:46,745 epoch 4 - iter 356/894 - loss 0.04904272 - time (sec): 18.14 - samples/sec: 1878.12 - lr: 0.000022 - momentum: 0.000000
125
+ 2023-10-13 13:29:51,714 epoch 4 - iter 445/894 - loss 0.05288936 - time (sec): 23.11 - samples/sec: 1892.72 - lr: 0.000022 - momentum: 0.000000
126
+ 2023-10-13 13:29:56,451 epoch 4 - iter 534/894 - loss 0.05337299 - time (sec): 27.84 - samples/sec: 1885.36 - lr: 0.000021 - momentum: 0.000000
127
+ 2023-10-13 13:30:01,001 epoch 4 - iter 623/894 - loss 0.05444078 - time (sec): 32.40 - samples/sec: 1864.41 - lr: 0.000021 - momentum: 0.000000
128
+ 2023-10-13 13:30:05,689 epoch 4 - iter 712/894 - loss 0.05523828 - time (sec): 37.08 - samples/sec: 1872.35 - lr: 0.000021 - momentum: 0.000000
129
+ 2023-10-13 13:30:10,379 epoch 4 - iter 801/894 - loss 0.05634012 - time (sec): 41.77 - samples/sec: 1870.31 - lr: 0.000020 - momentum: 0.000000
130
+ 2023-10-13 13:30:14,915 epoch 4 - iter 890/894 - loss 0.05893335 - time (sec): 46.31 - samples/sec: 1860.70 - lr: 0.000020 - momentum: 0.000000
131
+ 2023-10-13 13:30:15,109 ----------------------------------------------------------------------------------------------------
132
+ 2023-10-13 13:30:15,109 EPOCH 4 done: loss 0.0589 - lr: 0.000020
133
+ 2023-10-13 13:30:23,755 DEV : loss 0.17774192988872528 - f1-score (micro avg) 0.7457
134
+ 2023-10-13 13:30:23,789 saving best model
135
+ 2023-10-13 13:30:24,260 ----------------------------------------------------------------------------------------------------
136
+ 2023-10-13 13:30:28,888 epoch 5 - iter 89/894 - loss 0.04958656 - time (sec): 4.63 - samples/sec: 1958.63 - lr: 0.000020 - momentum: 0.000000
137
+ 2023-10-13 13:30:32,984 epoch 5 - iter 178/894 - loss 0.04439640 - time (sec): 8.72 - samples/sec: 1995.20 - lr: 0.000019 - momentum: 0.000000
138
+ 2023-10-13 13:30:37,156 epoch 5 - iter 267/894 - loss 0.04404992 - time (sec): 12.89 - samples/sec: 2038.74 - lr: 0.000019 - momentum: 0.000000
139
+ 2023-10-13 13:30:41,256 epoch 5 - iter 356/894 - loss 0.04148328 - time (sec): 16.99 - samples/sec: 2062.03 - lr: 0.000019 - momentum: 0.000000
140
+ 2023-10-13 13:30:45,339 epoch 5 - iter 445/894 - loss 0.04024971 - time (sec): 21.08 - samples/sec: 2053.98 - lr: 0.000018 - momentum: 0.000000
141
+ 2023-10-13 13:30:49,695 epoch 5 - iter 534/894 - loss 0.04182158 - time (sec): 25.43 - samples/sec: 2035.14 - lr: 0.000018 - momentum: 0.000000
142
+ 2023-10-13 13:30:54,179 epoch 5 - iter 623/894 - loss 0.03942623 - time (sec): 29.92 - samples/sec: 2047.61 - lr: 0.000018 - momentum: 0.000000
143
+ 2023-10-13 13:30:58,471 epoch 5 - iter 712/894 - loss 0.04026267 - time (sec): 34.21 - samples/sec: 2025.62 - lr: 0.000017 - momentum: 0.000000
144
+ 2023-10-13 13:31:02,647 epoch 5 - iter 801/894 - loss 0.03995825 - time (sec): 38.38 - samples/sec: 2024.13 - lr: 0.000017 - momentum: 0.000000
145
+ 2023-10-13 13:31:06,816 epoch 5 - iter 890/894 - loss 0.03960817 - time (sec): 42.55 - samples/sec: 2026.89 - lr: 0.000017 - momentum: 0.000000
146
+ 2023-10-13 13:31:07,001 ----------------------------------------------------------------------------------------------------
147
+ 2023-10-13 13:31:07,002 EPOCH 5 done: loss 0.0397 - lr: 0.000017
148
+ 2023-10-13 13:31:15,787 DEV : loss 0.2007289081811905 - f1-score (micro avg) 0.7645
149
+ 2023-10-13 13:31:15,820 saving best model
150
+ 2023-10-13 13:31:16,295 ----------------------------------------------------------------------------------------------------
151
+ 2023-10-13 13:31:20,481 epoch 6 - iter 89/894 - loss 0.02040547 - time (sec): 4.18 - samples/sec: 2093.62 - lr: 0.000016 - momentum: 0.000000
152
+ 2023-10-13 13:31:24,680 epoch 6 - iter 178/894 - loss 0.01964149 - time (sec): 8.38 - samples/sec: 2121.16 - lr: 0.000016 - momentum: 0.000000
153
+ 2023-10-13 13:31:28,716 epoch 6 - iter 267/894 - loss 0.02088570 - time (sec): 12.42 - samples/sec: 2112.13 - lr: 0.000016 - momentum: 0.000000
154
+ 2023-10-13 13:31:33,092 epoch 6 - iter 356/894 - loss 0.02324113 - time (sec): 16.79 - samples/sec: 2156.66 - lr: 0.000015 - momentum: 0.000000
155
+ 2023-10-13 13:31:37,174 epoch 6 - iter 445/894 - loss 0.02270645 - time (sec): 20.88 - samples/sec: 2094.45 - lr: 0.000015 - momentum: 0.000000
156
+ 2023-10-13 13:31:41,187 epoch 6 - iter 534/894 - loss 0.02337935 - time (sec): 24.89 - samples/sec: 2090.81 - lr: 0.000015 - momentum: 0.000000
157
+ 2023-10-13 13:31:45,296 epoch 6 - iter 623/894 - loss 0.02406282 - time (sec): 29.00 - samples/sec: 2087.74 - lr: 0.000014 - momentum: 0.000000
158
+ 2023-10-13 13:31:49,382 epoch 6 - iter 712/894 - loss 0.02361249 - time (sec): 33.08 - samples/sec: 2074.79 - lr: 0.000014 - momentum: 0.000000
159
+ 2023-10-13 13:31:53,449 epoch 6 - iter 801/894 - loss 0.02432036 - time (sec): 37.15 - samples/sec: 2088.20 - lr: 0.000014 - momentum: 0.000000
160
+ 2023-10-13 13:31:57,438 epoch 6 - iter 890/894 - loss 0.02630157 - time (sec): 41.14 - samples/sec: 2095.76 - lr: 0.000013 - momentum: 0.000000
161
+ 2023-10-13 13:31:57,617 ----------------------------------------------------------------------------------------------------
162
+ 2023-10-13 13:31:57,617 EPOCH 6 done: loss 0.0264 - lr: 0.000013
163
+ 2023-10-13 13:32:06,141 DEV : loss 0.20405448973178864 - f1-score (micro avg) 0.7684
164
+ 2023-10-13 13:32:06,171 saving best model
165
+ 2023-10-13 13:32:06,604 ----------------------------------------------------------------------------------------------------
166
+ 2023-10-13 13:32:11,208 epoch 7 - iter 89/894 - loss 0.01619462 - time (sec): 4.60 - samples/sec: 2178.62 - lr: 0.000013 - momentum: 0.000000
167
+ 2023-10-13 13:32:15,438 epoch 7 - iter 178/894 - loss 0.01588258 - time (sec): 8.83 - samples/sec: 2050.83 - lr: 0.000013 - momentum: 0.000000
168
+ 2023-10-13 13:32:19,895 epoch 7 - iter 267/894 - loss 0.01351689 - time (sec): 13.28 - samples/sec: 2029.35 - lr: 0.000012 - momentum: 0.000000
169
+ 2023-10-13 13:32:24,203 epoch 7 - iter 356/894 - loss 0.01461563 - time (sec): 17.59 - samples/sec: 2039.65 - lr: 0.000012 - momentum: 0.000000
170
+ 2023-10-13 13:32:28,462 epoch 7 - iter 445/894 - loss 0.01677621 - time (sec): 21.85 - samples/sec: 2033.09 - lr: 0.000012 - momentum: 0.000000
171
+ 2023-10-13 13:32:32,582 epoch 7 - iter 534/894 - loss 0.01693157 - time (sec): 25.97 - samples/sec: 2006.28 - lr: 0.000011 - momentum: 0.000000
172
+ 2023-10-13 13:32:36,783 epoch 7 - iter 623/894 - loss 0.01777874 - time (sec): 30.17 - samples/sec: 2017.12 - lr: 0.000011 - momentum: 0.000000
173
+ 2023-10-13 13:32:40,885 epoch 7 - iter 712/894 - loss 0.01692073 - time (sec): 34.27 - samples/sec: 2018.10 - lr: 0.000011 - momentum: 0.000000
174
+ 2023-10-13 13:32:44,949 epoch 7 - iter 801/894 - loss 0.01780651 - time (sec): 38.34 - samples/sec: 2021.61 - lr: 0.000010 - momentum: 0.000000
175
+ 2023-10-13 13:32:48,987 epoch 7 - iter 890/894 - loss 0.01723398 - time (sec): 42.38 - samples/sec: 2034.07 - lr: 0.000010 - momentum: 0.000000
176
+ 2023-10-13 13:32:49,163 ----------------------------------------------------------------------------------------------------
177
+ 2023-10-13 13:32:49,163 EPOCH 7 done: loss 0.0174 - lr: 0.000010
178
+ 2023-10-13 13:32:58,154 DEV : loss 0.22012537717819214 - f1-score (micro avg) 0.7795
179
+ 2023-10-13 13:32:58,194 saving best model
180
+ 2023-10-13 13:32:58,701 ----------------------------------------------------------------------------------------------------
181
+ 2023-10-13 13:33:03,051 epoch 8 - iter 89/894 - loss 0.00832136 - time (sec): 4.34 - samples/sec: 2013.98 - lr: 0.000010 - momentum: 0.000000
182
+ 2023-10-13 13:33:07,564 epoch 8 - iter 178/894 - loss 0.00808894 - time (sec): 8.86 - samples/sec: 1950.24 - lr: 0.000009 - momentum: 0.000000
183
+ 2023-10-13 13:33:11,637 epoch 8 - iter 267/894 - loss 0.01090431 - time (sec): 12.93 - samples/sec: 1983.97 - lr: 0.000009 - momentum: 0.000000
184
+ 2023-10-13 13:33:15,841 epoch 8 - iter 356/894 - loss 0.01087231 - time (sec): 17.13 - samples/sec: 1976.09 - lr: 0.000009 - momentum: 0.000000
185
+ 2023-10-13 13:33:20,134 epoch 8 - iter 445/894 - loss 0.01052967 - time (sec): 21.43 - samples/sec: 1971.66 - lr: 0.000008 - momentum: 0.000000
186
+ 2023-10-13 13:33:24,383 epoch 8 - iter 534/894 - loss 0.01078439 - time (sec): 25.68 - samples/sec: 1970.62 - lr: 0.000008 - momentum: 0.000000
187
+ 2023-10-13 13:33:28,522 epoch 8 - iter 623/894 - loss 0.01109453 - time (sec): 29.82 - samples/sec: 1983.29 - lr: 0.000008 - momentum: 0.000000
188
+ 2023-10-13 13:33:32,944 epoch 8 - iter 712/894 - loss 0.01240084 - time (sec): 34.24 - samples/sec: 1995.51 - lr: 0.000007 - momentum: 0.000000
189
+ 2023-10-13 13:33:37,178 epoch 8 - iter 801/894 - loss 0.01237060 - time (sec): 38.47 - samples/sec: 2016.12 - lr: 0.000007 - momentum: 0.000000
190
+ 2023-10-13 13:33:41,265 epoch 8 - iter 890/894 - loss 0.01219660 - time (sec): 42.56 - samples/sec: 2027.23 - lr: 0.000007 - momentum: 0.000000
191
+ 2023-10-13 13:33:41,440 ----------------------------------------------------------------------------------------------------
192
+ 2023-10-13 13:33:41,440 EPOCH 8 done: loss 0.0122 - lr: 0.000007
193
+ 2023-10-13 13:33:50,357 DEV : loss 0.2315651774406433 - f1-score (micro avg) 0.7705
194
+ 2023-10-13 13:33:50,387 ----------------------------------------------------------------------------------------------------
195
+ 2023-10-13 13:33:54,612 epoch 9 - iter 89/894 - loss 0.00348902 - time (sec): 4.22 - samples/sec: 1985.17 - lr: 0.000006 - momentum: 0.000000
196
+ 2023-10-13 13:33:58,941 epoch 9 - iter 178/894 - loss 0.00486255 - time (sec): 8.55 - samples/sec: 2008.41 - lr: 0.000006 - momentum: 0.000000
197
+ 2023-10-13 13:34:03,159 epoch 9 - iter 267/894 - loss 0.00642317 - time (sec): 12.77 - samples/sec: 1974.93 - lr: 0.000006 - momentum: 0.000000
198
+ 2023-10-13 13:34:07,390 epoch 9 - iter 356/894 - loss 0.01103944 - time (sec): 17.00 - samples/sec: 1971.80 - lr: 0.000005 - momentum: 0.000000
199
+ 2023-10-13 13:34:11,668 epoch 9 - iter 445/894 - loss 0.00897085 - time (sec): 21.28 - samples/sec: 1999.42 - lr: 0.000005 - momentum: 0.000000
200
+ 2023-10-13 13:34:15,923 epoch 9 - iter 534/894 - loss 0.00808237 - time (sec): 25.53 - samples/sec: 1998.01 - lr: 0.000005 - momentum: 0.000000
201
+ 2023-10-13 13:34:20,374 epoch 9 - iter 623/894 - loss 0.00817370 - time (sec): 29.99 - samples/sec: 2006.05 - lr: 0.000004 - momentum: 0.000000
202
+ 2023-10-13 13:34:24,751 epoch 9 - iter 712/894 - loss 0.00803605 - time (sec): 34.36 - samples/sec: 2035.31 - lr: 0.000004 - momentum: 0.000000
203
+ 2023-10-13 13:34:28,751 epoch 9 - iter 801/894 - loss 0.00806848 - time (sec): 38.36 - samples/sec: 2035.57 - lr: 0.000004 - momentum: 0.000000
204
+ 2023-10-13 13:34:32,862 epoch 9 - iter 890/894 - loss 0.00765097 - time (sec): 42.47 - samples/sec: 2032.01 - lr: 0.000003 - momentum: 0.000000
205
+ 2023-10-13 13:34:33,037 ----------------------------------------------------------------------------------------------------
206
+ 2023-10-13 13:34:33,038 EPOCH 9 done: loss 0.0076 - lr: 0.000003
207
+ 2023-10-13 13:34:41,714 DEV : loss 0.23668904602527618 - f1-score (micro avg) 0.7859
208
+ 2023-10-13 13:34:41,747 saving best model
209
+ 2023-10-13 13:34:42,200 ----------------------------------------------------------------------------------------------------
210
+ 2023-10-13 13:34:46,536 epoch 10 - iter 89/894 - loss 0.00697084 - time (sec): 4.33 - samples/sec: 2289.44 - lr: 0.000003 - momentum: 0.000000
211
+ 2023-10-13 13:34:50,883 epoch 10 - iter 178/894 - loss 0.00707544 - time (sec): 8.68 - samples/sec: 2158.02 - lr: 0.000003 - momentum: 0.000000
212
+ 2023-10-13 13:34:55,094 epoch 10 - iter 267/894 - loss 0.00741832 - time (sec): 12.89 - samples/sec: 2097.03 - lr: 0.000002 - momentum: 0.000000
213
+ 2023-10-13 13:34:59,329 epoch 10 - iter 356/894 - loss 0.00805166 - time (sec): 17.13 - samples/sec: 2052.37 - lr: 0.000002 - momentum: 0.000000
214
+ 2023-10-13 13:35:03,696 epoch 10 - iter 445/894 - loss 0.00661446 - time (sec): 21.49 - samples/sec: 2039.46 - lr: 0.000002 - momentum: 0.000000
215
+ 2023-10-13 13:35:07,973 epoch 10 - iter 534/894 - loss 0.00680689 - time (sec): 25.77 - samples/sec: 2014.68 - lr: 0.000001 - momentum: 0.000000
216
+ 2023-10-13 13:35:12,132 epoch 10 - iter 623/894 - loss 0.00641756 - time (sec): 29.93 - samples/sec: 2009.15 - lr: 0.000001 - momentum: 0.000000
217
+ 2023-10-13 13:35:16,236 epoch 10 - iter 712/894 - loss 0.00594203 - time (sec): 34.03 - samples/sec: 2025.05 - lr: 0.000001 - momentum: 0.000000
218
+ 2023-10-13 13:35:20,200 epoch 10 - iter 801/894 - loss 0.00534108 - time (sec): 38.00 - samples/sec: 2040.65 - lr: 0.000000 - momentum: 0.000000
219
+ 2023-10-13 13:35:24,171 epoch 10 - iter 890/894 - loss 0.00532176 - time (sec): 41.97 - samples/sec: 2054.23 - lr: 0.000000 - momentum: 0.000000
220
+ 2023-10-13 13:35:24,347 ----------------------------------------------------------------------------------------------------
221
+ 2023-10-13 13:35:24,348 EPOCH 10 done: loss 0.0053 - lr: 0.000000
222
+ 2023-10-13 13:35:33,250 DEV : loss 0.23821337521076202 - f1-score (micro avg) 0.7852
223
+ 2023-10-13 13:35:33,648 ----------------------------------------------------------------------------------------------------
224
+ 2023-10-13 13:35:33,649 Loading model from best epoch ...
225
+ 2023-10-13 13:35:35,424 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
226
+ 2023-10-13 13:35:40,059
227
+ Results:
228
+ - F-score (micro) 0.7491
229
+ - F-score (macro) 0.6783
230
+ - Accuracy 0.6188
231
+
232
+ By class:
233
+ precision recall f1-score support
234
+
235
+ loc 0.8225 0.8473 0.8347 596
236
+ pers 0.6649 0.7447 0.7025 333
237
+ org 0.6018 0.5152 0.5551 132
238
+ prod 0.6346 0.5000 0.5593 66
239
+ time 0.7255 0.7551 0.7400 49
240
+
241
+ micro avg 0.7406 0.7577 0.7491 1176
242
+ macro avg 0.6898 0.6725 0.6783 1176
243
+ weighted avg 0.7385 0.7577 0.7465 1176
244
+
245
+ 2023-10-13 13:35:40,059 ----------------------------------------------------------------------------------------------------