Hugging Face's logo

stefan-it
/
hmbench-icdar-nl-hmbert-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4

Token Classification
Flair
PyTorch
Dutch
sequence-tagger-model
Model card Files Community
1
stefan-it commited on
Commit
464d828
•
1 Parent(s): 6fee9a2

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 +239 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ff4d256e605b3ffa3b4e4a48002c7ef9c0adf88a0e33d1abd6ce69a9d153741a
3
+ size 443311111
dev.tsv ADDED
The diff for this file is too large to render. See raw diff
 
loss.tsv ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ EPOCH TIMESTAMP LEARNING_RATE TRAIN_LOSS DEV_LOSS DEV_PRECISION DEV_RECALL DEV_F1 DEV_ACCURACY
2
+ 1 10:45:12 0.0000 0.3125 0.1323 0.8156 0.5300 0.6425 0.4790
3
+ 2 10:46:31 0.0000 0.1053 0.1047 0.8031 0.6364 0.7101 0.5626
4
+ 3 10:47:51 0.0000 0.0730 0.0955 0.8315 0.7748 0.8021 0.6793
5
+ 4 10:49:09 0.0000 0.0541 0.1212 0.8000 0.7893 0.7946 0.6809
6
+ 5 10:50:27 0.0000 0.0393 0.1539 0.8672 0.7283 0.7917 0.6632
7
+ 6 10:51:43 0.0000 0.0317 0.1976 0.8851 0.7324 0.8016 0.6830
8
+ 7 10:53:00 0.0000 0.0209 0.1801 0.8503 0.7572 0.8011 0.6831
9
+ 8 10:54:17 0.0000 0.0151 0.2043 0.8443 0.7562 0.7978 0.6803
10
+ 9 10:55:34 0.0000 0.0105 0.1840 0.8485 0.7521 0.7974 0.6760
11
+ 10 10:56:52 0.0000 0.0070 0.1938 0.8676 0.7583 0.8093 0.6931
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,239 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-14 10:43:55,873 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-14 10:43:55,874 Model: "SequenceTagger(
3
+ (embeddings): TransformerWordEmbeddings(
4
+ (model): BertModel(
5
+ (embeddings): BertEmbeddings(
6
+ (word_embeddings): Embedding(32001, 768)
7
+ (position_embeddings): Embedding(512, 768)
8
+ (token_type_embeddings): Embedding(2, 768)
9
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
10
+ (dropout): Dropout(p=0.1, inplace=False)
11
+ )
12
+ (encoder): BertEncoder(
13
+ (layer): ModuleList(
14
+ (0-11): 12 x BertLayer(
15
+ (attention): BertAttention(
16
+ (self): BertSelfAttention(
17
+ (query): Linear(in_features=768, out_features=768, bias=True)
18
+ (key): Linear(in_features=768, out_features=768, bias=True)
19
+ (value): Linear(in_features=768, out_features=768, bias=True)
20
+ (dropout): Dropout(p=0.1, inplace=False)
21
+ )
22
+ (output): BertSelfOutput(
23
+ (dense): Linear(in_features=768, out_features=768, bias=True)
24
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
25
+ (dropout): Dropout(p=0.1, inplace=False)
26
+ )
27
+ )
28
+ (intermediate): BertIntermediate(
29
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
30
+ (intermediate_act_fn): GELUActivation()
31
+ )
32
+ (output): BertOutput(
33
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
34
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
35
+ (dropout): Dropout(p=0.1, inplace=False)
36
+ )
37
+ )
38
+ )
39
+ )
40
+ (pooler): BertPooler(
41
+ (dense): Linear(in_features=768, out_features=768, bias=True)
42
+ (activation): Tanh()
43
+ )
44
+ )
45
+ )
46
+ (locked_dropout): LockedDropout(p=0.5)
47
+ (linear): Linear(in_features=768, out_features=13, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-14 10:43:55,874 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-14 10:43:55,874 MultiCorpus: 5777 train + 722 dev + 723 test sentences
52
+ - NER_ICDAR_EUROPEANA Corpus: 5777 train + 722 dev + 723 test sentences - /root/.flair/datasets/ner_icdar_europeana/nl
53
+ 2023-10-14 10:43:55,874 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-14 10:43:55,874 Train: 5777 sentences
55
+ 2023-10-14 10:43:55,874 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-14 10:43:55,874 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-14 10:43:55,875 Training Params:
58
+ 2023-10-14 10:43:55,875 - learning_rate: "5e-05"
59
+ 2023-10-14 10:43:55,875 - mini_batch_size: "4"
60
+ 2023-10-14 10:43:55,875 - max_epochs: "10"
61
+ 2023-10-14 10:43:55,875 - shuffle: "True"
62
+ 2023-10-14 10:43:55,875 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-14 10:43:55,875 Plugins:
64
+ 2023-10-14 10:43:55,875 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-14 10:43:55,875 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-14 10:43:55,875 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-14 10:43:55,875 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-14 10:43:55,875 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-14 10:43:55,875 Computation:
70
+ 2023-10-14 10:43:55,875 - compute on device: cuda:0
71
+ 2023-10-14 10:43:55,875 - embedding storage: none
72
+ 2023-10-14 10:43:55,875 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-14 10:43:55,875 Model training base path: "hmbench-icdar/nl-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4"
74
+ 2023-10-14 10:43:55,875 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-14 10:43:55,875 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-14 10:44:03,371 epoch 1 - iter 144/1445 - loss 1.53614842 - time (sec): 7.49 - samples/sec: 2488.14 - lr: 0.000005 - momentum: 0.000000
77
+ 2023-10-14 10:44:10,666 epoch 1 - iter 288/1445 - loss 0.92543100 - time (sec): 14.79 - samples/sec: 2442.39 - lr: 0.000010 - momentum: 0.000000
78
+ 2023-10-14 10:44:18,047 epoch 1 - iter 432/1445 - loss 0.68772860 - time (sec): 22.17 - samples/sec: 2403.60 - lr: 0.000015 - momentum: 0.000000
79
+ 2023-10-14 10:44:25,279 epoch 1 - iter 576/1445 - loss 0.56173958 - time (sec): 29.40 - samples/sec: 2387.31 - lr: 0.000020 - momentum: 0.000000
80
+ 2023-10-14 10:44:32,595 epoch 1 - iter 720/1445 - loss 0.47974609 - time (sec): 36.72 - samples/sec: 2404.37 - lr: 0.000025 - momentum: 0.000000
81
+ 2023-10-14 10:44:39,880 epoch 1 - iter 864/1445 - loss 0.42590462 - time (sec): 44.00 - samples/sec: 2411.14 - lr: 0.000030 - momentum: 0.000000
82
+ 2023-10-14 10:44:47,180 epoch 1 - iter 1008/1445 - loss 0.38604397 - time (sec): 51.30 - samples/sec: 2409.43 - lr: 0.000035 - momentum: 0.000000
83
+ 2023-10-14 10:44:54,827 epoch 1 - iter 1152/1445 - loss 0.35511393 - time (sec): 58.95 - samples/sec: 2419.50 - lr: 0.000040 - momentum: 0.000000
84
+ 2023-10-14 10:45:01,879 epoch 1 - iter 1296/1445 - loss 0.33007466 - time (sec): 66.00 - samples/sec: 2423.12 - lr: 0.000045 - momentum: 0.000000
85
+ 2023-10-14 10:45:08,740 epoch 1 - iter 1440/1445 - loss 0.31318374 - time (sec): 72.86 - samples/sec: 2410.77 - lr: 0.000050 - momentum: 0.000000
86
+ 2023-10-14 10:45:08,985 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-14 10:45:08,986 EPOCH 1 done: loss 0.3125 - lr: 0.000050
88
+ 2023-10-14 10:45:12,844 DEV : loss 0.1323496401309967 - f1-score (micro avg) 0.6425
89
+ 2023-10-14 10:45:12,863 saving best model
90
+ 2023-10-14 10:45:13,235 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-14 10:45:20,499 epoch 2 - iter 144/1445 - loss 0.12655175 - time (sec): 7.26 - samples/sec: 2233.65 - lr: 0.000049 - momentum: 0.000000
92
+ 2023-10-14 10:45:28,359 epoch 2 - iter 288/1445 - loss 0.11664640 - time (sec): 15.12 - samples/sec: 2234.54 - lr: 0.000049 - momentum: 0.000000
93
+ 2023-10-14 10:45:35,769 epoch 2 - iter 432/1445 - loss 0.11764083 - time (sec): 22.53 - samples/sec: 2297.07 - lr: 0.000048 - momentum: 0.000000
94
+ 2023-10-14 10:45:43,380 epoch 2 - iter 576/1445 - loss 0.11288782 - time (sec): 30.14 - samples/sec: 2351.55 - lr: 0.000048 - momentum: 0.000000
95
+ 2023-10-14 10:45:50,679 epoch 2 - iter 720/1445 - loss 0.11005752 - time (sec): 37.44 - samples/sec: 2372.40 - lr: 0.000047 - momentum: 0.000000
96
+ 2023-10-14 10:45:57,858 epoch 2 - iter 864/1445 - loss 0.10861039 - time (sec): 44.62 - samples/sec: 2366.36 - lr: 0.000047 - momentum: 0.000000
97
+ 2023-10-14 10:46:04,905 epoch 2 - iter 1008/1445 - loss 0.11009884 - time (sec): 51.67 - samples/sec: 2363.61 - lr: 0.000046 - momentum: 0.000000
98
+ 2023-10-14 10:46:12,036 epoch 2 - iter 1152/1445 - loss 0.10756740 - time (sec): 58.80 - samples/sec: 2371.45 - lr: 0.000046 - momentum: 0.000000
99
+ 2023-10-14 10:46:19,797 epoch 2 - iter 1296/1445 - loss 0.10553537 - time (sec): 66.56 - samples/sec: 2362.48 - lr: 0.000045 - momentum: 0.000000
100
+ 2023-10-14 10:46:27,342 epoch 2 - iter 1440/1445 - loss 0.10547907 - time (sec): 74.10 - samples/sec: 2370.93 - lr: 0.000044 - momentum: 0.000000
101
+ 2023-10-14 10:46:27,595 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-14 10:46:27,595 EPOCH 2 done: loss 0.1053 - lr: 0.000044
103
+ 2023-10-14 10:46:31,969 DEV : loss 0.10467828810214996 - f1-score (micro avg) 0.7101
104
+ 2023-10-14 10:46:31,987 saving best model
105
+ 2023-10-14 10:46:32,513 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-14 10:46:40,845 epoch 3 - iter 144/1445 - loss 0.06046434 - time (sec): 8.33 - samples/sec: 2176.51 - lr: 0.000044 - momentum: 0.000000
107
+ 2023-10-14 10:46:49,225 epoch 3 - iter 288/1445 - loss 0.06004263 - time (sec): 16.71 - samples/sec: 2131.19 - lr: 0.000043 - momentum: 0.000000
108
+ 2023-10-14 10:46:56,492 epoch 3 - iter 432/1445 - loss 0.06632407 - time (sec): 23.98 - samples/sec: 2165.44 - lr: 0.000043 - momentum: 0.000000
109
+ 2023-10-14 10:47:03,797 epoch 3 - iter 576/1445 - loss 0.06736150 - time (sec): 31.28 - samples/sec: 2213.57 - lr: 0.000042 - momentum: 0.000000
110
+ 2023-10-14 10:47:11,200 epoch 3 - iter 720/1445 - loss 0.06752059 - time (sec): 38.68 - samples/sec: 2266.94 - lr: 0.000042 - momentum: 0.000000
111
+ 2023-10-14 10:47:18,528 epoch 3 - iter 864/1445 - loss 0.07237618 - time (sec): 46.01 - samples/sec: 2285.16 - lr: 0.000041 - momentum: 0.000000
112
+ 2023-10-14 10:47:26,048 epoch 3 - iter 1008/1445 - loss 0.07408386 - time (sec): 53.53 - samples/sec: 2314.03 - lr: 0.000041 - momentum: 0.000000
113
+ 2023-10-14 10:47:33,156 epoch 3 - iter 1152/1445 - loss 0.07290332 - time (sec): 60.64 - samples/sec: 2314.08 - lr: 0.000040 - momentum: 0.000000
114
+ 2023-10-14 10:47:40,398 epoch 3 - iter 1296/1445 - loss 0.07248165 - time (sec): 67.88 - samples/sec: 2319.56 - lr: 0.000039 - momentum: 0.000000
115
+ 2023-10-14 10:47:47,740 epoch 3 - iter 1440/1445 - loss 0.07325238 - time (sec): 75.22 - samples/sec: 2331.61 - lr: 0.000039 - momentum: 0.000000
116
+ 2023-10-14 10:47:48,047 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-14 10:47:48,048 EPOCH 3 done: loss 0.0730 - lr: 0.000039
118
+ 2023-10-14 10:47:51,583 DEV : loss 0.09553560614585876 - f1-score (micro avg) 0.8021
119
+ 2023-10-14 10:47:51,599 saving best model
120
+ 2023-10-14 10:47:52,265 ----------------------------------------------------------------------------------------------------
121
+ 2023-10-14 10:47:59,539 epoch 4 - iter 144/1445 - loss 0.04698177 - time (sec): 7.27 - samples/sec: 2415.48 - lr: 0.000038 - momentum: 0.000000
122
+ 2023-10-14 10:48:07,133 epoch 4 - iter 288/1445 - loss 0.06031754 - time (sec): 14.87 - samples/sec: 2411.66 - lr: 0.000038 - momentum: 0.000000
123
+ 2023-10-14 10:48:14,165 epoch 4 - iter 432/1445 - loss 0.06263096 - time (sec): 21.90 - samples/sec: 2410.81 - lr: 0.000037 - momentum: 0.000000
124
+ 2023-10-14 10:48:21,491 epoch 4 - iter 576/1445 - loss 0.05725339 - time (sec): 29.22 - samples/sec: 2417.83 - lr: 0.000037 - momentum: 0.000000
125
+ 2023-10-14 10:48:28,487 epoch 4 - iter 720/1445 - loss 0.05516418 - time (sec): 36.22 - samples/sec: 2406.40 - lr: 0.000036 - momentum: 0.000000
126
+ 2023-10-14 10:48:36,217 epoch 4 - iter 864/1445 - loss 0.05343786 - time (sec): 43.95 - samples/sec: 2404.89 - lr: 0.000036 - momentum: 0.000000
127
+ 2023-10-14 10:48:43,475 epoch 4 - iter 1008/1445 - loss 0.05296204 - time (sec): 51.21 - samples/sec: 2397.09 - lr: 0.000035 - momentum: 0.000000
128
+ 2023-10-14 10:48:50,864 epoch 4 - iter 1152/1445 - loss 0.05396480 - time (sec): 58.60 - samples/sec: 2397.71 - lr: 0.000034 - momentum: 0.000000
129
+ 2023-10-14 10:48:58,163 epoch 4 - iter 1296/1445 - loss 0.05465199 - time (sec): 65.90 - samples/sec: 2406.38 - lr: 0.000034 - momentum: 0.000000
130
+ 2023-10-14 10:49:05,509 epoch 4 - iter 1440/1445 - loss 0.05379926 - time (sec): 73.24 - samples/sec: 2397.56 - lr: 0.000033 - momentum: 0.000000
131
+ 2023-10-14 10:49:05,752 ----------------------------------------------------------------------------------------------------
132
+ 2023-10-14 10:49:05,752 EPOCH 4 done: loss 0.0541 - lr: 0.000033
133
+ 2023-10-14 10:49:09,309 DEV : loss 0.12118156254291534 - f1-score (micro avg) 0.7946
134
+ 2023-10-14 10:49:09,326 ----------------------------------------------------------------------------------------------------
135
+ 2023-10-14 10:49:16,954 epoch 5 - iter 144/1445 - loss 0.04673101 - time (sec): 7.63 - samples/sec: 2414.26 - lr: 0.000033 - momentum: 0.000000
136
+ 2023-10-14 10:49:24,004 epoch 5 - iter 288/1445 - loss 0.04488133 - time (sec): 14.68 - samples/sec: 2414.51 - lr: 0.000032 - momentum: 0.000000
137
+ 2023-10-14 10:49:31,520 epoch 5 - iter 432/1445 - loss 0.04016346 - time (sec): 22.19 - samples/sec: 2439.19 - lr: 0.000032 - momentum: 0.000000
138
+ 2023-10-14 10:49:38,723 epoch 5 - iter 576/1445 - loss 0.04147102 - time (sec): 29.40 - samples/sec: 2409.52 - lr: 0.000031 - momentum: 0.000000
139
+ 2023-10-14 10:49:46,016 epoch 5 - iter 720/1445 - loss 0.04117617 - time (sec): 36.69 - samples/sec: 2388.99 - lr: 0.000031 - momentum: 0.000000
140
+ 2023-10-14 10:49:53,185 epoch 5 - iter 864/1445 - loss 0.04145618 - time (sec): 43.86 - samples/sec: 2366.07 - lr: 0.000030 - momentum: 0.000000
141
+ 2023-10-14 10:50:00,848 epoch 5 - iter 1008/1445 - loss 0.03979828 - time (sec): 51.52 - samples/sec: 2367.21 - lr: 0.000029 - momentum: 0.000000
142
+ 2023-10-14 10:50:08,189 epoch 5 - iter 1152/1445 - loss 0.03982370 - time (sec): 58.86 - samples/sec: 2374.01 - lr: 0.000029 - momentum: 0.000000
143
+ 2023-10-14 10:50:15,584 epoch 5 - iter 1296/1445 - loss 0.04077806 - time (sec): 66.26 - samples/sec: 2388.64 - lr: 0.000028 - momentum: 0.000000
144
+ 2023-10-14 10:50:23,148 epoch 5 - iter 1440/1445 - loss 0.03927038 - time (sec): 73.82 - samples/sec: 2380.09 - lr: 0.000028 - momentum: 0.000000
145
+ 2023-10-14 10:50:23,369 ----------------------------------------------------------------------------------------------------
146
+ 2023-10-14 10:50:23,369 EPOCH 5 done: loss 0.0393 - lr: 0.000028
147
+ 2023-10-14 10:50:27,269 DEV : loss 0.15388108789920807 - f1-score (micro avg) 0.7917
148
+ 2023-10-14 10:50:27,287 ----------------------------------------------------------------------------------------------------
149
+ 2023-10-14 10:50:34,633 epoch 6 - iter 144/1445 - loss 0.03233428 - time (sec): 7.34 - samples/sec: 2362.50 - lr: 0.000027 - momentum: 0.000000
150
+ 2023-10-14 10:50:41,959 epoch 6 - iter 288/1445 - loss 0.03501451 - time (sec): 14.67 - samples/sec: 2390.85 - lr: 0.000027 - momentum: 0.000000
151
+ 2023-10-14 10:50:49,055 epoch 6 - iter 432/1445 - loss 0.03123834 - time (sec): 21.77 - samples/sec: 2411.58 - lr: 0.000026 - momentum: 0.000000
152
+ 2023-10-14 10:50:56,443 epoch 6 - iter 576/1445 - loss 0.03274913 - time (sec): 29.16 - samples/sec: 2419.30 - lr: 0.000026 - momentum: 0.000000
153
+ 2023-10-14 10:51:03,609 epoch 6 - iter 720/1445 - loss 0.03314690 - time (sec): 36.32 - samples/sec: 2406.97 - lr: 0.000025 - momentum: 0.000000
154
+ 2023-10-14 10:51:10,863 epoch 6 - iter 864/1445 - loss 0.03210885 - time (sec): 43.58 - samples/sec: 2410.60 - lr: 0.000024 - momentum: 0.000000
155
+ 2023-10-14 10:51:18,264 epoch 6 - iter 1008/1445 - loss 0.03115240 - time (sec): 50.98 - samples/sec: 2410.40 - lr: 0.000024 - momentum: 0.000000
156
+ 2023-10-14 10:51:25,820 epoch 6 - iter 1152/1445 - loss 0.03008407 - time (sec): 58.53 - samples/sec: 2430.27 - lr: 0.000023 - momentum: 0.000000
157
+ 2023-10-14 10:51:32,892 epoch 6 - iter 1296/1445 - loss 0.03095943 - time (sec): 65.60 - samples/sec: 2425.45 - lr: 0.000023 - momentum: 0.000000
158
+ 2023-10-14 10:51:39,991 epoch 6 - iter 1440/1445 - loss 0.03162133 - time (sec): 72.70 - samples/sec: 2417.44 - lr: 0.000022 - momentum: 0.000000
159
+ 2023-10-14 10:51:40,215 ----------------------------------------------------------------------------------------------------
160
+ 2023-10-14 10:51:40,215 EPOCH 6 done: loss 0.0317 - lr: 0.000022
161
+ 2023-10-14 10:51:43,805 DEV : loss 0.19756034016609192 - f1-score (micro avg) 0.8016
162
+ 2023-10-14 10:51:43,821 ----------------------------------------------------------------------------------------------------
163
+ 2023-10-14 10:51:51,088 epoch 7 - iter 144/1445 - loss 0.01816229 - time (sec): 7.27 - samples/sec: 2414.93 - lr: 0.000022 - momentum: 0.000000
164
+ 2023-10-14 10:51:58,649 epoch 7 - iter 288/1445 - loss 0.01954064 - time (sec): 14.83 - samples/sec: 2471.92 - lr: 0.000021 - momentum: 0.000000
165
+ 2023-10-14 10:52:05,759 epoch 7 - iter 432/1445 - loss 0.01867134 - time (sec): 21.94 - samples/sec: 2439.90 - lr: 0.000021 - momentum: 0.000000
166
+ 2023-10-14 10:52:13,556 epoch 7 - iter 576/1445 - loss 0.01895690 - time (sec): 29.73 - samples/sec: 2418.92 - lr: 0.000020 - momentum: 0.000000
167
+ 2023-10-14 10:52:20,679 epoch 7 - iter 720/1445 - loss 0.01836345 - time (sec): 36.86 - samples/sec: 2400.83 - lr: 0.000019 - momentum: 0.000000
168
+ 2023-10-14 10:52:27,667 epoch 7 - iter 864/1445 - loss 0.01877869 - time (sec): 43.84 - samples/sec: 2396.78 - lr: 0.000019 - momentum: 0.000000
169
+ 2023-10-14 10:52:35,009 epoch 7 - iter 1008/1445 - loss 0.02080686 - time (sec): 51.19 - samples/sec: 2419.50 - lr: 0.000018 - momentum: 0.000000
170
+ 2023-10-14 10:52:42,373 epoch 7 - iter 1152/1445 - loss 0.02134693 - time (sec): 58.55 - samples/sec: 2423.15 - lr: 0.000018 - momentum: 0.000000
171
+ 2023-10-14 10:52:49,576 epoch 7 - iter 1296/1445 - loss 0.02105516 - time (sec): 65.75 - samples/sec: 2420.11 - lr: 0.000017 - momentum: 0.000000
172
+ 2023-10-14 10:52:56,887 epoch 7 - iter 1440/1445 - loss 0.02073299 - time (sec): 73.06 - samples/sec: 2406.17 - lr: 0.000017 - momentum: 0.000000
173
+ 2023-10-14 10:52:57,114 ----------------------------------------------------------------------------------------------------
174
+ 2023-10-14 10:52:57,115 EPOCH 7 done: loss 0.0209 - lr: 0.000017
175
+ 2023-10-14 10:53:00,712 DEV : loss 0.18008936941623688 - f1-score (micro avg) 0.8011
176
+ 2023-10-14 10:53:00,729 ----------------------------------------------------------------------------------------------------
177
+ 2023-10-14 10:53:07,996 epoch 8 - iter 144/1445 - loss 0.01866497 - time (sec): 7.27 - samples/sec: 2300.59 - lr: 0.000016 - momentum: 0.000000
178
+ 2023-10-14 10:53:15,313 epoch 8 - iter 288/1445 - loss 0.01466923 - time (sec): 14.58 - samples/sec: 2372.09 - lr: 0.000016 - momentum: 0.000000
179
+ 2023-10-14 10:53:22,675 epoch 8 - iter 432/1445 - loss 0.01583407 - time (sec): 21.95 - samples/sec: 2384.78 - lr: 0.000015 - momentum: 0.000000
180
+ 2023-10-14 10:53:29,992 epoch 8 - iter 576/1445 - loss 0.01493921 - time (sec): 29.26 - samples/sec: 2381.30 - lr: 0.000014 - momentum: 0.000000
181
+ 2023-10-14 10:53:37,264 epoch 8 - iter 720/1445 - loss 0.01503871 - time (sec): 36.53 - samples/sec: 2407.26 - lr: 0.000014 - momentum: 0.000000
182
+ 2023-10-14 10:53:44,308 epoch 8 - iter 864/1445 - loss 0.01515293 - time (sec): 43.58 - samples/sec: 2399.36 - lr: 0.000013 - momentum: 0.000000
183
+ 2023-10-14 10:53:51,766 epoch 8 - iter 1008/1445 - loss 0.01576861 - time (sec): 51.04 - samples/sec: 2404.97 - lr: 0.000013 - momentum: 0.000000
184
+ 2023-10-14 10:53:59,034 epoch 8 - iter 1152/1445 - loss 0.01561594 - time (sec): 58.30 - samples/sec: 2408.13 - lr: 0.000012 - momentum: 0.000000
185
+ 2023-10-14 10:54:06,080 epoch 8 - iter 1296/1445 - loss 0.01534350 - time (sec): 65.35 - samples/sec: 2411.17 - lr: 0.000012 - momentum: 0.000000
186
+ 2023-10-14 10:54:13,533 epoch 8 - iter 1440/1445 - loss 0.01509812 - time (sec): 72.80 - samples/sec: 2409.56 - lr: 0.000011 - momentum: 0.000000
187
+ 2023-10-14 10:54:13,784 ----------------------------------------------------------------------------------------------------
188
+ 2023-10-14 10:54:13,784 EPOCH 8 done: loss 0.0151 - lr: 0.000011
189
+ 2023-10-14 10:54:17,824 DEV : loss 0.20430612564086914 - f1-score (micro avg) 0.7978
190
+ 2023-10-14 10:54:17,841 ----------------------------------------------------------------------------------------------------
191
+ 2023-10-14 10:54:25,386 epoch 9 - iter 144/1445 - loss 0.01055124 - time (sec): 7.54 - samples/sec: 2408.34 - lr: 0.000011 - momentum: 0.000000
192
+ 2023-10-14 10:54:32,660 epoch 9 - iter 288/1445 - loss 0.00857411 - time (sec): 14.82 - samples/sec: 2384.81 - lr: 0.000010 - momentum: 0.000000
193
+ 2023-10-14 10:54:39,915 epoch 9 - iter 432/1445 - loss 0.00889903 - time (sec): 22.07 - samples/sec: 2361.30 - lr: 0.000009 - momentum: 0.000000
194
+ 2023-10-14 10:54:47,643 epoch 9 - iter 576/1445 - loss 0.01030940 - time (sec): 29.80 - samples/sec: 2394.86 - lr: 0.000009 - momentum: 0.000000
195
+ 2023-10-14 10:54:54,820 epoch 9 - iter 720/1445 - loss 0.00993058 - time (sec): 36.98 - samples/sec: 2400.63 - lr: 0.000008 - momentum: 0.000000
196
+ 2023-10-14 10:55:02,175 epoch 9 - iter 864/1445 - loss 0.00936448 - time (sec): 44.33 - samples/sec: 2406.24 - lr: 0.000008 - momentum: 0.000000
197
+ 2023-10-14 10:55:09,201 epoch 9 - iter 1008/1445 - loss 0.00913328 - time (sec): 51.36 - samples/sec: 2404.99 - lr: 0.000007 - momentum: 0.000000
198
+ 2023-10-14 10:55:16,575 epoch 9 - iter 1152/1445 - loss 0.00941487 - time (sec): 58.73 - samples/sec: 2408.84 - lr: 0.000007 - momentum: 0.000000
199
+ 2023-10-14 10:55:23,825 epoch 9 - iter 1296/1445 - loss 0.01026341 - time (sec): 65.98 - samples/sec: 2401.83 - lr: 0.000006 - momentum: 0.000000
200
+ 2023-10-14 10:55:31,036 epoch 9 - iter 1440/1445 - loss 0.01047293 - time (sec): 73.19 - samples/sec: 2400.81 - lr: 0.000006 - momentum: 0.000000
201
+ 2023-10-14 10:55:31,280 ----------------------------------------------------------------------------------------------------
202
+ 2023-10-14 10:55:31,280 EPOCH 9 done: loss 0.0105 - lr: 0.000006
203
+ 2023-10-14 10:55:34,878 DEV : loss 0.1839500218629837 - f1-score (micro avg) 0.7974
204
+ 2023-10-14 10:55:34,896 ----------------------------------------------------------------------------------------------------
205
+ 2023-10-14 10:55:42,196 epoch 10 - iter 144/1445 - loss 0.00490526 - time (sec): 7.30 - samples/sec: 2287.22 - lr: 0.000005 - momentum: 0.000000
206
+ 2023-10-14 10:55:49,779 epoch 10 - iter 288/1445 - loss 0.00524218 - time (sec): 14.88 - samples/sec: 2333.86 - lr: 0.000004 - momentum: 0.000000
207
+ 2023-10-14 10:55:57,744 epoch 10 - iter 432/1445 - loss 0.00648842 - time (sec): 22.85 - samples/sec: 2293.26 - lr: 0.000004 - momentum: 0.000000
208
+ 2023-10-14 10:56:05,487 epoch 10 - iter 576/1445 - loss 0.00656819 - time (sec): 30.59 - samples/sec: 2325.72 - lr: 0.000003 - momentum: 0.000000
209
+ 2023-10-14 10:56:12,959 epoch 10 - iter 720/1445 - loss 0.00698945 - time (sec): 38.06 - samples/sec: 2354.45 - lr: 0.000003 - momentum: 0.000000
210
+ 2023-10-14 10:56:20,142 epoch 10 - iter 864/1445 - loss 0.00778598 - time (sec): 45.24 - samples/sec: 2362.14 - lr: 0.000002 - momentum: 0.000000
211
+ 2023-10-14 10:56:27,101 epoch 10 - iter 1008/1445 - loss 0.00737225 - time (sec): 52.20 - samples/sec: 2348.21 - lr: 0.000002 - momentum: 0.000000
212
+ 2023-10-14 10:56:34,107 epoch 10 - iter 1152/1445 - loss 0.00697948 - time (sec): 59.21 - samples/sec: 2352.75 - lr: 0.000001 - momentum: 0.000000
213
+ 2023-10-14 10:56:41,417 epoch 10 - iter 1296/1445 - loss 0.00700128 - time (sec): 66.52 - samples/sec: 2371.69 - lr: 0.000001 - momentum: 0.000000
214
+ 2023-10-14 10:56:48,753 epoch 10 - iter 1440/1445 - loss 0.00696742 - time (sec): 73.86 - samples/sec: 2376.15 - lr: 0.000000 - momentum: 0.000000
215
+ 2023-10-14 10:56:49,020 ----------------------------------------------------------------------------------------------------
216
+ 2023-10-14 10:56:49,021 EPOCH 10 done: loss 0.0070 - lr: 0.000000
217
+ 2023-10-14 10:56:52,539 DEV : loss 0.19384992122650146 - f1-score (micro avg) 0.8093
218
+ 2023-10-14 10:56:52,555 saving best model
219
+ 2023-10-14 10:56:53,410 ----------------------------------------------------------------------------------------------------
220
+ 2023-10-14 10:56:53,411 Loading model from best epoch ...
221
+ 2023-10-14 10:56:55,129 SequenceTagger predicts: Dictionary with 13 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-PER, B-PER, E-PER, I-PER, S-ORG, B-ORG, E-ORG, I-ORG
222
+ 2023-10-14 10:56:58,276
223
+ Results:
224
+ - F-score (micro) 0.7959
225
+ - F-score (macro) 0.6975
226
+ - Accuracy 0.6749
227
+
228
+ By class:
229
+ precision recall f1-score support
230
+
231
+ PER 0.8184 0.7759 0.7966 482
232
+ LOC 0.8949 0.7991 0.8443 458
233
+ ORG 0.5091 0.4058 0.4516 69
234
+
235
+ micro avg 0.8339 0.7611 0.7959 1009
236
+ macro avg 0.7408 0.6603 0.6975 1009
237
+ weighted avg 0.8319 0.7611 0.7947 1009
238
+
239
+ 2023-10-14 10:56:58,276 ----------------------------------------------------------------------------------------------------