Hugging Face's logo

stefan-it
/
hmbench-icdar-nl-hmbert-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-5

Token Classification
Flair
PyTorch
Dutch
sequence-tagger-model
Model card Files Community
1
stefan-it commited on
Commit
6cf3bbd
•
1 Parent(s): 8330f43

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 +241 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a414a7075bc4c0520a70e9f08bf80f58657a180612862b6467220b25b1b67ce7
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 11:48:13 0.0000 0.4072 0.1359 0.6752 0.6012 0.6361 0.4767
3
+ 2 11:49:18 0.0000 0.1023 0.0926 0.8108 0.7479 0.7781 0.6493
4
+ 3 11:50:24 0.0000 0.0641 0.0804 0.7747 0.8171 0.7954 0.6743
5
+ 4 11:51:28 0.0000 0.0424 0.0950 0.8391 0.7541 0.7943 0.6691
6
+ 5 11:52:33 0.0000 0.0319 0.1185 0.7883 0.8233 0.8055 0.6894
7
+ 6 11:53:38 0.0000 0.0246 0.1389 0.8565 0.7769 0.8147 0.7015
8
+ 7 11:54:43 0.0000 0.0172 0.1678 0.8600 0.7614 0.8077 0.6920
9
+ 8 11:55:49 0.0000 0.0133 0.1813 0.8783 0.7531 0.8109 0.6949
10
+ 9 11:56:53 0.0000 0.0103 0.1801 0.8533 0.7810 0.8155 0.7019
11
+ 10 11:57:57 0.0000 0.0080 0.1833 0.8508 0.7779 0.8127 0.6998
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,241 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-14 11:47:09,968 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-14 11:47:09,969 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 11:47:09,969 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-14 11:47:09,969 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 11:47:09,969 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-14 11:47:09,969 Train: 5777 sentences
55
+ 2023-10-14 11:47:09,969 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-14 11:47:09,969 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-14 11:47:09,969 Training Params:
58
+ 2023-10-14 11:47:09,969 - learning_rate: "3e-05"
59
+ 2023-10-14 11:47:09,970 - mini_batch_size: "8"
60
+ 2023-10-14 11:47:09,970 - max_epochs: "10"
61
+ 2023-10-14 11:47:09,970 - shuffle: "True"
62
+ 2023-10-14 11:47:09,970 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-14 11:47:09,970 Plugins:
64
+ 2023-10-14 11:47:09,970 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-14 11:47:09,970 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-14 11:47:09,970 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-14 11:47:09,970 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-14 11:47:09,970 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-14 11:47:09,970 Computation:
70
+ 2023-10-14 11:47:09,970 - compute on device: cuda:0
71
+ 2023-10-14 11:47:09,970 - embedding storage: none
72
+ 2023-10-14 11:47:09,970 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-14 11:47:09,970 Model training base path: "hmbench-icdar/nl-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-5"
74
+ 2023-10-14 11:47:09,970 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-14 11:47:09,970 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-14 11:47:16,044 epoch 1 - iter 72/723 - loss 2.17936848 - time (sec): 6.07 - samples/sec: 3053.65 - lr: 0.000003 - momentum: 0.000000
77
+ 2023-10-14 11:47:22,048 epoch 1 - iter 144/723 - loss 1.30488853 - time (sec): 12.08 - samples/sec: 2974.16 - lr: 0.000006 - momentum: 0.000000
78
+ 2023-10-14 11:47:27,819 epoch 1 - iter 216/723 - loss 0.96288539 - time (sec): 17.85 - samples/sec: 2988.26 - lr: 0.000009 - momentum: 0.000000
79
+ 2023-10-14 11:47:34,101 epoch 1 - iter 288/723 - loss 0.77877705 - time (sec): 24.13 - samples/sec: 2957.82 - lr: 0.000012 - momentum: 0.000000
80
+ 2023-10-14 11:47:40,166 epoch 1 - iter 360/723 - loss 0.66184851 - time (sec): 30.19 - samples/sec: 2937.05 - lr: 0.000015 - momentum: 0.000000
81
+ 2023-10-14 11:47:46,237 epoch 1 - iter 432/723 - loss 0.58058110 - time (sec): 36.27 - samples/sec: 2929.70 - lr: 0.000018 - momentum: 0.000000
82
+ 2023-10-14 11:47:52,424 epoch 1 - iter 504/723 - loss 0.52285800 - time (sec): 42.45 - samples/sec: 2920.34 - lr: 0.000021 - momentum: 0.000000
83
+ 2023-10-14 11:47:58,303 epoch 1 - iter 576/723 - loss 0.47873115 - time (sec): 48.33 - samples/sec: 2901.73 - lr: 0.000024 - momentum: 0.000000
84
+ 2023-10-14 11:48:04,095 epoch 1 - iter 648/723 - loss 0.43866282 - time (sec): 54.12 - samples/sec: 2916.97 - lr: 0.000027 - momentum: 0.000000
85
+ 2023-10-14 11:48:10,154 epoch 1 - iter 720/723 - loss 0.40791889 - time (sec): 60.18 - samples/sec: 2918.46 - lr: 0.000030 - momentum: 0.000000
86
+ 2023-10-14 11:48:10,356 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-14 11:48:10,356 EPOCH 1 done: loss 0.4072 - lr: 0.000030
88
+ 2023-10-14 11:48:13,522 DEV : loss 0.13588625192642212 - f1-score (micro avg) 0.6361
89
+ 2023-10-14 11:48:13,538 saving best model
90
+ 2023-10-14 11:48:13,953 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-14 11:48:19,937 epoch 2 - iter 72/723 - loss 0.11712269 - time (sec): 5.98 - samples/sec: 2859.68 - lr: 0.000030 - momentum: 0.000000
92
+ 2023-10-14 11:48:26,043 epoch 2 - iter 144/723 - loss 0.11668248 - time (sec): 12.09 - samples/sec: 2824.30 - lr: 0.000029 - momentum: 0.000000
93
+ 2023-10-14 11:48:32,261 epoch 2 - iter 216/723 - loss 0.11217901 - time (sec): 18.31 - samples/sec: 2863.21 - lr: 0.000029 - momentum: 0.000000
94
+ 2023-10-14 11:48:38,786 epoch 2 - iter 288/723 - loss 0.10925944 - time (sec): 24.83 - samples/sec: 2855.63 - lr: 0.000029 - momentum: 0.000000
95
+ 2023-10-14 11:48:44,379 epoch 2 - iter 360/723 - loss 0.10959979 - time (sec): 30.43 - samples/sec: 2891.40 - lr: 0.000028 - momentum: 0.000000
96
+ 2023-10-14 11:48:50,066 epoch 2 - iter 432/723 - loss 0.10627704 - time (sec): 36.11 - samples/sec: 2899.92 - lr: 0.000028 - momentum: 0.000000
97
+ 2023-10-14 11:48:56,429 epoch 2 - iter 504/723 - loss 0.10609633 - time (sec): 42.48 - samples/sec: 2880.18 - lr: 0.000028 - momentum: 0.000000
98
+ 2023-10-14 11:49:02,960 epoch 2 - iter 576/723 - loss 0.10378393 - time (sec): 49.01 - samples/sec: 2860.76 - lr: 0.000027 - momentum: 0.000000
99
+ 2023-10-14 11:49:09,150 epoch 2 - iter 648/723 - loss 0.10374395 - time (sec): 55.20 - samples/sec: 2866.12 - lr: 0.000027 - momentum: 0.000000
100
+ 2023-10-14 11:49:14,970 epoch 2 - iter 720/723 - loss 0.10227966 - time (sec): 61.02 - samples/sec: 2880.91 - lr: 0.000027 - momentum: 0.000000
101
+ 2023-10-14 11:49:15,136 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-14 11:49:15,136 EPOCH 2 done: loss 0.1023 - lr: 0.000027
103
+ 2023-10-14 11:49:18,826 DEV : loss 0.09264427423477173 - f1-score (micro avg) 0.7781
104
+ 2023-10-14 11:49:18,843 saving best model
105
+ 2023-10-14 11:49:19,322 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-14 11:49:25,171 epoch 3 - iter 72/723 - loss 0.07984702 - time (sec): 5.85 - samples/sec: 2901.93 - lr: 0.000026 - momentum: 0.000000
107
+ 2023-10-14 11:49:31,011 epoch 3 - iter 144/723 - loss 0.07300924 - time (sec): 11.69 - samples/sec: 2916.01 - lr: 0.000026 - momentum: 0.000000
108
+ 2023-10-14 11:49:37,186 epoch 3 - iter 216/723 - loss 0.06832100 - time (sec): 17.86 - samples/sec: 2865.29 - lr: 0.000026 - momentum: 0.000000
109
+ 2023-10-14 11:49:42,899 epoch 3 - iter 288/723 - loss 0.06940238 - time (sec): 23.57 - samples/sec: 2881.77 - lr: 0.000025 - momentum: 0.000000
110
+ 2023-10-14 11:49:48,505 epoch 3 - iter 360/723 - loss 0.06705455 - time (sec): 29.18 - samples/sec: 2889.13 - lr: 0.000025 - momentum: 0.000000
111
+ 2023-10-14 11:49:54,710 epoch 3 - iter 432/723 - loss 0.06525111 - time (sec): 35.39 - samples/sec: 2917.74 - lr: 0.000025 - momentum: 0.000000
112
+ 2023-10-14 11:50:00,599 epoch 3 - iter 504/723 - loss 0.06489497 - time (sec): 41.28 - samples/sec: 2917.45 - lr: 0.000024 - momentum: 0.000000
113
+ 2023-10-14 11:50:06,623 epoch 3 - iter 576/723 - loss 0.06635288 - time (sec): 47.30 - samples/sec: 2941.51 - lr: 0.000024 - momentum: 0.000000
114
+ 2023-10-14 11:50:12,929 epoch 3 - iter 648/723 - loss 0.06505342 - time (sec): 53.60 - samples/sec: 2926.74 - lr: 0.000024 - momentum: 0.000000
115
+ 2023-10-14 11:50:19,279 epoch 3 - iter 720/723 - loss 0.06394731 - time (sec): 59.95 - samples/sec: 2932.02 - lr: 0.000023 - momentum: 0.000000
116
+ 2023-10-14 11:50:19,444 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-14 11:50:19,444 EPOCH 3 done: loss 0.0641 - lr: 0.000023
118
+ 2023-10-14 11:50:24,040 DEV : loss 0.08042255789041519 - f1-score (micro avg) 0.7954
119
+ 2023-10-14 11:50:24,072 saving best model
120
+ 2023-10-14 11:50:24,642 ----------------------------------------------------------------------------------------------------
121
+ 2023-10-14 11:50:30,665 epoch 4 - iter 72/723 - loss 0.03888401 - time (sec): 6.02 - samples/sec: 2815.75 - lr: 0.000023 - momentum: 0.000000
122
+ 2023-10-14 11:50:36,602 epoch 4 - iter 144/723 - loss 0.04190461 - time (sec): 11.96 - samples/sec: 2998.16 - lr: 0.000023 - momentum: 0.000000
123
+ 2023-10-14 11:50:42,498 epoch 4 - iter 216/723 - loss 0.04041996 - time (sec): 17.85 - samples/sec: 2969.21 - lr: 0.000022 - momentum: 0.000000
124
+ 2023-10-14 11:50:48,871 epoch 4 - iter 288/723 - loss 0.04268068 - time (sec): 24.23 - samples/sec: 2917.23 - lr: 0.000022 - momentum: 0.000000
125
+ 2023-10-14 11:50:54,653 epoch 4 - iter 360/723 - loss 0.04323841 - time (sec): 30.01 - samples/sec: 2913.19 - lr: 0.000022 - momentum: 0.000000
126
+ 2023-10-14 11:51:00,881 epoch 4 - iter 432/723 - loss 0.04422879 - time (sec): 36.24 - samples/sec: 2899.02 - lr: 0.000021 - momentum: 0.000000
127
+ 2023-10-14 11:51:07,029 epoch 4 - iter 504/723 - loss 0.04346939 - time (sec): 42.38 - samples/sec: 2903.77 - lr: 0.000021 - momentum: 0.000000
128
+ 2023-10-14 11:51:12,738 epoch 4 - iter 576/723 - loss 0.04268000 - time (sec): 48.09 - samples/sec: 2905.38 - lr: 0.000021 - momentum: 0.000000
129
+ 2023-10-14 11:51:18,686 epoch 4 - iter 648/723 - loss 0.04192132 - time (sec): 54.04 - samples/sec: 2920.00 - lr: 0.000020 - momentum: 0.000000
130
+ 2023-10-14 11:51:24,862 epoch 4 - iter 720/723 - loss 0.04244714 - time (sec): 60.22 - samples/sec: 2919.39 - lr: 0.000020 - momentum: 0.000000
131
+ 2023-10-14 11:51:25,034 ----------------------------------------------------------------------------------------------------
132
+ 2023-10-14 11:51:25,034 EPOCH 4 done: loss 0.0424 - lr: 0.000020
133
+ 2023-10-14 11:51:28,568 DEV : loss 0.09498978406190872 - f1-score (micro avg) 0.7943
134
+ 2023-10-14 11:51:28,585 ----------------------------------------------------------------------------------------------------
135
+ 2023-10-14 11:51:34,533 epoch 5 - iter 72/723 - loss 0.02557454 - time (sec): 5.95 - samples/sec: 2796.46 - lr: 0.000020 - momentum: 0.000000
136
+ 2023-10-14 11:51:40,625 epoch 5 - iter 144/723 - loss 0.02813156 - time (sec): 12.04 - samples/sec: 2775.57 - lr: 0.000019 - momentum: 0.000000
137
+ 2023-10-14 11:51:47,161 epoch 5 - iter 216/723 - loss 0.03143363 - time (sec): 18.57 - samples/sec: 2722.72 - lr: 0.000019 - momentum: 0.000000
138
+ 2023-10-14 11:51:53,361 epoch 5 - iter 288/723 - loss 0.03092080 - time (sec): 24.77 - samples/sec: 2791.88 - lr: 0.000019 - momentum: 0.000000
139
+ 2023-10-14 11:51:59,575 epoch 5 - iter 360/723 - loss 0.03267232 - time (sec): 30.99 - samples/sec: 2813.37 - lr: 0.000018 - momentum: 0.000000
140
+ 2023-10-14 11:52:05,683 epoch 5 - iter 432/723 - loss 0.03342599 - time (sec): 37.10 - samples/sec: 2839.84 - lr: 0.000018 - momentum: 0.000000
141
+ 2023-10-14 11:52:12,172 epoch 5 - iter 504/723 - loss 0.03230870 - time (sec): 43.58 - samples/sec: 2848.04 - lr: 0.000018 - momentum: 0.000000
142
+ 2023-10-14 11:52:18,077 epoch 5 - iter 576/723 - loss 0.03186647 - time (sec): 49.49 - samples/sec: 2846.56 - lr: 0.000017 - momentum: 0.000000
143
+ 2023-10-14 11:52:23,851 epoch 5 - iter 648/723 - loss 0.03038680 - time (sec): 55.26 - samples/sec: 2860.04 - lr: 0.000017 - momentum: 0.000000
144
+ 2023-10-14 11:52:30,050 epoch 5 - iter 720/723 - loss 0.03171654 - time (sec): 61.46 - samples/sec: 2853.61 - lr: 0.000017 - momentum: 0.000000
145
+ 2023-10-14 11:52:30,317 ----------------------------------------------------------------------------------------------------
146
+ 2023-10-14 11:52:30,317 EPOCH 5 done: loss 0.0319 - lr: 0.000017
147
+ 2023-10-14 11:52:33,961 DEV : loss 0.11846506595611572 - f1-score (micro avg) 0.8055
148
+ 2023-10-14 11:52:33,977 saving best model
149
+ 2023-10-14 11:52:34,393 ----------------------------------------------------------------------------------------------------
150
+ 2023-10-14 11:52:40,195 epoch 6 - iter 72/723 - loss 0.02635934 - time (sec): 5.80 - samples/sec: 2915.02 - lr: 0.000016 - momentum: 0.000000
151
+ 2023-10-14 11:52:45,847 epoch 6 - iter 144/723 - loss 0.02567053 - time (sec): 11.45 - samples/sec: 2989.60 - lr: 0.000016 - momentum: 0.000000
152
+ 2023-10-14 11:52:52,001 epoch 6 - iter 216/723 - loss 0.02594002 - time (sec): 17.61 - samples/sec: 2958.72 - lr: 0.000016 - momentum: 0.000000
153
+ 2023-10-14 11:52:58,150 epoch 6 - iter 288/723 - loss 0.02839596 - time (sec): 23.76 - samples/sec: 2961.05 - lr: 0.000015 - momentum: 0.000000
154
+ 2023-10-14 11:53:04,782 epoch 6 - iter 360/723 - loss 0.02953778 - time (sec): 30.39 - samples/sec: 2945.44 - lr: 0.000015 - momentum: 0.000000
155
+ 2023-10-14 11:53:10,905 epoch 6 - iter 432/723 - loss 0.02804276 - time (sec): 36.51 - samples/sec: 2926.53 - lr: 0.000015 - momentum: 0.000000
156
+ 2023-10-14 11:53:16,320 epoch 6 - iter 504/723 - loss 0.02683702 - time (sec): 41.92 - samples/sec: 2948.93 - lr: 0.000014 - momentum: 0.000000
157
+ 2023-10-14 11:53:22,244 epoch 6 - iter 576/723 - loss 0.02562191 - time (sec): 47.85 - samples/sec: 2942.84 - lr: 0.000014 - momentum: 0.000000
158
+ 2023-10-14 11:53:27,883 epoch 6 - iter 648/723 - loss 0.02530607 - time (sec): 53.49 - samples/sec: 2958.77 - lr: 0.000014 - momentum: 0.000000
159
+ 2023-10-14 11:53:33,702 epoch 6 - iter 720/723 - loss 0.02457646 - time (sec): 59.31 - samples/sec: 2962.58 - lr: 0.000013 - momentum: 0.000000
160
+ 2023-10-14 11:53:33,918 ----------------------------------------------------------------------------------------------------
161
+ 2023-10-14 11:53:33,919 EPOCH 6 done: loss 0.0246 - lr: 0.000013
162
+ 2023-10-14 11:53:38,487 DEV : loss 0.13894404470920563 - f1-score (micro avg) 0.8147
163
+ 2023-10-14 11:53:38,512 saving best model
164
+ 2023-10-14 11:53:39,053 ----------------------------------------------------------------------------------------------------
165
+ 2023-10-14 11:53:45,100 epoch 7 - iter 72/723 - loss 0.01514089 - time (sec): 6.04 - samples/sec: 2816.37 - lr: 0.000013 - momentum: 0.000000
166
+ 2023-10-14 11:53:51,034 epoch 7 - iter 144/723 - loss 0.01504761 - time (sec): 11.98 - samples/sec: 2857.79 - lr: 0.000013 - momentum: 0.000000
167
+ 2023-10-14 11:53:57,277 epoch 7 - iter 216/723 - loss 0.01712868 - time (sec): 18.22 - samples/sec: 2888.95 - lr: 0.000012 - momentum: 0.000000
168
+ 2023-10-14 11:54:03,341 epoch 7 - iter 288/723 - loss 0.01831387 - time (sec): 24.29 - samples/sec: 2883.50 - lr: 0.000012 - momentum: 0.000000
169
+ 2023-10-14 11:54:09,339 epoch 7 - iter 360/723 - loss 0.01652720 - time (sec): 30.28 - samples/sec: 2908.98 - lr: 0.000012 - momentum: 0.000000
170
+ 2023-10-14 11:54:15,105 epoch 7 - iter 432/723 - loss 0.01612436 - time (sec): 36.05 - samples/sec: 2923.73 - lr: 0.000011 - momentum: 0.000000
171
+ 2023-10-14 11:54:21,245 epoch 7 - iter 504/723 - loss 0.01609187 - time (sec): 42.19 - samples/sec: 2914.96 - lr: 0.000011 - momentum: 0.000000
172
+ 2023-10-14 11:54:27,377 epoch 7 - iter 576/723 - loss 0.01654164 - time (sec): 48.32 - samples/sec: 2910.10 - lr: 0.000011 - momentum: 0.000000
173
+ 2023-10-14 11:54:33,068 epoch 7 - iter 648/723 - loss 0.01719631 - time (sec): 54.01 - samples/sec: 2909.29 - lr: 0.000010 - momentum: 0.000000
174
+ 2023-10-14 11:54:39,697 epoch 7 - iter 720/723 - loss 0.01721697 - time (sec): 60.64 - samples/sec: 2897.14 - lr: 0.000010 - momentum: 0.000000
175
+ 2023-10-14 11:54:39,934 ----------------------------------------------------------------------------------------------------
176
+ 2023-10-14 11:54:39,934 EPOCH 7 done: loss 0.0172 - lr: 0.000010
177
+ 2023-10-14 11:54:43,657 DEV : loss 0.16778483986854553 - f1-score (micro avg) 0.8077
178
+ 2023-10-14 11:54:43,680 ----------------------------------------------------------------------------------------------------
179
+ 2023-10-14 11:54:50,765 epoch 8 - iter 72/723 - loss 0.01257707 - time (sec): 7.08 - samples/sec: 2514.43 - lr: 0.000010 - momentum: 0.000000
180
+ 2023-10-14 11:54:56,831 epoch 8 - iter 144/723 - loss 0.01317721 - time (sec): 13.15 - samples/sec: 2673.46 - lr: 0.000009 - momentum: 0.000000
181
+ 2023-10-14 11:55:03,315 epoch 8 - iter 216/723 - loss 0.01362361 - time (sec): 19.63 - samples/sec: 2712.80 - lr: 0.000009 - momentum: 0.000000
182
+ 2023-10-14 11:55:09,075 epoch 8 - iter 288/723 - loss 0.01440453 - time (sec): 25.39 - samples/sec: 2780.28 - lr: 0.000009 - momentum: 0.000000
183
+ 2023-10-14 11:55:15,429 epoch 8 - iter 360/723 - loss 0.01453039 - time (sec): 31.75 - samples/sec: 2789.63 - lr: 0.000008 - momentum: 0.000000
184
+ 2023-10-14 11:55:21,249 epoch 8 - iter 432/723 - loss 0.01405394 - time (sec): 37.57 - samples/sec: 2820.91 - lr: 0.000008 - momentum: 0.000000
185
+ 2023-10-14 11:55:27,058 epoch 8 - iter 504/723 - loss 0.01345119 - time (sec): 43.38 - samples/sec: 2820.92 - lr: 0.000008 - momentum: 0.000000
186
+ 2023-10-14 11:55:33,497 epoch 8 - iter 576/723 - loss 0.01275603 - time (sec): 49.82 - samples/sec: 2817.53 - lr: 0.000007 - momentum: 0.000000
187
+ 2023-10-14 11:55:39,761 epoch 8 - iter 648/723 - loss 0.01366196 - time (sec): 56.08 - samples/sec: 2826.80 - lr: 0.000007 - momentum: 0.000000
188
+ 2023-10-14 11:55:45,544 epoch 8 - iter 720/723 - loss 0.01335513 - time (sec): 61.86 - samples/sec: 2836.20 - lr: 0.000007 - momentum: 0.000000
189
+ 2023-10-14 11:55:45,852 ----------------------------------------------------------------------------------------------------
190
+ 2023-10-14 11:55:45,852 EPOCH 8 done: loss 0.0133 - lr: 0.000007
191
+ 2023-10-14 11:55:49,362 DEV : loss 0.18129222095012665 - f1-score (micro avg) 0.8109
192
+ 2023-10-14 11:55:49,380 ----------------------------------------------------------------------------------------------------
193
+ 2023-10-14 11:55:55,373 epoch 9 - iter 72/723 - loss 0.00469641 - time (sec): 5.99 - samples/sec: 2909.93 - lr: 0.000006 - momentum: 0.000000
194
+ 2023-10-14 11:56:00,990 epoch 9 - iter 144/723 - loss 0.00558362 - time (sec): 11.61 - samples/sec: 2883.12 - lr: 0.000006 - momentum: 0.000000
195
+ 2023-10-14 11:56:07,785 epoch 9 - iter 216/723 - loss 0.00934811 - time (sec): 18.40 - samples/sec: 2914.61 - lr: 0.000006 - momentum: 0.000000
196
+ 2023-10-14 11:56:13,316 epoch 9 - iter 288/723 - loss 0.00967709 - time (sec): 23.93 - samples/sec: 2938.46 - lr: 0.000005 - momentum: 0.000000
197
+ 2023-10-14 11:56:19,287 epoch 9 - iter 360/723 - loss 0.00972270 - time (sec): 29.91 - samples/sec: 2955.84 - lr: 0.000005 - momentum: 0.000000
198
+ 2023-10-14 11:56:25,175 epoch 9 - iter 432/723 - loss 0.00996302 - time (sec): 35.79 - samples/sec: 2957.64 - lr: 0.000005 - momentum: 0.000000
199
+ 2023-10-14 11:56:31,060 epoch 9 - iter 504/723 - loss 0.00950339 - time (sec): 41.68 - samples/sec: 2959.25 - lr: 0.000004 - momentum: 0.000000
200
+ 2023-10-14 11:56:37,078 epoch 9 - iter 576/723 - loss 0.01010530 - time (sec): 47.70 - samples/sec: 2943.69 - lr: 0.000004 - momentum: 0.000000
201
+ 2023-10-14 11:56:42,906 epoch 9 - iter 648/723 - loss 0.01002914 - time (sec): 53.52 - samples/sec: 2946.96 - lr: 0.000004 - momentum: 0.000000
202
+ 2023-10-14 11:56:48,906 epoch 9 - iter 720/723 - loss 0.01036201 - time (sec): 59.52 - samples/sec: 2948.08 - lr: 0.000003 - momentum: 0.000000
203
+ 2023-10-14 11:56:49,182 ----------------------------------------------------------------------------------------------------
204
+ 2023-10-14 11:56:49,182 EPOCH 9 done: loss 0.0103 - lr: 0.000003
205
+ 2023-10-14 11:56:53,096 DEV : loss 0.1801426112651825 - f1-score (micro avg) 0.8155
206
+ 2023-10-14 11:56:53,111 saving best model
207
+ 2023-10-14 11:56:53,771 ----------------------------------------------------------------------------------------------------
208
+ 2023-10-14 11:57:00,146 epoch 10 - iter 72/723 - loss 0.00771571 - time (sec): 6.37 - samples/sec: 2841.33 - lr: 0.000003 - momentum: 0.000000
209
+ 2023-10-14 11:57:05,943 epoch 10 - iter 144/723 - loss 0.00799529 - time (sec): 12.17 - samples/sec: 2929.37 - lr: 0.000003 - momentum: 0.000000
210
+ 2023-10-14 11:57:12,088 epoch 10 - iter 216/723 - loss 0.00964900 - time (sec): 18.31 - samples/sec: 2909.61 - lr: 0.000002 - momentum: 0.000000
211
+ 2023-10-14 11:57:18,469 epoch 10 - iter 288/723 - loss 0.00910715 - time (sec): 24.69 - samples/sec: 2908.74 - lr: 0.000002 - momentum: 0.000000
212
+ 2023-10-14 11:57:24,261 epoch 10 - iter 360/723 - loss 0.00785801 - time (sec): 30.49 - samples/sec: 2927.64 - lr: 0.000002 - momentum: 0.000000
213
+ 2023-10-14 11:57:29,982 epoch 10 - iter 432/723 - loss 0.00782112 - time (sec): 36.21 - samples/sec: 2949.53 - lr: 0.000001 - momentum: 0.000000
214
+ 2023-10-14 11:57:36,136 epoch 10 - iter 504/723 - loss 0.00784393 - time (sec): 42.36 - samples/sec: 2934.43 - lr: 0.000001 - momentum: 0.000000
215
+ 2023-10-14 11:57:41,890 epoch 10 - iter 576/723 - loss 0.00793032 - time (sec): 48.12 - samples/sec: 2940.39 - lr: 0.000001 - momentum: 0.000000
216
+ 2023-10-14 11:57:47,590 epoch 10 - iter 648/723 - loss 0.00792955 - time (sec): 53.82 - samples/sec: 2936.65 - lr: 0.000000 - momentum: 0.000000
217
+ 2023-10-14 11:57:53,365 epoch 10 - iter 720/723 - loss 0.00802050 - time (sec): 59.59 - samples/sec: 2946.14 - lr: 0.000000 - momentum: 0.000000
218
+ 2023-10-14 11:57:53,666 ----------------------------------------------------------------------------------------------------
219
+ 2023-10-14 11:57:53,667 EPOCH 10 done: loss 0.0080 - lr: 0.000000
220
+ 2023-10-14 11:57:57,150 DEV : loss 0.1832604557275772 - f1-score (micro avg) 0.8127
221
+ 2023-10-14 11:57:57,618 ----------------------------------------------------------------------------------------------------
222
+ 2023-10-14 11:57:57,619 Loading model from best epoch ...
223
+ 2023-10-14 11:57:59,232 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
224
+ 2023-10-14 11:58:02,393
225
+ Results:
226
+ - F-score (micro) 0.8004
227
+ - F-score (macro) 0.6962
228
+ - Accuracy 0.6799
229
+
230
+ By class:
231
+ precision recall f1-score support
232
+
233
+ PER 0.8323 0.8237 0.8279 482
234
+ LOC 0.8741 0.7882 0.8289 458
235
+ ORG 0.4286 0.4348 0.4317 69
236
+
237
+ micro avg 0.8208 0.7810 0.8004 1009
238
+ macro avg 0.7116 0.6822 0.6962 1009
239
+ weighted avg 0.8237 0.7810 0.8013 1009
240
+
241
+ 2023-10-14 11:58:02,393 ----------------------------------------------------------------------------------------------------