Hugging Face's logo

stefan-it
/
hmbench-newseye-fr-hmbert_64k-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1

Token Classification
Flair
PyTorch
TensorBoard
French
sequence-tagger-model
Model card Files Metrics Community
stefan-it commited on
Commit
62185ba
·
1 Parent(s): 03936c9

Upload ./training.log with huggingface_hub

Browse files
Files changed (1) hide show
  1. training.log +245 -0
training.log ADDED
@@ -0,0 +1,245 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-25 14:50:02,212 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-25 14:50:02,212 Model: "SequenceTagger(
3
+ (embeddings): TransformerWordEmbeddings(
4
+ (model): BertModel(
5
+ (embeddings): BertEmbeddings(
6
+ (word_embeddings): Embedding(64001, 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=17, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-25 14:50:02,213 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-25 14:50:02,213 MultiCorpus: 7142 train + 698 dev + 2570 test sentences
52
+ - NER_HIPE_2022 Corpus: 7142 train + 698 dev + 2570 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fr/with_doc_seperator
53
+ 2023-10-25 14:50:02,213 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-25 14:50:02,213 Train: 7142 sentences
55
+ 2023-10-25 14:50:02,213 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-25 14:50:02,213 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-25 14:50:02,213 Training Params:
58
+ 2023-10-25 14:50:02,213 - learning_rate: "5e-05"
59
+ 2023-10-25 14:50:02,213 - mini_batch_size: "4"
60
+ 2023-10-25 14:50:02,213 - max_epochs: "10"
61
+ 2023-10-25 14:50:02,213 - shuffle: "True"
62
+ 2023-10-25 14:50:02,213 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-25 14:50:02,213 Plugins:
64
+ 2023-10-25 14:50:02,213 - TensorboardLogger
65
+ 2023-10-25 14:50:02,213 - LinearScheduler | warmup_fraction: '0.1'
66
+ 2023-10-25 14:50:02,213 ----------------------------------------------------------------------------------------------------
67
+ 2023-10-25 14:50:02,213 Final evaluation on model from best epoch (best-model.pt)
68
+ 2023-10-25 14:50:02,214 - metric: "('micro avg', 'f1-score')"
69
+ 2023-10-25 14:50:02,214 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-25 14:50:02,214 Computation:
71
+ 2023-10-25 14:50:02,214 - compute on device: cuda:0
72
+ 2023-10-25 14:50:02,214 - embedding storage: none
73
+ 2023-10-25 14:50:02,214 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-25 14:50:02,214 Model training base path: "hmbench-newseye/fr-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1"
75
+ 2023-10-25 14:50:02,214 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-25 14:50:02,214 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-25 14:50:02,214 Logging anything other than scalars to TensorBoard is currently not supported.
78
+ 2023-10-25 14:50:11,883 epoch 1 - iter 178/1786 - loss 1.84783869 - time (sec): 9.67 - samples/sec: 2583.61 - lr: 0.000005 - momentum: 0.000000
79
+ 2023-10-25 14:50:21,608 epoch 1 - iter 356/1786 - loss 1.11625612 - time (sec): 19.39 - samples/sec: 2603.59 - lr: 0.000010 - momentum: 0.000000
80
+ 2023-10-25 14:50:31,390 epoch 1 - iter 534/1786 - loss 0.83906654 - time (sec): 29.18 - samples/sec: 2568.92 - lr: 0.000015 - momentum: 0.000000
81
+ 2023-10-25 14:50:40,838 epoch 1 - iter 712/1786 - loss 0.68032105 - time (sec): 38.62 - samples/sec: 2597.39 - lr: 0.000020 - momentum: 0.000000
82
+ 2023-10-25 14:50:50,185 epoch 1 - iter 890/1786 - loss 0.58377808 - time (sec): 47.97 - samples/sec: 2608.11 - lr: 0.000025 - momentum: 0.000000
83
+ 2023-10-25 14:50:59,347 epoch 1 - iter 1068/1786 - loss 0.51796911 - time (sec): 57.13 - samples/sec: 2609.43 - lr: 0.000030 - momentum: 0.000000
84
+ 2023-10-25 14:51:08,975 epoch 1 - iter 1246/1786 - loss 0.46226937 - time (sec): 66.76 - samples/sec: 2625.70 - lr: 0.000035 - momentum: 0.000000
85
+ 2023-10-25 14:51:18,683 epoch 1 - iter 1424/1786 - loss 0.42498520 - time (sec): 76.47 - samples/sec: 2601.98 - lr: 0.000040 - momentum: 0.000000
86
+ 2023-10-25 14:51:27,888 epoch 1 - iter 1602/1786 - loss 0.39393828 - time (sec): 85.67 - samples/sec: 2609.03 - lr: 0.000045 - momentum: 0.000000
87
+ 2023-10-25 14:51:37,261 epoch 1 - iter 1780/1786 - loss 0.37225174 - time (sec): 95.05 - samples/sec: 2607.72 - lr: 0.000050 - momentum: 0.000000
88
+ 2023-10-25 14:51:37,609 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-25 14:51:37,610 EPOCH 1 done: loss 0.3712 - lr: 0.000050
90
+ 2023-10-25 14:51:41,345 DEV : loss 0.12383320182561874 - f1-score (micro avg) 0.692
91
+ 2023-10-25 14:51:41,367 saving best model
92
+ 2023-10-25 14:51:41,834 ----------------------------------------------------------------------------------------------------
93
+ 2023-10-25 14:51:50,898 epoch 2 - iter 178/1786 - loss 0.11440561 - time (sec): 9.06 - samples/sec: 2722.95 - lr: 0.000049 - momentum: 0.000000
94
+ 2023-10-25 14:51:59,829 epoch 2 - iter 356/1786 - loss 0.12055337 - time (sec): 17.99 - samples/sec: 2748.37 - lr: 0.000049 - momentum: 0.000000
95
+ 2023-10-25 14:52:09,589 epoch 2 - iter 534/1786 - loss 0.12495791 - time (sec): 27.75 - samples/sec: 2647.01 - lr: 0.000048 - momentum: 0.000000
96
+ 2023-10-25 14:52:19,211 epoch 2 - iter 712/1786 - loss 0.12559829 - time (sec): 37.38 - samples/sec: 2624.13 - lr: 0.000048 - momentum: 0.000000
97
+ 2023-10-25 14:52:28,850 epoch 2 - iter 890/1786 - loss 0.11968865 - time (sec): 47.01 - samples/sec: 2603.47 - lr: 0.000047 - momentum: 0.000000
98
+ 2023-10-25 14:52:38,347 epoch 2 - iter 1068/1786 - loss 0.12384761 - time (sec): 56.51 - samples/sec: 2593.27 - lr: 0.000047 - momentum: 0.000000
99
+ 2023-10-25 14:52:47,363 epoch 2 - iter 1246/1786 - loss 0.12564491 - time (sec): 65.53 - samples/sec: 2626.45 - lr: 0.000046 - momentum: 0.000000
100
+ 2023-10-25 14:52:56,330 epoch 2 - iter 1424/1786 - loss 0.12415748 - time (sec): 74.50 - samples/sec: 2659.00 - lr: 0.000046 - momentum: 0.000000
101
+ 2023-10-25 14:53:05,248 epoch 2 - iter 1602/1786 - loss 0.12176852 - time (sec): 83.41 - samples/sec: 2685.88 - lr: 0.000045 - momentum: 0.000000
102
+ 2023-10-25 14:53:14,581 epoch 2 - iter 1780/1786 - loss 0.12143837 - time (sec): 92.75 - samples/sec: 2672.87 - lr: 0.000044 - momentum: 0.000000
103
+ 2023-10-25 14:53:14,882 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-25 14:53:14,883 EPOCH 2 done: loss 0.1213 - lr: 0.000044
105
+ 2023-10-25 14:53:20,002 DEV : loss 0.15896384418010712 - f1-score (micro avg) 0.7482
106
+ 2023-10-25 14:53:20,025 saving best model
107
+ 2023-10-25 14:53:20,700 ----------------------------------------------------------------------------------------------------
108
+ 2023-10-25 14:53:29,708 epoch 3 - iter 178/1786 - loss 0.07747975 - time (sec): 9.01 - samples/sec: 2728.60 - lr: 0.000044 - momentum: 0.000000
109
+ 2023-10-25 14:53:39,305 epoch 3 - iter 356/1786 - loss 0.08659897 - time (sec): 18.60 - samples/sec: 2753.37 - lr: 0.000043 - momentum: 0.000000
110
+ 2023-10-25 14:53:48,734 epoch 3 - iter 534/1786 - loss 0.08333401 - time (sec): 28.03 - samples/sec: 2690.93 - lr: 0.000043 - momentum: 0.000000
111
+ 2023-10-25 14:53:58,266 epoch 3 - iter 712/1786 - loss 0.08421054 - time (sec): 37.56 - samples/sec: 2649.55 - lr: 0.000042 - momentum: 0.000000
112
+ 2023-10-25 14:54:07,433 epoch 3 - iter 890/1786 - loss 0.08466478 - time (sec): 46.73 - samples/sec: 2634.36 - lr: 0.000042 - momentum: 0.000000
113
+ 2023-10-25 14:54:16,555 epoch 3 - iter 1068/1786 - loss 0.08490863 - time (sec): 55.85 - samples/sec: 2649.10 - lr: 0.000041 - momentum: 0.000000
114
+ 2023-10-25 14:54:25,902 epoch 3 - iter 1246/1786 - loss 0.08422222 - time (sec): 65.20 - samples/sec: 2663.92 - lr: 0.000041 - momentum: 0.000000
115
+ 2023-10-25 14:54:34,984 epoch 3 - iter 1424/1786 - loss 0.08498648 - time (sec): 74.28 - samples/sec: 2675.02 - lr: 0.000040 - momentum: 0.000000
116
+ 2023-10-25 14:54:44,185 epoch 3 - iter 1602/1786 - loss 0.08479702 - time (sec): 83.48 - samples/sec: 2669.15 - lr: 0.000039 - momentum: 0.000000
117
+ 2023-10-25 14:54:53,654 epoch 3 - iter 1780/1786 - loss 0.08642108 - time (sec): 92.95 - samples/sec: 2669.53 - lr: 0.000039 - momentum: 0.000000
118
+ 2023-10-25 14:54:53,978 ----------------------------------------------------------------------------------------------------
119
+ 2023-10-25 14:54:53,978 EPOCH 3 done: loss 0.0865 - lr: 0.000039
120
+ 2023-10-25 14:54:57,810 DEV : loss 0.13499563932418823 - f1-score (micro avg) 0.7639
121
+ 2023-10-25 14:54:57,834 saving best model
122
+ 2023-10-25 14:54:58,513 ----------------------------------------------------------------------------------------------------
123
+ 2023-10-25 14:55:08,213 epoch 4 - iter 178/1786 - loss 0.07787612 - time (sec): 9.70 - samples/sec: 2662.18 - lr: 0.000038 - momentum: 0.000000
124
+ 2023-10-25 14:55:17,762 epoch 4 - iter 356/1786 - loss 0.07006565 - time (sec): 19.25 - samples/sec: 2597.01 - lr: 0.000038 - momentum: 0.000000
125
+ 2023-10-25 14:55:27,282 epoch 4 - iter 534/1786 - loss 0.06777304 - time (sec): 28.77 - samples/sec: 2561.15 - lr: 0.000037 - momentum: 0.000000
126
+ 2023-10-25 14:55:36,832 epoch 4 - iter 712/1786 - loss 0.06334353 - time (sec): 38.32 - samples/sec: 2605.60 - lr: 0.000037 - momentum: 0.000000
127
+ 2023-10-25 14:55:46,207 epoch 4 - iter 890/1786 - loss 0.06251813 - time (sec): 47.69 - samples/sec: 2628.71 - lr: 0.000036 - momentum: 0.000000
128
+ 2023-10-25 14:55:55,932 epoch 4 - iter 1068/1786 - loss 0.06251651 - time (sec): 57.42 - samples/sec: 2599.12 - lr: 0.000036 - momentum: 0.000000
129
+ 2023-10-25 14:56:05,686 epoch 4 - iter 1246/1786 - loss 0.06255099 - time (sec): 67.17 - samples/sec: 2585.14 - lr: 0.000035 - momentum: 0.000000
130
+ 2023-10-25 14:56:15,297 epoch 4 - iter 1424/1786 - loss 0.06202994 - time (sec): 76.78 - samples/sec: 2582.61 - lr: 0.000034 - momentum: 0.000000
131
+ 2023-10-25 14:56:25,232 epoch 4 - iter 1602/1786 - loss 0.06194081 - time (sec): 86.72 - samples/sec: 2583.68 - lr: 0.000034 - momentum: 0.000000
132
+ 2023-10-25 14:56:34,873 epoch 4 - iter 1780/1786 - loss 0.06270639 - time (sec): 96.36 - samples/sec: 2573.68 - lr: 0.000033 - momentum: 0.000000
133
+ 2023-10-25 14:56:35,198 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-25 14:56:35,198 EPOCH 4 done: loss 0.0628 - lr: 0.000033
135
+ 2023-10-25 14:56:39,818 DEV : loss 0.18497972190380096 - f1-score (micro avg) 0.7612
136
+ 2023-10-25 14:56:39,839 ----------------------------------------------------------------------------------------------------
137
+ 2023-10-25 14:56:49,615 epoch 5 - iter 178/1786 - loss 0.05664153 - time (sec): 9.77 - samples/sec: 2702.46 - lr: 0.000033 - momentum: 0.000000
138
+ 2023-10-25 14:56:59,207 epoch 5 - iter 356/1786 - loss 0.05522861 - time (sec): 19.37 - samples/sec: 2660.03 - lr: 0.000032 - momentum: 0.000000
139
+ 2023-10-25 14:57:09,051 epoch 5 - iter 534/1786 - loss 0.05116909 - time (sec): 29.21 - samples/sec: 2614.02 - lr: 0.000032 - momentum: 0.000000
140
+ 2023-10-25 14:57:18,747 epoch 5 - iter 712/1786 - loss 0.05214688 - time (sec): 38.91 - samples/sec: 2573.04 - lr: 0.000031 - momentum: 0.000000
141
+ 2023-10-25 14:57:28,444 epoch 5 - iter 890/1786 - loss 0.04958778 - time (sec): 48.60 - samples/sec: 2534.05 - lr: 0.000031 - momentum: 0.000000
142
+ 2023-10-25 14:57:38,250 epoch 5 - iter 1068/1786 - loss 0.04777163 - time (sec): 58.41 - samples/sec: 2521.25 - lr: 0.000030 - momentum: 0.000000
143
+ 2023-10-25 14:57:48,040 epoch 5 - iter 1246/1786 - loss 0.04818385 - time (sec): 68.20 - samples/sec: 2496.31 - lr: 0.000029 - momentum: 0.000000
144
+ 2023-10-25 14:57:57,629 epoch 5 - iter 1424/1786 - loss 0.04857465 - time (sec): 77.79 - samples/sec: 2538.86 - lr: 0.000029 - momentum: 0.000000
145
+ 2023-10-25 14:58:07,122 epoch 5 - iter 1602/1786 - loss 0.04773650 - time (sec): 87.28 - samples/sec: 2557.13 - lr: 0.000028 - momentum: 0.000000
146
+ 2023-10-25 14:58:16,910 epoch 5 - iter 1780/1786 - loss 0.04709670 - time (sec): 97.07 - samples/sec: 2554.87 - lr: 0.000028 - momentum: 0.000000
147
+ 2023-10-25 14:58:17,232 ----------------------------------------------------------------------------------------------------
148
+ 2023-10-25 14:58:17,233 EPOCH 5 done: loss 0.0470 - lr: 0.000028
149
+ 2023-10-25 14:58:21,821 DEV : loss 0.19018808007240295 - f1-score (micro avg) 0.7738
150
+ 2023-10-25 14:58:21,845 saving best model
151
+ 2023-10-25 14:58:24,127 ----------------------------------------------------------------------------------------------------
152
+ 2023-10-25 14:58:33,822 epoch 6 - iter 178/1786 - loss 0.02826424 - time (sec): 9.69 - samples/sec: 2757.71 - lr: 0.000027 - momentum: 0.000000
153
+ 2023-10-25 14:58:43,468 epoch 6 - iter 356/1786 - loss 0.03138743 - time (sec): 19.34 - samples/sec: 2673.84 - lr: 0.000027 - momentum: 0.000000
154
+ 2023-10-25 14:58:52,936 epoch 6 - iter 534/1786 - loss 0.03399004 - time (sec): 28.81 - samples/sec: 2617.71 - lr: 0.000026 - momentum: 0.000000
155
+ 2023-10-25 14:59:02,446 epoch 6 - iter 712/1786 - loss 0.03560434 - time (sec): 38.32 - samples/sec: 2650.36 - lr: 0.000026 - momentum: 0.000000
156
+ 2023-10-25 14:59:11,997 epoch 6 - iter 890/1786 - loss 0.03582739 - time (sec): 47.87 - samples/sec: 2636.12 - lr: 0.000025 - momentum: 0.000000
157
+ 2023-10-25 14:59:21,474 epoch 6 - iter 1068/1786 - loss 0.03714910 - time (sec): 57.35 - samples/sec: 2634.03 - lr: 0.000024 - momentum: 0.000000
158
+ 2023-10-25 14:59:30,970 epoch 6 - iter 1246/1786 - loss 0.03684768 - time (sec): 66.84 - samples/sec: 2621.31 - lr: 0.000024 - momentum: 0.000000
159
+ 2023-10-25 14:59:40,363 epoch 6 - iter 1424/1786 - loss 0.03615725 - time (sec): 76.23 - samples/sec: 2618.38 - lr: 0.000023 - momentum: 0.000000
160
+ 2023-10-25 14:59:49,929 epoch 6 - iter 1602/1786 - loss 0.03647106 - time (sec): 85.80 - samples/sec: 2621.52 - lr: 0.000023 - momentum: 0.000000
161
+ 2023-10-25 14:59:59,457 epoch 6 - iter 1780/1786 - loss 0.03647415 - time (sec): 95.33 - samples/sec: 2601.49 - lr: 0.000022 - momentum: 0.000000
162
+ 2023-10-25 14:59:59,775 ----------------------------------------------------------------------------------------------------
163
+ 2023-10-25 14:59:59,775 EPOCH 6 done: loss 0.0364 - lr: 0.000022
164
+ 2023-10-25 15:00:03,677 DEV : loss 0.20759737491607666 - f1-score (micro avg) 0.7809
165
+ 2023-10-25 15:00:03,700 saving best model
166
+ 2023-10-25 15:00:04,371 ----------------------------------------------------------------------------------------------------
167
+ 2023-10-25 15:00:13,891 epoch 7 - iter 178/1786 - loss 0.02523091 - time (sec): 9.52 - samples/sec: 2414.94 - lr: 0.000022 - momentum: 0.000000
168
+ 2023-10-25 15:00:23,459 epoch 7 - iter 356/1786 - loss 0.02954901 - time (sec): 19.08 - samples/sec: 2508.47 - lr: 0.000021 - momentum: 0.000000
169
+ 2023-10-25 15:00:33,286 epoch 7 - iter 534/1786 - loss 0.02745032 - time (sec): 28.91 - samples/sec: 2616.19 - lr: 0.000021 - momentum: 0.000000
170
+ 2023-10-25 15:00:43,077 epoch 7 - iter 712/1786 - loss 0.02791815 - time (sec): 38.70 - samples/sec: 2635.23 - lr: 0.000020 - momentum: 0.000000
171
+ 2023-10-25 15:00:52,413 epoch 7 - iter 890/1786 - loss 0.02773715 - time (sec): 48.04 - samples/sec: 2590.73 - lr: 0.000019 - momentum: 0.000000
172
+ 2023-10-25 15:01:01,227 epoch 7 - iter 1068/1786 - loss 0.02749024 - time (sec): 56.85 - samples/sec: 2611.10 - lr: 0.000019 - momentum: 0.000000
173
+ 2023-10-25 15:01:09,845 epoch 7 - iter 1246/1786 - loss 0.02786391 - time (sec): 65.47 - samples/sec: 2648.12 - lr: 0.000018 - momentum: 0.000000
174
+ 2023-10-25 15:01:18,505 epoch 7 - iter 1424/1786 - loss 0.02768994 - time (sec): 74.13 - samples/sec: 2674.88 - lr: 0.000018 - momentum: 0.000000
175
+ 2023-10-25 15:01:27,450 epoch 7 - iter 1602/1786 - loss 0.02880790 - time (sec): 83.08 - samples/sec: 2677.87 - lr: 0.000017 - momentum: 0.000000
176
+ 2023-10-25 15:01:36,479 epoch 7 - iter 1780/1786 - loss 0.02861625 - time (sec): 92.10 - samples/sec: 2693.30 - lr: 0.000017 - momentum: 0.000000
177
+ 2023-10-25 15:01:36,780 ----------------------------------------------------------------------------------------------------
178
+ 2023-10-25 15:01:36,781 EPOCH 7 done: loss 0.0287 - lr: 0.000017
179
+ 2023-10-25 15:01:41,727 DEV : loss 0.210079625248909 - f1-score (micro avg) 0.7967
180
+ 2023-10-25 15:01:41,752 saving best model
181
+ 2023-10-25 15:01:42,460 ----------------------------------------------------------------------------------------------------
182
+ 2023-10-25 15:01:51,741 epoch 8 - iter 178/1786 - loss 0.01706992 - time (sec): 9.28 - samples/sec: 2667.12 - lr: 0.000016 - momentum: 0.000000
183
+ 2023-10-25 15:02:00,894 epoch 8 - iter 356/1786 - loss 0.01935111 - time (sec): 18.43 - samples/sec: 2782.76 - lr: 0.000016 - momentum: 0.000000
184
+ 2023-10-25 15:02:10,126 epoch 8 - iter 534/1786 - loss 0.01874463 - time (sec): 27.66 - samples/sec: 2787.37 - lr: 0.000015 - momentum: 0.000000
185
+ 2023-10-25 15:02:19,255 epoch 8 - iter 712/1786 - loss 0.01879846 - time (sec): 36.79 - samples/sec: 2798.76 - lr: 0.000014 - momentum: 0.000000
186
+ 2023-10-25 15:02:28,045 epoch 8 - iter 890/1786 - loss 0.01977016 - time (sec): 45.58 - samples/sec: 2769.63 - lr: 0.000014 - momentum: 0.000000
187
+ 2023-10-25 15:02:37,385 epoch 8 - iter 1068/1786 - loss 0.02060995 - time (sec): 54.92 - samples/sec: 2734.37 - lr: 0.000013 - momentum: 0.000000
188
+ 2023-10-25 15:02:46,579 epoch 8 - iter 1246/1786 - loss 0.01992327 - time (sec): 64.12 - samples/sec: 2729.82 - lr: 0.000013 - momentum: 0.000000
189
+ 2023-10-25 15:02:55,628 epoch 8 - iter 1424/1786 - loss 0.01930754 - time (sec): 73.17 - samples/sec: 2730.50 - lr: 0.000012 - momentum: 0.000000
190
+ 2023-10-25 15:03:04,526 epoch 8 - iter 1602/1786 - loss 0.02003035 - time (sec): 82.06 - samples/sec: 2737.89 - lr: 0.000012 - momentum: 0.000000
191
+ 2023-10-25 15:03:13,732 epoch 8 - iter 1780/1786 - loss 0.01945620 - time (sec): 91.27 - samples/sec: 2717.14 - lr: 0.000011 - momentum: 0.000000
192
+ 2023-10-25 15:03:14,036 ----------------------------------------------------------------------------------------------------
193
+ 2023-10-25 15:03:14,037 EPOCH 8 done: loss 0.0194 - lr: 0.000011
194
+ 2023-10-25 15:03:17,868 DEV : loss 0.24006861448287964 - f1-score (micro avg) 0.8065
195
+ 2023-10-25 15:03:17,891 saving best model
196
+ 2023-10-25 15:03:18,584 ----------------------------------------------------------------------------------------------------
197
+ 2023-10-25 15:03:28,381 epoch 9 - iter 178/1786 - loss 0.01966686 - time (sec): 9.79 - samples/sec: 2497.75 - lr: 0.000011 - momentum: 0.000000
198
+ 2023-10-25 15:03:38,805 epoch 9 - iter 356/1786 - loss 0.01914487 - time (sec): 20.22 - samples/sec: 2486.87 - lr: 0.000010 - momentum: 0.000000
199
+ 2023-10-25 15:03:48,480 epoch 9 - iter 534/1786 - loss 0.01935434 - time (sec): 29.89 - samples/sec: 2548.37 - lr: 0.000009 - momentum: 0.000000
200
+ 2023-10-25 15:03:58,187 epoch 9 - iter 712/1786 - loss 0.02027442 - time (sec): 39.60 - samples/sec: 2527.51 - lr: 0.000009 - momentum: 0.000000
201
+ 2023-10-25 15:04:07,497 epoch 9 - iter 890/1786 - loss 0.01881800 - time (sec): 48.91 - samples/sec: 2574.08 - lr: 0.000008 - momentum: 0.000000
202
+ 2023-10-25 15:04:16,574 epoch 9 - iter 1068/1786 - loss 0.01810846 - time (sec): 57.99 - samples/sec: 2592.00 - lr: 0.000008 - momentum: 0.000000
203
+ 2023-10-25 15:04:25,884 epoch 9 - iter 1246/1786 - loss 0.01803578 - time (sec): 67.30 - samples/sec: 2607.25 - lr: 0.000007 - momentum: 0.000000
204
+ 2023-10-25 15:04:36,028 epoch 9 - iter 1424/1786 - loss 0.01811109 - time (sec): 77.44 - samples/sec: 2574.70 - lr: 0.000007 - momentum: 0.000000
205
+ 2023-10-25 15:04:45,582 epoch 9 - iter 1602/1786 - loss 0.01892874 - time (sec): 87.00 - samples/sec: 2579.09 - lr: 0.000006 - momentum: 0.000000
206
+ 2023-10-25 15:04:55,055 epoch 9 - iter 1780/1786 - loss 0.02267831 - time (sec): 96.47 - samples/sec: 2571.05 - lr: 0.000006 - momentum: 0.000000
207
+ 2023-10-25 15:04:55,366 ----------------------------------------------------------------------------------------------------
208
+ 2023-10-25 15:04:55,367 EPOCH 9 done: loss 0.0230 - lr: 0.000006
209
+ 2023-10-25 15:04:59,142 DEV : loss 0.25254154205322266 - f1-score (micro avg) 0.6079
210
+ 2023-10-25 15:04:59,166 ----------------------------------------------------------------------------------------------------
211
+ 2023-10-25 15:05:08,791 epoch 10 - iter 178/1786 - loss 0.11012665 - time (sec): 9.62 - samples/sec: 2684.65 - lr: 0.000005 - momentum: 0.000000
212
+ 2023-10-25 15:05:17,986 epoch 10 - iter 356/1786 - loss 0.07275891 - time (sec): 18.82 - samples/sec: 2728.54 - lr: 0.000004 - momentum: 0.000000
213
+ 2023-10-25 15:05:26,833 epoch 10 - iter 534/1786 - loss 0.06185256 - time (sec): 27.66 - samples/sec: 2718.84 - lr: 0.000004 - momentum: 0.000000
214
+ 2023-10-25 15:05:35,495 epoch 10 - iter 712/1786 - loss 0.06154437 - time (sec): 36.33 - samples/sec: 2766.27 - lr: 0.000003 - momentum: 0.000000
215
+ 2023-10-25 15:05:44,207 epoch 10 - iter 890/1786 - loss 0.06553764 - time (sec): 45.04 - samples/sec: 2755.02 - lr: 0.000003 - momentum: 0.000000
216
+ 2023-10-25 15:05:52,947 epoch 10 - iter 1068/1786 - loss 0.06476228 - time (sec): 53.78 - samples/sec: 2753.92 - lr: 0.000002 - momentum: 0.000000
217
+ 2023-10-25 15:06:01,545 epoch 10 - iter 1246/1786 - loss 0.06251026 - time (sec): 62.38 - samples/sec: 2780.55 - lr: 0.000002 - momentum: 0.000000
218
+ 2023-10-25 15:06:10,626 epoch 10 - iter 1424/1786 - loss 0.06067494 - time (sec): 71.46 - samples/sec: 2783.86 - lr: 0.000001 - momentum: 0.000000
219
+ 2023-10-25 15:06:19,695 epoch 10 - iter 1602/1786 - loss 0.05860883 - time (sec): 80.53 - samples/sec: 2772.55 - lr: 0.000001 - momentum: 0.000000
220
+ 2023-10-25 15:06:29,012 epoch 10 - iter 1780/1786 - loss 0.05709972 - time (sec): 89.84 - samples/sec: 2758.02 - lr: 0.000000 - momentum: 0.000000
221
+ 2023-10-25 15:06:29,337 ----------------------------------------------------------------------------------------------------
222
+ 2023-10-25 15:06:29,338 EPOCH 10 done: loss 0.0570 - lr: 0.000000
223
+ 2023-10-25 15:06:34,225 DEV : loss 0.2218979001045227 - f1-score (micro avg) 0.6822
224
+ 2023-10-25 15:06:34,730 ----------------------------------------------------------------------------------------------------
225
+ 2023-10-25 15:06:34,731 Loading model from best epoch ...
226
+ 2023-10-25 15:06:36,683 SequenceTagger predicts: Dictionary with 17 tags: O, S-PER, B-PER, E-PER, I-PER, S-LOC, B-LOC, E-LOC, I-LOC, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
227
+ 2023-10-25 15:06:48,731
228
+ Results:
229
+ - F-score (micro) 0.6719
230
+ - F-score (macro) 0.5777
231
+ - Accuracy 0.5222
232
+
233
+ By class:
234
+ precision recall f1-score support
235
+
236
+ LOC 0.6930 0.6721 0.6824 1095
237
+ PER 0.7336 0.7510 0.7422 1012
238
+ ORG 0.4452 0.5350 0.4860 357
239
+ HumanProd 0.3571 0.4545 0.4000 33
240
+
241
+ micro avg 0.6625 0.6816 0.6719 2497
242
+ macro avg 0.5572 0.6032 0.5777 2497
243
+ weighted avg 0.6696 0.6816 0.6748 2497
244
+
245
+ 2023-10-25 15:06:48,732 ----------------------------------------------------------------------------------------------------