Hugging Face's logo

hmbert-64k
/
flair-hipe-2022-newseye-de

Token Classification
Flair
PyTorch
TensorBoard
German
sequence-tagger-model
Model card Files Metrics Community
stefan-it commited on
Commit
c4499f6
•
1 Parent(s): f530d3e

Upload ./training.log with huggingface_hub

Browse files
Files changed (1) hide show
  1. training.log +242 -0
training.log ADDED
@@ -0,0 +1,242 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-25 11:39:23,291 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-25 11:39:23,292 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 11:39:23,292 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-25 11:39:23,292 MultiCorpus: 20847 train + 1123 dev + 3350 test sentences
52
+ - NER_HIPE_2022 Corpus: 20847 train + 1123 dev + 3350 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/de/with_doc_seperator
53
+ 2023-10-25 11:39:23,292 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-25 11:39:23,292 Train: 20847 sentences
55
+ 2023-10-25 11:39:23,292 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-25 11:39:23,292 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-25 11:39:23,292 Training Params:
58
+ 2023-10-25 11:39:23,292 - learning_rate: "3e-05"
59
+ 2023-10-25 11:39:23,292 - mini_batch_size: "8"
60
+ 2023-10-25 11:39:23,292 - max_epochs: "10"
61
+ 2023-10-25 11:39:23,292 - shuffle: "True"
62
+ 2023-10-25 11:39:23,292 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-25 11:39:23,292 Plugins:
64
+ 2023-10-25 11:39:23,292 - TensorboardLogger
65
+ 2023-10-25 11:39:23,292 - LinearScheduler | warmup_fraction: '0.1'
66
+ 2023-10-25 11:39:23,292 ----------------------------------------------------------------------------------------------------
67
+ 2023-10-25 11:39:23,292 Final evaluation on model from best epoch (best-model.pt)
68
+ 2023-10-25 11:39:23,293 - metric: "('micro avg', 'f1-score')"
69
+ 2023-10-25 11:39:23,293 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-25 11:39:23,293 Computation:
71
+ 2023-10-25 11:39:23,293 - compute on device: cuda:0
72
+ 2023-10-25 11:39:23,293 - embedding storage: none
73
+ 2023-10-25 11:39:23,293 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-25 11:39:23,293 Model training base path: "hmbench-newseye/de-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2"
75
+ 2023-10-25 11:39:23,293 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-25 11:39:23,293 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-25 11:39:23,293 Logging anything other than scalars to TensorBoard is currently not supported.
78
+ 2023-10-25 11:39:37,851 epoch 1 - iter 260/2606 - loss 1.57021077 - time (sec): 14.56 - samples/sec: 2556.73 - lr: 0.000003 - momentum: 0.000000
79
+ 2023-10-25 11:39:52,142 epoch 1 - iter 520/2606 - loss 0.95405161 - time (sec): 28.85 - samples/sec: 2664.38 - lr: 0.000006 - momentum: 0.000000
80
+ 2023-10-25 11:40:06,273 epoch 1 - iter 780/2606 - loss 0.73729341 - time (sec): 42.98 - samples/sec: 2655.02 - lr: 0.000009 - momentum: 0.000000
81
+ 2023-10-25 11:40:19,835 epoch 1 - iter 1040/2606 - loss 0.62532777 - time (sec): 56.54 - samples/sec: 2629.25 - lr: 0.000012 - momentum: 0.000000
82
+ 2023-10-25 11:40:33,679 epoch 1 - iter 1300/2606 - loss 0.54757678 - time (sec): 70.39 - samples/sec: 2616.46 - lr: 0.000015 - momentum: 0.000000
83
+ 2023-10-25 11:40:47,653 epoch 1 - iter 1560/2606 - loss 0.49066763 - time (sec): 84.36 - samples/sec: 2635.35 - lr: 0.000018 - momentum: 0.000000
84
+ 2023-10-25 11:41:01,175 epoch 1 - iter 1820/2606 - loss 0.45026054 - time (sec): 97.88 - samples/sec: 2620.86 - lr: 0.000021 - momentum: 0.000000
85
+ 2023-10-25 11:41:15,148 epoch 1 - iter 2080/2606 - loss 0.41942467 - time (sec): 111.85 - samples/sec: 2611.91 - lr: 0.000024 - momentum: 0.000000
86
+ 2023-10-25 11:41:29,102 epoch 1 - iter 2340/2606 - loss 0.39408020 - time (sec): 125.81 - samples/sec: 2607.23 - lr: 0.000027 - momentum: 0.000000
87
+ 2023-10-25 11:41:43,261 epoch 1 - iter 2600/2606 - loss 0.36990188 - time (sec): 139.97 - samples/sec: 2620.90 - lr: 0.000030 - momentum: 0.000000
88
+ 2023-10-25 11:41:43,543 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-25 11:41:43,544 EPOCH 1 done: loss 0.3696 - lr: 0.000030
90
+ 2023-10-25 11:41:47,256 DEV : loss 0.16545310616493225 - f1-score (micro avg) 0.3369
91
+ 2023-10-25 11:41:47,281 saving best model
92
+ 2023-10-25 11:41:47,676 ----------------------------------------------------------------------------------------------------
93
+ 2023-10-25 11:42:02,046 epoch 2 - iter 260/2606 - loss 0.16955557 - time (sec): 14.37 - samples/sec: 2689.29 - lr: 0.000030 - momentum: 0.000000
94
+ 2023-10-25 11:42:16,238 epoch 2 - iter 520/2606 - loss 0.16249568 - time (sec): 28.56 - samples/sec: 2660.42 - lr: 0.000029 - momentum: 0.000000
95
+ 2023-10-25 11:42:30,467 epoch 2 - iter 780/2606 - loss 0.16067408 - time (sec): 42.79 - samples/sec: 2662.51 - lr: 0.000029 - momentum: 0.000000
96
+ 2023-10-25 11:42:43,942 epoch 2 - iter 1040/2606 - loss 0.15460242 - time (sec): 56.26 - samples/sec: 2624.53 - lr: 0.000029 - momentum: 0.000000
97
+ 2023-10-25 11:42:57,089 epoch 2 - iter 1300/2606 - loss 0.15720241 - time (sec): 69.41 - samples/sec: 2628.78 - lr: 0.000028 - momentum: 0.000000
98
+ 2023-10-25 11:43:10,939 epoch 2 - iter 1560/2606 - loss 0.15583926 - time (sec): 83.26 - samples/sec: 2624.56 - lr: 0.000028 - momentum: 0.000000
99
+ 2023-10-25 11:43:25,093 epoch 2 - iter 1820/2606 - loss 0.15515707 - time (sec): 97.42 - samples/sec: 2626.00 - lr: 0.000028 - momentum: 0.000000
100
+ 2023-10-25 11:43:38,804 epoch 2 - iter 2080/2606 - loss 0.15504235 - time (sec): 111.13 - samples/sec: 2628.05 - lr: 0.000027 - momentum: 0.000000
101
+ 2023-10-25 11:43:52,956 epoch 2 - iter 2340/2606 - loss 0.15212685 - time (sec): 125.28 - samples/sec: 2629.04 - lr: 0.000027 - momentum: 0.000000
102
+ 2023-10-25 11:44:07,071 epoch 2 - iter 2600/2606 - loss 0.15026039 - time (sec): 139.39 - samples/sec: 2630.97 - lr: 0.000027 - momentum: 0.000000
103
+ 2023-10-25 11:44:07,353 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-25 11:44:07,354 EPOCH 2 done: loss 0.1505 - lr: 0.000027
105
+ 2023-10-25 11:44:14,184 DEV : loss 0.1197943240404129 - f1-score (micro avg) 0.3209
106
+ 2023-10-25 11:44:14,209 ----------------------------------------------------------------------------------------------------
107
+ 2023-10-25 11:44:27,824 epoch 3 - iter 260/2606 - loss 0.10481529 - time (sec): 13.61 - samples/sec: 2470.68 - lr: 0.000026 - momentum: 0.000000
108
+ 2023-10-25 11:44:41,595 epoch 3 - iter 520/2606 - loss 0.10487908 - time (sec): 27.38 - samples/sec: 2473.13 - lr: 0.000026 - momentum: 0.000000
109
+ 2023-10-25 11:44:55,771 epoch 3 - iter 780/2606 - loss 0.09618585 - time (sec): 41.56 - samples/sec: 2593.18 - lr: 0.000026 - momentum: 0.000000
110
+ 2023-10-25 11:45:09,666 epoch 3 - iter 1040/2606 - loss 0.09795426 - time (sec): 55.46 - samples/sec: 2580.25 - lr: 0.000025 - momentum: 0.000000
111
+ 2023-10-25 11:45:23,599 epoch 3 - iter 1300/2606 - loss 0.09869583 - time (sec): 69.39 - samples/sec: 2596.10 - lr: 0.000025 - momentum: 0.000000
112
+ 2023-10-25 11:45:37,541 epoch 3 - iter 1560/2606 - loss 0.09966936 - time (sec): 83.33 - samples/sec: 2604.12 - lr: 0.000025 - momentum: 0.000000
113
+ 2023-10-25 11:45:51,947 epoch 3 - iter 1820/2606 - loss 0.10055515 - time (sec): 97.74 - samples/sec: 2630.23 - lr: 0.000024 - momentum: 0.000000
114
+ 2023-10-25 11:46:05,429 epoch 3 - iter 2080/2606 - loss 0.09842991 - time (sec): 111.22 - samples/sec: 2645.66 - lr: 0.000024 - momentum: 0.000000
115
+ 2023-10-25 11:46:19,397 epoch 3 - iter 2340/2606 - loss 0.09950876 - time (sec): 125.19 - samples/sec: 2654.12 - lr: 0.000024 - momentum: 0.000000
116
+ 2023-10-25 11:46:32,626 epoch 3 - iter 2600/2606 - loss 0.09902161 - time (sec): 138.42 - samples/sec: 2648.89 - lr: 0.000023 - momentum: 0.000000
117
+ 2023-10-25 11:46:32,909 ----------------------------------------------------------------------------------------------------
118
+ 2023-10-25 11:46:32,909 EPOCH 3 done: loss 0.0991 - lr: 0.000023
119
+ 2023-10-25 11:46:39,755 DEV : loss 0.17354419827461243 - f1-score (micro avg) 0.3626
120
+ 2023-10-25 11:46:39,781 saving best model
121
+ 2023-10-25 11:46:40,474 ----------------------------------------------------------------------------------------------------
122
+ 2023-10-25 11:46:54,527 epoch 4 - iter 260/2606 - loss 0.06887745 - time (sec): 14.05 - samples/sec: 2623.65 - lr: 0.000023 - momentum: 0.000000
123
+ 2023-10-25 11:47:08,284 epoch 4 - iter 520/2606 - loss 0.06746623 - time (sec): 27.81 - samples/sec: 2633.87 - lr: 0.000023 - momentum: 0.000000
124
+ 2023-10-25 11:47:21,844 epoch 4 - iter 780/2606 - loss 0.06724212 - time (sec): 41.37 - samples/sec: 2648.90 - lr: 0.000022 - momentum: 0.000000
125
+ 2023-10-25 11:47:35,478 epoch 4 - iter 1040/2606 - loss 0.06775242 - time (sec): 55.00 - samples/sec: 2592.82 - lr: 0.000022 - momentum: 0.000000
126
+ 2023-10-25 11:47:49,494 epoch 4 - iter 1300/2606 - loss 0.06853685 - time (sec): 69.02 - samples/sec: 2618.05 - lr: 0.000022 - momentum: 0.000000
127
+ 2023-10-25 11:48:03,215 epoch 4 - iter 1560/2606 - loss 0.06832555 - time (sec): 82.74 - samples/sec: 2619.28 - lr: 0.000021 - momentum: 0.000000
128
+ 2023-10-25 11:48:17,331 epoch 4 - iter 1820/2606 - loss 0.06715608 - time (sec): 96.85 - samples/sec: 2620.90 - lr: 0.000021 - momentum: 0.000000
129
+ 2023-10-25 11:48:31,507 epoch 4 - iter 2080/2606 - loss 0.06797360 - time (sec): 111.03 - samples/sec: 2642.22 - lr: 0.000021 - momentum: 0.000000
130
+ 2023-10-25 11:48:45,148 epoch 4 - iter 2340/2606 - loss 0.06732721 - time (sec): 124.67 - samples/sec: 2628.22 - lr: 0.000020 - momentum: 0.000000
131
+ 2023-10-25 11:48:59,012 epoch 4 - iter 2600/2606 - loss 0.06658621 - time (sec): 138.53 - samples/sec: 2644.43 - lr: 0.000020 - momentum: 0.000000
132
+ 2023-10-25 11:48:59,328 ----------------------------------------------------------------------------------------------------
133
+ 2023-10-25 11:48:59,328 EPOCH 4 done: loss 0.0666 - lr: 0.000020
134
+ 2023-10-25 11:49:05,589 DEV : loss 0.30783411860466003 - f1-score (micro avg) 0.3413
135
+ 2023-10-25 11:49:05,614 ----------------------------------------------------------------------------------------------------
136
+ 2023-10-25 11:49:20,140 epoch 5 - iter 260/2606 - loss 0.04604953 - time (sec): 14.52 - samples/sec: 2436.13 - lr: 0.000020 - momentum: 0.000000
137
+ 2023-10-25 11:49:33,623 epoch 5 - iter 520/2606 - loss 0.04697664 - time (sec): 28.01 - samples/sec: 2540.14 - lr: 0.000019 - momentum: 0.000000
138
+ 2023-10-25 11:49:47,511 epoch 5 - iter 780/2606 - loss 0.04775602 - time (sec): 41.90 - samples/sec: 2623.00 - lr: 0.000019 - momentum: 0.000000
139
+ 2023-10-25 11:50:01,505 epoch 5 - iter 1040/2606 - loss 0.04621368 - time (sec): 55.89 - samples/sec: 2641.15 - lr: 0.000019 - momentum: 0.000000
140
+ 2023-10-25 11:50:15,262 epoch 5 - iter 1300/2606 - loss 0.04643873 - time (sec): 69.65 - samples/sec: 2665.18 - lr: 0.000018 - momentum: 0.000000
141
+ 2023-10-25 11:50:29,597 epoch 5 - iter 1560/2606 - loss 0.04688888 - time (sec): 83.98 - samples/sec: 2672.97 - lr: 0.000018 - momentum: 0.000000
142
+ 2023-10-25 11:50:43,060 epoch 5 - iter 1820/2606 - loss 0.04703940 - time (sec): 97.44 - samples/sec: 2673.71 - lr: 0.000018 - momentum: 0.000000
143
+ 2023-10-25 11:50:56,340 epoch 5 - iter 2080/2606 - loss 0.04639153 - time (sec): 110.72 - samples/sec: 2664.01 - lr: 0.000017 - momentum: 0.000000
144
+ 2023-10-25 11:51:09,859 epoch 5 - iter 2340/2606 - loss 0.04516071 - time (sec): 124.24 - samples/sec: 2649.51 - lr: 0.000017 - momentum: 0.000000
145
+ 2023-10-25 11:51:24,025 epoch 5 - iter 2600/2606 - loss 0.04552710 - time (sec): 138.41 - samples/sec: 2647.47 - lr: 0.000017 - momentum: 0.000000
146
+ 2023-10-25 11:51:24,357 ----------------------------------------------------------------------------------------------------
147
+ 2023-10-25 11:51:24,357 EPOCH 5 done: loss 0.0455 - lr: 0.000017
148
+ 2023-10-25 11:51:30,590 DEV : loss 0.3278255760669708 - f1-score (micro avg) 0.4015
149
+ 2023-10-25 11:51:30,616 saving best model
150
+ 2023-10-25 11:51:31,314 ----------------------------------------------------------------------------------------------------
151
+ 2023-10-25 11:51:45,515 epoch 6 - iter 260/2606 - loss 0.03373944 - time (sec): 14.20 - samples/sec: 2672.10 - lr: 0.000016 - momentum: 0.000000
152
+ 2023-10-25 11:51:59,610 epoch 6 - iter 520/2606 - loss 0.03199852 - time (sec): 28.29 - samples/sec: 2723.33 - lr: 0.000016 - momentum: 0.000000
153
+ 2023-10-25 11:52:13,794 epoch 6 - iter 780/2606 - loss 0.03393438 - time (sec): 42.48 - samples/sec: 2694.52 - lr: 0.000016 - momentum: 0.000000
154
+ 2023-10-25 11:52:28,592 epoch 6 - iter 1040/2606 - loss 0.03480030 - time (sec): 57.28 - samples/sec: 2659.23 - lr: 0.000015 - momentum: 0.000000
155
+ 2023-10-25 11:52:42,170 epoch 6 - iter 1300/2606 - loss 0.03442460 - time (sec): 70.85 - samples/sec: 2627.20 - lr: 0.000015 - momentum: 0.000000
156
+ 2023-10-25 11:52:55,763 epoch 6 - iter 1560/2606 - loss 0.03670962 - time (sec): 84.45 - samples/sec: 2631.63 - lr: 0.000015 - momentum: 0.000000
157
+ 2023-10-25 11:53:09,342 epoch 6 - iter 1820/2606 - loss 0.03715112 - time (sec): 98.03 - samples/sec: 2618.16 - lr: 0.000014 - momentum: 0.000000
158
+ 2023-10-25 11:53:23,179 epoch 6 - iter 2080/2606 - loss 0.03633915 - time (sec): 111.86 - samples/sec: 2616.67 - lr: 0.000014 - momentum: 0.000000
159
+ 2023-10-25 11:53:37,063 epoch 6 - iter 2340/2606 - loss 0.03702732 - time (sec): 125.75 - samples/sec: 2617.58 - lr: 0.000014 - momentum: 0.000000
160
+ 2023-10-25 11:53:51,255 epoch 6 - iter 2600/2606 - loss 0.03728123 - time (sec): 139.94 - samples/sec: 2619.81 - lr: 0.000013 - momentum: 0.000000
161
+ 2023-10-25 11:53:51,569 ----------------------------------------------------------------------------------------------------
162
+ 2023-10-25 11:53:51,569 EPOCH 6 done: loss 0.0373 - lr: 0.000013
163
+ 2023-10-25 11:53:57,772 DEV : loss 0.34462153911590576 - f1-score (micro avg) 0.3964
164
+ 2023-10-25 11:53:57,797 ----------------------------------------------------------------------------------------------------
165
+ 2023-10-25 11:54:11,682 epoch 7 - iter 260/2606 - loss 0.01979562 - time (sec): 13.88 - samples/sec: 2671.15 - lr: 0.000013 - momentum: 0.000000
166
+ 2023-10-25 11:54:25,692 epoch 7 - iter 520/2606 - loss 0.02087209 - time (sec): 27.89 - samples/sec: 2702.05 - lr: 0.000013 - momentum: 0.000000
167
+ 2023-10-25 11:54:39,839 epoch 7 - iter 780/2606 - loss 0.02290501 - time (sec): 42.04 - samples/sec: 2625.40 - lr: 0.000012 - momentum: 0.000000
168
+ 2023-10-25 11:54:54,087 epoch 7 - iter 1040/2606 - loss 0.02416505 - time (sec): 56.29 - samples/sec: 2625.41 - lr: 0.000012 - momentum: 0.000000
169
+ 2023-10-25 11:55:08,140 epoch 7 - iter 1300/2606 - loss 0.02420180 - time (sec): 70.34 - samples/sec: 2626.94 - lr: 0.000012 - momentum: 0.000000
170
+ 2023-10-25 11:55:23,022 epoch 7 - iter 1560/2606 - loss 0.02360409 - time (sec): 85.22 - samples/sec: 2626.26 - lr: 0.000011 - momentum: 0.000000
171
+ 2023-10-25 11:55:38,226 epoch 7 - iter 1820/2606 - loss 0.02405417 - time (sec): 100.43 - samples/sec: 2591.22 - lr: 0.000011 - momentum: 0.000000
172
+ 2023-10-25 11:55:51,845 epoch 7 - iter 2080/2606 - loss 0.02451150 - time (sec): 114.05 - samples/sec: 2587.49 - lr: 0.000011 - momentum: 0.000000
173
+ 2023-10-25 11:56:06,068 epoch 7 - iter 2340/2606 - loss 0.02449885 - time (sec): 128.27 - samples/sec: 2578.50 - lr: 0.000010 - momentum: 0.000000
174
+ 2023-10-25 11:56:19,845 epoch 7 - iter 2600/2606 - loss 0.02439320 - time (sec): 142.05 - samples/sec: 2578.34 - lr: 0.000010 - momentum: 0.000000
175
+ 2023-10-25 11:56:20,236 ----------------------------------------------------------------------------------------------------
176
+ 2023-10-25 11:56:20,236 EPOCH 7 done: loss 0.0244 - lr: 0.000010
177
+ 2023-10-25 11:56:26,581 DEV : loss 0.32091233134269714 - f1-score (micro avg) 0.4248
178
+ 2023-10-25 11:56:26,606 saving best model
179
+ 2023-10-25 11:56:27,151 ----------------------------------------------------------------------------------------------------
180
+ 2023-10-25 11:56:41,384 epoch 8 - iter 260/2606 - loss 0.02252562 - time (sec): 14.23 - samples/sec: 2609.58 - lr: 0.000010 - momentum: 0.000000
181
+ 2023-10-25 11:56:55,898 epoch 8 - iter 520/2606 - loss 0.01930306 - time (sec): 28.74 - samples/sec: 2625.60 - lr: 0.000009 - momentum: 0.000000
182
+ 2023-10-25 11:57:10,032 epoch 8 - iter 780/2606 - loss 0.01870884 - time (sec): 42.88 - samples/sec: 2618.88 - lr: 0.000009 - momentum: 0.000000
183
+ 2023-10-25 11:57:24,578 epoch 8 - iter 1040/2606 - loss 0.01915372 - time (sec): 57.42 - samples/sec: 2612.96 - lr: 0.000009 - momentum: 0.000000
184
+ 2023-10-25 11:57:40,398 epoch 8 - iter 1300/2606 - loss 0.01860075 - time (sec): 73.24 - samples/sec: 2595.58 - lr: 0.000008 - momentum: 0.000000
185
+ 2023-10-25 11:57:54,134 epoch 8 - iter 1560/2606 - loss 0.01949918 - time (sec): 86.98 - samples/sec: 2586.84 - lr: 0.000008 - momentum: 0.000000
186
+ 2023-10-25 11:58:08,121 epoch 8 - iter 1820/2606 - loss 0.01962360 - time (sec): 100.97 - samples/sec: 2558.62 - lr: 0.000008 - momentum: 0.000000
187
+ 2023-10-25 11:58:23,529 epoch 8 - iter 2080/2606 - loss 0.01974142 - time (sec): 116.37 - samples/sec: 2547.11 - lr: 0.000007 - momentum: 0.000000
188
+ 2023-10-25 11:58:37,973 epoch 8 - iter 2340/2606 - loss 0.01965665 - time (sec): 130.82 - samples/sec: 2541.92 - lr: 0.000007 - momentum: 0.000000
189
+ 2023-10-25 11:58:52,332 epoch 8 - iter 2600/2606 - loss 0.01992311 - time (sec): 145.18 - samples/sec: 2525.82 - lr: 0.000007 - momentum: 0.000000
190
+ 2023-10-25 11:58:52,624 ----------------------------------------------------------------------------------------------------
191
+ 2023-10-25 11:58:52,625 EPOCH 8 done: loss 0.0199 - lr: 0.000007
192
+ 2023-10-25 11:58:59,774 DEV : loss 0.4752597510814667 - f1-score (micro avg) 0.3824
193
+ 2023-10-25 11:58:59,801 ----------------------------------------------------------------------------------------------------
194
+ 2023-10-25 11:59:13,984 epoch 9 - iter 260/2606 - loss 0.01615083 - time (sec): 14.18 - samples/sec: 2541.15 - lr: 0.000006 - momentum: 0.000000
195
+ 2023-10-25 11:59:27,976 epoch 9 - iter 520/2606 - loss 0.01461464 - time (sec): 28.17 - samples/sec: 2585.33 - lr: 0.000006 - momentum: 0.000000
196
+ 2023-10-25 11:59:42,170 epoch 9 - iter 780/2606 - loss 0.01350285 - time (sec): 42.37 - samples/sec: 2579.29 - lr: 0.000006 - momentum: 0.000000
197
+ 2023-10-25 11:59:56,028 epoch 9 - iter 1040/2606 - loss 0.01384109 - time (sec): 56.23 - samples/sec: 2587.65 - lr: 0.000005 - momentum: 0.000000
198
+ 2023-10-25 12:00:10,480 epoch 9 - iter 1300/2606 - loss 0.01413513 - time (sec): 70.68 - samples/sec: 2595.62 - lr: 0.000005 - momentum: 0.000000
199
+ 2023-10-25 12:00:24,644 epoch 9 - iter 1560/2606 - loss 0.01406109 - time (sec): 84.84 - samples/sec: 2600.34 - lr: 0.000005 - momentum: 0.000000
200
+ 2023-10-25 12:00:38,961 epoch 9 - iter 1820/2606 - loss 0.01421979 - time (sec): 99.16 - samples/sec: 2603.67 - lr: 0.000004 - momentum: 0.000000
201
+ 2023-10-25 12:00:53,372 epoch 9 - iter 2080/2606 - loss 0.01397783 - time (sec): 113.57 - samples/sec: 2582.73 - lr: 0.000004 - momentum: 0.000000
202
+ 2023-10-25 12:01:08,292 epoch 9 - iter 2340/2606 - loss 0.01361878 - time (sec): 128.49 - samples/sec: 2573.71 - lr: 0.000004 - momentum: 0.000000
203
+ 2023-10-25 12:01:22,562 epoch 9 - iter 2600/2606 - loss 0.01331532 - time (sec): 142.76 - samples/sec: 2566.49 - lr: 0.000003 - momentum: 0.000000
204
+ 2023-10-25 12:01:22,891 ----------------------------------------------------------------------------------------------------
205
+ 2023-10-25 12:01:22,891 EPOCH 9 done: loss 0.0133 - lr: 0.000003
206
+ 2023-10-25 12:01:29,986 DEV : loss 0.4919085204601288 - f1-score (micro avg) 0.3897
207
+ 2023-10-25 12:01:30,013 ----------------------------------------------------------------------------------------------------
208
+ 2023-10-25 12:01:44,440 epoch 10 - iter 260/2606 - loss 0.00534808 - time (sec): 14.43 - samples/sec: 2579.04 - lr: 0.000003 - momentum: 0.000000
209
+ 2023-10-25 12:01:58,548 epoch 10 - iter 520/2606 - loss 0.00657158 - time (sec): 28.53 - samples/sec: 2522.87 - lr: 0.000003 - momentum: 0.000000
210
+ 2023-10-25 12:02:14,114 epoch 10 - iter 780/2606 - loss 0.00801464 - time (sec): 44.10 - samples/sec: 2529.83 - lr: 0.000002 - momentum: 0.000000
211
+ 2023-10-25 12:02:28,519 epoch 10 - iter 1040/2606 - loss 0.00927196 - time (sec): 58.50 - samples/sec: 2551.25 - lr: 0.000002 - momentum: 0.000000
212
+ 2023-10-25 12:02:42,788 epoch 10 - iter 1300/2606 - loss 0.00938321 - time (sec): 72.77 - samples/sec: 2518.72 - lr: 0.000002 - momentum: 0.000000
213
+ 2023-10-25 12:02:57,574 epoch 10 - iter 1560/2606 - loss 0.01027436 - time (sec): 87.56 - samples/sec: 2506.66 - lr: 0.000001 - momentum: 0.000000
214
+ 2023-10-25 12:03:11,726 epoch 10 - iter 1820/2606 - loss 0.01005900 - time (sec): 101.71 - samples/sec: 2509.51 - lr: 0.000001 - momentum: 0.000000
215
+ 2023-10-25 12:03:25,658 epoch 10 - iter 2080/2606 - loss 0.00995170 - time (sec): 115.64 - samples/sec: 2529.85 - lr: 0.000001 - momentum: 0.000000
216
+ 2023-10-25 12:03:39,660 epoch 10 - iter 2340/2606 - loss 0.00964683 - time (sec): 129.65 - samples/sec: 2546.93 - lr: 0.000000 - momentum: 0.000000
217
+ 2023-10-25 12:03:53,974 epoch 10 - iter 2600/2606 - loss 0.00939974 - time (sec): 143.96 - samples/sec: 2548.58 - lr: 0.000000 - momentum: 0.000000
218
+ 2023-10-25 12:03:54,247 ----------------------------------------------------------------------------------------------------
219
+ 2023-10-25 12:03:54,247 EPOCH 10 done: loss 0.0094 - lr: 0.000000
220
+ 2023-10-25 12:04:01,161 DEV : loss 0.4912854731082916 - f1-score (micro avg) 0.3849
221
+ 2023-10-25 12:04:01,808 ----------------------------------------------------------------------------------------------------
222
+ 2023-10-25 12:04:01,809 Loading model from best epoch ...
223
+ 2023-10-25 12:04:03,671 SequenceTagger predicts: Dictionary with 17 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, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
224
+ 2023-10-25 12:04:13,558
225
+ Results:
226
+ - F-score (micro) 0.4363
227
+ - F-score (macro) 0.2969
228
+ - Accuracy 0.2841
229
+
230
+ By class:
231
+ precision recall f1-score support
232
+
233
+ LOC 0.4646 0.4967 0.4801 1214
234
+ PER 0.4357 0.4567 0.4459 808
235
+ ORG 0.2788 0.2465 0.2617 353
236
+ HumanProd 0.0000 0.0000 0.0000 15
237
+
238
+ micro avg 0.4298 0.4431 0.4363 2390
239
+ macro avg 0.2948 0.3000 0.2969 2390
240
+ weighted avg 0.4244 0.4431 0.4333 2390
241
+
242
+ 2023-10-25 12:04:13,558 ----------------------------------------------------------------------------------------------------