Hugging Face's logo

stefan-it
/
hmbench-newseye-de-hmbert_64k-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4

Token Classification
Flair
PyTorch
TensorBoard
German
sequence-tagger-model
Model card Files Metrics Community
stefan-it commited on
Commit
00c677b
·
1 Parent(s): e21169d

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 16:45:15,823 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-25 16:45:15,823 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 16:45:15,824 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-25 16:45:15,824 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 16:45:15,824 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-25 16:45:15,824 Train: 20847 sentences
55
+ 2023-10-25 16:45:15,824 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-25 16:45:15,824 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-25 16:45:15,824 Training Params:
58
+ 2023-10-25 16:45:15,824 - learning_rate: "5e-05"
59
+ 2023-10-25 16:45:15,824 - mini_batch_size: "8"
60
+ 2023-10-25 16:45:15,824 - max_epochs: "10"
61
+ 2023-10-25 16:45:15,824 - shuffle: "True"
62
+ 2023-10-25 16:45:15,824 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-25 16:45:15,824 Plugins:
64
+ 2023-10-25 16:45:15,824 - TensorboardLogger
65
+ 2023-10-25 16:45:15,824 - LinearScheduler | warmup_fraction: '0.1'
66
+ 2023-10-25 16:45:15,824 ----------------------------------------------------------------------------------------------------
67
+ 2023-10-25 16:45:15,824 Final evaluation on model from best epoch (best-model.pt)
68
+ 2023-10-25 16:45:15,824 - metric: "('micro avg', 'f1-score')"
69
+ 2023-10-25 16:45:15,824 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-25 16:45:15,824 Computation:
71
+ 2023-10-25 16:45:15,824 - compute on device: cuda:0
72
+ 2023-10-25 16:45:15,824 - embedding storage: none
73
+ 2023-10-25 16:45:15,824 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-25 16:45:15,824 Model training base path: "hmbench-newseye/de-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4"
75
+ 2023-10-25 16:45:15,824 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-25 16:45:15,824 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-25 16:45:15,824 Logging anything other than scalars to TensorBoard is currently not supported.
78
+ 2023-10-25 16:45:29,924 epoch 1 - iter 260/2606 - loss 1.44327200 - time (sec): 14.10 - samples/sec: 2479.77 - lr: 0.000005 - momentum: 0.000000
79
+ 2023-10-25 16:45:44,833 epoch 1 - iter 520/2606 - loss 0.87020391 - time (sec): 29.01 - samples/sec: 2516.71 - lr: 0.000010 - momentum: 0.000000
80
+ 2023-10-25 16:45:58,506 epoch 1 - iter 780/2606 - loss 0.67200000 - time (sec): 42.68 - samples/sec: 2488.08 - lr: 0.000015 - momentum: 0.000000
81
+ 2023-10-25 16:46:12,807 epoch 1 - iter 1040/2606 - loss 0.56002942 - time (sec): 56.98 - samples/sec: 2552.72 - lr: 0.000020 - momentum: 0.000000
82
+ 2023-10-25 16:46:27,006 epoch 1 - iter 1300/2606 - loss 0.48938374 - time (sec): 71.18 - samples/sec: 2564.16 - lr: 0.000025 - momentum: 0.000000
83
+ 2023-10-25 16:46:40,890 epoch 1 - iter 1560/2606 - loss 0.43900299 - time (sec): 85.06 - samples/sec: 2585.58 - lr: 0.000030 - momentum: 0.000000
84
+ 2023-10-25 16:46:55,169 epoch 1 - iter 1820/2606 - loss 0.39943791 - time (sec): 99.34 - samples/sec: 2600.59 - lr: 0.000035 - momentum: 0.000000
85
+ 2023-10-25 16:47:09,222 epoch 1 - iter 2080/2606 - loss 0.37314820 - time (sec): 113.40 - samples/sec: 2595.66 - lr: 0.000040 - momentum: 0.000000
86
+ 2023-10-25 16:47:23,358 epoch 1 - iter 2340/2606 - loss 0.35456131 - time (sec): 127.53 - samples/sec: 2592.17 - lr: 0.000045 - momentum: 0.000000
87
+ 2023-10-25 16:47:37,439 epoch 1 - iter 2600/2606 - loss 0.33854746 - time (sec): 141.61 - samples/sec: 2592.16 - lr: 0.000050 - momentum: 0.000000
88
+ 2023-10-25 16:47:37,706 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-25 16:47:37,707 EPOCH 1 done: loss 0.3387 - lr: 0.000050
90
+ 2023-10-25 16:47:41,324 DEV : loss 0.118812695145607 - f1-score (micro avg) 0.1625
91
+ 2023-10-25 16:47:41,349 saving best model
92
+ 2023-10-25 16:47:41,819 ----------------------------------------------------------------------------------------------------
93
+ 2023-10-25 16:47:55,809 epoch 2 - iter 260/2606 - loss 0.17564576 - time (sec): 13.99 - samples/sec: 2630.19 - lr: 0.000049 - momentum: 0.000000
94
+ 2023-10-25 16:48:10,903 epoch 2 - iter 520/2606 - loss 0.16339489 - time (sec): 29.08 - samples/sec: 2633.99 - lr: 0.000049 - momentum: 0.000000
95
+ 2023-10-25 16:48:25,129 epoch 2 - iter 780/2606 - loss 0.16367825 - time (sec): 43.31 - samples/sec: 2648.13 - lr: 0.000048 - momentum: 0.000000
96
+ 2023-10-25 16:48:39,023 epoch 2 - iter 1040/2606 - loss 0.16465426 - time (sec): 57.20 - samples/sec: 2644.39 - lr: 0.000048 - momentum: 0.000000
97
+ 2023-10-25 16:48:52,698 epoch 2 - iter 1300/2606 - loss 0.16691656 - time (sec): 70.88 - samples/sec: 2618.17 - lr: 0.000047 - momentum: 0.000000
98
+ 2023-10-25 16:49:07,060 epoch 2 - iter 1560/2606 - loss 0.16504912 - time (sec): 85.24 - samples/sec: 2615.67 - lr: 0.000047 - momentum: 0.000000
99
+ 2023-10-25 16:49:21,199 epoch 2 - iter 1820/2606 - loss 0.16040319 - time (sec): 99.38 - samples/sec: 2625.20 - lr: 0.000046 - momentum: 0.000000
100
+ 2023-10-25 16:49:35,311 epoch 2 - iter 2080/2606 - loss 0.15902007 - time (sec): 113.49 - samples/sec: 2637.30 - lr: 0.000046 - momentum: 0.000000
101
+ 2023-10-25 16:49:48,475 epoch 2 - iter 2340/2606 - loss 0.15709971 - time (sec): 126.65 - samples/sec: 2628.74 - lr: 0.000045 - momentum: 0.000000
102
+ 2023-10-25 16:50:02,559 epoch 2 - iter 2600/2606 - loss 0.15647373 - time (sec): 140.74 - samples/sec: 2604.39 - lr: 0.000044 - momentum: 0.000000
103
+ 2023-10-25 16:50:02,869 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-25 16:50:02,869 EPOCH 2 done: loss 0.1566 - lr: 0.000044
105
+ 2023-10-25 16:50:09,872 DEV : loss 0.1611776500940323 - f1-score (micro avg) 0.3391
106
+ 2023-10-25 16:50:09,896 saving best model
107
+ 2023-10-25 16:50:10,500 ----------------------------------------------------------------------------------------------------
108
+ 2023-10-25 16:50:24,874 epoch 3 - iter 260/2606 - loss 0.10670473 - time (sec): 14.37 - samples/sec: 2724.91 - lr: 0.000044 - momentum: 0.000000
109
+ 2023-10-25 16:50:38,583 epoch 3 - iter 520/2606 - loss 0.10898329 - time (sec): 28.08 - samples/sec: 2706.78 - lr: 0.000043 - momentum: 0.000000
110
+ 2023-10-25 16:50:52,379 epoch 3 - iter 780/2606 - loss 0.11166505 - time (sec): 41.88 - samples/sec: 2643.92 - lr: 0.000043 - momentum: 0.000000
111
+ 2023-10-25 16:51:06,834 epoch 3 - iter 1040/2606 - loss 0.10939807 - time (sec): 56.33 - samples/sec: 2663.11 - lr: 0.000042 - momentum: 0.000000
112
+ 2023-10-25 16:51:20,398 epoch 3 - iter 1300/2606 - loss 0.10997620 - time (sec): 69.90 - samples/sec: 2648.46 - lr: 0.000042 - momentum: 0.000000
113
+ 2023-10-25 16:51:34,567 epoch 3 - iter 1560/2606 - loss 0.11114702 - time (sec): 84.07 - samples/sec: 2627.29 - lr: 0.000041 - momentum: 0.000000
114
+ 2023-10-25 16:51:48,566 epoch 3 - iter 1820/2606 - loss 0.10915658 - time (sec): 98.06 - samples/sec: 2619.10 - lr: 0.000041 - momentum: 0.000000
115
+ 2023-10-25 16:52:02,818 epoch 3 - iter 2080/2606 - loss 0.10966065 - time (sec): 112.32 - samples/sec: 2620.09 - lr: 0.000040 - momentum: 0.000000
116
+ 2023-10-25 16:52:16,817 epoch 3 - iter 2340/2606 - loss 0.11109113 - time (sec): 126.31 - samples/sec: 2612.60 - lr: 0.000039 - momentum: 0.000000
117
+ 2023-10-25 16:52:30,695 epoch 3 - iter 2600/2606 - loss 0.10926086 - time (sec): 140.19 - samples/sec: 2616.49 - lr: 0.000039 - momentum: 0.000000
118
+ 2023-10-25 16:52:30,986 ----------------------------------------------------------------------------------------------------
119
+ 2023-10-25 16:52:30,986 EPOCH 3 done: loss 0.1093 - lr: 0.000039
120
+ 2023-10-25 16:52:38,212 DEV : loss 0.23014920949935913 - f1-score (micro avg) 0.3626
121
+ 2023-10-25 16:52:38,236 saving best model
122
+ 2023-10-25 16:52:38,689 ----------------------------------------------------------------------------------------------------
123
+ 2023-10-25 16:52:52,399 epoch 4 - iter 260/2606 - loss 0.08026601 - time (sec): 13.71 - samples/sec: 2578.59 - lr: 0.000038 - momentum: 0.000000
124
+ 2023-10-25 16:53:06,053 epoch 4 - iter 520/2606 - loss 0.07866155 - time (sec): 27.36 - samples/sec: 2540.99 - lr: 0.000038 - momentum: 0.000000
125
+ 2023-10-25 16:53:20,595 epoch 4 - iter 780/2606 - loss 0.07694070 - time (sec): 41.90 - samples/sec: 2599.38 - lr: 0.000037 - momentum: 0.000000
126
+ 2023-10-25 16:53:34,797 epoch 4 - iter 1040/2606 - loss 0.07572775 - time (sec): 56.11 - samples/sec: 2612.43 - lr: 0.000037 - momentum: 0.000000
127
+ 2023-10-25 16:53:48,644 epoch 4 - iter 1300/2606 - loss 0.07947650 - time (sec): 69.95 - samples/sec: 2613.93 - lr: 0.000036 - momentum: 0.000000
128
+ 2023-10-25 16:54:02,886 epoch 4 - iter 1560/2606 - loss 0.07880117 - time (sec): 84.20 - samples/sec: 2609.19 - lr: 0.000036 - momentum: 0.000000
129
+ 2023-10-25 16:54:16,960 epoch 4 - iter 1820/2606 - loss 0.07887358 - time (sec): 98.27 - samples/sec: 2611.18 - lr: 0.000035 - momentum: 0.000000
130
+ 2023-10-25 16:54:31,164 epoch 4 - iter 2080/2606 - loss 0.07742549 - time (sec): 112.47 - samples/sec: 2620.31 - lr: 0.000034 - momentum: 0.000000
131
+ 2023-10-25 16:54:44,982 epoch 4 - iter 2340/2606 - loss 0.07826687 - time (sec): 126.29 - samples/sec: 2623.98 - lr: 0.000034 - momentum: 0.000000
132
+ 2023-10-25 16:54:58,682 epoch 4 - iter 2600/2606 - loss 0.07860251 - time (sec): 139.99 - samples/sec: 2620.12 - lr: 0.000033 - momentum: 0.000000
133
+ 2023-10-25 16:54:58,958 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-25 16:54:58,958 EPOCH 4 done: loss 0.0788 - lr: 0.000033
135
+ 2023-10-25 16:55:05,179 DEV : loss 0.24600794911384583 - f1-score (micro avg) 0.358
136
+ 2023-10-25 16:55:05,203 ----------------------------------------------------------------------------------------------------
137
+ 2023-10-25 16:55:19,990 epoch 5 - iter 260/2606 - loss 0.05672130 - time (sec): 14.79 - samples/sec: 2594.82 - lr: 0.000033 - momentum: 0.000000
138
+ 2023-10-25 16:55:34,219 epoch 5 - iter 520/2606 - loss 0.06166018 - time (sec): 29.01 - samples/sec: 2621.33 - lr: 0.000032 - momentum: 0.000000
139
+ 2023-10-25 16:55:48,173 epoch 5 - iter 780/2606 - loss 0.06054051 - time (sec): 42.97 - samples/sec: 2619.67 - lr: 0.000032 - momentum: 0.000000
140
+ 2023-10-25 16:56:02,180 epoch 5 - iter 1040/2606 - loss 0.05994699 - time (sec): 56.98 - samples/sec: 2592.31 - lr: 0.000031 - momentum: 0.000000
141
+ 2023-10-25 16:56:16,341 epoch 5 - iter 1300/2606 - loss 0.06032593 - time (sec): 71.14 - samples/sec: 2605.83 - lr: 0.000031 - momentum: 0.000000
142
+ 2023-10-25 16:56:30,625 epoch 5 - iter 1560/2606 - loss 0.05835629 - time (sec): 85.42 - samples/sec: 2598.27 - lr: 0.000030 - momentum: 0.000000
143
+ 2023-10-25 16:56:44,827 epoch 5 - iter 1820/2606 - loss 0.06037339 - time (sec): 99.62 - samples/sec: 2592.66 - lr: 0.000029 - momentum: 0.000000
144
+ 2023-10-25 16:56:58,661 epoch 5 - iter 2080/2606 - loss 0.05960227 - time (sec): 113.46 - samples/sec: 2600.76 - lr: 0.000029 - momentum: 0.000000
145
+ 2023-10-25 16:57:11,848 epoch 5 - iter 2340/2606 - loss 0.05877380 - time (sec): 126.64 - samples/sec: 2617.27 - lr: 0.000028 - momentum: 0.000000
146
+ 2023-10-25 16:57:26,094 epoch 5 - iter 2600/2606 - loss 0.05813652 - time (sec): 140.89 - samples/sec: 2605.23 - lr: 0.000028 - momentum: 0.000000
147
+ 2023-10-25 16:57:26,431 ----------------------------------------------------------------------------------------------------
148
+ 2023-10-25 16:57:26,431 EPOCH 5 done: loss 0.0581 - lr: 0.000028
149
+ 2023-10-25 16:57:32,774 DEV : loss 0.29553988575935364 - f1-score (micro avg) 0.4099
150
+ 2023-10-25 16:57:32,799 saving best model
151
+ 2023-10-25 16:57:33,294 ----------------------------------------------------------------------------------------------------
152
+ 2023-10-25 16:57:48,554 epoch 6 - iter 260/2606 - loss 0.04797258 - time (sec): 15.26 - samples/sec: 2573.89 - lr: 0.000027 - momentum: 0.000000
153
+ 2023-10-25 16:58:02,779 epoch 6 - iter 520/2606 - loss 0.04837650 - time (sec): 29.48 - samples/sec: 2595.28 - lr: 0.000027 - momentum: 0.000000
154
+ 2023-10-25 16:58:17,011 epoch 6 - iter 780/2606 - loss 0.04578472 - time (sec): 43.71 - samples/sec: 2609.95 - lr: 0.000026 - momentum: 0.000000
155
+ 2023-10-25 16:58:31,560 epoch 6 - iter 1040/2606 - loss 0.04760570 - time (sec): 58.26 - samples/sec: 2572.96 - lr: 0.000026 - momentum: 0.000000
156
+ 2023-10-25 16:58:45,137 epoch 6 - iter 1300/2606 - loss 0.04892803 - time (sec): 71.84 - samples/sec: 2569.00 - lr: 0.000025 - momentum: 0.000000
157
+ 2023-10-25 16:58:58,688 epoch 6 - iter 1560/2606 - loss 0.05121577 - time (sec): 85.39 - samples/sec: 2567.67 - lr: 0.000024 - momentum: 0.000000
158
+ 2023-10-25 16:59:13,684 epoch 6 - iter 1820/2606 - loss 0.05294998 - time (sec): 100.39 - samples/sec: 2576.77 - lr: 0.000024 - momentum: 0.000000
159
+ 2023-10-25 16:59:27,754 epoch 6 - iter 2080/2606 - loss 0.05640148 - time (sec): 114.46 - samples/sec: 2573.59 - lr: 0.000023 - momentum: 0.000000
160
+ 2023-10-25 16:59:41,576 epoch 6 - iter 2340/2606 - loss 0.05593627 - time (sec): 128.28 - samples/sec: 2576.66 - lr: 0.000023 - momentum: 0.000000
161
+ 2023-10-25 16:59:55,997 epoch 6 - iter 2600/2606 - loss 0.05483941 - time (sec): 142.70 - samples/sec: 2565.84 - lr: 0.000022 - momentum: 0.000000
162
+ 2023-10-25 16:59:56,369 ----------------------------------------------------------------------------------------------------
163
+ 2023-10-25 16:59:56,369 EPOCH 6 done: loss 0.0547 - lr: 0.000022
164
+ 2023-10-25 17:00:02,624 DEV : loss 0.33623284101486206 - f1-score (micro avg) 0.3687
165
+ 2023-10-25 17:00:02,649 ----------------------------------------------------------------------------------------------------
166
+ 2023-10-25 17:00:16,615 epoch 7 - iter 260/2606 - loss 0.03858880 - time (sec): 13.97 - samples/sec: 2649.66 - lr: 0.000022 - momentum: 0.000000
167
+ 2023-10-25 17:00:30,568 epoch 7 - iter 520/2606 - loss 0.04139851 - time (sec): 27.92 - samples/sec: 2635.80 - lr: 0.000021 - momentum: 0.000000
168
+ 2023-10-25 17:00:45,290 epoch 7 - iter 780/2606 - loss 0.04241529 - time (sec): 42.64 - samples/sec: 2604.64 - lr: 0.000021 - momentum: 0.000000
169
+ 2023-10-25 17:00:59,285 epoch 7 - iter 1040/2606 - loss 0.04833894 - time (sec): 56.64 - samples/sec: 2601.68 - lr: 0.000020 - momentum: 0.000000
170
+ 2023-10-25 17:01:13,271 epoch 7 - iter 1300/2606 - loss 0.04856922 - time (sec): 70.62 - samples/sec: 2592.87 - lr: 0.000019 - momentum: 0.000000
171
+ 2023-10-25 17:01:28,375 epoch 7 - iter 1560/2606 - loss 0.05096085 - time (sec): 85.73 - samples/sec: 2579.70 - lr: 0.000019 - momentum: 0.000000
172
+ 2023-10-25 17:01:43,143 epoch 7 - iter 1820/2606 - loss 0.05958653 - time (sec): 100.49 - samples/sec: 2603.24 - lr: 0.000018 - momentum: 0.000000
173
+ 2023-10-25 17:01:56,751 epoch 7 - iter 2080/2606 - loss 0.06793201 - time (sec): 114.10 - samples/sec: 2605.36 - lr: 0.000018 - momentum: 0.000000
174
+ 2023-10-25 17:02:09,950 epoch 7 - iter 2340/2606 - loss 0.06878116 - time (sec): 127.30 - samples/sec: 2607.52 - lr: 0.000017 - momentum: 0.000000
175
+ 2023-10-25 17:02:23,915 epoch 7 - iter 2600/2606 - loss 0.06941224 - time (sec): 141.27 - samples/sec: 2596.78 - lr: 0.000017 - momentum: 0.000000
176
+ 2023-10-25 17:02:24,220 ----------------------------------------------------------------------------------------------------
177
+ 2023-10-25 17:02:24,220 EPOCH 7 done: loss 0.0694 - lr: 0.000017
178
+ 2023-10-25 17:02:30,444 DEV : loss 0.35152772068977356 - f1-score (micro avg) 0.3214
179
+ 2023-10-25 17:02:30,469 ----------------------------------------------------------------------------------------------------
180
+ 2023-10-25 17:02:44,163 epoch 8 - iter 260/2606 - loss 0.07588476 - time (sec): 13.69 - samples/sec: 2635.53 - lr: 0.000016 - momentum: 0.000000
181
+ 2023-10-25 17:02:57,982 epoch 8 - iter 520/2606 - loss 0.10043230 - time (sec): 27.51 - samples/sec: 2649.59 - lr: 0.000016 - momentum: 0.000000
182
+ 2023-10-25 17:03:11,756 epoch 8 - iter 780/2606 - loss 0.13203650 - time (sec): 41.29 - samples/sec: 2628.01 - lr: 0.000015 - momentum: 0.000000
183
+ 2023-10-25 17:03:25,598 epoch 8 - iter 1040/2606 - loss 0.12385530 - time (sec): 55.13 - samples/sec: 2623.26 - lr: 0.000014 - momentum: 0.000000
184
+ 2023-10-25 17:03:39,705 epoch 8 - iter 1300/2606 - loss 0.12813762 - time (sec): 69.23 - samples/sec: 2618.67 - lr: 0.000014 - momentum: 0.000000
185
+ 2023-10-25 17:03:53,694 epoch 8 - iter 1560/2606 - loss 0.13153397 - time (sec): 83.22 - samples/sec: 2621.37 - lr: 0.000013 - momentum: 0.000000
186
+ 2023-10-25 17:04:08,336 epoch 8 - iter 1820/2606 - loss 0.12905434 - time (sec): 97.87 - samples/sec: 2630.44 - lr: 0.000013 - momentum: 0.000000
187
+ 2023-10-25 17:04:22,400 epoch 8 - iter 2080/2606 - loss 0.13389258 - time (sec): 111.93 - samples/sec: 2623.41 - lr: 0.000012 - momentum: 0.000000
188
+ 2023-10-25 17:04:36,755 epoch 8 - iter 2340/2606 - loss 0.13480518 - time (sec): 126.28 - samples/sec: 2606.93 - lr: 0.000012 - momentum: 0.000000
189
+ 2023-10-25 17:04:50,583 epoch 8 - iter 2600/2606 - loss 0.13531880 - time (sec): 140.11 - samples/sec: 2616.35 - lr: 0.000011 - momentum: 0.000000
190
+ 2023-10-25 17:04:50,912 ----------------------------------------------------------------------------------------------------
191
+ 2023-10-25 17:04:50,912 EPOCH 8 done: loss 0.1352 - lr: 0.000011
192
+ 2023-10-25 17:04:57,141 DEV : loss 0.2633623480796814 - f1-score (micro avg) 0.2342
193
+ 2023-10-25 17:04:57,166 ----------------------------------------------------------------------------------------------------
194
+ 2023-10-25 17:05:11,033 epoch 9 - iter 260/2606 - loss 0.08967090 - time (sec): 13.87 - samples/sec: 2713.58 - lr: 0.000011 - momentum: 0.000000
195
+ 2023-10-25 17:05:25,132 epoch 9 - iter 520/2606 - loss 0.09331506 - time (sec): 27.97 - samples/sec: 2651.52 - lr: 0.000010 - momentum: 0.000000
196
+ 2023-10-25 17:05:38,813 epoch 9 - iter 780/2606 - loss 0.09092922 - time (sec): 41.65 - samples/sec: 2643.14 - lr: 0.000009 - momentum: 0.000000
197
+ 2023-10-25 17:05:52,488 epoch 9 - iter 1040/2606 - loss 0.09774521 - time (sec): 55.32 - samples/sec: 2686.25 - lr: 0.000009 - momentum: 0.000000
198
+ 2023-10-25 17:06:06,110 epoch 9 - iter 1300/2606 - loss 0.10814172 - time (sec): 68.94 - samples/sec: 2685.63 - lr: 0.000008 - momentum: 0.000000
199
+ 2023-10-25 17:06:19,697 epoch 9 - iter 1560/2606 - loss 0.11196481 - time (sec): 82.53 - samples/sec: 2668.35 - lr: 0.000008 - momentum: 0.000000
200
+ 2023-10-25 17:06:33,655 epoch 9 - iter 1820/2606 - loss 0.11081450 - time (sec): 96.49 - samples/sec: 2673.87 - lr: 0.000007 - momentum: 0.000000
201
+ 2023-10-25 17:06:47,477 epoch 9 - iter 2080/2606 - loss 0.11167705 - time (sec): 110.31 - samples/sec: 2663.33 - lr: 0.000007 - momentum: 0.000000
202
+ 2023-10-25 17:07:01,705 epoch 9 - iter 2340/2606 - loss 0.11066523 - time (sec): 124.54 - samples/sec: 2664.34 - lr: 0.000006 - momentum: 0.000000
203
+ 2023-10-25 17:07:15,529 epoch 9 - iter 2600/2606 - loss 0.11158714 - time (sec): 138.36 - samples/sec: 2647.26 - lr: 0.000006 - momentum: 0.000000
204
+ 2023-10-25 17:07:15,945 ----------------------------------------------------------------------------------------------------
205
+ 2023-10-25 17:07:15,946 EPOCH 9 done: loss 0.1115 - lr: 0.000006
206
+ 2023-10-25 17:07:22,880 DEV : loss 0.2682478427886963 - f1-score (micro avg) 0.2293
207
+ 2023-10-25 17:07:22,913 ----------------------------------------------------------------------------------------------------
208
+ 2023-10-25 17:07:37,193 epoch 10 - iter 260/2606 - loss 0.09408029 - time (sec): 14.28 - samples/sec: 2595.12 - lr: 0.000005 - momentum: 0.000000
209
+ 2023-10-25 17:07:51,010 epoch 10 - iter 520/2606 - loss 0.09512229 - time (sec): 28.09 - samples/sec: 2593.26 - lr: 0.000004 - momentum: 0.000000
210
+ 2023-10-25 17:08:05,607 epoch 10 - iter 780/2606 - loss 0.08964442 - time (sec): 42.69 - samples/sec: 2609.86 - lr: 0.000004 - momentum: 0.000000
211
+ 2023-10-25 17:08:20,011 epoch 10 - iter 1040/2606 - loss 0.08987618 - time (sec): 57.10 - samples/sec: 2633.35 - lr: 0.000003 - momentum: 0.000000
212
+ 2023-10-25 17:08:33,885 epoch 10 - iter 1300/2606 - loss 0.08788357 - time (sec): 70.97 - samples/sec: 2663.52 - lr: 0.000003 - momentum: 0.000000
213
+ 2023-10-25 17:08:47,392 epoch 10 - iter 1560/2606 - loss 0.08724918 - time (sec): 84.48 - samples/sec: 2641.55 - lr: 0.000002 - momentum: 0.000000
214
+ 2023-10-25 17:09:01,218 epoch 10 - iter 1820/2606 - loss 0.08700501 - time (sec): 98.30 - samples/sec: 2620.75 - lr: 0.000002 - momentum: 0.000000
215
+ 2023-10-25 17:09:14,991 epoch 10 - iter 2080/2606 - loss 0.08812984 - time (sec): 112.08 - samples/sec: 2606.24 - lr: 0.000001 - momentum: 0.000000
216
+ 2023-10-25 17:09:29,100 epoch 10 - iter 2340/2606 - loss 0.09074088 - time (sec): 126.18 - samples/sec: 2614.01 - lr: 0.000001 - momentum: 0.000000
217
+ 2023-10-25 17:09:43,532 epoch 10 - iter 2600/2606 - loss 0.09068240 - time (sec): 140.62 - samples/sec: 2606.43 - lr: 0.000000 - momentum: 0.000000
218
+ 2023-10-25 17:09:43,851 ----------------------------------------------------------------------------------------------------
219
+ 2023-10-25 17:09:43,851 EPOCH 10 done: loss 0.0905 - lr: 0.000000
220
+ 2023-10-25 17:09:50,726 DEV : loss 0.27786970138549805 - f1-score (micro avg) 0.2168
221
+ 2023-10-25 17:09:51,221 ----------------------------------------------------------------------------------------------------
222
+ 2023-10-25 17:09:51,222 Loading model from best epoch ...
223
+ 2023-10-25 17:09:52,830 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 17:10:02,533
225
+ Results:
226
+ - F-score (micro) 0.4446
227
+ - F-score (macro) 0.2829
228
+ - Accuracy 0.2912
229
+
230
+ By class:
231
+ precision recall f1-score support
232
+
233
+ LOC 0.5264 0.5840 0.5537 1214
234
+ PER 0.4000 0.3490 0.3728 808
235
+ ORG 0.2194 0.1926 0.2051 353
236
+ HumanProd 0.0000 0.0000 0.0000 15
237
+
238
+ micro avg 0.4461 0.4431 0.4446 2390
239
+ macro avg 0.2864 0.2814 0.2829 2390
240
+ weighted avg 0.4350 0.4431 0.4376 2390
241
+
242
+ 2023-10-25 17:10:02,533 ----------------------------------------------------------------------------------------------------