Hugging Face's logo

stefan-it
/
hmbench-ajmc-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
1
stefan-it commited on
Commit
f3c5a65
•
1 Parent(s): d55a104

Upload ./training.log with huggingface_hub

Browse files
Files changed (1) hide show
  1. training.log +248 -0
training.log ADDED
@@ -0,0 +1,248 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-23 15:47:11,119 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-23 15:47:11,120 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=25, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-23 15:47:11,120 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-23 15:47:11,120 MultiCorpus: 1100 train + 206 dev + 240 test sentences
52
+ - NER_HIPE_2022 Corpus: 1100 train + 206 dev + 240 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/de/with_doc_seperator
53
+ 2023-10-23 15:47:11,120 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-23 15:47:11,120 Train: 1100 sentences
55
+ 2023-10-23 15:47:11,120 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-23 15:47:11,121 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-23 15:47:11,121 Training Params:
58
+ 2023-10-23 15:47:11,121 - learning_rate: "5e-05"
59
+ 2023-10-23 15:47:11,121 - mini_batch_size: "8"
60
+ 2023-10-23 15:47:11,121 - max_epochs: "10"
61
+ 2023-10-23 15:47:11,121 - shuffle: "True"
62
+ 2023-10-23 15:47:11,121 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-23 15:47:11,121 Plugins:
64
+ 2023-10-23 15:47:11,121 - TensorboardLogger
65
+ 2023-10-23 15:47:11,121 - LinearScheduler | warmup_fraction: '0.1'
66
+ 2023-10-23 15:47:11,121 ----------------------------------------------------------------------------------------------------
67
+ 2023-10-23 15:47:11,121 Final evaluation on model from best epoch (best-model.pt)
68
+ 2023-10-23 15:47:11,121 - metric: "('micro avg', 'f1-score')"
69
+ 2023-10-23 15:47:11,121 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-23 15:47:11,121 Computation:
71
+ 2023-10-23 15:47:11,121 - compute on device: cuda:0
72
+ 2023-10-23 15:47:11,121 - embedding storage: none
73
+ 2023-10-23 15:47:11,121 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-23 15:47:11,121 Model training base path: "hmbench-ajmc/de-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4"
75
+ 2023-10-23 15:47:11,121 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-23 15:47:11,121 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-23 15:47:11,121 Logging anything other than scalars to TensorBoard is currently not supported.
78
+ 2023-10-23 15:47:11,830 epoch 1 - iter 13/138 - loss 3.11505708 - time (sec): 0.71 - samples/sec: 2973.30 - lr: 0.000004 - momentum: 0.000000
79
+ 2023-10-23 15:47:12,567 epoch 1 - iter 26/138 - loss 2.46744691 - time (sec): 1.44 - samples/sec: 2948.36 - lr: 0.000009 - momentum: 0.000000
80
+ 2023-10-23 15:47:13,294 epoch 1 - iter 39/138 - loss 2.00064272 - time (sec): 2.17 - samples/sec: 2848.83 - lr: 0.000014 - momentum: 0.000000
81
+ 2023-10-23 15:47:14,059 epoch 1 - iter 52/138 - loss 1.65199002 - time (sec): 2.94 - samples/sec: 2947.01 - lr: 0.000018 - momentum: 0.000000
82
+ 2023-10-23 15:47:14,778 epoch 1 - iter 65/138 - loss 1.48169540 - time (sec): 3.66 - samples/sec: 2901.54 - lr: 0.000023 - momentum: 0.000000
83
+ 2023-10-23 15:47:15,519 epoch 1 - iter 78/138 - loss 1.29756776 - time (sec): 4.40 - samples/sec: 2920.72 - lr: 0.000028 - momentum: 0.000000
84
+ 2023-10-23 15:47:16,270 epoch 1 - iter 91/138 - loss 1.16855120 - time (sec): 5.15 - samples/sec: 2865.24 - lr: 0.000033 - momentum: 0.000000
85
+ 2023-10-23 15:47:17,053 epoch 1 - iter 104/138 - loss 1.03837897 - time (sec): 5.93 - samples/sec: 2921.34 - lr: 0.000037 - momentum: 0.000000
86
+ 2023-10-23 15:47:17,867 epoch 1 - iter 117/138 - loss 0.94938763 - time (sec): 6.75 - samples/sec: 2886.32 - lr: 0.000042 - momentum: 0.000000
87
+ 2023-10-23 15:47:18,654 epoch 1 - iter 130/138 - loss 0.88540734 - time (sec): 7.53 - samples/sec: 2878.25 - lr: 0.000047 - momentum: 0.000000
88
+ 2023-10-23 15:47:19,132 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-23 15:47:19,132 EPOCH 1 done: loss 0.8523 - lr: 0.000047
90
+ 2023-10-23 15:47:19,554 DEV : loss 0.2265705019235611 - f1-score (micro avg) 0.6643
91
+ 2023-10-23 15:47:19,560 saving best model
92
+ 2023-10-23 15:47:19,955 ----------------------------------------------------------------------------------------------------
93
+ 2023-10-23 15:47:20,667 epoch 2 - iter 13/138 - loss 0.26476026 - time (sec): 0.71 - samples/sec: 2982.33 - lr: 0.000050 - momentum: 0.000000
94
+ 2023-10-23 15:47:21,385 epoch 2 - iter 26/138 - loss 0.21464055 - time (sec): 1.43 - samples/sec: 3104.30 - lr: 0.000049 - momentum: 0.000000
95
+ 2023-10-23 15:47:22,113 epoch 2 - iter 39/138 - loss 0.20326127 - time (sec): 2.16 - samples/sec: 3134.81 - lr: 0.000048 - momentum: 0.000000
96
+ 2023-10-23 15:47:22,857 epoch 2 - iter 52/138 - loss 0.18189690 - time (sec): 2.90 - samples/sec: 3124.57 - lr: 0.000048 - momentum: 0.000000
97
+ 2023-10-23 15:47:23,666 epoch 2 - iter 65/138 - loss 0.18181936 - time (sec): 3.71 - samples/sec: 3005.53 - lr: 0.000047 - momentum: 0.000000
98
+ 2023-10-23 15:47:24,458 epoch 2 - iter 78/138 - loss 0.17939422 - time (sec): 4.50 - samples/sec: 2972.13 - lr: 0.000047 - momentum: 0.000000
99
+ 2023-10-23 15:47:25,250 epoch 2 - iter 91/138 - loss 0.17730142 - time (sec): 5.29 - samples/sec: 2935.32 - lr: 0.000046 - momentum: 0.000000
100
+ 2023-10-23 15:47:26,046 epoch 2 - iter 104/138 - loss 0.17262996 - time (sec): 6.09 - samples/sec: 2894.92 - lr: 0.000046 - momentum: 0.000000
101
+ 2023-10-23 15:47:26,839 epoch 2 - iter 117/138 - loss 0.16894373 - time (sec): 6.88 - samples/sec: 2840.58 - lr: 0.000045 - momentum: 0.000000
102
+ 2023-10-23 15:47:27,618 epoch 2 - iter 130/138 - loss 0.16407567 - time (sec): 7.66 - samples/sec: 2819.51 - lr: 0.000045 - momentum: 0.000000
103
+ 2023-10-23 15:47:28,102 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-23 15:47:28,102 EPOCH 2 done: loss 0.1596 - lr: 0.000045
105
+ 2023-10-23 15:47:28,638 DEV : loss 0.11940671503543854 - f1-score (micro avg) 0.8465
106
+ 2023-10-23 15:47:28,644 saving best model
107
+ 2023-10-23 15:47:29,202 ----------------------------------------------------------------------------------------------------
108
+ 2023-10-23 15:47:30,003 epoch 3 - iter 13/138 - loss 0.07988993 - time (sec): 0.80 - samples/sec: 2593.24 - lr: 0.000044 - momentum: 0.000000
109
+ 2023-10-23 15:47:30,798 epoch 3 - iter 26/138 - loss 0.10888248 - time (sec): 1.59 - samples/sec: 2556.50 - lr: 0.000043 - momentum: 0.000000
110
+ 2023-10-23 15:47:31,608 epoch 3 - iter 39/138 - loss 0.09016497 - time (sec): 2.40 - samples/sec: 2621.42 - lr: 0.000043 - momentum: 0.000000
111
+ 2023-10-23 15:47:32,396 epoch 3 - iter 52/138 - loss 0.08999953 - time (sec): 3.19 - samples/sec: 2617.65 - lr: 0.000042 - momentum: 0.000000
112
+ 2023-10-23 15:47:33,173 epoch 3 - iter 65/138 - loss 0.09109642 - time (sec): 3.96 - samples/sec: 2592.58 - lr: 0.000042 - momentum: 0.000000
113
+ 2023-10-23 15:47:33,952 epoch 3 - iter 78/138 - loss 0.08731252 - time (sec): 4.74 - samples/sec: 2642.90 - lr: 0.000041 - momentum: 0.000000
114
+ 2023-10-23 15:47:34,766 epoch 3 - iter 91/138 - loss 0.09236665 - time (sec): 5.56 - samples/sec: 2646.20 - lr: 0.000041 - momentum: 0.000000
115
+ 2023-10-23 15:47:35,556 epoch 3 - iter 104/138 - loss 0.08968405 - time (sec): 6.35 - samples/sec: 2661.88 - lr: 0.000040 - momentum: 0.000000
116
+ 2023-10-23 15:47:36,367 epoch 3 - iter 117/138 - loss 0.09102394 - time (sec): 7.16 - samples/sec: 2676.05 - lr: 0.000040 - momentum: 0.000000
117
+ 2023-10-23 15:47:37,174 epoch 3 - iter 130/138 - loss 0.09080745 - time (sec): 7.97 - samples/sec: 2709.68 - lr: 0.000039 - momentum: 0.000000
118
+ 2023-10-23 15:47:37,664 ----------------------------------------------------------------------------------------------------
119
+ 2023-10-23 15:47:37,664 EPOCH 3 done: loss 0.0903 - lr: 0.000039
120
+ 2023-10-23 15:47:38,358 DEV : loss 0.1325010061264038 - f1-score (micro avg) 0.839
121
+ 2023-10-23 15:47:38,364 ----------------------------------------------------------------------------------------------------
122
+ 2023-10-23 15:47:39,107 epoch 4 - iter 13/138 - loss 0.04646751 - time (sec): 0.74 - samples/sec: 2994.83 - lr: 0.000038 - momentum: 0.000000
123
+ 2023-10-23 15:47:39,891 epoch 4 - iter 26/138 - loss 0.05610819 - time (sec): 1.53 - samples/sec: 2739.53 - lr: 0.000038 - momentum: 0.000000
124
+ 2023-10-23 15:47:40,684 epoch 4 - iter 39/138 - loss 0.05405158 - time (sec): 2.32 - samples/sec: 2691.38 - lr: 0.000037 - momentum: 0.000000
125
+ 2023-10-23 15:47:41,500 epoch 4 - iter 52/138 - loss 0.05580227 - time (sec): 3.14 - samples/sec: 2655.31 - lr: 0.000037 - momentum: 0.000000
126
+ 2023-10-23 15:47:42,296 epoch 4 - iter 65/138 - loss 0.04970762 - time (sec): 3.93 - samples/sec: 2609.10 - lr: 0.000036 - momentum: 0.000000
127
+ 2023-10-23 15:47:43,110 epoch 4 - iter 78/138 - loss 0.05294991 - time (sec): 4.75 - samples/sec: 2629.20 - lr: 0.000036 - momentum: 0.000000
128
+ 2023-10-23 15:47:43,916 epoch 4 - iter 91/138 - loss 0.05649414 - time (sec): 5.55 - samples/sec: 2658.90 - lr: 0.000035 - momentum: 0.000000
129
+ 2023-10-23 15:47:44,734 epoch 4 - iter 104/138 - loss 0.05731326 - time (sec): 6.37 - samples/sec: 2672.85 - lr: 0.000035 - momentum: 0.000000
130
+ 2023-10-23 15:47:45,535 epoch 4 - iter 117/138 - loss 0.05895584 - time (sec): 7.17 - samples/sec: 2691.66 - lr: 0.000034 - momentum: 0.000000
131
+ 2023-10-23 15:47:46,328 epoch 4 - iter 130/138 - loss 0.06194870 - time (sec): 7.96 - samples/sec: 2686.50 - lr: 0.000034 - momentum: 0.000000
132
+ 2023-10-23 15:47:46,822 ----------------------------------------------------------------------------------------------------
133
+ 2023-10-23 15:47:46,822 EPOCH 4 done: loss 0.0612 - lr: 0.000034
134
+ 2023-10-23 15:47:47,354 DEV : loss 0.12988846004009247 - f1-score (micro avg) 0.8554
135
+ 2023-10-23 15:47:47,360 saving best model
136
+ 2023-10-23 15:47:47,915 ----------------------------------------------------------------------------------------------------
137
+ 2023-10-23 15:47:48,628 epoch 5 - iter 13/138 - loss 0.04927867 - time (sec): 0.71 - samples/sec: 2938.01 - lr: 0.000033 - momentum: 0.000000
138
+ 2023-10-23 15:47:49,431 epoch 5 - iter 26/138 - loss 0.03675946 - time (sec): 1.51 - samples/sec: 2794.05 - lr: 0.000032 - momentum: 0.000000
139
+ 2023-10-23 15:47:50,208 epoch 5 - iter 39/138 - loss 0.04209356 - time (sec): 2.29 - samples/sec: 2798.76 - lr: 0.000032 - momentum: 0.000000
140
+ 2023-10-23 15:47:50,987 epoch 5 - iter 52/138 - loss 0.04151665 - time (sec): 3.07 - samples/sec: 2811.44 - lr: 0.000031 - momentum: 0.000000
141
+ 2023-10-23 15:47:51,704 epoch 5 - iter 65/138 - loss 0.03815920 - time (sec): 3.78 - samples/sec: 2832.74 - lr: 0.000031 - momentum: 0.000000
142
+ 2023-10-23 15:47:52,432 epoch 5 - iter 78/138 - loss 0.04408756 - time (sec): 4.51 - samples/sec: 2896.67 - lr: 0.000030 - momentum: 0.000000
143
+ 2023-10-23 15:47:53,147 epoch 5 - iter 91/138 - loss 0.04668946 - time (sec): 5.23 - samples/sec: 2880.16 - lr: 0.000030 - momentum: 0.000000
144
+ 2023-10-23 15:47:53,836 epoch 5 - iter 104/138 - loss 0.04823073 - time (sec): 5.92 - samples/sec: 2872.66 - lr: 0.000029 - momentum: 0.000000
145
+ 2023-10-23 15:47:54,576 epoch 5 - iter 117/138 - loss 0.04762912 - time (sec): 6.66 - samples/sec: 2917.07 - lr: 0.000029 - momentum: 0.000000
146
+ 2023-10-23 15:47:55,286 epoch 5 - iter 130/138 - loss 0.04742132 - time (sec): 7.37 - samples/sec: 2933.92 - lr: 0.000028 - momentum: 0.000000
147
+ 2023-10-23 15:47:55,736 ----------------------------------------------------------------------------------------------------
148
+ 2023-10-23 15:47:55,736 EPOCH 5 done: loss 0.0454 - lr: 0.000028
149
+ 2023-10-23 15:47:56,268 DEV : loss 0.15986910462379456 - f1-score (micro avg) 0.8606
150
+ 2023-10-23 15:47:56,274 saving best model
151
+ 2023-10-23 15:47:56,818 ----------------------------------------------------------------------------------------------------
152
+ 2023-10-23 15:47:57,521 epoch 6 - iter 13/138 - loss 0.04161822 - time (sec): 0.70 - samples/sec: 3012.41 - lr: 0.000027 - momentum: 0.000000
153
+ 2023-10-23 15:47:58,239 epoch 6 - iter 26/138 - loss 0.05124527 - time (sec): 1.42 - samples/sec: 3101.86 - lr: 0.000027 - momentum: 0.000000
154
+ 2023-10-23 15:47:58,975 epoch 6 - iter 39/138 - loss 0.04212149 - time (sec): 2.15 - samples/sec: 3019.44 - lr: 0.000026 - momentum: 0.000000
155
+ 2023-10-23 15:47:59,690 epoch 6 - iter 52/138 - loss 0.03547030 - time (sec): 2.87 - samples/sec: 2992.94 - lr: 0.000026 - momentum: 0.000000
156
+ 2023-10-23 15:48:00,386 epoch 6 - iter 65/138 - loss 0.03220072 - time (sec): 3.56 - samples/sec: 3004.40 - lr: 0.000025 - momentum: 0.000000
157
+ 2023-10-23 15:48:01,097 epoch 6 - iter 78/138 - loss 0.02840167 - time (sec): 4.27 - samples/sec: 3103.60 - lr: 0.000025 - momentum: 0.000000
158
+ 2023-10-23 15:48:01,795 epoch 6 - iter 91/138 - loss 0.02684088 - time (sec): 4.97 - samples/sec: 3106.79 - lr: 0.000024 - momentum: 0.000000
159
+ 2023-10-23 15:48:02,516 epoch 6 - iter 104/138 - loss 0.03197223 - time (sec): 5.69 - samples/sec: 3058.56 - lr: 0.000024 - momentum: 0.000000
160
+ 2023-10-23 15:48:03,217 epoch 6 - iter 117/138 - loss 0.03171220 - time (sec): 6.39 - samples/sec: 3068.40 - lr: 0.000023 - momentum: 0.000000
161
+ 2023-10-23 15:48:03,901 epoch 6 - iter 130/138 - loss 0.03129985 - time (sec): 7.08 - samples/sec: 3033.14 - lr: 0.000023 - momentum: 0.000000
162
+ 2023-10-23 15:48:04,322 ----------------------------------------------------------------------------------------------------
163
+ 2023-10-23 15:48:04,323 EPOCH 6 done: loss 0.0310 - lr: 0.000023
164
+ 2023-10-23 15:48:04,853 DEV : loss 0.1651483029127121 - f1-score (micro avg) 0.8617
165
+ 2023-10-23 15:48:04,859 saving best model
166
+ 2023-10-23 15:48:05,386 ----------------------------------------------------------------------------------------------------
167
+ 2023-10-23 15:48:06,114 epoch 7 - iter 13/138 - loss 0.00533927 - time (sec): 0.72 - samples/sec: 2942.35 - lr: 0.000022 - momentum: 0.000000
168
+ 2023-10-23 15:48:06,817 epoch 7 - iter 26/138 - loss 0.01162164 - time (sec): 1.43 - samples/sec: 3030.67 - lr: 0.000021 - momentum: 0.000000
169
+ 2023-10-23 15:48:07,522 epoch 7 - iter 39/138 - loss 0.02842342 - time (sec): 2.13 - samples/sec: 3046.05 - lr: 0.000021 - momentum: 0.000000
170
+ 2023-10-23 15:48:08,284 epoch 7 - iter 52/138 - loss 0.02496210 - time (sec): 2.89 - samples/sec: 3099.66 - lr: 0.000020 - momentum: 0.000000
171
+ 2023-10-23 15:48:09,011 epoch 7 - iter 65/138 - loss 0.02352598 - time (sec): 3.62 - samples/sec: 3028.85 - lr: 0.000020 - momentum: 0.000000
172
+ 2023-10-23 15:48:09,752 epoch 7 - iter 78/138 - loss 0.02146960 - time (sec): 4.36 - samples/sec: 3021.91 - lr: 0.000019 - momentum: 0.000000
173
+ 2023-10-23 15:48:10,477 epoch 7 - iter 91/138 - loss 0.02311809 - time (sec): 5.09 - samples/sec: 2983.84 - lr: 0.000019 - momentum: 0.000000
174
+ 2023-10-23 15:48:11,224 epoch 7 - iter 104/138 - loss 0.02189737 - time (sec): 5.83 - samples/sec: 2999.38 - lr: 0.000018 - momentum: 0.000000
175
+ 2023-10-23 15:48:11,967 epoch 7 - iter 117/138 - loss 0.02004301 - time (sec): 6.58 - samples/sec: 2995.24 - lr: 0.000018 - momentum: 0.000000
176
+ 2023-10-23 15:48:12,695 epoch 7 - iter 130/138 - loss 0.01962060 - time (sec): 7.30 - samples/sec: 2966.80 - lr: 0.000017 - momentum: 0.000000
177
+ 2023-10-23 15:48:13,127 ----------------------------------------------------------------------------------------------------
178
+ 2023-10-23 15:48:13,127 EPOCH 7 done: loss 0.0200 - lr: 0.000017
179
+ 2023-10-23 15:48:13,671 DEV : loss 0.1657860279083252 - f1-score (micro avg) 0.8644
180
+ 2023-10-23 15:48:13,678 saving best model
181
+ 2023-10-23 15:48:14,230 ----------------------------------------------------------------------------------------------------
182
+ 2023-10-23 15:48:14,928 epoch 8 - iter 13/138 - loss 0.00589471 - time (sec): 0.69 - samples/sec: 2934.05 - lr: 0.000016 - momentum: 0.000000
183
+ 2023-10-23 15:48:15,622 epoch 8 - iter 26/138 - loss 0.00765983 - time (sec): 1.39 - samples/sec: 3195.34 - lr: 0.000016 - momentum: 0.000000
184
+ 2023-10-23 15:48:16,335 epoch 8 - iter 39/138 - loss 0.00720803 - time (sec): 2.10 - samples/sec: 3132.26 - lr: 0.000015 - momentum: 0.000000
185
+ 2023-10-23 15:48:17,042 epoch 8 - iter 52/138 - loss 0.01376241 - time (sec): 2.81 - samples/sec: 3114.13 - lr: 0.000015 - momentum: 0.000000
186
+ 2023-10-23 15:48:17,757 epoch 8 - iter 65/138 - loss 0.01729594 - time (sec): 3.52 - samples/sec: 3040.45 - lr: 0.000014 - momentum: 0.000000
187
+ 2023-10-23 15:48:18,467 epoch 8 - iter 78/138 - loss 0.01653361 - time (sec): 4.23 - samples/sec: 3061.06 - lr: 0.000014 - momentum: 0.000000
188
+ 2023-10-23 15:48:19,184 epoch 8 - iter 91/138 - loss 0.01734789 - time (sec): 4.95 - samples/sec: 3064.51 - lr: 0.000013 - momentum: 0.000000
189
+ 2023-10-23 15:48:19,884 epoch 8 - iter 104/138 - loss 0.01590998 - time (sec): 5.65 - samples/sec: 3029.29 - lr: 0.000013 - momentum: 0.000000
190
+ 2023-10-23 15:48:20,592 epoch 8 - iter 117/138 - loss 0.01576961 - time (sec): 6.36 - samples/sec: 3025.51 - lr: 0.000012 - momentum: 0.000000
191
+ 2023-10-23 15:48:21,291 epoch 8 - iter 130/138 - loss 0.01454689 - time (sec): 7.06 - samples/sec: 3046.09 - lr: 0.000012 - momentum: 0.000000
192
+ 2023-10-23 15:48:21,722 ----------------------------------------------------------------------------------------------------
193
+ 2023-10-23 15:48:21,722 EPOCH 8 done: loss 0.0141 - lr: 0.000012
194
+ 2023-10-23 15:48:22,253 DEV : loss 0.1713750660419464 - f1-score (micro avg) 0.8809
195
+ 2023-10-23 15:48:22,259 saving best model
196
+ 2023-10-23 15:48:22,785 ----------------------------------------------------------------------------------------------------
197
+ 2023-10-23 15:48:23,573 epoch 9 - iter 13/138 - loss 0.00064167 - time (sec): 0.78 - samples/sec: 2662.24 - lr: 0.000011 - momentum: 0.000000
198
+ 2023-10-23 15:48:24,379 epoch 9 - iter 26/138 - loss 0.00090612 - time (sec): 1.59 - samples/sec: 2591.74 - lr: 0.000010 - momentum: 0.000000
199
+ 2023-10-23 15:48:25,156 epoch 9 - iter 39/138 - loss 0.00356920 - time (sec): 2.37 - samples/sec: 2661.44 - lr: 0.000010 - momentum: 0.000000
200
+ 2023-10-23 15:48:25,905 epoch 9 - iter 52/138 - loss 0.00694166 - time (sec): 3.12 - samples/sec: 2703.28 - lr: 0.000009 - momentum: 0.000000
201
+ 2023-10-23 15:48:26,662 epoch 9 - iter 65/138 - loss 0.00634685 - time (sec): 3.87 - samples/sec: 2750.75 - lr: 0.000009 - momentum: 0.000000
202
+ 2023-10-23 15:48:27,371 epoch 9 - iter 78/138 - loss 0.00630311 - time (sec): 4.58 - samples/sec: 2811.30 - lr: 0.000008 - momentum: 0.000000
203
+ 2023-10-23 15:48:28,085 epoch 9 - iter 91/138 - loss 0.00702119 - time (sec): 5.29 - samples/sec: 2857.61 - lr: 0.000008 - momentum: 0.000000
204
+ 2023-10-23 15:48:28,793 epoch 9 - iter 104/138 - loss 0.00633296 - time (sec): 6.00 - samples/sec: 2881.34 - lr: 0.000007 - momentum: 0.000000
205
+ 2023-10-23 15:48:29,494 epoch 9 - iter 117/138 - loss 0.00628241 - time (sec): 6.70 - samples/sec: 2893.94 - lr: 0.000007 - momentum: 0.000000
206
+ 2023-10-23 15:48:30,188 epoch 9 - iter 130/138 - loss 0.00742424 - time (sec): 7.40 - samples/sec: 2912.60 - lr: 0.000006 - momentum: 0.000000
207
+ 2023-10-23 15:48:30,618 ----------------------------------------------------------------------------------------------------
208
+ 2023-10-23 15:48:30,619 EPOCH 9 done: loss 0.0081 - lr: 0.000006
209
+ 2023-10-23 15:48:31,149 DEV : loss 0.168824702501297 - f1-score (micro avg) 0.8878
210
+ 2023-10-23 15:48:31,155 saving best model
211
+ 2023-10-23 15:48:31,685 ----------------------------------------------------------------------------------------------------
212
+ 2023-10-23 15:48:32,447 epoch 10 - iter 13/138 - loss 0.00118623 - time (sec): 0.76 - samples/sec: 2828.23 - lr: 0.000005 - momentum: 0.000000
213
+ 2023-10-23 15:48:33,213 epoch 10 - iter 26/138 - loss 0.00065035 - time (sec): 1.53 - samples/sec: 2780.14 - lr: 0.000005 - momentum: 0.000000
214
+ 2023-10-23 15:48:33,980 epoch 10 - iter 39/138 - loss 0.00146356 - time (sec): 2.29 - samples/sec: 2887.96 - lr: 0.000004 - momentum: 0.000000
215
+ 2023-10-23 15:48:34,716 epoch 10 - iter 52/138 - loss 0.00148726 - time (sec): 3.03 - samples/sec: 2858.71 - lr: 0.000004 - momentum: 0.000000
216
+ 2023-10-23 15:48:35,468 epoch 10 - iter 65/138 - loss 0.00601163 - time (sec): 3.78 - samples/sec: 2811.73 - lr: 0.000003 - momentum: 0.000000
217
+ 2023-10-23 15:48:36,217 epoch 10 - iter 78/138 - loss 0.00550576 - time (sec): 4.53 - samples/sec: 2841.53 - lr: 0.000003 - momentum: 0.000000
218
+ 2023-10-23 15:48:36,937 epoch 10 - iter 91/138 - loss 0.00571920 - time (sec): 5.25 - samples/sec: 2850.58 - lr: 0.000002 - momentum: 0.000000
219
+ 2023-10-23 15:48:37,659 epoch 10 - iter 104/138 - loss 0.00554833 - time (sec): 5.97 - samples/sec: 2870.23 - lr: 0.000002 - momentum: 0.000000
220
+ 2023-10-23 15:48:38,411 epoch 10 - iter 117/138 - loss 0.00516556 - time (sec): 6.72 - samples/sec: 2864.43 - lr: 0.000001 - momentum: 0.000000
221
+ 2023-10-23 15:48:39,166 epoch 10 - iter 130/138 - loss 0.00497822 - time (sec): 7.48 - samples/sec: 2875.52 - lr: 0.000000 - momentum: 0.000000
222
+ 2023-10-23 15:48:39,615 ----------------------------------------------------------------------------------------------------
223
+ 2023-10-23 15:48:39,615 EPOCH 10 done: loss 0.0052 - lr: 0.000000
224
+ 2023-10-23 15:48:40,154 DEV : loss 0.16836421191692352 - f1-score (micro avg) 0.8892
225
+ 2023-10-23 15:48:40,160 saving best model
226
+ 2023-10-23 15:48:41,107 ----------------------------------------------------------------------------------------------------
227
+ 2023-10-23 15:48:41,109 Loading model from best epoch ...
228
+ 2023-10-23 15:48:42,759 SequenceTagger predicts: Dictionary with 25 tags: O, S-scope, B-scope, E-scope, I-scope, S-pers, B-pers, E-pers, I-pers, S-work, B-work, E-work, I-work, S-loc, B-loc, E-loc, I-loc, S-object, B-object, E-object, I-object, S-date, B-date, E-date, I-date
229
+ 2023-10-23 15:48:43,434
230
+ Results:
231
+ - F-score (micro) 0.9098
232
+ - F-score (macro) 0.8239
233
+ - Accuracy 0.8426
234
+
235
+ By class:
236
+ precision recall f1-score support
237
+
238
+ scope 0.8883 0.9034 0.8958 176
239
+ pers 0.9840 0.9609 0.9723 128
240
+ work 0.8514 0.8514 0.8514 74
241
+ loc 0.3333 0.5000 0.4000 2
242
+ object 1.0000 1.0000 1.0000 2
243
+
244
+ micro avg 0.9086 0.9110 0.9098 382
245
+ macro avg 0.8114 0.8431 0.8239 382
246
+ weighted avg 0.9109 0.9110 0.9108 382
247
+
248
+ 2023-10-23 15:48:43,435 ----------------------------------------------------------------------------------------------------