Hugging Face's logo

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

Token Classification
Flair
PyTorch
TensorBoard
German
sequence-tagger-model
Model card Files Metrics Community
1
stefan-it commited on
Commit
87b5f20
•
1 Parent(s): 5ccfe2f

Upload ./training.log with huggingface_hub

Browse files
Files changed (1) hide show
  1. training.log +246 -0
training.log ADDED
@@ -0,0 +1,246 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-23 14:59:24,306 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-23 14:59:24,307 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 14:59:24,307 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-23 14:59:24,307 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 14:59:24,307 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-23 14:59:24,307 Train: 1100 sentences
55
+ 2023-10-23 14:59:24,307 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-23 14:59:24,307 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-23 14:59:24,307 Training Params:
58
+ 2023-10-23 14:59:24,307 - learning_rate: "5e-05"
59
+ 2023-10-23 14:59:24,307 - mini_batch_size: "8"
60
+ 2023-10-23 14:59:24,307 - max_epochs: "10"
61
+ 2023-10-23 14:59:24,307 - shuffle: "True"
62
+ 2023-10-23 14:59:24,307 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-23 14:59:24,308 Plugins:
64
+ 2023-10-23 14:59:24,308 - TensorboardLogger
65
+ 2023-10-23 14:59:24,308 - LinearScheduler | warmup_fraction: '0.1'
66
+ 2023-10-23 14:59:24,308 ----------------------------------------------------------------------------------------------------
67
+ 2023-10-23 14:59:24,308 Final evaluation on model from best epoch (best-model.pt)
68
+ 2023-10-23 14:59:24,308 - metric: "('micro avg', 'f1-score')"
69
+ 2023-10-23 14:59:24,308 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-23 14:59:24,308 Computation:
71
+ 2023-10-23 14:59:24,308 - compute on device: cuda:0
72
+ 2023-10-23 14:59:24,308 - embedding storage: none
73
+ 2023-10-23 14:59:24,308 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-23 14:59:24,308 Model training base path: "hmbench-ajmc/de-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1"
75
+ 2023-10-23 14:59:24,308 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-23 14:59:24,308 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-23 14:59:24,308 Logging anything other than scalars to TensorBoard is currently not supported.
78
+ 2023-10-23 14:59:25,027 epoch 1 - iter 13/138 - loss 3.66543780 - time (sec): 0.72 - samples/sec: 3026.20 - lr: 0.000004 - momentum: 0.000000
79
+ 2023-10-23 14:59:25,745 epoch 1 - iter 26/138 - loss 3.00984760 - time (sec): 1.44 - samples/sec: 2969.12 - lr: 0.000009 - momentum: 0.000000
80
+ 2023-10-23 14:59:26,450 epoch 1 - iter 39/138 - loss 2.39180029 - time (sec): 2.14 - samples/sec: 2875.04 - lr: 0.000014 - momentum: 0.000000
81
+ 2023-10-23 14:59:27,243 epoch 1 - iter 52/138 - loss 1.93634022 - time (sec): 2.93 - samples/sec: 2845.79 - lr: 0.000018 - momentum: 0.000000
82
+ 2023-10-23 14:59:27,962 epoch 1 - iter 65/138 - loss 1.68757988 - time (sec): 3.65 - samples/sec: 2879.97 - lr: 0.000023 - momentum: 0.000000
83
+ 2023-10-23 14:59:28,690 epoch 1 - iter 78/138 - loss 1.48023065 - time (sec): 4.38 - samples/sec: 2886.24 - lr: 0.000028 - momentum: 0.000000
84
+ 2023-10-23 14:59:29,408 epoch 1 - iter 91/138 - loss 1.32188926 - time (sec): 5.10 - samples/sec: 2900.66 - lr: 0.000033 - momentum: 0.000000
85
+ 2023-10-23 14:59:30,130 epoch 1 - iter 104/138 - loss 1.18393768 - time (sec): 5.82 - samples/sec: 2915.39 - lr: 0.000037 - momentum: 0.000000
86
+ 2023-10-23 14:59:30,841 epoch 1 - iter 117/138 - loss 1.07131939 - time (sec): 6.53 - samples/sec: 2945.55 - lr: 0.000042 - momentum: 0.000000
87
+ 2023-10-23 14:59:31,603 epoch 1 - iter 130/138 - loss 0.98230483 - time (sec): 7.29 - samples/sec: 2950.92 - lr: 0.000047 - momentum: 0.000000
88
+ 2023-10-23 14:59:32,041 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-23 14:59:32,041 EPOCH 1 done: loss 0.9408 - lr: 0.000047
90
+ 2023-10-23 14:59:32,466 DEV : loss 0.17025014758110046 - f1-score (micro avg) 0.7773
91
+ 2023-10-23 14:59:32,472 saving best model
92
+ 2023-10-23 14:59:32,868 ----------------------------------------------------------------------------------------------------
93
+ 2023-10-23 14:59:33,590 epoch 2 - iter 13/138 - loss 0.14784199 - time (sec): 0.72 - samples/sec: 2802.71 - lr: 0.000050 - momentum: 0.000000
94
+ 2023-10-23 14:59:34,321 epoch 2 - iter 26/138 - loss 0.11871101 - time (sec): 1.45 - samples/sec: 2935.30 - lr: 0.000049 - momentum: 0.000000
95
+ 2023-10-23 14:59:35,042 epoch 2 - iter 39/138 - loss 0.14691239 - time (sec): 2.17 - samples/sec: 2897.04 - lr: 0.000048 - momentum: 0.000000
96
+ 2023-10-23 14:59:35,847 epoch 2 - iter 52/138 - loss 0.15325754 - time (sec): 2.98 - samples/sec: 2875.85 - lr: 0.000048 - momentum: 0.000000
97
+ 2023-10-23 14:59:36,572 epoch 2 - iter 65/138 - loss 0.15032886 - time (sec): 3.70 - samples/sec: 2879.84 - lr: 0.000047 - momentum: 0.000000
98
+ 2023-10-23 14:59:37,291 epoch 2 - iter 78/138 - loss 0.15293320 - time (sec): 4.42 - samples/sec: 2905.21 - lr: 0.000047 - momentum: 0.000000
99
+ 2023-10-23 14:59:38,012 epoch 2 - iter 91/138 - loss 0.14945325 - time (sec): 5.14 - samples/sec: 2934.83 - lr: 0.000046 - momentum: 0.000000
100
+ 2023-10-23 14:59:38,796 epoch 2 - iter 104/138 - loss 0.15342696 - time (sec): 5.93 - samples/sec: 2916.57 - lr: 0.000046 - momentum: 0.000000
101
+ 2023-10-23 14:59:39,528 epoch 2 - iter 117/138 - loss 0.14987829 - time (sec): 6.66 - samples/sec: 2915.77 - lr: 0.000045 - momentum: 0.000000
102
+ 2023-10-23 14:59:40,253 epoch 2 - iter 130/138 - loss 0.15091087 - time (sec): 7.38 - samples/sec: 2922.83 - lr: 0.000045 - momentum: 0.000000
103
+ 2023-10-23 14:59:40,709 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-23 14:59:40,709 EPOCH 2 done: loss 0.1508 - lr: 0.000045
105
+ 2023-10-23 14:59:41,247 DEV : loss 0.12781710922718048 - f1-score (micro avg) 0.8111
106
+ 2023-10-23 14:59:41,252 saving best model
107
+ 2023-10-23 14:59:41,793 ----------------------------------------------------------------------------------------------------
108
+ 2023-10-23 14:59:42,587 epoch 3 - iter 13/138 - loss 0.08645294 - time (sec): 0.79 - samples/sec: 2604.66 - lr: 0.000044 - momentum: 0.000000
109
+ 2023-10-23 14:59:43,329 epoch 3 - iter 26/138 - loss 0.08482620 - time (sec): 1.53 - samples/sec: 2803.15 - lr: 0.000043 - momentum: 0.000000
110
+ 2023-10-23 14:59:44,087 epoch 3 - iter 39/138 - loss 0.07421142 - time (sec): 2.29 - samples/sec: 2951.16 - lr: 0.000043 - momentum: 0.000000
111
+ 2023-10-23 14:59:44,821 epoch 3 - iter 52/138 - loss 0.07684756 - time (sec): 3.02 - samples/sec: 2890.92 - lr: 0.000042 - momentum: 0.000000
112
+ 2023-10-23 14:59:45,575 epoch 3 - iter 65/138 - loss 0.07988944 - time (sec): 3.78 - samples/sec: 2892.17 - lr: 0.000042 - momentum: 0.000000
113
+ 2023-10-23 14:59:46,359 epoch 3 - iter 78/138 - loss 0.08173290 - time (sec): 4.56 - samples/sec: 2809.17 - lr: 0.000041 - momentum: 0.000000
114
+ 2023-10-23 14:59:47,106 epoch 3 - iter 91/138 - loss 0.08449295 - time (sec): 5.31 - samples/sec: 2813.00 - lr: 0.000041 - momentum: 0.000000
115
+ 2023-10-23 14:59:47,859 epoch 3 - iter 104/138 - loss 0.08365887 - time (sec): 6.06 - samples/sec: 2815.61 - lr: 0.000040 - momentum: 0.000000
116
+ 2023-10-23 14:59:48,597 epoch 3 - iter 117/138 - loss 0.08635961 - time (sec): 6.80 - samples/sec: 2814.13 - lr: 0.000040 - momentum: 0.000000
117
+ 2023-10-23 14:59:49,362 epoch 3 - iter 130/138 - loss 0.08894879 - time (sec): 7.57 - samples/sec: 2840.59 - lr: 0.000039 - momentum: 0.000000
118
+ 2023-10-23 14:59:49,833 ----------------------------------------------------------------------------------------------------
119
+ 2023-10-23 14:59:49,833 EPOCH 3 done: loss 0.0867 - lr: 0.000039
120
+ 2023-10-23 14:59:50,542 DEV : loss 0.12280590832233429 - f1-score (micro avg) 0.8447
121
+ 2023-10-23 14:59:50,547 saving best model
122
+ 2023-10-23 14:59:51,053 ----------------------------------------------------------------------------------------------------
123
+ 2023-10-23 14:59:51,872 epoch 4 - iter 13/138 - loss 0.05732552 - time (sec): 0.82 - samples/sec: 2575.68 - lr: 0.000038 - momentum: 0.000000
124
+ 2023-10-23 14:59:52,594 epoch 4 - iter 26/138 - loss 0.05523775 - time (sec): 1.54 - samples/sec: 2899.82 - lr: 0.000038 - momentum: 0.000000
125
+ 2023-10-23 14:59:53,306 epoch 4 - iter 39/138 - loss 0.05726177 - time (sec): 2.25 - samples/sec: 2827.15 - lr: 0.000037 - momentum: 0.000000
126
+ 2023-10-23 14:59:54,023 epoch 4 - iter 52/138 - loss 0.05790800 - time (sec): 2.97 - samples/sec: 2893.97 - lr: 0.000037 - momentum: 0.000000
127
+ 2023-10-23 14:59:54,753 epoch 4 - iter 65/138 - loss 0.05696300 - time (sec): 3.70 - samples/sec: 2898.53 - lr: 0.000036 - momentum: 0.000000
128
+ 2023-10-23 14:59:55,463 epoch 4 - iter 78/138 - loss 0.05800066 - time (sec): 4.41 - samples/sec: 2861.98 - lr: 0.000036 - momentum: 0.000000
129
+ 2023-10-23 14:59:56,181 epoch 4 - iter 91/138 - loss 0.05510131 - time (sec): 5.12 - samples/sec: 2878.87 - lr: 0.000035 - momentum: 0.000000
130
+ 2023-10-23 14:59:56,894 epoch 4 - iter 104/138 - loss 0.05698036 - time (sec): 5.84 - samples/sec: 2905.22 - lr: 0.000035 - momentum: 0.000000
131
+ 2023-10-23 14:59:57,614 epoch 4 - iter 117/138 - loss 0.05516748 - time (sec): 6.56 - samples/sec: 2934.96 - lr: 0.000034 - momentum: 0.000000
132
+ 2023-10-23 14:59:58,324 epoch 4 - iter 130/138 - loss 0.05554848 - time (sec): 7.27 - samples/sec: 2943.58 - lr: 0.000034 - momentum: 0.000000
133
+ 2023-10-23 14:59:58,765 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-23 14:59:58,765 EPOCH 4 done: loss 0.0548 - lr: 0.000034
135
+ 2023-10-23 14:59:59,315 DEV : loss 0.13340476155281067 - f1-score (micro avg) 0.8653
136
+ 2023-10-23 14:59:59,320 saving best model
137
+ 2023-10-23 14:59:59,850 ----------------------------------------------------------------------------------------------------
138
+ 2023-10-23 15:00:00,561 epoch 5 - iter 13/138 - loss 0.02190968 - time (sec): 0.71 - samples/sec: 2839.29 - lr: 0.000033 - momentum: 0.000000
139
+ 2023-10-23 15:00:01,278 epoch 5 - iter 26/138 - loss 0.03255026 - time (sec): 1.42 - samples/sec: 2948.57 - lr: 0.000032 - momentum: 0.000000
140
+ 2023-10-23 15:00:01,994 epoch 5 - iter 39/138 - loss 0.04017228 - time (sec): 2.14 - samples/sec: 2944.89 - lr: 0.000032 - momentum: 0.000000
141
+ 2023-10-23 15:00:02,723 epoch 5 - iter 52/138 - loss 0.05156922 - time (sec): 2.87 - samples/sec: 2939.16 - lr: 0.000031 - momentum: 0.000000
142
+ 2023-10-23 15:00:03,439 epoch 5 - iter 65/138 - loss 0.05092015 - time (sec): 3.58 - samples/sec: 2910.39 - lr: 0.000031 - momentum: 0.000000
143
+ 2023-10-23 15:00:04,172 epoch 5 - iter 78/138 - loss 0.04808824 - time (sec): 4.32 - samples/sec: 2931.55 - lr: 0.000030 - momentum: 0.000000
144
+ 2023-10-23 15:00:04,888 epoch 5 - iter 91/138 - loss 0.04539131 - time (sec): 5.03 - samples/sec: 2964.95 - lr: 0.000030 - momentum: 0.000000
145
+ 2023-10-23 15:00:05,600 epoch 5 - iter 104/138 - loss 0.04367704 - time (sec): 5.75 - samples/sec: 2983.60 - lr: 0.000029 - momentum: 0.000000
146
+ 2023-10-23 15:00:06,328 epoch 5 - iter 117/138 - loss 0.04343764 - time (sec): 6.47 - samples/sec: 2981.65 - lr: 0.000029 - momentum: 0.000000
147
+ 2023-10-23 15:00:07,061 epoch 5 - iter 130/138 - loss 0.04444769 - time (sec): 7.21 - samples/sec: 2988.59 - lr: 0.000028 - momentum: 0.000000
148
+ 2023-10-23 15:00:07,496 ----------------------------------------------------------------------------------------------------
149
+ 2023-10-23 15:00:07,496 EPOCH 5 done: loss 0.0447 - lr: 0.000028
150
+ 2023-10-23 15:00:08,032 DEV : loss 0.13237358629703522 - f1-score (micro avg) 0.8854
151
+ 2023-10-23 15:00:08,038 saving best model
152
+ 2023-10-23 15:00:08,590 ----------------------------------------------------------------------------------------------------
153
+ 2023-10-23 15:00:09,309 epoch 6 - iter 13/138 - loss 0.03222312 - time (sec): 0.72 - samples/sec: 3156.71 - lr: 0.000027 - momentum: 0.000000
154
+ 2023-10-23 15:00:10,022 epoch 6 - iter 26/138 - loss 0.02407024 - time (sec): 1.43 - samples/sec: 2975.50 - lr: 0.000027 - momentum: 0.000000
155
+ 2023-10-23 15:00:10,738 epoch 6 - iter 39/138 - loss 0.02068903 - time (sec): 2.15 - samples/sec: 3069.60 - lr: 0.000026 - momentum: 0.000000
156
+ 2023-10-23 15:00:11,452 epoch 6 - iter 52/138 - loss 0.03086218 - time (sec): 2.86 - samples/sec: 3087.93 - lr: 0.000026 - momentum: 0.000000
157
+ 2023-10-23 15:00:12,173 epoch 6 - iter 65/138 - loss 0.03106766 - time (sec): 3.58 - samples/sec: 3080.79 - lr: 0.000025 - momentum: 0.000000
158
+ 2023-10-23 15:00:12,884 epoch 6 - iter 78/138 - loss 0.03356739 - time (sec): 4.29 - samples/sec: 3029.68 - lr: 0.000025 - momentum: 0.000000
159
+ 2023-10-23 15:00:13,599 epoch 6 - iter 91/138 - loss 0.03089337 - time (sec): 5.01 - samples/sec: 3050.62 - lr: 0.000024 - momentum: 0.000000
160
+ 2023-10-23 15:00:14,300 epoch 6 - iter 104/138 - loss 0.03442090 - time (sec): 5.71 - samples/sec: 3025.17 - lr: 0.000024 - momentum: 0.000000
161
+ 2023-10-23 15:00:15,035 epoch 6 - iter 117/138 - loss 0.03202038 - time (sec): 6.44 - samples/sec: 3021.98 - lr: 0.000023 - momentum: 0.000000
162
+ 2023-10-23 15:00:15,771 epoch 6 - iter 130/138 - loss 0.03533535 - time (sec): 7.18 - samples/sec: 3006.32 - lr: 0.000023 - momentum: 0.000000
163
+ 2023-10-23 15:00:16,197 ----------------------------------------------------------------------------------------------------
164
+ 2023-10-23 15:00:16,197 EPOCH 6 done: loss 0.0364 - lr: 0.000023
165
+ 2023-10-23 15:00:16,729 DEV : loss 0.15275059640407562 - f1-score (micro avg) 0.8938
166
+ 2023-10-23 15:00:16,735 saving best model
167
+ 2023-10-23 15:00:17,279 ----------------------------------------------------------------------------------------------------
168
+ 2023-10-23 15:00:18,001 epoch 7 - iter 13/138 - loss 0.00527848 - time (sec): 0.72 - samples/sec: 2744.44 - lr: 0.000022 - momentum: 0.000000
169
+ 2023-10-23 15:00:18,717 epoch 7 - iter 26/138 - loss 0.01820801 - time (sec): 1.44 - samples/sec: 2926.12 - lr: 0.000021 - momentum: 0.000000
170
+ 2023-10-23 15:00:19,427 epoch 7 - iter 39/138 - loss 0.01831966 - time (sec): 2.15 - samples/sec: 2962.30 - lr: 0.000021 - momentum: 0.000000
171
+ 2023-10-23 15:00:20,142 epoch 7 - iter 52/138 - loss 0.02742555 - time (sec): 2.86 - samples/sec: 3044.56 - lr: 0.000020 - momentum: 0.000000
172
+ 2023-10-23 15:00:20,875 epoch 7 - iter 65/138 - loss 0.02380596 - time (sec): 3.59 - samples/sec: 3024.39 - lr: 0.000020 - momentum: 0.000000
173
+ 2023-10-23 15:00:21,601 epoch 7 - iter 78/138 - loss 0.02668274 - time (sec): 4.32 - samples/sec: 3007.31 - lr: 0.000019 - momentum: 0.000000
174
+ 2023-10-23 15:00:22,323 epoch 7 - iter 91/138 - loss 0.02478362 - time (sec): 5.04 - samples/sec: 3037.01 - lr: 0.000019 - momentum: 0.000000
175
+ 2023-10-23 15:00:23,037 epoch 7 - iter 104/138 - loss 0.02409141 - time (sec): 5.76 - samples/sec: 3014.13 - lr: 0.000018 - momentum: 0.000000
176
+ 2023-10-23 15:00:23,746 epoch 7 - iter 117/138 - loss 0.02390261 - time (sec): 6.46 - samples/sec: 3022.82 - lr: 0.000018 - momentum: 0.000000
177
+ 2023-10-23 15:00:24,451 epoch 7 - iter 130/138 - loss 0.02278232 - time (sec): 7.17 - samples/sec: 3011.96 - lr: 0.000017 - momentum: 0.000000
178
+ 2023-10-23 15:00:24,880 ----------------------------------------------------------------------------------------------------
179
+ 2023-10-23 15:00:24,881 EPOCH 7 done: loss 0.0232 - lr: 0.000017
180
+ 2023-10-23 15:00:25,413 DEV : loss 0.1645553708076477 - f1-score (micro avg) 0.883
181
+ 2023-10-23 15:00:25,418 ----------------------------------------------------------------------------------------------------
182
+ 2023-10-23 15:00:26,141 epoch 8 - iter 13/138 - loss 0.03132831 - time (sec): 0.72 - samples/sec: 3036.44 - lr: 0.000016 - momentum: 0.000000
183
+ 2023-10-23 15:00:26,851 epoch 8 - iter 26/138 - loss 0.03228913 - time (sec): 1.43 - samples/sec: 3076.10 - lr: 0.000016 - momentum: 0.000000
184
+ 2023-10-23 15:00:27,565 epoch 8 - iter 39/138 - loss 0.02750764 - time (sec): 2.15 - samples/sec: 2978.19 - lr: 0.000015 - momentum: 0.000000
185
+ 2023-10-23 15:00:28,292 epoch 8 - iter 52/138 - loss 0.02386624 - time (sec): 2.87 - samples/sec: 3066.68 - lr: 0.000015 - momentum: 0.000000
186
+ 2023-10-23 15:00:29,005 epoch 8 - iter 65/138 - loss 0.02192605 - time (sec): 3.59 - samples/sec: 3071.37 - lr: 0.000014 - momentum: 0.000000
187
+ 2023-10-23 15:00:29,709 epoch 8 - iter 78/138 - loss 0.01935845 - time (sec): 4.29 - samples/sec: 3028.11 - lr: 0.000014 - momentum: 0.000000
188
+ 2023-10-23 15:00:30,419 epoch 8 - iter 91/138 - loss 0.01854524 - time (sec): 5.00 - samples/sec: 3005.90 - lr: 0.000013 - momentum: 0.000000
189
+ 2023-10-23 15:00:31,127 epoch 8 - iter 104/138 - loss 0.01754654 - time (sec): 5.71 - samples/sec: 3037.43 - lr: 0.000013 - momentum: 0.000000
190
+ 2023-10-23 15:00:31,844 epoch 8 - iter 117/138 - loss 0.01644821 - time (sec): 6.42 - samples/sec: 3044.82 - lr: 0.000012 - momentum: 0.000000
191
+ 2023-10-23 15:00:32,546 epoch 8 - iter 130/138 - loss 0.01595208 - time (sec): 7.13 - samples/sec: 3027.96 - lr: 0.000012 - momentum: 0.000000
192
+ 2023-10-23 15:00:32,977 ----------------------------------------------------------------------------------------------------
193
+ 2023-10-23 15:00:32,977 EPOCH 8 done: loss 0.0162 - lr: 0.000012
194
+ 2023-10-23 15:00:33,512 DEV : loss 0.16538017988204956 - f1-score (micro avg) 0.8924
195
+ 2023-10-23 15:00:33,517 ----------------------------------------------------------------------------------------------------
196
+ 2023-10-23 15:00:34,230 epoch 9 - iter 13/138 - loss 0.02788626 - time (sec): 0.71 - samples/sec: 2745.49 - lr: 0.000011 - momentum: 0.000000
197
+ 2023-10-23 15:00:34,951 epoch 9 - iter 26/138 - loss 0.01954824 - time (sec): 1.43 - samples/sec: 3002.40 - lr: 0.000010 - momentum: 0.000000
198
+ 2023-10-23 15:00:35,659 epoch 9 - iter 39/138 - loss 0.01360116 - time (sec): 2.14 - samples/sec: 2984.85 - lr: 0.000010 - momentum: 0.000000
199
+ 2023-10-23 15:00:36,369 epoch 9 - iter 52/138 - loss 0.01102750 - time (sec): 2.85 - samples/sec: 2994.57 - lr: 0.000009 - momentum: 0.000000
200
+ 2023-10-23 15:00:37,087 epoch 9 - iter 65/138 - loss 0.01089929 - time (sec): 3.57 - samples/sec: 3021.95 - lr: 0.000009 - momentum: 0.000000
201
+ 2023-10-23 15:00:37,808 epoch 9 - iter 78/138 - loss 0.01437039 - time (sec): 4.29 - samples/sec: 3072.07 - lr: 0.000008 - momentum: 0.000000
202
+ 2023-10-23 15:00:38,526 epoch 9 - iter 91/138 - loss 0.01440842 - time (sec): 5.01 - samples/sec: 3068.49 - lr: 0.000008 - momentum: 0.000000
203
+ 2023-10-23 15:00:39,241 epoch 9 - iter 104/138 - loss 0.01322011 - time (sec): 5.72 - samples/sec: 3013.70 - lr: 0.000007 - momentum: 0.000000
204
+ 2023-10-23 15:00:39,951 epoch 9 - iter 117/138 - loss 0.01204757 - time (sec): 6.43 - samples/sec: 2979.71 - lr: 0.000007 - momentum: 0.000000
205
+ 2023-10-23 15:00:40,696 epoch 9 - iter 130/138 - loss 0.01145643 - time (sec): 7.18 - samples/sec: 3030.74 - lr: 0.000006 - momentum: 0.000000
206
+ 2023-10-23 15:00:41,121 ----------------------------------------------------------------------------------------------------
207
+ 2023-10-23 15:00:41,121 EPOCH 9 done: loss 0.0113 - lr: 0.000006
208
+ 2023-10-23 15:00:41,656 DEV : loss 0.17684748768806458 - f1-score (micro avg) 0.8894
209
+ 2023-10-23 15:00:41,661 ----------------------------------------------------------------------------------------------------
210
+ 2023-10-23 15:00:42,380 epoch 10 - iter 13/138 - loss 0.00794737 - time (sec): 0.72 - samples/sec: 2884.96 - lr: 0.000005 - momentum: 0.000000
211
+ 2023-10-23 15:00:43,098 epoch 10 - iter 26/138 - loss 0.00943552 - time (sec): 1.44 - samples/sec: 2951.74 - lr: 0.000005 - momentum: 0.000000
212
+ 2023-10-23 15:00:43,822 epoch 10 - iter 39/138 - loss 0.00724063 - time (sec): 2.16 - samples/sec: 2948.75 - lr: 0.000004 - momentum: 0.000000
213
+ 2023-10-23 15:00:44,535 epoch 10 - iter 52/138 - loss 0.00610859 - time (sec): 2.87 - samples/sec: 2948.72 - lr: 0.000004 - momentum: 0.000000
214
+ 2023-10-23 15:00:45,237 epoch 10 - iter 65/138 - loss 0.00718335 - time (sec): 3.58 - samples/sec: 2912.38 - lr: 0.000003 - momentum: 0.000000
215
+ 2023-10-23 15:00:45,949 epoch 10 - iter 78/138 - loss 0.00609465 - time (sec): 4.29 - samples/sec: 2974.64 - lr: 0.000003 - momentum: 0.000000
216
+ 2023-10-23 15:00:46,661 epoch 10 - iter 91/138 - loss 0.00724821 - time (sec): 5.00 - samples/sec: 2992.73 - lr: 0.000002 - momentum: 0.000000
217
+ 2023-10-23 15:00:47,375 epoch 10 - iter 104/138 - loss 0.00669768 - time (sec): 5.71 - samples/sec: 3002.23 - lr: 0.000002 - momentum: 0.000000
218
+ 2023-10-23 15:00:48,096 epoch 10 - iter 117/138 - loss 0.01007155 - time (sec): 6.43 - samples/sec: 3012.47 - lr: 0.000001 - momentum: 0.000000
219
+ 2023-10-23 15:00:48,811 epoch 10 - iter 130/138 - loss 0.00905841 - time (sec): 7.15 - samples/sec: 3019.54 - lr: 0.000000 - momentum: 0.000000
220
+ 2023-10-23 15:00:49,255 ----------------------------------------------------------------------------------------------------
221
+ 2023-10-23 15:00:49,256 EPOCH 10 done: loss 0.0087 - lr: 0.000000
222
+ 2023-10-23 15:00:49,787 DEV : loss 0.17404906451702118 - f1-score (micro avg) 0.8951
223
+ 2023-10-23 15:00:49,792 saving best model
224
+ 2023-10-23 15:00:50,737 ----------------------------------------------------------------------------------------------------
225
+ 2023-10-23 15:00:50,738 Loading model from best epoch ...
226
+ 2023-10-23 15:00:52,533 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
227
+ 2023-10-23 15:00:53,190
228
+ Results:
229
+ - F-score (micro) 0.9026
230
+ - F-score (macro) 0.8381
231
+ - Accuracy 0.8407
232
+
233
+ By class:
234
+ precision recall f1-score support
235
+
236
+ scope 0.8876 0.8977 0.8927 176
237
+ pers 1.0000 0.9297 0.9636 128
238
+ work 0.8182 0.8514 0.8344 74
239
+ object 1.0000 1.0000 1.0000 2
240
+ loc 0.5000 0.5000 0.5000 2
241
+
242
+ micro avg 0.9074 0.8979 0.9026 382
243
+ macro avg 0.8412 0.8358 0.8381 382
244
+ weighted avg 0.9104 0.8979 0.9036 382
245
+
246
+ 2023-10-23 15:00:53,190 ----------------------------------------------------------------------------------------------------