Hugging Face's logo

stefan-it
/
hmbench-hipe2020-de-hmbert-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2

Token Classification
Flair
PyTorch
German
sequence-tagger-model
Model card Files Community
1
stefan-it commited on
Commit
7fed36d
·
1 Parent(s): 976a8a7

Upload folder using huggingface_hub

Browse files
Files changed (5) hide show
  1. best-model.pt +3 -0
  2. dev.tsv +0 -0
  3. loss.tsv +11 -0
  4. test.tsv +0 -0
  5. training.log +246 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6670141c7c21f7224fa938b927c4d484514c1691f3cbb1392e14960e35845485
3
+ size 443335879
dev.tsv ADDED
The diff for this file is too large to render. See raw diff
 
loss.tsv ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ EPOCH TIMESTAMP LEARNING_RATE TRAIN_LOSS DEV_LOSS DEV_PRECISION DEV_RECALL DEV_F1 DEV_ACCURACY
2
+ 1 12:33:11 0.0000 0.5533 0.1821 0.6872 0.5754 0.6264 0.4655
3
+ 2 12:34:02 0.0000 0.1651 0.1484 0.6739 0.7076 0.6903 0.5455
4
+ 3 12:34:54 0.0000 0.0970 0.2167 0.7806 0.7037 0.7401 0.5992
5
+ 4 12:35:44 0.0000 0.0639 0.1947 0.7330 0.7428 0.7379 0.6028
6
+ 5 12:36:35 0.0000 0.0406 0.2318 0.7648 0.7576 0.7612 0.6304
7
+ 6 12:37:27 0.0000 0.0343 0.2459 0.7658 0.7670 0.7664 0.6395
8
+ 7 12:38:19 0.0000 0.0206 0.2359 0.7541 0.7819 0.7678 0.6402
9
+ 8 12:39:10 0.0000 0.0133 0.2714 0.7726 0.7678 0.7702 0.6456
10
+ 9 12:40:03 0.0000 0.0098 0.2518 0.7923 0.7756 0.7839 0.6609
11
+ 10 12:40:56 0.0000 0.0048 0.2590 0.7894 0.7826 0.7860 0.6629
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,246 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-13 12:32:19,278 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-13 12:32:19,279 Model: "SequenceTagger(
3
+ (embeddings): TransformerWordEmbeddings(
4
+ (model): BertModel(
5
+ (embeddings): BertEmbeddings(
6
+ (word_embeddings): Embedding(32001, 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=21, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-13 12:32:19,279 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-13 12:32:19,279 MultiCorpus: 3575 train + 1235 dev + 1266 test sentences
52
+ - NER_HIPE_2022 Corpus: 3575 train + 1235 dev + 1266 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/hipe2020/de/with_doc_seperator
53
+ 2023-10-13 12:32:19,279 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-13 12:32:19,279 Train: 3575 sentences
55
+ 2023-10-13 12:32:19,279 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-13 12:32:19,279 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-13 12:32:19,279 Training Params:
58
+ 2023-10-13 12:32:19,279 - learning_rate: "5e-05"
59
+ 2023-10-13 12:32:19,279 - mini_batch_size: "4"
60
+ 2023-10-13 12:32:19,279 - max_epochs: "10"
61
+ 2023-10-13 12:32:19,279 - shuffle: "True"
62
+ 2023-10-13 12:32:19,279 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-13 12:32:19,279 Plugins:
64
+ 2023-10-13 12:32:19,280 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-13 12:32:19,280 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-13 12:32:19,280 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-13 12:32:19,280 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-13 12:32:19,280 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-13 12:32:19,280 Computation:
70
+ 2023-10-13 12:32:19,280 - compute on device: cuda:0
71
+ 2023-10-13 12:32:19,280 - embedding storage: none
72
+ 2023-10-13 12:32:19,280 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-13 12:32:19,280 Model training base path: "hmbench-hipe2020/de-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2"
74
+ 2023-10-13 12:32:19,280 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-13 12:32:19,280 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-13 12:32:23,751 epoch 1 - iter 89/894 - loss 2.85983461 - time (sec): 4.47 - samples/sec: 1792.74 - lr: 0.000005 - momentum: 0.000000
77
+ 2023-10-13 12:32:28,441 epoch 1 - iter 178/894 - loss 1.72346897 - time (sec): 9.16 - samples/sec: 1753.15 - lr: 0.000010 - momentum: 0.000000
78
+ 2023-10-13 12:32:33,078 epoch 1 - iter 267/894 - loss 1.24732200 - time (sec): 13.80 - samples/sec: 1807.21 - lr: 0.000015 - momentum: 0.000000
79
+ 2023-10-13 12:32:37,557 epoch 1 - iter 356/894 - loss 1.03004553 - time (sec): 18.28 - samples/sec: 1807.56 - lr: 0.000020 - momentum: 0.000000
80
+ 2023-10-13 12:32:42,240 epoch 1 - iter 445/894 - loss 0.87776144 - time (sec): 22.96 - samples/sec: 1824.47 - lr: 0.000025 - momentum: 0.000000
81
+ 2023-10-13 12:32:47,145 epoch 1 - iter 534/894 - loss 0.76230451 - time (sec): 27.86 - samples/sec: 1873.14 - lr: 0.000030 - momentum: 0.000000
82
+ 2023-10-13 12:32:52,025 epoch 1 - iter 623/894 - loss 0.69436195 - time (sec): 32.74 - samples/sec: 1846.87 - lr: 0.000035 - momentum: 0.000000
83
+ 2023-10-13 12:32:56,787 epoch 1 - iter 712/894 - loss 0.63775811 - time (sec): 37.51 - samples/sec: 1846.05 - lr: 0.000040 - momentum: 0.000000
84
+ 2023-10-13 12:33:01,489 epoch 1 - iter 801/894 - loss 0.59768895 - time (sec): 42.21 - samples/sec: 1829.53 - lr: 0.000045 - momentum: 0.000000
85
+ 2023-10-13 12:33:06,315 epoch 1 - iter 890/894 - loss 0.55580460 - time (sec): 47.03 - samples/sec: 1829.34 - lr: 0.000050 - momentum: 0.000000
86
+ 2023-10-13 12:33:06,492 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-13 12:33:06,493 EPOCH 1 done: loss 0.5533 - lr: 0.000050
88
+ 2023-10-13 12:33:11,352 DEV : loss 0.18206389248371124 - f1-score (micro avg) 0.6264
89
+ 2023-10-13 12:33:11,379 saving best model
90
+ 2023-10-13 12:33:11,748 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-13 12:33:15,811 epoch 2 - iter 89/894 - loss 0.20901231 - time (sec): 4.06 - samples/sec: 2117.99 - lr: 0.000049 - momentum: 0.000000
92
+ 2023-10-13 12:33:19,898 epoch 2 - iter 178/894 - loss 0.19760510 - time (sec): 8.15 - samples/sec: 2090.27 - lr: 0.000049 - momentum: 0.000000
93
+ 2023-10-13 12:33:23,982 epoch 2 - iter 267/894 - loss 0.19189895 - time (sec): 12.23 - samples/sec: 2062.14 - lr: 0.000048 - momentum: 0.000000
94
+ 2023-10-13 12:33:28,119 epoch 2 - iter 356/894 - loss 0.18761752 - time (sec): 16.37 - samples/sec: 2080.74 - lr: 0.000048 - momentum: 0.000000
95
+ 2023-10-13 12:33:32,271 epoch 2 - iter 445/894 - loss 0.18116435 - time (sec): 20.52 - samples/sec: 2055.03 - lr: 0.000047 - momentum: 0.000000
96
+ 2023-10-13 12:33:36,461 epoch 2 - iter 534/894 - loss 0.17830011 - time (sec): 24.71 - samples/sec: 2083.01 - lr: 0.000047 - momentum: 0.000000
97
+ 2023-10-13 12:33:40,787 epoch 2 - iter 623/894 - loss 0.17220591 - time (sec): 29.04 - samples/sec: 2059.70 - lr: 0.000046 - momentum: 0.000000
98
+ 2023-10-13 12:33:45,142 epoch 2 - iter 712/894 - loss 0.16726249 - time (sec): 33.39 - samples/sec: 2070.36 - lr: 0.000046 - momentum: 0.000000
99
+ 2023-10-13 12:33:49,272 epoch 2 - iter 801/894 - loss 0.16456063 - time (sec): 37.52 - samples/sec: 2075.68 - lr: 0.000045 - momentum: 0.000000
100
+ 2023-10-13 12:33:53,170 epoch 2 - iter 890/894 - loss 0.16515628 - time (sec): 41.42 - samples/sec: 2079.86 - lr: 0.000044 - momentum: 0.000000
101
+ 2023-10-13 12:33:53,343 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-13 12:33:53,343 EPOCH 2 done: loss 0.1651 - lr: 0.000044
103
+ 2023-10-13 12:34:02,186 DEV : loss 0.14844626188278198 - f1-score (micro avg) 0.6903
104
+ 2023-10-13 12:34:02,216 saving best model
105
+ 2023-10-13 12:34:02,669 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-13 12:34:07,048 epoch 3 - iter 89/894 - loss 0.08584339 - time (sec): 4.38 - samples/sec: 1971.04 - lr: 0.000044 - momentum: 0.000000
107
+ 2023-10-13 12:34:11,408 epoch 3 - iter 178/894 - loss 0.08458851 - time (sec): 8.74 - samples/sec: 2090.91 - lr: 0.000043 - momentum: 0.000000
108
+ 2023-10-13 12:34:15,589 epoch 3 - iter 267/894 - loss 0.08982643 - time (sec): 12.92 - samples/sec: 2129.73 - lr: 0.000043 - momentum: 0.000000
109
+ 2023-10-13 12:34:19,889 epoch 3 - iter 356/894 - loss 0.08300527 - time (sec): 17.22 - samples/sec: 2118.70 - lr: 0.000042 - momentum: 0.000000
110
+ 2023-10-13 12:34:24,349 epoch 3 - iter 445/894 - loss 0.09461899 - time (sec): 21.68 - samples/sec: 2097.72 - lr: 0.000042 - momentum: 0.000000
111
+ 2023-10-13 12:34:28,556 epoch 3 - iter 534/894 - loss 0.09779057 - time (sec): 25.89 - samples/sec: 2053.48 - lr: 0.000041 - momentum: 0.000000
112
+ 2023-10-13 12:34:32,893 epoch 3 - iter 623/894 - loss 0.09502757 - time (sec): 30.22 - samples/sec: 2036.10 - lr: 0.000041 - momentum: 0.000000
113
+ 2023-10-13 12:34:37,045 epoch 3 - iter 712/894 - loss 0.09421525 - time (sec): 34.37 - samples/sec: 2022.03 - lr: 0.000040 - momentum: 0.000000
114
+ 2023-10-13 12:34:41,215 epoch 3 - iter 801/894 - loss 0.09728203 - time (sec): 38.54 - samples/sec: 2020.67 - lr: 0.000039 - momentum: 0.000000
115
+ 2023-10-13 12:34:45,220 epoch 3 - iter 890/894 - loss 0.09737888 - time (sec): 42.55 - samples/sec: 2024.31 - lr: 0.000039 - momentum: 0.000000
116
+ 2023-10-13 12:34:45,395 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-13 12:34:45,395 EPOCH 3 done: loss 0.0970 - lr: 0.000039
118
+ 2023-10-13 12:34:54,092 DEV : loss 0.21671560406684875 - f1-score (micro avg) 0.7401
119
+ 2023-10-13 12:34:54,121 saving best model
120
+ 2023-10-13 12:34:54,564 ----------------------------------------------------------------------------------------------------
121
+ 2023-10-13 12:34:58,507 epoch 4 - iter 89/894 - loss 0.06239486 - time (sec): 3.94 - samples/sec: 1934.14 - lr: 0.000038 - momentum: 0.000000
122
+ 2023-10-13 12:35:02,740 epoch 4 - iter 178/894 - loss 0.05659001 - time (sec): 8.17 - samples/sec: 2076.86 - lr: 0.000038 - momentum: 0.000000
123
+ 2023-10-13 12:35:06,896 epoch 4 - iter 267/894 - loss 0.07314064 - time (sec): 12.33 - samples/sec: 2052.34 - lr: 0.000037 - momentum: 0.000000
124
+ 2023-10-13 12:35:11,114 epoch 4 - iter 356/894 - loss 0.06967169 - time (sec): 16.55 - samples/sec: 2051.60 - lr: 0.000037 - momentum: 0.000000
125
+ 2023-10-13 12:35:15,115 epoch 4 - iter 445/894 - loss 0.06814429 - time (sec): 20.55 - samples/sec: 2023.19 - lr: 0.000036 - momentum: 0.000000
126
+ 2023-10-13 12:35:19,600 epoch 4 - iter 534/894 - loss 0.06312653 - time (sec): 25.04 - samples/sec: 2074.53 - lr: 0.000036 - momentum: 0.000000
127
+ 2023-10-13 12:35:23,863 epoch 4 - iter 623/894 - loss 0.06341301 - time (sec): 29.30 - samples/sec: 2064.67 - lr: 0.000035 - momentum: 0.000000
128
+ 2023-10-13 12:35:27,776 epoch 4 - iter 712/894 - loss 0.06424281 - time (sec): 33.21 - samples/sec: 2065.66 - lr: 0.000034 - momentum: 0.000000
129
+ 2023-10-13 12:35:31,743 epoch 4 - iter 801/894 - loss 0.06543736 - time (sec): 37.18 - samples/sec: 2092.22 - lr: 0.000034 - momentum: 0.000000
130
+ 2023-10-13 12:35:35,761 epoch 4 - iter 890/894 - loss 0.06405722 - time (sec): 41.20 - samples/sec: 2093.77 - lr: 0.000033 - momentum: 0.000000
131
+ 2023-10-13 12:35:35,945 ----------------------------------------------------------------------------------------------------
132
+ 2023-10-13 12:35:35,945 EPOCH 4 done: loss 0.0639 - lr: 0.000033
133
+ 2023-10-13 12:35:44,595 DEV : loss 0.1947321891784668 - f1-score (micro avg) 0.7379
134
+ 2023-10-13 12:35:44,623 ----------------------------------------------------------------------------------------------------
135
+ 2023-10-13 12:35:48,550 epoch 5 - iter 89/894 - loss 0.05096803 - time (sec): 3.93 - samples/sec: 2072.36 - lr: 0.000033 - momentum: 0.000000
136
+ 2023-10-13 12:35:52,426 epoch 5 - iter 178/894 - loss 0.04532651 - time (sec): 7.80 - samples/sec: 2056.55 - lr: 0.000032 - momentum: 0.000000
137
+ 2023-10-13 12:35:56,464 epoch 5 - iter 267/894 - loss 0.04436703 - time (sec): 11.84 - samples/sec: 2090.11 - lr: 0.000032 - momentum: 0.000000
138
+ 2023-10-13 12:36:00,970 epoch 5 - iter 356/894 - loss 0.04213125 - time (sec): 16.35 - samples/sec: 2086.23 - lr: 0.000031 - momentum: 0.000000
139
+ 2023-10-13 12:36:05,425 epoch 5 - iter 445/894 - loss 0.04195800 - time (sec): 20.80 - samples/sec: 2067.23 - lr: 0.000031 - momentum: 0.000000
140
+ 2023-10-13 12:36:09,548 epoch 5 - iter 534/894 - loss 0.04254711 - time (sec): 24.92 - samples/sec: 2062.82 - lr: 0.000030 - momentum: 0.000000
141
+ 2023-10-13 12:36:13,943 epoch 5 - iter 623/894 - loss 0.04283399 - time (sec): 29.32 - samples/sec: 2062.46 - lr: 0.000029 - momentum: 0.000000
142
+ 2023-10-13 12:36:18,296 epoch 5 - iter 712/894 - loss 0.04335760 - time (sec): 33.67 - samples/sec: 2073.39 - lr: 0.000029 - momentum: 0.000000
143
+ 2023-10-13 12:36:22,427 epoch 5 - iter 801/894 - loss 0.04107727 - time (sec): 37.80 - samples/sec: 2074.95 - lr: 0.000028 - momentum: 0.000000
144
+ 2023-10-13 12:36:26,403 epoch 5 - iter 890/894 - loss 0.04070854 - time (sec): 41.78 - samples/sec: 2063.32 - lr: 0.000028 - momentum: 0.000000
145
+ 2023-10-13 12:36:26,582 ----------------------------------------------------------------------------------------------------
146
+ 2023-10-13 12:36:26,582 EPOCH 5 done: loss 0.0406 - lr: 0.000028
147
+ 2023-10-13 12:36:35,756 DEV : loss 0.2318100780248642 - f1-score (micro avg) 0.7612
148
+ 2023-10-13 12:36:35,787 saving best model
149
+ 2023-10-13 12:36:36,173 ----------------------------------------------------------------------------------------------------
150
+ 2023-10-13 12:36:40,528 epoch 6 - iter 89/894 - loss 0.02285241 - time (sec): 4.35 - samples/sec: 1993.67 - lr: 0.000027 - momentum: 0.000000
151
+ 2023-10-13 12:36:44,740 epoch 6 - iter 178/894 - loss 0.02994531 - time (sec): 8.57 - samples/sec: 1977.19 - lr: 0.000027 - momentum: 0.000000
152
+ 2023-10-13 12:36:48,963 epoch 6 - iter 267/894 - loss 0.03140204 - time (sec): 12.79 - samples/sec: 1961.50 - lr: 0.000026 - momentum: 0.000000
153
+ 2023-10-13 12:36:53,037 epoch 6 - iter 356/894 - loss 0.03042671 - time (sec): 16.86 - samples/sec: 2000.69 - lr: 0.000026 - momentum: 0.000000
154
+ 2023-10-13 12:36:57,485 epoch 6 - iter 445/894 - loss 0.03182242 - time (sec): 21.31 - samples/sec: 1970.17 - lr: 0.000025 - momentum: 0.000000
155
+ 2023-10-13 12:37:01,561 epoch 6 - iter 534/894 - loss 0.03205269 - time (sec): 25.39 - samples/sec: 1986.79 - lr: 0.000024 - momentum: 0.000000
156
+ 2023-10-13 12:37:05,724 epoch 6 - iter 623/894 - loss 0.03194324 - time (sec): 29.55 - samples/sec: 1982.14 - lr: 0.000024 - momentum: 0.000000
157
+ 2023-10-13 12:37:10,061 epoch 6 - iter 712/894 - loss 0.03214632 - time (sec): 33.89 - samples/sec: 1976.66 - lr: 0.000023 - momentum: 0.000000
158
+ 2023-10-13 12:37:14,206 epoch 6 - iter 801/894 - loss 0.03639544 - time (sec): 38.03 - samples/sec: 1997.06 - lr: 0.000023 - momentum: 0.000000
159
+ 2023-10-13 12:37:18,693 epoch 6 - iter 890/894 - loss 0.03436798 - time (sec): 42.52 - samples/sec: 2022.52 - lr: 0.000022 - momentum: 0.000000
160
+ 2023-10-13 12:37:18,891 ----------------------------------------------------------------------------------------------------
161
+ 2023-10-13 12:37:18,891 EPOCH 6 done: loss 0.0343 - lr: 0.000022
162
+ 2023-10-13 12:37:27,725 DEV : loss 0.2459443360567093 - f1-score (micro avg) 0.7664
163
+ 2023-10-13 12:37:27,753 saving best model
164
+ 2023-10-13 12:37:28,192 ----------------------------------------------------------------------------------------------------
165
+ 2023-10-13 12:37:32,358 epoch 7 - iter 89/894 - loss 0.02901541 - time (sec): 4.16 - samples/sec: 2086.25 - lr: 0.000022 - momentum: 0.000000
166
+ 2023-10-13 12:37:37,075 epoch 7 - iter 178/894 - loss 0.02057225 - time (sec): 8.88 - samples/sec: 1942.97 - lr: 0.000021 - momentum: 0.000000
167
+ 2023-10-13 12:37:41,718 epoch 7 - iter 267/894 - loss 0.02355731 - time (sec): 13.52 - samples/sec: 1944.79 - lr: 0.000021 - momentum: 0.000000
168
+ 2023-10-13 12:37:45,887 epoch 7 - iter 356/894 - loss 0.02226925 - time (sec): 17.69 - samples/sec: 1983.66 - lr: 0.000020 - momentum: 0.000000
169
+ 2023-10-13 12:37:49,779 epoch 7 - iter 445/894 - loss 0.02176302 - time (sec): 21.58 - samples/sec: 2001.58 - lr: 0.000019 - momentum: 0.000000
170
+ 2023-10-13 12:37:54,061 epoch 7 - iter 534/894 - loss 0.02088338 - time (sec): 25.87 - samples/sec: 2002.74 - lr: 0.000019 - momentum: 0.000000
171
+ 2023-10-13 12:37:58,011 epoch 7 - iter 623/894 - loss 0.02136378 - time (sec): 29.82 - samples/sec: 2014.53 - lr: 0.000018 - momentum: 0.000000
172
+ 2023-10-13 12:38:01,910 epoch 7 - iter 712/894 - loss 0.02093973 - time (sec): 33.72 - samples/sec: 2030.45 - lr: 0.000018 - momentum: 0.000000
173
+ 2023-10-13 12:38:05,780 epoch 7 - iter 801/894 - loss 0.02131084 - time (sec): 37.59 - samples/sec: 2027.97 - lr: 0.000017 - momentum: 0.000000
174
+ 2023-10-13 12:38:10,100 epoch 7 - iter 890/894 - loss 0.02072021 - time (sec): 41.91 - samples/sec: 2053.61 - lr: 0.000017 - momentum: 0.000000
175
+ 2023-10-13 12:38:10,297 ----------------------------------------------------------------------------------------------------
176
+ 2023-10-13 12:38:10,297 EPOCH 7 done: loss 0.0206 - lr: 0.000017
177
+ 2023-10-13 12:38:19,433 DEV : loss 0.23593628406524658 - f1-score (micro avg) 0.7678
178
+ 2023-10-13 12:38:19,472 saving best model
179
+ 2023-10-13 12:38:20,009 ----------------------------------------------------------------------------------------------------
180
+ 2023-10-13 12:38:24,729 epoch 8 - iter 89/894 - loss 0.01361398 - time (sec): 4.72 - samples/sec: 1779.60 - lr: 0.000016 - momentum: 0.000000
181
+ 2023-10-13 12:38:29,248 epoch 8 - iter 178/894 - loss 0.01742511 - time (sec): 9.24 - samples/sec: 1957.94 - lr: 0.000016 - momentum: 0.000000
182
+ 2023-10-13 12:38:33,322 epoch 8 - iter 267/894 - loss 0.01587200 - time (sec): 13.31 - samples/sec: 1983.09 - lr: 0.000015 - momentum: 0.000000
183
+ 2023-10-13 12:38:37,457 epoch 8 - iter 356/894 - loss 0.01499465 - time (sec): 17.45 - samples/sec: 2007.73 - lr: 0.000014 - momentum: 0.000000
184
+ 2023-10-13 12:38:41,570 epoch 8 - iter 445/894 - loss 0.01522297 - time (sec): 21.56 - samples/sec: 1985.39 - lr: 0.000014 - momentum: 0.000000
185
+ 2023-10-13 12:38:45,637 epoch 8 - iter 534/894 - loss 0.01336557 - time (sec): 25.63 - samples/sec: 2019.99 - lr: 0.000013 - momentum: 0.000000
186
+ 2023-10-13 12:38:49,596 epoch 8 - iter 623/894 - loss 0.01367776 - time (sec): 29.58 - samples/sec: 2048.84 - lr: 0.000013 - momentum: 0.000000
187
+ 2023-10-13 12:38:53,562 epoch 8 - iter 712/894 - loss 0.01386286 - time (sec): 33.55 - samples/sec: 2053.20 - lr: 0.000012 - momentum: 0.000000
188
+ 2023-10-13 12:38:57,493 epoch 8 - iter 801/894 - loss 0.01362775 - time (sec): 37.48 - samples/sec: 2067.50 - lr: 0.000012 - momentum: 0.000000
189
+ 2023-10-13 12:39:01,513 epoch 8 - iter 890/894 - loss 0.01332056 - time (sec): 41.50 - samples/sec: 2077.00 - lr: 0.000011 - momentum: 0.000000
190
+ 2023-10-13 12:39:01,685 ----------------------------------------------------------------------------------------------------
191
+ 2023-10-13 12:39:01,685 EPOCH 8 done: loss 0.0133 - lr: 0.000011
192
+ 2023-10-13 12:39:10,171 DEV : loss 0.2714207172393799 - f1-score (micro avg) 0.7702
193
+ 2023-10-13 12:39:10,205 saving best model
194
+ 2023-10-13 12:39:10,686 ----------------------------------------------------------------------------------------------------
195
+ 2023-10-13 12:39:15,078 epoch 9 - iter 89/894 - loss 0.00547130 - time (sec): 4.39 - samples/sec: 1971.39 - lr: 0.000011 - momentum: 0.000000
196
+ 2023-10-13 12:39:19,244 epoch 9 - iter 178/894 - loss 0.00963629 - time (sec): 8.56 - samples/sec: 1993.56 - lr: 0.000010 - momentum: 0.000000
197
+ 2023-10-13 12:39:23,663 epoch 9 - iter 267/894 - loss 0.01066901 - time (sec): 12.97 - samples/sec: 1953.09 - lr: 0.000009 - momentum: 0.000000
198
+ 2023-10-13 12:39:27,955 epoch 9 - iter 356/894 - loss 0.01008720 - time (sec): 17.27 - samples/sec: 1972.23 - lr: 0.000009 - momentum: 0.000000
199
+ 2023-10-13 12:39:32,716 epoch 9 - iter 445/894 - loss 0.00899087 - time (sec): 22.03 - samples/sec: 1989.31 - lr: 0.000008 - momentum: 0.000000
200
+ 2023-10-13 12:39:37,235 epoch 9 - iter 534/894 - loss 0.00832282 - time (sec): 26.55 - samples/sec: 1967.26 - lr: 0.000008 - momentum: 0.000000
201
+ 2023-10-13 12:39:41,682 epoch 9 - iter 623/894 - loss 0.00840774 - time (sec): 30.99 - samples/sec: 1954.70 - lr: 0.000007 - momentum: 0.000000
202
+ 2023-10-13 12:39:46,042 epoch 9 - iter 712/894 - loss 0.00907033 - time (sec): 35.35 - samples/sec: 1965.05 - lr: 0.000007 - momentum: 0.000000
203
+ 2023-10-13 12:39:50,191 epoch 9 - iter 801/894 - loss 0.00943423 - time (sec): 39.50 - samples/sec: 1962.69 - lr: 0.000006 - momentum: 0.000000
204
+ 2023-10-13 12:39:54,425 epoch 9 - iter 890/894 - loss 0.00956840 - time (sec): 43.74 - samples/sec: 1970.62 - lr: 0.000006 - momentum: 0.000000
205
+ 2023-10-13 12:39:54,626 ----------------------------------------------------------------------------------------------------
206
+ 2023-10-13 12:39:54,627 EPOCH 9 done: loss 0.0098 - lr: 0.000006
207
+ 2023-10-13 12:40:03,411 DEV : loss 0.25177356600761414 - f1-score (micro avg) 0.7839
208
+ 2023-10-13 12:40:03,439 saving best model
209
+ 2023-10-13 12:40:03,879 ----------------------------------------------------------------------------------------------------
210
+ 2023-10-13 12:40:08,201 epoch 10 - iter 89/894 - loss 0.00028415 - time (sec): 4.32 - samples/sec: 2139.25 - lr: 0.000005 - momentum: 0.000000
211
+ 2023-10-13 12:40:12,369 epoch 10 - iter 178/894 - loss 0.00242940 - time (sec): 8.49 - samples/sec: 2042.29 - lr: 0.000004 - momentum: 0.000000
212
+ 2023-10-13 12:40:16,627 epoch 10 - iter 267/894 - loss 0.00404990 - time (sec): 12.75 - samples/sec: 2020.69 - lr: 0.000004 - momentum: 0.000000
213
+ 2023-10-13 12:40:20,968 epoch 10 - iter 356/894 - loss 0.00350026 - time (sec): 17.09 - samples/sec: 2079.62 - lr: 0.000003 - momentum: 0.000000
214
+ 2023-10-13 12:40:25,142 epoch 10 - iter 445/894 - loss 0.00462522 - time (sec): 21.26 - samples/sec: 2064.72 - lr: 0.000003 - momentum: 0.000000
215
+ 2023-10-13 12:40:29,316 epoch 10 - iter 534/894 - loss 0.00559551 - time (sec): 25.44 - samples/sec: 2060.31 - lr: 0.000002 - momentum: 0.000000
216
+ 2023-10-13 12:40:33,584 epoch 10 - iter 623/894 - loss 0.00545572 - time (sec): 29.70 - samples/sec: 2024.32 - lr: 0.000002 - momentum: 0.000000
217
+ 2023-10-13 12:40:38,084 epoch 10 - iter 712/894 - loss 0.00477820 - time (sec): 34.20 - samples/sec: 2010.58 - lr: 0.000001 - momentum: 0.000000
218
+ 2023-10-13 12:40:42,361 epoch 10 - iter 801/894 - loss 0.00505387 - time (sec): 38.48 - samples/sec: 1999.45 - lr: 0.000001 - momentum: 0.000000
219
+ 2023-10-13 12:40:46,742 epoch 10 - iter 890/894 - loss 0.00486550 - time (sec): 42.86 - samples/sec: 2012.22 - lr: 0.000000 - momentum: 0.000000
220
+ 2023-10-13 12:40:46,927 ----------------------------------------------------------------------------------------------------
221
+ 2023-10-13 12:40:46,927 EPOCH 10 done: loss 0.0048 - lr: 0.000000
222
+ 2023-10-13 12:40:56,024 DEV : loss 0.2590446472167969 - f1-score (micro avg) 0.786
223
+ 2023-10-13 12:40:56,051 saving best model
224
+ 2023-10-13 12:40:56,820 ----------------------------------------------------------------------------------------------------
225
+ 2023-10-13 12:40:56,822 Loading model from best epoch ...
226
+ 2023-10-13 12:40:58,285 SequenceTagger predicts: Dictionary with 21 tags: O, S-loc, B-loc, E-loc, I-loc, S-pers, B-pers, E-pers, I-pers, S-org, B-org, E-org, I-org, S-prod, B-prod, E-prod, I-prod, S-time, B-time, E-time, I-time
227
+ 2023-10-13 12:41:03,540
228
+ Results:
229
+ - F-score (micro) 0.7304
230
+ - F-score (macro) 0.6539
231
+ - Accuracy 0.5962
232
+
233
+ By class:
234
+ precision recall f1-score support
235
+
236
+ loc 0.8065 0.8322 0.8192 596
237
+ pers 0.6715 0.6937 0.6824 333
238
+ org 0.5800 0.4394 0.5000 132
239
+ prod 0.6735 0.5000 0.5739 66
240
+ time 0.6939 0.6939 0.6939 49
241
+
242
+ micro avg 0.7364 0.7245 0.7304 1176
243
+ macro avg 0.6851 0.6318 0.6539 1176
244
+ weighted avg 0.7307 0.7245 0.7256 1176
245
+
246
+ 2023-10-13 12:41:03,540 ----------------------------------------------------------------------------------------------------