Hugging Face's logo

stefan-it
/
hmbench-newseye-de-hmbert-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4

Token Classification
Flair
PyTorch
German
sequence-tagger-model
Model card Files Community
2
stefan-it commited on
Commit
635abce
•
1 Parent(s): 5531109

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 +240 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1993b3b3ca00bbabd116839d0c9e2d8c6f84f0bf3df92e96d52243c18dad5107
3
+ size 443323527
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 20:29:01 0.0000 0.3902 0.1364 0.1861 0.4924 0.2701 0.1577
3
+ 2 20:33:21 0.0000 0.1697 0.1600 0.2863 0.5189 0.3690 0.2276
4
+ 3 20:37:43 0.0000 0.1166 0.2202 0.2583 0.4110 0.3173 0.1895
5
+ 4 20:42:03 0.0000 0.0858 0.2818 0.2502 0.5701 0.3478 0.2114
6
+ 5 20:46:25 0.0000 0.0600 0.2872 0.2758 0.5417 0.3655 0.2245
7
+ 6 20:50:47 0.0000 0.0446 0.3866 0.2624 0.6004 0.3652 0.2247
8
+ 7 20:55:09 0.0000 0.0318 0.4148 0.2722 0.5871 0.3719 0.2293
9
+ 8 20:59:30 0.0000 0.0234 0.4218 0.2725 0.5966 0.3741 0.2313
10
+ 9 21:03:52 0.0000 0.0148 0.4862 0.2630 0.5833 0.3626 0.2227
11
+ 10 21:08:12 0.0000 0.0110 0.4902 0.2607 0.6023 0.3638 0.2238
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,240 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-15 20:24:44,673 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-15 20:24:44,674 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=17, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-15 20:24:44,674 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-15 20:24:44,675 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
+ 2023-10-15 20:24:44,675 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-15 20:24:44,675 Train: 20847 sentences
55
+ 2023-10-15 20:24:44,675 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-15 20:24:44,675 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-15 20:24:44,675 Training Params:
58
+ 2023-10-15 20:24:44,675 - learning_rate: "3e-05"
59
+ 2023-10-15 20:24:44,675 - mini_batch_size: "4"
60
+ 2023-10-15 20:24:44,675 - max_epochs: "10"
61
+ 2023-10-15 20:24:44,675 - shuffle: "True"
62
+ 2023-10-15 20:24:44,675 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-15 20:24:44,675 Plugins:
64
+ 2023-10-15 20:24:44,675 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-15 20:24:44,675 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-15 20:24:44,675 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-15 20:24:44,675 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-15 20:24:44,675 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-15 20:24:44,675 Computation:
70
+ 2023-10-15 20:24:44,675 - compute on device: cuda:0
71
+ 2023-10-15 20:24:44,675 - embedding storage: none
72
+ 2023-10-15 20:24:44,675 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-15 20:24:44,675 Model training base path: "hmbench-newseye/de-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4"
74
+ 2023-10-15 20:24:44,675 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-15 20:24:44,675 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-15 20:25:09,747 epoch 1 - iter 521/5212 - loss 1.56165169 - time (sec): 25.07 - samples/sec: 1394.68 - lr: 0.000003 - momentum: 0.000000
77
+ 2023-10-15 20:25:35,194 epoch 1 - iter 1042/5212 - loss 0.99845314 - time (sec): 50.52 - samples/sec: 1454.90 - lr: 0.000006 - momentum: 0.000000
78
+ 2023-10-15 20:26:00,007 epoch 1 - iter 1563/5212 - loss 0.77898021 - time (sec): 75.33 - samples/sec: 1445.20 - lr: 0.000009 - momentum: 0.000000
79
+ 2023-10-15 20:26:25,375 epoch 1 - iter 2084/5212 - loss 0.65478772 - time (sec): 100.70 - samples/sec: 1439.43 - lr: 0.000012 - momentum: 0.000000
80
+ 2023-10-15 20:26:50,728 epoch 1 - iter 2605/5212 - loss 0.57005146 - time (sec): 126.05 - samples/sec: 1457.86 - lr: 0.000015 - momentum: 0.000000
81
+ 2023-10-15 20:27:15,361 epoch 1 - iter 3126/5212 - loss 0.51705867 - time (sec): 150.68 - samples/sec: 1452.88 - lr: 0.000018 - momentum: 0.000000
82
+ 2023-10-15 20:27:40,262 epoch 1 - iter 3647/5212 - loss 0.47447550 - time (sec): 175.59 - samples/sec: 1451.31 - lr: 0.000021 - momentum: 0.000000
83
+ 2023-10-15 20:28:05,433 epoch 1 - iter 4168/5212 - loss 0.43833701 - time (sec): 200.76 - samples/sec: 1455.45 - lr: 0.000024 - momentum: 0.000000
84
+ 2023-10-15 20:28:30,449 epoch 1 - iter 4689/5212 - loss 0.41315675 - time (sec): 225.77 - samples/sec: 1455.37 - lr: 0.000027 - momentum: 0.000000
85
+ 2023-10-15 20:28:55,738 epoch 1 - iter 5210/5212 - loss 0.39022797 - time (sec): 251.06 - samples/sec: 1463.30 - lr: 0.000030 - momentum: 0.000000
86
+ 2023-10-15 20:28:55,827 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-15 20:28:55,827 EPOCH 1 done: loss 0.3902 - lr: 0.000030
88
+ 2023-10-15 20:29:01,505 DEV : loss 0.13643652200698853 - f1-score (micro avg) 0.2701
89
+ 2023-10-15 20:29:01,533 saving best model
90
+ 2023-10-15 20:29:01,913 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-15 20:29:27,078 epoch 2 - iter 521/5212 - loss 0.19544052 - time (sec): 25.16 - samples/sec: 1514.57 - lr: 0.000030 - momentum: 0.000000
92
+ 2023-10-15 20:29:52,579 epoch 2 - iter 1042/5212 - loss 0.17228581 - time (sec): 50.66 - samples/sec: 1505.74 - lr: 0.000029 - momentum: 0.000000
93
+ 2023-10-15 20:30:18,392 epoch 2 - iter 1563/5212 - loss 0.17467943 - time (sec): 76.48 - samples/sec: 1484.11 - lr: 0.000029 - momentum: 0.000000
94
+ 2023-10-15 20:30:43,402 epoch 2 - iter 2084/5212 - loss 0.17589538 - time (sec): 101.49 - samples/sec: 1461.85 - lr: 0.000029 - momentum: 0.000000
95
+ 2023-10-15 20:31:08,492 epoch 2 - iter 2605/5212 - loss 0.17733174 - time (sec): 126.58 - samples/sec: 1466.95 - lr: 0.000028 - momentum: 0.000000
96
+ 2023-10-15 20:31:33,239 epoch 2 - iter 3126/5212 - loss 0.17324370 - time (sec): 151.32 - samples/sec: 1462.37 - lr: 0.000028 - momentum: 0.000000
97
+ 2023-10-15 20:31:58,539 epoch 2 - iter 3647/5212 - loss 0.16913298 - time (sec): 176.62 - samples/sec: 1471.29 - lr: 0.000028 - momentum: 0.000000
98
+ 2023-10-15 20:32:23,133 epoch 2 - iter 4168/5212 - loss 0.17368275 - time (sec): 201.22 - samples/sec: 1467.38 - lr: 0.000027 - momentum: 0.000000
99
+ 2023-10-15 20:32:48,579 epoch 2 - iter 4689/5212 - loss 0.17139323 - time (sec): 226.66 - samples/sec: 1472.89 - lr: 0.000027 - momentum: 0.000000
100
+ 2023-10-15 20:33:13,124 epoch 2 - iter 5210/5212 - loss 0.16968841 - time (sec): 251.21 - samples/sec: 1462.47 - lr: 0.000027 - momentum: 0.000000
101
+ 2023-10-15 20:33:13,213 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-15 20:33:13,213 EPOCH 2 done: loss 0.1697 - lr: 0.000027
103
+ 2023-10-15 20:33:21,474 DEV : loss 0.1600484699010849 - f1-score (micro avg) 0.369
104
+ 2023-10-15 20:33:21,503 saving best model
105
+ 2023-10-15 20:33:21,991 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-15 20:33:46,977 epoch 3 - iter 521/5212 - loss 0.11625189 - time (sec): 24.98 - samples/sec: 1460.19 - lr: 0.000026 - momentum: 0.000000
107
+ 2023-10-15 20:34:12,053 epoch 3 - iter 1042/5212 - loss 0.11731984 - time (sec): 50.06 - samples/sec: 1460.97 - lr: 0.000026 - momentum: 0.000000
108
+ 2023-10-15 20:34:37,239 epoch 3 - iter 1563/5212 - loss 0.11673671 - time (sec): 75.24 - samples/sec: 1468.35 - lr: 0.000026 - momentum: 0.000000
109
+ 2023-10-15 20:35:03,000 epoch 3 - iter 2084/5212 - loss 0.12067893 - time (sec): 101.01 - samples/sec: 1469.90 - lr: 0.000025 - momentum: 0.000000
110
+ 2023-10-15 20:35:28,469 epoch 3 - iter 2605/5212 - loss 0.11697764 - time (sec): 126.47 - samples/sec: 1462.26 - lr: 0.000025 - momentum: 0.000000
111
+ 2023-10-15 20:35:53,355 epoch 3 - iter 3126/5212 - loss 0.11647880 - time (sec): 151.36 - samples/sec: 1458.11 - lr: 0.000025 - momentum: 0.000000
112
+ 2023-10-15 20:36:19,263 epoch 3 - iter 3647/5212 - loss 0.11797656 - time (sec): 177.27 - samples/sec: 1448.30 - lr: 0.000024 - momentum: 0.000000
113
+ 2023-10-15 20:36:44,778 epoch 3 - iter 4168/5212 - loss 0.11532415 - time (sec): 202.78 - samples/sec: 1457.64 - lr: 0.000024 - momentum: 0.000000
114
+ 2023-10-15 20:37:09,921 epoch 3 - iter 4689/5212 - loss 0.11597611 - time (sec): 227.93 - samples/sec: 1458.52 - lr: 0.000024 - momentum: 0.000000
115
+ 2023-10-15 20:37:34,894 epoch 3 - iter 5210/5212 - loss 0.11657433 - time (sec): 252.90 - samples/sec: 1452.74 - lr: 0.000023 - momentum: 0.000000
116
+ 2023-10-15 20:37:34,980 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-15 20:37:34,980 EPOCH 3 done: loss 0.1166 - lr: 0.000023
118
+ 2023-10-15 20:37:43,168 DEV : loss 0.22019526362419128 - f1-score (micro avg) 0.3173
119
+ 2023-10-15 20:37:43,195 ----------------------------------------------------------------------------------------------------
120
+ 2023-10-15 20:38:08,655 epoch 4 - iter 521/5212 - loss 0.09517878 - time (sec): 25.46 - samples/sec: 1442.09 - lr: 0.000023 - momentum: 0.000000
121
+ 2023-10-15 20:38:33,547 epoch 4 - iter 1042/5212 - loss 0.08447031 - time (sec): 50.35 - samples/sec: 1416.98 - lr: 0.000023 - momentum: 0.000000
122
+ 2023-10-15 20:38:58,367 epoch 4 - iter 1563/5212 - loss 0.08700323 - time (sec): 75.17 - samples/sec: 1429.05 - lr: 0.000022 - momentum: 0.000000
123
+ 2023-10-15 20:39:23,658 epoch 4 - iter 2084/5212 - loss 0.08555122 - time (sec): 100.46 - samples/sec: 1439.08 - lr: 0.000022 - momentum: 0.000000
124
+ 2023-10-15 20:39:48,375 epoch 4 - iter 2605/5212 - loss 0.08406529 - time (sec): 125.18 - samples/sec: 1438.20 - lr: 0.000022 - momentum: 0.000000
125
+ 2023-10-15 20:40:13,097 epoch 4 - iter 3126/5212 - loss 0.08735544 - time (sec): 149.90 - samples/sec: 1440.09 - lr: 0.000021 - momentum: 0.000000
126
+ 2023-10-15 20:40:38,425 epoch 4 - iter 3647/5212 - loss 0.08619736 - time (sec): 175.23 - samples/sec: 1451.28 - lr: 0.000021 - momentum: 0.000000
127
+ 2023-10-15 20:41:03,491 epoch 4 - iter 4168/5212 - loss 0.08622321 - time (sec): 200.30 - samples/sec: 1447.99 - lr: 0.000021 - momentum: 0.000000
128
+ 2023-10-15 20:41:28,778 epoch 4 - iter 4689/5212 - loss 0.08588628 - time (sec): 225.58 - samples/sec: 1454.70 - lr: 0.000020 - momentum: 0.000000
129
+ 2023-10-15 20:41:54,667 epoch 4 - iter 5210/5212 - loss 0.08579389 - time (sec): 251.47 - samples/sec: 1460.90 - lr: 0.000020 - momentum: 0.000000
130
+ 2023-10-15 20:41:54,753 ----------------------------------------------------------------------------------------------------
131
+ 2023-10-15 20:41:54,753 EPOCH 4 done: loss 0.0858 - lr: 0.000020
132
+ 2023-10-15 20:42:03,901 DEV : loss 0.28183305263519287 - f1-score (micro avg) 0.3478
133
+ 2023-10-15 20:42:03,930 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-15 20:42:28,675 epoch 5 - iter 521/5212 - loss 0.05241740 - time (sec): 24.74 - samples/sec: 1389.53 - lr: 0.000020 - momentum: 0.000000
135
+ 2023-10-15 20:42:53,630 epoch 5 - iter 1042/5212 - loss 0.06367030 - time (sec): 49.70 - samples/sec: 1399.06 - lr: 0.000019 - momentum: 0.000000
136
+ 2023-10-15 20:43:18,460 epoch 5 - iter 1563/5212 - loss 0.06209608 - time (sec): 74.53 - samples/sec: 1410.98 - lr: 0.000019 - momentum: 0.000000
137
+ 2023-10-15 20:43:43,337 epoch 5 - iter 2084/5212 - loss 0.06304772 - time (sec): 99.41 - samples/sec: 1431.86 - lr: 0.000019 - momentum: 0.000000
138
+ 2023-10-15 20:44:08,826 epoch 5 - iter 2605/5212 - loss 0.06257189 - time (sec): 124.89 - samples/sec: 1441.74 - lr: 0.000018 - momentum: 0.000000
139
+ 2023-10-15 20:44:34,138 epoch 5 - iter 3126/5212 - loss 0.06061267 - time (sec): 150.21 - samples/sec: 1435.46 - lr: 0.000018 - momentum: 0.000000
140
+ 2023-10-15 20:44:59,301 epoch 5 - iter 3647/5212 - loss 0.06083740 - time (sec): 175.37 - samples/sec: 1443.21 - lr: 0.000018 - momentum: 0.000000
141
+ 2023-10-15 20:45:25,043 epoch 5 - iter 4168/5212 - loss 0.05950603 - time (sec): 201.11 - samples/sec: 1448.76 - lr: 0.000017 - momentum: 0.000000
142
+ 2023-10-15 20:45:50,728 epoch 5 - iter 4689/5212 - loss 0.05942891 - time (sec): 226.80 - samples/sec: 1451.75 - lr: 0.000017 - momentum: 0.000000
143
+ 2023-10-15 20:46:16,246 epoch 5 - iter 5210/5212 - loss 0.05999733 - time (sec): 252.31 - samples/sec: 1455.71 - lr: 0.000017 - momentum: 0.000000
144
+ 2023-10-15 20:46:16,342 ----------------------------------------------------------------------------------------------------
145
+ 2023-10-15 20:46:16,342 EPOCH 5 done: loss 0.0600 - lr: 0.000017
146
+ 2023-10-15 20:46:25,496 DEV : loss 0.28720250725746155 - f1-score (micro avg) 0.3655
147
+ 2023-10-15 20:46:25,523 ----------------------------------------------------------------------------------------------------
148
+ 2023-10-15 20:46:50,940 epoch 6 - iter 521/5212 - loss 0.04731132 - time (sec): 25.42 - samples/sec: 1475.22 - lr: 0.000016 - momentum: 0.000000
149
+ 2023-10-15 20:47:16,372 epoch 6 - iter 1042/5212 - loss 0.04736510 - time (sec): 50.85 - samples/sec: 1494.69 - lr: 0.000016 - momentum: 0.000000
150
+ 2023-10-15 20:47:41,147 epoch 6 - iter 1563/5212 - loss 0.04572425 - time (sec): 75.62 - samples/sec: 1473.37 - lr: 0.000016 - momentum: 0.000000
151
+ 2023-10-15 20:48:06,986 epoch 6 - iter 2084/5212 - loss 0.04405961 - time (sec): 101.46 - samples/sec: 1485.66 - lr: 0.000015 - momentum: 0.000000
152
+ 2023-10-15 20:48:32,376 epoch 6 - iter 2605/5212 - loss 0.04285739 - time (sec): 126.85 - samples/sec: 1474.66 - lr: 0.000015 - momentum: 0.000000
153
+ 2023-10-15 20:48:57,879 epoch 6 - iter 3126/5212 - loss 0.04390113 - time (sec): 152.35 - samples/sec: 1465.47 - lr: 0.000015 - momentum: 0.000000
154
+ 2023-10-15 20:49:22,624 epoch 6 - iter 3647/5212 - loss 0.04383399 - time (sec): 177.10 - samples/sec: 1456.57 - lr: 0.000014 - momentum: 0.000000
155
+ 2023-10-15 20:49:47,726 epoch 6 - iter 4168/5212 - loss 0.04360406 - time (sec): 202.20 - samples/sec: 1451.05 - lr: 0.000014 - momentum: 0.000000
156
+ 2023-10-15 20:50:12,908 epoch 6 - iter 4689/5212 - loss 0.04400334 - time (sec): 227.38 - samples/sec: 1444.97 - lr: 0.000014 - momentum: 0.000000
157
+ 2023-10-15 20:50:38,364 epoch 6 - iter 5210/5212 - loss 0.04464236 - time (sec): 252.84 - samples/sec: 1451.74 - lr: 0.000013 - momentum: 0.000000
158
+ 2023-10-15 20:50:38,495 ----------------------------------------------------------------------------------------------------
159
+ 2023-10-15 20:50:38,495 EPOCH 6 done: loss 0.0446 - lr: 0.000013
160
+ 2023-10-15 20:50:47,694 DEV : loss 0.3866041302680969 - f1-score (micro avg) 0.3652
161
+ 2023-10-15 20:50:47,722 ----------------------------------------------------------------------------------------------------
162
+ 2023-10-15 20:51:13,075 epoch 7 - iter 521/5212 - loss 0.03788773 - time (sec): 25.35 - samples/sec: 1488.17 - lr: 0.000013 - momentum: 0.000000
163
+ 2023-10-15 20:51:38,952 epoch 7 - iter 1042/5212 - loss 0.03141041 - time (sec): 51.23 - samples/sec: 1481.15 - lr: 0.000013 - momentum: 0.000000
164
+ 2023-10-15 20:52:04,004 epoch 7 - iter 1563/5212 - loss 0.03509067 - time (sec): 76.28 - samples/sec: 1464.43 - lr: 0.000012 - momentum: 0.000000
165
+ 2023-10-15 20:52:29,033 epoch 7 - iter 2084/5212 - loss 0.03393019 - time (sec): 101.31 - samples/sec: 1436.78 - lr: 0.000012 - momentum: 0.000000
166
+ 2023-10-15 20:52:54,318 epoch 7 - iter 2605/5212 - loss 0.03383844 - time (sec): 126.60 - samples/sec: 1448.48 - lr: 0.000012 - momentum: 0.000000
167
+ 2023-10-15 20:53:19,320 epoch 7 - iter 3126/5212 - loss 0.03348699 - time (sec): 151.60 - samples/sec: 1441.89 - lr: 0.000011 - momentum: 0.000000
168
+ 2023-10-15 20:53:45,054 epoch 7 - iter 3647/5212 - loss 0.03338128 - time (sec): 177.33 - samples/sec: 1455.10 - lr: 0.000011 - momentum: 0.000000
169
+ 2023-10-15 20:54:09,864 epoch 7 - iter 4168/5212 - loss 0.03278061 - time (sec): 202.14 - samples/sec: 1446.19 - lr: 0.000011 - momentum: 0.000000
170
+ 2023-10-15 20:54:35,559 epoch 7 - iter 4689/5212 - loss 0.03275384 - time (sec): 227.84 - samples/sec: 1453.23 - lr: 0.000010 - momentum: 0.000000
171
+ 2023-10-15 20:55:00,583 epoch 7 - iter 5210/5212 - loss 0.03182261 - time (sec): 252.86 - samples/sec: 1452.89 - lr: 0.000010 - momentum: 0.000000
172
+ 2023-10-15 20:55:00,671 ----------------------------------------------------------------------------------------------------
173
+ 2023-10-15 20:55:00,671 EPOCH 7 done: loss 0.0318 - lr: 0.000010
174
+ 2023-10-15 20:55:09,772 DEV : loss 0.41479361057281494 - f1-score (micro avg) 0.3719
175
+ 2023-10-15 20:55:09,801 saving best model
176
+ 2023-10-15 20:55:10,286 ----------------------------------------------------------------------------------------------------
177
+ 2023-10-15 20:55:34,976 epoch 8 - iter 521/5212 - loss 0.02874987 - time (sec): 24.69 - samples/sec: 1382.35 - lr: 0.000010 - momentum: 0.000000
178
+ 2023-10-15 20:56:00,292 epoch 8 - iter 1042/5212 - loss 0.02974470 - time (sec): 50.00 - samples/sec: 1456.42 - lr: 0.000009 - momentum: 0.000000
179
+ 2023-10-15 20:56:25,436 epoch 8 - iter 1563/5212 - loss 0.02516966 - time (sec): 75.15 - samples/sec: 1449.58 - lr: 0.000009 - momentum: 0.000000
180
+ 2023-10-15 20:56:50,608 epoch 8 - iter 2084/5212 - loss 0.02526379 - time (sec): 100.32 - samples/sec: 1446.32 - lr: 0.000009 - momentum: 0.000000
181
+ 2023-10-15 20:57:16,033 epoch 8 - iter 2605/5212 - loss 0.02435580 - time (sec): 125.74 - samples/sec: 1454.15 - lr: 0.000008 - momentum: 0.000000
182
+ 2023-10-15 20:57:41,452 epoch 8 - iter 3126/5212 - loss 0.02388305 - time (sec): 151.16 - samples/sec: 1460.49 - lr: 0.000008 - momentum: 0.000000
183
+ 2023-10-15 20:58:06,478 epoch 8 - iter 3647/5212 - loss 0.02325843 - time (sec): 176.19 - samples/sec: 1462.53 - lr: 0.000008 - momentum: 0.000000
184
+ 2023-10-15 20:58:31,384 epoch 8 - iter 4168/5212 - loss 0.02355797 - time (sec): 201.10 - samples/sec: 1462.23 - lr: 0.000007 - momentum: 0.000000
185
+ 2023-10-15 20:58:56,660 epoch 8 - iter 4689/5212 - loss 0.02339324 - time (sec): 226.37 - samples/sec: 1462.32 - lr: 0.000007 - momentum: 0.000000
186
+ 2023-10-15 20:59:21,654 epoch 8 - iter 5210/5212 - loss 0.02336518 - time (sec): 251.37 - samples/sec: 1460.88 - lr: 0.000007 - momentum: 0.000000
187
+ 2023-10-15 20:59:21,754 ----------------------------------------------------------------------------------------------------
188
+ 2023-10-15 20:59:21,754 EPOCH 8 done: loss 0.0234 - lr: 0.000007
189
+ 2023-10-15 20:59:30,902 DEV : loss 0.42180541157722473 - f1-score (micro avg) 0.3741
190
+ 2023-10-15 20:59:30,930 saving best model
191
+ 2023-10-15 20:59:31,418 ----------------------------------------------------------------------------------------------------
192
+ 2023-10-15 20:59:57,079 epoch 9 - iter 521/5212 - loss 0.01219894 - time (sec): 25.66 - samples/sec: 1547.72 - lr: 0.000006 - momentum: 0.000000
193
+ 2023-10-15 21:00:22,613 epoch 9 - iter 1042/5212 - loss 0.01351483 - time (sec): 51.19 - samples/sec: 1525.14 - lr: 0.000006 - momentum: 0.000000
194
+ 2023-10-15 21:00:47,949 epoch 9 - iter 1563/5212 - loss 0.01411036 - time (sec): 76.53 - samples/sec: 1511.90 - lr: 0.000006 - momentum: 0.000000
195
+ 2023-10-15 21:01:12,968 epoch 9 - iter 2084/5212 - loss 0.01442847 - time (sec): 101.55 - samples/sec: 1496.00 - lr: 0.000005 - momentum: 0.000000
196
+ 2023-10-15 21:01:38,021 epoch 9 - iter 2605/5212 - loss 0.01486565 - time (sec): 126.60 - samples/sec: 1486.57 - lr: 0.000005 - momentum: 0.000000
197
+ 2023-10-15 21:02:02,741 epoch 9 - iter 3126/5212 - loss 0.01496493 - time (sec): 151.32 - samples/sec: 1460.15 - lr: 0.000005 - momentum: 0.000000
198
+ 2023-10-15 21:02:28,122 epoch 9 - iter 3647/5212 - loss 0.01502140 - time (sec): 176.70 - samples/sec: 1462.06 - lr: 0.000004 - momentum: 0.000000
199
+ 2023-10-15 21:02:52,991 epoch 9 - iter 4168/5212 - loss 0.01484723 - time (sec): 201.57 - samples/sec: 1461.58 - lr: 0.000004 - momentum: 0.000000
200
+ 2023-10-15 21:03:18,183 epoch 9 - iter 4689/5212 - loss 0.01495762 - time (sec): 226.76 - samples/sec: 1460.46 - lr: 0.000004 - momentum: 0.000000
201
+ 2023-10-15 21:03:43,356 epoch 9 - iter 5210/5212 - loss 0.01475329 - time (sec): 251.93 - samples/sec: 1458.16 - lr: 0.000003 - momentum: 0.000000
202
+ 2023-10-15 21:03:43,445 ----------------------------------------------------------------------------------------------------
203
+ 2023-10-15 21:03:43,445 EPOCH 9 done: loss 0.0148 - lr: 0.000003
204
+ 2023-10-15 21:03:52,572 DEV : loss 0.4861517548561096 - f1-score (micro avg) 0.3626
205
+ 2023-10-15 21:03:52,599 ----------------------------------------------------------------------------------------------------
206
+ 2023-10-15 21:04:17,540 epoch 10 - iter 521/5212 - loss 0.00895462 - time (sec): 24.94 - samples/sec: 1423.74 - lr: 0.000003 - momentum: 0.000000
207
+ 2023-10-15 21:04:42,571 epoch 10 - iter 1042/5212 - loss 0.01126073 - time (sec): 49.97 - samples/sec: 1388.03 - lr: 0.000003 - momentum: 0.000000
208
+ 2023-10-15 21:05:07,649 epoch 10 - iter 1563/5212 - loss 0.01142583 - time (sec): 75.05 - samples/sec: 1420.18 - lr: 0.000002 - momentum: 0.000000
209
+ 2023-10-15 21:05:32,542 epoch 10 - iter 2084/5212 - loss 0.01092774 - time (sec): 99.94 - samples/sec: 1426.42 - lr: 0.000002 - momentum: 0.000000
210
+ 2023-10-15 21:05:57,462 epoch 10 - iter 2605/5212 - loss 0.01134639 - time (sec): 124.86 - samples/sec: 1432.03 - lr: 0.000002 - momentum: 0.000000
211
+ 2023-10-15 21:06:22,580 epoch 10 - iter 3126/5212 - loss 0.01131287 - time (sec): 149.98 - samples/sec: 1436.28 - lr: 0.000001 - momentum: 0.000000
212
+ 2023-10-15 21:06:47,748 epoch 10 - iter 3647/5212 - loss 0.01122214 - time (sec): 175.15 - samples/sec: 1442.52 - lr: 0.000001 - momentum: 0.000000
213
+ 2023-10-15 21:07:13,252 epoch 10 - iter 4168/5212 - loss 0.01080290 - time (sec): 200.65 - samples/sec: 1452.51 - lr: 0.000001 - momentum: 0.000000
214
+ 2023-10-15 21:07:38,300 epoch 10 - iter 4689/5212 - loss 0.01104492 - time (sec): 225.70 - samples/sec: 1450.67 - lr: 0.000000 - momentum: 0.000000
215
+ 2023-10-15 21:08:03,995 epoch 10 - iter 5210/5212 - loss 0.01096840 - time (sec): 251.39 - samples/sec: 1461.12 - lr: 0.000000 - momentum: 0.000000
216
+ 2023-10-15 21:08:04,082 ----------------------------------------------------------------------------------------------------
217
+ 2023-10-15 21:08:04,082 EPOCH 10 done: loss 0.0110 - lr: 0.000000
218
+ 2023-10-15 21:08:12,303 DEV : loss 0.4902326166629791 - f1-score (micro avg) 0.3638
219
+ 2023-10-15 21:08:12,728 ----------------------------------------------------------------------------------------------------
220
+ 2023-10-15 21:08:12,729 Loading model from best epoch ...
221
+ 2023-10-15 21:08:14,255 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
222
+ 2023-10-15 21:08:30,524
223
+ Results:
224
+ - F-score (micro) 0.4907
225
+ - F-score (macro) 0.3334
226
+ - Accuracy 0.3301
227
+
228
+ By class:
229
+ precision recall f1-score support
230
+
231
+ LOC 0.5226 0.6203 0.5672 1214
232
+ PER 0.4030 0.5371 0.4605 808
233
+ ORG 0.3028 0.3088 0.3058 353
234
+ HumanProd 0.0000 0.0000 0.0000 15
235
+
236
+ micro avg 0.4481 0.5423 0.4907 2390
237
+ macro avg 0.3071 0.3665 0.3334 2390
238
+ weighted avg 0.4464 0.5423 0.4890 2390
239
+
240
+ 2023-10-15 21:08:30,524 ----------------------------------------------------------------------------------------------------