Hugging Face's logo

stefan-it
/
hmbench-newseye-de-hmbert-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-3

Token Classification
Flair
PyTorch
German
sequence-tagger-model
Model card Files Community
2
stefan-it commited on
Commit
794b77b
•
1 Parent(s): 54f4065

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 +239 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c9f19a565661edc6b6e0aa3dd9a68d818c0d2743a272006cc9a4009f39e2651a
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 17:50:42 0.0000 0.3887 0.2012 0.1862 0.6174 0.2861 0.1682
3
+ 2 17:55:04 0.0000 0.1735 0.1790 0.2738 0.5511 0.3658 0.2244
4
+ 3 17:59:25 0.0000 0.1185 0.2483 0.2251 0.5568 0.3206 0.1917
5
+ 4 18:03:48 0.0000 0.0825 0.3585 0.2498 0.6307 0.3579 0.2186
6
+ 5 18:08:10 0.0000 0.0592 0.2955 0.2966 0.5758 0.3915 0.2442
7
+ 6 18:12:34 0.0000 0.0441 0.3935 0.2719 0.5928 0.3728 0.2307
8
+ 7 18:16:58 0.0000 0.0324 0.4698 0.2398 0.6117 0.3445 0.2092
9
+ 8 18:21:21 0.0000 0.0227 0.4874 0.2504 0.6496 0.3614 0.2217
10
+ 9 18:25:43 0.0000 0.0152 0.5062 0.2573 0.6307 0.3655 0.2247
11
+ 10 18:30:05 0.0000 0.0102 0.5002 0.2648 0.6364 0.3740 0.2311
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,239 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-15 17:46:21,046 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-15 17:46:21,047 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 17:46:21,047 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-15 17:46:21,047 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 17:46:21,047 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-15 17:46:21,047 Train: 20847 sentences
55
+ 2023-10-15 17:46:21,047 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-15 17:46:21,048 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-15 17:46:21,048 Training Params:
58
+ 2023-10-15 17:46:21,048 - learning_rate: "3e-05"
59
+ 2023-10-15 17:46:21,048 - mini_batch_size: "4"
60
+ 2023-10-15 17:46:21,048 - max_epochs: "10"
61
+ 2023-10-15 17:46:21,048 - shuffle: "True"
62
+ 2023-10-15 17:46:21,048 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-15 17:46:21,048 Plugins:
64
+ 2023-10-15 17:46:21,048 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-15 17:46:21,048 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-15 17:46:21,048 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-15 17:46:21,048 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-15 17:46:21,048 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-15 17:46:21,048 Computation:
70
+ 2023-10-15 17:46:21,048 - compute on device: cuda:0
71
+ 2023-10-15 17:46:21,048 - embedding storage: none
72
+ 2023-10-15 17:46:21,048 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-15 17:46:21,048 Model training base path: "hmbench-newseye/de-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-3"
74
+ 2023-10-15 17:46:21,048 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-15 17:46:21,048 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-15 17:46:46,375 epoch 1 - iter 521/5212 - loss 1.50097394 - time (sec): 25.33 - samples/sec: 1383.16 - lr: 0.000003 - momentum: 0.000000
77
+ 2023-10-15 17:47:11,339 epoch 1 - iter 1042/5212 - loss 1.01650844 - time (sec): 50.29 - samples/sec: 1411.09 - lr: 0.000006 - momentum: 0.000000
78
+ 2023-10-15 17:47:37,066 epoch 1 - iter 1563/5212 - loss 0.77214666 - time (sec): 76.02 - samples/sec: 1423.69 - lr: 0.000009 - momentum: 0.000000
79
+ 2023-10-15 17:48:02,847 epoch 1 - iter 2084/5212 - loss 0.64040683 - time (sec): 101.80 - samples/sec: 1431.64 - lr: 0.000012 - momentum: 0.000000
80
+ 2023-10-15 17:48:28,551 epoch 1 - iter 2605/5212 - loss 0.56192130 - time (sec): 127.50 - samples/sec: 1434.16 - lr: 0.000015 - momentum: 0.000000
81
+ 2023-10-15 17:48:54,484 epoch 1 - iter 3126/5212 - loss 0.49948839 - time (sec): 153.44 - samples/sec: 1449.35 - lr: 0.000018 - momentum: 0.000000
82
+ 2023-10-15 17:49:20,094 epoch 1 - iter 3647/5212 - loss 0.46113087 - time (sec): 179.04 - samples/sec: 1446.19 - lr: 0.000021 - momentum: 0.000000
83
+ 2023-10-15 17:49:45,443 epoch 1 - iter 4168/5212 - loss 0.43061045 - time (sec): 204.39 - samples/sec: 1446.65 - lr: 0.000024 - momentum: 0.000000
84
+ 2023-10-15 17:50:10,290 epoch 1 - iter 4689/5212 - loss 0.41123436 - time (sec): 229.24 - samples/sec: 1435.64 - lr: 0.000027 - momentum: 0.000000
85
+ 2023-10-15 17:50:36,050 epoch 1 - iter 5210/5212 - loss 0.38867196 - time (sec): 255.00 - samples/sec: 1440.57 - lr: 0.000030 - momentum: 0.000000
86
+ 2023-10-15 17:50:36,136 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-15 17:50:36,137 EPOCH 1 done: loss 0.3887 - lr: 0.000030
88
+ 2023-10-15 17:50:42,081 DEV : loss 0.20117585361003876 - f1-score (micro avg) 0.2861
89
+ 2023-10-15 17:50:42,110 saving best model
90
+ 2023-10-15 17:50:42,473 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-15 17:51:07,318 epoch 2 - iter 521/5212 - loss 0.22240788 - time (sec): 24.84 - samples/sec: 1352.80 - lr: 0.000030 - momentum: 0.000000
92
+ 2023-10-15 17:51:32,709 epoch 2 - iter 1042/5212 - loss 0.20503019 - time (sec): 50.24 - samples/sec: 1409.19 - lr: 0.000029 - momentum: 0.000000
93
+ 2023-10-15 17:51:57,674 epoch 2 - iter 1563/5212 - loss 0.19541278 - time (sec): 75.20 - samples/sec: 1411.83 - lr: 0.000029 - momentum: 0.000000
94
+ 2023-10-15 17:52:23,171 epoch 2 - iter 2084/5212 - loss 0.18582207 - time (sec): 100.70 - samples/sec: 1423.80 - lr: 0.000029 - momentum: 0.000000
95
+ 2023-10-15 17:52:48,033 epoch 2 - iter 2605/5212 - loss 0.18632288 - time (sec): 125.56 - samples/sec: 1427.21 - lr: 0.000028 - momentum: 0.000000
96
+ 2023-10-15 17:53:13,971 epoch 2 - iter 3126/5212 - loss 0.18323477 - time (sec): 151.50 - samples/sec: 1443.16 - lr: 0.000028 - momentum: 0.000000
97
+ 2023-10-15 17:53:39,204 epoch 2 - iter 3647/5212 - loss 0.18006233 - time (sec): 176.73 - samples/sec: 1439.74 - lr: 0.000028 - momentum: 0.000000
98
+ 2023-10-15 17:54:04,624 epoch 2 - iter 4168/5212 - loss 0.17990351 - time (sec): 202.15 - samples/sec: 1439.47 - lr: 0.000027 - momentum: 0.000000
99
+ 2023-10-15 17:54:30,039 epoch 2 - iter 4689/5212 - loss 0.17515115 - time (sec): 227.56 - samples/sec: 1442.75 - lr: 0.000027 - momentum: 0.000000
100
+ 2023-10-15 17:54:55,742 epoch 2 - iter 5210/5212 - loss 0.17338933 - time (sec): 253.27 - samples/sec: 1448.53 - lr: 0.000027 - momentum: 0.000000
101
+ 2023-10-15 17:54:55,905 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-15 17:54:55,905 EPOCH 2 done: loss 0.1735 - lr: 0.000027
103
+ 2023-10-15 17:55:04,829 DEV : loss 0.17898787558078766 - f1-score (micro avg) 0.3658
104
+ 2023-10-15 17:55:04,856 saving best model
105
+ 2023-10-15 17:55:05,299 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-15 17:55:30,120 epoch 3 - iter 521/5212 - loss 0.12227339 - time (sec): 24.82 - samples/sec: 1351.59 - lr: 0.000026 - momentum: 0.000000
107
+ 2023-10-15 17:55:55,409 epoch 3 - iter 1042/5212 - loss 0.12395854 - time (sec): 50.11 - samples/sec: 1373.02 - lr: 0.000026 - momentum: 0.000000
108
+ 2023-10-15 17:56:21,481 epoch 3 - iter 1563/5212 - loss 0.11709392 - time (sec): 76.18 - samples/sec: 1451.09 - lr: 0.000026 - momentum: 0.000000
109
+ 2023-10-15 17:56:47,235 epoch 3 - iter 2084/5212 - loss 0.11296261 - time (sec): 101.93 - samples/sec: 1450.16 - lr: 0.000025 - momentum: 0.000000
110
+ 2023-10-15 17:57:12,531 epoch 3 - iter 2605/5212 - loss 0.11412983 - time (sec): 127.23 - samples/sec: 1452.46 - lr: 0.000025 - momentum: 0.000000
111
+ 2023-10-15 17:57:36,931 epoch 3 - iter 3126/5212 - loss 0.11505627 - time (sec): 151.63 - samples/sec: 1462.07 - lr: 0.000025 - momentum: 0.000000
112
+ 2023-10-15 17:58:01,267 epoch 3 - iter 3647/5212 - loss 0.11553900 - time (sec): 175.97 - samples/sec: 1465.80 - lr: 0.000024 - momentum: 0.000000
113
+ 2023-10-15 17:58:26,228 epoch 3 - iter 4168/5212 - loss 0.11759207 - time (sec): 200.93 - samples/sec: 1457.48 - lr: 0.000024 - momentum: 0.000000
114
+ 2023-10-15 17:58:51,153 epoch 3 - iter 4689/5212 - loss 0.11866732 - time (sec): 225.85 - samples/sec: 1457.93 - lr: 0.000024 - momentum: 0.000000
115
+ 2023-10-15 17:59:16,200 epoch 3 - iter 5210/5212 - loss 0.11854949 - time (sec): 250.90 - samples/sec: 1464.12 - lr: 0.000023 - momentum: 0.000000
116
+ 2023-10-15 17:59:16,294 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-15 17:59:16,294 EPOCH 3 done: loss 0.1185 - lr: 0.000023
118
+ 2023-10-15 17:59:25,325 DEV : loss 0.24826383590698242 - f1-score (micro avg) 0.3206
119
+ 2023-10-15 17:59:25,351 ----------------------------------------------------------------------------------------------------
120
+ 2023-10-15 17:59:50,986 epoch 4 - iter 521/5212 - loss 0.07086836 - time (sec): 25.63 - samples/sec: 1446.52 - lr: 0.000023 - momentum: 0.000000
121
+ 2023-10-15 18:00:16,603 epoch 4 - iter 1042/5212 - loss 0.07461997 - time (sec): 51.25 - samples/sec: 1467.28 - lr: 0.000023 - momentum: 0.000000
122
+ 2023-10-15 18:00:42,316 epoch 4 - iter 1563/5212 - loss 0.07589647 - time (sec): 76.96 - samples/sec: 1444.37 - lr: 0.000022 - momentum: 0.000000
123
+ 2023-10-15 18:01:08,183 epoch 4 - iter 2084/5212 - loss 0.07950376 - time (sec): 102.83 - samples/sec: 1433.87 - lr: 0.000022 - momentum: 0.000000
124
+ 2023-10-15 18:01:33,328 epoch 4 - iter 2605/5212 - loss 0.07820170 - time (sec): 127.98 - samples/sec: 1437.64 - lr: 0.000022 - momentum: 0.000000
125
+ 2023-10-15 18:01:59,013 epoch 4 - iter 3126/5212 - loss 0.08102218 - time (sec): 153.66 - samples/sec: 1442.95 - lr: 0.000021 - momentum: 0.000000
126
+ 2023-10-15 18:02:24,785 epoch 4 - iter 3647/5212 - loss 0.08157332 - time (sec): 179.43 - samples/sec: 1446.50 - lr: 0.000021 - momentum: 0.000000
127
+ 2023-10-15 18:02:50,143 epoch 4 - iter 4168/5212 - loss 0.08239732 - time (sec): 204.79 - samples/sec: 1440.24 - lr: 0.000021 - momentum: 0.000000
128
+ 2023-10-15 18:03:15,106 epoch 4 - iter 4689/5212 - loss 0.08200821 - time (sec): 229.75 - samples/sec: 1437.08 - lr: 0.000020 - momentum: 0.000000
129
+ 2023-10-15 18:03:40,271 epoch 4 - iter 5210/5212 - loss 0.08251544 - time (sec): 254.92 - samples/sec: 1441.21 - lr: 0.000020 - momentum: 0.000000
130
+ 2023-10-15 18:03:40,369 ----------------------------------------------------------------------------------------------------
131
+ 2023-10-15 18:03:40,369 EPOCH 4 done: loss 0.0825 - lr: 0.000020
132
+ 2023-10-15 18:03:48,838 DEV : loss 0.35853323340415955 - f1-score (micro avg) 0.3579
133
+ 2023-10-15 18:03:48,867 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-15 18:04:15,944 epoch 5 - iter 521/5212 - loss 0.06074459 - time (sec): 27.07 - samples/sec: 1474.06 - lr: 0.000020 - momentum: 0.000000
135
+ 2023-10-15 18:04:41,310 epoch 5 - iter 1042/5212 - loss 0.05750324 - time (sec): 52.44 - samples/sec: 1450.08 - lr: 0.000019 - momentum: 0.000000
136
+ 2023-10-15 18:05:06,545 epoch 5 - iter 1563/5212 - loss 0.06019844 - time (sec): 77.68 - samples/sec: 1440.58 - lr: 0.000019 - momentum: 0.000000
137
+ 2023-10-15 18:05:31,865 epoch 5 - iter 2084/5212 - loss 0.05797686 - time (sec): 103.00 - samples/sec: 1457.52 - lr: 0.000019 - momentum: 0.000000
138
+ 2023-10-15 18:05:56,701 epoch 5 - iter 2605/5212 - loss 0.05779824 - time (sec): 127.83 - samples/sec: 1463.83 - lr: 0.000018 - momentum: 0.000000
139
+ 2023-10-15 18:06:21,368 epoch 5 - iter 3126/5212 - loss 0.05796520 - time (sec): 152.50 - samples/sec: 1458.09 - lr: 0.000018 - momentum: 0.000000
140
+ 2023-10-15 18:06:46,164 epoch 5 - iter 3647/5212 - loss 0.05749706 - time (sec): 177.30 - samples/sec: 1447.84 - lr: 0.000018 - momentum: 0.000000
141
+ 2023-10-15 18:07:11,414 epoch 5 - iter 4168/5212 - loss 0.05748009 - time (sec): 202.55 - samples/sec: 1451.08 - lr: 0.000017 - momentum: 0.000000
142
+ 2023-10-15 18:07:36,488 epoch 5 - iter 4689/5212 - loss 0.05835980 - time (sec): 227.62 - samples/sec: 1453.58 - lr: 0.000017 - momentum: 0.000000
143
+ 2023-10-15 18:08:02,017 epoch 5 - iter 5210/5212 - loss 0.05917565 - time (sec): 253.15 - samples/sec: 1451.21 - lr: 0.000017 - momentum: 0.000000
144
+ 2023-10-15 18:08:02,105 ----------------------------------------------------------------------------------------------------
145
+ 2023-10-15 18:08:02,106 EPOCH 5 done: loss 0.0592 - lr: 0.000017
146
+ 2023-10-15 18:08:10,370 DEV : loss 0.2954880893230438 - f1-score (micro avg) 0.3915
147
+ 2023-10-15 18:08:10,397 saving best model
148
+ 2023-10-15 18:08:10,852 ----------------------------------------------------------------------------------------------------
149
+ 2023-10-15 18:08:36,370 epoch 6 - iter 521/5212 - loss 0.03566624 - time (sec): 25.51 - samples/sec: 1407.17 - lr: 0.000016 - momentum: 0.000000
150
+ 2023-10-15 18:09:01,776 epoch 6 - iter 1042/5212 - loss 0.03735819 - time (sec): 50.92 - samples/sec: 1459.40 - lr: 0.000016 - momentum: 0.000000
151
+ 2023-10-15 18:09:27,672 epoch 6 - iter 1563/5212 - loss 0.04010312 - time (sec): 76.82 - samples/sec: 1461.32 - lr: 0.000016 - momentum: 0.000000
152
+ 2023-10-15 18:09:53,920 epoch 6 - iter 2084/5212 - loss 0.04272252 - time (sec): 103.06 - samples/sec: 1440.65 - lr: 0.000015 - momentum: 0.000000
153
+ 2023-10-15 18:10:18,891 epoch 6 - iter 2605/5212 - loss 0.04658724 - time (sec): 128.04 - samples/sec: 1423.31 - lr: 0.000015 - momentum: 0.000000
154
+ 2023-10-15 18:10:44,769 epoch 6 - iter 3126/5212 - loss 0.04608168 - time (sec): 153.91 - samples/sec: 1436.73 - lr: 0.000015 - momentum: 0.000000
155
+ 2023-10-15 18:11:09,948 epoch 6 - iter 3647/5212 - loss 0.04595678 - time (sec): 179.09 - samples/sec: 1434.36 - lr: 0.000014 - momentum: 0.000000
156
+ 2023-10-15 18:11:35,711 epoch 6 - iter 4168/5212 - loss 0.04599344 - time (sec): 204.85 - samples/sec: 1436.63 - lr: 0.000014 - momentum: 0.000000
157
+ 2023-10-15 18:12:00,745 epoch 6 - iter 4689/5212 - loss 0.04455311 - time (sec): 229.89 - samples/sec: 1439.85 - lr: 0.000014 - momentum: 0.000000
158
+ 2023-10-15 18:12:26,059 epoch 6 - iter 5210/5212 - loss 0.04410825 - time (sec): 255.20 - samples/sec: 1438.27 - lr: 0.000013 - momentum: 0.000000
159
+ 2023-10-15 18:12:26,260 ----------------------------------------------------------------------------------------------------
160
+ 2023-10-15 18:12:26,260 EPOCH 6 done: loss 0.0441 - lr: 0.000013
161
+ 2023-10-15 18:12:34,571 DEV : loss 0.39353951811790466 - f1-score (micro avg) 0.3728
162
+ 2023-10-15 18:12:34,601 ----------------------------------------------------------------------------------------------------
163
+ 2023-10-15 18:12:59,435 epoch 7 - iter 521/5212 - loss 0.03179235 - time (sec): 24.83 - samples/sec: 1364.97 - lr: 0.000013 - momentum: 0.000000
164
+ 2023-10-15 18:13:24,802 epoch 7 - iter 1042/5212 - loss 0.02701866 - time (sec): 50.20 - samples/sec: 1413.92 - lr: 0.000013 - momentum: 0.000000
165
+ 2023-10-15 18:13:49,752 epoch 7 - iter 1563/5212 - loss 0.02758616 - time (sec): 75.15 - samples/sec: 1397.11 - lr: 0.000012 - momentum: 0.000000
166
+ 2023-10-15 18:14:15,112 epoch 7 - iter 2084/5212 - loss 0.02597876 - time (sec): 100.51 - samples/sec: 1423.38 - lr: 0.000012 - momentum: 0.000000
167
+ 2023-10-15 18:14:40,889 epoch 7 - iter 2605/5212 - loss 0.03022671 - time (sec): 126.29 - samples/sec: 1432.09 - lr: 0.000012 - momentum: 0.000000
168
+ 2023-10-15 18:15:06,162 epoch 7 - iter 3126/5212 - loss 0.03044416 - time (sec): 151.56 - samples/sec: 1431.88 - lr: 0.000011 - momentum: 0.000000
169
+ 2023-10-15 18:15:31,480 epoch 7 - iter 3647/5212 - loss 0.03128317 - time (sec): 176.88 - samples/sec: 1442.88 - lr: 0.000011 - momentum: 0.000000
170
+ 2023-10-15 18:15:57,689 epoch 7 - iter 4168/5212 - loss 0.03231927 - time (sec): 203.09 - samples/sec: 1441.69 - lr: 0.000011 - momentum: 0.000000
171
+ 2023-10-15 18:16:24,016 epoch 7 - iter 4689/5212 - loss 0.03123865 - time (sec): 229.41 - samples/sec: 1435.67 - lr: 0.000010 - momentum: 0.000000
172
+ 2023-10-15 18:16:49,654 epoch 7 - iter 5210/5212 - loss 0.03238048 - time (sec): 255.05 - samples/sec: 1440.47 - lr: 0.000010 - momentum: 0.000000
173
+ 2023-10-15 18:16:49,745 ----------------------------------------------------------------------------------------------------
174
+ 2023-10-15 18:16:49,745 EPOCH 7 done: loss 0.0324 - lr: 0.000010
175
+ 2023-10-15 18:16:58,188 DEV : loss 0.4698057472705841 - f1-score (micro avg) 0.3445
176
+ 2023-10-15 18:16:58,221 ----------------------------------------------------------------------------------------------------
177
+ 2023-10-15 18:17:24,331 epoch 8 - iter 521/5212 - loss 0.02118882 - time (sec): 26.11 - samples/sec: 1507.33 - lr: 0.000010 - momentum: 0.000000
178
+ 2023-10-15 18:17:50,035 epoch 8 - iter 1042/5212 - loss 0.02174828 - time (sec): 51.81 - samples/sec: 1501.03 - lr: 0.000009 - momentum: 0.000000
179
+ 2023-10-15 18:18:15,674 epoch 8 - iter 1563/5212 - loss 0.02240042 - time (sec): 77.45 - samples/sec: 1484.23 - lr: 0.000009 - momentum: 0.000000
180
+ 2023-10-15 18:18:40,960 epoch 8 - iter 2084/5212 - loss 0.02307800 - time (sec): 102.74 - samples/sec: 1458.93 - lr: 0.000009 - momentum: 0.000000
181
+ 2023-10-15 18:19:06,094 epoch 8 - iter 2605/5212 - loss 0.02270835 - time (sec): 127.87 - samples/sec: 1448.41 - lr: 0.000008 - momentum: 0.000000
182
+ 2023-10-15 18:19:31,710 epoch 8 - iter 3126/5212 - loss 0.02207908 - time (sec): 153.49 - samples/sec: 1443.14 - lr: 0.000008 - momentum: 0.000000
183
+ 2023-10-15 18:19:57,514 epoch 8 - iter 3647/5212 - loss 0.02166312 - time (sec): 179.29 - samples/sec: 1447.94 - lr: 0.000008 - momentum: 0.000000
184
+ 2023-10-15 18:20:22,251 epoch 8 - iter 4168/5212 - loss 0.02233639 - time (sec): 204.03 - samples/sec: 1441.81 - lr: 0.000007 - momentum: 0.000000
185
+ 2023-10-15 18:20:47,299 epoch 8 - iter 4689/5212 - loss 0.02246766 - time (sec): 229.08 - samples/sec: 1441.54 - lr: 0.000007 - momentum: 0.000000
186
+ 2023-10-15 18:21:11,946 epoch 8 - iter 5210/5212 - loss 0.02273107 - time (sec): 253.72 - samples/sec: 1446.48 - lr: 0.000007 - momentum: 0.000000
187
+ 2023-10-15 18:21:12,056 ----------------------------------------------------------------------------------------------------
188
+ 2023-10-15 18:21:12,056 EPOCH 8 done: loss 0.0227 - lr: 0.000007
189
+ 2023-10-15 18:21:21,123 DEV : loss 0.48737385869026184 - f1-score (micro avg) 0.3614
190
+ 2023-10-15 18:21:21,155 ----------------------------------------------------------------------------------------------------
191
+ 2023-10-15 18:21:45,925 epoch 9 - iter 521/5212 - loss 0.01657988 - time (sec): 24.77 - samples/sec: 1436.79 - lr: 0.000006 - momentum: 0.000000
192
+ 2023-10-15 18:22:11,071 epoch 9 - iter 1042/5212 - loss 0.01825306 - time (sec): 49.92 - samples/sec: 1435.80 - lr: 0.000006 - momentum: 0.000000
193
+ 2023-10-15 18:22:35,983 epoch 9 - iter 1563/5212 - loss 0.01537751 - time (sec): 74.83 - samples/sec: 1446.59 - lr: 0.000006 - momentum: 0.000000
194
+ 2023-10-15 18:23:01,098 epoch 9 - iter 2084/5212 - loss 0.01556559 - time (sec): 99.94 - samples/sec: 1448.93 - lr: 0.000005 - momentum: 0.000000
195
+ 2023-10-15 18:23:26,191 epoch 9 - iter 2605/5212 - loss 0.01597811 - time (sec): 125.04 - samples/sec: 1441.42 - lr: 0.000005 - momentum: 0.000000
196
+ 2023-10-15 18:23:52,046 epoch 9 - iter 3126/5212 - loss 0.01561272 - time (sec): 150.89 - samples/sec: 1451.99 - lr: 0.000005 - momentum: 0.000000
197
+ 2023-10-15 18:24:16,900 epoch 9 - iter 3647/5212 - loss 0.01574407 - time (sec): 175.74 - samples/sec: 1453.65 - lr: 0.000004 - momentum: 0.000000
198
+ 2023-10-15 18:24:42,461 epoch 9 - iter 4168/5212 - loss 0.01612832 - time (sec): 201.31 - samples/sec: 1448.57 - lr: 0.000004 - momentum: 0.000000
199
+ 2023-10-15 18:25:08,164 epoch 9 - iter 4689/5212 - loss 0.01585164 - time (sec): 227.01 - samples/sec: 1452.94 - lr: 0.000004 - momentum: 0.000000
200
+ 2023-10-15 18:25:33,956 epoch 9 - iter 5210/5212 - loss 0.01523723 - time (sec): 252.80 - samples/sec: 1452.97 - lr: 0.000003 - momentum: 0.000000
201
+ 2023-10-15 18:25:34,044 ----------------------------------------------------------------------------------------------------
202
+ 2023-10-15 18:25:34,044 EPOCH 9 done: loss 0.0152 - lr: 0.000003
203
+ 2023-10-15 18:25:43,065 DEV : loss 0.5062488317489624 - f1-score (micro avg) 0.3655
204
+ 2023-10-15 18:25:43,093 ----------------------------------------------------------------------------------------------------
205
+ 2023-10-15 18:26:08,163 epoch 10 - iter 521/5212 - loss 0.00931166 - time (sec): 25.07 - samples/sec: 1432.29 - lr: 0.000003 - momentum: 0.000000
206
+ 2023-10-15 18:26:33,323 epoch 10 - iter 1042/5212 - loss 0.01081890 - time (sec): 50.23 - samples/sec: 1458.64 - lr: 0.000003 - momentum: 0.000000
207
+ 2023-10-15 18:26:58,431 epoch 10 - iter 1563/5212 - loss 0.01002542 - time (sec): 75.34 - samples/sec: 1467.77 - lr: 0.000002 - momentum: 0.000000
208
+ 2023-10-15 18:27:24,060 epoch 10 - iter 2084/5212 - loss 0.00973675 - time (sec): 100.97 - samples/sec: 1456.67 - lr: 0.000002 - momentum: 0.000000
209
+ 2023-10-15 18:27:49,708 epoch 10 - iter 2605/5212 - loss 0.01027887 - time (sec): 126.61 - samples/sec: 1466.36 - lr: 0.000002 - momentum: 0.000000
210
+ 2023-10-15 18:28:15,215 epoch 10 - iter 3126/5212 - loss 0.01040973 - time (sec): 152.12 - samples/sec: 1460.95 - lr: 0.000001 - momentum: 0.000000
211
+ 2023-10-15 18:28:40,464 epoch 10 - iter 3647/5212 - loss 0.01009036 - time (sec): 177.37 - samples/sec: 1459.94 - lr: 0.000001 - momentum: 0.000000
212
+ 2023-10-15 18:29:05,142 epoch 10 - iter 4168/5212 - loss 0.01032693 - time (sec): 202.05 - samples/sec: 1454.33 - lr: 0.000001 - momentum: 0.000000
213
+ 2023-10-15 18:29:30,640 epoch 10 - iter 4689/5212 - loss 0.01050964 - time (sec): 227.55 - samples/sec: 1448.72 - lr: 0.000000 - momentum: 0.000000
214
+ 2023-10-15 18:29:55,882 epoch 10 - iter 5210/5212 - loss 0.01022552 - time (sec): 252.79 - samples/sec: 1452.20 - lr: 0.000000 - momentum: 0.000000
215
+ 2023-10-15 18:29:56,017 ----------------------------------------------------------------------------------------------------
216
+ 2023-10-15 18:29:56,017 EPOCH 10 done: loss 0.0102 - lr: 0.000000
217
+ 2023-10-15 18:30:05,188 DEV : loss 0.5002045631408691 - f1-score (micro avg) 0.374
218
+ 2023-10-15 18:30:05,584 ----------------------------------------------------------------------------------------------------
219
+ 2023-10-15 18:30:05,585 Loading model from best epoch ...
220
+ 2023-10-15 18:30:07,219 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
221
+ 2023-10-15 18:30:22,520
222
+ Results:
223
+ - F-score (micro) 0.4665
224
+ - F-score (macro) 0.3023
225
+ - Accuracy 0.3084
226
+
227
+ By class:
228
+ precision recall f1-score support
229
+
230
+ LOC 0.5410 0.5865 0.5628 1214
231
+ PER 0.4010 0.4035 0.4022 808
232
+ ORG 0.3038 0.2040 0.2441 353
233
+ HumanProd 0.0000 0.0000 0.0000 15
234
+
235
+ micro avg 0.4686 0.4644 0.4665 2390
236
+ macro avg 0.3115 0.2985 0.3023 2390
237
+ weighted avg 0.4553 0.4644 0.4579 2390
238
+
239
+ 2023-10-15 18:30:22,520 ----------------------------------------------------------------------------------------------------