Hugging Face's logo

stefan-it
/
hmbench-newseye-de-hmbert_tiny-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2

Token Classification
Flair
PyTorch
TensorBoard
German
sequence-tagger-model
Model card Files Metrics Community
stefan-it commited on
Commit
f70f6c9
•
1 Parent(s): 88f0b2b

Upload folder using huggingface_hub

Browse files
Files changed (6) hide show
  1. best-model.pt +3 -0
  2. dev.tsv +0 -0
  3. loss.tsv +11 -0
  4. runs/events.out.tfevents.1697711900.46dc0c540dd0.4319.4 +3 -0
  5. test.tsv +0 -0
  6. training.log +241 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:57fc736ac5a97c6c3ce431c66d549d3255a68216c3e541edd3679a6e7ad50dd8
3
+ size 19048098
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 10:39:46 0.0000 0.8426 0.1434 0.3810 0.0152 0.0291 0.0148
3
+ 2 10:41:14 0.0000 0.3656 0.1391 0.3213 0.2689 0.2928 0.1721
4
+ 3 10:42:42 0.0000 0.3119 0.1372 0.2820 0.3466 0.3110 0.1848
5
+ 4 10:44:11 0.0000 0.2782 0.1481 0.2199 0.3352 0.2656 0.1540
6
+ 5 10:45:39 0.0000 0.2535 0.1490 0.2427 0.3466 0.2855 0.1676
7
+ 6 10:47:07 0.0000 0.2364 0.1648 0.2011 0.4072 0.2693 0.1567
8
+ 7 10:48:35 0.0000 0.2220 0.1679 0.2145 0.3693 0.2714 0.1583
9
+ 8 10:50:04 0.0000 0.2161 0.1719 0.2079 0.3693 0.2660 0.1545
10
+ 9 10:51:32 0.0000 0.2094 0.1816 0.2071 0.3958 0.2720 0.1583
11
+ 10 10:53:01 0.0000 0.2072 0.1781 0.2076 0.3617 0.2638 0.1528
runs/events.out.tfevents.1697711900.46dc0c540dd0.4319.4 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2d6e0ab3e878c2cfd0c06ff68e3191663621e8a9c98f7fcffcdde7cd827b2e6f
3
+ size 2923780
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,241 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-19 10:38:20,282 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-19 10:38:20,282 Model: "SequenceTagger(
3
+ (embeddings): TransformerWordEmbeddings(
4
+ (model): BertModel(
5
+ (embeddings): BertEmbeddings(
6
+ (word_embeddings): Embedding(32001, 128)
7
+ (position_embeddings): Embedding(512, 128)
8
+ (token_type_embeddings): Embedding(2, 128)
9
+ (LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
10
+ (dropout): Dropout(p=0.1, inplace=False)
11
+ )
12
+ (encoder): BertEncoder(
13
+ (layer): ModuleList(
14
+ (0-1): 2 x BertLayer(
15
+ (attention): BertAttention(
16
+ (self): BertSelfAttention(
17
+ (query): Linear(in_features=128, out_features=128, bias=True)
18
+ (key): Linear(in_features=128, out_features=128, bias=True)
19
+ (value): Linear(in_features=128, out_features=128, bias=True)
20
+ (dropout): Dropout(p=0.1, inplace=False)
21
+ )
22
+ (output): BertSelfOutput(
23
+ (dense): Linear(in_features=128, out_features=128, bias=True)
24
+ (LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
25
+ (dropout): Dropout(p=0.1, inplace=False)
26
+ )
27
+ )
28
+ (intermediate): BertIntermediate(
29
+ (dense): Linear(in_features=128, out_features=512, bias=True)
30
+ (intermediate_act_fn): GELUActivation()
31
+ )
32
+ (output): BertOutput(
33
+ (dense): Linear(in_features=512, out_features=128, bias=True)
34
+ (LayerNorm): LayerNorm((128,), 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=128, out_features=128, bias=True)
42
+ (activation): Tanh()
43
+ )
44
+ )
45
+ )
46
+ (locked_dropout): LockedDropout(p=0.5)
47
+ (linear): Linear(in_features=128, out_features=17, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-19 10:38:20,282 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-19 10:38:20,282 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
+ 2023-10-19 10:38:20,282 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-19 10:38:20,282 Train: 20847 sentences
55
+ 2023-10-19 10:38:20,282 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-19 10:38:20,282 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-19 10:38:20,282 Training Params:
58
+ 2023-10-19 10:38:20,282 - learning_rate: "3e-05"
59
+ 2023-10-19 10:38:20,283 - mini_batch_size: "4"
60
+ 2023-10-19 10:38:20,283 - max_epochs: "10"
61
+ 2023-10-19 10:38:20,283 - shuffle: "True"
62
+ 2023-10-19 10:38:20,283 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-19 10:38:20,283 Plugins:
64
+ 2023-10-19 10:38:20,283 - TensorboardLogger
65
+ 2023-10-19 10:38:20,283 - LinearScheduler | warmup_fraction: '0.1'
66
+ 2023-10-19 10:38:20,283 ----------------------------------------------------------------------------------------------------
67
+ 2023-10-19 10:38:20,283 Final evaluation on model from best epoch (best-model.pt)
68
+ 2023-10-19 10:38:20,283 - metric: "('micro avg', 'f1-score')"
69
+ 2023-10-19 10:38:20,283 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-19 10:38:20,283 Computation:
71
+ 2023-10-19 10:38:20,283 - compute on device: cuda:0
72
+ 2023-10-19 10:38:20,283 - embedding storage: none
73
+ 2023-10-19 10:38:20,283 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-19 10:38:20,283 Model training base path: "hmbench-newseye/de-dbmdz/bert-tiny-historic-multilingual-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2"
75
+ 2023-10-19 10:38:20,283 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-19 10:38:20,283 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-19 10:38:20,283 Logging anything other than scalars to TensorBoard is currently not supported.
78
+ 2023-10-19 10:38:29,211 epoch 1 - iter 521/5212 - loss 2.72371222 - time (sec): 8.93 - samples/sec: 4096.81 - lr: 0.000003 - momentum: 0.000000
79
+ 2023-10-19 10:38:37,636 epoch 1 - iter 1042/5212 - loss 2.05502782 - time (sec): 17.35 - samples/sec: 4193.38 - lr: 0.000006 - momentum: 0.000000
80
+ 2023-10-19 10:38:45,775 epoch 1 - iter 1563/5212 - loss 1.58883158 - time (sec): 25.49 - samples/sec: 4304.23 - lr: 0.000009 - momentum: 0.000000
81
+ 2023-10-19 10:38:54,422 epoch 1 - iter 2084/5212 - loss 1.32512439 - time (sec): 34.14 - samples/sec: 4326.46 - lr: 0.000012 - momentum: 0.000000
82
+ 2023-10-19 10:39:02,797 epoch 1 - iter 2605/5212 - loss 1.17407642 - time (sec): 42.51 - samples/sec: 4404.91 - lr: 0.000015 - momentum: 0.000000
83
+ 2023-10-19 10:39:11,106 epoch 1 - iter 3126/5212 - loss 1.08043972 - time (sec): 50.82 - samples/sec: 4390.07 - lr: 0.000018 - momentum: 0.000000
84
+ 2023-10-19 10:39:19,406 epoch 1 - iter 3647/5212 - loss 1.01154260 - time (sec): 59.12 - samples/sec: 4394.12 - lr: 0.000021 - momentum: 0.000000
85
+ 2023-10-19 10:39:27,380 epoch 1 - iter 4168/5212 - loss 0.94868657 - time (sec): 67.10 - samples/sec: 4397.60 - lr: 0.000024 - momentum: 0.000000
86
+ 2023-10-19 10:39:35,900 epoch 1 - iter 4689/5212 - loss 0.88980347 - time (sec): 75.62 - samples/sec: 4380.26 - lr: 0.000027 - momentum: 0.000000
87
+ 2023-10-19 10:39:44,202 epoch 1 - iter 5210/5212 - loss 0.84260434 - time (sec): 83.92 - samples/sec: 4377.91 - lr: 0.000030 - momentum: 0.000000
88
+ 2023-10-19 10:39:44,234 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-19 10:39:44,234 EPOCH 1 done: loss 0.8426 - lr: 0.000030
90
+ 2023-10-19 10:39:46,475 DEV : loss 0.1433752328157425 - f1-score (micro avg) 0.0291
91
+ 2023-10-19 10:39:46,497 saving best model
92
+ 2023-10-19 10:39:46,526 ----------------------------------------------------------------------------------------------------
93
+ 2023-10-19 10:39:54,611 epoch 2 - iter 521/5212 - loss 0.40879602 - time (sec): 8.09 - samples/sec: 4230.25 - lr: 0.000030 - momentum: 0.000000
94
+ 2023-10-19 10:40:02,957 epoch 2 - iter 1042/5212 - loss 0.41373491 - time (sec): 16.43 - samples/sec: 4320.03 - lr: 0.000029 - momentum: 0.000000
95
+ 2023-10-19 10:40:11,068 epoch 2 - iter 1563/5212 - loss 0.41110618 - time (sec): 24.54 - samples/sec: 4287.77 - lr: 0.000029 - momentum: 0.000000
96
+ 2023-10-19 10:40:19,346 epoch 2 - iter 2084/5212 - loss 0.39790099 - time (sec): 32.82 - samples/sec: 4380.84 - lr: 0.000029 - momentum: 0.000000
97
+ 2023-10-19 10:40:27,777 epoch 2 - iter 2605/5212 - loss 0.39498142 - time (sec): 41.25 - samples/sec: 4415.85 - lr: 0.000028 - momentum: 0.000000
98
+ 2023-10-19 10:40:36,056 epoch 2 - iter 3126/5212 - loss 0.38607733 - time (sec): 49.53 - samples/sec: 4425.77 - lr: 0.000028 - momentum: 0.000000
99
+ 2023-10-19 10:40:44,415 epoch 2 - iter 3647/5212 - loss 0.37832413 - time (sec): 57.89 - samples/sec: 4438.26 - lr: 0.000028 - momentum: 0.000000
100
+ 2023-10-19 10:40:53,047 epoch 2 - iter 4168/5212 - loss 0.37306540 - time (sec): 66.52 - samples/sec: 4423.81 - lr: 0.000027 - momentum: 0.000000
101
+ 2023-10-19 10:41:01,276 epoch 2 - iter 4689/5212 - loss 0.36784622 - time (sec): 74.75 - samples/sec: 4421.81 - lr: 0.000027 - momentum: 0.000000
102
+ 2023-10-19 10:41:09,649 epoch 2 - iter 5210/5212 - loss 0.36564089 - time (sec): 83.12 - samples/sec: 4419.15 - lr: 0.000027 - momentum: 0.000000
103
+ 2023-10-19 10:41:09,684 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-19 10:41:09,684 EPOCH 2 done: loss 0.3656 - lr: 0.000027
105
+ 2023-10-19 10:41:14,780 DEV : loss 0.1391393542289734 - f1-score (micro avg) 0.2928
106
+ 2023-10-19 10:41:14,804 saving best model
107
+ 2023-10-19 10:41:14,840 ----------------------------------------------------------------------------------------------------
108
+ 2023-10-19 10:41:23,076 epoch 3 - iter 521/5212 - loss 0.32466550 - time (sec): 8.24 - samples/sec: 4448.41 - lr: 0.000026 - momentum: 0.000000
109
+ 2023-10-19 10:41:31,416 epoch 3 - iter 1042/5212 - loss 0.31821062 - time (sec): 16.58 - samples/sec: 4451.66 - lr: 0.000026 - momentum: 0.000000
110
+ 2023-10-19 10:41:39,462 epoch 3 - iter 1563/5212 - loss 0.32728696 - time (sec): 24.62 - samples/sec: 4451.18 - lr: 0.000026 - momentum: 0.000000
111
+ 2023-10-19 10:41:47,786 epoch 3 - iter 2084/5212 - loss 0.31607836 - time (sec): 32.95 - samples/sec: 4487.01 - lr: 0.000025 - momentum: 0.000000
112
+ 2023-10-19 10:41:56,154 epoch 3 - iter 2605/5212 - loss 0.31480478 - time (sec): 41.31 - samples/sec: 4494.25 - lr: 0.000025 - momentum: 0.000000
113
+ 2023-10-19 10:42:04,737 epoch 3 - iter 3126/5212 - loss 0.31038250 - time (sec): 49.90 - samples/sec: 4473.56 - lr: 0.000025 - momentum: 0.000000
114
+ 2023-10-19 10:42:12,965 epoch 3 - iter 3647/5212 - loss 0.31026995 - time (sec): 58.12 - samples/sec: 4453.95 - lr: 0.000024 - momentum: 0.000000
115
+ 2023-10-19 10:42:21,291 epoch 3 - iter 4168/5212 - loss 0.31015311 - time (sec): 66.45 - samples/sec: 4444.22 - lr: 0.000024 - momentum: 0.000000
116
+ 2023-10-19 10:42:29,486 epoch 3 - iter 4689/5212 - loss 0.31207638 - time (sec): 74.65 - samples/sec: 4429.21 - lr: 0.000024 - momentum: 0.000000
117
+ 2023-10-19 10:42:37,697 epoch 3 - iter 5210/5212 - loss 0.31193129 - time (sec): 82.86 - samples/sec: 4433.61 - lr: 0.000023 - momentum: 0.000000
118
+ 2023-10-19 10:42:37,733 ----------------------------------------------------------------------------------------------------
119
+ 2023-10-19 10:42:37,733 EPOCH 3 done: loss 0.3119 - lr: 0.000023
120
+ 2023-10-19 10:42:42,852 DEV : loss 0.13720223307609558 - f1-score (micro avg) 0.311
121
+ 2023-10-19 10:42:42,876 saving best model
122
+ 2023-10-19 10:42:42,916 ----------------------------------------------------------------------------------------------------
123
+ 2023-10-19 10:42:51,333 epoch 4 - iter 521/5212 - loss 0.25709768 - time (sec): 8.42 - samples/sec: 4514.44 - lr: 0.000023 - momentum: 0.000000
124
+ 2023-10-19 10:42:59,518 epoch 4 - iter 1042/5212 - loss 0.26137660 - time (sec): 16.60 - samples/sec: 4367.48 - lr: 0.000023 - momentum: 0.000000
125
+ 2023-10-19 10:43:07,636 epoch 4 - iter 1563/5212 - loss 0.27629460 - time (sec): 24.72 - samples/sec: 4311.54 - lr: 0.000022 - momentum: 0.000000
126
+ 2023-10-19 10:43:15,953 epoch 4 - iter 2084/5212 - loss 0.27504062 - time (sec): 33.04 - samples/sec: 4377.33 - lr: 0.000022 - momentum: 0.000000
127
+ 2023-10-19 10:43:24,401 epoch 4 - iter 2605/5212 - loss 0.27571513 - time (sec): 41.48 - samples/sec: 4448.97 - lr: 0.000022 - momentum: 0.000000
128
+ 2023-10-19 10:43:32,741 epoch 4 - iter 3126/5212 - loss 0.27824346 - time (sec): 49.82 - samples/sec: 4441.02 - lr: 0.000021 - momentum: 0.000000
129
+ 2023-10-19 10:43:41,037 epoch 4 - iter 3647/5212 - loss 0.27812047 - time (sec): 58.12 - samples/sec: 4436.80 - lr: 0.000021 - momentum: 0.000000
130
+ 2023-10-19 10:43:49,174 epoch 4 - iter 4168/5212 - loss 0.28260826 - time (sec): 66.26 - samples/sec: 4408.59 - lr: 0.000021 - momentum: 0.000000
131
+ 2023-10-19 10:43:57,417 epoch 4 - iter 4689/5212 - loss 0.28076208 - time (sec): 74.50 - samples/sec: 4423.38 - lr: 0.000020 - momentum: 0.000000
132
+ 2023-10-19 10:44:05,823 epoch 4 - iter 5210/5212 - loss 0.27822833 - time (sec): 82.91 - samples/sec: 4430.02 - lr: 0.000020 - momentum: 0.000000
133
+ 2023-10-19 10:44:05,856 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-19 10:44:05,856 EPOCH 4 done: loss 0.2782 - lr: 0.000020
135
+ 2023-10-19 10:44:11,009 DEV : loss 0.14805111289024353 - f1-score (micro avg) 0.2656
136
+ 2023-10-19 10:44:11,033 ----------------------------------------------------------------------------------------------------
137
+ 2023-10-19 10:44:19,265 epoch 5 - iter 521/5212 - loss 0.25458992 - time (sec): 8.23 - samples/sec: 4663.96 - lr: 0.000020 - momentum: 0.000000
138
+ 2023-10-19 10:44:27,618 epoch 5 - iter 1042/5212 - loss 0.23812930 - time (sec): 16.58 - samples/sec: 4645.30 - lr: 0.000019 - momentum: 0.000000
139
+ 2023-10-19 10:44:35,825 epoch 5 - iter 1563/5212 - loss 0.23909797 - time (sec): 24.79 - samples/sec: 4518.74 - lr: 0.000019 - momentum: 0.000000
140
+ 2023-10-19 10:44:44,069 epoch 5 - iter 2084/5212 - loss 0.24733360 - time (sec): 33.04 - samples/sec: 4498.14 - lr: 0.000019 - momentum: 0.000000
141
+ 2023-10-19 10:44:52,390 epoch 5 - iter 2605/5212 - loss 0.24607385 - time (sec): 41.36 - samples/sec: 4479.65 - lr: 0.000018 - momentum: 0.000000
142
+ 2023-10-19 10:45:00,692 epoch 5 - iter 3126/5212 - loss 0.25227478 - time (sec): 49.66 - samples/sec: 4455.06 - lr: 0.000018 - momentum: 0.000000
143
+ 2023-10-19 10:45:08,913 epoch 5 - iter 3647/5212 - loss 0.25331088 - time (sec): 57.88 - samples/sec: 4439.80 - lr: 0.000018 - momentum: 0.000000
144
+ 2023-10-19 10:45:17,372 epoch 5 - iter 4168/5212 - loss 0.25226388 - time (sec): 66.34 - samples/sec: 4444.21 - lr: 0.000017 - momentum: 0.000000
145
+ 2023-10-19 10:45:25,748 epoch 5 - iter 4689/5212 - loss 0.25437720 - time (sec): 74.71 - samples/sec: 4442.17 - lr: 0.000017 - momentum: 0.000000
146
+ 2023-10-19 10:45:34,004 epoch 5 - iter 5210/5212 - loss 0.25346983 - time (sec): 82.97 - samples/sec: 4428.05 - lr: 0.000017 - momentum: 0.000000
147
+ 2023-10-19 10:45:34,030 ----------------------------------------------------------------------------------------------------
148
+ 2023-10-19 10:45:34,030 EPOCH 5 done: loss 0.2535 - lr: 0.000017
149
+ 2023-10-19 10:45:39,162 DEV : loss 0.1490069180727005 - f1-score (micro avg) 0.2855
150
+ 2023-10-19 10:45:39,197 ----------------------------------------------------------------------------------------------------
151
+ 2023-10-19 10:45:47,655 epoch 6 - iter 521/5212 - loss 0.26624853 - time (sec): 8.46 - samples/sec: 3991.54 - lr: 0.000016 - momentum: 0.000000
152
+ 2023-10-19 10:45:56,023 epoch 6 - iter 1042/5212 - loss 0.25453009 - time (sec): 16.82 - samples/sec: 4277.48 - lr: 0.000016 - momentum: 0.000000
153
+ 2023-10-19 10:46:04,383 epoch 6 - iter 1563/5212 - loss 0.24457171 - time (sec): 25.18 - samples/sec: 4370.58 - lr: 0.000016 - momentum: 0.000000
154
+ 2023-10-19 10:46:12,858 epoch 6 - iter 2084/5212 - loss 0.23560982 - time (sec): 33.66 - samples/sec: 4419.97 - lr: 0.000015 - momentum: 0.000000
155
+ 2023-10-19 10:46:21,179 epoch 6 - iter 2605/5212 - loss 0.23329802 - time (sec): 41.98 - samples/sec: 4434.39 - lr: 0.000015 - momentum: 0.000000
156
+ 2023-10-19 10:46:29,142 epoch 6 - iter 3126/5212 - loss 0.23211029 - time (sec): 49.94 - samples/sec: 4480.71 - lr: 0.000015 - momentum: 0.000000
157
+ 2023-10-19 10:46:37,489 epoch 6 - iter 3647/5212 - loss 0.23718572 - time (sec): 58.29 - samples/sec: 4445.74 - lr: 0.000014 - momentum: 0.000000
158
+ 2023-10-19 10:46:45,775 epoch 6 - iter 4168/5212 - loss 0.23673427 - time (sec): 66.58 - samples/sec: 4422.59 - lr: 0.000014 - momentum: 0.000000
159
+ 2023-10-19 10:46:53,971 epoch 6 - iter 4689/5212 - loss 0.23213792 - time (sec): 74.77 - samples/sec: 4423.57 - lr: 0.000014 - momentum: 0.000000
160
+ 2023-10-19 10:47:02,844 epoch 6 - iter 5210/5212 - loss 0.23638178 - time (sec): 83.65 - samples/sec: 4391.69 - lr: 0.000013 - momentum: 0.000000
161
+ 2023-10-19 10:47:02,878 ----------------------------------------------------------------------------------------------------
162
+ 2023-10-19 10:47:02,879 EPOCH 6 done: loss 0.2364 - lr: 0.000013
163
+ 2023-10-19 10:47:07,421 DEV : loss 0.1647791564464569 - f1-score (micro avg) 0.2693
164
+ 2023-10-19 10:47:07,444 ----------------------------------------------------------------------------------------------------
165
+ 2023-10-19 10:47:15,620 epoch 7 - iter 521/5212 - loss 0.23817423 - time (sec): 8.18 - samples/sec: 4509.93 - lr: 0.000013 - momentum: 0.000000
166
+ 2023-10-19 10:47:23,915 epoch 7 - iter 1042/5212 - loss 0.22435065 - time (sec): 16.47 - samples/sec: 4529.78 - lr: 0.000013 - momentum: 0.000000
167
+ 2023-10-19 10:47:32,101 epoch 7 - iter 1563/5212 - loss 0.22347997 - time (sec): 24.66 - samples/sec: 4503.56 - lr: 0.000012 - momentum: 0.000000
168
+ 2023-10-19 10:47:40,302 epoch 7 - iter 2084/5212 - loss 0.22299657 - time (sec): 32.86 - samples/sec: 4508.96 - lr: 0.000012 - momentum: 0.000000
169
+ 2023-10-19 10:47:49,223 epoch 7 - iter 2605/5212 - loss 0.21747021 - time (sec): 41.78 - samples/sec: 4483.66 - lr: 0.000012 - momentum: 0.000000
170
+ 2023-10-19 10:47:57,646 epoch 7 - iter 3126/5212 - loss 0.21824678 - time (sec): 50.20 - samples/sec: 4446.54 - lr: 0.000011 - momentum: 0.000000
171
+ 2023-10-19 10:48:05,716 epoch 7 - iter 3647/5212 - loss 0.22182997 - time (sec): 58.27 - samples/sec: 4432.18 - lr: 0.000011 - momentum: 0.000000
172
+ 2023-10-19 10:48:14,327 epoch 7 - iter 4168/5212 - loss 0.22075907 - time (sec): 66.88 - samples/sec: 4401.35 - lr: 0.000011 - momentum: 0.000000
173
+ 2023-10-19 10:48:22,791 epoch 7 - iter 4689/5212 - loss 0.22067660 - time (sec): 75.35 - samples/sec: 4403.88 - lr: 0.000010 - momentum: 0.000000
174
+ 2023-10-19 10:48:31,053 epoch 7 - iter 5210/5212 - loss 0.22219262 - time (sec): 83.61 - samples/sec: 4391.66 - lr: 0.000010 - momentum: 0.000000
175
+ 2023-10-19 10:48:31,096 ----------------------------------------------------------------------------------------------------
176
+ 2023-10-19 10:48:31,096 EPOCH 7 done: loss 0.2220 - lr: 0.000010
177
+ 2023-10-19 10:48:35,618 DEV : loss 0.16794627904891968 - f1-score (micro avg) 0.2714
178
+ 2023-10-19 10:48:35,641 ----------------------------------------------------------------------------------------------------
179
+ 2023-10-19 10:48:44,018 epoch 8 - iter 521/5212 - loss 0.24024885 - time (sec): 8.38 - samples/sec: 4188.92 - lr: 0.000010 - momentum: 0.000000
180
+ 2023-10-19 10:48:52,244 epoch 8 - iter 1042/5212 - loss 0.23380355 - time (sec): 16.60 - samples/sec: 4239.48 - lr: 0.000009 - momentum: 0.000000
181
+ 2023-10-19 10:49:00,438 epoch 8 - iter 1563/5212 - loss 0.22497629 - time (sec): 24.80 - samples/sec: 4319.11 - lr: 0.000009 - momentum: 0.000000
182
+ 2023-10-19 10:49:08,803 epoch 8 - iter 2084/5212 - loss 0.23065970 - time (sec): 33.16 - samples/sec: 4388.00 - lr: 0.000009 - momentum: 0.000000
183
+ 2023-10-19 10:49:17,008 epoch 8 - iter 2605/5212 - loss 0.22269701 - time (sec): 41.37 - samples/sec: 4419.51 - lr: 0.000008 - momentum: 0.000000
184
+ 2023-10-19 10:49:25,407 epoch 8 - iter 3126/5212 - loss 0.21867285 - time (sec): 49.77 - samples/sec: 4421.06 - lr: 0.000008 - momentum: 0.000000
185
+ 2023-10-19 10:49:33,743 epoch 8 - iter 3647/5212 - loss 0.21580111 - time (sec): 58.10 - samples/sec: 4448.90 - lr: 0.000008 - momentum: 0.000000
186
+ 2023-10-19 10:49:42,116 epoch 8 - iter 4168/5212 - loss 0.21326174 - time (sec): 66.47 - samples/sec: 4465.53 - lr: 0.000007 - momentum: 0.000000
187
+ 2023-10-19 10:49:50,650 epoch 8 - iter 4689/5212 - loss 0.21523254 - time (sec): 75.01 - samples/sec: 4435.87 - lr: 0.000007 - momentum: 0.000000
188
+ 2023-10-19 10:49:58,898 epoch 8 - iter 5210/5212 - loss 0.21610124 - time (sec): 83.26 - samples/sec: 4412.87 - lr: 0.000007 - momentum: 0.000000
189
+ 2023-10-19 10:49:58,930 ----------------------------------------------------------------------------------------------------
190
+ 2023-10-19 10:49:58,930 EPOCH 8 done: loss 0.2161 - lr: 0.000007
191
+ 2023-10-19 10:50:04,152 DEV : loss 0.17187514901161194 - f1-score (micro avg) 0.266
192
+ 2023-10-19 10:50:04,179 ----------------------------------------------------------------------------------------------------
193
+ 2023-10-19 10:50:12,398 epoch 9 - iter 521/5212 - loss 0.21708792 - time (sec): 8.22 - samples/sec: 4102.45 - lr: 0.000006 - momentum: 0.000000
194
+ 2023-10-19 10:50:20,747 epoch 9 - iter 1042/5212 - loss 0.19390674 - time (sec): 16.57 - samples/sec: 4312.08 - lr: 0.000006 - momentum: 0.000000
195
+ 2023-10-19 10:50:29,009 epoch 9 - iter 1563/5212 - loss 0.20054262 - time (sec): 24.83 - samples/sec: 4341.33 - lr: 0.000006 - momentum: 0.000000
196
+ 2023-10-19 10:50:37,304 epoch 9 - iter 2084/5212 - loss 0.21128490 - time (sec): 33.12 - samples/sec: 4338.21 - lr: 0.000005 - momentum: 0.000000
197
+ 2023-10-19 10:50:45,779 epoch 9 - iter 2605/5212 - loss 0.21144254 - time (sec): 41.60 - samples/sec: 4433.69 - lr: 0.000005 - momentum: 0.000000
198
+ 2023-10-19 10:50:54,001 epoch 9 - iter 3126/5212 - loss 0.20955816 - time (sec): 49.82 - samples/sec: 4423.37 - lr: 0.000005 - momentum: 0.000000
199
+ 2023-10-19 10:51:02,390 epoch 9 - iter 3647/5212 - loss 0.21385600 - time (sec): 58.21 - samples/sec: 4438.87 - lr: 0.000004 - momentum: 0.000000
200
+ 2023-10-19 10:51:10,644 epoch 9 - iter 4168/5212 - loss 0.21026236 - time (sec): 66.46 - samples/sec: 4440.93 - lr: 0.000004 - momentum: 0.000000
201
+ 2023-10-19 10:51:19,014 epoch 9 - iter 4689/5212 - loss 0.20931466 - time (sec): 74.83 - samples/sec: 4411.30 - lr: 0.000004 - momentum: 0.000000
202
+ 2023-10-19 10:51:27,385 epoch 9 - iter 5210/5212 - loss 0.20947495 - time (sec): 83.20 - samples/sec: 4414.82 - lr: 0.000003 - momentum: 0.000000
203
+ 2023-10-19 10:51:27,418 ----------------------------------------------------------------------------------------------------
204
+ 2023-10-19 10:51:27,418 EPOCH 9 done: loss 0.2094 - lr: 0.000003
205
+ 2023-10-19 10:51:32,597 DEV : loss 0.1816394329071045 - f1-score (micro avg) 0.272
206
+ 2023-10-19 10:51:32,621 ----------------------------------------------------------------------------------------------------
207
+ 2023-10-19 10:51:41,163 epoch 10 - iter 521/5212 - loss 0.20993504 - time (sec): 8.54 - samples/sec: 4199.80 - lr: 0.000003 - momentum: 0.000000
208
+ 2023-10-19 10:51:49,336 epoch 10 - iter 1042/5212 - loss 0.19721566 - time (sec): 16.71 - samples/sec: 4408.14 - lr: 0.000003 - momentum: 0.000000
209
+ 2023-10-19 10:51:57,596 epoch 10 - iter 1563/5212 - loss 0.19886392 - time (sec): 24.97 - samples/sec: 4399.09 - lr: 0.000002 - momentum: 0.000000
210
+ 2023-10-19 10:52:06,090 epoch 10 - iter 2084/5212 - loss 0.20201347 - time (sec): 33.47 - samples/sec: 4376.06 - lr: 0.000002 - momentum: 0.000000
211
+ 2023-10-19 10:52:14,589 epoch 10 - iter 2605/5212 - loss 0.20351706 - time (sec): 41.97 - samples/sec: 4365.97 - lr: 0.000002 - momentum: 0.000000
212
+ 2023-10-19 10:52:23,017 epoch 10 - iter 3126/5212 - loss 0.20748962 - time (sec): 50.40 - samples/sec: 4427.54 - lr: 0.000001 - momentum: 0.000000
213
+ 2023-10-19 10:52:31,397 epoch 10 - iter 3647/5212 - loss 0.20967361 - time (sec): 58.78 - samples/sec: 4432.91 - lr: 0.000001 - momentum: 0.000000
214
+ 2023-10-19 10:52:39,724 epoch 10 - iter 4168/5212 - loss 0.20825590 - time (sec): 67.10 - samples/sec: 4426.18 - lr: 0.000001 - momentum: 0.000000
215
+ 2023-10-19 10:52:48,049 epoch 10 - iter 4689/5212 - loss 0.20604718 - time (sec): 75.43 - samples/sec: 4423.79 - lr: 0.000000 - momentum: 0.000000
216
+ 2023-10-19 10:52:56,230 epoch 10 - iter 5210/5212 - loss 0.20716771 - time (sec): 83.61 - samples/sec: 4391.22 - lr: 0.000000 - momentum: 0.000000
217
+ 2023-10-19 10:52:56,272 ----------------------------------------------------------------------------------------------------
218
+ 2023-10-19 10:52:56,272 EPOCH 10 done: loss 0.2072 - lr: 0.000000
219
+ 2023-10-19 10:53:01,417 DEV : loss 0.17811782658100128 - f1-score (micro avg) 0.2638
220
+ 2023-10-19 10:53:01,469 ----------------------------------------------------------------------------------------------------
221
+ 2023-10-19 10:53:01,470 Loading model from best epoch ...
222
+ 2023-10-19 10:53:01,547 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
223
+ 2023-10-19 10:53:07,781
224
+ Results:
225
+ - F-score (micro) 0.299
226
+ - F-score (macro) 0.1455
227
+ - Accuracy 0.177
228
+
229
+ By class:
230
+ precision recall f1-score support
231
+
232
+ LOC 0.4586 0.4605 0.4595 1214
233
+ PER 0.1394 0.0718 0.0948 808
234
+ ORG 0.0470 0.0198 0.0279 353
235
+ HumanProd 0.0000 0.0000 0.0000 15
236
+
237
+ micro avg 0.3498 0.2611 0.2990 2390
238
+ macro avg 0.1612 0.1380 0.1455 2390
239
+ weighted avg 0.2870 0.2611 0.2696 2390
240
+
241
+ 2023-10-19 10:53:07,782 ----------------------------------------------------------------------------------------------------