Hugging Face's logo

stefan-it
/
hmbench-topres19th-en-hmbert_tiny-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4

Token Classification
Flair
PyTorch
TensorBoard
English
sequence-tagger-model
Model card Files Metrics Community
stefan-it commited on
Commit
8a56398
·
1 Parent(s): 956f53f

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.1697796036.46dc0c540dd0.5704.14 +3 -0
  5. test.tsv +0 -0
  6. training.log +245 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e0d2a49ca68e06779ddc8722efac9a3926d6d4a5d7e3d9e1999e7fef1f82d760
3
+ size 19045986
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 10:00:54 0.0000 1.3710 0.1529 0.0000 0.0000 0.0000 0.0000
3
+ 2 10:01:12 0.0000 0.2189 0.1070 0.6383 0.1266 0.2113 0.1195
4
+ 3 10:01:31 0.0000 0.1759 0.0952 0.5963 0.2743 0.3757 0.2372
5
+ 4 10:01:50 0.0000 0.1598 0.0869 0.5657 0.4177 0.4806 0.3278
6
+ 5 10:02:09 0.0000 0.1495 0.0850 0.5795 0.4304 0.4939 0.3400
7
+ 6 10:02:27 0.0000 0.1404 0.0833 0.5471 0.5148 0.5304 0.3754
8
+ 7 10:02:46 0.0000 0.1368 0.0845 0.6124 0.4599 0.5253 0.3707
9
+ 8 10:03:04 0.0000 0.1312 0.0827 0.5882 0.5063 0.5442 0.3883
10
+ 9 10:03:23 0.0000 0.1270 0.0825 0.5837 0.5148 0.5471 0.3910
11
+ 10 10:03:42 0.0000 0.1262 0.0823 0.5865 0.5148 0.5483 0.3935
runs/events.out.tfevents.1697796036.46dc0c540dd0.5704.14 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7a3548765f556305e219919ba05a1ab8b74686d7453be60663e401541e02320b
3
+ size 434848
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,245 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-20 10:00:36,344 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-20 10:00:36,344 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=13, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-20 10:00:36,344 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-20 10:00:36,344 MultiCorpus: 6183 train + 680 dev + 2113 test sentences
52
+ - NER_HIPE_2022 Corpus: 6183 train + 680 dev + 2113 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/topres19th/en/with_doc_seperator
53
+ 2023-10-20 10:00:36,345 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-20 10:00:36,345 Train: 6183 sentences
55
+ 2023-10-20 10:00:36,345 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-20 10:00:36,345 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-20 10:00:36,345 Training Params:
58
+ 2023-10-20 10:00:36,345 - learning_rate: "3e-05"
59
+ 2023-10-20 10:00:36,345 - mini_batch_size: "8"
60
+ 2023-10-20 10:00:36,345 - max_epochs: "10"
61
+ 2023-10-20 10:00:36,345 - shuffle: "True"
62
+ 2023-10-20 10:00:36,345 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-20 10:00:36,345 Plugins:
64
+ 2023-10-20 10:00:36,345 - TensorboardLogger
65
+ 2023-10-20 10:00:36,345 - LinearScheduler | warmup_fraction: '0.1'
66
+ 2023-10-20 10:00:36,345 ----------------------------------------------------------------------------------------------------
67
+ 2023-10-20 10:00:36,345 Final evaluation on model from best epoch (best-model.pt)
68
+ 2023-10-20 10:00:36,345 - metric: "('micro avg', 'f1-score')"
69
+ 2023-10-20 10:00:36,345 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-20 10:00:36,345 Computation:
71
+ 2023-10-20 10:00:36,345 - compute on device: cuda:0
72
+ 2023-10-20 10:00:36,345 - embedding storage: none
73
+ 2023-10-20 10:00:36,345 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-20 10:00:36,345 Model training base path: "hmbench-topres19th/en-dbmdz/bert-tiny-historic-multilingual-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4"
75
+ 2023-10-20 10:00:36,345 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-20 10:00:36,345 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-20 10:00:36,345 Logging anything other than scalars to TensorBoard is currently not supported.
78
+ 2023-10-20 10:00:38,005 epoch 1 - iter 77/773 - loss 3.81241056 - time (sec): 1.66 - samples/sec: 7452.07 - lr: 0.000003 - momentum: 0.000000
79
+ 2023-10-20 10:00:39,655 epoch 1 - iter 154/773 - loss 3.61951758 - time (sec): 3.31 - samples/sec: 7488.42 - lr: 0.000006 - momentum: 0.000000
80
+ 2023-10-20 10:00:41,384 epoch 1 - iter 231/773 - loss 3.30275246 - time (sec): 5.04 - samples/sec: 7254.09 - lr: 0.000009 - momentum: 0.000000
81
+ 2023-10-20 10:00:43,171 epoch 1 - iter 308/773 - loss 2.89747658 - time (sec): 6.82 - samples/sec: 7201.04 - lr: 0.000012 - momentum: 0.000000
82
+ 2023-10-20 10:00:44,906 epoch 1 - iter 385/773 - loss 2.50629080 - time (sec): 8.56 - samples/sec: 7072.38 - lr: 0.000015 - momentum: 0.000000
83
+ 2023-10-20 10:00:46,625 epoch 1 - iter 462/773 - loss 2.15258275 - time (sec): 10.28 - samples/sec: 7082.99 - lr: 0.000018 - momentum: 0.000000
84
+ 2023-10-20 10:00:48,309 epoch 1 - iter 539/773 - loss 1.88716852 - time (sec): 11.96 - samples/sec: 7097.11 - lr: 0.000021 - momentum: 0.000000
85
+ 2023-10-20 10:00:50,102 epoch 1 - iter 616/773 - loss 1.66647173 - time (sec): 13.76 - samples/sec: 7139.05 - lr: 0.000024 - momentum: 0.000000
86
+ 2023-10-20 10:00:51,814 epoch 1 - iter 693/773 - loss 1.50059755 - time (sec): 15.47 - samples/sec: 7193.98 - lr: 0.000027 - momentum: 0.000000
87
+ 2023-10-20 10:00:53,502 epoch 1 - iter 770/773 - loss 1.37496153 - time (sec): 17.16 - samples/sec: 7219.32 - lr: 0.000030 - momentum: 0.000000
88
+ 2023-10-20 10:00:53,558 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-20 10:00:53,558 EPOCH 1 done: loss 1.3710 - lr: 0.000030
90
+ 2023-10-20 10:00:54,265 DEV : loss 0.15286844968795776 - f1-score (micro avg) 0.0
91
+ 2023-10-20 10:00:54,278 ----------------------------------------------------------------------------------------------------
92
+ 2023-10-20 10:00:55,770 epoch 2 - iter 77/773 - loss 0.23214451 - time (sec): 1.49 - samples/sec: 8859.76 - lr: 0.000030 - momentum: 0.000000
93
+ 2023-10-20 10:00:57,781 epoch 2 - iter 154/773 - loss 0.24097431 - time (sec): 3.50 - samples/sec: 7562.15 - lr: 0.000029 - momentum: 0.000000
94
+ 2023-10-20 10:00:59,442 epoch 2 - iter 231/773 - loss 0.24356049 - time (sec): 5.16 - samples/sec: 7206.01 - lr: 0.000029 - momentum: 0.000000
95
+ 2023-10-20 10:01:01,167 epoch 2 - iter 308/773 - loss 0.24470903 - time (sec): 6.89 - samples/sec: 7125.02 - lr: 0.000029 - momentum: 0.000000
96
+ 2023-10-20 10:01:02,897 epoch 2 - iter 385/773 - loss 0.24262456 - time (sec): 8.62 - samples/sec: 7071.31 - lr: 0.000028 - momentum: 0.000000
97
+ 2023-10-20 10:01:04,655 epoch 2 - iter 462/773 - loss 0.23888426 - time (sec): 10.38 - samples/sec: 7082.59 - lr: 0.000028 - momentum: 0.000000
98
+ 2023-10-20 10:01:06,429 epoch 2 - iter 539/773 - loss 0.22964243 - time (sec): 12.15 - samples/sec: 7132.08 - lr: 0.000028 - momentum: 0.000000
99
+ 2023-10-20 10:01:08,166 epoch 2 - iter 616/773 - loss 0.23048232 - time (sec): 13.89 - samples/sec: 7075.56 - lr: 0.000027 - momentum: 0.000000
100
+ 2023-10-20 10:01:09,923 epoch 2 - iter 693/773 - loss 0.22376378 - time (sec): 15.64 - samples/sec: 7042.07 - lr: 0.000027 - momentum: 0.000000
101
+ 2023-10-20 10:01:11,712 epoch 2 - iter 770/773 - loss 0.21921775 - time (sec): 17.43 - samples/sec: 7098.42 - lr: 0.000027 - momentum: 0.000000
102
+ 2023-10-20 10:01:11,774 ----------------------------------------------------------------------------------------------------
103
+ 2023-10-20 10:01:11,774 EPOCH 2 done: loss 0.2189 - lr: 0.000027
104
+ 2023-10-20 10:01:12,842 DEV : loss 0.10697879642248154 - f1-score (micro avg) 0.2113
105
+ 2023-10-20 10:01:12,854 saving best model
106
+ 2023-10-20 10:01:12,882 ----------------------------------------------------------------------------------------------------
107
+ 2023-10-20 10:01:14,585 epoch 3 - iter 77/773 - loss 0.17960081 - time (sec): 1.70 - samples/sec: 7327.95 - lr: 0.000026 - momentum: 0.000000
108
+ 2023-10-20 10:01:16,295 epoch 3 - iter 154/773 - loss 0.17934017 - time (sec): 3.41 - samples/sec: 7047.87 - lr: 0.000026 - momentum: 0.000000
109
+ 2023-10-20 10:01:18,025 epoch 3 - iter 231/773 - loss 0.17558716 - time (sec): 5.14 - samples/sec: 7188.59 - lr: 0.000026 - momentum: 0.000000
110
+ 2023-10-20 10:01:19,753 epoch 3 - iter 308/773 - loss 0.17212991 - time (sec): 6.87 - samples/sec: 7197.64 - lr: 0.000025 - momentum: 0.000000
111
+ 2023-10-20 10:01:21,516 epoch 3 - iter 385/773 - loss 0.17346036 - time (sec): 8.63 - samples/sec: 7219.64 - lr: 0.000025 - momentum: 0.000000
112
+ 2023-10-20 10:01:23,201 epoch 3 - iter 462/773 - loss 0.17673051 - time (sec): 10.32 - samples/sec: 7092.46 - lr: 0.000025 - momentum: 0.000000
113
+ 2023-10-20 10:01:24,962 epoch 3 - iter 539/773 - loss 0.17649304 - time (sec): 12.08 - samples/sec: 7123.71 - lr: 0.000024 - momentum: 0.000000
114
+ 2023-10-20 10:01:26,682 epoch 3 - iter 616/773 - loss 0.17628556 - time (sec): 13.80 - samples/sec: 7172.27 - lr: 0.000024 - momentum: 0.000000
115
+ 2023-10-20 10:01:28,419 epoch 3 - iter 693/773 - loss 0.17756531 - time (sec): 15.54 - samples/sec: 7157.36 - lr: 0.000024 - momentum: 0.000000
116
+ 2023-10-20 10:01:30,165 epoch 3 - iter 770/773 - loss 0.17579288 - time (sec): 17.28 - samples/sec: 7169.81 - lr: 0.000023 - momentum: 0.000000
117
+ 2023-10-20 10:01:30,225 ----------------------------------------------------------------------------------------------------
118
+ 2023-10-20 10:01:30,225 EPOCH 3 done: loss 0.1759 - lr: 0.000023
119
+ 2023-10-20 10:01:31,314 DEV : loss 0.0952216237783432 - f1-score (micro avg) 0.3757
120
+ 2023-10-20 10:01:31,327 saving best model
121
+ 2023-10-20 10:01:31,361 ----------------------------------------------------------------------------------------------------
122
+ 2023-10-20 10:01:33,173 epoch 4 - iter 77/773 - loss 0.18477387 - time (sec): 1.81 - samples/sec: 7281.37 - lr: 0.000023 - momentum: 0.000000
123
+ 2023-10-20 10:01:34,897 epoch 4 - iter 154/773 - loss 0.18279801 - time (sec): 3.53 - samples/sec: 7010.16 - lr: 0.000023 - momentum: 0.000000
124
+ 2023-10-20 10:01:36,632 epoch 4 - iter 231/773 - loss 0.17577730 - time (sec): 5.27 - samples/sec: 6871.73 - lr: 0.000022 - momentum: 0.000000
125
+ 2023-10-20 10:01:38,395 epoch 4 - iter 308/773 - loss 0.17374644 - time (sec): 7.03 - samples/sec: 7028.03 - lr: 0.000022 - momentum: 0.000000
126
+ 2023-10-20 10:01:40,155 epoch 4 - iter 385/773 - loss 0.17036912 - time (sec): 8.79 - samples/sec: 6952.04 - lr: 0.000022 - momentum: 0.000000
127
+ 2023-10-20 10:01:41,895 epoch 4 - iter 462/773 - loss 0.16676008 - time (sec): 10.53 - samples/sec: 6967.86 - lr: 0.000021 - momentum: 0.000000
128
+ 2023-10-20 10:01:43,678 epoch 4 - iter 539/773 - loss 0.16364124 - time (sec): 12.32 - samples/sec: 6973.48 - lr: 0.000021 - momentum: 0.000000
129
+ 2023-10-20 10:01:45,415 epoch 4 - iter 616/773 - loss 0.16324635 - time (sec): 14.05 - samples/sec: 7019.45 - lr: 0.000021 - momentum: 0.000000
130
+ 2023-10-20 10:01:47,202 epoch 4 - iter 693/773 - loss 0.16215188 - time (sec): 15.84 - samples/sec: 7030.24 - lr: 0.000020 - momentum: 0.000000
131
+ 2023-10-20 10:01:48,917 epoch 4 - iter 770/773 - loss 0.15946543 - time (sec): 17.56 - samples/sec: 7046.10 - lr: 0.000020 - momentum: 0.000000
132
+ 2023-10-20 10:01:48,986 ----------------------------------------------------------------------------------------------------
133
+ 2023-10-20 10:01:48,986 EPOCH 4 done: loss 0.1598 - lr: 0.000020
134
+ 2023-10-20 10:01:50,067 DEV : loss 0.08685088157653809 - f1-score (micro avg) 0.4806
135
+ 2023-10-20 10:01:50,079 saving best model
136
+ 2023-10-20 10:01:50,119 ----------------------------------------------------------------------------------------------------
137
+ 2023-10-20 10:01:51,873 epoch 5 - iter 77/773 - loss 0.14344519 - time (sec): 1.75 - samples/sec: 6814.44 - lr: 0.000020 - momentum: 0.000000
138
+ 2023-10-20 10:01:53,683 epoch 5 - iter 154/773 - loss 0.15258628 - time (sec): 3.56 - samples/sec: 7121.76 - lr: 0.000019 - momentum: 0.000000
139
+ 2023-10-20 10:01:55,431 epoch 5 - iter 231/773 - loss 0.15926291 - time (sec): 5.31 - samples/sec: 7141.00 - lr: 0.000019 - momentum: 0.000000
140
+ 2023-10-20 10:01:57,170 epoch 5 - iter 308/773 - loss 0.15525040 - time (sec): 7.05 - samples/sec: 7002.21 - lr: 0.000019 - momentum: 0.000000
141
+ 2023-10-20 10:01:59,041 epoch 5 - iter 385/773 - loss 0.15298376 - time (sec): 8.92 - samples/sec: 6989.00 - lr: 0.000018 - momentum: 0.000000
142
+ 2023-10-20 10:02:00,849 epoch 5 - iter 462/773 - loss 0.14900655 - time (sec): 10.73 - samples/sec: 6924.76 - lr: 0.000018 - momentum: 0.000000
143
+ 2023-10-20 10:02:02,653 epoch 5 - iter 539/773 - loss 0.14781168 - time (sec): 12.53 - samples/sec: 6941.72 - lr: 0.000018 - momentum: 0.000000
144
+ 2023-10-20 10:02:04,495 epoch 5 - iter 616/773 - loss 0.15062811 - time (sec): 14.38 - samples/sec: 6896.32 - lr: 0.000017 - momentum: 0.000000
145
+ 2023-10-20 10:02:06,282 epoch 5 - iter 693/773 - loss 0.15115235 - time (sec): 16.16 - samples/sec: 6910.24 - lr: 0.000017 - momentum: 0.000000
146
+ 2023-10-20 10:02:08,002 epoch 5 - iter 770/773 - loss 0.14968908 - time (sec): 17.88 - samples/sec: 6919.99 - lr: 0.000017 - momentum: 0.000000
147
+ 2023-10-20 10:02:08,074 ----------------------------------------------------------------------------------------------------
148
+ 2023-10-20 10:02:08,074 EPOCH 5 done: loss 0.1495 - lr: 0.000017
149
+ 2023-10-20 10:02:09,173 DEV : loss 0.08496666699647903 - f1-score (micro avg) 0.4939
150
+ 2023-10-20 10:02:09,185 saving best model
151
+ 2023-10-20 10:02:09,219 ----------------------------------------------------------------------------------------------------
152
+ 2023-10-20 10:02:10,915 epoch 6 - iter 77/773 - loss 0.13683614 - time (sec): 1.70 - samples/sec: 6668.03 - lr: 0.000016 - momentum: 0.000000
153
+ 2023-10-20 10:02:12,711 epoch 6 - iter 154/773 - loss 0.14211735 - time (sec): 3.49 - samples/sec: 6801.54 - lr: 0.000016 - momentum: 0.000000
154
+ 2023-10-20 10:02:14,513 epoch 6 - iter 231/773 - loss 0.14128117 - time (sec): 5.29 - samples/sec: 6853.01 - lr: 0.000016 - momentum: 0.000000
155
+ 2023-10-20 10:02:16,289 epoch 6 - iter 308/773 - loss 0.13590630 - time (sec): 7.07 - samples/sec: 6891.57 - lr: 0.000015 - momentum: 0.000000
156
+ 2023-10-20 10:02:17,903 epoch 6 - iter 385/773 - loss 0.14259420 - time (sec): 8.68 - samples/sec: 7003.45 - lr: 0.000015 - momentum: 0.000000
157
+ 2023-10-20 10:02:19,585 epoch 6 - iter 462/773 - loss 0.14552202 - time (sec): 10.37 - samples/sec: 7037.70 - lr: 0.000015 - momentum: 0.000000
158
+ 2023-10-20 10:02:21,375 epoch 6 - iter 539/773 - loss 0.14254133 - time (sec): 12.16 - samples/sec: 7031.44 - lr: 0.000014 - momentum: 0.000000
159
+ 2023-10-20 10:02:23,107 epoch 6 - iter 616/773 - loss 0.14102050 - time (sec): 13.89 - samples/sec: 7092.37 - lr: 0.000014 - momentum: 0.000000
160
+ 2023-10-20 10:02:24,871 epoch 6 - iter 693/773 - loss 0.13907785 - time (sec): 15.65 - samples/sec: 7098.85 - lr: 0.000014 - momentum: 0.000000
161
+ 2023-10-20 10:02:26,636 epoch 6 - iter 770/773 - loss 0.14100570 - time (sec): 17.42 - samples/sec: 7095.47 - lr: 0.000013 - momentum: 0.000000
162
+ 2023-10-20 10:02:26,717 ----------------------------------------------------------------------------------------------------
163
+ 2023-10-20 10:02:26,717 EPOCH 6 done: loss 0.1404 - lr: 0.000013
164
+ 2023-10-20 10:02:27,791 DEV : loss 0.08332625776529312 - f1-score (micro avg) 0.5304
165
+ 2023-10-20 10:02:27,802 saving best model
166
+ 2023-10-20 10:02:27,836 ----------------------------------------------------------------------------------------------------
167
+ 2023-10-20 10:02:29,593 epoch 7 - iter 77/773 - loss 0.14476114 - time (sec): 1.76 - samples/sec: 7188.18 - lr: 0.000013 - momentum: 0.000000
168
+ 2023-10-20 10:02:31,331 epoch 7 - iter 154/773 - loss 0.14402242 - time (sec): 3.49 - samples/sec: 7109.92 - lr: 0.000013 - momentum: 0.000000
169
+ 2023-10-20 10:02:33,063 epoch 7 - iter 231/773 - loss 0.14036025 - time (sec): 5.23 - samples/sec: 7082.29 - lr: 0.000012 - momentum: 0.000000
170
+ 2023-10-20 10:02:34,756 epoch 7 - iter 308/773 - loss 0.14534767 - time (sec): 6.92 - samples/sec: 7173.84 - lr: 0.000012 - momentum: 0.000000
171
+ 2023-10-20 10:02:36,559 epoch 7 - iter 385/773 - loss 0.14110367 - time (sec): 8.72 - samples/sec: 7275.71 - lr: 0.000012 - momentum: 0.000000
172
+ 2023-10-20 10:02:38,203 epoch 7 - iter 462/773 - loss 0.13753663 - time (sec): 10.37 - samples/sec: 7309.11 - lr: 0.000011 - momentum: 0.000000
173
+ 2023-10-20 10:02:39,822 epoch 7 - iter 539/773 - loss 0.13875122 - time (sec): 11.99 - samples/sec: 7318.44 - lr: 0.000011 - momentum: 0.000000
174
+ 2023-10-20 10:02:41,511 epoch 7 - iter 616/773 - loss 0.13955186 - time (sec): 13.67 - samples/sec: 7296.14 - lr: 0.000011 - momentum: 0.000000
175
+ 2023-10-20 10:02:43,257 epoch 7 - iter 693/773 - loss 0.13895365 - time (sec): 15.42 - samples/sec: 7257.89 - lr: 0.000010 - momentum: 0.000000
176
+ 2023-10-20 10:02:44,971 epoch 7 - iter 770/773 - loss 0.13686198 - time (sec): 17.13 - samples/sec: 7227.08 - lr: 0.000010 - momentum: 0.000000
177
+ 2023-10-20 10:02:45,031 ----------------------------------------------------------------------------------------------------
178
+ 2023-10-20 10:02:45,031 EPOCH 7 done: loss 0.1368 - lr: 0.000010
179
+ 2023-10-20 10:02:46,107 DEV : loss 0.08445805311203003 - f1-score (micro avg) 0.5253
180
+ 2023-10-20 10:02:46,118 ----------------------------------------------------------------------------------------------------
181
+ 2023-10-20 10:02:47,862 epoch 8 - iter 77/773 - loss 0.12222683 - time (sec): 1.74 - samples/sec: 7201.77 - lr: 0.000010 - momentum: 0.000000
182
+ 2023-10-20 10:02:49,623 epoch 8 - iter 154/773 - loss 0.12160289 - time (sec): 3.50 - samples/sec: 7112.73 - lr: 0.000009 - momentum: 0.000000
183
+ 2023-10-20 10:02:51,352 epoch 8 - iter 231/773 - loss 0.12601210 - time (sec): 5.23 - samples/sec: 7119.34 - lr: 0.000009 - momentum: 0.000000
184
+ 2023-10-20 10:02:53,104 epoch 8 - iter 308/773 - loss 0.12825557 - time (sec): 6.99 - samples/sec: 7114.87 - lr: 0.000009 - momentum: 0.000000
185
+ 2023-10-20 10:02:54,802 epoch 8 - iter 385/773 - loss 0.13149787 - time (sec): 8.68 - samples/sec: 7182.75 - lr: 0.000008 - momentum: 0.000000
186
+ 2023-10-20 10:02:56,573 epoch 8 - iter 462/773 - loss 0.13242894 - time (sec): 10.45 - samples/sec: 7128.04 - lr: 0.000008 - momentum: 0.000000
187
+ 2023-10-20 10:02:58,284 epoch 8 - iter 539/773 - loss 0.13472536 - time (sec): 12.17 - samples/sec: 7127.85 - lr: 0.000008 - momentum: 0.000000
188
+ 2023-10-20 10:03:00,085 epoch 8 - iter 616/773 - loss 0.13365370 - time (sec): 13.97 - samples/sec: 7098.48 - lr: 0.000007 - momentum: 0.000000
189
+ 2023-10-20 10:03:01,833 epoch 8 - iter 693/773 - loss 0.13290360 - time (sec): 15.71 - samples/sec: 7091.52 - lr: 0.000007 - momentum: 0.000000
190
+ 2023-10-20 10:03:03,535 epoch 8 - iter 770/773 - loss 0.13146102 - time (sec): 17.42 - samples/sec: 7107.06 - lr: 0.000007 - momentum: 0.000000
191
+ 2023-10-20 10:03:03,599 ----------------------------------------------------------------------------------------------------
192
+ 2023-10-20 10:03:03,599 EPOCH 8 done: loss 0.1312 - lr: 0.000007
193
+ 2023-10-20 10:03:04,675 DEV : loss 0.08267096430063248 - f1-score (micro avg) 0.5442
194
+ 2023-10-20 10:03:04,687 saving best model
195
+ 2023-10-20 10:03:04,730 ----------------------------------------------------------------------------------------------------
196
+ 2023-10-20 10:03:06,377 epoch 9 - iter 77/773 - loss 0.13426845 - time (sec): 1.65 - samples/sec: 7221.24 - lr: 0.000006 - momentum: 0.000000
197
+ 2023-10-20 10:03:08,121 epoch 9 - iter 154/773 - loss 0.12764335 - time (sec): 3.39 - samples/sec: 7073.83 - lr: 0.000006 - momentum: 0.000000
198
+ 2023-10-20 10:03:09,885 epoch 9 - iter 231/773 - loss 0.13064958 - time (sec): 5.15 - samples/sec: 7054.91 - lr: 0.000006 - momentum: 0.000000
199
+ 2023-10-20 10:03:11,689 epoch 9 - iter 308/773 - loss 0.12547793 - time (sec): 6.96 - samples/sec: 7098.59 - lr: 0.000005 - momentum: 0.000000
200
+ 2023-10-20 10:03:13,457 epoch 9 - iter 385/773 - loss 0.12185205 - time (sec): 8.73 - samples/sec: 7038.94 - lr: 0.000005 - momentum: 0.000000
201
+ 2023-10-20 10:03:15,360 epoch 9 - iter 462/773 - loss 0.12534457 - time (sec): 10.63 - samples/sec: 6904.04 - lr: 0.000005 - momentum: 0.000000
202
+ 2023-10-20 10:03:17,212 epoch 9 - iter 539/773 - loss 0.12499457 - time (sec): 12.48 - samples/sec: 6875.73 - lr: 0.000004 - momentum: 0.000000
203
+ 2023-10-20 10:03:19,029 epoch 9 - iter 616/773 - loss 0.12754088 - time (sec): 14.30 - samples/sec: 6896.18 - lr: 0.000004 - momentum: 0.000000
204
+ 2023-10-20 10:03:20,794 epoch 9 - iter 693/773 - loss 0.12629823 - time (sec): 16.06 - samples/sec: 6965.13 - lr: 0.000004 - momentum: 0.000000
205
+ 2023-10-20 10:03:22,545 epoch 9 - iter 770/773 - loss 0.12671079 - time (sec): 17.81 - samples/sec: 6951.05 - lr: 0.000003 - momentum: 0.000000
206
+ 2023-10-20 10:03:22,608 ----------------------------------------------------------------------------------------------------
207
+ 2023-10-20 10:03:22,609 EPOCH 9 done: loss 0.1270 - lr: 0.000003
208
+ 2023-10-20 10:03:23,687 DEV : loss 0.08247760683298111 - f1-score (micro avg) 0.5471
209
+ 2023-10-20 10:03:23,698 saving best model
210
+ 2023-10-20 10:03:23,736 ----------------------------------------------------------------------------------------------------
211
+ 2023-10-20 10:03:25,433 epoch 10 - iter 77/773 - loss 0.14764006 - time (sec): 1.70 - samples/sec: 6905.70 - lr: 0.000003 - momentum: 0.000000
212
+ 2023-10-20 10:03:27,260 epoch 10 - iter 154/773 - loss 0.13385632 - time (sec): 3.52 - samples/sec: 6919.85 - lr: 0.000003 - momentum: 0.000000
213
+ 2023-10-20 10:03:28,975 epoch 10 - iter 231/773 - loss 0.12965434 - time (sec): 5.24 - samples/sec: 7108.23 - lr: 0.000002 - momentum: 0.000000
214
+ 2023-10-20 10:03:30,805 epoch 10 - iter 308/773 - loss 0.12967758 - time (sec): 7.07 - samples/sec: 7041.74 - lr: 0.000002 - momentum: 0.000000
215
+ 2023-10-20 10:03:32,607 epoch 10 - iter 385/773 - loss 0.13165440 - time (sec): 8.87 - samples/sec: 6989.40 - lr: 0.000002 - momentum: 0.000000
216
+ 2023-10-20 10:03:34,348 epoch 10 - iter 462/773 - loss 0.13289038 - time (sec): 10.61 - samples/sec: 6946.57 - lr: 0.000001 - momentum: 0.000000
217
+ 2023-10-20 10:03:36,176 epoch 10 - iter 539/773 - loss 0.13066958 - time (sec): 12.44 - samples/sec: 6936.41 - lr: 0.000001 - momentum: 0.000000
218
+ 2023-10-20 10:03:37,989 epoch 10 - iter 616/773 - loss 0.12645987 - time (sec): 14.25 - samples/sec: 6961.31 - lr: 0.000001 - momentum: 0.000000
219
+ 2023-10-20 10:03:39,744 epoch 10 - iter 693/773 - loss 0.12677426 - time (sec): 16.01 - samples/sec: 6945.06 - lr: 0.000000 - momentum: 0.000000
220
+ 2023-10-20 10:03:41,586 epoch 10 - iter 770/773 - loss 0.12613690 - time (sec): 17.85 - samples/sec: 6927.38 - lr: 0.000000 - momentum: 0.000000
221
+ 2023-10-20 10:03:41,655 ----------------------------------------------------------------------------------------------------
222
+ 2023-10-20 10:03:41,655 EPOCH 10 done: loss 0.1262 - lr: 0.000000
223
+ 2023-10-20 10:03:42,740 DEV : loss 0.08230794966220856 - f1-score (micro avg) 0.5483
224
+ 2023-10-20 10:03:42,752 saving best model
225
+ 2023-10-20 10:03:42,817 ----------------------------------------------------------------------------------------------------
226
+ 2023-10-20 10:03:42,817 Loading model from best epoch ...
227
+ 2023-10-20 10:03:42,890 SequenceTagger predicts: Dictionary with 13 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-BUILDING, B-BUILDING, E-BUILDING, I-BUILDING, S-STREET, B-STREET, E-STREET, I-STREET
228
+ 2023-10-20 10:03:45,758
229
+ Results:
230
+ - F-score (micro) 0.5024
231
+ - F-score (macro) 0.189
232
+ - Accuracy 0.3412
233
+
234
+ By class:
235
+ precision recall f1-score support
236
+
237
+ LOC 0.5714 0.5624 0.5669 946
238
+ BUILDING 0.0000 0.0000 0.0000 185
239
+ STREET 0.0000 0.0000 0.0000 56
240
+
241
+ micro avg 0.5714 0.4482 0.5024 1187
242
+ macro avg 0.1905 0.1875 0.1890 1187
243
+ weighted avg 0.4554 0.4482 0.4518 1187
244
+
245
+ 2023-10-20 10:03:45,758 ----------------------------------------------------------------------------------------------------