Hugging Face's logo

stefan-it
/
hmbench-topres19th-en-hmbert_tiny-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3

Token Classification
Flair
PyTorch
TensorBoard
English
sequence-tagger-model
Model card Files Metrics Community
stefan-it commited on
Commit
32e542b
·
1 Parent(s): 104d8b5

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.1697795322.46dc0c540dd0.5704.11 +3 -0
  5. test.tsv +0 -0
  6. training.log +244 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6964c9bdd5e14c44e6a5dc22cc120e16fb3c485995e2ee93abf0ed8864946aa1
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 09:49:00 0.0000 0.8589 0.1377 0.0000 0.0000 0.0000 0.0000
3
+ 2 09:49:19 0.0000 0.1896 0.0960 0.5323 0.2785 0.3657 0.2308
4
+ 3 09:49:38 0.0000 0.1552 0.0913 0.6204 0.3586 0.4545 0.3036
5
+ 4 09:49:56 0.0000 0.1362 0.0875 0.6080 0.4515 0.5182 0.3627
6
+ 5 09:50:15 0.0000 0.1256 0.0846 0.5455 0.5316 0.5385 0.3841
7
+ 6 09:50:33 0.0000 0.1160 0.0856 0.5721 0.5021 0.5348 0.3839
8
+ 7 09:50:52 0.0000 0.1107 0.0840 0.5708 0.5443 0.5572 0.4044
9
+ 8 09:51:10 0.0000 0.1044 0.0872 0.5799 0.5359 0.5570 0.4032
10
+ 9 09:51:29 0.0000 0.1020 0.0875 0.6019 0.5232 0.5598 0.4092
11
+ 10 09:51:48 0.0000 0.1004 0.0881 0.6029 0.5316 0.5650 0.4145
runs/events.out.tfevents.1697795322.46dc0c540dd0.5704.11 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c058b225d9333e9ee208c6213e2bef8f5e46166152a82fa039567557f851a7fd
3
+ size 434848
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,244 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-20 09:48:42,765 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-20 09:48:42,765 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 09:48:42,765 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-20 09:48:42,765 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 09:48:42,765 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-20 09:48:42,765 Train: 6183 sentences
55
+ 2023-10-20 09:48:42,765 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-20 09:48:42,765 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-20 09:48:42,765 Training Params:
58
+ 2023-10-20 09:48:42,765 - learning_rate: "5e-05"
59
+ 2023-10-20 09:48:42,765 - mini_batch_size: "8"
60
+ 2023-10-20 09:48:42,765 - max_epochs: "10"
61
+ 2023-10-20 09:48:42,765 - shuffle: "True"
62
+ 2023-10-20 09:48:42,765 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-20 09:48:42,766 Plugins:
64
+ 2023-10-20 09:48:42,766 - TensorboardLogger
65
+ 2023-10-20 09:48:42,766 - LinearScheduler | warmup_fraction: '0.1'
66
+ 2023-10-20 09:48:42,766 ----------------------------------------------------------------------------------------------------
67
+ 2023-10-20 09:48:42,766 Final evaluation on model from best epoch (best-model.pt)
68
+ 2023-10-20 09:48:42,766 - metric: "('micro avg', 'f1-score')"
69
+ 2023-10-20 09:48:42,766 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-20 09:48:42,766 Computation:
71
+ 2023-10-20 09:48:42,766 - compute on device: cuda:0
72
+ 2023-10-20 09:48:42,766 - embedding storage: none
73
+ 2023-10-20 09:48:42,766 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-20 09:48:42,766 Model training base path: "hmbench-topres19th/en-dbmdz/bert-tiny-historic-multilingual-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3"
75
+ 2023-10-20 09:48:42,766 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-20 09:48:42,766 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-20 09:48:42,766 Logging anything other than scalars to TensorBoard is currently not supported.
78
+ 2023-10-20 09:48:44,557 epoch 1 - iter 77/773 - loss 2.87926452 - time (sec): 1.79 - samples/sec: 6926.45 - lr: 0.000005 - momentum: 0.000000
79
+ 2023-10-20 09:48:46,323 epoch 1 - iter 154/773 - loss 2.58090791 - time (sec): 3.56 - samples/sec: 7242.62 - lr: 0.000010 - momentum: 0.000000
80
+ 2023-10-20 09:48:48,037 epoch 1 - iter 231/773 - loss 2.19347159 - time (sec): 5.27 - samples/sec: 7066.32 - lr: 0.000015 - momentum: 0.000000
81
+ 2023-10-20 09:48:49,762 epoch 1 - iter 308/773 - loss 1.79530466 - time (sec): 7.00 - samples/sec: 6939.16 - lr: 0.000020 - momentum: 0.000000
82
+ 2023-10-20 09:48:51,483 epoch 1 - iter 385/773 - loss 1.49005800 - time (sec): 8.72 - samples/sec: 6982.79 - lr: 0.000025 - momentum: 0.000000
83
+ 2023-10-20 09:48:53,275 epoch 1 - iter 462/773 - loss 1.28696614 - time (sec): 10.51 - samples/sec: 6966.72 - lr: 0.000030 - momentum: 0.000000
84
+ 2023-10-20 09:48:55,026 epoch 1 - iter 539/773 - loss 1.12674983 - time (sec): 12.26 - samples/sec: 7093.68 - lr: 0.000035 - momentum: 0.000000
85
+ 2023-10-20 09:48:56,741 epoch 1 - iter 616/773 - loss 1.02283949 - time (sec): 13.97 - samples/sec: 7070.20 - lr: 0.000040 - momentum: 0.000000
86
+ 2023-10-20 09:48:58,434 epoch 1 - iter 693/773 - loss 0.93498457 - time (sec): 15.67 - samples/sec: 7095.90 - lr: 0.000045 - momentum: 0.000000
87
+ 2023-10-20 09:49:00,161 epoch 1 - iter 770/773 - loss 0.86157976 - time (sec): 17.39 - samples/sec: 7118.20 - lr: 0.000050 - momentum: 0.000000
88
+ 2023-10-20 09:49:00,225 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-20 09:49:00,225 EPOCH 1 done: loss 0.8589 - lr: 0.000050
90
+ 2023-10-20 09:49:00,920 DEV : loss 0.13765180110931396 - f1-score (micro avg) 0.0
91
+ 2023-10-20 09:49:00,931 ----------------------------------------------------------------------------------------------------
92
+ 2023-10-20 09:49:02,640 epoch 2 - iter 77/773 - loss 0.21698829 - time (sec): 1.71 - samples/sec: 7968.37 - lr: 0.000049 - momentum: 0.000000
93
+ 2023-10-20 09:49:04,280 epoch 2 - iter 154/773 - loss 0.20811745 - time (sec): 3.35 - samples/sec: 7749.56 - lr: 0.000049 - momentum: 0.000000
94
+ 2023-10-20 09:49:06,063 epoch 2 - iter 231/773 - loss 0.20796981 - time (sec): 5.13 - samples/sec: 7473.80 - lr: 0.000048 - momentum: 0.000000
95
+ 2023-10-20 09:49:08,149 epoch 2 - iter 308/773 - loss 0.20289547 - time (sec): 7.22 - samples/sec: 6999.97 - lr: 0.000048 - momentum: 0.000000
96
+ 2023-10-20 09:49:09,929 epoch 2 - iter 385/773 - loss 0.20413434 - time (sec): 9.00 - samples/sec: 6985.69 - lr: 0.000047 - momentum: 0.000000
97
+ 2023-10-20 09:49:11,712 epoch 2 - iter 462/773 - loss 0.19808681 - time (sec): 10.78 - samples/sec: 7019.02 - lr: 0.000047 - momentum: 0.000000
98
+ 2023-10-20 09:49:13,398 epoch 2 - iter 539/773 - loss 0.19621468 - time (sec): 12.47 - samples/sec: 7043.99 - lr: 0.000046 - momentum: 0.000000
99
+ 2023-10-20 09:49:15,069 epoch 2 - iter 616/773 - loss 0.19508540 - time (sec): 14.14 - samples/sec: 7028.74 - lr: 0.000046 - momentum: 0.000000
100
+ 2023-10-20 09:49:16,802 epoch 2 - iter 693/773 - loss 0.19324075 - time (sec): 15.87 - samples/sec: 7033.26 - lr: 0.000045 - momentum: 0.000000
101
+ 2023-10-20 09:49:18,554 epoch 2 - iter 770/773 - loss 0.18967567 - time (sec): 17.62 - samples/sec: 7025.86 - lr: 0.000044 - momentum: 0.000000
102
+ 2023-10-20 09:49:18,621 ----------------------------------------------------------------------------------------------------
103
+ 2023-10-20 09:49:18,621 EPOCH 2 done: loss 0.1896 - lr: 0.000044
104
+ 2023-10-20 09:49:19,698 DEV : loss 0.09595523029565811 - f1-score (micro avg) 0.3657
105
+ 2023-10-20 09:49:19,709 saving best model
106
+ 2023-10-20 09:49:19,738 ----------------------------------------------------------------------------------------------------
107
+ 2023-10-20 09:49:21,498 epoch 3 - iter 77/773 - loss 0.18262781 - time (sec): 1.76 - samples/sec: 6333.46 - lr: 0.000044 - momentum: 0.000000
108
+ 2023-10-20 09:49:23,277 epoch 3 - iter 154/773 - loss 0.16257290 - time (sec): 3.54 - samples/sec: 6881.12 - lr: 0.000043 - momentum: 0.000000
109
+ 2023-10-20 09:49:25,002 epoch 3 - iter 231/773 - loss 0.16928954 - time (sec): 5.26 - samples/sec: 6976.00 - lr: 0.000043 - momentum: 0.000000
110
+ 2023-10-20 09:49:26,737 epoch 3 - iter 308/773 - loss 0.16039613 - time (sec): 7.00 - samples/sec: 6906.62 - lr: 0.000042 - momentum: 0.000000
111
+ 2023-10-20 09:49:28,481 epoch 3 - iter 385/773 - loss 0.15828599 - time (sec): 8.74 - samples/sec: 6981.46 - lr: 0.000042 - momentum: 0.000000
112
+ 2023-10-20 09:49:30,238 epoch 3 - iter 462/773 - loss 0.15435750 - time (sec): 10.50 - samples/sec: 7057.56 - lr: 0.000041 - momentum: 0.000000
113
+ 2023-10-20 09:49:31,993 epoch 3 - iter 539/773 - loss 0.15526689 - time (sec): 12.25 - samples/sec: 7053.55 - lr: 0.000041 - momentum: 0.000000
114
+ 2023-10-20 09:49:33,650 epoch 3 - iter 616/773 - loss 0.15528455 - time (sec): 13.91 - samples/sec: 7090.48 - lr: 0.000040 - momentum: 0.000000
115
+ 2023-10-20 09:49:35,406 epoch 3 - iter 693/773 - loss 0.15441732 - time (sec): 15.67 - samples/sec: 7124.04 - lr: 0.000039 - momentum: 0.000000
116
+ 2023-10-20 09:49:37,244 epoch 3 - iter 770/773 - loss 0.15466267 - time (sec): 17.50 - samples/sec: 7073.24 - lr: 0.000039 - momentum: 0.000000
117
+ 2023-10-20 09:49:37,308 ----------------------------------------------------------------------------------------------------
118
+ 2023-10-20 09:49:37,309 EPOCH 3 done: loss 0.1552 - lr: 0.000039
119
+ 2023-10-20 09:49:38,391 DEV : loss 0.09129446744918823 - f1-score (micro avg) 0.4545
120
+ 2023-10-20 09:49:38,402 saving best model
121
+ 2023-10-20 09:49:38,437 ----------------------------------------------------------------------------------------------------
122
+ 2023-10-20 09:49:40,135 epoch 4 - iter 77/773 - loss 0.12897817 - time (sec): 1.70 - samples/sec: 8025.57 - lr: 0.000038 - momentum: 0.000000
123
+ 2023-10-20 09:49:41,847 epoch 4 - iter 154/773 - loss 0.14162760 - time (sec): 3.41 - samples/sec: 7389.73 - lr: 0.000038 - momentum: 0.000000
124
+ 2023-10-20 09:49:43,541 epoch 4 - iter 231/773 - loss 0.13190156 - time (sec): 5.10 - samples/sec: 7334.53 - lr: 0.000037 - momentum: 0.000000
125
+ 2023-10-20 09:49:45,309 epoch 4 - iter 308/773 - loss 0.13315426 - time (sec): 6.87 - samples/sec: 7213.08 - lr: 0.000037 - momentum: 0.000000
126
+ 2023-10-20 09:49:47,080 epoch 4 - iter 385/773 - loss 0.13213692 - time (sec): 8.64 - samples/sec: 7243.47 - lr: 0.000036 - momentum: 0.000000
127
+ 2023-10-20 09:49:48,813 epoch 4 - iter 462/773 - loss 0.13554983 - time (sec): 10.38 - samples/sec: 7207.39 - lr: 0.000036 - momentum: 0.000000
128
+ 2023-10-20 09:49:50,578 epoch 4 - iter 539/773 - loss 0.13667052 - time (sec): 12.14 - samples/sec: 7200.13 - lr: 0.000035 - momentum: 0.000000
129
+ 2023-10-20 09:49:52,341 epoch 4 - iter 616/773 - loss 0.13714767 - time (sec): 13.90 - samples/sec: 7204.78 - lr: 0.000034 - momentum: 0.000000
130
+ 2023-10-20 09:49:54,053 epoch 4 - iter 693/773 - loss 0.13504563 - time (sec): 15.62 - samples/sec: 7204.58 - lr: 0.000034 - momentum: 0.000000
131
+ 2023-10-20 09:49:55,783 epoch 4 - iter 770/773 - loss 0.13630022 - time (sec): 17.35 - samples/sec: 7139.71 - lr: 0.000033 - momentum: 0.000000
132
+ 2023-10-20 09:49:55,846 ----------------------------------------------------------------------------------------------------
133
+ 2023-10-20 09:49:55,847 EPOCH 4 done: loss 0.1362 - lr: 0.000033
134
+ 2023-10-20 09:49:56,915 DEV : loss 0.08750458806753159 - f1-score (micro avg) 0.5182
135
+ 2023-10-20 09:49:56,927 saving best model
136
+ 2023-10-20 09:49:56,969 ----------------------------------------------------------------------------------------------------
137
+ 2023-10-20 09:49:58,649 epoch 5 - iter 77/773 - loss 0.11837622 - time (sec): 1.68 - samples/sec: 7426.52 - lr: 0.000033 - momentum: 0.000000
138
+ 2023-10-20 09:50:00,345 epoch 5 - iter 154/773 - loss 0.11950342 - time (sec): 3.38 - samples/sec: 7326.73 - lr: 0.000032 - momentum: 0.000000
139
+ 2023-10-20 09:50:02,083 epoch 5 - iter 231/773 - loss 0.12281718 - time (sec): 5.11 - samples/sec: 7176.13 - lr: 0.000032 - momentum: 0.000000
140
+ 2023-10-20 09:50:03,759 epoch 5 - iter 308/773 - loss 0.12163778 - time (sec): 6.79 - samples/sec: 7307.34 - lr: 0.000031 - momentum: 0.000000
141
+ 2023-10-20 09:50:05,449 epoch 5 - iter 385/773 - loss 0.11961392 - time (sec): 8.48 - samples/sec: 7367.32 - lr: 0.000031 - momentum: 0.000000
142
+ 2023-10-20 09:50:07,225 epoch 5 - iter 462/773 - loss 0.12444845 - time (sec): 10.26 - samples/sec: 7314.02 - lr: 0.000030 - momentum: 0.000000
143
+ 2023-10-20 09:50:08,940 epoch 5 - iter 539/773 - loss 0.12636329 - time (sec): 11.97 - samples/sec: 7281.68 - lr: 0.000029 - momentum: 0.000000
144
+ 2023-10-20 09:50:10,627 epoch 5 - iter 616/773 - loss 0.12908626 - time (sec): 13.66 - samples/sec: 7262.66 - lr: 0.000029 - momentum: 0.000000
145
+ 2023-10-20 09:50:12,304 epoch 5 - iter 693/773 - loss 0.12873301 - time (sec): 15.33 - samples/sec: 7273.24 - lr: 0.000028 - momentum: 0.000000
146
+ 2023-10-20 09:50:14,033 epoch 5 - iter 770/773 - loss 0.12550489 - time (sec): 17.06 - samples/sec: 7257.85 - lr: 0.000028 - momentum: 0.000000
147
+ 2023-10-20 09:50:14,093 ----------------------------------------------------------------------------------------------------
148
+ 2023-10-20 09:50:14,093 EPOCH 5 done: loss 0.1256 - lr: 0.000028
149
+ 2023-10-20 09:50:15,180 DEV : loss 0.08463426679372787 - f1-score (micro avg) 0.5385
150
+ 2023-10-20 09:50:15,192 saving best model
151
+ 2023-10-20 09:50:15,232 ----------------------------------------------------------------------------------------------------
152
+ 2023-10-20 09:50:16,956 epoch 6 - iter 77/773 - loss 0.15202072 - time (sec): 1.72 - samples/sec: 7215.52 - lr: 0.000027 - momentum: 0.000000
153
+ 2023-10-20 09:50:18,717 epoch 6 - iter 154/773 - loss 0.12842673 - time (sec): 3.48 - samples/sec: 7212.32 - lr: 0.000027 - momentum: 0.000000
154
+ 2023-10-20 09:50:20,453 epoch 6 - iter 231/773 - loss 0.12246865 - time (sec): 5.22 - samples/sec: 7160.36 - lr: 0.000026 - momentum: 0.000000
155
+ 2023-10-20 09:50:22,137 epoch 6 - iter 308/773 - loss 0.12273321 - time (sec): 6.90 - samples/sec: 7074.89 - lr: 0.000026 - momentum: 0.000000
156
+ 2023-10-20 09:50:23,834 epoch 6 - iter 385/773 - loss 0.12023376 - time (sec): 8.60 - samples/sec: 7223.55 - lr: 0.000025 - momentum: 0.000000
157
+ 2023-10-20 09:50:25,485 epoch 6 - iter 462/773 - loss 0.11889169 - time (sec): 10.25 - samples/sec: 7235.65 - lr: 0.000024 - momentum: 0.000000
158
+ 2023-10-20 09:50:27,185 epoch 6 - iter 539/773 - loss 0.11869253 - time (sec): 11.95 - samples/sec: 7243.32 - lr: 0.000024 - momentum: 0.000000
159
+ 2023-10-20 09:50:28,935 epoch 6 - iter 616/773 - loss 0.11634542 - time (sec): 13.70 - samples/sec: 7221.14 - lr: 0.000023 - momentum: 0.000000
160
+ 2023-10-20 09:50:30,696 epoch 6 - iter 693/773 - loss 0.11714691 - time (sec): 15.46 - samples/sec: 7245.60 - lr: 0.000023 - momentum: 0.000000
161
+ 2023-10-20 09:50:32,382 epoch 6 - iter 770/773 - loss 0.11633289 - time (sec): 17.15 - samples/sec: 7219.80 - lr: 0.000022 - momentum: 0.000000
162
+ 2023-10-20 09:50:32,450 ----------------------------------------------------------------------------------------------------
163
+ 2023-10-20 09:50:32,451 EPOCH 6 done: loss 0.1160 - lr: 0.000022
164
+ 2023-10-20 09:50:33,540 DEV : loss 0.08559702336788177 - f1-score (micro avg) 0.5348
165
+ 2023-10-20 09:50:33,551 ----------------------------------------------------------------------------------------------------
166
+ 2023-10-20 09:50:35,193 epoch 7 - iter 77/773 - loss 0.13418942 - time (sec): 1.64 - samples/sec: 7322.35 - lr: 0.000022 - momentum: 0.000000
167
+ 2023-10-20 09:50:36,886 epoch 7 - iter 154/773 - loss 0.11605993 - time (sec): 3.33 - samples/sec: 7552.83 - lr: 0.000021 - momentum: 0.000000
168
+ 2023-10-20 09:50:38,581 epoch 7 - iter 231/773 - loss 0.11974916 - time (sec): 5.03 - samples/sec: 7346.15 - lr: 0.000021 - momentum: 0.000000
169
+ 2023-10-20 09:50:40,361 epoch 7 - iter 308/773 - loss 0.11273530 - time (sec): 6.81 - samples/sec: 7309.01 - lr: 0.000020 - momentum: 0.000000
170
+ 2023-10-20 09:50:42,109 epoch 7 - iter 385/773 - loss 0.11096649 - time (sec): 8.56 - samples/sec: 7100.74 - lr: 0.000019 - momentum: 0.000000
171
+ 2023-10-20 09:50:43,841 epoch 7 - iter 462/773 - loss 0.11189813 - time (sec): 10.29 - samples/sec: 7072.68 - lr: 0.000019 - momentum: 0.000000
172
+ 2023-10-20 09:50:45,715 epoch 7 - iter 539/773 - loss 0.11121427 - time (sec): 12.16 - samples/sec: 7043.16 - lr: 0.000018 - momentum: 0.000000
173
+ 2023-10-20 09:50:47,508 epoch 7 - iter 616/773 - loss 0.11017605 - time (sec): 13.96 - samples/sec: 7059.63 - lr: 0.000018 - momentum: 0.000000
174
+ 2023-10-20 09:50:49,317 epoch 7 - iter 693/773 - loss 0.10910748 - time (sec): 15.76 - samples/sec: 7072.09 - lr: 0.000017 - momentum: 0.000000
175
+ 2023-10-20 09:50:51,038 epoch 7 - iter 770/773 - loss 0.11083517 - time (sec): 17.49 - samples/sec: 7079.64 - lr: 0.000017 - momentum: 0.000000
176
+ 2023-10-20 09:50:51,101 ----------------------------------------------------------------------------------------------------
177
+ 2023-10-20 09:50:51,101 EPOCH 7 done: loss 0.1107 - lr: 0.000017
178
+ 2023-10-20 09:50:52,197 DEV : loss 0.08398106694221497 - f1-score (micro avg) 0.5572
179
+ 2023-10-20 09:50:52,209 saving best model
180
+ 2023-10-20 09:50:52,248 ----------------------------------------------------------------------------------------------------
181
+ 2023-10-20 09:50:53,959 epoch 8 - iter 77/773 - loss 0.07735418 - time (sec): 1.71 - samples/sec: 7032.51 - lr: 0.000016 - momentum: 0.000000
182
+ 2023-10-20 09:50:55,671 epoch 8 - iter 154/773 - loss 0.09799317 - time (sec): 3.42 - samples/sec: 6976.98 - lr: 0.000016 - momentum: 0.000000
183
+ 2023-10-20 09:50:57,435 epoch 8 - iter 231/773 - loss 0.10334869 - time (sec): 5.19 - samples/sec: 7043.50 - lr: 0.000015 - momentum: 0.000000
184
+ 2023-10-20 09:50:59,177 epoch 8 - iter 308/773 - loss 0.09645731 - time (sec): 6.93 - samples/sec: 7215.93 - lr: 0.000014 - momentum: 0.000000
185
+ 2023-10-20 09:51:00,995 epoch 8 - iter 385/773 - loss 0.09864950 - time (sec): 8.75 - samples/sec: 7181.08 - lr: 0.000014 - momentum: 0.000000
186
+ 2023-10-20 09:51:02,709 epoch 8 - iter 462/773 - loss 0.10127673 - time (sec): 10.46 - samples/sec: 7190.72 - lr: 0.000013 - momentum: 0.000000
187
+ 2023-10-20 09:51:04,451 epoch 8 - iter 539/773 - loss 0.10595667 - time (sec): 12.20 - samples/sec: 7126.05 - lr: 0.000013 - momentum: 0.000000
188
+ 2023-10-20 09:51:06,293 epoch 8 - iter 616/773 - loss 0.10535196 - time (sec): 14.05 - samples/sec: 7091.97 - lr: 0.000012 - momentum: 0.000000
189
+ 2023-10-20 09:51:08,042 epoch 8 - iter 693/773 - loss 0.10603763 - time (sec): 15.79 - samples/sec: 7039.09 - lr: 0.000012 - momentum: 0.000000
190
+ 2023-10-20 09:51:09,809 epoch 8 - iter 770/773 - loss 0.10450961 - time (sec): 17.56 - samples/sec: 7048.84 - lr: 0.000011 - momentum: 0.000000
191
+ 2023-10-20 09:51:09,879 ----------------------------------------------------------------------------------------------------
192
+ 2023-10-20 09:51:09,880 EPOCH 8 done: loss 0.1044 - lr: 0.000011
193
+ 2023-10-20 09:51:10,964 DEV : loss 0.08721306174993515 - f1-score (micro avg) 0.557
194
+ 2023-10-20 09:51:10,977 ----------------------------------------------------------------------------------------------------
195
+ 2023-10-20 09:51:12,693 epoch 9 - iter 77/773 - loss 0.08508022 - time (sec): 1.72 - samples/sec: 6831.62 - lr: 0.000011 - momentum: 0.000000
196
+ 2023-10-20 09:51:14,449 epoch 9 - iter 154/773 - loss 0.10086226 - time (sec): 3.47 - samples/sec: 7068.70 - lr: 0.000010 - momentum: 0.000000
197
+ 2023-10-20 09:51:16,184 epoch 9 - iter 231/773 - loss 0.10411885 - time (sec): 5.21 - samples/sec: 7041.88 - lr: 0.000009 - momentum: 0.000000
198
+ 2023-10-20 09:51:17,985 epoch 9 - iter 308/773 - loss 0.10360436 - time (sec): 7.01 - samples/sec: 7036.34 - lr: 0.000009 - momentum: 0.000000
199
+ 2023-10-20 09:51:19,766 epoch 9 - iter 385/773 - loss 0.10607327 - time (sec): 8.79 - samples/sec: 7123.12 - lr: 0.000008 - momentum: 0.000000
200
+ 2023-10-20 09:51:21,608 epoch 9 - iter 462/773 - loss 0.10462746 - time (sec): 10.63 - samples/sec: 6995.10 - lr: 0.000008 - momentum: 0.000000
201
+ 2023-10-20 09:51:23,423 epoch 9 - iter 539/773 - loss 0.10355217 - time (sec): 12.45 - samples/sec: 6945.51 - lr: 0.000007 - momentum: 0.000000
202
+ 2023-10-20 09:51:25,251 epoch 9 - iter 616/773 - loss 0.10298710 - time (sec): 14.27 - samples/sec: 6947.34 - lr: 0.000007 - momentum: 0.000000
203
+ 2023-10-20 09:51:26,965 epoch 9 - iter 693/773 - loss 0.10206767 - time (sec): 15.99 - samples/sec: 6974.89 - lr: 0.000006 - momentum: 0.000000
204
+ 2023-10-20 09:51:28,689 epoch 9 - iter 770/773 - loss 0.10193084 - time (sec): 17.71 - samples/sec: 6991.53 - lr: 0.000006 - momentum: 0.000000
205
+ 2023-10-20 09:51:28,754 ----------------------------------------------------------------------------------------------------
206
+ 2023-10-20 09:51:28,755 EPOCH 9 done: loss 0.1020 - lr: 0.000006
207
+ 2023-10-20 09:51:29,845 DEV : loss 0.08746206760406494 - f1-score (micro avg) 0.5598
208
+ 2023-10-20 09:51:29,857 saving best model
209
+ 2023-10-20 09:51:29,898 ----------------------------------------------------------------------------------------------------
210
+ 2023-10-20 09:51:31,632 epoch 10 - iter 77/773 - loss 0.09052757 - time (sec): 1.73 - samples/sec: 7036.26 - lr: 0.000005 - momentum: 0.000000
211
+ 2023-10-20 09:51:33,295 epoch 10 - iter 154/773 - loss 0.08568932 - time (sec): 3.40 - samples/sec: 6743.47 - lr: 0.000005 - momentum: 0.000000
212
+ 2023-10-20 09:51:35,057 epoch 10 - iter 231/773 - loss 0.09090122 - time (sec): 5.16 - samples/sec: 7114.36 - lr: 0.000004 - momentum: 0.000000
213
+ 2023-10-20 09:51:36,834 epoch 10 - iter 308/773 - loss 0.09401948 - time (sec): 6.94 - samples/sec: 7126.06 - lr: 0.000003 - momentum: 0.000000
214
+ 2023-10-20 09:51:38,537 epoch 10 - iter 385/773 - loss 0.09521242 - time (sec): 8.64 - samples/sec: 7168.24 - lr: 0.000003 - momentum: 0.000000
215
+ 2023-10-20 09:51:40,283 epoch 10 - iter 462/773 - loss 0.09812942 - time (sec): 10.38 - samples/sec: 7175.35 - lr: 0.000002 - momentum: 0.000000
216
+ 2023-10-20 09:51:42,013 epoch 10 - iter 539/773 - loss 0.09933101 - time (sec): 12.11 - samples/sec: 7131.17 - lr: 0.000002 - momentum: 0.000000
217
+ 2023-10-20 09:51:43,770 epoch 10 - iter 616/773 - loss 0.09792858 - time (sec): 13.87 - samples/sec: 7186.32 - lr: 0.000001 - momentum: 0.000000
218
+ 2023-10-20 09:51:45,469 epoch 10 - iter 693/773 - loss 0.09892283 - time (sec): 15.57 - samples/sec: 7150.93 - lr: 0.000001 - momentum: 0.000000
219
+ 2023-10-20 09:51:47,214 epoch 10 - iter 770/773 - loss 0.10051481 - time (sec): 17.32 - samples/sec: 7154.09 - lr: 0.000000 - momentum: 0.000000
220
+ 2023-10-20 09:51:47,278 ----------------------------------------------------------------------------------------------------
221
+ 2023-10-20 09:51:47,278 EPOCH 10 done: loss 0.1004 - lr: 0.000000
222
+ 2023-10-20 09:51:48,362 DEV : loss 0.08812259882688522 - f1-score (micro avg) 0.565
223
+ 2023-10-20 09:51:48,374 saving best model
224
+ 2023-10-20 09:51:48,448 ----------------------------------------------------------------------------------------------------
225
+ 2023-10-20 09:51:48,449 Loading model from best epoch ...
226
+ 2023-10-20 09:51:48,522 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
227
+ 2023-10-20 09:51:51,447
228
+ Results:
229
+ - F-score (micro) 0.5627
230
+ - F-score (macro) 0.3215
231
+ - Accuracy 0.4019
232
+
233
+ By class:
234
+ precision recall f1-score support
235
+
236
+ LOC 0.6176 0.6438 0.6304 946
237
+ BUILDING 0.2535 0.0973 0.1406 185
238
+ STREET 1.0000 0.1071 0.1935 56
239
+
240
+ micro avg 0.5955 0.5333 0.5627 1187
241
+ macro avg 0.6237 0.2827 0.3215 1187
242
+ weighted avg 0.5789 0.5333 0.5335 1187
243
+
244
+ 2023-10-20 09:51:51,447 ----------------------------------------------------------------------------------------------------