Hugging Face's logo

stefan-it
/
hmbench-icdar-nl-hmteams-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2

Token Classification
Flair
PyTorch
TensorBoard
Dutch
sequence-tagger-model
Model card Files Metrics Community
stefan-it commited on
Commit
9a1dea6
•
1 Parent(s): c7d3992

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.1697559529.bce904bcef33.2251.5 +3 -0
  5. test.tsv +0 -0
  6. training.log +240 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9c65c9d0585e4ef0cd4aa7d8bcbbb2f6b0a72b1d49e9c6947d44ce8e511d437f
3
+ size 440941957
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 16:20:02 0.0000 0.3666 0.0854 0.8399 0.8130 0.8262 0.7148
3
+ 2 16:21:19 0.0000 0.0963 0.0902 0.8866 0.7913 0.8362 0.7288
4
+ 3 16:22:34 0.0000 0.0723 0.0939 0.8844 0.8223 0.8522 0.7545
5
+ 4 16:23:49 0.0000 0.0541 0.1042 0.8690 0.8430 0.8558 0.7619
6
+ 5 16:25:04 0.0000 0.0434 0.1496 0.8838 0.7231 0.7955 0.6699
7
+ 6 16:26:17 0.0000 0.0289 0.1501 0.8960 0.7562 0.8202 0.7086
8
+ 7 16:27:31 0.0000 0.0204 0.1493 0.8920 0.8275 0.8585 0.7636
9
+ 8 16:28:46 0.0000 0.0134 0.1695 0.8944 0.8140 0.8524 0.7562
10
+ 9 16:30:04 0.0000 0.0096 0.1593 0.8959 0.8357 0.8648 0.7712
11
+ 10 16:31:25 0.0000 0.0061 0.1579 0.8966 0.8419 0.8684 0.7784
runs/events.out.tfevents.1697559529.bce904bcef33.2251.5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:532b837be18a0fcfb810252f6aa6cf7ebdc5b56ccab46472abafd5d182e6ddbf
3
+ size 808480
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,240 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-17 16:18:49,833 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-17 16:18:49,834 Model: "SequenceTagger(
3
+ (embeddings): TransformerWordEmbeddings(
4
+ (model): ElectraModel(
5
+ (embeddings): ElectraEmbeddings(
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): ElectraEncoder(
13
+ (layer): ModuleList(
14
+ (0-11): 12 x ElectraLayer(
15
+ (attention): ElectraAttention(
16
+ (self): ElectraSelfAttention(
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): ElectraSelfOutput(
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): ElectraIntermediate(
29
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
30
+ (intermediate_act_fn): GELUActivation()
31
+ )
32
+ (output): ElectraOutput(
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
+ )
41
+ )
42
+ (locked_dropout): LockedDropout(p=0.5)
43
+ (linear): Linear(in_features=768, out_features=13, bias=True)
44
+ (loss_function): CrossEntropyLoss()
45
+ )"
46
+ 2023-10-17 16:18:49,834 ----------------------------------------------------------------------------------------------------
47
+ 2023-10-17 16:18:49,834 MultiCorpus: 5777 train + 722 dev + 723 test sentences
48
+ - NER_ICDAR_EUROPEANA Corpus: 5777 train + 722 dev + 723 test sentences - /root/.flair/datasets/ner_icdar_europeana/nl
49
+ 2023-10-17 16:18:49,835 ----------------------------------------------------------------------------------------------------
50
+ 2023-10-17 16:18:49,835 Train: 5777 sentences
51
+ 2023-10-17 16:18:49,835 (train_with_dev=False, train_with_test=False)
52
+ 2023-10-17 16:18:49,835 ----------------------------------------------------------------------------------------------------
53
+ 2023-10-17 16:18:49,835 Training Params:
54
+ 2023-10-17 16:18:49,835 - learning_rate: "5e-05"
55
+ 2023-10-17 16:18:49,835 - mini_batch_size: "4"
56
+ 2023-10-17 16:18:49,835 - max_epochs: "10"
57
+ 2023-10-17 16:18:49,835 - shuffle: "True"
58
+ 2023-10-17 16:18:49,835 ----------------------------------------------------------------------------------------------------
59
+ 2023-10-17 16:18:49,835 Plugins:
60
+ 2023-10-17 16:18:49,835 - TensorboardLogger
61
+ 2023-10-17 16:18:49,835 - LinearScheduler | warmup_fraction: '0.1'
62
+ 2023-10-17 16:18:49,835 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-17 16:18:49,835 Final evaluation on model from best epoch (best-model.pt)
64
+ 2023-10-17 16:18:49,835 - metric: "('micro avg', 'f1-score')"
65
+ 2023-10-17 16:18:49,835 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-17 16:18:49,835 Computation:
67
+ 2023-10-17 16:18:49,835 - compute on device: cuda:0
68
+ 2023-10-17 16:18:49,835 - embedding storage: none
69
+ 2023-10-17 16:18:49,835 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-17 16:18:49,835 Model training base path: "hmbench-icdar/nl-hmteams/teams-base-historic-multilingual-discriminator-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2"
71
+ 2023-10-17 16:18:49,835 ----------------------------------------------------------------------------------------------------
72
+ 2023-10-17 16:18:49,835 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-17 16:18:49,835 Logging anything other than scalars to TensorBoard is currently not supported.
74
+ 2023-10-17 16:18:56,849 epoch 1 - iter 144/1445 - loss 2.13705809 - time (sec): 7.01 - samples/sec: 2602.15 - lr: 0.000005 - momentum: 0.000000
75
+ 2023-10-17 16:19:03,750 epoch 1 - iter 288/1445 - loss 1.21878406 - time (sec): 13.91 - samples/sec: 2562.25 - lr: 0.000010 - momentum: 0.000000
76
+ 2023-10-17 16:19:10,804 epoch 1 - iter 432/1445 - loss 0.90739784 - time (sec): 20.97 - samples/sec: 2498.56 - lr: 0.000015 - momentum: 0.000000
77
+ 2023-10-17 16:19:17,974 epoch 1 - iter 576/1445 - loss 0.72044535 - time (sec): 28.14 - samples/sec: 2490.68 - lr: 0.000020 - momentum: 0.000000
78
+ 2023-10-17 16:19:25,043 epoch 1 - iter 720/1445 - loss 0.60080964 - time (sec): 35.21 - samples/sec: 2512.23 - lr: 0.000025 - momentum: 0.000000
79
+ 2023-10-17 16:19:31,904 epoch 1 - iter 864/1445 - loss 0.52771699 - time (sec): 42.07 - samples/sec: 2513.55 - lr: 0.000030 - momentum: 0.000000
80
+ 2023-10-17 16:19:38,848 epoch 1 - iter 1008/1445 - loss 0.47241003 - time (sec): 49.01 - samples/sec: 2521.27 - lr: 0.000035 - momentum: 0.000000
81
+ 2023-10-17 16:19:45,917 epoch 1 - iter 1152/1445 - loss 0.42720095 - time (sec): 56.08 - samples/sec: 2519.40 - lr: 0.000040 - momentum: 0.000000
82
+ 2023-10-17 16:19:52,870 epoch 1 - iter 1296/1445 - loss 0.39375757 - time (sec): 63.03 - samples/sec: 2509.51 - lr: 0.000045 - momentum: 0.000000
83
+ 2023-10-17 16:19:59,869 epoch 1 - iter 1440/1445 - loss 0.36718281 - time (sec): 70.03 - samples/sec: 2509.84 - lr: 0.000050 - momentum: 0.000000
84
+ 2023-10-17 16:20:00,093 ----------------------------------------------------------------------------------------------------
85
+ 2023-10-17 16:20:00,093 EPOCH 1 done: loss 0.3666 - lr: 0.000050
86
+ 2023-10-17 16:20:02,889 DEV : loss 0.0853525772690773 - f1-score (micro avg) 0.8262
87
+ 2023-10-17 16:20:02,908 saving best model
88
+ 2023-10-17 16:20:03,241 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-17 16:20:09,898 epoch 2 - iter 144/1445 - loss 0.13163265 - time (sec): 6.65 - samples/sec: 2487.09 - lr: 0.000049 - momentum: 0.000000
90
+ 2023-10-17 16:20:16,952 epoch 2 - iter 288/1445 - loss 0.11859313 - time (sec): 13.71 - samples/sec: 2444.56 - lr: 0.000049 - momentum: 0.000000
91
+ 2023-10-17 16:20:24,250 epoch 2 - iter 432/1445 - loss 0.10993065 - time (sec): 21.01 - samples/sec: 2427.99 - lr: 0.000048 - momentum: 0.000000
92
+ 2023-10-17 16:20:31,667 epoch 2 - iter 576/1445 - loss 0.10512844 - time (sec): 28.42 - samples/sec: 2425.61 - lr: 0.000048 - momentum: 0.000000
93
+ 2023-10-17 16:20:38,916 epoch 2 - iter 720/1445 - loss 0.10132998 - time (sec): 35.67 - samples/sec: 2409.57 - lr: 0.000047 - momentum: 0.000000
94
+ 2023-10-17 16:20:46,181 epoch 2 - iter 864/1445 - loss 0.09827432 - time (sec): 42.94 - samples/sec: 2449.22 - lr: 0.000047 - momentum: 0.000000
95
+ 2023-10-17 16:20:53,243 epoch 2 - iter 1008/1445 - loss 0.09734789 - time (sec): 50.00 - samples/sec: 2461.18 - lr: 0.000046 - momentum: 0.000000
96
+ 2023-10-17 16:21:00,233 epoch 2 - iter 1152/1445 - loss 0.09616555 - time (sec): 56.99 - samples/sec: 2456.82 - lr: 0.000046 - momentum: 0.000000
97
+ 2023-10-17 16:21:07,398 epoch 2 - iter 1296/1445 - loss 0.09677640 - time (sec): 64.15 - samples/sec: 2465.71 - lr: 0.000045 - momentum: 0.000000
98
+ 2023-10-17 16:21:14,472 epoch 2 - iter 1440/1445 - loss 0.09633877 - time (sec): 71.23 - samples/sec: 2467.41 - lr: 0.000044 - momentum: 0.000000
99
+ 2023-10-17 16:21:14,703 ----------------------------------------------------------------------------------------------------
100
+ 2023-10-17 16:21:14,703 EPOCH 2 done: loss 0.0963 - lr: 0.000044
101
+ 2023-10-17 16:21:19,008 DEV : loss 0.09023821353912354 - f1-score (micro avg) 0.8362
102
+ 2023-10-17 16:21:19,040 saving best model
103
+ 2023-10-17 16:21:19,492 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-17 16:21:26,867 epoch 3 - iter 144/1445 - loss 0.07639569 - time (sec): 7.37 - samples/sec: 2469.85 - lr: 0.000044 - momentum: 0.000000
105
+ 2023-10-17 16:21:33,792 epoch 3 - iter 288/1445 - loss 0.07289280 - time (sec): 14.30 - samples/sec: 2474.87 - lr: 0.000043 - momentum: 0.000000
106
+ 2023-10-17 16:21:41,036 epoch 3 - iter 432/1445 - loss 0.07369285 - time (sec): 21.54 - samples/sec: 2526.94 - lr: 0.000043 - momentum: 0.000000
107
+ 2023-10-17 16:21:47,971 epoch 3 - iter 576/1445 - loss 0.07352607 - time (sec): 28.48 - samples/sec: 2503.52 - lr: 0.000042 - momentum: 0.000000
108
+ 2023-10-17 16:21:55,171 epoch 3 - iter 720/1445 - loss 0.07225890 - time (sec): 35.68 - samples/sec: 2486.81 - lr: 0.000042 - momentum: 0.000000
109
+ 2023-10-17 16:22:02,381 epoch 3 - iter 864/1445 - loss 0.06938398 - time (sec): 42.88 - samples/sec: 2494.71 - lr: 0.000041 - momentum: 0.000000
110
+ 2023-10-17 16:22:09,345 epoch 3 - iter 1008/1445 - loss 0.07054082 - time (sec): 49.85 - samples/sec: 2475.72 - lr: 0.000041 - momentum: 0.000000
111
+ 2023-10-17 16:22:16,271 epoch 3 - iter 1152/1445 - loss 0.07020152 - time (sec): 56.78 - samples/sec: 2463.08 - lr: 0.000040 - momentum: 0.000000
112
+ 2023-10-17 16:22:23,487 epoch 3 - iter 1296/1445 - loss 0.07093905 - time (sec): 63.99 - samples/sec: 2465.29 - lr: 0.000039 - momentum: 0.000000
113
+ 2023-10-17 16:22:30,787 epoch 3 - iter 1440/1445 - loss 0.07220694 - time (sec): 71.29 - samples/sec: 2462.25 - lr: 0.000039 - momentum: 0.000000
114
+ 2023-10-17 16:22:31,037 ----------------------------------------------------------------------------------------------------
115
+ 2023-10-17 16:22:31,038 EPOCH 3 done: loss 0.0723 - lr: 0.000039
116
+ 2023-10-17 16:22:34,317 DEV : loss 0.09389135241508484 - f1-score (micro avg) 0.8522
117
+ 2023-10-17 16:22:34,334 saving best model
118
+ 2023-10-17 16:22:34,794 ----------------------------------------------------------------------------------------------------
119
+ 2023-10-17 16:22:41,916 epoch 4 - iter 144/1445 - loss 0.03934107 - time (sec): 7.12 - samples/sec: 2501.87 - lr: 0.000038 - momentum: 0.000000
120
+ 2023-10-17 16:22:48,884 epoch 4 - iter 288/1445 - loss 0.04960378 - time (sec): 14.09 - samples/sec: 2474.30 - lr: 0.000038 - momentum: 0.000000
121
+ 2023-10-17 16:22:55,742 epoch 4 - iter 432/1445 - loss 0.04967966 - time (sec): 20.95 - samples/sec: 2464.10 - lr: 0.000037 - momentum: 0.000000
122
+ 2023-10-17 16:23:02,583 epoch 4 - iter 576/1445 - loss 0.05239601 - time (sec): 27.79 - samples/sec: 2472.89 - lr: 0.000037 - momentum: 0.000000
123
+ 2023-10-17 16:23:10,050 epoch 4 - iter 720/1445 - loss 0.05481533 - time (sec): 35.25 - samples/sec: 2459.94 - lr: 0.000036 - momentum: 0.000000
124
+ 2023-10-17 16:23:17,233 epoch 4 - iter 864/1445 - loss 0.05681795 - time (sec): 42.44 - samples/sec: 2473.57 - lr: 0.000036 - momentum: 0.000000
125
+ 2023-10-17 16:23:24,313 epoch 4 - iter 1008/1445 - loss 0.05672883 - time (sec): 49.52 - samples/sec: 2478.78 - lr: 0.000035 - momentum: 0.000000
126
+ 2023-10-17 16:23:31,477 epoch 4 - iter 1152/1445 - loss 0.05481502 - time (sec): 56.68 - samples/sec: 2477.63 - lr: 0.000034 - momentum: 0.000000
127
+ 2023-10-17 16:23:38,544 epoch 4 - iter 1296/1445 - loss 0.05437627 - time (sec): 63.75 - samples/sec: 2479.36 - lr: 0.000034 - momentum: 0.000000
128
+ 2023-10-17 16:23:45,784 epoch 4 - iter 1440/1445 - loss 0.05419519 - time (sec): 70.99 - samples/sec: 2475.65 - lr: 0.000033 - momentum: 0.000000
129
+ 2023-10-17 16:23:46,031 ----------------------------------------------------------------------------------------------------
130
+ 2023-10-17 16:23:46,032 EPOCH 4 done: loss 0.0541 - lr: 0.000033
131
+ 2023-10-17 16:23:49,286 DEV : loss 0.10415765643119812 - f1-score (micro avg) 0.8558
132
+ 2023-10-17 16:23:49,302 saving best model
133
+ 2023-10-17 16:23:49,759 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-17 16:23:56,689 epoch 5 - iter 144/1445 - loss 0.05670271 - time (sec): 6.93 - samples/sec: 2374.53 - lr: 0.000033 - momentum: 0.000000
135
+ 2023-10-17 16:24:03,688 epoch 5 - iter 288/1445 - loss 0.04436221 - time (sec): 13.93 - samples/sec: 2443.03 - lr: 0.000032 - momentum: 0.000000
136
+ 2023-10-17 16:24:10,773 epoch 5 - iter 432/1445 - loss 0.04616850 - time (sec): 21.01 - samples/sec: 2458.36 - lr: 0.000032 - momentum: 0.000000
137
+ 2023-10-17 16:24:17,711 epoch 5 - iter 576/1445 - loss 0.04352124 - time (sec): 27.95 - samples/sec: 2426.81 - lr: 0.000031 - momentum: 0.000000
138
+ 2023-10-17 16:24:25,343 epoch 5 - iter 720/1445 - loss 0.04411388 - time (sec): 35.58 - samples/sec: 2418.81 - lr: 0.000031 - momentum: 0.000000
139
+ 2023-10-17 16:24:32,451 epoch 5 - iter 864/1445 - loss 0.04402856 - time (sec): 42.69 - samples/sec: 2426.02 - lr: 0.000030 - momentum: 0.000000
140
+ 2023-10-17 16:24:39,891 epoch 5 - iter 1008/1445 - loss 0.04635845 - time (sec): 50.13 - samples/sec: 2442.79 - lr: 0.000029 - momentum: 0.000000
141
+ 2023-10-17 16:24:47,269 epoch 5 - iter 1152/1445 - loss 0.04510679 - time (sec): 57.51 - samples/sec: 2459.29 - lr: 0.000029 - momentum: 0.000000
142
+ 2023-10-17 16:24:54,254 epoch 5 - iter 1296/1445 - loss 0.04387905 - time (sec): 64.49 - samples/sec: 2467.17 - lr: 0.000028 - momentum: 0.000000
143
+ 2023-10-17 16:25:00,944 epoch 5 - iter 1440/1445 - loss 0.04340737 - time (sec): 71.18 - samples/sec: 2466.49 - lr: 0.000028 - momentum: 0.000000
144
+ 2023-10-17 16:25:01,206 ----------------------------------------------------------------------------------------------------
145
+ 2023-10-17 16:25:01,206 EPOCH 5 done: loss 0.0434 - lr: 0.000028
146
+ 2023-10-17 16:25:04,542 DEV : loss 0.14961402118206024 - f1-score (micro avg) 0.7955
147
+ 2023-10-17 16:25:04,563 ----------------------------------------------------------------------------------------------------
148
+ 2023-10-17 16:25:11,913 epoch 6 - iter 144/1445 - loss 0.05786667 - time (sec): 7.35 - samples/sec: 2503.05 - lr: 0.000027 - momentum: 0.000000
149
+ 2023-10-17 16:25:18,998 epoch 6 - iter 288/1445 - loss 0.03941575 - time (sec): 14.43 - samples/sec: 2463.13 - lr: 0.000027 - momentum: 0.000000
150
+ 2023-10-17 16:25:25,910 epoch 6 - iter 432/1445 - loss 0.03463529 - time (sec): 21.35 - samples/sec: 2477.92 - lr: 0.000026 - momentum: 0.000000
151
+ 2023-10-17 16:25:32,830 epoch 6 - iter 576/1445 - loss 0.03145162 - time (sec): 28.27 - samples/sec: 2487.21 - lr: 0.000026 - momentum: 0.000000
152
+ 2023-10-17 16:25:39,923 epoch 6 - iter 720/1445 - loss 0.02986732 - time (sec): 35.36 - samples/sec: 2501.74 - lr: 0.000025 - momentum: 0.000000
153
+ 2023-10-17 16:25:46,695 epoch 6 - iter 864/1445 - loss 0.02915652 - time (sec): 42.13 - samples/sec: 2533.84 - lr: 0.000024 - momentum: 0.000000
154
+ 2023-10-17 16:25:53,428 epoch 6 - iter 1008/1445 - loss 0.02815790 - time (sec): 48.86 - samples/sec: 2551.88 - lr: 0.000024 - momentum: 0.000000
155
+ 2023-10-17 16:26:00,418 epoch 6 - iter 1152/1445 - loss 0.02863677 - time (sec): 55.85 - samples/sec: 2527.10 - lr: 0.000023 - momentum: 0.000000
156
+ 2023-10-17 16:26:07,287 epoch 6 - iter 1296/1445 - loss 0.02857277 - time (sec): 62.72 - samples/sec: 2519.48 - lr: 0.000023 - momentum: 0.000000
157
+ 2023-10-17 16:26:14,266 epoch 6 - iter 1440/1445 - loss 0.02901616 - time (sec): 69.70 - samples/sec: 2518.37 - lr: 0.000022 - momentum: 0.000000
158
+ 2023-10-17 16:26:14,559 ----------------------------------------------------------------------------------------------------
159
+ 2023-10-17 16:26:14,559 EPOCH 6 done: loss 0.0289 - lr: 0.000022
160
+ 2023-10-17 16:26:17,754 DEV : loss 0.15008436143398285 - f1-score (micro avg) 0.8202
161
+ 2023-10-17 16:26:17,770 ----------------------------------------------------------------------------------------------------
162
+ 2023-10-17 16:26:24,700 epoch 7 - iter 144/1445 - loss 0.02000359 - time (sec): 6.93 - samples/sec: 2656.70 - lr: 0.000022 - momentum: 0.000000
163
+ 2023-10-17 16:26:31,753 epoch 7 - iter 288/1445 - loss 0.01569030 - time (sec): 13.98 - samples/sec: 2590.17 - lr: 0.000021 - momentum: 0.000000
164
+ 2023-10-17 16:26:39,001 epoch 7 - iter 432/1445 - loss 0.01782813 - time (sec): 21.23 - samples/sec: 2505.61 - lr: 0.000021 - momentum: 0.000000
165
+ 2023-10-17 16:26:46,180 epoch 7 - iter 576/1445 - loss 0.01964390 - time (sec): 28.41 - samples/sec: 2499.50 - lr: 0.000020 - momentum: 0.000000
166
+ 2023-10-17 16:26:53,381 epoch 7 - iter 720/1445 - loss 0.01835474 - time (sec): 35.61 - samples/sec: 2499.50 - lr: 0.000019 - momentum: 0.000000
167
+ 2023-10-17 16:27:00,040 epoch 7 - iter 864/1445 - loss 0.01866023 - time (sec): 42.27 - samples/sec: 2508.78 - lr: 0.000019 - momentum: 0.000000
168
+ 2023-10-17 16:27:07,184 epoch 7 - iter 1008/1445 - loss 0.01877361 - time (sec): 49.41 - samples/sec: 2486.61 - lr: 0.000018 - momentum: 0.000000
169
+ 2023-10-17 16:27:13,862 epoch 7 - iter 1152/1445 - loss 0.01962775 - time (sec): 56.09 - samples/sec: 2498.38 - lr: 0.000018 - momentum: 0.000000
170
+ 2023-10-17 16:27:20,488 epoch 7 - iter 1296/1445 - loss 0.01955537 - time (sec): 62.72 - samples/sec: 2507.44 - lr: 0.000017 - momentum: 0.000000
171
+ 2023-10-17 16:27:27,403 epoch 7 - iter 1440/1445 - loss 0.02047548 - time (sec): 69.63 - samples/sec: 2522.27 - lr: 0.000017 - momentum: 0.000000
172
+ 2023-10-17 16:27:27,637 ----------------------------------------------------------------------------------------------------
173
+ 2023-10-17 16:27:27,637 EPOCH 7 done: loss 0.0204 - lr: 0.000017
174
+ 2023-10-17 16:27:31,016 DEV : loss 0.14934930205345154 - f1-score (micro avg) 0.8585
175
+ 2023-10-17 16:27:31,032 saving best model
176
+ 2023-10-17 16:27:31,496 ----------------------------------------------------------------------------------------------------
177
+ 2023-10-17 16:27:38,491 epoch 8 - iter 144/1445 - loss 0.01993635 - time (sec): 6.99 - samples/sec: 2446.12 - lr: 0.000016 - momentum: 0.000000
178
+ 2023-10-17 16:27:45,557 epoch 8 - iter 288/1445 - loss 0.01529310 - time (sec): 14.06 - samples/sec: 2434.12 - lr: 0.000016 - momentum: 0.000000
179
+ 2023-10-17 16:27:52,729 epoch 8 - iter 432/1445 - loss 0.01421649 - time (sec): 21.23 - samples/sec: 2451.36 - lr: 0.000015 - momentum: 0.000000
180
+ 2023-10-17 16:27:59,839 epoch 8 - iter 576/1445 - loss 0.01275077 - time (sec): 28.34 - samples/sec: 2445.92 - lr: 0.000014 - momentum: 0.000000
181
+ 2023-10-17 16:28:06,792 epoch 8 - iter 720/1445 - loss 0.01099475 - time (sec): 35.29 - samples/sec: 2421.95 - lr: 0.000014 - momentum: 0.000000
182
+ 2023-10-17 16:28:14,048 epoch 8 - iter 864/1445 - loss 0.01053576 - time (sec): 42.55 - samples/sec: 2442.96 - lr: 0.000013 - momentum: 0.000000
183
+ 2023-10-17 16:28:21,295 epoch 8 - iter 1008/1445 - loss 0.01107931 - time (sec): 49.80 - samples/sec: 2455.05 - lr: 0.000013 - momentum: 0.000000
184
+ 2023-10-17 16:28:28,360 epoch 8 - iter 1152/1445 - loss 0.01295286 - time (sec): 56.86 - samples/sec: 2453.26 - lr: 0.000012 - momentum: 0.000000
185
+ 2023-10-17 16:28:35,774 epoch 8 - iter 1296/1445 - loss 0.01309587 - time (sec): 64.28 - samples/sec: 2455.43 - lr: 0.000012 - momentum: 0.000000
186
+ 2023-10-17 16:28:43,046 epoch 8 - iter 1440/1445 - loss 0.01319627 - time (sec): 71.55 - samples/sec: 2454.86 - lr: 0.000011 - momentum: 0.000000
187
+ 2023-10-17 16:28:43,274 ----------------------------------------------------------------------------------------------------
188
+ 2023-10-17 16:28:43,274 EPOCH 8 done: loss 0.0134 - lr: 0.000011
189
+ 2023-10-17 16:28:46,569 DEV : loss 0.16947199404239655 - f1-score (micro avg) 0.8524
190
+ 2023-10-17 16:28:46,602 ----------------------------------------------------------------------------------------------------
191
+ 2023-10-17 16:28:53,623 epoch 9 - iter 144/1445 - loss 0.00899269 - time (sec): 7.02 - samples/sec: 2577.61 - lr: 0.000011 - momentum: 0.000000
192
+ 2023-10-17 16:29:01,060 epoch 9 - iter 288/1445 - loss 0.01406469 - time (sec): 14.46 - samples/sec: 2548.89 - lr: 0.000010 - momentum: 0.000000
193
+ 2023-10-17 16:29:08,313 epoch 9 - iter 432/1445 - loss 0.01182979 - time (sec): 21.71 - samples/sec: 2487.70 - lr: 0.000009 - momentum: 0.000000
194
+ 2023-10-17 16:29:15,658 epoch 9 - iter 576/1445 - loss 0.01105186 - time (sec): 29.06 - samples/sec: 2453.24 - lr: 0.000009 - momentum: 0.000000
195
+ 2023-10-17 16:29:22,816 epoch 9 - iter 720/1445 - loss 0.01186549 - time (sec): 36.21 - samples/sec: 2470.18 - lr: 0.000008 - momentum: 0.000000
196
+ 2023-10-17 16:29:30,151 epoch 9 - iter 864/1445 - loss 0.01072237 - time (sec): 43.55 - samples/sec: 2460.49 - lr: 0.000008 - momentum: 0.000000
197
+ 2023-10-17 16:29:37,289 epoch 9 - iter 1008/1445 - loss 0.01098163 - time (sec): 50.69 - samples/sec: 2446.11 - lr: 0.000007 - momentum: 0.000000
198
+ 2023-10-17 16:29:44,787 epoch 9 - iter 1152/1445 - loss 0.01023991 - time (sec): 58.18 - samples/sec: 2426.10 - lr: 0.000007 - momentum: 0.000000
199
+ 2023-10-17 16:29:52,556 epoch 9 - iter 1296/1445 - loss 0.01014140 - time (sec): 65.95 - samples/sec: 2401.36 - lr: 0.000006 - momentum: 0.000000
200
+ 2023-10-17 16:30:00,822 epoch 9 - iter 1440/1445 - loss 0.00966526 - time (sec): 74.22 - samples/sec: 2365.25 - lr: 0.000006 - momentum: 0.000000
201
+ 2023-10-17 16:30:01,060 ----------------------------------------------------------------------------------------------------
202
+ 2023-10-17 16:30:01,060 EPOCH 9 done: loss 0.0096 - lr: 0.000006
203
+ 2023-10-17 16:30:04,778 DEV : loss 0.15933012962341309 - f1-score (micro avg) 0.8648
204
+ 2023-10-17 16:30:04,794 saving best model
205
+ 2023-10-17 16:30:05,250 ----------------------------------------------------------------------------------------------------
206
+ 2023-10-17 16:30:13,087 epoch 10 - iter 144/1445 - loss 0.00391485 - time (sec): 7.83 - samples/sec: 2316.86 - lr: 0.000005 - momentum: 0.000000
207
+ 2023-10-17 16:30:20,360 epoch 10 - iter 288/1445 - loss 0.00316501 - time (sec): 15.11 - samples/sec: 2309.41 - lr: 0.000004 - momentum: 0.000000
208
+ 2023-10-17 16:30:27,734 epoch 10 - iter 432/1445 - loss 0.00358503 - time (sec): 22.48 - samples/sec: 2231.35 - lr: 0.000004 - momentum: 0.000000
209
+ 2023-10-17 16:30:35,048 epoch 10 - iter 576/1445 - loss 0.00564098 - time (sec): 29.80 - samples/sec: 2274.25 - lr: 0.000003 - momentum: 0.000000
210
+ 2023-10-17 16:30:43,390 epoch 10 - iter 720/1445 - loss 0.00548837 - time (sec): 38.14 - samples/sec: 2249.40 - lr: 0.000003 - momentum: 0.000000
211
+ 2023-10-17 16:30:51,087 epoch 10 - iter 864/1445 - loss 0.00552389 - time (sec): 45.83 - samples/sec: 2243.93 - lr: 0.000002 - momentum: 0.000000
212
+ 2023-10-17 16:30:59,119 epoch 10 - iter 1008/1445 - loss 0.00560035 - time (sec): 53.87 - samples/sec: 2241.21 - lr: 0.000002 - momentum: 0.000000
213
+ 2023-10-17 16:31:07,014 epoch 10 - iter 1152/1445 - loss 0.00546225 - time (sec): 61.76 - samples/sec: 2241.91 - lr: 0.000001 - momentum: 0.000000
214
+ 2023-10-17 16:31:14,728 epoch 10 - iter 1296/1445 - loss 0.00603445 - time (sec): 69.48 - samples/sec: 2258.14 - lr: 0.000001 - momentum: 0.000000
215
+ 2023-10-17 16:31:21,998 epoch 10 - iter 1440/1445 - loss 0.00608040 - time (sec): 76.75 - samples/sec: 2289.15 - lr: 0.000000 - momentum: 0.000000
216
+ 2023-10-17 16:31:22,229 ----------------------------------------------------------------------------------------------------
217
+ 2023-10-17 16:31:22,230 EPOCH 10 done: loss 0.0061 - lr: 0.000000
218
+ 2023-10-17 16:31:25,469 DEV : loss 0.15785089135169983 - f1-score (micro avg) 0.8684
219
+ 2023-10-17 16:31:25,485 saving best model
220
+ 2023-10-17 16:31:26,269 ----------------------------------------------------------------------------------------------------
221
+ 2023-10-17 16:31:26,270 Loading model from best epoch ...
222
+ 2023-10-17 16:31:27,631 SequenceTagger predicts: Dictionary with 13 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
223
+ 2023-10-17 16:31:30,393
224
+ Results:
225
+ - F-score (micro) 0.8444
226
+ - F-score (macro) 0.7376
227
+ - Accuracy 0.7372
228
+
229
+ By class:
230
+ precision recall f1-score support
231
+
232
+ PER 0.8584 0.8299 0.8439 482
233
+ LOC 0.9402 0.8581 0.8973 458
234
+ ORG 0.5370 0.4203 0.4715 69
235
+
236
+ micro avg 0.8763 0.8147 0.8444 1009
237
+ macro avg 0.7785 0.7027 0.7376 1009
238
+ weighted avg 0.8735 0.8147 0.8426 1009
239
+
240
+ 2023-10-17 16:31:30,393 ----------------------------------------------------------------------------------------------------