Hugging Face's logo

stefan-it
/
hmbench-icdar-nl-hmbyt5-bs4-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-2

Token Classification
Flair
PyTorch
TensorBoard
Dutch
sequence-tagger-model
Model card Files Metrics Community
2
stefan-it commited on
Commit
f5ddf2b
•
1 Parent(s): 247c6c4

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. best-model.pt +3 -0
  2. dev.tsv +0 -0
  3. final-model.pt +3 -0
  4. loss.tsv +11 -0
  5. runs/events.out.tfevents.1697060628.c8b2203b18a8.1914.6 +3 -0
  6. test.tsv +0 -0
  7. training.log +262 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:503d569227e3385fcdfc219d12601dfd503924674c63dc25a701970a6f540ca0
3
+ size 870793839
dev.tsv ADDED
The diff for this file is too large to render. See raw diff
 
final-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dd1c09b8bd46463ee48ed7c259ec1570180d33868751d7f67e03ddd01f774d38
3
+ size 870793956
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 21:51:43 0.0001 0.9247 0.1899 0.5098 0.2955 0.3741 0.2449
3
+ 2 21:59:25 0.0001 0.1133 0.0921 0.7909 0.7893 0.7901 0.6655
4
+ 3 22:07:08 0.0001 0.0666 0.0813 0.8765 0.8068 0.8402 0.7340
5
+ 4 22:14:40 0.0001 0.0441 0.0811 0.8598 0.8554 0.8576 0.7610
6
+ 5 22:22:15 0.0001 0.0352 0.1032 0.8885 0.8068 0.8457 0.7438
7
+ 6 22:29:54 0.0001 0.0271 0.1073 0.8879 0.8430 0.8649 0.7727
8
+ 7 22:38:07 0.0001 0.0201 0.1264 0.8930 0.8275 0.8590 0.7672
9
+ 8 22:45:48 0.0000 0.0145 0.1457 0.9133 0.7944 0.8497 0.7473
10
+ 9 22:53:12 0.0000 0.0108 0.1438 0.8953 0.8130 0.8522 0.7531
11
+ 10 23:00:41 0.0000 0.0084 0.1489 0.9005 0.8130 0.8545 0.7567
runs/events.out.tfevents.1697060628.c8b2203b18a8.1914.6 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3994a45322a13de7433abc0146e4e7dc283a120bc13b625f7aeb7f11bbc80d5d
3
+ size 808480
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,262 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-11 21:43:48,273 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-11 21:43:48,275 Model: "SequenceTagger(
3
+ (embeddings): ByT5Embeddings(
4
+ (model): T5EncoderModel(
5
+ (shared): Embedding(384, 1472)
6
+ (encoder): T5Stack(
7
+ (embed_tokens): Embedding(384, 1472)
8
+ (block): ModuleList(
9
+ (0): T5Block(
10
+ (layer): ModuleList(
11
+ (0): T5LayerSelfAttention(
12
+ (SelfAttention): T5Attention(
13
+ (q): Linear(in_features=1472, out_features=384, bias=False)
14
+ (k): Linear(in_features=1472, out_features=384, bias=False)
15
+ (v): Linear(in_features=1472, out_features=384, bias=False)
16
+ (o): Linear(in_features=384, out_features=1472, bias=False)
17
+ (relative_attention_bias): Embedding(32, 6)
18
+ )
19
+ (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
20
+ (dropout): Dropout(p=0.1, inplace=False)
21
+ )
22
+ (1): T5LayerFF(
23
+ (DenseReluDense): T5DenseGatedActDense(
24
+ (wi_0): Linear(in_features=1472, out_features=3584, bias=False)
25
+ (wi_1): Linear(in_features=1472, out_features=3584, bias=False)
26
+ (wo): Linear(in_features=3584, out_features=1472, bias=False)
27
+ (dropout): Dropout(p=0.1, inplace=False)
28
+ (act): NewGELUActivation()
29
+ )
30
+ (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
31
+ (dropout): Dropout(p=0.1, inplace=False)
32
+ )
33
+ )
34
+ )
35
+ (1-11): 11 x T5Block(
36
+ (layer): ModuleList(
37
+ (0): T5LayerSelfAttention(
38
+ (SelfAttention): T5Attention(
39
+ (q): Linear(in_features=1472, out_features=384, bias=False)
40
+ (k): Linear(in_features=1472, out_features=384, bias=False)
41
+ (v): Linear(in_features=1472, out_features=384, bias=False)
42
+ (o): Linear(in_features=384, out_features=1472, bias=False)
43
+ )
44
+ (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
45
+ (dropout): Dropout(p=0.1, inplace=False)
46
+ )
47
+ (1): T5LayerFF(
48
+ (DenseReluDense): T5DenseGatedActDense(
49
+ (wi_0): Linear(in_features=1472, out_features=3584, bias=False)
50
+ (wi_1): Linear(in_features=1472, out_features=3584, bias=False)
51
+ (wo): Linear(in_features=3584, out_features=1472, bias=False)
52
+ (dropout): Dropout(p=0.1, inplace=False)
53
+ (act): NewGELUActivation()
54
+ )
55
+ (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
56
+ (dropout): Dropout(p=0.1, inplace=False)
57
+ )
58
+ )
59
+ )
60
+ )
61
+ (final_layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
62
+ (dropout): Dropout(p=0.1, inplace=False)
63
+ )
64
+ )
65
+ )
66
+ (locked_dropout): LockedDropout(p=0.5)
67
+ (linear): Linear(in_features=1472, out_features=13, bias=True)
68
+ (loss_function): CrossEntropyLoss()
69
+ )"
70
+ 2023-10-11 21:43:48,275 ----------------------------------------------------------------------------------------------------
71
+ 2023-10-11 21:43:48,275 MultiCorpus: 5777 train + 722 dev + 723 test sentences
72
+ - NER_ICDAR_EUROPEANA Corpus: 5777 train + 722 dev + 723 test sentences - /root/.flair/datasets/ner_icdar_europeana/nl
73
+ 2023-10-11 21:43:48,275 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-11 21:43:48,275 Train: 5777 sentences
75
+ 2023-10-11 21:43:48,275 (train_with_dev=False, train_with_test=False)
76
+ 2023-10-11 21:43:48,275 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-11 21:43:48,276 Training Params:
78
+ 2023-10-11 21:43:48,276 - learning_rate: "0.00015"
79
+ 2023-10-11 21:43:48,276 - mini_batch_size: "4"
80
+ 2023-10-11 21:43:48,276 - max_epochs: "10"
81
+ 2023-10-11 21:43:48,276 - shuffle: "True"
82
+ 2023-10-11 21:43:48,276 ----------------------------------------------------------------------------------------------------
83
+ 2023-10-11 21:43:48,276 Plugins:
84
+ 2023-10-11 21:43:48,276 - TensorboardLogger
85
+ 2023-10-11 21:43:48,276 - LinearScheduler | warmup_fraction: '0.1'
86
+ 2023-10-11 21:43:48,276 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-11 21:43:48,276 Final evaluation on model from best epoch (best-model.pt)
88
+ 2023-10-11 21:43:48,276 - metric: "('micro avg', 'f1-score')"
89
+ 2023-10-11 21:43:48,276 ----------------------------------------------------------------------------------------------------
90
+ 2023-10-11 21:43:48,276 Computation:
91
+ 2023-10-11 21:43:48,276 - compute on device: cuda:0
92
+ 2023-10-11 21:43:48,276 - embedding storage: none
93
+ 2023-10-11 21:43:48,276 ----------------------------------------------------------------------------------------------------
94
+ 2023-10-11 21:43:48,277 Model training base path: "hmbench-icdar/nl-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-2"
95
+ 2023-10-11 21:43:48,277 ----------------------------------------------------------------------------------------------------
96
+ 2023-10-11 21:43:48,277 ----------------------------------------------------------------------------------------------------
97
+ 2023-10-11 21:43:48,277 Logging anything other than scalars to TensorBoard is currently not supported.
98
+ 2023-10-11 21:44:31,372 epoch 1 - iter 144/1445 - loss 2.56613189 - time (sec): 43.09 - samples/sec: 431.43 - lr: 0.000015 - momentum: 0.000000
99
+ 2023-10-11 21:45:13,718 epoch 1 - iter 288/1445 - loss 2.45511613 - time (sec): 85.44 - samples/sec: 417.54 - lr: 0.000030 - momentum: 0.000000
100
+ 2023-10-11 21:46:03,679 epoch 1 - iter 432/1445 - loss 2.18917637 - time (sec): 135.40 - samples/sec: 398.18 - lr: 0.000045 - momentum: 0.000000
101
+ 2023-10-11 21:46:51,899 epoch 1 - iter 576/1445 - loss 1.90175372 - time (sec): 183.62 - samples/sec: 388.46 - lr: 0.000060 - momentum: 0.000000
102
+ 2023-10-11 21:47:35,925 epoch 1 - iter 720/1445 - loss 1.62081885 - time (sec): 227.65 - samples/sec: 391.46 - lr: 0.000075 - momentum: 0.000000
103
+ 2023-10-11 21:48:19,686 epoch 1 - iter 864/1445 - loss 1.39933345 - time (sec): 271.41 - samples/sec: 393.50 - lr: 0.000090 - momentum: 0.000000
104
+ 2023-10-11 21:49:04,987 epoch 1 - iter 1008/1445 - loss 1.23677883 - time (sec): 316.71 - samples/sec: 393.63 - lr: 0.000105 - momentum: 0.000000
105
+ 2023-10-11 21:49:52,040 epoch 1 - iter 1152/1445 - loss 1.11004262 - time (sec): 363.76 - samples/sec: 389.52 - lr: 0.000119 - momentum: 0.000000
106
+ 2023-10-11 21:50:38,645 epoch 1 - iter 1296/1445 - loss 1.00603691 - time (sec): 410.37 - samples/sec: 388.26 - lr: 0.000134 - momentum: 0.000000
107
+ 2023-10-11 21:51:22,208 epoch 1 - iter 1440/1445 - loss 0.92641471 - time (sec): 453.93 - samples/sec: 387.25 - lr: 0.000149 - momentum: 0.000000
108
+ 2023-10-11 21:51:23,376 ----------------------------------------------------------------------------------------------------
109
+ 2023-10-11 21:51:23,376 EPOCH 1 done: loss 0.9247 - lr: 0.000149
110
+ 2023-10-11 21:51:43,816 DEV : loss 0.18985705077648163 - f1-score (micro avg) 0.3741
111
+ 2023-10-11 21:51:43,847 saving best model
112
+ 2023-10-11 21:51:44,770 ----------------------------------------------------------------------------------------------------
113
+ 2023-10-11 21:52:27,449 epoch 2 - iter 144/1445 - loss 0.14481160 - time (sec): 42.68 - samples/sec: 399.05 - lr: 0.000148 - momentum: 0.000000
114
+ 2023-10-11 21:53:11,274 epoch 2 - iter 288/1445 - loss 0.14361979 - time (sec): 86.50 - samples/sec: 401.10 - lr: 0.000147 - momentum: 0.000000
115
+ 2023-10-11 21:53:54,180 epoch 2 - iter 432/1445 - loss 0.13638655 - time (sec): 129.41 - samples/sec: 409.63 - lr: 0.000145 - momentum: 0.000000
116
+ 2023-10-11 21:54:37,286 epoch 2 - iter 576/1445 - loss 0.13335780 - time (sec): 172.51 - samples/sec: 412.26 - lr: 0.000143 - momentum: 0.000000
117
+ 2023-10-11 21:55:21,432 epoch 2 - iter 720/1445 - loss 0.12713278 - time (sec): 216.66 - samples/sec: 409.48 - lr: 0.000142 - momentum: 0.000000
118
+ 2023-10-11 21:56:03,104 epoch 2 - iter 864/1445 - loss 0.12400258 - time (sec): 258.33 - samples/sec: 409.74 - lr: 0.000140 - momentum: 0.000000
119
+ 2023-10-11 21:56:47,352 epoch 2 - iter 1008/1445 - loss 0.12005883 - time (sec): 302.58 - samples/sec: 407.29 - lr: 0.000138 - momentum: 0.000000
120
+ 2023-10-11 21:57:31,108 epoch 2 - iter 1152/1445 - loss 0.11751552 - time (sec): 346.34 - samples/sec: 406.59 - lr: 0.000137 - momentum: 0.000000
121
+ 2023-10-11 21:58:13,976 epoch 2 - iter 1296/1445 - loss 0.11540253 - time (sec): 389.20 - samples/sec: 404.69 - lr: 0.000135 - momentum: 0.000000
122
+ 2023-10-11 21:59:01,670 epoch 2 - iter 1440/1445 - loss 0.11322182 - time (sec): 436.90 - samples/sec: 402.21 - lr: 0.000133 - momentum: 0.000000
123
+ 2023-10-11 21:59:03,067 ----------------------------------------------------------------------------------------------------
124
+ 2023-10-11 21:59:03,067 EPOCH 2 done: loss 0.1133 - lr: 0.000133
125
+ 2023-10-11 21:59:25,392 DEV : loss 0.09209223836660385 - f1-score (micro avg) 0.7901
126
+ 2023-10-11 21:59:25,426 saving best model
127
+ 2023-10-11 21:59:34,741 ----------------------------------------------------------------------------------------------------
128
+ 2023-10-11 22:00:17,606 epoch 3 - iter 144/1445 - loss 0.08766300 - time (sec): 42.86 - samples/sec: 424.42 - lr: 0.000132 - momentum: 0.000000
129
+ 2023-10-11 22:01:00,870 epoch 3 - iter 288/1445 - loss 0.07859983 - time (sec): 86.12 - samples/sec: 409.94 - lr: 0.000130 - momentum: 0.000000
130
+ 2023-10-11 22:01:45,034 epoch 3 - iter 432/1445 - loss 0.07368106 - time (sec): 130.29 - samples/sec: 402.84 - lr: 0.000128 - momentum: 0.000000
131
+ 2023-10-11 22:02:28,697 epoch 3 - iter 576/1445 - loss 0.07207079 - time (sec): 173.95 - samples/sec: 398.48 - lr: 0.000127 - momentum: 0.000000
132
+ 2023-10-11 22:03:10,644 epoch 3 - iter 720/1445 - loss 0.06966165 - time (sec): 215.90 - samples/sec: 401.78 - lr: 0.000125 - momentum: 0.000000
133
+ 2023-10-11 22:03:55,232 epoch 3 - iter 864/1445 - loss 0.07048286 - time (sec): 260.49 - samples/sec: 406.23 - lr: 0.000123 - momentum: 0.000000
134
+ 2023-10-11 22:04:37,732 epoch 3 - iter 1008/1445 - loss 0.07055815 - time (sec): 302.99 - samples/sec: 405.74 - lr: 0.000122 - momentum: 0.000000
135
+ 2023-10-11 22:05:19,641 epoch 3 - iter 1152/1445 - loss 0.06860905 - time (sec): 344.90 - samples/sec: 406.49 - lr: 0.000120 - momentum: 0.000000
136
+ 2023-10-11 22:06:02,514 epoch 3 - iter 1296/1445 - loss 0.06746734 - time (sec): 387.77 - samples/sec: 405.33 - lr: 0.000118 - momentum: 0.000000
137
+ 2023-10-11 22:06:45,691 epoch 3 - iter 1440/1445 - loss 0.06670716 - time (sec): 430.95 - samples/sec: 407.69 - lr: 0.000117 - momentum: 0.000000
138
+ 2023-10-11 22:06:46,939 ----------------------------------------------------------------------------------------------------
139
+ 2023-10-11 22:06:46,939 EPOCH 3 done: loss 0.0666 - lr: 0.000117
140
+ 2023-10-11 22:07:08,004 DEV : loss 0.08132287114858627 - f1-score (micro avg) 0.8402
141
+ 2023-10-11 22:07:08,033 saving best model
142
+ 2023-10-11 22:07:10,846 ----------------------------------------------------------------------------------------------------
143
+ 2023-10-11 22:07:56,901 epoch 4 - iter 144/1445 - loss 0.05158645 - time (sec): 46.05 - samples/sec: 380.50 - lr: 0.000115 - momentum: 0.000000
144
+ 2023-10-11 22:08:39,454 epoch 4 - iter 288/1445 - loss 0.04322719 - time (sec): 88.60 - samples/sec: 401.07 - lr: 0.000113 - momentum: 0.000000
145
+ 2023-10-11 22:09:22,353 epoch 4 - iter 432/1445 - loss 0.04502248 - time (sec): 131.50 - samples/sec: 401.44 - lr: 0.000112 - momentum: 0.000000
146
+ 2023-10-11 22:10:05,063 epoch 4 - iter 576/1445 - loss 0.04533770 - time (sec): 174.21 - samples/sec: 403.27 - lr: 0.000110 - momentum: 0.000000
147
+ 2023-10-11 22:10:48,013 epoch 4 - iter 720/1445 - loss 0.04823559 - time (sec): 217.16 - samples/sec: 398.48 - lr: 0.000108 - momentum: 0.000000
148
+ 2023-10-11 22:11:31,776 epoch 4 - iter 864/1445 - loss 0.04734272 - time (sec): 260.93 - samples/sec: 401.28 - lr: 0.000107 - momentum: 0.000000
149
+ 2023-10-11 22:12:14,790 epoch 4 - iter 1008/1445 - loss 0.04796593 - time (sec): 303.94 - samples/sec: 404.98 - lr: 0.000105 - momentum: 0.000000
150
+ 2023-10-11 22:12:57,535 epoch 4 - iter 1152/1445 - loss 0.04550591 - time (sec): 346.68 - samples/sec: 405.64 - lr: 0.000103 - momentum: 0.000000
151
+ 2023-10-11 22:13:40,031 epoch 4 - iter 1296/1445 - loss 0.04326301 - time (sec): 389.18 - samples/sec: 410.29 - lr: 0.000102 - momentum: 0.000000
152
+ 2023-10-11 22:14:19,474 epoch 4 - iter 1440/1445 - loss 0.04418604 - time (sec): 428.62 - samples/sec: 410.30 - lr: 0.000100 - momentum: 0.000000
153
+ 2023-10-11 22:14:20,585 ----------------------------------------------------------------------------------------------------
154
+ 2023-10-11 22:14:20,586 EPOCH 4 done: loss 0.0441 - lr: 0.000100
155
+ 2023-10-11 22:14:40,862 DEV : loss 0.08113545924425125 - f1-score (micro avg) 0.8576
156
+ 2023-10-11 22:14:40,893 saving best model
157
+ 2023-10-11 22:14:43,508 ----------------------------------------------------------------------------------------------------
158
+ 2023-10-11 22:15:28,377 epoch 5 - iter 144/1445 - loss 0.04035968 - time (sec): 44.86 - samples/sec: 393.65 - lr: 0.000098 - momentum: 0.000000
159
+ 2023-10-11 22:16:14,096 epoch 5 - iter 288/1445 - loss 0.03196018 - time (sec): 90.58 - samples/sec: 389.38 - lr: 0.000097 - momentum: 0.000000
160
+ 2023-10-11 22:16:55,049 epoch 5 - iter 432/1445 - loss 0.03344470 - time (sec): 131.54 - samples/sec: 406.68 - lr: 0.000095 - momentum: 0.000000
161
+ 2023-10-11 22:17:35,393 epoch 5 - iter 576/1445 - loss 0.03320723 - time (sec): 171.88 - samples/sec: 411.81 - lr: 0.000093 - momentum: 0.000000
162
+ 2023-10-11 22:18:17,041 epoch 5 - iter 720/1445 - loss 0.03057888 - time (sec): 213.53 - samples/sec: 419.12 - lr: 0.000092 - momentum: 0.000000
163
+ 2023-10-11 22:18:57,338 epoch 5 - iter 864/1445 - loss 0.03031577 - time (sec): 253.83 - samples/sec: 416.95 - lr: 0.000090 - momentum: 0.000000
164
+ 2023-10-11 22:19:39,454 epoch 5 - iter 1008/1445 - loss 0.03257003 - time (sec): 295.94 - samples/sec: 417.92 - lr: 0.000088 - momentum: 0.000000
165
+ 2023-10-11 22:20:22,551 epoch 5 - iter 1152/1445 - loss 0.03228254 - time (sec): 339.04 - samples/sec: 416.03 - lr: 0.000087 - momentum: 0.000000
166
+ 2023-10-11 22:21:06,771 epoch 5 - iter 1296/1445 - loss 0.03578187 - time (sec): 383.26 - samples/sec: 413.65 - lr: 0.000085 - momentum: 0.000000
167
+ 2023-10-11 22:21:51,227 epoch 5 - iter 1440/1445 - loss 0.03531510 - time (sec): 427.72 - samples/sec: 410.54 - lr: 0.000083 - momentum: 0.000000
168
+ 2023-10-11 22:21:52,682 ----------------------------------------------------------------------------------------------------
169
+ 2023-10-11 22:21:52,683 EPOCH 5 done: loss 0.0352 - lr: 0.000083
170
+ 2023-10-11 22:22:15,675 DEV : loss 0.10317305475473404 - f1-score (micro avg) 0.8457
171
+ 2023-10-11 22:22:15,704 ----------------------------------------------------------------------------------------------------
172
+ 2023-10-11 22:22:59,890 epoch 6 - iter 144/1445 - loss 0.02205187 - time (sec): 44.18 - samples/sec: 421.08 - lr: 0.000082 - momentum: 0.000000
173
+ 2023-10-11 22:23:43,468 epoch 6 - iter 288/1445 - loss 0.01777970 - time (sec): 87.76 - samples/sec: 419.85 - lr: 0.000080 - momentum: 0.000000
174
+ 2023-10-11 22:24:23,986 epoch 6 - iter 432/1445 - loss 0.02038655 - time (sec): 128.28 - samples/sec: 413.43 - lr: 0.000078 - momentum: 0.000000
175
+ 2023-10-11 22:25:07,319 epoch 6 - iter 576/1445 - loss 0.02196497 - time (sec): 171.61 - samples/sec: 411.13 - lr: 0.000077 - momentum: 0.000000
176
+ 2023-10-11 22:25:50,124 epoch 6 - iter 720/1445 - loss 0.02171133 - time (sec): 214.42 - samples/sec: 404.91 - lr: 0.000075 - momentum: 0.000000
177
+ 2023-10-11 22:26:33,517 epoch 6 - iter 864/1445 - loss 0.02164208 - time (sec): 257.81 - samples/sec: 406.70 - lr: 0.000073 - momentum: 0.000000
178
+ 2023-10-11 22:27:18,879 epoch 6 - iter 1008/1445 - loss 0.02727128 - time (sec): 303.17 - samples/sec: 407.42 - lr: 0.000072 - momentum: 0.000000
179
+ 2023-10-11 22:28:02,229 epoch 6 - iter 1152/1445 - loss 0.02637994 - time (sec): 346.52 - samples/sec: 407.96 - lr: 0.000070 - momentum: 0.000000
180
+ 2023-10-11 22:28:43,733 epoch 6 - iter 1296/1445 - loss 0.02664456 - time (sec): 388.03 - samples/sec: 406.53 - lr: 0.000068 - momentum: 0.000000
181
+ 2023-10-11 22:29:30,599 epoch 6 - iter 1440/1445 - loss 0.02710892 - time (sec): 434.89 - samples/sec: 403.31 - lr: 0.000067 - momentum: 0.000000
182
+ 2023-10-11 22:29:32,275 ----------------------------------------------------------------------------------------------------
183
+ 2023-10-11 22:29:32,276 EPOCH 6 done: loss 0.0271 - lr: 0.000067
184
+ 2023-10-11 22:29:54,267 DEV : loss 0.1072750836610794 - f1-score (micro avg) 0.8649
185
+ 2023-10-11 22:29:54,297 saving best model
186
+ 2023-10-11 22:29:56,969 ----------------------------------------------------------------------------------------------------
187
+ 2023-10-11 22:30:46,657 epoch 7 - iter 144/1445 - loss 0.02635537 - time (sec): 49.68 - samples/sec: 357.49 - lr: 0.000065 - momentum: 0.000000
188
+ 2023-10-11 22:31:33,184 epoch 7 - iter 288/1445 - loss 0.02023054 - time (sec): 96.21 - samples/sec: 355.67 - lr: 0.000063 - momentum: 0.000000
189
+ 2023-10-11 22:32:20,494 epoch 7 - iter 432/1445 - loss 0.01764895 - time (sec): 143.52 - samples/sec: 351.82 - lr: 0.000062 - momentum: 0.000000
190
+ 2023-10-11 22:33:09,609 epoch 7 - iter 576/1445 - loss 0.01906232 - time (sec): 192.64 - samples/sec: 362.35 - lr: 0.000060 - momentum: 0.000000
191
+ 2023-10-11 22:33:55,335 epoch 7 - iter 720/1445 - loss 0.02408914 - time (sec): 238.36 - samples/sec: 368.13 - lr: 0.000058 - momentum: 0.000000
192
+ 2023-10-11 22:34:39,801 epoch 7 - iter 864/1445 - loss 0.02394349 - time (sec): 282.83 - samples/sec: 374.43 - lr: 0.000057 - momentum: 0.000000
193
+ 2023-10-11 22:35:25,999 epoch 7 - iter 1008/1445 - loss 0.02201986 - time (sec): 329.03 - samples/sec: 372.91 - lr: 0.000055 - momentum: 0.000000
194
+ 2023-10-11 22:36:14,321 epoch 7 - iter 1152/1445 - loss 0.02087086 - time (sec): 377.35 - samples/sec: 372.68 - lr: 0.000053 - momentum: 0.000000
195
+ 2023-10-11 22:37:01,701 epoch 7 - iter 1296/1445 - loss 0.02084656 - time (sec): 424.73 - samples/sec: 372.61 - lr: 0.000052 - momentum: 0.000000
196
+ 2023-10-11 22:37:44,560 epoch 7 - iter 1440/1445 - loss 0.02014863 - time (sec): 467.59 - samples/sec: 375.29 - lr: 0.000050 - momentum: 0.000000
197
+ 2023-10-11 22:37:45,902 ----------------------------------------------------------------------------------------------------
198
+ 2023-10-11 22:37:45,902 EPOCH 7 done: loss 0.0201 - lr: 0.000050
199
+ 2023-10-11 22:38:07,040 DEV : loss 0.12641826272010803 - f1-score (micro avg) 0.859
200
+ 2023-10-11 22:38:07,074 ----------------------------------------------------------------------------------------------------
201
+ 2023-10-11 22:38:53,705 epoch 8 - iter 144/1445 - loss 0.00670521 - time (sec): 46.63 - samples/sec: 385.04 - lr: 0.000048 - momentum: 0.000000
202
+ 2023-10-11 22:39:39,103 epoch 8 - iter 288/1445 - loss 0.01176368 - time (sec): 92.03 - samples/sec: 371.49 - lr: 0.000047 - momentum: 0.000000
203
+ 2023-10-11 22:40:21,149 epoch 8 - iter 432/1445 - loss 0.01224252 - time (sec): 134.07 - samples/sec: 378.66 - lr: 0.000045 - momentum: 0.000000
204
+ 2023-10-11 22:41:03,327 epoch 8 - iter 576/1445 - loss 0.01088605 - time (sec): 176.25 - samples/sec: 384.80 - lr: 0.000043 - momentum: 0.000000
205
+ 2023-10-11 22:41:45,603 epoch 8 - iter 720/1445 - loss 0.01170222 - time (sec): 218.53 - samples/sec: 390.97 - lr: 0.000042 - momentum: 0.000000
206
+ 2023-10-11 22:42:29,081 epoch 8 - iter 864/1445 - loss 0.01255260 - time (sec): 262.00 - samples/sec: 396.33 - lr: 0.000040 - momentum: 0.000000
207
+ 2023-10-11 22:43:13,554 epoch 8 - iter 1008/1445 - loss 0.01365201 - time (sec): 306.48 - samples/sec: 400.25 - lr: 0.000038 - momentum: 0.000000
208
+ 2023-10-11 22:43:56,080 epoch 8 - iter 1152/1445 - loss 0.01389711 - time (sec): 349.00 - samples/sec: 400.74 - lr: 0.000037 - momentum: 0.000000
209
+ 2023-10-11 22:44:42,051 epoch 8 - iter 1296/1445 - loss 0.01404508 - time (sec): 394.97 - samples/sec: 400.50 - lr: 0.000035 - momentum: 0.000000
210
+ 2023-10-11 22:45:23,596 epoch 8 - iter 1440/1445 - loss 0.01455041 - time (sec): 436.52 - samples/sec: 402.48 - lr: 0.000033 - momentum: 0.000000
211
+ 2023-10-11 22:45:24,969 ----------------------------------------------------------------------------------------------------
212
+ 2023-10-11 22:45:24,969 EPOCH 8 done: loss 0.0145 - lr: 0.000033
213
+ 2023-10-11 22:45:48,608 DEV : loss 0.14573527872562408 - f1-score (micro avg) 0.8497
214
+ 2023-10-11 22:45:48,654 ----------------------------------------------------------------------------------------------------
215
+ 2023-10-11 22:46:30,779 epoch 9 - iter 144/1445 - loss 0.00824711 - time (sec): 42.12 - samples/sec: 407.35 - lr: 0.000032 - momentum: 0.000000
216
+ 2023-10-11 22:47:11,606 epoch 9 - iter 288/1445 - loss 0.01013820 - time (sec): 82.95 - samples/sec: 410.28 - lr: 0.000030 - momentum: 0.000000
217
+ 2023-10-11 22:47:53,881 epoch 9 - iter 432/1445 - loss 0.01054407 - time (sec): 125.23 - samples/sec: 408.55 - lr: 0.000028 - momentum: 0.000000
218
+ 2023-10-11 22:48:37,536 epoch 9 - iter 576/1445 - loss 0.01159667 - time (sec): 168.88 - samples/sec: 413.26 - lr: 0.000027 - momentum: 0.000000
219
+ 2023-10-11 22:49:20,546 epoch 9 - iter 720/1445 - loss 0.01246631 - time (sec): 211.89 - samples/sec: 418.08 - lr: 0.000025 - momentum: 0.000000
220
+ 2023-10-11 22:50:01,940 epoch 9 - iter 864/1445 - loss 0.01199709 - time (sec): 253.28 - samples/sec: 419.72 - lr: 0.000023 - momentum: 0.000000
221
+ 2023-10-11 22:50:44,345 epoch 9 - iter 1008/1445 - loss 0.01176200 - time (sec): 295.69 - samples/sec: 419.73 - lr: 0.000022 - momentum: 0.000000
222
+ 2023-10-11 22:51:26,657 epoch 9 - iter 1152/1445 - loss 0.01140876 - time (sec): 338.00 - samples/sec: 418.90 - lr: 0.000020 - momentum: 0.000000
223
+ 2023-10-11 22:52:08,330 epoch 9 - iter 1296/1445 - loss 0.01128398 - time (sec): 379.67 - samples/sec: 417.47 - lr: 0.000018 - momentum: 0.000000
224
+ 2023-10-11 22:52:49,772 epoch 9 - iter 1440/1445 - loss 0.01080269 - time (sec): 421.12 - samples/sec: 417.01 - lr: 0.000017 - momentum: 0.000000
225
+ 2023-10-11 22:52:51,055 ----------------------------------------------------------------------------------------------------
226
+ 2023-10-11 22:52:51,056 EPOCH 9 done: loss 0.0108 - lr: 0.000017
227
+ 2023-10-11 22:53:12,104 DEV : loss 0.14379249513149261 - f1-score (micro avg) 0.8522
228
+ 2023-10-11 22:53:12,135 ----------------------------------------------------------------------------------------------------
229
+ 2023-10-11 22:53:54,142 epoch 10 - iter 144/1445 - loss 0.00870221 - time (sec): 42.01 - samples/sec: 428.73 - lr: 0.000015 - momentum: 0.000000
230
+ 2023-10-11 22:54:35,576 epoch 10 - iter 288/1445 - loss 0.00818486 - time (sec): 83.44 - samples/sec: 434.53 - lr: 0.000013 - momentum: 0.000000
231
+ 2023-10-11 22:55:19,109 epoch 10 - iter 432/1445 - loss 0.01027080 - time (sec): 126.97 - samples/sec: 436.28 - lr: 0.000012 - momentum: 0.000000
232
+ 2023-10-11 22:56:03,228 epoch 10 - iter 576/1445 - loss 0.00825222 - time (sec): 171.09 - samples/sec: 426.55 - lr: 0.000010 - momentum: 0.000000
233
+ 2023-10-11 22:56:51,934 epoch 10 - iter 720/1445 - loss 0.00778683 - time (sec): 219.80 - samples/sec: 414.50 - lr: 0.000008 - momentum: 0.000000
234
+ 2023-10-11 22:57:36,523 epoch 10 - iter 864/1445 - loss 0.00795167 - time (sec): 264.39 - samples/sec: 406.20 - lr: 0.000007 - momentum: 0.000000
235
+ 2023-10-11 22:58:17,410 epoch 10 - iter 1008/1445 - loss 0.00851367 - time (sec): 305.27 - samples/sec: 407.63 - lr: 0.000005 - momentum: 0.000000
236
+ 2023-10-11 22:58:58,536 epoch 10 - iter 1152/1445 - loss 0.00819079 - time (sec): 346.40 - samples/sec: 409.12 - lr: 0.000003 - momentum: 0.000000
237
+ 2023-10-11 22:59:39,221 epoch 10 - iter 1296/1445 - loss 0.00816272 - time (sec): 387.08 - samples/sec: 410.50 - lr: 0.000002 - momentum: 0.000000
238
+ 2023-10-11 23:00:19,487 epoch 10 - iter 1440/1445 - loss 0.00838020 - time (sec): 427.35 - samples/sec: 411.19 - lr: 0.000000 - momentum: 0.000000
239
+ 2023-10-11 23:00:20,673 ----------------------------------------------------------------------------------------------------
240
+ 2023-10-11 23:00:20,674 EPOCH 10 done: loss 0.0084 - lr: 0.000000
241
+ 2023-10-11 23:00:41,686 DEV : loss 0.1489063799381256 - f1-score (micro avg) 0.8545
242
+ 2023-10-11 23:00:42,634 ----------------------------------------------------------------------------------------------------
243
+ 2023-10-11 23:00:42,636 Loading model from best epoch ...
244
+ 2023-10-11 23:00:46,258 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
245
+ 2023-10-11 23:01:06,639
246
+ Results:
247
+ - F-score (micro) 0.8388
248
+ - F-score (macro) 0.7554
249
+ - Accuracy 0.7402
250
+
251
+ By class:
252
+ precision recall f1-score support
253
+
254
+ PER 0.8427 0.8444 0.8435 482
255
+ LOC 0.9051 0.8537 0.8787 458
256
+ ORG 0.5522 0.5362 0.5441 69
257
+
258
+ micro avg 0.8503 0.8276 0.8388 1009
259
+ macro avg 0.7667 0.7448 0.7554 1009
260
+ weighted avg 0.8511 0.8276 0.8390 1009
261
+
262
+ 2023-10-11 23:01:06,640 ----------------------------------------------------------------------------------------------------