Hugging Face's logo

stefan-it
/
hmbench-icdar-nl-hmbyt5-bs8-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-1

Token Classification
Flair
PyTorch
TensorBoard
Dutch
sequence-tagger-model
Model card Files Metrics Community
2
stefan-it commited on
Commit
bcc5022
•
1 Parent(s): 084a798

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.1697033130.c8b2203b18a8.1914.0 +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:f8840803db8c3518413c5da3ce77f3480d0ca6c63eac496a64fc959aed829303
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 sha256:3612c9264c3b0cdc3d5f14e9394974ea2b779f06bf7a05fd41e901835b755c1e
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 14:12:23 0.0001 1.1808 0.2482 0.0000 0.0000 0.0000 0.0000
3
+ 2 14:19:26 0.0001 0.1366 0.1194 0.7333 0.7500 0.7416 0.6153
4
+ 3 14:26:39 0.0001 0.0810 0.0933 0.8440 0.7324 0.7843 0.6547
5
+ 4 14:33:43 0.0001 0.0522 0.0754 0.8515 0.8295 0.8404 0.7401
6
+ 5 14:41:04 0.0001 0.0356 0.0821 0.8588 0.8233 0.8407 0.7366
7
+ 6 14:48:26 0.0001 0.0266 0.0841 0.8774 0.8647 0.8710 0.7852
8
+ 7 14:55:45 0.0001 0.0225 0.0974 0.8834 0.8378 0.8600 0.7673
9
+ 8 15:02:54 0.0000 0.0179 0.1137 0.8673 0.8171 0.8415 0.7399
10
+ 9 15:10:26 0.0000 0.0145 0.1123 0.8687 0.8409 0.8546 0.7600
11
+ 10 15:17:41 0.0000 0.0121 0.1201 0.8753 0.8264 0.8502 0.7526
runs/events.out.tfevents.1697033130.c8b2203b18a8.1914.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6acffe66fd4948ead9b2fc7be01c97d72e8a0277afd9188362a27e05aa6897c5
3
+ size 407048
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 14:05:30,951 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-11 14:05:30,953 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 14:05:30,953 ----------------------------------------------------------------------------------------------------
71
+ 2023-10-11 14:05:30,954 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 14:05:30,954 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-11 14:05:30,954 Train: 5777 sentences
75
+ 2023-10-11 14:05:30,954 (train_with_dev=False, train_with_test=False)
76
+ 2023-10-11 14:05:30,954 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-11 14:05:30,954 Training Params:
78
+ 2023-10-11 14:05:30,954 - learning_rate: "0.00015"
79
+ 2023-10-11 14:05:30,954 - mini_batch_size: "8"
80
+ 2023-10-11 14:05:30,954 - max_epochs: "10"
81
+ 2023-10-11 14:05:30,954 - shuffle: "True"
82
+ 2023-10-11 14:05:30,954 ----------------------------------------------------------------------------------------------------
83
+ 2023-10-11 14:05:30,954 Plugins:
84
+ 2023-10-11 14:05:30,954 - TensorboardLogger
85
+ 2023-10-11 14:05:30,954 - LinearScheduler | warmup_fraction: '0.1'
86
+ 2023-10-11 14:05:30,955 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-11 14:05:30,955 Final evaluation on model from best epoch (best-model.pt)
88
+ 2023-10-11 14:05:30,955 - metric: "('micro avg', 'f1-score')"
89
+ 2023-10-11 14:05:30,955 ----------------------------------------------------------------------------------------------------
90
+ 2023-10-11 14:05:30,955 Computation:
91
+ 2023-10-11 14:05:30,955 - compute on device: cuda:0
92
+ 2023-10-11 14:05:30,955 - embedding storage: none
93
+ 2023-10-11 14:05:30,955 ----------------------------------------------------------------------------------------------------
94
+ 2023-10-11 14:05:30,955 Model training base path: "hmbench-icdar/nl-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-1"
95
+ 2023-10-11 14:05:30,955 ----------------------------------------------------------------------------------------------------
96
+ 2023-10-11 14:05:30,955 ----------------------------------------------------------------------------------------------------
97
+ 2023-10-11 14:05:30,956 Logging anything other than scalars to TensorBoard is currently not supported.
98
+ 2023-10-11 14:06:12,814 epoch 1 - iter 72/723 - loss 2.58705621 - time (sec): 41.86 - samples/sec: 420.87 - lr: 0.000015 - momentum: 0.000000
99
+ 2023-10-11 14:06:55,356 epoch 1 - iter 144/723 - loss 2.55125919 - time (sec): 84.40 - samples/sec: 440.37 - lr: 0.000030 - momentum: 0.000000
100
+ 2023-10-11 14:07:32,736 epoch 1 - iter 216/723 - loss 2.42208641 - time (sec): 121.78 - samples/sec: 439.68 - lr: 0.000045 - momentum: 0.000000
101
+ 2023-10-11 14:08:11,674 epoch 1 - iter 288/723 - loss 2.21128263 - time (sec): 160.72 - samples/sec: 445.55 - lr: 0.000060 - momentum: 0.000000
102
+ 2023-10-11 14:08:52,057 epoch 1 - iter 360/723 - loss 1.96849538 - time (sec): 201.10 - samples/sec: 453.32 - lr: 0.000074 - momentum: 0.000000
103
+ 2023-10-11 14:09:30,023 epoch 1 - iter 432/723 - loss 1.75669262 - time (sec): 239.07 - samples/sec: 452.37 - lr: 0.000089 - momentum: 0.000000
104
+ 2023-10-11 14:10:08,540 epoch 1 - iter 504/723 - loss 1.56586690 - time (sec): 277.58 - samples/sec: 452.47 - lr: 0.000104 - momentum: 0.000000
105
+ 2023-10-11 14:10:46,586 epoch 1 - iter 576/723 - loss 1.41230121 - time (sec): 315.63 - samples/sec: 451.25 - lr: 0.000119 - momentum: 0.000000
106
+ 2023-10-11 14:11:23,583 epoch 1 - iter 648/723 - loss 1.29317716 - time (sec): 352.63 - samples/sec: 448.46 - lr: 0.000134 - momentum: 0.000000
107
+ 2023-10-11 14:12:02,646 epoch 1 - iter 720/723 - loss 1.18252757 - time (sec): 391.69 - samples/sec: 448.88 - lr: 0.000149 - momentum: 0.000000
108
+ 2023-10-11 14:12:03,721 ----------------------------------------------------------------------------------------------------
109
+ 2023-10-11 14:12:03,721 EPOCH 1 done: loss 1.1808 - lr: 0.000149
110
+ 2023-10-11 14:12:23,449 DEV : loss 0.2481938898563385 - f1-score (micro avg) 0.0
111
+ 2023-10-11 14:12:23,483 ----------------------------------------------------------------------------------------------------
112
+ 2023-10-11 14:13:02,418 epoch 2 - iter 72/723 - loss 0.19665532 - time (sec): 38.93 - samples/sec: 463.18 - lr: 0.000148 - momentum: 0.000000
113
+ 2023-10-11 14:13:43,718 epoch 2 - iter 144/723 - loss 0.17259289 - time (sec): 80.23 - samples/sec: 442.69 - lr: 0.000147 - momentum: 0.000000
114
+ 2023-10-11 14:14:22,701 epoch 2 - iter 216/723 - loss 0.17115371 - time (sec): 119.22 - samples/sec: 433.56 - lr: 0.000145 - momentum: 0.000000
115
+ 2023-10-11 14:15:03,146 epoch 2 - iter 288/723 - loss 0.16331197 - time (sec): 159.66 - samples/sec: 434.40 - lr: 0.000143 - momentum: 0.000000
116
+ 2023-10-11 14:15:43,402 epoch 2 - iter 360/723 - loss 0.16009303 - time (sec): 199.92 - samples/sec: 437.18 - lr: 0.000142 - momentum: 0.000000
117
+ 2023-10-11 14:16:23,005 epoch 2 - iter 432/723 - loss 0.15669479 - time (sec): 239.52 - samples/sec: 437.77 - lr: 0.000140 - momentum: 0.000000
118
+ 2023-10-11 14:17:02,620 epoch 2 - iter 504/723 - loss 0.14903751 - time (sec): 279.14 - samples/sec: 441.76 - lr: 0.000138 - momentum: 0.000000
119
+ 2023-10-11 14:17:43,533 epoch 2 - iter 576/723 - loss 0.14173836 - time (sec): 320.05 - samples/sec: 446.76 - lr: 0.000137 - momentum: 0.000000
120
+ 2023-10-11 14:18:21,648 epoch 2 - iter 648/723 - loss 0.13875105 - time (sec): 358.16 - samples/sec: 444.18 - lr: 0.000135 - momentum: 0.000000
121
+ 2023-10-11 14:19:03,265 epoch 2 - iter 720/723 - loss 0.13671767 - time (sec): 399.78 - samples/sec: 438.93 - lr: 0.000133 - momentum: 0.000000
122
+ 2023-10-11 14:19:04,918 ----------------------------------------------------------------------------------------------------
123
+ 2023-10-11 14:19:04,919 EPOCH 2 done: loss 0.1366 - lr: 0.000133
124
+ 2023-10-11 14:19:26,427 DEV : loss 0.11942420899868011 - f1-score (micro avg) 0.7416
125
+ 2023-10-11 14:19:26,458 saving best model
126
+ 2023-10-11 14:19:27,827 ----------------------------------------------------------------------------------------------------
127
+ 2023-10-11 14:20:12,005 epoch 3 - iter 72/723 - loss 0.08259658 - time (sec): 44.18 - samples/sec: 390.95 - lr: 0.000132 - momentum: 0.000000
128
+ 2023-10-11 14:20:53,017 epoch 3 - iter 144/723 - loss 0.09033580 - time (sec): 85.19 - samples/sec: 404.11 - lr: 0.000130 - momentum: 0.000000
129
+ 2023-10-11 14:21:34,442 epoch 3 - iter 216/723 - loss 0.08477045 - time (sec): 126.61 - samples/sec: 410.89 - lr: 0.000128 - momentum: 0.000000
130
+ 2023-10-11 14:22:14,549 epoch 3 - iter 288/723 - loss 0.09192864 - time (sec): 166.72 - samples/sec: 411.77 - lr: 0.000127 - momentum: 0.000000
131
+ 2023-10-11 14:22:55,726 epoch 3 - iter 360/723 - loss 0.08964205 - time (sec): 207.90 - samples/sec: 419.60 - lr: 0.000125 - momentum: 0.000000
132
+ 2023-10-11 14:23:36,234 epoch 3 - iter 432/723 - loss 0.08993680 - time (sec): 248.40 - samples/sec: 418.75 - lr: 0.000123 - momentum: 0.000000
133
+ 2023-10-11 14:24:14,710 epoch 3 - iter 504/723 - loss 0.08843282 - time (sec): 286.88 - samples/sec: 423.56 - lr: 0.000122 - momentum: 0.000000
134
+ 2023-10-11 14:24:54,374 epoch 3 - iter 576/723 - loss 0.08469597 - time (sec): 326.55 - samples/sec: 427.94 - lr: 0.000120 - momentum: 0.000000
135
+ 2023-10-11 14:25:33,840 epoch 3 - iter 648/723 - loss 0.08177926 - time (sec): 366.01 - samples/sec: 431.84 - lr: 0.000118 - momentum: 0.000000
136
+ 2023-10-11 14:26:15,057 epoch 3 - iter 720/723 - loss 0.08098319 - time (sec): 407.23 - samples/sec: 431.32 - lr: 0.000117 - momentum: 0.000000
137
+ 2023-10-11 14:26:16,272 ----------------------------------------------------------------------------------------------------
138
+ 2023-10-11 14:26:16,272 EPOCH 3 done: loss 0.0810 - lr: 0.000117
139
+ 2023-10-11 14:26:39,318 DEV : loss 0.0933394655585289 - f1-score (micro avg) 0.7843
140
+ 2023-10-11 14:26:39,365 saving best model
141
+ 2023-10-11 14:26:42,488 ----------------------------------------------------------------------------------------------------
142
+ 2023-10-11 14:27:23,951 epoch 4 - iter 72/723 - loss 0.06794431 - time (sec): 41.46 - samples/sec: 436.19 - lr: 0.000115 - momentum: 0.000000
143
+ 2023-10-11 14:28:04,485 epoch 4 - iter 144/723 - loss 0.05806820 - time (sec): 81.99 - samples/sec: 427.75 - lr: 0.000113 - momentum: 0.000000
144
+ 2023-10-11 14:28:44,946 epoch 4 - iter 216/723 - loss 0.05532765 - time (sec): 122.45 - samples/sec: 431.87 - lr: 0.000112 - momentum: 0.000000
145
+ 2023-10-11 14:29:25,877 epoch 4 - iter 288/723 - loss 0.05495337 - time (sec): 163.38 - samples/sec: 430.00 - lr: 0.000110 - momentum: 0.000000
146
+ 2023-10-11 14:30:06,043 epoch 4 - iter 360/723 - loss 0.05464785 - time (sec): 203.55 - samples/sec: 428.59 - lr: 0.000108 - momentum: 0.000000
147
+ 2023-10-11 14:30:43,057 epoch 4 - iter 432/723 - loss 0.05467701 - time (sec): 240.56 - samples/sec: 431.40 - lr: 0.000107 - momentum: 0.000000
148
+ 2023-10-11 14:31:22,681 epoch 4 - iter 504/723 - loss 0.05477446 - time (sec): 280.19 - samples/sec: 431.78 - lr: 0.000105 - momentum: 0.000000
149
+ 2023-10-11 14:32:04,286 epoch 4 - iter 576/723 - loss 0.05566990 - time (sec): 321.79 - samples/sec: 432.98 - lr: 0.000103 - momentum: 0.000000
150
+ 2023-10-11 14:32:42,989 epoch 4 - iter 648/723 - loss 0.05417209 - time (sec): 360.49 - samples/sec: 436.05 - lr: 0.000102 - momentum: 0.000000
151
+ 2023-10-11 14:33:21,891 epoch 4 - iter 720/723 - loss 0.05222244 - time (sec): 399.40 - samples/sec: 440.11 - lr: 0.000100 - momentum: 0.000000
152
+ 2023-10-11 14:33:22,951 ----------------------------------------------------------------------------------------------------
153
+ 2023-10-11 14:33:22,952 EPOCH 4 done: loss 0.0522 - lr: 0.000100
154
+ 2023-10-11 14:33:43,757 DEV : loss 0.0753728598356247 - f1-score (micro avg) 0.8404
155
+ 2023-10-11 14:33:43,792 saving best model
156
+ 2023-10-11 14:33:46,481 ----------------------------------------------------------------------------------------------------
157
+ 2023-10-11 14:34:26,467 epoch 5 - iter 72/723 - loss 0.02836811 - time (sec): 39.98 - samples/sec: 446.76 - lr: 0.000098 - momentum: 0.000000
158
+ 2023-10-11 14:35:04,451 epoch 5 - iter 144/723 - loss 0.03014743 - time (sec): 77.96 - samples/sec: 443.42 - lr: 0.000097 - momentum: 0.000000
159
+ 2023-10-11 14:35:43,263 epoch 5 - iter 216/723 - loss 0.03649224 - time (sec): 116.77 - samples/sec: 439.91 - lr: 0.000095 - momentum: 0.000000
160
+ 2023-10-11 14:36:24,885 epoch 5 - iter 288/723 - loss 0.03442032 - time (sec): 158.40 - samples/sec: 436.97 - lr: 0.000093 - momentum: 0.000000
161
+ 2023-10-11 14:37:11,469 epoch 5 - iter 360/723 - loss 0.03751789 - time (sec): 204.98 - samples/sec: 428.52 - lr: 0.000092 - momentum: 0.000000
162
+ 2023-10-11 14:37:56,871 epoch 5 - iter 432/723 - loss 0.03539090 - time (sec): 250.38 - samples/sec: 416.37 - lr: 0.000090 - momentum: 0.000000
163
+ 2023-10-11 14:38:39,139 epoch 5 - iter 504/723 - loss 0.03487207 - time (sec): 292.65 - samples/sec: 416.32 - lr: 0.000088 - momentum: 0.000000
164
+ 2023-10-11 14:39:21,228 epoch 5 - iter 576/723 - loss 0.03534212 - time (sec): 334.74 - samples/sec: 420.39 - lr: 0.000087 - momentum: 0.000000
165
+ 2023-10-11 14:40:01,428 epoch 5 - iter 648/723 - loss 0.03542313 - time (sec): 374.94 - samples/sec: 420.10 - lr: 0.000085 - momentum: 0.000000
166
+ 2023-10-11 14:40:41,071 epoch 5 - iter 720/723 - loss 0.03559847 - time (sec): 414.58 - samples/sec: 423.82 - lr: 0.000083 - momentum: 0.000000
167
+ 2023-10-11 14:40:42,254 ----------------------------------------------------------------------------------------------------
168
+ 2023-10-11 14:40:42,254 EPOCH 5 done: loss 0.0356 - lr: 0.000083
169
+ 2023-10-11 14:41:04,297 DEV : loss 0.08206792920827866 - f1-score (micro avg) 0.8407
170
+ 2023-10-11 14:41:04,332 saving best model
171
+ 2023-10-11 14:41:06,999 ----------------------------------------------------------------------------------------------------
172
+ 2023-10-11 14:41:44,849 epoch 6 - iter 72/723 - loss 0.02906877 - time (sec): 37.84 - samples/sec: 444.50 - lr: 0.000082 - momentum: 0.000000
173
+ 2023-10-11 14:42:25,331 epoch 6 - iter 144/723 - loss 0.02818642 - time (sec): 78.32 - samples/sec: 434.40 - lr: 0.000080 - momentum: 0.000000
174
+ 2023-10-11 14:43:06,681 epoch 6 - iter 216/723 - loss 0.02990584 - time (sec): 119.67 - samples/sec: 433.58 - lr: 0.000078 - momentum: 0.000000
175
+ 2023-10-11 14:43:47,544 epoch 6 - iter 288/723 - loss 0.02714357 - time (sec): 160.53 - samples/sec: 436.14 - lr: 0.000077 - momentum: 0.000000
176
+ 2023-10-11 14:44:29,330 epoch 6 - iter 360/723 - loss 0.02747624 - time (sec): 202.32 - samples/sec: 436.90 - lr: 0.000075 - momentum: 0.000000
177
+ 2023-10-11 14:45:10,018 epoch 6 - iter 432/723 - loss 0.02687441 - time (sec): 243.01 - samples/sec: 433.75 - lr: 0.000073 - momentum: 0.000000
178
+ 2023-10-11 14:45:52,721 epoch 6 - iter 504/723 - loss 0.02594408 - time (sec): 285.71 - samples/sec: 427.66 - lr: 0.000072 - momentum: 0.000000
179
+ 2023-10-11 14:46:37,946 epoch 6 - iter 576/723 - loss 0.02542891 - time (sec): 330.94 - samples/sec: 422.42 - lr: 0.000070 - momentum: 0.000000
180
+ 2023-10-11 14:47:23,370 epoch 6 - iter 648/723 - loss 0.02627547 - time (sec): 376.36 - samples/sec: 423.51 - lr: 0.000068 - momentum: 0.000000
181
+ 2023-10-11 14:48:05,278 epoch 6 - iter 720/723 - loss 0.02645835 - time (sec): 418.27 - samples/sec: 420.22 - lr: 0.000067 - momentum: 0.000000
182
+ 2023-10-11 14:48:06,364 ----------------------------------------------------------------------------------------------------
183
+ 2023-10-11 14:48:06,364 EPOCH 6 done: loss 0.0266 - lr: 0.000067
184
+ 2023-10-11 14:48:26,664 DEV : loss 0.08414550870656967 - f1-score (micro avg) 0.871
185
+ 2023-10-11 14:48:26,696 saving best model
186
+ 2023-10-11 14:48:29,329 ----------------------------------------------------------------------------------------------------
187
+ 2023-10-11 14:49:10,549 epoch 7 - iter 72/723 - loss 0.02559871 - time (sec): 41.22 - samples/sec: 434.37 - lr: 0.000065 - momentum: 0.000000
188
+ 2023-10-11 14:49:50,528 epoch 7 - iter 144/723 - loss 0.02202443 - time (sec): 81.19 - samples/sec: 429.79 - lr: 0.000063 - momentum: 0.000000
189
+ 2023-10-11 14:50:33,114 epoch 7 - iter 216/723 - loss 0.01970189 - time (sec): 123.78 - samples/sec: 418.63 - lr: 0.000062 - momentum: 0.000000
190
+ 2023-10-11 14:51:15,905 epoch 7 - iter 288/723 - loss 0.02107156 - time (sec): 166.57 - samples/sec: 413.39 - lr: 0.000060 - momentum: 0.000000
191
+ 2023-10-11 14:51:59,500 epoch 7 - iter 360/723 - loss 0.02138550 - time (sec): 210.17 - samples/sec: 411.59 - lr: 0.000058 - momentum: 0.000000
192
+ 2023-10-11 14:52:41,711 epoch 7 - iter 432/723 - loss 0.02195180 - time (sec): 252.38 - samples/sec: 413.71 - lr: 0.000057 - momentum: 0.000000
193
+ 2023-10-11 14:53:22,689 epoch 7 - iter 504/723 - loss 0.02270591 - time (sec): 293.36 - samples/sec: 417.57 - lr: 0.000055 - momentum: 0.000000
194
+ 2023-10-11 14:54:02,112 epoch 7 - iter 576/723 - loss 0.02192802 - time (sec): 332.78 - samples/sec: 420.35 - lr: 0.000053 - momentum: 0.000000
195
+ 2023-10-11 14:54:41,762 epoch 7 - iter 648/723 - loss 0.02326251 - time (sec): 372.43 - samples/sec: 422.01 - lr: 0.000052 - momentum: 0.000000
196
+ 2023-10-11 14:55:21,919 epoch 7 - iter 720/723 - loss 0.02251924 - time (sec): 412.59 - samples/sec: 425.38 - lr: 0.000050 - momentum: 0.000000
197
+ 2023-10-11 14:55:23,238 ----------------------------------------------------------------------------------------------------
198
+ 2023-10-11 14:55:23,239 EPOCH 7 done: loss 0.0225 - lr: 0.000050
199
+ 2023-10-11 14:55:45,123 DEV : loss 0.09741368144750595 - f1-score (micro avg) 0.86
200
+ 2023-10-11 14:55:45,161 ----------------------------------------------------------------------------------------------------
201
+ 2023-10-11 14:56:25,043 epoch 8 - iter 72/723 - loss 0.01566425 - time (sec): 39.88 - samples/sec: 470.64 - lr: 0.000048 - momentum: 0.000000
202
+ 2023-10-11 14:57:03,296 epoch 8 - iter 144/723 - loss 0.01425224 - time (sec): 78.13 - samples/sec: 460.33 - lr: 0.000047 - momentum: 0.000000
203
+ 2023-10-11 14:57:41,996 epoch 8 - iter 216/723 - loss 0.01421036 - time (sec): 116.83 - samples/sec: 454.02 - lr: 0.000045 - momentum: 0.000000
204
+ 2023-10-11 14:58:20,720 epoch 8 - iter 288/723 - loss 0.01334895 - time (sec): 155.56 - samples/sec: 449.16 - lr: 0.000043 - momentum: 0.000000
205
+ 2023-10-11 14:59:01,048 epoch 8 - iter 360/723 - loss 0.01534408 - time (sec): 195.89 - samples/sec: 448.40 - lr: 0.000042 - momentum: 0.000000
206
+ 2023-10-11 14:59:42,233 epoch 8 - iter 432/723 - loss 0.01520775 - time (sec): 237.07 - samples/sec: 447.96 - lr: 0.000040 - momentum: 0.000000
207
+ 2023-10-11 15:00:22,379 epoch 8 - iter 504/723 - loss 0.01628644 - time (sec): 277.22 - samples/sec: 447.92 - lr: 0.000038 - momentum: 0.000000
208
+ 2023-10-11 15:01:03,684 epoch 8 - iter 576/723 - loss 0.01780027 - time (sec): 318.52 - samples/sec: 447.22 - lr: 0.000037 - momentum: 0.000000
209
+ 2023-10-11 15:01:46,996 epoch 8 - iter 648/723 - loss 0.01751548 - time (sec): 361.83 - samples/sec: 437.74 - lr: 0.000035 - momentum: 0.000000
210
+ 2023-10-11 15:02:30,881 epoch 8 - iter 720/723 - loss 0.01788380 - time (sec): 405.72 - samples/sec: 432.99 - lr: 0.000033 - momentum: 0.000000
211
+ 2023-10-11 15:02:32,251 ----------------------------------------------------------------------------------------------------
212
+ 2023-10-11 15:02:32,251 EPOCH 8 done: loss 0.0179 - lr: 0.000033
213
+ 2023-10-11 15:02:54,685 DEV : loss 0.11365482956171036 - f1-score (micro avg) 0.8415
214
+ 2023-10-11 15:02:54,716 ----------------------------------------------------------------------------------------------------
215
+ 2023-10-11 15:03:41,064 epoch 9 - iter 72/723 - loss 0.02258103 - time (sec): 46.35 - samples/sec: 404.98 - lr: 0.000032 - momentum: 0.000000
216
+ 2023-10-11 15:04:22,261 epoch 9 - iter 144/723 - loss 0.01587706 - time (sec): 87.54 - samples/sec: 402.96 - lr: 0.000030 - momentum: 0.000000
217
+ 2023-10-11 15:05:06,366 epoch 9 - iter 216/723 - loss 0.01536952 - time (sec): 131.65 - samples/sec: 409.42 - lr: 0.000028 - momentum: 0.000000
218
+ 2023-10-11 15:05:51,100 epoch 9 - iter 288/723 - loss 0.01464115 - time (sec): 176.38 - samples/sec: 404.84 - lr: 0.000027 - momentum: 0.000000
219
+ 2023-10-11 15:06:36,197 epoch 9 - iter 360/723 - loss 0.01451980 - time (sec): 221.48 - samples/sec: 400.55 - lr: 0.000025 - momentum: 0.000000
220
+ 2023-10-11 15:07:21,999 epoch 9 - iter 432/723 - loss 0.01468506 - time (sec): 267.28 - samples/sec: 397.94 - lr: 0.000023 - momentum: 0.000000
221
+ 2023-10-11 15:08:04,206 epoch 9 - iter 504/723 - loss 0.01473622 - time (sec): 309.49 - samples/sec: 403.27 - lr: 0.000022 - momentum: 0.000000
222
+ 2023-10-11 15:08:45,094 epoch 9 - iter 576/723 - loss 0.01522224 - time (sec): 350.38 - samples/sec: 403.83 - lr: 0.000020 - momentum: 0.000000
223
+ 2023-10-11 15:09:24,875 epoch 9 - iter 648/723 - loss 0.01442489 - time (sec): 390.16 - samples/sec: 406.21 - lr: 0.000018 - momentum: 0.000000
224
+ 2023-10-11 15:10:04,792 epoch 9 - iter 720/723 - loss 0.01450651 - time (sec): 430.07 - samples/sec: 408.82 - lr: 0.000017 - momentum: 0.000000
225
+ 2023-10-11 15:10:05,942 ----------------------------------------------------------------------------------------------------
226
+ 2023-10-11 15:10:05,942 EPOCH 9 done: loss 0.0145 - lr: 0.000017
227
+ 2023-10-11 15:10:26,729 DEV : loss 0.11227083206176758 - f1-score (micro avg) 0.8546
228
+ 2023-10-11 15:10:26,766 ----------------------------------------------------------------------------------------------------
229
+ 2023-10-11 15:11:05,747 epoch 10 - iter 72/723 - loss 0.00745596 - time (sec): 38.98 - samples/sec: 430.39 - lr: 0.000015 - momentum: 0.000000
230
+ 2023-10-11 15:11:47,734 epoch 10 - iter 144/723 - loss 0.01171360 - time (sec): 80.96 - samples/sec: 435.46 - lr: 0.000013 - momentum: 0.000000
231
+ 2023-10-11 15:12:31,352 epoch 10 - iter 216/723 - loss 0.01187809 - time (sec): 124.58 - samples/sec: 420.80 - lr: 0.000012 - momentum: 0.000000
232
+ 2023-10-11 15:13:13,766 epoch 10 - iter 288/723 - loss 0.01086224 - time (sec): 167.00 - samples/sec: 414.49 - lr: 0.000010 - momentum: 0.000000
233
+ 2023-10-11 15:13:58,505 epoch 10 - iter 360/723 - loss 0.01143704 - time (sec): 211.74 - samples/sec: 414.27 - lr: 0.000008 - momentum: 0.000000
234
+ 2023-10-11 15:14:39,915 epoch 10 - iter 432/723 - loss 0.01105902 - time (sec): 253.15 - samples/sec: 418.84 - lr: 0.000007 - momentum: 0.000000
235
+ 2023-10-11 15:15:20,796 epoch 10 - iter 504/723 - loss 0.01259970 - time (sec): 294.03 - samples/sec: 422.08 - lr: 0.000005 - momentum: 0.000000
236
+ 2023-10-11 15:16:00,222 epoch 10 - iter 576/723 - loss 0.01185845 - time (sec): 333.45 - samples/sec: 422.00 - lr: 0.000003 - momentum: 0.000000
237
+ 2023-10-11 15:16:40,161 epoch 10 - iter 648/723 - loss 0.01209669 - time (sec): 373.39 - samples/sec: 424.20 - lr: 0.000002 - momentum: 0.000000
238
+ 2023-10-11 15:17:19,030 epoch 10 - iter 720/723 - loss 0.01209148 - time (sec): 412.26 - samples/sec: 425.91 - lr: 0.000000 - momentum: 0.000000
239
+ 2023-10-11 15:17:20,270 ----------------------------------------------------------------------------------------------------
240
+ 2023-10-11 15:17:20,270 EPOCH 10 done: loss 0.0121 - lr: 0.000000
241
+ 2023-10-11 15:17:41,620 DEV : loss 0.12011167407035828 - f1-score (micro avg) 0.8502
242
+ 2023-10-11 15:17:42,629 ----------------------------------------------------------------------------------------------------
243
+ 2023-10-11 15:17:42,632 Loading model from best epoch ...
244
+ 2023-10-11 15:17:48,592 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 15:18:08,765
246
+ Results:
247
+ - F-score (micro) 0.8516
248
+ - F-score (macro) 0.7648
249
+ - Accuracy 0.7513
250
+
251
+ By class:
252
+ precision recall f1-score support
253
+
254
+ PER 0.8173 0.8817 0.8483 482
255
+ LOC 0.9240 0.8755 0.8991 458
256
+ ORG 0.5932 0.5072 0.5469 69
257
+
258
+ micro avg 0.8500 0.8533 0.8516 1009
259
+ macro avg 0.7782 0.7548 0.7648 1009
260
+ weighted avg 0.8504 0.8533 0.8507 1009
261
+
262
+ 2023-10-11 15:18:08,765 ----------------------------------------------------------------------------------------------------