Hugging Face's logo

stefan-it
/
hmbench-icdar-nl-hmbyt5-bs4-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-4

Token Classification
Flair
PyTorch
TensorBoard
Dutch
sequence-tagger-model
Model card Files Metrics Community
2
stefan-it commited on
Commit
23eecbb
•
1 Parent(s): 847338c

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.1697128399.c8b2203b18a8.2408.7 +3 -0
  6. test.tsv +0 -0
  7. training.log +261 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:73f5f57a3dc67597a1bb455241b7a6bae92130f172cbed05ac260de8f16f15da
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 sha256:51d6bf4f5ed9188aceebdde17d5372e1f3ba6327e63f1408814e1da63ca776a2
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 16:40:48 0.0002 0.8825 0.1898 0.6960 0.2696 0.3887 0.2462
3
+ 2 16:48:12 0.0001 0.1131 0.0912 0.8539 0.8089 0.8308 0.7203
4
+ 3 16:55:32 0.0001 0.0644 0.0662 0.8834 0.8450 0.8638 0.7710
5
+ 4 17:02:44 0.0001 0.0441 0.0877 0.8758 0.8378 0.8564 0.7594
6
+ 5 17:09:49 0.0001 0.0319 0.1074 0.8659 0.8337 0.8495 0.7535
7
+ 6 17:16:56 0.0001 0.0248 0.1121 0.8866 0.8316 0.8582 0.7652
8
+ 7 17:24:20 0.0001 0.0191 0.1289 0.8909 0.8182 0.8530 0.7557
9
+ 8 17:31:35 0.0000 0.0138 0.1225 0.8843 0.8450 0.8642 0.7724
10
+ 9 17:38:52 0.0000 0.0096 0.1408 0.8974 0.8316 0.8633 0.7703
11
+ 10 17:46:30 0.0000 0.0075 0.1429 0.8923 0.8306 0.8604 0.7664
runs/events.out.tfevents.1697128399.c8b2203b18a8.2408.7 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:06ac0c705d9173cc405d46ad853b8463bb5b64183eb29c3968380e5d5e97bc0b
3
+ size 808480
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,261 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-12 16:33:19,371 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-12 16:33:19,373 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
+ 2023-10-12 16:33:19,373 ----------------------------------------------------------------------------------------------------
71
+ 2023-10-12 16:33:19,374 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
+ 2023-10-12 16:33:19,374 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-12 16:33:19,374 Train: 5777 sentences
75
+ 2023-10-12 16:33:19,374 (train_with_dev=False, train_with_test=False)
76
+ 2023-10-12 16:33:19,374 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-12 16:33:19,374 Training Params:
78
+ 2023-10-12 16:33:19,374 - learning_rate: "0.00016"
79
+ 2023-10-12 16:33:19,374 - mini_batch_size: "4"
80
+ 2023-10-12 16:33:19,374 - max_epochs: "10"
81
+ 2023-10-12 16:33:19,374 - shuffle: "True"
82
+ 2023-10-12 16:33:19,374 ----------------------------------------------------------------------------------------------------
83
+ 2023-10-12 16:33:19,374 Plugins:
84
+ 2023-10-12 16:33:19,374 - TensorboardLogger
85
+ 2023-10-12 16:33:19,375 - LinearScheduler | warmup_fraction: '0.1'
86
+ 2023-10-12 16:33:19,375 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-12 16:33:19,375 Final evaluation on model from best epoch (best-model.pt)
88
+ 2023-10-12 16:33:19,375 - metric: "('micro avg', 'f1-score')"
89
+ 2023-10-12 16:33:19,375 ----------------------------------------------------------------------------------------------------
90
+ 2023-10-12 16:33:19,375 Computation:
91
+ 2023-10-12 16:33:19,375 - compute on device: cuda:0
92
+ 2023-10-12 16:33:19,375 - embedding storage: none
93
+ 2023-10-12 16:33:19,375 ----------------------------------------------------------------------------------------------------
94
+ 2023-10-12 16:33:19,375 Model training base path: "hmbench-icdar/nl-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-4"
95
+ 2023-10-12 16:33:19,375 ----------------------------------------------------------------------------------------------------
96
+ 2023-10-12 16:33:19,375 ----------------------------------------------------------------------------------------------------
97
+ 2023-10-12 16:33:19,375 Logging anything other than scalars to TensorBoard is currently not supported.
98
+ 2023-10-12 16:34:03,482 epoch 1 - iter 144/1445 - loss 2.56300254 - time (sec): 44.10 - samples/sec: 405.02 - lr: 0.000016 - momentum: 0.000000
99
+ 2023-10-12 16:34:47,886 epoch 1 - iter 288/1445 - loss 2.38526191 - time (sec): 88.51 - samples/sec: 407.78 - lr: 0.000032 - momentum: 0.000000
100
+ 2023-10-12 16:35:30,464 epoch 1 - iter 432/1445 - loss 2.13289859 - time (sec): 131.09 - samples/sec: 405.11 - lr: 0.000048 - momentum: 0.000000
101
+ 2023-10-12 16:36:13,934 epoch 1 - iter 576/1445 - loss 1.84024411 - time (sec): 174.56 - samples/sec: 405.11 - lr: 0.000064 - momentum: 0.000000
102
+ 2023-10-12 16:36:56,656 epoch 1 - iter 720/1445 - loss 1.55933001 - time (sec): 217.28 - samples/sec: 406.68 - lr: 0.000080 - momentum: 0.000000
103
+ 2023-10-12 16:37:37,250 epoch 1 - iter 864/1445 - loss 1.35312862 - time (sec): 257.87 - samples/sec: 407.98 - lr: 0.000096 - momentum: 0.000000
104
+ 2023-10-12 16:38:19,681 epoch 1 - iter 1008/1445 - loss 1.19495096 - time (sec): 300.30 - samples/sec: 407.48 - lr: 0.000112 - momentum: 0.000000
105
+ 2023-10-12 16:39:02,528 epoch 1 - iter 1152/1445 - loss 1.06319855 - time (sec): 343.15 - samples/sec: 409.47 - lr: 0.000127 - momentum: 0.000000
106
+ 2023-10-12 16:39:44,972 epoch 1 - iter 1296/1445 - loss 0.96042465 - time (sec): 385.59 - samples/sec: 412.03 - lr: 0.000143 - momentum: 0.000000
107
+ 2023-10-12 16:40:27,593 epoch 1 - iter 1440/1445 - loss 0.88488446 - time (sec): 428.22 - samples/sec: 410.24 - lr: 0.000159 - momentum: 0.000000
108
+ 2023-10-12 16:40:28,827 ----------------------------------------------------------------------------------------------------
109
+ 2023-10-12 16:40:28,828 EPOCH 1 done: loss 0.8825 - lr: 0.000159
110
+ 2023-10-12 16:40:48,696 DEV : loss 0.18981926143169403 - f1-score (micro avg) 0.3887
111
+ 2023-10-12 16:40:48,730 saving best model
112
+ 2023-10-12 16:40:49,705 ----------------------------------------------------------------------------------------------------
113
+ 2023-10-12 16:41:32,871 epoch 2 - iter 144/1445 - loss 0.13981313 - time (sec): 43.16 - samples/sec: 410.92 - lr: 0.000158 - momentum: 0.000000
114
+ 2023-10-12 16:42:13,700 epoch 2 - iter 288/1445 - loss 0.13956803 - time (sec): 83.99 - samples/sec: 415.01 - lr: 0.000156 - momentum: 0.000000
115
+ 2023-10-12 16:42:54,732 epoch 2 - iter 432/1445 - loss 0.13836300 - time (sec): 125.02 - samples/sec: 417.55 - lr: 0.000155 - momentum: 0.000000
116
+ 2023-10-12 16:43:36,244 epoch 2 - iter 576/1445 - loss 0.13112203 - time (sec): 166.54 - samples/sec: 417.27 - lr: 0.000153 - momentum: 0.000000
117
+ 2023-10-12 16:44:19,336 epoch 2 - iter 720/1445 - loss 0.12498734 - time (sec): 209.63 - samples/sec: 409.49 - lr: 0.000151 - momentum: 0.000000
118
+ 2023-10-12 16:45:03,660 epoch 2 - iter 864/1445 - loss 0.12218588 - time (sec): 253.95 - samples/sec: 407.93 - lr: 0.000149 - momentum: 0.000000
119
+ 2023-10-12 16:45:46,648 epoch 2 - iter 1008/1445 - loss 0.12188956 - time (sec): 296.94 - samples/sec: 410.26 - lr: 0.000148 - momentum: 0.000000
120
+ 2023-10-12 16:46:27,790 epoch 2 - iter 1152/1445 - loss 0.11964934 - time (sec): 338.08 - samples/sec: 413.86 - lr: 0.000146 - momentum: 0.000000
121
+ 2023-10-12 16:47:09,340 epoch 2 - iter 1296/1445 - loss 0.11521379 - time (sec): 379.63 - samples/sec: 415.90 - lr: 0.000144 - momentum: 0.000000
122
+ 2023-10-12 16:47:50,210 epoch 2 - iter 1440/1445 - loss 0.11336959 - time (sec): 420.50 - samples/sec: 417.31 - lr: 0.000142 - momentum: 0.000000
123
+ 2023-10-12 16:47:51,614 ----------------------------------------------------------------------------------------------------
124
+ 2023-10-12 16:47:51,614 EPOCH 2 done: loss 0.1131 - lr: 0.000142
125
+ 2023-10-12 16:48:12,874 DEV : loss 0.09119933843612671 - f1-score (micro avg) 0.8308
126
+ 2023-10-12 16:48:12,904 saving best model
127
+ 2023-10-12 16:48:15,491 ----------------------------------------------------------------------------------------------------
128
+ 2023-10-12 16:48:57,020 epoch 3 - iter 144/1445 - loss 0.07985533 - time (sec): 41.52 - samples/sec: 405.49 - lr: 0.000140 - momentum: 0.000000
129
+ 2023-10-12 16:49:38,239 epoch 3 - iter 288/1445 - loss 0.07523997 - time (sec): 82.74 - samples/sec: 418.79 - lr: 0.000139 - momentum: 0.000000
130
+ 2023-10-12 16:50:18,976 epoch 3 - iter 432/1445 - loss 0.07488360 - time (sec): 123.48 - samples/sec: 419.10 - lr: 0.000137 - momentum: 0.000000
131
+ 2023-10-12 16:50:59,095 epoch 3 - iter 576/1445 - loss 0.07112822 - time (sec): 163.60 - samples/sec: 421.60 - lr: 0.000135 - momentum: 0.000000
132
+ 2023-10-12 16:51:39,727 epoch 3 - iter 720/1445 - loss 0.07107570 - time (sec): 204.23 - samples/sec: 423.37 - lr: 0.000133 - momentum: 0.000000
133
+ 2023-10-12 16:52:23,561 epoch 3 - iter 864/1445 - loss 0.07062393 - time (sec): 248.06 - samples/sec: 425.40 - lr: 0.000132 - momentum: 0.000000
134
+ 2023-10-12 16:53:06,138 epoch 3 - iter 1008/1445 - loss 0.06769948 - time (sec): 290.64 - samples/sec: 424.70 - lr: 0.000130 - momentum: 0.000000
135
+ 2023-10-12 16:53:47,855 epoch 3 - iter 1152/1445 - loss 0.06676501 - time (sec): 332.36 - samples/sec: 423.10 - lr: 0.000128 - momentum: 0.000000
136
+ 2023-10-12 16:54:29,300 epoch 3 - iter 1296/1445 - loss 0.06508146 - time (sec): 373.80 - samples/sec: 422.23 - lr: 0.000126 - momentum: 0.000000
137
+ 2023-10-12 16:55:11,121 epoch 3 - iter 1440/1445 - loss 0.06435452 - time (sec): 415.62 - samples/sec: 422.69 - lr: 0.000125 - momentum: 0.000000
138
+ 2023-10-12 16:55:12,348 ----------------------------------------------------------------------------------------------------
139
+ 2023-10-12 16:55:12,349 EPOCH 3 done: loss 0.0644 - lr: 0.000125
140
+ 2023-10-12 16:55:32,850 DEV : loss 0.06621405482292175 - f1-score (micro avg) 0.8638
141
+ 2023-10-12 16:55:32,879 saving best model
142
+ 2023-10-12 16:55:35,449 ----------------------------------------------------------------------------------------------------
143
+ 2023-10-12 16:56:16,669 epoch 4 - iter 144/1445 - loss 0.03491763 - time (sec): 41.22 - samples/sec: 458.62 - lr: 0.000123 - momentum: 0.000000
144
+ 2023-10-12 16:56:55,734 epoch 4 - iter 288/1445 - loss 0.03857362 - time (sec): 80.28 - samples/sec: 436.36 - lr: 0.000121 - momentum: 0.000000
145
+ 2023-10-12 16:57:35,889 epoch 4 - iter 432/1445 - loss 0.03858829 - time (sec): 120.43 - samples/sec: 429.12 - lr: 0.000119 - momentum: 0.000000
146
+ 2023-10-12 16:58:16,860 epoch 4 - iter 576/1445 - loss 0.04126412 - time (sec): 161.41 - samples/sec: 427.20 - lr: 0.000117 - momentum: 0.000000
147
+ 2023-10-12 16:58:57,647 epoch 4 - iter 720/1445 - loss 0.04127926 - time (sec): 202.19 - samples/sec: 425.21 - lr: 0.000116 - momentum: 0.000000
148
+ 2023-10-12 16:59:40,966 epoch 4 - iter 864/1445 - loss 0.04331915 - time (sec): 245.51 - samples/sec: 423.87 - lr: 0.000114 - momentum: 0.000000
149
+ 2023-10-12 17:00:22,245 epoch 4 - iter 1008/1445 - loss 0.04325634 - time (sec): 286.79 - samples/sec: 429.29 - lr: 0.000112 - momentum: 0.000000
150
+ 2023-10-12 17:01:01,932 epoch 4 - iter 1152/1445 - loss 0.04376823 - time (sec): 326.48 - samples/sec: 429.38 - lr: 0.000110 - momentum: 0.000000
151
+ 2023-10-12 17:01:41,831 epoch 4 - iter 1296/1445 - loss 0.04471474 - time (sec): 366.38 - samples/sec: 430.38 - lr: 0.000109 - momentum: 0.000000
152
+ 2023-10-12 17:02:22,332 epoch 4 - iter 1440/1445 - loss 0.04407434 - time (sec): 406.88 - samples/sec: 431.94 - lr: 0.000107 - momentum: 0.000000
153
+ 2023-10-12 17:02:23,481 ----------------------------------------------------------------------------------------------------
154
+ 2023-10-12 17:02:23,481 EPOCH 4 done: loss 0.0441 - lr: 0.000107
155
+ 2023-10-12 17:02:44,195 DEV : loss 0.08768334984779358 - f1-score (micro avg) 0.8564
156
+ 2023-10-12 17:02:44,226 ----------------------------------------------------------------------------------------------------
157
+ 2023-10-12 17:03:24,221 epoch 5 - iter 144/1445 - loss 0.01928414 - time (sec): 39.99 - samples/sec: 424.43 - lr: 0.000105 - momentum: 0.000000
158
+ 2023-10-12 17:04:04,655 epoch 5 - iter 288/1445 - loss 0.02782982 - time (sec): 80.43 - samples/sec: 427.60 - lr: 0.000103 - momentum: 0.000000
159
+ 2023-10-12 17:04:44,485 epoch 5 - iter 432/1445 - loss 0.02750831 - time (sec): 120.26 - samples/sec: 433.41 - lr: 0.000101 - momentum: 0.000000
160
+ 2023-10-12 17:05:25,173 epoch 5 - iter 576/1445 - loss 0.02988547 - time (sec): 160.95 - samples/sec: 438.34 - lr: 0.000100 - momentum: 0.000000
161
+ 2023-10-12 17:06:05,256 epoch 5 - iter 720/1445 - loss 0.03160699 - time (sec): 201.03 - samples/sec: 438.23 - lr: 0.000098 - momentum: 0.000000
162
+ 2023-10-12 17:06:45,729 epoch 5 - iter 864/1445 - loss 0.03162838 - time (sec): 241.50 - samples/sec: 437.51 - lr: 0.000096 - momentum: 0.000000
163
+ 2023-10-12 17:07:24,736 epoch 5 - iter 1008/1445 - loss 0.03117068 - time (sec): 280.51 - samples/sec: 435.41 - lr: 0.000094 - momentum: 0.000000
164
+ 2023-10-12 17:08:05,615 epoch 5 - iter 1152/1445 - loss 0.03085839 - time (sec): 321.39 - samples/sec: 435.33 - lr: 0.000093 - momentum: 0.000000
165
+ 2023-10-12 17:08:46,790 epoch 5 - iter 1296/1445 - loss 0.03070485 - time (sec): 362.56 - samples/sec: 434.52 - lr: 0.000091 - momentum: 0.000000
166
+ 2023-10-12 17:09:27,677 epoch 5 - iter 1440/1445 - loss 0.03200208 - time (sec): 403.45 - samples/sec: 435.47 - lr: 0.000089 - momentum: 0.000000
167
+ 2023-10-12 17:09:28,879 ----------------------------------------------------------------------------------------------------
168
+ 2023-10-12 17:09:28,879 EPOCH 5 done: loss 0.0319 - lr: 0.000089
169
+ 2023-10-12 17:09:49,304 DEV : loss 0.10739118605852127 - f1-score (micro avg) 0.8495
170
+ 2023-10-12 17:09:49,334 ----------------------------------------------------------------------------------------------------
171
+ 2023-10-12 17:10:30,193 epoch 6 - iter 144/1445 - loss 0.03926076 - time (sec): 40.86 - samples/sec: 439.95 - lr: 0.000087 - momentum: 0.000000
172
+ 2023-10-12 17:11:09,843 epoch 6 - iter 288/1445 - loss 0.03195608 - time (sec): 80.51 - samples/sec: 434.47 - lr: 0.000085 - momentum: 0.000000
173
+ 2023-10-12 17:11:50,454 epoch 6 - iter 432/1445 - loss 0.03002443 - time (sec): 121.12 - samples/sec: 436.80 - lr: 0.000084 - momentum: 0.000000
174
+ 2023-10-12 17:12:31,958 epoch 6 - iter 576/1445 - loss 0.02767328 - time (sec): 162.62 - samples/sec: 441.01 - lr: 0.000082 - momentum: 0.000000
175
+ 2023-10-12 17:13:10,662 epoch 6 - iter 720/1445 - loss 0.02642560 - time (sec): 201.33 - samples/sec: 434.28 - lr: 0.000080 - momentum: 0.000000
176
+ 2023-10-12 17:13:52,438 epoch 6 - iter 864/1445 - loss 0.02552455 - time (sec): 243.10 - samples/sec: 439.35 - lr: 0.000078 - momentum: 0.000000
177
+ 2023-10-12 17:14:31,894 epoch 6 - iter 1008/1445 - loss 0.02476366 - time (sec): 282.56 - samples/sec: 437.47 - lr: 0.000076 - momentum: 0.000000
178
+ 2023-10-12 17:15:12,546 epoch 6 - iter 1152/1445 - loss 0.02469782 - time (sec): 323.21 - samples/sec: 436.99 - lr: 0.000075 - momentum: 0.000000
179
+ 2023-10-12 17:15:54,048 epoch 6 - iter 1296/1445 - loss 0.02478677 - time (sec): 364.71 - samples/sec: 435.29 - lr: 0.000073 - momentum: 0.000000
180
+ 2023-10-12 17:16:33,885 epoch 6 - iter 1440/1445 - loss 0.02484737 - time (sec): 404.55 - samples/sec: 434.17 - lr: 0.000071 - momentum: 0.000000
181
+ 2023-10-12 17:16:35,141 ----------------------------------------------------------------------------------------------------
182
+ 2023-10-12 17:16:35,141 EPOCH 6 done: loss 0.0248 - lr: 0.000071
183
+ 2023-10-12 17:16:56,074 DEV : loss 0.11206483840942383 - f1-score (micro avg) 0.8582
184
+ 2023-10-12 17:16:56,105 ----------------------------------------------------------------------------------------------------
185
+ 2023-10-12 17:17:37,350 epoch 7 - iter 144/1445 - loss 0.03196159 - time (sec): 41.24 - samples/sec: 405.21 - lr: 0.000069 - momentum: 0.000000
186
+ 2023-10-12 17:18:21,847 epoch 7 - iter 288/1445 - loss 0.02494608 - time (sec): 85.74 - samples/sec: 417.58 - lr: 0.000068 - momentum: 0.000000
187
+ 2023-10-12 17:19:04,928 epoch 7 - iter 432/1445 - loss 0.02528055 - time (sec): 128.82 - samples/sec: 416.07 - lr: 0.000066 - momentum: 0.000000
188
+ 2023-10-12 17:19:47,446 epoch 7 - iter 576/1445 - loss 0.02201332 - time (sec): 171.34 - samples/sec: 414.68 - lr: 0.000064 - momentum: 0.000000
189
+ 2023-10-12 17:20:30,277 epoch 7 - iter 720/1445 - loss 0.02177977 - time (sec): 214.17 - samples/sec: 414.34 - lr: 0.000062 - momentum: 0.000000
190
+ 2023-10-12 17:21:12,767 epoch 7 - iter 864/1445 - loss 0.02046133 - time (sec): 256.66 - samples/sec: 416.64 - lr: 0.000060 - momentum: 0.000000
191
+ 2023-10-12 17:21:54,758 epoch 7 - iter 1008/1445 - loss 0.02080315 - time (sec): 298.65 - samples/sec: 419.29 - lr: 0.000059 - momentum: 0.000000
192
+ 2023-10-12 17:22:35,231 epoch 7 - iter 1152/1445 - loss 0.01995810 - time (sec): 339.12 - samples/sec: 420.44 - lr: 0.000057 - momentum: 0.000000
193
+ 2023-10-12 17:23:15,074 epoch 7 - iter 1296/1445 - loss 0.01959775 - time (sec): 378.97 - samples/sec: 419.20 - lr: 0.000055 - momentum: 0.000000
194
+ 2023-10-12 17:23:57,171 epoch 7 - iter 1440/1445 - loss 0.01919288 - time (sec): 421.06 - samples/sec: 417.48 - lr: 0.000053 - momentum: 0.000000
195
+ 2023-10-12 17:23:58,329 ----------------------------------------------------------------------------------------------------
196
+ 2023-10-12 17:23:58,330 EPOCH 7 done: loss 0.0191 - lr: 0.000053
197
+ 2023-10-12 17:24:20,533 DEV : loss 0.12889063358306885 - f1-score (micro avg) 0.853
198
+ 2023-10-12 17:24:20,562 ----------------------------------------------------------------------------------------------------
199
+ 2023-10-12 17:25:01,698 epoch 8 - iter 144/1445 - loss 0.01021579 - time (sec): 41.13 - samples/sec: 434.73 - lr: 0.000052 - momentum: 0.000000
200
+ 2023-10-12 17:25:42,787 epoch 8 - iter 288/1445 - loss 0.01004474 - time (sec): 82.22 - samples/sec: 437.18 - lr: 0.000050 - momentum: 0.000000
201
+ 2023-10-12 17:26:23,418 epoch 8 - iter 432/1445 - loss 0.01117270 - time (sec): 122.85 - samples/sec: 435.97 - lr: 0.000048 - momentum: 0.000000
202
+ 2023-10-12 17:27:05,836 epoch 8 - iter 576/1445 - loss 0.01033342 - time (sec): 165.27 - samples/sec: 437.03 - lr: 0.000046 - momentum: 0.000000
203
+ 2023-10-12 17:27:45,951 epoch 8 - iter 720/1445 - loss 0.01317246 - time (sec): 205.39 - samples/sec: 428.88 - lr: 0.000044 - momentum: 0.000000
204
+ 2023-10-12 17:28:26,943 epoch 8 - iter 864/1445 - loss 0.01310077 - time (sec): 246.38 - samples/sec: 426.90 - lr: 0.000043 - momentum: 0.000000
205
+ 2023-10-12 17:29:07,910 epoch 8 - iter 1008/1445 - loss 0.01364110 - time (sec): 287.35 - samples/sec: 426.32 - lr: 0.000041 - momentum: 0.000000
206
+ 2023-10-12 17:29:50,319 epoch 8 - iter 1152/1445 - loss 0.01391426 - time (sec): 329.75 - samples/sec: 425.34 - lr: 0.000039 - momentum: 0.000000
207
+ 2023-10-12 17:30:32,149 epoch 8 - iter 1296/1445 - loss 0.01330664 - time (sec): 371.58 - samples/sec: 424.88 - lr: 0.000037 - momentum: 0.000000
208
+ 2023-10-12 17:31:14,018 epoch 8 - iter 1440/1445 - loss 0.01380146 - time (sec): 413.45 - samples/sec: 425.19 - lr: 0.000036 - momentum: 0.000000
209
+ 2023-10-12 17:31:15,188 ----------------------------------------------------------------------------------------------------
210
+ 2023-10-12 17:31:15,188 EPOCH 8 done: loss 0.0138 - lr: 0.000036
211
+ 2023-10-12 17:31:35,933 DEV : loss 0.12251030653715134 - f1-score (micro avg) 0.8642
212
+ 2023-10-12 17:31:35,963 saving best model
213
+ 2023-10-12 17:31:36,938 ----------------------------------------------------------------------------------------------------
214
+ 2023-10-12 17:32:18,958 epoch 9 - iter 144/1445 - loss 0.00745630 - time (sec): 42.02 - samples/sec: 437.46 - lr: 0.000034 - momentum: 0.000000
215
+ 2023-10-12 17:33:00,161 epoch 9 - iter 288/1445 - loss 0.00761895 - time (sec): 83.22 - samples/sec: 422.78 - lr: 0.000032 - momentum: 0.000000
216
+ 2023-10-12 17:33:40,433 epoch 9 - iter 432/1445 - loss 0.00995441 - time (sec): 123.49 - samples/sec: 422.87 - lr: 0.000030 - momentum: 0.000000
217
+ 2023-10-12 17:34:21,456 epoch 9 - iter 576/1445 - loss 0.00880978 - time (sec): 164.52 - samples/sec: 423.26 - lr: 0.000028 - momentum: 0.000000
218
+ 2023-10-12 17:35:02,766 epoch 9 - iter 720/1445 - loss 0.00785433 - time (sec): 205.83 - samples/sec: 427.36 - lr: 0.000027 - momentum: 0.000000
219
+ 2023-10-12 17:35:43,814 epoch 9 - iter 864/1445 - loss 0.00778198 - time (sec): 246.87 - samples/sec: 427.54 - lr: 0.000025 - momentum: 0.000000
220
+ 2023-10-12 17:36:24,978 epoch 9 - iter 1008/1445 - loss 0.00817747 - time (sec): 288.04 - samples/sec: 430.09 - lr: 0.000023 - momentum: 0.000000
221
+ 2023-10-12 17:37:05,337 epoch 9 - iter 1152/1445 - loss 0.00847105 - time (sec): 328.40 - samples/sec: 429.48 - lr: 0.000021 - momentum: 0.000000
222
+ 2023-10-12 17:37:45,918 epoch 9 - iter 1296/1445 - loss 0.00923305 - time (sec): 368.98 - samples/sec: 426.79 - lr: 0.000020 - momentum: 0.000000
223
+ 2023-10-12 17:38:28,699 epoch 9 - iter 1440/1445 - loss 0.00900821 - time (sec): 411.76 - samples/sec: 425.30 - lr: 0.000018 - momentum: 0.000000
224
+ 2023-10-12 17:38:30,718 ----------------------------------------------------------------------------------------------------
225
+ 2023-10-12 17:38:30,718 EPOCH 9 done: loss 0.0096 - lr: 0.000018
226
+ 2023-10-12 17:38:52,760 DEV : loss 0.14077246189117432 - f1-score (micro avg) 0.8633
227
+ 2023-10-12 17:38:52,801 ----------------------------------------------------------------------------------------------------
228
+ 2023-10-12 17:39:37,018 epoch 10 - iter 144/1445 - loss 0.01174630 - time (sec): 44.21 - samples/sec: 426.17 - lr: 0.000016 - momentum: 0.000000
229
+ 2023-10-12 17:40:20,486 epoch 10 - iter 288/1445 - loss 0.01002501 - time (sec): 87.68 - samples/sec: 415.57 - lr: 0.000014 - momentum: 0.000000
230
+ 2023-10-12 17:41:04,188 epoch 10 - iter 432/1445 - loss 0.00910839 - time (sec): 131.39 - samples/sec: 407.70 - lr: 0.000012 - momentum: 0.000000
231
+ 2023-10-12 17:41:48,780 epoch 10 - iter 576/1445 - loss 0.00824494 - time (sec): 175.98 - samples/sec: 405.69 - lr: 0.000011 - momentum: 0.000000
232
+ 2023-10-12 17:42:33,468 epoch 10 - iter 720/1445 - loss 0.00769188 - time (sec): 220.67 - samples/sec: 404.45 - lr: 0.000009 - momentum: 0.000000
233
+ 2023-10-12 17:43:20,283 epoch 10 - iter 864/1445 - loss 0.00752313 - time (sec): 267.48 - samples/sec: 403.45 - lr: 0.000007 - momentum: 0.000000
234
+ 2023-10-12 17:44:01,105 epoch 10 - iter 1008/1445 - loss 0.00804398 - time (sec): 308.30 - samples/sec: 401.60 - lr: 0.000005 - momentum: 0.000000
235
+ 2023-10-12 17:44:43,927 epoch 10 - iter 1152/1445 - loss 0.00739720 - time (sec): 351.12 - samples/sec: 405.21 - lr: 0.000004 - momentum: 0.000000
236
+ 2023-10-12 17:45:25,058 epoch 10 - iter 1296/1445 - loss 0.00730540 - time (sec): 392.26 - samples/sec: 404.88 - lr: 0.000002 - momentum: 0.000000
237
+ 2023-10-12 17:46:07,507 epoch 10 - iter 1440/1445 - loss 0.00754475 - time (sec): 434.70 - samples/sec: 404.40 - lr: 0.000000 - momentum: 0.000000
238
+ 2023-10-12 17:46:08,714 ----------------------------------------------------------------------------------------------------
239
+ 2023-10-12 17:46:08,715 EPOCH 10 done: loss 0.0075 - lr: 0.000000
240
+ 2023-10-12 17:46:30,653 DEV : loss 0.14291653037071228 - f1-score (micro avg) 0.8604
241
+ 2023-10-12 17:46:31,574 ----------------------------------------------------------------------------------------------------
242
+ 2023-10-12 17:46:31,576 Loading model from best epoch ...
243
+ 2023-10-12 17:46:35,453 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
244
+ 2023-10-12 17:46:59,023
245
+ Results:
246
+ - F-score (micro) 0.851
247
+ - F-score (macro) 0.744
248
+ - Accuracy 0.7524
249
+
250
+ By class:
251
+ precision recall f1-score support
252
+
253
+ PER 0.8621 0.8693 0.8657 482
254
+ LOC 0.9277 0.8690 0.8974 458
255
+ ORG 0.4474 0.4928 0.4690 69
256
+
257
+ micro avg 0.8587 0.8434 0.8510 1009
258
+ macro avg 0.7457 0.7437 0.7440 1009
259
+ weighted avg 0.8636 0.8434 0.8530 1009
260
+
261
+ 2023-10-12 17:46:59,023 ----------------------------------------------------------------------------------------------------