Hugging Face's logo

hmbyt5-preliminary
/
flair-icdar-nl

Token Classification
Flair
PyTorch
TensorBoard
Dutch
sequence-tagger-model
Model card Files Metrics Community
2
stefan-it commited on
Commit
9df1844
•
1 Parent(s): 7d3fc51

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.1697051639.c8b2203b18a8.1914.4 +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 sha256:95bd9fb31d7d280fd56ae719a0b612cc218ab110e2fda4e28f603484fd8ad61e
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 sha256:190af2de64a4227c8ef9b5bf21ab9705d9e1c28e59cf354925d8be39584ec0cb
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 19:21:17 0.0001 1.1488 0.2165 0.6000 0.0062 0.0123 0.0062
3
+ 2 19:28:45 0.0001 0.1278 0.1181 0.7631 0.5888 0.6647 0.5022
4
+ 3 19:36:12 0.0001 0.0769 0.0809 0.8581 0.8182 0.8377 0.7354
5
+ 4 19:43:48 0.0001 0.0519 0.0700 0.8745 0.8709 0.8727 0.7879
6
+ 5 19:51:16 0.0001 0.0369 0.0790 0.8852 0.8368 0.8603 0.7656
7
+ 6 19:59:07 0.0001 0.0270 0.0944 0.8963 0.8213 0.8571 0.7600
8
+ 7 20:06:40 0.0001 0.0208 0.0988 0.8925 0.8492 0.8703 0.7821
9
+ 8 20:13:55 0.0000 0.0165 0.1050 0.8970 0.8543 0.8751 0.7876
10
+ 9 20:21:30 0.0000 0.0136 0.1160 0.8884 0.8306 0.8585 0.7643
11
+ 10 20:28:49 0.0000 0.0119 0.1147 0.8874 0.8388 0.8625 0.7711
runs/events.out.tfevents.1697051639.c8b2203b18a8.1914.4 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:418b420d35f5e9944b513b97bc782ff23cbc551afbf669b13f7996ef59c31943
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 19:13:59,349 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-11 19:13:59,352 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 19:13:59,352 ----------------------------------------------------------------------------------------------------
71
+ 2023-10-11 19:13:59,352 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 19:13:59,352 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-11 19:13:59,352 Train: 5777 sentences
75
+ 2023-10-11 19:13:59,352 (train_with_dev=False, train_with_test=False)
76
+ 2023-10-11 19:13:59,352 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-11 19:13:59,352 Training Params:
78
+ 2023-10-11 19:13:59,352 - learning_rate: "0.00015"
79
+ 2023-10-11 19:13:59,352 - mini_batch_size: "8"
80
+ 2023-10-11 19:13:59,353 - max_epochs: "10"
81
+ 2023-10-11 19:13:59,353 - shuffle: "True"
82
+ 2023-10-11 19:13:59,353 ----------------------------------------------------------------------------------------------------
83
+ 2023-10-11 19:13:59,353 Plugins:
84
+ 2023-10-11 19:13:59,353 - TensorboardLogger
85
+ 2023-10-11 19:13:59,353 - LinearScheduler | warmup_fraction: '0.1'
86
+ 2023-10-11 19:13:59,353 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-11 19:13:59,353 Final evaluation on model from best epoch (best-model.pt)
88
+ 2023-10-11 19:13:59,353 - metric: "('micro avg', 'f1-score')"
89
+ 2023-10-11 19:13:59,353 ----------------------------------------------------------------------------------------------------
90
+ 2023-10-11 19:13:59,353 Computation:
91
+ 2023-10-11 19:13:59,353 - compute on device: cuda:0
92
+ 2023-10-11 19:13:59,353 - embedding storage: none
93
+ 2023-10-11 19:13:59,353 ----------------------------------------------------------------------------------------------------
94
+ 2023-10-11 19:13:59,353 Model training base path: "hmbench-icdar/nl-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-2"
95
+ 2023-10-11 19:13:59,354 ----------------------------------------------------------------------------------------------------
96
+ 2023-10-11 19:13:59,354 ----------------------------------------------------------------------------------------------------
97
+ 2023-10-11 19:13:59,354 Logging anything other than scalars to TensorBoard is currently not supported.
98
+ 2023-10-11 19:14:44,985 epoch 1 - iter 72/723 - loss 2.57086872 - time (sec): 45.63 - samples/sec: 407.45 - lr: 0.000015 - momentum: 0.000000
99
+ 2023-10-11 19:15:23,902 epoch 1 - iter 144/723 - loss 2.53330840 - time (sec): 84.55 - samples/sec: 421.96 - lr: 0.000030 - momentum: 0.000000
100
+ 2023-10-11 19:16:07,025 epoch 1 - iter 216/723 - loss 2.37416485 - time (sec): 127.67 - samples/sec: 422.29 - lr: 0.000045 - momentum: 0.000000
101
+ 2023-10-11 19:16:48,488 epoch 1 - iter 288/723 - loss 2.16789438 - time (sec): 169.13 - samples/sec: 421.74 - lr: 0.000060 - momentum: 0.000000
102
+ 2023-10-11 19:17:29,619 epoch 1 - iter 360/723 - loss 1.94703601 - time (sec): 210.26 - samples/sec: 423.82 - lr: 0.000074 - momentum: 0.000000
103
+ 2023-10-11 19:18:12,121 epoch 1 - iter 432/723 - loss 1.72485908 - time (sec): 252.77 - samples/sec: 422.53 - lr: 0.000089 - momentum: 0.000000
104
+ 2023-10-11 19:18:54,982 epoch 1 - iter 504/723 - loss 1.52985339 - time (sec): 295.63 - samples/sec: 421.70 - lr: 0.000104 - momentum: 0.000000
105
+ 2023-10-11 19:19:35,885 epoch 1 - iter 576/723 - loss 1.37738975 - time (sec): 336.53 - samples/sec: 421.04 - lr: 0.000119 - momentum: 0.000000
106
+ 2023-10-11 19:20:16,171 epoch 1 - iter 648/723 - loss 1.25078985 - time (sec): 376.81 - samples/sec: 422.83 - lr: 0.000134 - momentum: 0.000000
107
+ 2023-10-11 19:20:56,215 epoch 1 - iter 720/723 - loss 1.15095694 - time (sec): 416.86 - samples/sec: 421.68 - lr: 0.000149 - momentum: 0.000000
108
+ 2023-10-11 19:20:57,395 ----------------------------------------------------------------------------------------------------
109
+ 2023-10-11 19:20:57,396 EPOCH 1 done: loss 1.1488 - lr: 0.000149
110
+ 2023-10-11 19:21:17,705 DEV : loss 0.21648679673671722 - f1-score (micro avg) 0.0123
111
+ 2023-10-11 19:21:17,740 saving best model
112
+ 2023-10-11 19:21:18,688 ----------------------------------------------------------------------------------------------------
113
+ 2023-10-11 19:22:01,303 epoch 2 - iter 72/723 - loss 0.16183662 - time (sec): 42.61 - samples/sec: 399.65 - lr: 0.000148 - momentum: 0.000000
114
+ 2023-10-11 19:22:42,923 epoch 2 - iter 144/723 - loss 0.16103628 - time (sec): 84.23 - samples/sec: 411.91 - lr: 0.000147 - momentum: 0.000000
115
+ 2023-10-11 19:23:25,027 epoch 2 - iter 216/723 - loss 0.15369470 - time (sec): 126.34 - samples/sec: 419.60 - lr: 0.000145 - momentum: 0.000000
116
+ 2023-10-11 19:24:10,544 epoch 2 - iter 288/723 - loss 0.14990715 - time (sec): 171.85 - samples/sec: 413.84 - lr: 0.000143 - momentum: 0.000000
117
+ 2023-10-11 19:24:54,049 epoch 2 - iter 360/723 - loss 0.14344494 - time (sec): 215.36 - samples/sec: 411.96 - lr: 0.000142 - momentum: 0.000000
118
+ 2023-10-11 19:25:35,378 epoch 2 - iter 432/723 - loss 0.14132544 - time (sec): 256.69 - samples/sec: 412.37 - lr: 0.000140 - momentum: 0.000000
119
+ 2023-10-11 19:26:17,204 epoch 2 - iter 504/723 - loss 0.13701095 - time (sec): 298.51 - samples/sec: 412.84 - lr: 0.000138 - momentum: 0.000000
120
+ 2023-10-11 19:26:59,167 epoch 2 - iter 576/723 - loss 0.13330675 - time (sec): 340.48 - samples/sec: 413.59 - lr: 0.000137 - momentum: 0.000000
121
+ 2023-10-11 19:27:42,057 epoch 2 - iter 648/723 - loss 0.13052442 - time (sec): 383.37 - samples/sec: 410.86 - lr: 0.000135 - momentum: 0.000000
122
+ 2023-10-11 19:28:23,642 epoch 2 - iter 720/723 - loss 0.12780591 - time (sec): 424.95 - samples/sec: 413.52 - lr: 0.000133 - momentum: 0.000000
123
+ 2023-10-11 19:28:24,842 ----------------------------------------------------------------------------------------------------
124
+ 2023-10-11 19:28:24,843 EPOCH 2 done: loss 0.1278 - lr: 0.000133
125
+ 2023-10-11 19:28:45,421 DEV : loss 0.1180637776851654 - f1-score (micro avg) 0.6647
126
+ 2023-10-11 19:28:45,451 saving best model
127
+ 2023-10-11 19:28:52,913 ----------------------------------------------------------------------------------------------------
128
+ 2023-10-11 19:29:33,850 epoch 3 - iter 72/723 - loss 0.10015265 - time (sec): 40.91 - samples/sec: 444.71 - lr: 0.000132 - momentum: 0.000000
129
+ 2023-10-11 19:30:13,514 epoch 3 - iter 144/723 - loss 0.08972155 - time (sec): 80.57 - samples/sec: 438.21 - lr: 0.000130 - momentum: 0.000000
130
+ 2023-10-11 19:30:52,539 epoch 3 - iter 216/723 - loss 0.08709046 - time (sec): 119.59 - samples/sec: 438.87 - lr: 0.000128 - momentum: 0.000000
131
+ 2023-10-11 19:31:33,908 epoch 3 - iter 288/723 - loss 0.08496174 - time (sec): 160.96 - samples/sec: 430.63 - lr: 0.000127 - momentum: 0.000000
132
+ 2023-10-11 19:32:13,347 epoch 3 - iter 360/723 - loss 0.08081415 - time (sec): 200.40 - samples/sec: 432.84 - lr: 0.000125 - momentum: 0.000000
133
+ 2023-10-11 19:32:54,772 epoch 3 - iter 432/723 - loss 0.08200482 - time (sec): 241.83 - samples/sec: 437.58 - lr: 0.000123 - momentum: 0.000000
134
+ 2023-10-11 19:33:38,370 epoch 3 - iter 504/723 - loss 0.08119768 - time (sec): 285.42 - samples/sec: 430.71 - lr: 0.000122 - momentum: 0.000000
135
+ 2023-10-11 19:34:22,240 epoch 3 - iter 576/723 - loss 0.07951784 - time (sec): 329.29 - samples/sec: 425.75 - lr: 0.000120 - momentum: 0.000000
136
+ 2023-10-11 19:35:04,706 epoch 3 - iter 648/723 - loss 0.07780987 - time (sec): 371.76 - samples/sec: 422.78 - lr: 0.000118 - momentum: 0.000000
137
+ 2023-10-11 19:35:48,845 epoch 3 - iter 720/723 - loss 0.07699412 - time (sec): 415.90 - samples/sec: 422.44 - lr: 0.000117 - momentum: 0.000000
138
+ 2023-10-11 19:35:50,187 ----------------------------------------------------------------------------------------------------
139
+ 2023-10-11 19:35:50,188 EPOCH 3 done: loss 0.0769 - lr: 0.000117
140
+ 2023-10-11 19:36:12,306 DEV : loss 0.08086864650249481 - f1-score (micro avg) 0.8377
141
+ 2023-10-11 19:36:12,336 saving best model
142
+ 2023-10-11 19:36:23,238 ----------------------------------------------------------------------------------------------------
143
+ 2023-10-11 19:37:04,812 epoch 4 - iter 72/723 - loss 0.06706830 - time (sec): 41.57 - samples/sec: 421.51 - lr: 0.000115 - momentum: 0.000000
144
+ 2023-10-11 19:37:47,107 epoch 4 - iter 144/723 - loss 0.05669850 - time (sec): 83.86 - samples/sec: 423.73 - lr: 0.000113 - momentum: 0.000000
145
+ 2023-10-11 19:38:29,958 epoch 4 - iter 216/723 - loss 0.05685534 - time (sec): 126.72 - samples/sec: 416.60 - lr: 0.000112 - momentum: 0.000000
146
+ 2023-10-11 19:39:11,998 epoch 4 - iter 288/723 - loss 0.05541904 - time (sec): 168.76 - samples/sec: 416.31 - lr: 0.000110 - momentum: 0.000000
147
+ 2023-10-11 19:39:54,241 epoch 4 - iter 360/723 - loss 0.05734549 - time (sec): 211.00 - samples/sec: 410.12 - lr: 0.000108 - momentum: 0.000000
148
+ 2023-10-11 19:40:37,348 epoch 4 - iter 432/723 - loss 0.05576876 - time (sec): 254.10 - samples/sec: 412.05 - lr: 0.000107 - momentum: 0.000000
149
+ 2023-10-11 19:41:20,058 epoch 4 - iter 504/723 - loss 0.05631673 - time (sec): 296.82 - samples/sec: 414.70 - lr: 0.000105 - momentum: 0.000000
150
+ 2023-10-11 19:42:01,084 epoch 4 - iter 576/723 - loss 0.05413009 - time (sec): 337.84 - samples/sec: 416.26 - lr: 0.000103 - momentum: 0.000000
151
+ 2023-10-11 19:42:44,112 epoch 4 - iter 648/723 - loss 0.05175789 - time (sec): 380.87 - samples/sec: 419.25 - lr: 0.000102 - momentum: 0.000000
152
+ 2023-10-11 19:43:24,791 epoch 4 - iter 720/723 - loss 0.05190640 - time (sec): 421.55 - samples/sec: 417.18 - lr: 0.000100 - momentum: 0.000000
153
+ 2023-10-11 19:43:26,016 ----------------------------------------------------------------------------------------------------
154
+ 2023-10-11 19:43:26,016 EPOCH 4 done: loss 0.0519 - lr: 0.000100
155
+ 2023-10-11 19:43:48,629 DEV : loss 0.07001111656427383 - f1-score (micro avg) 0.8727
156
+ 2023-10-11 19:43:48,665 saving best model
157
+ 2023-10-11 19:43:52,185 ----------------------------------------------------------------------------------------------------
158
+ 2023-10-11 19:44:35,248 epoch 5 - iter 72/723 - loss 0.03577280 - time (sec): 43.06 - samples/sec: 410.16 - lr: 0.000098 - momentum: 0.000000
159
+ 2023-10-11 19:45:15,914 epoch 5 - iter 144/723 - loss 0.03264401 - time (sec): 83.72 - samples/sec: 421.27 - lr: 0.000097 - momentum: 0.000000
160
+ 2023-10-11 19:45:57,621 epoch 5 - iter 216/723 - loss 0.03662028 - time (sec): 125.43 - samples/sec: 426.47 - lr: 0.000095 - momentum: 0.000000
161
+ 2023-10-11 19:46:40,431 epoch 5 - iter 288/723 - loss 0.03545328 - time (sec): 168.24 - samples/sec: 420.72 - lr: 0.000093 - momentum: 0.000000
162
+ 2023-10-11 19:47:22,294 epoch 5 - iter 360/723 - loss 0.03467577 - time (sec): 210.10 - samples/sec: 425.95 - lr: 0.000092 - momentum: 0.000000
163
+ 2023-10-11 19:48:03,651 epoch 5 - iter 432/723 - loss 0.03412429 - time (sec): 251.46 - samples/sec: 420.87 - lr: 0.000090 - momentum: 0.000000
164
+ 2023-10-11 19:48:47,048 epoch 5 - iter 504/723 - loss 0.03625783 - time (sec): 294.86 - samples/sec: 419.45 - lr: 0.000088 - momentum: 0.000000
165
+ 2023-10-11 19:49:28,818 epoch 5 - iter 576/723 - loss 0.03576595 - time (sec): 336.63 - samples/sec: 419.01 - lr: 0.000087 - momentum: 0.000000
166
+ 2023-10-11 19:50:09,866 epoch 5 - iter 648/723 - loss 0.03742931 - time (sec): 377.68 - samples/sec: 419.77 - lr: 0.000085 - momentum: 0.000000
167
+ 2023-10-11 19:50:50,937 epoch 5 - iter 720/723 - loss 0.03701053 - time (sec): 418.75 - samples/sec: 419.33 - lr: 0.000083 - momentum: 0.000000
168
+ 2023-10-11 19:50:52,310 ----------------------------------------------------------------------------------------------------
169
+ 2023-10-11 19:50:52,311 EPOCH 5 done: loss 0.0369 - lr: 0.000083
170
+ 2023-10-11 19:51:16,106 DEV : loss 0.07903970032930374 - f1-score (micro avg) 0.8603
171
+ 2023-10-11 19:51:16,143 ----------------------------------------------------------------------------------------------------
172
+ 2023-10-11 19:52:00,787 epoch 6 - iter 72/723 - loss 0.02447853 - time (sec): 44.64 - samples/sec: 416.76 - lr: 0.000082 - momentum: 0.000000
173
+ 2023-10-11 19:52:44,152 epoch 6 - iter 144/723 - loss 0.02205426 - time (sec): 88.01 - samples/sec: 418.68 - lr: 0.000080 - momentum: 0.000000
174
+ 2023-10-11 19:53:25,211 epoch 6 - iter 216/723 - loss 0.02750277 - time (sec): 129.07 - samples/sec: 410.91 - lr: 0.000078 - momentum: 0.000000
175
+ 2023-10-11 19:54:08,739 epoch 6 - iter 288/723 - loss 0.02591301 - time (sec): 172.59 - samples/sec: 408.80 - lr: 0.000077 - momentum: 0.000000
176
+ 2023-10-11 19:54:51,542 epoch 6 - iter 360/723 - loss 0.02561029 - time (sec): 215.40 - samples/sec: 403.07 - lr: 0.000075 - momentum: 0.000000
177
+ 2023-10-11 19:55:37,784 epoch 6 - iter 432/723 - loss 0.02497769 - time (sec): 261.64 - samples/sec: 400.75 - lr: 0.000073 - momentum: 0.000000
178
+ 2023-10-11 19:56:24,102 epoch 6 - iter 504/723 - loss 0.02676010 - time (sec): 307.96 - samples/sec: 401.09 - lr: 0.000072 - momentum: 0.000000
179
+ 2023-10-11 19:57:11,010 epoch 6 - iter 576/723 - loss 0.02633032 - time (sec): 354.86 - samples/sec: 398.37 - lr: 0.000070 - momentum: 0.000000
180
+ 2023-10-11 19:57:54,453 epoch 6 - iter 648/723 - loss 0.02681762 - time (sec): 398.31 - samples/sec: 396.04 - lr: 0.000068 - momentum: 0.000000
181
+ 2023-10-11 19:58:40,721 epoch 6 - iter 720/723 - loss 0.02707283 - time (sec): 444.58 - samples/sec: 394.53 - lr: 0.000067 - momentum: 0.000000
182
+ 2023-10-11 19:58:42,172 ----------------------------------------------------------------------------------------------------
183
+ 2023-10-11 19:58:42,172 EPOCH 6 done: loss 0.0270 - lr: 0.000067
184
+ 2023-10-11 19:59:07,335 DEV : loss 0.09442799538373947 - f1-score (micro avg) 0.8571
185
+ 2023-10-11 19:59:07,373 ----------------------------------------------------------------------------------------------------
186
+ 2023-10-11 19:59:52,725 epoch 7 - iter 72/723 - loss 0.02094561 - time (sec): 45.35 - samples/sec: 391.65 - lr: 0.000065 - momentum: 0.000000
187
+ 2023-10-11 20:00:33,300 epoch 7 - iter 144/723 - loss 0.01875934 - time (sec): 85.92 - samples/sec: 398.24 - lr: 0.000063 - momentum: 0.000000
188
+ 2023-10-11 20:01:14,107 epoch 7 - iter 216/723 - loss 0.01681267 - time (sec): 126.73 - samples/sec: 398.42 - lr: 0.000062 - momentum: 0.000000
189
+ 2023-10-11 20:01:57,563 epoch 7 - iter 288/723 - loss 0.01766153 - time (sec): 170.19 - samples/sec: 410.15 - lr: 0.000060 - momentum: 0.000000
190
+ 2023-10-11 20:02:43,950 epoch 7 - iter 360/723 - loss 0.02292811 - time (sec): 216.58 - samples/sec: 405.16 - lr: 0.000058 - momentum: 0.000000
191
+ 2023-10-11 20:03:27,340 epoch 7 - iter 432/723 - loss 0.02258017 - time (sec): 259.97 - samples/sec: 407.35 - lr: 0.000057 - momentum: 0.000000
192
+ 2023-10-11 20:04:08,335 epoch 7 - iter 504/723 - loss 0.02155383 - time (sec): 300.96 - samples/sec: 407.69 - lr: 0.000055 - momentum: 0.000000
193
+ 2023-10-11 20:04:50,397 epoch 7 - iter 576/723 - loss 0.02107573 - time (sec): 343.02 - samples/sec: 409.98 - lr: 0.000053 - momentum: 0.000000
194
+ 2023-10-11 20:05:32,273 epoch 7 - iter 648/723 - loss 0.02110528 - time (sec): 384.90 - samples/sec: 411.17 - lr: 0.000052 - momentum: 0.000000
195
+ 2023-10-11 20:06:15,185 epoch 7 - iter 720/723 - loss 0.02085093 - time (sec): 427.81 - samples/sec: 410.19 - lr: 0.000050 - momentum: 0.000000
196
+ 2023-10-11 20:06:16,648 ----------------------------------------------------------------------------------------------------
197
+ 2023-10-11 20:06:16,649 EPOCH 7 done: loss 0.0208 - lr: 0.000050
198
+ 2023-10-11 20:06:40,108 DEV : loss 0.09876430779695511 - f1-score (micro avg) 0.8703
199
+ 2023-10-11 20:06:40,152 ----------------------------------------------------------------------------------------------------
200
+ 2023-10-11 20:07:22,836 epoch 8 - iter 72/723 - loss 0.00933933 - time (sec): 42.68 - samples/sec: 420.64 - lr: 0.000048 - momentum: 0.000000
201
+ 2023-10-11 20:08:03,853 epoch 8 - iter 144/723 - loss 0.01292328 - time (sec): 83.70 - samples/sec: 408.45 - lr: 0.000047 - momentum: 0.000000
202
+ 2023-10-11 20:08:43,013 epoch 8 - iter 216/723 - loss 0.01314725 - time (sec): 122.86 - samples/sec: 413.22 - lr: 0.000045 - momentum: 0.000000
203
+ 2023-10-11 20:09:23,514 epoch 8 - iter 288/723 - loss 0.01277868 - time (sec): 163.36 - samples/sec: 415.17 - lr: 0.000043 - momentum: 0.000000
204
+ 2023-10-11 20:10:04,265 epoch 8 - iter 360/723 - loss 0.01439634 - time (sec): 204.11 - samples/sec: 418.58 - lr: 0.000042 - momentum: 0.000000
205
+ 2023-10-11 20:10:45,239 epoch 8 - iter 432/723 - loss 0.01543636 - time (sec): 245.09 - samples/sec: 423.69 - lr: 0.000040 - momentum: 0.000000
206
+ 2023-10-11 20:11:26,857 epoch 8 - iter 504/723 - loss 0.01613212 - time (sec): 286.70 - samples/sec: 427.85 - lr: 0.000038 - momentum: 0.000000
207
+ 2023-10-11 20:12:06,133 epoch 8 - iter 576/723 - loss 0.01588643 - time (sec): 325.98 - samples/sec: 429.05 - lr: 0.000037 - momentum: 0.000000
208
+ 2023-10-11 20:12:48,288 epoch 8 - iter 648/723 - loss 0.01595246 - time (sec): 368.13 - samples/sec: 429.70 - lr: 0.000035 - momentum: 0.000000
209
+ 2023-10-11 20:13:31,496 epoch 8 - iter 720/723 - loss 0.01649749 - time (sec): 411.34 - samples/sec: 427.11 - lr: 0.000033 - momentum: 0.000000
210
+ 2023-10-11 20:13:32,739 ----------------------------------------------------------------------------------------------------
211
+ 2023-10-11 20:13:32,740 EPOCH 8 done: loss 0.0165 - lr: 0.000033
212
+ 2023-10-11 20:13:55,278 DEV : loss 0.10501116514205933 - f1-score (micro avg) 0.8751
213
+ 2023-10-11 20:13:55,327 saving best model
214
+ 2023-10-11 20:13:56,488 ----------------------------------------------------------------------------------------------------
215
+ 2023-10-11 20:14:41,953 epoch 9 - iter 72/723 - loss 0.00692157 - time (sec): 45.46 - samples/sec: 377.43 - lr: 0.000032 - momentum: 0.000000
216
+ 2023-10-11 20:15:23,711 epoch 9 - iter 144/723 - loss 0.00987217 - time (sec): 87.22 - samples/sec: 390.20 - lr: 0.000030 - momentum: 0.000000
217
+ 2023-10-11 20:16:06,341 epoch 9 - iter 216/723 - loss 0.01091573 - time (sec): 129.85 - samples/sec: 394.00 - lr: 0.000028 - momentum: 0.000000
218
+ 2023-10-11 20:16:50,893 epoch 9 - iter 288/723 - loss 0.01283372 - time (sec): 174.40 - samples/sec: 400.17 - lr: 0.000027 - momentum: 0.000000
219
+ 2023-10-11 20:17:36,364 epoch 9 - iter 360/723 - loss 0.01390500 - time (sec): 219.87 - samples/sec: 402.90 - lr: 0.000025 - momentum: 0.000000
220
+ 2023-10-11 20:18:19,862 epoch 9 - iter 432/723 - loss 0.01345060 - time (sec): 263.37 - samples/sec: 403.65 - lr: 0.000023 - momentum: 0.000000
221
+ 2023-10-11 20:19:02,380 epoch 9 - iter 504/723 - loss 0.01441546 - time (sec): 305.89 - samples/sec: 405.73 - lr: 0.000022 - momentum: 0.000000
222
+ 2023-10-11 20:19:45,060 epoch 9 - iter 576/723 - loss 0.01394066 - time (sec): 348.57 - samples/sec: 406.20 - lr: 0.000020 - momentum: 0.000000
223
+ 2023-10-11 20:20:25,766 epoch 9 - iter 648/723 - loss 0.01417785 - time (sec): 389.28 - samples/sec: 407.17 - lr: 0.000018 - momentum: 0.000000
224
+ 2023-10-11 20:21:06,100 epoch 9 - iter 720/723 - loss 0.01360125 - time (sec): 429.61 - samples/sec: 408.77 - lr: 0.000017 - momentum: 0.000000
225
+ 2023-10-11 20:21:07,399 ----------------------------------------------------------------------------------------------------
226
+ 2023-10-11 20:21:07,399 EPOCH 9 done: loss 0.0136 - lr: 0.000017
227
+ 2023-10-11 20:21:30,023 DEV : loss 0.11595697700977325 - f1-score (micro avg) 0.8585
228
+ 2023-10-11 20:21:30,064 ----------------------------------------------------------------------------------------------------
229
+ 2023-10-11 20:22:12,604 epoch 10 - iter 72/723 - loss 0.01406246 - time (sec): 42.54 - samples/sec: 423.37 - lr: 0.000015 - momentum: 0.000000
230
+ 2023-10-11 20:22:55,515 epoch 10 - iter 144/723 - loss 0.01245519 - time (sec): 85.45 - samples/sec: 424.31 - lr: 0.000013 - momentum: 0.000000
231
+ 2023-10-11 20:23:38,829 epoch 10 - iter 216/723 - loss 0.01322961 - time (sec): 128.76 - samples/sec: 430.22 - lr: 0.000012 - momentum: 0.000000
232
+ 2023-10-11 20:24:20,473 epoch 10 - iter 288/723 - loss 0.01274717 - time (sec): 170.41 - samples/sec: 428.27 - lr: 0.000010 - momentum: 0.000000
233
+ 2023-10-11 20:25:01,572 epoch 10 - iter 360/723 - loss 0.01269929 - time (sec): 211.51 - samples/sec: 430.74 - lr: 0.000008 - momentum: 0.000000
234
+ 2023-10-11 20:25:43,691 epoch 10 - iter 432/723 - loss 0.01264156 - time (sec): 253.62 - samples/sec: 423.44 - lr: 0.000007 - momentum: 0.000000
235
+ 2023-10-11 20:26:25,384 epoch 10 - iter 504/723 - loss 0.01254290 - time (sec): 295.32 - samples/sec: 421.37 - lr: 0.000005 - momentum: 0.000000
236
+ 2023-10-11 20:27:06,979 epoch 10 - iter 576/723 - loss 0.01252330 - time (sec): 336.91 - samples/sec: 420.64 - lr: 0.000003 - momentum: 0.000000
237
+ 2023-10-11 20:27:46,703 epoch 10 - iter 648/723 - loss 0.01221020 - time (sec): 376.64 - samples/sec: 421.88 - lr: 0.000002 - momentum: 0.000000
238
+ 2023-10-11 20:28:26,795 epoch 10 - iter 720/723 - loss 0.01195603 - time (sec): 416.73 - samples/sec: 421.67 - lr: 0.000000 - momentum: 0.000000
239
+ 2023-10-11 20:28:27,946 ----------------------------------------------------------------------------------------------------
240
+ 2023-10-11 20:28:27,946 EPOCH 10 done: loss 0.0119 - lr: 0.000000
241
+ 2023-10-11 20:28:49,635 DEV : loss 0.11474814265966415 - f1-score (micro avg) 0.8625
242
+ 2023-10-11 20:28:50,709 ----------------------------------------------------------------------------------------------------
243
+ 2023-10-11 20:28:50,711 Loading model from best epoch ...
244
+ 2023-10-11 20:28:55,094 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 20:29:19,250
246
+ Results:
247
+ - F-score (micro) 0.8422
248
+ - F-score (macro) 0.7359
249
+ - Accuracy 0.7429
250
+
251
+ By class:
252
+ precision recall f1-score support
253
+
254
+ PER 0.8480 0.8797 0.8635 482
255
+ LOC 0.9159 0.8319 0.8719 458
256
+ ORG 0.5172 0.4348 0.4724 69
257
+
258
+ micro avg 0.8573 0.8276 0.8422 1009
259
+ macro avg 0.7604 0.7154 0.7359 1009
260
+ weighted avg 0.8562 0.8276 0.8406 1009
261
+
262
+ 2023-10-11 20:29:19,250 ----------------------------------------------------------------------------------------------------