Hugging Face's logo

stefan-it
/
hmbench-icdar-fr-hmbyt5-bs8-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-1

Token Classification
Flair
PyTorch
TensorBoard
French
sequence-tagger-model
Model card Files Metrics Community
2
stefan-it commited on
Commit
7733fc7
•
1 Parent(s): 519fe4e

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.1697103327.de2e83fddbee.1952.1 +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 sha256:5e560dc88807a57a870bb732ec2da0af4c922d8f0cda3b71adc1a550bc46a38a
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 sha256:154a853357d3ef55b825e03b09279358e0403ce34035823dcbe7935a6cb86312
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 09:44:01 0.0002 1.0274 0.1744 0.3156 0.3552 0.3342 0.2373
3
+ 2 09:52:45 0.0001 0.1506 0.0922 0.6882 0.7115 0.6997 0.5656
4
+ 3 10:02:21 0.0001 0.0843 0.0892 0.6984 0.7885 0.7407 0.6077
5
+ 4 10:10:54 0.0001 0.0560 0.1052 0.7508 0.7873 0.7686 0.6462
6
+ 5 10:19:46 0.0001 0.0413 0.1187 0.7371 0.7896 0.7624 0.6380
7
+ 6 10:29:00 0.0001 0.0313 0.1380 0.7432 0.7726 0.7576 0.6289
8
+ 7 10:38:22 0.0001 0.0237 0.1812 0.7309 0.7681 0.7490 0.6184
9
+ 8 10:47:29 0.0000 0.0189 0.1817 0.7269 0.7828 0.7538 0.6251
10
+ 9 10:56:32 0.0000 0.0152 0.1933 0.7357 0.7873 0.7607 0.6339
11
+ 10 11:05:20 0.0000 0.0129 0.1981 0.7377 0.7828 0.7596 0.6325
runs/events.out.tfevents.1697103327.de2e83fddbee.1952.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f383e12b39ab9f370ac6c434b8b31d46bded69e17d6af2e9083521d313e5b15f
3
+ size 556612
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 09:35:27,482 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-12 09:35:27,484 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 09:35:27,484 ----------------------------------------------------------------------------------------------------
71
+ 2023-10-12 09:35:27,484 MultiCorpus: 7936 train + 992 dev + 992 test sentences
72
+ - NER_ICDAR_EUROPEANA Corpus: 7936 train + 992 dev + 992 test sentences - /root/.flair/datasets/ner_icdar_europeana/fr
73
+ 2023-10-12 09:35:27,484 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-12 09:35:27,485 Train: 7936 sentences
75
+ 2023-10-12 09:35:27,485 (train_with_dev=False, train_with_test=False)
76
+ 2023-10-12 09:35:27,485 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-12 09:35:27,485 Training Params:
78
+ 2023-10-12 09:35:27,485 - learning_rate: "0.00016"
79
+ 2023-10-12 09:35:27,485 - mini_batch_size: "8"
80
+ 2023-10-12 09:35:27,485 - max_epochs: "10"
81
+ 2023-10-12 09:35:27,485 - shuffle: "True"
82
+ 2023-10-12 09:35:27,485 ----------------------------------------------------------------------------------------------------
83
+ 2023-10-12 09:35:27,485 Plugins:
84
+ 2023-10-12 09:35:27,485 - TensorboardLogger
85
+ 2023-10-12 09:35:27,485 - LinearScheduler | warmup_fraction: '0.1'
86
+ 2023-10-12 09:35:27,485 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-12 09:35:27,485 Final evaluation on model from best epoch (best-model.pt)
88
+ 2023-10-12 09:35:27,485 - metric: "('micro avg', 'f1-score')"
89
+ 2023-10-12 09:35:27,486 ----------------------------------------------------------------------------------------------------
90
+ 2023-10-12 09:35:27,486 Computation:
91
+ 2023-10-12 09:35:27,486 - compute on device: cuda:0
92
+ 2023-10-12 09:35:27,486 - embedding storage: none
93
+ 2023-10-12 09:35:27,486 ----------------------------------------------------------------------------------------------------
94
+ 2023-10-12 09:35:27,486 Model training base path: "hmbench-icdar/fr-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-1"
95
+ 2023-10-12 09:35:27,486 ----------------------------------------------------------------------------------------------------
96
+ 2023-10-12 09:35:27,486 ----------------------------------------------------------------------------------------------------
97
+ 2023-10-12 09:35:27,486 Logging anything other than scalars to TensorBoard is currently not supported.
98
+ 2023-10-12 09:36:15,430 epoch 1 - iter 99/992 - loss 2.58479086 - time (sec): 47.94 - samples/sec: 322.62 - lr: 0.000016 - momentum: 0.000000
99
+ 2023-10-12 09:37:05,745 epoch 1 - iter 198/992 - loss 2.52870939 - time (sec): 98.26 - samples/sec: 320.85 - lr: 0.000032 - momentum: 0.000000
100
+ 2023-10-12 09:37:55,606 epoch 1 - iter 297/992 - loss 2.31835059 - time (sec): 148.12 - samples/sec: 323.43 - lr: 0.000048 - momentum: 0.000000
101
+ 2023-10-12 09:38:44,520 epoch 1 - iter 396/992 - loss 2.04948183 - time (sec): 197.03 - samples/sec: 326.45 - lr: 0.000064 - momentum: 0.000000
102
+ 2023-10-12 09:39:33,893 epoch 1 - iter 495/992 - loss 1.78919303 - time (sec): 246.40 - samples/sec: 325.79 - lr: 0.000080 - momentum: 0.000000
103
+ 2023-10-12 09:40:23,131 epoch 1 - iter 594/992 - loss 1.55565919 - time (sec): 295.64 - samples/sec: 328.11 - lr: 0.000096 - momentum: 0.000000
104
+ 2023-10-12 09:41:12,279 epoch 1 - iter 693/992 - loss 1.36235321 - time (sec): 344.79 - samples/sec: 331.68 - lr: 0.000112 - momentum: 0.000000
105
+ 2023-10-12 09:41:59,986 epoch 1 - iter 792/992 - loss 1.22392122 - time (sec): 392.50 - samples/sec: 333.92 - lr: 0.000128 - momentum: 0.000000
106
+ 2023-10-12 09:42:46,949 epoch 1 - iter 891/992 - loss 1.11828998 - time (sec): 439.46 - samples/sec: 335.27 - lr: 0.000144 - momentum: 0.000000
107
+ 2023-10-12 09:43:36,228 epoch 1 - iter 990/992 - loss 1.02891630 - time (sec): 488.74 - samples/sec: 334.90 - lr: 0.000160 - momentum: 0.000000
108
+ 2023-10-12 09:43:37,265 ----------------------------------------------------------------------------------------------------
109
+ 2023-10-12 09:43:37,265 EPOCH 1 done: loss 1.0274 - lr: 0.000160
110
+ 2023-10-12 09:44:01,623 DEV : loss 0.1744040548801422 - f1-score (micro avg) 0.3342
111
+ 2023-10-12 09:44:01,663 saving best model
112
+ 2023-10-12 09:44:02,545 ----------------------------------------------------------------------------------------------------
113
+ 2023-10-12 09:44:50,545 epoch 2 - iter 99/992 - loss 0.22753274 - time (sec): 48.00 - samples/sec: 343.18 - lr: 0.000158 - momentum: 0.000000
114
+ 2023-10-12 09:45:38,087 epoch 2 - iter 198/992 - loss 0.19608680 - time (sec): 95.54 - samples/sec: 342.17 - lr: 0.000156 - momentum: 0.000000
115
+ 2023-10-12 09:46:26,577 epoch 2 - iter 297/992 - loss 0.18358912 - time (sec): 144.03 - samples/sec: 340.53 - lr: 0.000155 - momentum: 0.000000
116
+ 2023-10-12 09:47:15,707 epoch 2 - iter 396/992 - loss 0.17888904 - time (sec): 193.16 - samples/sec: 340.25 - lr: 0.000153 - momentum: 0.000000
117
+ 2023-10-12 09:48:02,749 epoch 2 - iter 495/992 - loss 0.17228318 - time (sec): 240.20 - samples/sec: 342.93 - lr: 0.000151 - momentum: 0.000000
118
+ 2023-10-12 09:48:49,148 epoch 2 - iter 594/992 - loss 0.16801476 - time (sec): 286.60 - samples/sec: 343.83 - lr: 0.000149 - momentum: 0.000000
119
+ 2023-10-12 09:49:36,285 epoch 2 - iter 693/992 - loss 0.16526240 - time (sec): 333.74 - samples/sec: 344.83 - lr: 0.000148 - momentum: 0.000000
120
+ 2023-10-12 09:50:26,071 epoch 2 - iter 792/992 - loss 0.15957310 - time (sec): 383.52 - samples/sec: 341.06 - lr: 0.000146 - momentum: 0.000000
121
+ 2023-10-12 09:51:20,596 epoch 2 - iter 891/992 - loss 0.15487382 - time (sec): 438.05 - samples/sec: 335.55 - lr: 0.000144 - momentum: 0.000000
122
+ 2023-10-12 09:52:15,328 epoch 2 - iter 990/992 - loss 0.15093391 - time (sec): 492.78 - samples/sec: 331.80 - lr: 0.000142 - momentum: 0.000000
123
+ 2023-10-12 09:52:16,520 ----------------------------------------------------------------------------------------------------
124
+ 2023-10-12 09:52:16,520 EPOCH 2 done: loss 0.1506 - lr: 0.000142
125
+ 2023-10-12 09:52:45,153 DEV : loss 0.09215505421161652 - f1-score (micro avg) 0.6997
126
+ 2023-10-12 09:52:45,206 saving best model
127
+ 2023-10-12 09:52:53,862 ----------------------------------------------------------------------------------------------------
128
+ 2023-10-12 09:53:51,412 epoch 3 - iter 99/992 - loss 0.09122045 - time (sec): 57.54 - samples/sec: 273.51 - lr: 0.000140 - momentum: 0.000000
129
+ 2023-10-12 09:54:44,654 epoch 3 - iter 198/992 - loss 0.09329663 - time (sec): 110.79 - samples/sec: 288.45 - lr: 0.000139 - momentum: 0.000000
130
+ 2023-10-12 09:55:42,468 epoch 3 - iter 297/992 - loss 0.09230889 - time (sec): 168.60 - samples/sec: 289.46 - lr: 0.000137 - momentum: 0.000000
131
+ 2023-10-12 09:56:37,292 epoch 3 - iter 396/992 - loss 0.09112678 - time (sec): 223.42 - samples/sec: 291.75 - lr: 0.000135 - momentum: 0.000000
132
+ 2023-10-12 09:57:29,861 epoch 3 - iter 495/992 - loss 0.08931256 - time (sec): 275.99 - samples/sec: 294.90 - lr: 0.000133 - momentum: 0.000000
133
+ 2023-10-12 09:58:26,690 epoch 3 - iter 594/992 - loss 0.08792155 - time (sec): 332.82 - samples/sec: 292.83 - lr: 0.000132 - momentum: 0.000000
134
+ 2023-10-12 09:59:24,075 epoch 3 - iter 693/992 - loss 0.08786202 - time (sec): 390.21 - samples/sec: 290.12 - lr: 0.000130 - momentum: 0.000000
135
+ 2023-10-12 10:00:15,500 epoch 3 - iter 792/992 - loss 0.08568713 - time (sec): 441.63 - samples/sec: 296.32 - lr: 0.000128 - momentum: 0.000000
136
+ 2023-10-12 10:01:05,466 epoch 3 - iter 891/992 - loss 0.08435593 - time (sec): 491.60 - samples/sec: 300.53 - lr: 0.000126 - momentum: 0.000000
137
+ 2023-10-12 10:01:52,852 epoch 3 - iter 990/992 - loss 0.08429677 - time (sec): 538.98 - samples/sec: 303.70 - lr: 0.000125 - momentum: 0.000000
138
+ 2023-10-12 10:01:53,890 ----------------------------------------------------------------------------------------------------
139
+ 2023-10-12 10:01:53,890 EPOCH 3 done: loss 0.0843 - lr: 0.000125
140
+ 2023-10-12 10:02:21,183 DEV : loss 0.08918626606464386 - f1-score (micro avg) 0.7407
141
+ 2023-10-12 10:02:21,233 saving best model
142
+ 2023-10-12 10:02:23,914 ----------------------------------------------------------------------------------------------------
143
+ 2023-10-12 10:03:16,084 epoch 4 - iter 99/992 - loss 0.05874315 - time (sec): 52.17 - samples/sec: 328.59 - lr: 0.000123 - momentum: 0.000000
144
+ 2023-10-12 10:04:07,859 epoch 4 - iter 198/992 - loss 0.05797209 - time (sec): 103.94 - samples/sec: 327.68 - lr: 0.000121 - momentum: 0.000000
145
+ 2023-10-12 10:04:56,031 epoch 4 - iter 297/992 - loss 0.06035790 - time (sec): 152.11 - samples/sec: 328.57 - lr: 0.000119 - momentum: 0.000000
146
+ 2023-10-12 10:05:43,678 epoch 4 - iter 396/992 - loss 0.05838440 - time (sec): 199.76 - samples/sec: 328.96 - lr: 0.000117 - momentum: 0.000000
147
+ 2023-10-12 10:06:31,630 epoch 4 - iter 495/992 - loss 0.05780847 - time (sec): 247.71 - samples/sec: 330.93 - lr: 0.000116 - momentum: 0.000000
148
+ 2023-10-12 10:07:19,579 epoch 4 - iter 594/992 - loss 0.05754794 - time (sec): 295.66 - samples/sec: 331.61 - lr: 0.000114 - momentum: 0.000000
149
+ 2023-10-12 10:08:06,828 epoch 4 - iter 693/992 - loss 0.05712259 - time (sec): 342.91 - samples/sec: 334.16 - lr: 0.000112 - momentum: 0.000000
150
+ 2023-10-12 10:08:54,022 epoch 4 - iter 792/992 - loss 0.05741937 - time (sec): 390.10 - samples/sec: 335.15 - lr: 0.000110 - momentum: 0.000000
151
+ 2023-10-12 10:09:42,447 epoch 4 - iter 891/992 - loss 0.05589883 - time (sec): 438.53 - samples/sec: 336.91 - lr: 0.000109 - momentum: 0.000000
152
+ 2023-10-12 10:10:28,979 epoch 4 - iter 990/992 - loss 0.05609350 - time (sec): 485.06 - samples/sec: 337.58 - lr: 0.000107 - momentum: 0.000000
153
+ 2023-10-12 10:10:29,864 ----------------------------------------------------------------------------------------------------
154
+ 2023-10-12 10:10:29,865 EPOCH 4 done: loss 0.0560 - lr: 0.000107
155
+ 2023-10-12 10:10:54,352 DEV : loss 0.10516904294490814 - f1-score (micro avg) 0.7686
156
+ 2023-10-12 10:10:54,392 saving best model
157
+ 2023-10-12 10:10:56,993 ----------------------------------------------------------------------------------------------------
158
+ 2023-10-12 10:11:51,597 epoch 5 - iter 99/992 - loss 0.04339909 - time (sec): 54.60 - samples/sec: 295.88 - lr: 0.000105 - momentum: 0.000000
159
+ 2023-10-12 10:12:41,528 epoch 5 - iter 198/992 - loss 0.03593089 - time (sec): 104.53 - samples/sec: 309.37 - lr: 0.000103 - momentum: 0.000000
160
+ 2023-10-12 10:13:29,901 epoch 5 - iter 297/992 - loss 0.03787890 - time (sec): 152.90 - samples/sec: 318.18 - lr: 0.000101 - momentum: 0.000000
161
+ 2023-10-12 10:14:18,612 epoch 5 - iter 396/992 - loss 0.03841376 - time (sec): 201.61 - samples/sec: 322.88 - lr: 0.000100 - momentum: 0.000000
162
+ 2023-10-12 10:15:09,736 epoch 5 - iter 495/992 - loss 0.03858312 - time (sec): 252.74 - samples/sec: 321.68 - lr: 0.000098 - momentum: 0.000000
163
+ 2023-10-12 10:16:02,866 epoch 5 - iter 594/992 - loss 0.04020965 - time (sec): 305.87 - samples/sec: 320.15 - lr: 0.000096 - momentum: 0.000000
164
+ 2023-10-12 10:16:51,758 epoch 5 - iter 693/992 - loss 0.04044582 - time (sec): 354.76 - samples/sec: 322.58 - lr: 0.000094 - momentum: 0.000000
165
+ 2023-10-12 10:17:41,098 epoch 5 - iter 792/992 - loss 0.04037223 - time (sec): 404.10 - samples/sec: 325.00 - lr: 0.000093 - momentum: 0.000000
166
+ 2023-10-12 10:18:30,954 epoch 5 - iter 891/992 - loss 0.04070099 - time (sec): 453.96 - samples/sec: 325.68 - lr: 0.000091 - momentum: 0.000000
167
+ 2023-10-12 10:19:19,136 epoch 5 - iter 990/992 - loss 0.04132693 - time (sec): 502.14 - samples/sec: 325.85 - lr: 0.000089 - momentum: 0.000000
168
+ 2023-10-12 10:19:20,111 ----------------------------------------------------------------------------------------------------
169
+ 2023-10-12 10:19:20,111 EPOCH 5 done: loss 0.0413 - lr: 0.000089
170
+ 2023-10-12 10:19:46,676 DEV : loss 0.11871492117643356 - f1-score (micro avg) 0.7624
171
+ 2023-10-12 10:19:46,717 ----------------------------------------------------------------------------------------------------
172
+ 2023-10-12 10:20:35,147 epoch 6 - iter 99/992 - loss 0.02568349 - time (sec): 48.43 - samples/sec: 323.47 - lr: 0.000087 - momentum: 0.000000
173
+ 2023-10-12 10:21:29,969 epoch 6 - iter 198/992 - loss 0.02791147 - time (sec): 103.25 - samples/sec: 310.13 - lr: 0.000085 - momentum: 0.000000
174
+ 2023-10-12 10:22:20,524 epoch 6 - iter 297/992 - loss 0.02795523 - time (sec): 153.80 - samples/sec: 313.23 - lr: 0.000084 - momentum: 0.000000
175
+ 2023-10-12 10:23:15,727 epoch 6 - iter 396/992 - loss 0.02853544 - time (sec): 209.01 - samples/sec: 311.81 - lr: 0.000082 - momentum: 0.000000
176
+ 2023-10-12 10:24:10,797 epoch 6 - iter 495/992 - loss 0.02779838 - time (sec): 264.08 - samples/sec: 307.25 - lr: 0.000080 - momentum: 0.000000
177
+ 2023-10-12 10:25:04,606 epoch 6 - iter 594/992 - loss 0.02783298 - time (sec): 317.89 - samples/sec: 307.75 - lr: 0.000078 - momentum: 0.000000
178
+ 2023-10-12 10:25:57,328 epoch 6 - iter 693/992 - loss 0.02926439 - time (sec): 370.61 - samples/sec: 309.48 - lr: 0.000077 - momentum: 0.000000
179
+ 2023-10-12 10:26:49,073 epoch 6 - iter 792/992 - loss 0.03085251 - time (sec): 422.35 - samples/sec: 309.52 - lr: 0.000075 - momentum: 0.000000
180
+ 2023-10-12 10:27:40,834 epoch 6 - iter 891/992 - loss 0.03137567 - time (sec): 474.12 - samples/sec: 310.75 - lr: 0.000073 - momentum: 0.000000
181
+ 2023-10-12 10:28:32,177 epoch 6 - iter 990/992 - loss 0.03128091 - time (sec): 525.46 - samples/sec: 311.37 - lr: 0.000071 - momentum: 0.000000
182
+ 2023-10-12 10:28:33,245 ----------------------------------------------------------------------------------------------------
183
+ 2023-10-12 10:28:33,245 EPOCH 6 done: loss 0.0313 - lr: 0.000071
184
+ 2023-10-12 10:29:00,108 DEV : loss 0.13804303109645844 - f1-score (micro avg) 0.7576
185
+ 2023-10-12 10:29:00,151 ----------------------------------------------------------------------------------------------------
186
+ 2023-10-12 10:29:54,368 epoch 7 - iter 99/992 - loss 0.01668401 - time (sec): 54.21 - samples/sec: 300.78 - lr: 0.000069 - momentum: 0.000000
187
+ 2023-10-12 10:30:54,123 epoch 7 - iter 198/992 - loss 0.02161418 - time (sec): 113.97 - samples/sec: 289.29 - lr: 0.000068 - momentum: 0.000000
188
+ 2023-10-12 10:31:45,039 epoch 7 - iter 297/992 - loss 0.02234921 - time (sec): 164.89 - samples/sec: 296.81 - lr: 0.000066 - momentum: 0.000000
189
+ 2023-10-12 10:32:41,367 epoch 7 - iter 396/992 - loss 0.02304129 - time (sec): 221.21 - samples/sec: 296.76 - lr: 0.000064 - momentum: 0.000000
190
+ 2023-10-12 10:33:33,523 epoch 7 - iter 495/992 - loss 0.02267395 - time (sec): 273.37 - samples/sec: 298.35 - lr: 0.000062 - momentum: 0.000000
191
+ 2023-10-12 10:34:22,325 epoch 7 - iter 594/992 - loss 0.02287761 - time (sec): 322.17 - samples/sec: 303.73 - lr: 0.000061 - momentum: 0.000000
192
+ 2023-10-12 10:35:13,730 epoch 7 - iter 693/992 - loss 0.02354855 - time (sec): 373.58 - samples/sec: 307.18 - lr: 0.000059 - momentum: 0.000000
193
+ 2023-10-12 10:36:05,992 epoch 7 - iter 792/992 - loss 0.02437092 - time (sec): 425.84 - samples/sec: 304.43 - lr: 0.000057 - momentum: 0.000000
194
+ 2023-10-12 10:36:56,191 epoch 7 - iter 891/992 - loss 0.02374473 - time (sec): 476.04 - samples/sec: 307.35 - lr: 0.000055 - momentum: 0.000000
195
+ 2023-10-12 10:37:49,617 epoch 7 - iter 990/992 - loss 0.02363874 - time (sec): 529.46 - samples/sec: 309.00 - lr: 0.000053 - momentum: 0.000000
196
+ 2023-10-12 10:37:50,718 ----------------------------------------------------------------------------------------------------
197
+ 2023-10-12 10:37:50,719 EPOCH 7 done: loss 0.0237 - lr: 0.000053
198
+ 2023-10-12 10:38:22,287 DEV : loss 0.1811920553445816 - f1-score (micro avg) 0.749
199
+ 2023-10-12 10:38:22,339 ----------------------------------------------------------------------------------------------------
200
+ 2023-10-12 10:39:16,794 epoch 8 - iter 99/992 - loss 0.02146333 - time (sec): 54.45 - samples/sec: 310.86 - lr: 0.000052 - momentum: 0.000000
201
+ 2023-10-12 10:40:06,997 epoch 8 - iter 198/992 - loss 0.01992038 - time (sec): 104.66 - samples/sec: 308.15 - lr: 0.000050 - momentum: 0.000000
202
+ 2023-10-12 10:40:54,436 epoch 8 - iter 297/992 - loss 0.02020086 - time (sec): 152.09 - samples/sec: 313.96 - lr: 0.000048 - momentum: 0.000000
203
+ 2023-10-12 10:41:44,381 epoch 8 - iter 396/992 - loss 0.01943430 - time (sec): 202.04 - samples/sec: 315.64 - lr: 0.000046 - momentum: 0.000000
204
+ 2023-10-12 10:42:34,397 epoch 8 - iter 495/992 - loss 0.01928764 - time (sec): 252.06 - samples/sec: 319.49 - lr: 0.000045 - momentum: 0.000000
205
+ 2023-10-12 10:43:25,292 epoch 8 - iter 594/992 - loss 0.01982911 - time (sec): 302.95 - samples/sec: 323.08 - lr: 0.000043 - momentum: 0.000000
206
+ 2023-10-12 10:44:22,487 epoch 8 - iter 693/992 - loss 0.01964615 - time (sec): 360.15 - samples/sec: 316.37 - lr: 0.000041 - momentum: 0.000000
207
+ 2023-10-12 10:45:13,212 epoch 8 - iter 792/992 - loss 0.01935565 - time (sec): 410.87 - samples/sec: 317.94 - lr: 0.000039 - momentum: 0.000000
208
+ 2023-10-12 10:46:03,197 epoch 8 - iter 891/992 - loss 0.01971031 - time (sec): 460.86 - samples/sec: 318.23 - lr: 0.000037 - momentum: 0.000000
209
+ 2023-10-12 10:47:03,208 epoch 8 - iter 990/992 - loss 0.01892606 - time (sec): 520.87 - samples/sec: 314.37 - lr: 0.000036 - momentum: 0.000000
210
+ 2023-10-12 10:47:04,250 ----------------------------------------------------------------------------------------------------
211
+ 2023-10-12 10:47:04,250 EPOCH 8 done: loss 0.0189 - lr: 0.000036
212
+ 2023-10-12 10:47:29,935 DEV : loss 0.18165574967861176 - f1-score (micro avg) 0.7538
213
+ 2023-10-12 10:47:29,980 ----------------------------------------------------------------------------------------------------
214
+ 2023-10-12 10:48:24,334 epoch 9 - iter 99/992 - loss 0.02135843 - time (sec): 54.35 - samples/sec: 316.68 - lr: 0.000034 - momentum: 0.000000
215
+ 2023-10-12 10:49:15,051 epoch 9 - iter 198/992 - loss 0.01908305 - time (sec): 105.07 - samples/sec: 320.27 - lr: 0.000032 - momentum: 0.000000
216
+ 2023-10-12 10:50:05,328 epoch 9 - iter 297/992 - loss 0.01680344 - time (sec): 155.35 - samples/sec: 325.86 - lr: 0.000030 - momentum: 0.000000
217
+ 2023-10-12 10:50:54,195 epoch 9 - iter 396/992 - loss 0.01730348 - time (sec): 204.21 - samples/sec: 324.06 - lr: 0.000029 - momentum: 0.000000
218
+ 2023-10-12 10:51:45,215 epoch 9 - iter 495/992 - loss 0.01563250 - time (sec): 255.23 - samples/sec: 323.92 - lr: 0.000027 - momentum: 0.000000
219
+ 2023-10-12 10:52:35,904 epoch 9 - iter 594/992 - loss 0.01487404 - time (sec): 305.92 - samples/sec: 323.84 - lr: 0.000025 - momentum: 0.000000
220
+ 2023-10-12 10:53:26,780 epoch 9 - iter 693/992 - loss 0.01457903 - time (sec): 356.80 - samples/sec: 324.34 - lr: 0.000023 - momentum: 0.000000
221
+ 2023-10-12 10:54:19,486 epoch 9 - iter 792/992 - loss 0.01480671 - time (sec): 409.50 - samples/sec: 320.45 - lr: 0.000022 - momentum: 0.000000
222
+ 2023-10-12 10:55:09,652 epoch 9 - iter 891/992 - loss 0.01542691 - time (sec): 459.67 - samples/sec: 320.55 - lr: 0.000020 - momentum: 0.000000
223
+ 2023-10-12 10:56:00,454 epoch 9 - iter 990/992 - loss 0.01512327 - time (sec): 510.47 - samples/sec: 320.57 - lr: 0.000018 - momentum: 0.000000
224
+ 2023-10-12 10:56:01,434 ----------------------------------------------------------------------------------------------------
225
+ 2023-10-12 10:56:01,434 EPOCH 9 done: loss 0.0152 - lr: 0.000018
226
+ 2023-10-12 10:56:32,551 DEV : loss 0.19330915808677673 - f1-score (micro avg) 0.7607
227
+ 2023-10-12 10:56:32,595 ----------------------------------------------------------------------------------------------------
228
+ 2023-10-12 10:57:21,108 epoch 10 - iter 99/992 - loss 0.01035239 - time (sec): 48.51 - samples/sec: 344.03 - lr: 0.000016 - momentum: 0.000000
229
+ 2023-10-12 10:58:11,337 epoch 10 - iter 198/992 - loss 0.01080075 - time (sec): 98.74 - samples/sec: 331.86 - lr: 0.000014 - momentum: 0.000000
230
+ 2023-10-12 10:59:05,409 epoch 10 - iter 297/992 - loss 0.01162795 - time (sec): 152.81 - samples/sec: 323.25 - lr: 0.000013 - momentum: 0.000000
231
+ 2023-10-12 10:59:56,254 epoch 10 - iter 396/992 - loss 0.01246594 - time (sec): 203.66 - samples/sec: 323.79 - lr: 0.000011 - momentum: 0.000000
232
+ 2023-10-12 11:00:46,032 epoch 10 - iter 495/992 - loss 0.01213316 - time (sec): 253.43 - samples/sec: 325.65 - lr: 0.000009 - momentum: 0.000000
233
+ 2023-10-12 11:01:34,559 epoch 10 - iter 594/992 - loss 0.01206470 - time (sec): 301.96 - samples/sec: 325.89 - lr: 0.000007 - momentum: 0.000000
234
+ 2023-10-12 11:02:23,481 epoch 10 - iter 693/992 - loss 0.01218444 - time (sec): 350.88 - samples/sec: 325.35 - lr: 0.000006 - momentum: 0.000000
235
+ 2023-10-12 11:03:16,546 epoch 10 - iter 792/992 - loss 0.01226512 - time (sec): 403.95 - samples/sec: 323.31 - lr: 0.000004 - momentum: 0.000000
236
+ 2023-10-12 11:04:05,167 epoch 10 - iter 891/992 - loss 0.01227256 - time (sec): 452.57 - samples/sec: 325.33 - lr: 0.000002 - momentum: 0.000000
237
+ 2023-10-12 11:04:54,115 epoch 10 - iter 990/992 - loss 0.01280459 - time (sec): 501.52 - samples/sec: 326.55 - lr: 0.000000 - momentum: 0.000000
238
+ 2023-10-12 11:04:54,990 ----------------------------------------------------------------------------------------------------
239
+ 2023-10-12 11:04:54,990 EPOCH 10 done: loss 0.0129 - lr: 0.000000
240
+ 2023-10-12 11:05:20,207 DEV : loss 0.1981041431427002 - f1-score (micro avg) 0.7596
241
+ 2023-10-12 11:05:21,142 ----------------------------------------------------------------------------------------------------
242
+ 2023-10-12 11:05:21,144 Loading model from best epoch ...
243
+ 2023-10-12 11:05:25,056 SequenceTagger predicts: Dictionary with 13 tags: O, S-PER, B-PER, E-PER, I-PER, S-LOC, B-LOC, E-LOC, I-LOC, S-ORG, B-ORG, E-ORG, I-ORG
244
+ 2023-10-12 11:05:50,281
245
+ Results:
246
+ - F-score (micro) 0.7433
247
+ - F-score (macro) 0.6567
248
+ - Accuracy 0.6209
249
+
250
+ By class:
251
+ precision recall f1-score support
252
+
253
+ LOC 0.8058 0.8107 0.8082 655
254
+ PER 0.7102 0.7803 0.7436 223
255
+ ORG 0.4044 0.4331 0.4183 127
256
+
257
+ micro avg 0.7308 0.7562 0.7433 1005
258
+ macro avg 0.6401 0.6747 0.6567 1005
259
+ weighted avg 0.7338 0.7562 0.7446 1005
260
+
261
+ 2023-10-12 11:05:50,281 ----------------------------------------------------------------------------------------------------