Hugging Face's logo

stefan-it
/
hmbench-icdar-fr-hmbert-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-5

Token Classification
Flair
PyTorch
French
sequence-tagger-model
Model card Files Community
1
stefan-it commited on
Commit
f745937
•
1 Parent(s): 41b9aa8

Upload folder using huggingface_hub

Browse files
Files changed (5) hide show
  1. best-model.pt +3 -0
  2. dev.tsv +0 -0
  3. loss.tsv +11 -0
  4. test.tsv +0 -0
  5. training.log +242 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:99f8f16fba2046c9736e175d861ce0cb42c5f1e6aae867250460fec90748b933
3
+ size 443311111
dev.tsv ADDED
The diff for this file is too large to render. See raw diff
 
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 01:07:05 0.0000 0.3930 0.0962 0.6706 0.7002 0.6851 0.5359
3
+ 2 01:08:08 0.0000 0.1025 0.0840 0.7229 0.7613 0.7416 0.6047
4
+ 3 01:09:10 0.0000 0.0696 0.1156 0.7109 0.7817 0.7446 0.6153
5
+ 4 01:10:13 0.0000 0.0517 0.1159 0.7292 0.7738 0.7508 0.6184
6
+ 5 01:11:15 0.0000 0.0396 0.1494 0.7330 0.7704 0.7512 0.6197
7
+ 6 01:12:18 0.0000 0.0311 0.1660 0.7221 0.7907 0.7549 0.6247
8
+ 7 01:13:21 0.0000 0.0228 0.1981 0.7157 0.7919 0.7519 0.6222
9
+ 8 01:14:22 0.0000 0.0169 0.2040 0.7220 0.7873 0.7532 0.6253
10
+ 9 01:15:24 0.0000 0.0125 0.2083 0.7356 0.7805 0.7574 0.6273
11
+ 10 01:16:27 0.0000 0.0088 0.2199 0.7254 0.7828 0.7530 0.6234
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,242 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-14 01:06:03,859 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-14 01:06:03,860 Model: "SequenceTagger(
3
+ (embeddings): TransformerWordEmbeddings(
4
+ (model): BertModel(
5
+ (embeddings): BertEmbeddings(
6
+ (word_embeddings): Embedding(32001, 768)
7
+ (position_embeddings): Embedding(512, 768)
8
+ (token_type_embeddings): Embedding(2, 768)
9
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
10
+ (dropout): Dropout(p=0.1, inplace=False)
11
+ )
12
+ (encoder): BertEncoder(
13
+ (layer): ModuleList(
14
+ (0-11): 12 x BertLayer(
15
+ (attention): BertAttention(
16
+ (self): BertSelfAttention(
17
+ (query): Linear(in_features=768, out_features=768, bias=True)
18
+ (key): Linear(in_features=768, out_features=768, bias=True)
19
+ (value): Linear(in_features=768, out_features=768, bias=True)
20
+ (dropout): Dropout(p=0.1, inplace=False)
21
+ )
22
+ (output): BertSelfOutput(
23
+ (dense): Linear(in_features=768, out_features=768, bias=True)
24
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
25
+ (dropout): Dropout(p=0.1, inplace=False)
26
+ )
27
+ )
28
+ (intermediate): BertIntermediate(
29
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
30
+ (intermediate_act_fn): GELUActivation()
31
+ )
32
+ (output): BertOutput(
33
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
34
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
35
+ (dropout): Dropout(p=0.1, inplace=False)
36
+ )
37
+ )
38
+ )
39
+ )
40
+ (pooler): BertPooler(
41
+ (dense): Linear(in_features=768, out_features=768, bias=True)
42
+ (activation): Tanh()
43
+ )
44
+ )
45
+ )
46
+ (locked_dropout): LockedDropout(p=0.5)
47
+ (linear): Linear(in_features=768, out_features=13, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-14 01:06:03,860 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-14 01:06:03,860 MultiCorpus: 7936 train + 992 dev + 992 test sentences
52
+ - NER_ICDAR_EUROPEANA Corpus: 7936 train + 992 dev + 992 test sentences - /root/.flair/datasets/ner_icdar_europeana/fr
53
+ 2023-10-14 01:06:03,860 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-14 01:06:03,860 Train: 7936 sentences
55
+ 2023-10-14 01:06:03,860 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-14 01:06:03,860 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-14 01:06:03,860 Training Params:
58
+ 2023-10-14 01:06:03,860 - learning_rate: "3e-05"
59
+ 2023-10-14 01:06:03,860 - mini_batch_size: "8"
60
+ 2023-10-14 01:06:03,860 - max_epochs: "10"
61
+ 2023-10-14 01:06:03,860 - shuffle: "True"
62
+ 2023-10-14 01:06:03,860 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-14 01:06:03,860 Plugins:
64
+ 2023-10-14 01:06:03,860 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-14 01:06:03,860 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-14 01:06:03,861 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-14 01:06:03,861 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-14 01:06:03,861 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-14 01:06:03,861 Computation:
70
+ 2023-10-14 01:06:03,861 - compute on device: cuda:0
71
+ 2023-10-14 01:06:03,861 - embedding storage: none
72
+ 2023-10-14 01:06:03,861 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-14 01:06:03,861 Model training base path: "hmbench-icdar/fr-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-5"
74
+ 2023-10-14 01:06:03,861 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-14 01:06:03,861 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-14 01:06:09,367 epoch 1 - iter 99/992 - loss 2.17471007 - time (sec): 5.51 - samples/sec: 2804.90 - lr: 0.000003 - momentum: 0.000000
77
+ 2023-10-14 01:06:15,065 epoch 1 - iter 198/992 - loss 1.30549124 - time (sec): 11.20 - samples/sec: 2813.64 - lr: 0.000006 - momentum: 0.000000
78
+ 2023-10-14 01:06:20,977 epoch 1 - iter 297/992 - loss 0.95601849 - time (sec): 17.11 - samples/sec: 2810.76 - lr: 0.000009 - momentum: 0.000000
79
+ 2023-10-14 01:06:26,541 epoch 1 - iter 396/992 - loss 0.76749053 - time (sec): 22.68 - samples/sec: 2838.05 - lr: 0.000012 - momentum: 0.000000
80
+ 2023-10-14 01:06:32,406 epoch 1 - iter 495/992 - loss 0.65224931 - time (sec): 28.54 - samples/sec: 2831.49 - lr: 0.000015 - momentum: 0.000000
81
+ 2023-10-14 01:06:38,300 epoch 1 - iter 594/992 - loss 0.56475688 - time (sec): 34.44 - samples/sec: 2841.74 - lr: 0.000018 - momentum: 0.000000
82
+ 2023-10-14 01:06:44,094 epoch 1 - iter 693/992 - loss 0.50715690 - time (sec): 40.23 - samples/sec: 2827.09 - lr: 0.000021 - momentum: 0.000000
83
+ 2023-10-14 01:06:50,014 epoch 1 - iter 792/992 - loss 0.46053485 - time (sec): 46.15 - samples/sec: 2818.19 - lr: 0.000024 - momentum: 0.000000
84
+ 2023-10-14 01:06:55,862 epoch 1 - iter 891/992 - loss 0.42443987 - time (sec): 52.00 - samples/sec: 2813.04 - lr: 0.000027 - momentum: 0.000000
85
+ 2023-10-14 01:07:01,950 epoch 1 - iter 990/992 - loss 0.39381793 - time (sec): 58.09 - samples/sec: 2813.05 - lr: 0.000030 - momentum: 0.000000
86
+ 2023-10-14 01:07:02,157 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-14 01:07:02,157 EPOCH 1 done: loss 0.3930 - lr: 0.000030
88
+ 2023-10-14 01:07:05,244 DEV : loss 0.09621600061655045 - f1-score (micro avg) 0.6851
89
+ 2023-10-14 01:07:05,264 saving best model
90
+ 2023-10-14 01:07:05,661 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-14 01:07:11,275 epoch 2 - iter 99/992 - loss 0.13490577 - time (sec): 5.61 - samples/sec: 2709.42 - lr: 0.000030 - momentum: 0.000000
92
+ 2023-10-14 01:07:17,108 epoch 2 - iter 198/992 - loss 0.11610667 - time (sec): 11.45 - samples/sec: 2725.35 - lr: 0.000029 - momentum: 0.000000
93
+ 2023-10-14 01:07:22,696 epoch 2 - iter 297/992 - loss 0.11416877 - time (sec): 17.03 - samples/sec: 2766.88 - lr: 0.000029 - momentum: 0.000000
94
+ 2023-10-14 01:07:28,612 epoch 2 - iter 396/992 - loss 0.10769612 - time (sec): 22.95 - samples/sec: 2773.46 - lr: 0.000029 - momentum: 0.000000
95
+ 2023-10-14 01:07:34,383 epoch 2 - iter 495/992 - loss 0.10656848 - time (sec): 28.72 - samples/sec: 2814.53 - lr: 0.000028 - momentum: 0.000000
96
+ 2023-10-14 01:07:40,280 epoch 2 - iter 594/992 - loss 0.10563479 - time (sec): 34.62 - samples/sec: 2823.23 - lr: 0.000028 - momentum: 0.000000
97
+ 2023-10-14 01:07:46,122 epoch 2 - iter 693/992 - loss 0.10497520 - time (sec): 40.46 - samples/sec: 2823.34 - lr: 0.000028 - momentum: 0.000000
98
+ 2023-10-14 01:07:51,931 epoch 2 - iter 792/992 - loss 0.10311384 - time (sec): 46.27 - samples/sec: 2814.64 - lr: 0.000027 - momentum: 0.000000
99
+ 2023-10-14 01:07:58,133 epoch 2 - iter 891/992 - loss 0.10194024 - time (sec): 52.47 - samples/sec: 2802.48 - lr: 0.000027 - momentum: 0.000000
100
+ 2023-10-14 01:08:03,960 epoch 2 - iter 990/992 - loss 0.10262038 - time (sec): 58.30 - samples/sec: 2804.64 - lr: 0.000027 - momentum: 0.000000
101
+ 2023-10-14 01:08:04,121 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-14 01:08:04,121 EPOCH 2 done: loss 0.1025 - lr: 0.000027
103
+ 2023-10-14 01:08:07,983 DEV : loss 0.08396855741739273 - f1-score (micro avg) 0.7416
104
+ 2023-10-14 01:08:08,004 saving best model
105
+ 2023-10-14 01:08:08,517 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-14 01:08:14,189 epoch 3 - iter 99/992 - loss 0.06551401 - time (sec): 5.67 - samples/sec: 2662.64 - lr: 0.000026 - momentum: 0.000000
107
+ 2023-10-14 01:08:20,258 epoch 3 - iter 198/992 - loss 0.06924244 - time (sec): 11.74 - samples/sec: 2763.88 - lr: 0.000026 - momentum: 0.000000
108
+ 2023-10-14 01:08:25,798 epoch 3 - iter 297/992 - loss 0.07056573 - time (sec): 17.28 - samples/sec: 2771.16 - lr: 0.000026 - momentum: 0.000000
109
+ 2023-10-14 01:08:31,708 epoch 3 - iter 396/992 - loss 0.07055584 - time (sec): 23.19 - samples/sec: 2748.52 - lr: 0.000025 - momentum: 0.000000
110
+ 2023-10-14 01:08:37,722 epoch 3 - iter 495/992 - loss 0.06873774 - time (sec): 29.20 - samples/sec: 2788.98 - lr: 0.000025 - momentum: 0.000000
111
+ 2023-10-14 01:08:43,529 epoch 3 - iter 594/992 - loss 0.07045008 - time (sec): 35.01 - samples/sec: 2794.24 - lr: 0.000025 - momentum: 0.000000
112
+ 2023-10-14 01:08:49,391 epoch 3 - iter 693/992 - loss 0.07037162 - time (sec): 40.87 - samples/sec: 2803.95 - lr: 0.000024 - momentum: 0.000000
113
+ 2023-10-14 01:08:55,506 epoch 3 - iter 792/992 - loss 0.07021216 - time (sec): 46.99 - samples/sec: 2796.17 - lr: 0.000024 - momentum: 0.000000
114
+ 2023-10-14 01:09:01,195 epoch 3 - iter 891/992 - loss 0.06994634 - time (sec): 52.68 - samples/sec: 2790.15 - lr: 0.000024 - momentum: 0.000000
115
+ 2023-10-14 01:09:06,919 epoch 3 - iter 990/992 - loss 0.06967297 - time (sec): 58.40 - samples/sec: 2801.51 - lr: 0.000023 - momentum: 0.000000
116
+ 2023-10-14 01:09:07,048 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-14 01:09:07,048 EPOCH 3 done: loss 0.0696 - lr: 0.000023
118
+ 2023-10-14 01:09:10,503 DEV : loss 0.11555210500955582 - f1-score (micro avg) 0.7446
119
+ 2023-10-14 01:09:10,523 saving best model
120
+ 2023-10-14 01:09:11,025 ----------------------------------------------------------------------------------------------------
121
+ 2023-10-14 01:09:16,953 epoch 4 - iter 99/992 - loss 0.03972797 - time (sec): 5.93 - samples/sec: 2955.76 - lr: 0.000023 - momentum: 0.000000
122
+ 2023-10-14 01:09:22,805 epoch 4 - iter 198/992 - loss 0.04570840 - time (sec): 11.78 - samples/sec: 2867.88 - lr: 0.000023 - momentum: 0.000000
123
+ 2023-10-14 01:09:28,524 epoch 4 - iter 297/992 - loss 0.04904627 - time (sec): 17.50 - samples/sec: 2862.88 - lr: 0.000022 - momentum: 0.000000
124
+ 2023-10-14 01:09:34,474 epoch 4 - iter 396/992 - loss 0.04945405 - time (sec): 23.45 - samples/sec: 2817.49 - lr: 0.000022 - momentum: 0.000000
125
+ 2023-10-14 01:09:40,476 epoch 4 - iter 495/992 - loss 0.04831346 - time (sec): 29.45 - samples/sec: 2809.35 - lr: 0.000022 - momentum: 0.000000
126
+ 2023-10-14 01:09:46,365 epoch 4 - iter 594/992 - loss 0.04825902 - time (sec): 35.34 - samples/sec: 2794.59 - lr: 0.000021 - momentum: 0.000000
127
+ 2023-10-14 01:09:52,103 epoch 4 - iter 693/992 - loss 0.04856953 - time (sec): 41.08 - samples/sec: 2782.77 - lr: 0.000021 - momentum: 0.000000
128
+ 2023-10-14 01:09:57,796 epoch 4 - iter 792/992 - loss 0.04919667 - time (sec): 46.77 - samples/sec: 2790.01 - lr: 0.000021 - momentum: 0.000000
129
+ 2023-10-14 01:10:03,524 epoch 4 - iter 891/992 - loss 0.04895947 - time (sec): 52.50 - samples/sec: 2784.48 - lr: 0.000020 - momentum: 0.000000
130
+ 2023-10-14 01:10:09,641 epoch 4 - iter 990/992 - loss 0.05170296 - time (sec): 58.62 - samples/sec: 2792.06 - lr: 0.000020 - momentum: 0.000000
131
+ 2023-10-14 01:10:09,813 ----------------------------------------------------------------------------------------------------
132
+ 2023-10-14 01:10:09,813 EPOCH 4 done: loss 0.0517 - lr: 0.000020
133
+ 2023-10-14 01:10:13,733 DEV : loss 0.11588922142982483 - f1-score (micro avg) 0.7508
134
+ 2023-10-14 01:10:13,754 saving best model
135
+ 2023-10-14 01:10:14,263 ----------------------------------------------------------------------------------------------------
136
+ 2023-10-14 01:10:20,108 epoch 5 - iter 99/992 - loss 0.03385090 - time (sec): 5.84 - samples/sec: 2833.18 - lr: 0.000020 - momentum: 0.000000
137
+ 2023-10-14 01:10:25,922 epoch 5 - iter 198/992 - loss 0.03715409 - time (sec): 11.66 - samples/sec: 2861.38 - lr: 0.000019 - momentum: 0.000000
138
+ 2023-10-14 01:10:31,745 epoch 5 - iter 297/992 - loss 0.03954664 - time (sec): 17.48 - samples/sec: 2820.44 - lr: 0.000019 - momentum: 0.000000
139
+ 2023-10-14 01:10:37,565 epoch 5 - iter 396/992 - loss 0.03785856 - time (sec): 23.30 - samples/sec: 2828.39 - lr: 0.000019 - momentum: 0.000000
140
+ 2023-10-14 01:10:43,326 epoch 5 - iter 495/992 - loss 0.03770491 - time (sec): 29.06 - samples/sec: 2832.89 - lr: 0.000018 - momentum: 0.000000
141
+ 2023-10-14 01:10:49,025 epoch 5 - iter 594/992 - loss 0.03881175 - time (sec): 34.76 - samples/sec: 2844.93 - lr: 0.000018 - momentum: 0.000000
142
+ 2023-10-14 01:10:54,554 epoch 5 - iter 693/992 - loss 0.04006727 - time (sec): 40.29 - samples/sec: 2834.78 - lr: 0.000018 - momentum: 0.000000
143
+ 2023-10-14 01:11:00,484 epoch 5 - iter 792/992 - loss 0.04022835 - time (sec): 46.22 - samples/sec: 2839.18 - lr: 0.000017 - momentum: 0.000000
144
+ 2023-10-14 01:11:06,349 epoch 5 - iter 891/992 - loss 0.03964020 - time (sec): 52.08 - samples/sec: 2827.37 - lr: 0.000017 - momentum: 0.000000
145
+ 2023-10-14 01:11:12,062 epoch 5 - iter 990/992 - loss 0.03963685 - time (sec): 57.80 - samples/sec: 2832.44 - lr: 0.000017 - momentum: 0.000000
146
+ 2023-10-14 01:11:12,174 ----------------------------------------------------------------------------------------------------
147
+ 2023-10-14 01:11:12,174 EPOCH 5 done: loss 0.0396 - lr: 0.000017
148
+ 2023-10-14 01:11:15,549 DEV : loss 0.149429589509964 - f1-score (micro avg) 0.7512
149
+ 2023-10-14 01:11:15,571 saving best model
150
+ 2023-10-14 01:11:16,077 ----------------------------------------------------------------------------------------------------
151
+ 2023-10-14 01:11:22,503 epoch 6 - iter 99/992 - loss 0.03207299 - time (sec): 6.42 - samples/sec: 2699.08 - lr: 0.000016 - momentum: 0.000000
152
+ 2023-10-14 01:11:28,059 epoch 6 - iter 198/992 - loss 0.03319848 - time (sec): 11.98 - samples/sec: 2789.57 - lr: 0.000016 - momentum: 0.000000
153
+ 2023-10-14 01:11:33,724 epoch 6 - iter 297/992 - loss 0.03023090 - time (sec): 17.64 - samples/sec: 2790.07 - lr: 0.000016 - momentum: 0.000000
154
+ 2023-10-14 01:11:39,537 epoch 6 - iter 396/992 - loss 0.03113076 - time (sec): 23.45 - samples/sec: 2808.14 - lr: 0.000015 - momentum: 0.000000
155
+ 2023-10-14 01:11:45,406 epoch 6 - iter 495/992 - loss 0.03125986 - time (sec): 29.32 - samples/sec: 2812.25 - lr: 0.000015 - momentum: 0.000000
156
+ 2023-10-14 01:11:51,038 epoch 6 - iter 594/992 - loss 0.03070883 - time (sec): 34.95 - samples/sec: 2815.40 - lr: 0.000015 - momentum: 0.000000
157
+ 2023-10-14 01:11:56,908 epoch 6 - iter 693/992 - loss 0.03147804 - time (sec): 40.82 - samples/sec: 2809.72 - lr: 0.000014 - momentum: 0.000000
158
+ 2023-10-14 01:12:02,821 epoch 6 - iter 792/992 - loss 0.03114004 - time (sec): 46.74 - samples/sec: 2805.07 - lr: 0.000014 - momentum: 0.000000
159
+ 2023-10-14 01:12:08,954 epoch 6 - iter 891/992 - loss 0.03098304 - time (sec): 52.87 - samples/sec: 2800.80 - lr: 0.000014 - momentum: 0.000000
160
+ 2023-10-14 01:12:14,742 epoch 6 - iter 990/992 - loss 0.03115872 - time (sec): 58.66 - samples/sec: 2790.74 - lr: 0.000013 - momentum: 0.000000
161
+ 2023-10-14 01:12:14,852 ----------------------------------------------------------------------------------------------------
162
+ 2023-10-14 01:12:14,852 EPOCH 6 done: loss 0.0311 - lr: 0.000013
163
+ 2023-10-14 01:12:18,275 DEV : loss 0.1660223752260208 - f1-score (micro avg) 0.7549
164
+ 2023-10-14 01:12:18,296 saving best model
165
+ 2023-10-14 01:12:18,723 ----------------------------------------------------------------------------------------------------
166
+ 2023-10-14 01:12:24,554 epoch 7 - iter 99/992 - loss 0.02003484 - time (sec): 5.83 - samples/sec: 2773.72 - lr: 0.000013 - momentum: 0.000000
167
+ 2023-10-14 01:12:30,403 epoch 7 - iter 198/992 - loss 0.02219555 - time (sec): 11.68 - samples/sec: 2749.50 - lr: 0.000013 - momentum: 0.000000
168
+ 2023-10-14 01:12:36,359 epoch 7 - iter 297/992 - loss 0.02044054 - time (sec): 17.63 - samples/sec: 2783.24 - lr: 0.000012 - momentum: 0.000000
169
+ 2023-10-14 01:12:42,275 epoch 7 - iter 396/992 - loss 0.02115285 - time (sec): 23.55 - samples/sec: 2787.47 - lr: 0.000012 - momentum: 0.000000
170
+ 2023-10-14 01:12:47,979 epoch 7 - iter 495/992 - loss 0.02111835 - time (sec): 29.25 - samples/sec: 2790.21 - lr: 0.000012 - momentum: 0.000000
171
+ 2023-10-14 01:12:53,906 epoch 7 - iter 594/992 - loss 0.02231115 - time (sec): 35.18 - samples/sec: 2794.36 - lr: 0.000011 - momentum: 0.000000
172
+ 2023-10-14 01:12:59,944 epoch 7 - iter 693/992 - loss 0.02257417 - time (sec): 41.22 - samples/sec: 2790.76 - lr: 0.000011 - momentum: 0.000000
173
+ 2023-10-14 01:13:05,688 epoch 7 - iter 792/992 - loss 0.02390230 - time (sec): 46.96 - samples/sec: 2792.08 - lr: 0.000011 - momentum: 0.000000
174
+ 2023-10-14 01:13:11,978 epoch 7 - iter 891/992 - loss 0.02331735 - time (sec): 53.25 - samples/sec: 2769.06 - lr: 0.000010 - momentum: 0.000000
175
+ 2023-10-14 01:13:17,717 epoch 7 - iter 990/992 - loss 0.02276912 - time (sec): 58.99 - samples/sec: 2774.18 - lr: 0.000010 - momentum: 0.000000
176
+ 2023-10-14 01:13:17,823 ----------------------------------------------------------------------------------------------------
177
+ 2023-10-14 01:13:17,823 EPOCH 7 done: loss 0.0228 - lr: 0.000010
178
+ 2023-10-14 01:13:21,216 DEV : loss 0.19811701774597168 - f1-score (micro avg) 0.7519
179
+ 2023-10-14 01:13:21,240 ----------------------------------------------------------------------------------------------------
180
+ 2023-10-14 01:13:26,990 epoch 8 - iter 99/992 - loss 0.01904585 - time (sec): 5.75 - samples/sec: 2988.62 - lr: 0.000010 - momentum: 0.000000
181
+ 2023-10-14 01:13:32,631 epoch 8 - iter 198/992 - loss 0.01503292 - time (sec): 11.39 - samples/sec: 2919.52 - lr: 0.000009 - momentum: 0.000000
182
+ 2023-10-14 01:13:38,169 epoch 8 - iter 297/992 - loss 0.01613248 - time (sec): 16.93 - samples/sec: 2884.32 - lr: 0.000009 - momentum: 0.000000
183
+ 2023-10-14 01:13:44,251 epoch 8 - iter 396/992 - loss 0.01565979 - time (sec): 23.01 - samples/sec: 2873.38 - lr: 0.000009 - momentum: 0.000000
184
+ 2023-10-14 01:13:50,113 epoch 8 - iter 495/992 - loss 0.01515949 - time (sec): 28.87 - samples/sec: 2875.24 - lr: 0.000008 - momentum: 0.000000
185
+ 2023-10-14 01:13:56,010 epoch 8 - iter 594/992 - loss 0.01545050 - time (sec): 34.77 - samples/sec: 2873.67 - lr: 0.000008 - momentum: 0.000000
186
+ 2023-10-14 01:14:01,470 epoch 8 - iter 693/992 - loss 0.01547792 - time (sec): 40.23 - samples/sec: 2886.91 - lr: 0.000008 - momentum: 0.000000
187
+ 2023-10-14 01:14:07,035 epoch 8 - iter 792/992 - loss 0.01620138 - time (sec): 45.79 - samples/sec: 2880.33 - lr: 0.000007 - momentum: 0.000000
188
+ 2023-10-14 01:14:12,743 epoch 8 - iter 891/992 - loss 0.01643518 - time (sec): 51.50 - samples/sec: 2871.73 - lr: 0.000007 - momentum: 0.000000
189
+ 2023-10-14 01:14:18,495 epoch 8 - iter 990/992 - loss 0.01691683 - time (sec): 57.25 - samples/sec: 2860.38 - lr: 0.000007 - momentum: 0.000000
190
+ 2023-10-14 01:14:18,596 ----------------------------------------------------------------------------------------------------
191
+ 2023-10-14 01:14:18,596 EPOCH 8 done: loss 0.0169 - lr: 0.000007
192
+ 2023-10-14 01:14:22,033 DEV : loss 0.2040073573589325 - f1-score (micro avg) 0.7532
193
+ 2023-10-14 01:14:22,053 ----------------------------------------------------------------------------------------------------
194
+ 2023-10-14 01:14:27,795 epoch 9 - iter 99/992 - loss 0.01058698 - time (sec): 5.74 - samples/sec: 2818.02 - lr: 0.000006 - momentum: 0.000000
195
+ 2023-10-14 01:14:33,771 epoch 9 - iter 198/992 - loss 0.01076997 - time (sec): 11.72 - samples/sec: 2835.15 - lr: 0.000006 - momentum: 0.000000
196
+ 2023-10-14 01:14:39,862 epoch 9 - iter 297/992 - loss 0.01232413 - time (sec): 17.81 - samples/sec: 2809.95 - lr: 0.000006 - momentum: 0.000000
197
+ 2023-10-14 01:14:45,579 epoch 9 - iter 396/992 - loss 0.01181418 - time (sec): 23.52 - samples/sec: 2791.33 - lr: 0.000005 - momentum: 0.000000
198
+ 2023-10-14 01:14:51,506 epoch 9 - iter 495/992 - loss 0.01146147 - time (sec): 29.45 - samples/sec: 2793.73 - lr: 0.000005 - momentum: 0.000000
199
+ 2023-10-14 01:14:57,263 epoch 9 - iter 594/992 - loss 0.01205104 - time (sec): 35.21 - samples/sec: 2802.96 - lr: 0.000005 - momentum: 0.000000
200
+ 2023-10-14 01:15:03,302 epoch 9 - iter 693/992 - loss 0.01212528 - time (sec): 41.25 - samples/sec: 2782.13 - lr: 0.000004 - momentum: 0.000000
201
+ 2023-10-14 01:15:09,297 epoch 9 - iter 792/992 - loss 0.01203877 - time (sec): 47.24 - samples/sec: 2788.87 - lr: 0.000004 - momentum: 0.000000
202
+ 2023-10-14 01:15:14,909 epoch 9 - iter 891/992 - loss 0.01226298 - time (sec): 52.85 - samples/sec: 2793.28 - lr: 0.000004 - momentum: 0.000000
203
+ 2023-10-14 01:15:20,595 epoch 9 - iter 990/992 - loss 0.01257149 - time (sec): 58.54 - samples/sec: 2793.66 - lr: 0.000003 - momentum: 0.000000
204
+ 2023-10-14 01:15:20,749 ----------------------------------------------------------------------------------------------------
205
+ 2023-10-14 01:15:20,750 EPOCH 9 done: loss 0.0125 - lr: 0.000003
206
+ 2023-10-14 01:15:24,734 DEV : loss 0.20826229453086853 - f1-score (micro avg) 0.7574
207
+ 2023-10-14 01:15:24,754 saving best model
208
+ 2023-10-14 01:15:25,268 ----------------------------------------------------------------------------------------------------
209
+ 2023-10-14 01:15:31,404 epoch 10 - iter 99/992 - loss 0.00769297 - time (sec): 6.13 - samples/sec: 2865.70 - lr: 0.000003 - momentum: 0.000000
210
+ 2023-10-14 01:15:37,404 epoch 10 - iter 198/992 - loss 0.00697383 - time (sec): 12.13 - samples/sec: 2809.85 - lr: 0.000003 - momentum: 0.000000
211
+ 2023-10-14 01:15:42,992 epoch 10 - iter 297/992 - loss 0.00764415 - time (sec): 17.72 - samples/sec: 2775.21 - lr: 0.000002 - momentum: 0.000000
212
+ 2023-10-14 01:15:48,942 epoch 10 - iter 396/992 - loss 0.00827819 - time (sec): 23.67 - samples/sec: 2778.87 - lr: 0.000002 - momentum: 0.000000
213
+ 2023-10-14 01:15:54,828 epoch 10 - iter 495/992 - loss 0.00802192 - time (sec): 29.56 - samples/sec: 2786.17 - lr: 0.000002 - momentum: 0.000000
214
+ 2023-10-14 01:16:00,688 epoch 10 - iter 594/992 - loss 0.00768562 - time (sec): 35.42 - samples/sec: 2778.12 - lr: 0.000001 - momentum: 0.000000
215
+ 2023-10-14 01:16:06,479 epoch 10 - iter 693/992 - loss 0.00852248 - time (sec): 41.21 - samples/sec: 2781.76 - lr: 0.000001 - momentum: 0.000000
216
+ 2023-10-14 01:16:12,477 epoch 10 - iter 792/992 - loss 0.00836690 - time (sec): 47.20 - samples/sec: 2781.41 - lr: 0.000001 - momentum: 0.000000
217
+ 2023-10-14 01:16:18,131 epoch 10 - iter 891/992 - loss 0.00856003 - time (sec): 52.86 - samples/sec: 2797.47 - lr: 0.000000 - momentum: 0.000000
218
+ 2023-10-14 01:16:23,891 epoch 10 - iter 990/992 - loss 0.00877170 - time (sec): 58.62 - samples/sec: 2792.44 - lr: 0.000000 - momentum: 0.000000
219
+ 2023-10-14 01:16:23,999 ----------------------------------------------------------------------------------------------------
220
+ 2023-10-14 01:16:24,000 EPOCH 10 done: loss 0.0088 - lr: 0.000000
221
+ 2023-10-14 01:16:27,448 DEV : loss 0.21987785398960114 - f1-score (micro avg) 0.753
222
+ 2023-10-14 01:16:27,906 ----------------------------------------------------------------------------------------------------
223
+ 2023-10-14 01:16:27,907 Loading model from best epoch ...
224
+ 2023-10-14 01:16:29,242 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
225
+ 2023-10-14 01:16:32,497
226
+ Results:
227
+ - F-score (micro) 0.7723
228
+ - F-score (macro) 0.6898
229
+ - Accuracy 0.6513
230
+
231
+ By class:
232
+ precision recall f1-score support
233
+
234
+ LOC 0.8118 0.8427 0.8270 655
235
+ PER 0.7379 0.8206 0.7771 223
236
+ ORG 0.4831 0.4488 0.4653 127
237
+
238
+ micro avg 0.7572 0.7881 0.7723 1005
239
+ macro avg 0.6776 0.7041 0.6898 1005
240
+ weighted avg 0.7538 0.7881 0.7702 1005
241
+
242
+ 2023-10-14 01:16:32,497 ----------------------------------------------------------------------------------------------------