Hugging Face's logo

stefan-it
/
hmbench-newseye-fr-hmbert-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2

Token Classification
Flair
PyTorch
French
sequence-tagger-model
Model card Files Community
stefan-it commited on
Commit
27d8c6c
·
1 Parent(s): 2aea0f8

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 +243 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f4ddfc6eb5f67f5db2cdebc7ff8e8afe8028a394f9ba5c8d9679435a88055928
3
+ size 443323527
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 09:59:09 0.0000 0.4118 0.1314 0.6353 0.7347 0.6814 0.5395
3
+ 2 10:00:43 0.0000 0.1258 0.1379 0.7477 0.7741 0.7607 0.6273
4
+ 3 10:02:16 0.0000 0.0909 0.1591 0.7686 0.7864 0.7774 0.6561
5
+ 4 10:03:50 0.0000 0.0686 0.1592 0.7594 0.7946 0.7766 0.6532
6
+ 5 10:05:23 0.0000 0.0529 0.2099 0.7709 0.7918 0.7812 0.6599
7
+ 6 10:06:56 0.0000 0.0405 0.1888 0.7901 0.7837 0.7869 0.6713
8
+ 7 10:08:29 0.0000 0.0302 0.2058 0.7523 0.7932 0.7722 0.6543
9
+ 8 10:10:02 0.0000 0.0235 0.2024 0.7968 0.8054 0.8011 0.6876
10
+ 9 10:11:34 0.0000 0.0170 0.2027 0.7981 0.8014 0.7997 0.6857
11
+ 10 10:13:07 0.0000 0.0116 0.2077 0.8008 0.8150 0.8078 0.6957
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,243 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-16 09:57:38,090 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-16 09:57:38,091 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=17, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-16 09:57:38,091 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-16 09:57:38,092 MultiCorpus: 7142 train + 698 dev + 2570 test sentences
52
+ - NER_HIPE_2022 Corpus: 7142 train + 698 dev + 2570 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fr/with_doc_seperator
53
+ 2023-10-16 09:57:38,092 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-16 09:57:38,092 Train: 7142 sentences
55
+ 2023-10-16 09:57:38,092 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-16 09:57:38,092 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-16 09:57:38,092 Training Params:
58
+ 2023-10-16 09:57:38,092 - learning_rate: "5e-05"
59
+ 2023-10-16 09:57:38,092 - mini_batch_size: "4"
60
+ 2023-10-16 09:57:38,092 - max_epochs: "10"
61
+ 2023-10-16 09:57:38,092 - shuffle: "True"
62
+ 2023-10-16 09:57:38,092 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-16 09:57:38,092 Plugins:
64
+ 2023-10-16 09:57:38,092 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-16 09:57:38,092 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-16 09:57:38,092 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-16 09:57:38,092 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-16 09:57:38,092 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-16 09:57:38,092 Computation:
70
+ 2023-10-16 09:57:38,092 - compute on device: cuda:0
71
+ 2023-10-16 09:57:38,092 - embedding storage: none
72
+ 2023-10-16 09:57:38,092 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-16 09:57:38,092 Model training base path: "hmbench-newseye/fr-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2"
74
+ 2023-10-16 09:57:38,093 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-16 09:57:38,093 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-16 09:57:46,981 epoch 1 - iter 178/1786 - loss 1.94742290 - time (sec): 8.89 - samples/sec: 2937.79 - lr: 0.000005 - momentum: 0.000000
77
+ 2023-10-16 09:57:55,593 epoch 1 - iter 356/1786 - loss 1.24408736 - time (sec): 17.50 - samples/sec: 2890.19 - lr: 0.000010 - momentum: 0.000000
78
+ 2023-10-16 09:58:04,150 epoch 1 - iter 534/1786 - loss 0.94076449 - time (sec): 26.06 - samples/sec: 2887.12 - lr: 0.000015 - momentum: 0.000000
79
+ 2023-10-16 09:58:12,740 epoch 1 - iter 712/1786 - loss 0.77543030 - time (sec): 34.65 - samples/sec: 2872.30 - lr: 0.000020 - momentum: 0.000000
80
+ 2023-10-16 09:58:21,321 epoch 1 - iter 890/1786 - loss 0.66770568 - time (sec): 43.23 - samples/sec: 2843.87 - lr: 0.000025 - momentum: 0.000000
81
+ 2023-10-16 09:58:30,044 epoch 1 - iter 1068/1786 - loss 0.59121842 - time (sec): 51.95 - samples/sec: 2822.92 - lr: 0.000030 - momentum: 0.000000
82
+ 2023-10-16 09:58:39,285 epoch 1 - iter 1246/1786 - loss 0.52805819 - time (sec): 61.19 - samples/sec: 2824.23 - lr: 0.000035 - momentum: 0.000000
83
+ 2023-10-16 09:58:48,299 epoch 1 - iter 1424/1786 - loss 0.47799529 - time (sec): 70.21 - samples/sec: 2826.86 - lr: 0.000040 - momentum: 0.000000
84
+ 2023-10-16 09:58:57,179 epoch 1 - iter 1602/1786 - loss 0.44205077 - time (sec): 79.09 - samples/sec: 2820.65 - lr: 0.000045 - momentum: 0.000000
85
+ 2023-10-16 09:59:06,126 epoch 1 - iter 1780/1786 - loss 0.41271469 - time (sec): 88.03 - samples/sec: 2816.87 - lr: 0.000050 - momentum: 0.000000
86
+ 2023-10-16 09:59:06,417 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-16 09:59:06,417 EPOCH 1 done: loss 0.4118 - lr: 0.000050
88
+ 2023-10-16 09:59:09,619 DEV : loss 0.13136854767799377 - f1-score (micro avg) 0.6814
89
+ 2023-10-16 09:59:09,638 saving best model
90
+ 2023-10-16 09:59:10,047 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-16 09:59:19,266 epoch 2 - iter 178/1786 - loss 0.10867454 - time (sec): 9.22 - samples/sec: 2781.86 - lr: 0.000049 - momentum: 0.000000
92
+ 2023-10-16 09:59:28,367 epoch 2 - iter 356/1786 - loss 0.11195066 - time (sec): 18.32 - samples/sec: 2755.44 - lr: 0.000049 - momentum: 0.000000
93
+ 2023-10-16 09:59:37,194 epoch 2 - iter 534/1786 - loss 0.12303775 - time (sec): 27.14 - samples/sec: 2719.96 - lr: 0.000048 - momentum: 0.000000
94
+ 2023-10-16 09:59:45,992 epoch 2 - iter 712/1786 - loss 0.12365229 - time (sec): 35.94 - samples/sec: 2757.21 - lr: 0.000048 - momentum: 0.000000
95
+ 2023-10-16 09:59:54,863 epoch 2 - iter 890/1786 - loss 0.12778904 - time (sec): 44.81 - samples/sec: 2792.71 - lr: 0.000047 - momentum: 0.000000
96
+ 2023-10-16 10:00:03,600 epoch 2 - iter 1068/1786 - loss 0.12649898 - time (sec): 53.55 - samples/sec: 2790.61 - lr: 0.000047 - momentum: 0.000000
97
+ 2023-10-16 10:00:12,465 epoch 2 - iter 1246/1786 - loss 0.12771009 - time (sec): 62.42 - samples/sec: 2777.44 - lr: 0.000046 - momentum: 0.000000
98
+ 2023-10-16 10:00:21,267 epoch 2 - iter 1424/1786 - loss 0.12802324 - time (sec): 71.22 - samples/sec: 2780.94 - lr: 0.000046 - momentum: 0.000000
99
+ 2023-10-16 10:00:30,154 epoch 2 - iter 1602/1786 - loss 0.12740373 - time (sec): 80.10 - samples/sec: 2800.22 - lr: 0.000045 - momentum: 0.000000
100
+ 2023-10-16 10:00:38,805 epoch 2 - iter 1780/1786 - loss 0.12594981 - time (sec): 88.76 - samples/sec: 2796.84 - lr: 0.000044 - momentum: 0.000000
101
+ 2023-10-16 10:00:39,067 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-16 10:00:39,067 EPOCH 2 done: loss 0.1258 - lr: 0.000044
103
+ 2023-10-16 10:00:43,284 DEV : loss 0.13792423903942108 - f1-score (micro avg) 0.7607
104
+ 2023-10-16 10:00:43,300 saving best model
105
+ 2023-10-16 10:00:44,350 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-16 10:00:53,160 epoch 3 - iter 178/1786 - loss 0.08110664 - time (sec): 8.81 - samples/sec: 2699.52 - lr: 0.000044 - momentum: 0.000000
107
+ 2023-10-16 10:01:01,652 epoch 3 - iter 356/1786 - loss 0.08864351 - time (sec): 17.30 - samples/sec: 2858.43 - lr: 0.000043 - momentum: 0.000000
108
+ 2023-10-16 10:01:10,237 epoch 3 - iter 534/1786 - loss 0.09228133 - time (sec): 25.89 - samples/sec: 2872.89 - lr: 0.000043 - momentum: 0.000000
109
+ 2023-10-16 10:01:19,143 epoch 3 - iter 712/1786 - loss 0.09148226 - time (sec): 34.79 - samples/sec: 2895.87 - lr: 0.000042 - momentum: 0.000000
110
+ 2023-10-16 10:01:28,075 epoch 3 - iter 890/1786 - loss 0.09059160 - time (sec): 43.72 - samples/sec: 2862.22 - lr: 0.000042 - momentum: 0.000000
111
+ 2023-10-16 10:01:36,653 epoch 3 - iter 1068/1786 - loss 0.09136882 - time (sec): 52.30 - samples/sec: 2844.87 - lr: 0.000041 - momentum: 0.000000
112
+ 2023-10-16 10:01:45,467 epoch 3 - iter 1246/1786 - loss 0.09087475 - time (sec): 61.12 - samples/sec: 2825.63 - lr: 0.000041 - momentum: 0.000000
113
+ 2023-10-16 10:01:54,159 epoch 3 - iter 1424/1786 - loss 0.09230921 - time (sec): 69.81 - samples/sec: 2828.78 - lr: 0.000040 - momentum: 0.000000
114
+ 2023-10-16 10:02:02,750 epoch 3 - iter 1602/1786 - loss 0.09190419 - time (sec): 78.40 - samples/sec: 2829.31 - lr: 0.000039 - momentum: 0.000000
115
+ 2023-10-16 10:02:11,661 epoch 3 - iter 1780/1786 - loss 0.09100255 - time (sec): 87.31 - samples/sec: 2838.16 - lr: 0.000039 - momentum: 0.000000
116
+ 2023-10-16 10:02:12,015 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-16 10:02:12,016 EPOCH 3 done: loss 0.0909 - lr: 0.000039
118
+ 2023-10-16 10:02:16,212 DEV : loss 0.15914735198020935 - f1-score (micro avg) 0.7774
119
+ 2023-10-16 10:02:16,228 saving best model
120
+ 2023-10-16 10:02:16,730 ----------------------------------------------------------------------------------------------------
121
+ 2023-10-16 10:02:25,530 epoch 4 - iter 178/1786 - loss 0.05552608 - time (sec): 8.80 - samples/sec: 2812.79 - lr: 0.000038 - momentum: 0.000000
122
+ 2023-10-16 10:02:34,417 epoch 4 - iter 356/1786 - loss 0.06436274 - time (sec): 17.69 - samples/sec: 2799.71 - lr: 0.000038 - momentum: 0.000000
123
+ 2023-10-16 10:02:43,378 epoch 4 - iter 534/1786 - loss 0.06377877 - time (sec): 26.65 - samples/sec: 2766.31 - lr: 0.000037 - momentum: 0.000000
124
+ 2023-10-16 10:02:52,524 epoch 4 - iter 712/1786 - loss 0.06390155 - time (sec): 35.79 - samples/sec: 2794.47 - lr: 0.000037 - momentum: 0.000000
125
+ 2023-10-16 10:03:01,362 epoch 4 - iter 890/1786 - loss 0.06628737 - time (sec): 44.63 - samples/sec: 2774.34 - lr: 0.000036 - momentum: 0.000000
126
+ 2023-10-16 10:03:10,171 epoch 4 - iter 1068/1786 - loss 0.06610693 - time (sec): 53.44 - samples/sec: 2767.56 - lr: 0.000036 - momentum: 0.000000
127
+ 2023-10-16 10:03:19,114 epoch 4 - iter 1246/1786 - loss 0.06561388 - time (sec): 62.38 - samples/sec: 2769.21 - lr: 0.000035 - momentum: 0.000000
128
+ 2023-10-16 10:03:27,722 epoch 4 - iter 1424/1786 - loss 0.06713646 - time (sec): 70.99 - samples/sec: 2773.70 - lr: 0.000034 - momentum: 0.000000
129
+ 2023-10-16 10:03:36,190 epoch 4 - iter 1602/1786 - loss 0.06817480 - time (sec): 79.46 - samples/sec: 2771.49 - lr: 0.000034 - momentum: 0.000000
130
+ 2023-10-16 10:03:45,416 epoch 4 - iter 1780/1786 - loss 0.06843290 - time (sec): 88.69 - samples/sec: 2796.41 - lr: 0.000033 - momentum: 0.000000
131
+ 2023-10-16 10:03:45,716 ----------------------------------------------------------------------------------------------------
132
+ 2023-10-16 10:03:45,716 EPOCH 4 done: loss 0.0686 - lr: 0.000033
133
+ 2023-10-16 10:03:50,390 DEV : loss 0.15915921330451965 - f1-score (micro avg) 0.7766
134
+ 2023-10-16 10:03:50,406 ----------------------------------------------------------------------------------------------------
135
+ 2023-10-16 10:03:59,089 epoch 5 - iter 178/1786 - loss 0.05022154 - time (sec): 8.68 - samples/sec: 2613.18 - lr: 0.000033 - momentum: 0.000000
136
+ 2023-10-16 10:04:07,970 epoch 5 - iter 356/1786 - loss 0.05090393 - time (sec): 17.56 - samples/sec: 2741.71 - lr: 0.000032 - momentum: 0.000000
137
+ 2023-10-16 10:04:16,988 epoch 5 - iter 534/1786 - loss 0.05106929 - time (sec): 26.58 - samples/sec: 2801.87 - lr: 0.000032 - momentum: 0.000000
138
+ 2023-10-16 10:04:25,594 epoch 5 - iter 712/1786 - loss 0.04974481 - time (sec): 35.19 - samples/sec: 2783.34 - lr: 0.000031 - momentum: 0.000000
139
+ 2023-10-16 10:04:34,524 epoch 5 - iter 890/1786 - loss 0.05195749 - time (sec): 44.12 - samples/sec: 2808.00 - lr: 0.000031 - momentum: 0.000000
140
+ 2023-10-16 10:04:43,144 epoch 5 - iter 1068/1786 - loss 0.05146219 - time (sec): 52.74 - samples/sec: 2780.35 - lr: 0.000030 - momentum: 0.000000
141
+ 2023-10-16 10:04:52,122 epoch 5 - iter 1246/1786 - loss 0.05143701 - time (sec): 61.72 - samples/sec: 2790.81 - lr: 0.000029 - momentum: 0.000000
142
+ 2023-10-16 10:05:01,020 epoch 5 - iter 1424/1786 - loss 0.05105708 - time (sec): 70.61 - samples/sec: 2786.63 - lr: 0.000029 - momentum: 0.000000
143
+ 2023-10-16 10:05:09,953 epoch 5 - iter 1602/1786 - loss 0.05263520 - time (sec): 79.55 - samples/sec: 2805.39 - lr: 0.000028 - momentum: 0.000000
144
+ 2023-10-16 10:05:18,626 epoch 5 - iter 1780/1786 - loss 0.05290404 - time (sec): 88.22 - samples/sec: 2811.00 - lr: 0.000028 - momentum: 0.000000
145
+ 2023-10-16 10:05:18,934 ----------------------------------------------------------------------------------------------------
146
+ 2023-10-16 10:05:18,935 EPOCH 5 done: loss 0.0529 - lr: 0.000028
147
+ 2023-10-16 10:05:23,061 DEV : loss 0.20985299348831177 - f1-score (micro avg) 0.7812
148
+ 2023-10-16 10:05:23,077 saving best model
149
+ 2023-10-16 10:05:23,560 ----------------------------------------------------------------------------------------------------
150
+ 2023-10-16 10:05:32,536 epoch 6 - iter 178/1786 - loss 0.04741637 - time (sec): 8.97 - samples/sec: 2965.00 - lr: 0.000027 - momentum: 0.000000
151
+ 2023-10-16 10:05:41,275 epoch 6 - iter 356/1786 - loss 0.04275169 - time (sec): 17.71 - samples/sec: 2872.22 - lr: 0.000027 - momentum: 0.000000
152
+ 2023-10-16 10:05:50,210 epoch 6 - iter 534/1786 - loss 0.03942651 - time (sec): 26.65 - samples/sec: 2818.41 - lr: 0.000026 - momentum: 0.000000
153
+ 2023-10-16 10:05:58,662 epoch 6 - iter 712/1786 - loss 0.04059826 - time (sec): 35.10 - samples/sec: 2830.73 - lr: 0.000026 - momentum: 0.000000
154
+ 2023-10-16 10:06:07,484 epoch 6 - iter 890/1786 - loss 0.04004597 - time (sec): 43.92 - samples/sec: 2814.49 - lr: 0.000025 - momentum: 0.000000
155
+ 2023-10-16 10:06:16,218 epoch 6 - iter 1068/1786 - loss 0.03891575 - time (sec): 52.66 - samples/sec: 2844.00 - lr: 0.000024 - momentum: 0.000000
156
+ 2023-10-16 10:06:24,675 epoch 6 - iter 1246/1786 - loss 0.03960494 - time (sec): 61.11 - samples/sec: 2830.47 - lr: 0.000024 - momentum: 0.000000
157
+ 2023-10-16 10:06:33,645 epoch 6 - iter 1424/1786 - loss 0.03949478 - time (sec): 70.08 - samples/sec: 2816.99 - lr: 0.000023 - momentum: 0.000000
158
+ 2023-10-16 10:06:43,058 epoch 6 - iter 1602/1786 - loss 0.03917810 - time (sec): 79.50 - samples/sec: 2797.23 - lr: 0.000023 - momentum: 0.000000
159
+ 2023-10-16 10:06:51,989 epoch 6 - iter 1780/1786 - loss 0.04050214 - time (sec): 88.43 - samples/sec: 2802.76 - lr: 0.000022 - momentum: 0.000000
160
+ 2023-10-16 10:06:52,248 ----------------------------------------------------------------------------------------------------
161
+ 2023-10-16 10:06:52,248 EPOCH 6 done: loss 0.0405 - lr: 0.000022
162
+ 2023-10-16 10:06:56,411 DEV : loss 0.1887538582086563 - f1-score (micro avg) 0.7869
163
+ 2023-10-16 10:06:56,427 saving best model
164
+ 2023-10-16 10:06:56,933 ----------------------------------------------------------------------------------------------------
165
+ 2023-10-16 10:07:05,796 epoch 7 - iter 178/1786 - loss 0.03361585 - time (sec): 8.86 - samples/sec: 2795.40 - lr: 0.000022 - momentum: 0.000000
166
+ 2023-10-16 10:07:14,528 epoch 7 - iter 356/1786 - loss 0.03329973 - time (sec): 17.59 - samples/sec: 2848.14 - lr: 0.000021 - momentum: 0.000000
167
+ 2023-10-16 10:07:23,285 epoch 7 - iter 534/1786 - loss 0.03326424 - time (sec): 26.35 - samples/sec: 2855.10 - lr: 0.000021 - momentum: 0.000000
168
+ 2023-10-16 10:07:31,995 epoch 7 - iter 712/1786 - loss 0.03160995 - time (sec): 35.06 - samples/sec: 2818.73 - lr: 0.000020 - momentum: 0.000000
169
+ 2023-10-16 10:07:40,594 epoch 7 - iter 890/1786 - loss 0.03005408 - time (sec): 43.66 - samples/sec: 2835.09 - lr: 0.000019 - momentum: 0.000000
170
+ 2023-10-16 10:07:49,327 epoch 7 - iter 1068/1786 - loss 0.03084739 - time (sec): 52.39 - samples/sec: 2832.70 - lr: 0.000019 - momentum: 0.000000
171
+ 2023-10-16 10:07:58,392 epoch 7 - iter 1246/1786 - loss 0.03032454 - time (sec): 61.45 - samples/sec: 2831.30 - lr: 0.000018 - momentum: 0.000000
172
+ 2023-10-16 10:08:06,931 epoch 7 - iter 1424/1786 - loss 0.03088217 - time (sec): 69.99 - samples/sec: 2818.38 - lr: 0.000018 - momentum: 0.000000
173
+ 2023-10-16 10:08:15,826 epoch 7 - iter 1602/1786 - loss 0.03073768 - time (sec): 78.89 - samples/sec: 2831.34 - lr: 0.000017 - momentum: 0.000000
174
+ 2023-10-16 10:08:24,577 epoch 7 - iter 1780/1786 - loss 0.03023599 - time (sec): 87.64 - samples/sec: 2829.60 - lr: 0.000017 - momentum: 0.000000
175
+ 2023-10-16 10:08:24,844 ----------------------------------------------------------------------------------------------------
176
+ 2023-10-16 10:08:24,844 EPOCH 7 done: loss 0.0302 - lr: 0.000017
177
+ 2023-10-16 10:08:29,610 DEV : loss 0.2058480978012085 - f1-score (micro avg) 0.7722
178
+ 2023-10-16 10:08:29,627 ----------------------------------------------------------------------------------------------------
179
+ 2023-10-16 10:08:38,694 epoch 8 - iter 178/1786 - loss 0.04352804 - time (sec): 9.07 - samples/sec: 2988.38 - lr: 0.000016 - momentum: 0.000000
180
+ 2023-10-16 10:08:47,551 epoch 8 - iter 356/1786 - loss 0.03156578 - time (sec): 17.92 - samples/sec: 2921.28 - lr: 0.000016 - momentum: 0.000000
181
+ 2023-10-16 10:08:56,308 epoch 8 - iter 534/1786 - loss 0.02706534 - time (sec): 26.68 - samples/sec: 2895.29 - lr: 0.000015 - momentum: 0.000000
182
+ 2023-10-16 10:09:04,820 epoch 8 - iter 712/1786 - loss 0.02702835 - time (sec): 35.19 - samples/sec: 2867.00 - lr: 0.000014 - momentum: 0.000000
183
+ 2023-10-16 10:09:13,393 epoch 8 - iter 890/1786 - loss 0.02599787 - time (sec): 43.76 - samples/sec: 2854.41 - lr: 0.000014 - momentum: 0.000000
184
+ 2023-10-16 10:09:22,260 epoch 8 - iter 1068/1786 - loss 0.02445266 - time (sec): 52.63 - samples/sec: 2819.22 - lr: 0.000013 - momentum: 0.000000
185
+ 2023-10-16 10:09:31,228 epoch 8 - iter 1246/1786 - loss 0.02430149 - time (sec): 61.60 - samples/sec: 2840.93 - lr: 0.000013 - momentum: 0.000000
186
+ 2023-10-16 10:09:40,085 epoch 8 - iter 1424/1786 - loss 0.02378568 - time (sec): 70.46 - samples/sec: 2827.21 - lr: 0.000012 - momentum: 0.000000
187
+ 2023-10-16 10:09:48,717 epoch 8 - iter 1602/1786 - loss 0.02346589 - time (sec): 79.09 - samples/sec: 2807.17 - lr: 0.000012 - momentum: 0.000000
188
+ 2023-10-16 10:09:57,557 epoch 8 - iter 1780/1786 - loss 0.02359126 - time (sec): 87.93 - samples/sec: 2821.74 - lr: 0.000011 - momentum: 0.000000
189
+ 2023-10-16 10:09:57,838 ----------------------------------------------------------------------------------------------------
190
+ 2023-10-16 10:09:57,839 EPOCH 8 done: loss 0.0235 - lr: 0.000011
191
+ 2023-10-16 10:10:02,026 DEV : loss 0.20242059230804443 - f1-score (micro avg) 0.8011
192
+ 2023-10-16 10:10:02,043 saving best model
193
+ 2023-10-16 10:10:02,569 ----------------------------------------------------------------------------------------------------
194
+ 2023-10-16 10:10:11,249 epoch 9 - iter 178/1786 - loss 0.02199362 - time (sec): 8.67 - samples/sec: 2936.70 - lr: 0.000011 - momentum: 0.000000
195
+ 2023-10-16 10:10:19,860 epoch 9 - iter 356/1786 - loss 0.01834317 - time (sec): 17.29 - samples/sec: 2843.56 - lr: 0.000010 - momentum: 0.000000
196
+ 2023-10-16 10:10:28,500 epoch 9 - iter 534/1786 - loss 0.01636593 - time (sec): 25.93 - samples/sec: 2848.80 - lr: 0.000009 - momentum: 0.000000
197
+ 2023-10-16 10:10:37,237 epoch 9 - iter 712/1786 - loss 0.01554274 - time (sec): 34.66 - samples/sec: 2869.85 - lr: 0.000009 - momentum: 0.000000
198
+ 2023-10-16 10:10:45,796 epoch 9 - iter 890/1786 - loss 0.01668982 - time (sec): 43.22 - samples/sec: 2843.23 - lr: 0.000008 - momentum: 0.000000
199
+ 2023-10-16 10:10:54,295 epoch 9 - iter 1068/1786 - loss 0.01736686 - time (sec): 51.72 - samples/sec: 2846.50 - lr: 0.000008 - momentum: 0.000000
200
+ 2023-10-16 10:11:03,452 epoch 9 - iter 1246/1786 - loss 0.01717748 - time (sec): 60.88 - samples/sec: 2822.73 - lr: 0.000007 - momentum: 0.000000
201
+ 2023-10-16 10:11:12,044 epoch 9 - iter 1424/1786 - loss 0.01657150 - time (sec): 69.47 - samples/sec: 2830.65 - lr: 0.000007 - momentum: 0.000000
202
+ 2023-10-16 10:11:20,876 epoch 9 - iter 1602/1786 - loss 0.01682811 - time (sec): 78.30 - samples/sec: 2823.29 - lr: 0.000006 - momentum: 0.000000
203
+ 2023-10-16 10:11:29,889 epoch 9 - iter 1780/1786 - loss 0.01701446 - time (sec): 87.31 - samples/sec: 2839.70 - lr: 0.000006 - momentum: 0.000000
204
+ 2023-10-16 10:11:30,162 ----------------------------------------------------------------------------------------------------
205
+ 2023-10-16 10:11:30,163 EPOCH 9 done: loss 0.0170 - lr: 0.000006
206
+ 2023-10-16 10:11:34,352 DEV : loss 0.20268410444259644 - f1-score (micro avg) 0.7997
207
+ 2023-10-16 10:11:34,368 ----------------------------------------------------------------------------------------------------
208
+ 2023-10-16 10:11:43,098 epoch 10 - iter 178/1786 - loss 0.01040220 - time (sec): 8.73 - samples/sec: 2763.04 - lr: 0.000005 - momentum: 0.000000
209
+ 2023-10-16 10:11:51,917 epoch 10 - iter 356/1786 - loss 0.00841270 - time (sec): 17.55 - samples/sec: 2818.53 - lr: 0.000004 - momentum: 0.000000
210
+ 2023-10-16 10:12:00,800 epoch 10 - iter 534/1786 - loss 0.01266186 - time (sec): 26.43 - samples/sec: 2818.85 - lr: 0.000004 - momentum: 0.000000
211
+ 2023-10-16 10:12:09,453 epoch 10 - iter 712/1786 - loss 0.01314607 - time (sec): 35.08 - samples/sec: 2838.64 - lr: 0.000003 - momentum: 0.000000
212
+ 2023-10-16 10:12:18,110 epoch 10 - iter 890/1786 - loss 0.01322097 - time (sec): 43.74 - samples/sec: 2848.20 - lr: 0.000003 - momentum: 0.000000
213
+ 2023-10-16 10:12:26,982 epoch 10 - iter 1068/1786 - loss 0.01308993 - time (sec): 52.61 - samples/sec: 2851.51 - lr: 0.000002 - momentum: 0.000000
214
+ 2023-10-16 10:12:35,888 epoch 10 - iter 1246/1786 - loss 0.01241075 - time (sec): 61.52 - samples/sec: 2855.40 - lr: 0.000002 - momentum: 0.000000
215
+ 2023-10-16 10:12:44,553 epoch 10 - iter 1424/1786 - loss 0.01264047 - time (sec): 70.18 - samples/sec: 2847.46 - lr: 0.000001 - momentum: 0.000000
216
+ 2023-10-16 10:12:53,476 epoch 10 - iter 1602/1786 - loss 0.01178041 - time (sec): 79.11 - samples/sec: 2839.80 - lr: 0.000001 - momentum: 0.000000
217
+ 2023-10-16 10:13:02,172 epoch 10 - iter 1780/1786 - loss 0.01160785 - time (sec): 87.80 - samples/sec: 2826.86 - lr: 0.000000 - momentum: 0.000000
218
+ 2023-10-16 10:13:02,437 ----------------------------------------------------------------------------------------------------
219
+ 2023-10-16 10:13:02,438 EPOCH 10 done: loss 0.0116 - lr: 0.000000
220
+ 2023-10-16 10:13:07,862 DEV : loss 0.207749143242836 - f1-score (micro avg) 0.8078
221
+ 2023-10-16 10:13:07,880 saving best model
222
+ 2023-10-16 10:13:09,016 ----------------------------------------------------------------------------------------------------
223
+ 2023-10-16 10:13:09,018 Loading model from best epoch ...
224
+ 2023-10-16 10:13:11,262 SequenceTagger predicts: Dictionary with 17 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, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
225
+ 2023-10-16 10:13:21,169
226
+ Results:
227
+ - F-score (micro) 0.6891
228
+ - F-score (macro) 0.607
229
+ - Accuracy 0.5469
230
+
231
+ By class:
232
+ precision recall f1-score support
233
+
234
+ LOC 0.7334 0.6758 0.7034 1095
235
+ PER 0.7644 0.7727 0.7686 1012
236
+ ORG 0.4248 0.5462 0.4779 357
237
+ HumanProd 0.3729 0.6667 0.4783 33
238
+
239
+ micro avg 0.6820 0.6964 0.6891 2497
240
+ macro avg 0.5739 0.6654 0.6070 2497
241
+ weighted avg 0.6971 0.6964 0.6946 2497
242
+
243
+ 2023-10-16 10:13:21,170 ----------------------------------------------------------------------------------------------------