Hugging Face's logo

stefan-it
/
hmbench-letemps-fr-hmbert-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1

Token Classification
Flair
PyTorch
French
sequence-tagger-model
Model card Files Community
1
stefan-it commited on
Commit
d712a65
•
1 Parent(s): 0b4b492

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 +241 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:af855e2e64c9fa9cfc481f3bb28dd3fe76546e5dfc1a3e7fc6690f8eccd32326
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 19:56:40 0.0000 0.2674 0.1105 0.4971 0.6922 0.5787 0.4158
3
+ 2 19:58:41 0.0000 0.0850 0.1115 0.5731 0.6911 0.6266 0.4593
4
+ 3 20:00:44 0.0000 0.0612 0.1642 0.5412 0.7666 0.6345 0.4722
5
+ 4 20:02:45 0.0000 0.0439 0.2740 0.5250 0.8181 0.6395 0.4799
6
+ 5 20:04:44 0.0000 0.0326 0.3013 0.5185 0.7689 0.6194 0.4590
7
+ 6 20:06:41 0.0000 0.0229 0.3020 0.5663 0.7277 0.6370 0.4753
8
+ 7 20:08:38 0.0000 0.0159 0.3790 0.5544 0.7872 0.6506 0.4893
9
+ 8 20:10:35 0.0000 0.0102 0.3863 0.5637 0.7746 0.6525 0.4920
10
+ 9 20:12:33 0.0000 0.0067 0.4053 0.5465 0.7792 0.6425 0.4826
11
+ 10 20:14:31 0.0000 0.0040 0.4170 0.5482 0.7746 0.6420 0.4822
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,241 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-14 19:54:42,062 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-14 19:54:42,063 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 19:54:42,063 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-14 19:54:42,064 MultiCorpus: 14465 train + 1392 dev + 2432 test sentences
52
+ - NER_HIPE_2022 Corpus: 14465 train + 1392 dev + 2432 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/letemps/fr/with_doc_seperator
53
+ 2023-10-14 19:54:42,064 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-14 19:54:42,064 Train: 14465 sentences
55
+ 2023-10-14 19:54:42,064 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-14 19:54:42,064 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-14 19:54:42,064 Training Params:
58
+ 2023-10-14 19:54:42,064 - learning_rate: "5e-05"
59
+ 2023-10-14 19:54:42,064 - mini_batch_size: "8"
60
+ 2023-10-14 19:54:42,064 - max_epochs: "10"
61
+ 2023-10-14 19:54:42,064 - shuffle: "True"
62
+ 2023-10-14 19:54:42,064 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-14 19:54:42,064 Plugins:
64
+ 2023-10-14 19:54:42,064 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-14 19:54:42,064 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-14 19:54:42,064 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-14 19:54:42,064 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-14 19:54:42,064 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-14 19:54:42,064 Computation:
70
+ 2023-10-14 19:54:42,064 - compute on device: cuda:0
71
+ 2023-10-14 19:54:42,064 - embedding storage: none
72
+ 2023-10-14 19:54:42,064 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-14 19:54:42,064 Model training base path: "hmbench-letemps/fr-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1"
74
+ 2023-10-14 19:54:42,064 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-14 19:54:42,064 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-14 19:54:53,413 epoch 1 - iter 180/1809 - loss 1.52200951 - time (sec): 11.35 - samples/sec: 3372.05 - lr: 0.000005 - momentum: 0.000000
77
+ 2023-10-14 19:55:04,616 epoch 1 - iter 360/1809 - loss 0.86831072 - time (sec): 22.55 - samples/sec: 3369.01 - lr: 0.000010 - momentum: 0.000000
78
+ 2023-10-14 19:55:16,043 epoch 1 - iter 540/1809 - loss 0.63472441 - time (sec): 33.98 - samples/sec: 3341.10 - lr: 0.000015 - momentum: 0.000000
79
+ 2023-10-14 19:55:27,315 epoch 1 - iter 720/1809 - loss 0.50459405 - time (sec): 45.25 - samples/sec: 3369.27 - lr: 0.000020 - momentum: 0.000000
80
+ 2023-10-14 19:55:38,557 epoch 1 - iter 900/1809 - loss 0.42878307 - time (sec): 56.49 - samples/sec: 3361.57 - lr: 0.000025 - momentum: 0.000000
81
+ 2023-10-14 19:55:49,637 epoch 1 - iter 1080/1809 - loss 0.37737077 - time (sec): 67.57 - samples/sec: 3365.97 - lr: 0.000030 - momentum: 0.000000
82
+ 2023-10-14 19:56:00,758 epoch 1 - iter 1260/1809 - loss 0.33903493 - time (sec): 78.69 - samples/sec: 3358.41 - lr: 0.000035 - momentum: 0.000000
83
+ 2023-10-14 19:56:11,947 epoch 1 - iter 1440/1809 - loss 0.30859986 - time (sec): 89.88 - samples/sec: 3380.66 - lr: 0.000040 - momentum: 0.000000
84
+ 2023-10-14 19:56:23,049 epoch 1 - iter 1620/1809 - loss 0.28703340 - time (sec): 100.98 - samples/sec: 3376.84 - lr: 0.000045 - momentum: 0.000000
85
+ 2023-10-14 19:56:34,342 epoch 1 - iter 1800/1809 - loss 0.26826484 - time (sec): 112.28 - samples/sec: 3369.33 - lr: 0.000050 - momentum: 0.000000
86
+ 2023-10-14 19:56:34,848 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-14 19:56:34,848 EPOCH 1 done: loss 0.2674 - lr: 0.000050
88
+ 2023-10-14 19:56:40,094 DEV : loss 0.11045825481414795 - f1-score (micro avg) 0.5787
89
+ 2023-10-14 19:56:40,133 saving best model
90
+ 2023-10-14 19:56:40,519 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-14 19:56:52,597 epoch 2 - iter 180/1809 - loss 0.08253900 - time (sec): 12.08 - samples/sec: 3148.77 - lr: 0.000049 - momentum: 0.000000
92
+ 2023-10-14 19:57:03,849 epoch 2 - iter 360/1809 - loss 0.08121285 - time (sec): 23.33 - samples/sec: 3260.77 - lr: 0.000049 - momentum: 0.000000
93
+ 2023-10-14 19:57:15,405 epoch 2 - iter 540/1809 - loss 0.08522193 - time (sec): 34.88 - samples/sec: 3285.78 - lr: 0.000048 - momentum: 0.000000
94
+ 2023-10-14 19:57:26,475 epoch 2 - iter 720/1809 - loss 0.08730972 - time (sec): 45.95 - samples/sec: 3299.17 - lr: 0.000048 - momentum: 0.000000
95
+ 2023-10-14 19:57:37,666 epoch 2 - iter 900/1809 - loss 0.08799984 - time (sec): 57.15 - samples/sec: 3299.11 - lr: 0.000047 - momentum: 0.000000
96
+ 2023-10-14 19:57:49,179 epoch 2 - iter 1080/1809 - loss 0.08647646 - time (sec): 68.66 - samples/sec: 3309.86 - lr: 0.000047 - momentum: 0.000000
97
+ 2023-10-14 19:58:00,879 epoch 2 - iter 1260/1809 - loss 0.08797357 - time (sec): 80.36 - samples/sec: 3303.56 - lr: 0.000046 - momentum: 0.000000
98
+ 2023-10-14 19:58:12,292 epoch 2 - iter 1440/1809 - loss 0.08580293 - time (sec): 91.77 - samples/sec: 3298.04 - lr: 0.000046 - momentum: 0.000000
99
+ 2023-10-14 19:58:23,555 epoch 2 - iter 1620/1809 - loss 0.08520800 - time (sec): 103.03 - samples/sec: 3307.44 - lr: 0.000045 - momentum: 0.000000
100
+ 2023-10-14 19:58:34,969 epoch 2 - iter 1800/1809 - loss 0.08503746 - time (sec): 114.45 - samples/sec: 3303.56 - lr: 0.000044 - momentum: 0.000000
101
+ 2023-10-14 19:58:35,478 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-14 19:58:35,478 EPOCH 2 done: loss 0.0850 - lr: 0.000044
103
+ 2023-10-14 19:58:41,049 DEV : loss 0.1115032285451889 - f1-score (micro avg) 0.6266
104
+ 2023-10-14 19:58:41,090 saving best model
105
+ 2023-10-14 19:58:41,553 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-14 19:58:52,347 epoch 3 - iter 180/1809 - loss 0.06275948 - time (sec): 10.79 - samples/sec: 3279.08 - lr: 0.000044 - momentum: 0.000000
107
+ 2023-10-14 19:59:03,595 epoch 3 - iter 360/1809 - loss 0.05970878 - time (sec): 22.04 - samples/sec: 3333.30 - lr: 0.000043 - momentum: 0.000000
108
+ 2023-10-14 19:59:14,873 epoch 3 - iter 540/1809 - loss 0.05985749 - time (sec): 33.31 - samples/sec: 3337.20 - lr: 0.000043 - momentum: 0.000000
109
+ 2023-10-14 19:59:26,115 epoch 3 - iter 720/1809 - loss 0.05962676 - time (sec): 44.55 - samples/sec: 3373.45 - lr: 0.000042 - momentum: 0.000000
110
+ 2023-10-14 19:59:38,946 epoch 3 - iter 900/1809 - loss 0.05858862 - time (sec): 57.39 - samples/sec: 3272.64 - lr: 0.000042 - momentum: 0.000000
111
+ 2023-10-14 19:59:51,020 epoch 3 - iter 1080/1809 - loss 0.06104670 - time (sec): 69.46 - samples/sec: 3263.46 - lr: 0.000041 - momentum: 0.000000
112
+ 2023-10-14 20:00:02,728 epoch 3 - iter 1260/1809 - loss 0.06035191 - time (sec): 81.17 - samples/sec: 3266.60 - lr: 0.000041 - momentum: 0.000000
113
+ 2023-10-14 20:00:14,305 epoch 3 - iter 1440/1809 - loss 0.06077928 - time (sec): 92.75 - samples/sec: 3263.60 - lr: 0.000040 - momentum: 0.000000
114
+ 2023-10-14 20:00:25,873 epoch 3 - iter 1620/1809 - loss 0.06099591 - time (sec): 104.31 - samples/sec: 3258.39 - lr: 0.000039 - momentum: 0.000000
115
+ 2023-10-14 20:00:38,096 epoch 3 - iter 1800/1809 - loss 0.06131526 - time (sec): 116.54 - samples/sec: 3247.12 - lr: 0.000039 - momentum: 0.000000
116
+ 2023-10-14 20:00:38,612 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-14 20:00:38,613 EPOCH 3 done: loss 0.0612 - lr: 0.000039
118
+ 2023-10-14 20:00:44,254 DEV : loss 0.16415372490882874 - f1-score (micro avg) 0.6345
119
+ 2023-10-14 20:00:44,285 saving best model
120
+ 2023-10-14 20:00:44,797 ----------------------------------------------------------------------------------------------------
121
+ 2023-10-14 20:00:56,704 epoch 4 - iter 180/1809 - loss 0.04050032 - time (sec): 11.90 - samples/sec: 3289.62 - lr: 0.000038 - momentum: 0.000000
122
+ 2023-10-14 20:01:08,146 epoch 4 - iter 360/1809 - loss 0.04638374 - time (sec): 23.35 - samples/sec: 3270.50 - lr: 0.000038 - momentum: 0.000000
123
+ 2023-10-14 20:01:19,817 epoch 4 - iter 540/1809 - loss 0.04256396 - time (sec): 35.02 - samples/sec: 3250.14 - lr: 0.000037 - momentum: 0.000000
124
+ 2023-10-14 20:01:30,997 epoch 4 - iter 720/1809 - loss 0.04168535 - time (sec): 46.20 - samples/sec: 3253.48 - lr: 0.000037 - momentum: 0.000000
125
+ 2023-10-14 20:01:42,070 epoch 4 - iter 900/1809 - loss 0.04138962 - time (sec): 57.27 - samples/sec: 3294.92 - lr: 0.000036 - momentum: 0.000000
126
+ 2023-10-14 20:01:53,028 epoch 4 - iter 1080/1809 - loss 0.04200716 - time (sec): 68.23 - samples/sec: 3313.41 - lr: 0.000036 - momentum: 0.000000
127
+ 2023-10-14 20:02:04,495 epoch 4 - iter 1260/1809 - loss 0.04192659 - time (sec): 79.69 - samples/sec: 3316.82 - lr: 0.000035 - momentum: 0.000000
128
+ 2023-10-14 20:02:15,701 epoch 4 - iter 1440/1809 - loss 0.04206329 - time (sec): 90.90 - samples/sec: 3329.95 - lr: 0.000034 - momentum: 0.000000
129
+ 2023-10-14 20:02:26,693 epoch 4 - iter 1620/1809 - loss 0.04322145 - time (sec): 101.89 - samples/sec: 3341.18 - lr: 0.000034 - momentum: 0.000000
130
+ 2023-10-14 20:02:37,872 epoch 4 - iter 1800/1809 - loss 0.04389061 - time (sec): 113.07 - samples/sec: 3343.24 - lr: 0.000033 - momentum: 0.000000
131
+ 2023-10-14 20:02:38,448 ----------------------------------------------------------------------------------------------------
132
+ 2023-10-14 20:02:38,448 EPOCH 4 done: loss 0.0439 - lr: 0.000033
133
+ 2023-10-14 20:02:45,397 DEV : loss 0.2739905118942261 - f1-score (micro avg) 0.6395
134
+ 2023-10-14 20:02:45,428 saving best model
135
+ 2023-10-14 20:02:45,967 ----------------------------------------------------------------------------------------------------
136
+ 2023-10-14 20:02:57,970 epoch 5 - iter 180/1809 - loss 0.03274883 - time (sec): 12.00 - samples/sec: 3275.44 - lr: 0.000033 - momentum: 0.000000
137
+ 2023-10-14 20:03:09,178 epoch 5 - iter 360/1809 - loss 0.03376456 - time (sec): 23.21 - samples/sec: 3310.77 - lr: 0.000032 - momentum: 0.000000
138
+ 2023-10-14 20:03:20,340 epoch 5 - iter 540/1809 - loss 0.03372631 - time (sec): 34.37 - samples/sec: 3343.03 - lr: 0.000032 - momentum: 0.000000
139
+ 2023-10-14 20:03:31,437 epoch 5 - iter 720/1809 - loss 0.03416861 - time (sec): 45.47 - samples/sec: 3332.71 - lr: 0.000031 - momentum: 0.000000
140
+ 2023-10-14 20:03:42,821 epoch 5 - iter 900/1809 - loss 0.03305077 - time (sec): 56.85 - samples/sec: 3350.33 - lr: 0.000031 - momentum: 0.000000
141
+ 2023-10-14 20:03:54,085 epoch 5 - iter 1080/1809 - loss 0.03328099 - time (sec): 68.11 - samples/sec: 3360.47 - lr: 0.000030 - momentum: 0.000000
142
+ 2023-10-14 20:04:05,270 epoch 5 - iter 1260/1809 - loss 0.03312214 - time (sec): 79.30 - samples/sec: 3368.80 - lr: 0.000029 - momentum: 0.000000
143
+ 2023-10-14 20:04:16,232 epoch 5 - iter 1440/1809 - loss 0.03276691 - time (sec): 90.26 - samples/sec: 3368.73 - lr: 0.000029 - momentum: 0.000000
144
+ 2023-10-14 20:04:26,972 epoch 5 - iter 1620/1809 - loss 0.03238180 - time (sec): 101.00 - samples/sec: 3385.30 - lr: 0.000028 - momentum: 0.000000
145
+ 2023-10-14 20:04:37,686 epoch 5 - iter 1800/1809 - loss 0.03256257 - time (sec): 111.71 - samples/sec: 3385.96 - lr: 0.000028 - momentum: 0.000000
146
+ 2023-10-14 20:04:38,165 ----------------------------------------------------------------------------------------------------
147
+ 2023-10-14 20:04:38,165 EPOCH 5 done: loss 0.0326 - lr: 0.000028
148
+ 2023-10-14 20:04:44,446 DEV : loss 0.30132314562797546 - f1-score (micro avg) 0.6194
149
+ 2023-10-14 20:04:44,476 ----------------------------------------------------------------------------------------------------
150
+ 2023-10-14 20:04:55,618 epoch 6 - iter 180/1809 - loss 0.02000950 - time (sec): 11.14 - samples/sec: 3392.54 - lr: 0.000027 - momentum: 0.000000
151
+ 2023-10-14 20:05:06,487 epoch 6 - iter 360/1809 - loss 0.02102558 - time (sec): 22.01 - samples/sec: 3410.42 - lr: 0.000027 - momentum: 0.000000
152
+ 2023-10-14 20:05:17,381 epoch 6 - iter 540/1809 - loss 0.02110462 - time (sec): 32.90 - samples/sec: 3392.68 - lr: 0.000026 - momentum: 0.000000
153
+ 2023-10-14 20:05:28,393 epoch 6 - iter 720/1809 - loss 0.02298616 - time (sec): 43.92 - samples/sec: 3398.32 - lr: 0.000026 - momentum: 0.000000
154
+ 2023-10-14 20:05:39,567 epoch 6 - iter 900/1809 - loss 0.02202486 - time (sec): 55.09 - samples/sec: 3409.55 - lr: 0.000025 - momentum: 0.000000
155
+ 2023-10-14 20:05:50,592 epoch 6 - iter 1080/1809 - loss 0.02290144 - time (sec): 66.11 - samples/sec: 3416.70 - lr: 0.000024 - momentum: 0.000000
156
+ 2023-10-14 20:06:01,594 epoch 6 - iter 1260/1809 - loss 0.02273488 - time (sec): 77.12 - samples/sec: 3416.38 - lr: 0.000024 - momentum: 0.000000
157
+ 2023-10-14 20:06:12,951 epoch 6 - iter 1440/1809 - loss 0.02310618 - time (sec): 88.47 - samples/sec: 3424.89 - lr: 0.000023 - momentum: 0.000000
158
+ 2023-10-14 20:06:24,004 epoch 6 - iter 1620/1809 - loss 0.02272973 - time (sec): 99.53 - samples/sec: 3414.79 - lr: 0.000023 - momentum: 0.000000
159
+ 2023-10-14 20:06:34,954 epoch 6 - iter 1800/1809 - loss 0.02299704 - time (sec): 110.48 - samples/sec: 3422.82 - lr: 0.000022 - momentum: 0.000000
160
+ 2023-10-14 20:06:35,473 ----------------------------------------------------------------------------------------------------
161
+ 2023-10-14 20:06:35,473 EPOCH 6 done: loss 0.0229 - lr: 0.000022
162
+ 2023-10-14 20:06:41,937 DEV : loss 0.30204930901527405 - f1-score (micro avg) 0.637
163
+ 2023-10-14 20:06:41,970 ----------------------------------------------------------------------------------------------------
164
+ 2023-10-14 20:06:53,076 epoch 7 - iter 180/1809 - loss 0.01920066 - time (sec): 11.10 - samples/sec: 3480.61 - lr: 0.000022 - momentum: 0.000000
165
+ 2023-10-14 20:07:04,403 epoch 7 - iter 360/1809 - loss 0.01732531 - time (sec): 22.43 - samples/sec: 3423.94 - lr: 0.000021 - momentum: 0.000000
166
+ 2023-10-14 20:07:15,376 epoch 7 - iter 540/1809 - loss 0.01693909 - time (sec): 33.40 - samples/sec: 3427.04 - lr: 0.000021 - momentum: 0.000000
167
+ 2023-10-14 20:07:26,346 epoch 7 - iter 720/1809 - loss 0.01681882 - time (sec): 44.37 - samples/sec: 3438.61 - lr: 0.000020 - momentum: 0.000000
168
+ 2023-10-14 20:07:37,415 epoch 7 - iter 900/1809 - loss 0.01628210 - time (sec): 55.44 - samples/sec: 3429.77 - lr: 0.000019 - momentum: 0.000000
169
+ 2023-10-14 20:07:48,365 epoch 7 - iter 1080/1809 - loss 0.01625405 - time (sec): 66.39 - samples/sec: 3429.63 - lr: 0.000019 - momentum: 0.000000
170
+ 2023-10-14 20:07:59,296 epoch 7 - iter 1260/1809 - loss 0.01605560 - time (sec): 77.33 - samples/sec: 3432.14 - lr: 0.000018 - momentum: 0.000000
171
+ 2023-10-14 20:08:10,313 epoch 7 - iter 1440/1809 - loss 0.01618800 - time (sec): 88.34 - samples/sec: 3433.08 - lr: 0.000018 - momentum: 0.000000
172
+ 2023-10-14 20:08:21,163 epoch 7 - iter 1620/1809 - loss 0.01610626 - time (sec): 99.19 - samples/sec: 3433.41 - lr: 0.000017 - momentum: 0.000000
173
+ 2023-10-14 20:08:31,856 epoch 7 - iter 1800/1809 - loss 0.01585667 - time (sec): 109.88 - samples/sec: 3441.41 - lr: 0.000017 - momentum: 0.000000
174
+ 2023-10-14 20:08:32,342 ----------------------------------------------------------------------------------------------------
175
+ 2023-10-14 20:08:32,342 EPOCH 7 done: loss 0.0159 - lr: 0.000017
176
+ 2023-10-14 20:08:38,752 DEV : loss 0.37900328636169434 - f1-score (micro avg) 0.6506
177
+ 2023-10-14 20:08:38,787 saving best model
178
+ 2023-10-14 20:08:39,292 ----------------------------------------------------------------------------------------------------
179
+ 2023-10-14 20:08:50,043 epoch 8 - iter 180/1809 - loss 0.01074333 - time (sec): 10.75 - samples/sec: 3429.58 - lr: 0.000016 - momentum: 0.000000
180
+ 2023-10-14 20:09:01,139 epoch 8 - iter 360/1809 - loss 0.01024638 - time (sec): 21.85 - samples/sec: 3426.84 - lr: 0.000016 - momentum: 0.000000
181
+ 2023-10-14 20:09:12,250 epoch 8 - iter 540/1809 - loss 0.00991737 - time (sec): 32.96 - samples/sec: 3443.87 - lr: 0.000015 - momentum: 0.000000
182
+ 2023-10-14 20:09:23,339 epoch 8 - iter 720/1809 - loss 0.01001735 - time (sec): 44.05 - samples/sec: 3434.64 - lr: 0.000014 - momentum: 0.000000
183
+ 2023-10-14 20:09:34,066 epoch 8 - iter 900/1809 - loss 0.01029062 - time (sec): 54.77 - samples/sec: 3440.98 - lr: 0.000014 - momentum: 0.000000
184
+ 2023-10-14 20:09:44,738 epoch 8 - iter 1080/1809 - loss 0.01096769 - time (sec): 65.44 - samples/sec: 3422.83 - lr: 0.000013 - momentum: 0.000000
185
+ 2023-10-14 20:09:56,318 epoch 8 - iter 1260/1809 - loss 0.01096529 - time (sec): 77.02 - samples/sec: 3418.87 - lr: 0.000013 - momentum: 0.000000
186
+ 2023-10-14 20:10:07,341 epoch 8 - iter 1440/1809 - loss 0.01043110 - time (sec): 88.05 - samples/sec: 3414.98 - lr: 0.000012 - momentum: 0.000000
187
+ 2023-10-14 20:10:18,516 epoch 8 - iter 1620/1809 - loss 0.01040260 - time (sec): 99.22 - samples/sec: 3419.48 - lr: 0.000012 - momentum: 0.000000
188
+ 2023-10-14 20:10:29,408 epoch 8 - iter 1800/1809 - loss 0.01021177 - time (sec): 110.11 - samples/sec: 3430.44 - lr: 0.000011 - momentum: 0.000000
189
+ 2023-10-14 20:10:30,023 ----------------------------------------------------------------------------------------------------
190
+ 2023-10-14 20:10:30,023 EPOCH 8 done: loss 0.0102 - lr: 0.000011
191
+ 2023-10-14 20:10:35,669 DEV : loss 0.386315256357193 - f1-score (micro avg) 0.6525
192
+ 2023-10-14 20:10:35,702 saving best model
193
+ 2023-10-14 20:10:36,220 ----------------------------------------------------------------------------------------------------
194
+ 2023-10-14 20:10:48,463 epoch 9 - iter 180/1809 - loss 0.00502273 - time (sec): 12.24 - samples/sec: 3086.65 - lr: 0.000011 - momentum: 0.000000
195
+ 2023-10-14 20:10:59,887 epoch 9 - iter 360/1809 - loss 0.00783396 - time (sec): 23.66 - samples/sec: 3250.37 - lr: 0.000010 - momentum: 0.000000
196
+ 2023-10-14 20:11:11,027 epoch 9 - iter 540/1809 - loss 0.00740246 - time (sec): 34.80 - samples/sec: 3308.71 - lr: 0.000009 - momentum: 0.000000
197
+ 2023-10-14 20:11:21,961 epoch 9 - iter 720/1809 - loss 0.00813486 - time (sec): 45.74 - samples/sec: 3318.19 - lr: 0.000009 - momentum: 0.000000
198
+ 2023-10-14 20:11:32,832 epoch 9 - iter 900/1809 - loss 0.00721541 - time (sec): 56.61 - samples/sec: 3340.81 - lr: 0.000008 - momentum: 0.000000
199
+ 2023-10-14 20:11:43,695 epoch 9 - iter 1080/1809 - loss 0.00729423 - time (sec): 67.47 - samples/sec: 3357.17 - lr: 0.000008 - momentum: 0.000000
200
+ 2023-10-14 20:11:54,696 epoch 9 - iter 1260/1809 - loss 0.00715794 - time (sec): 78.47 - samples/sec: 3355.65 - lr: 0.000007 - momentum: 0.000000
201
+ 2023-10-14 20:12:05,736 epoch 9 - iter 1440/1809 - loss 0.00685492 - time (sec): 89.51 - samples/sec: 3361.23 - lr: 0.000007 - momentum: 0.000000
202
+ 2023-10-14 20:12:16,672 epoch 9 - iter 1620/1809 - loss 0.00683880 - time (sec): 100.45 - samples/sec: 3379.37 - lr: 0.000006 - momentum: 0.000000
203
+ 2023-10-14 20:12:27,522 epoch 9 - iter 1800/1809 - loss 0.00673882 - time (sec): 111.30 - samples/sec: 3400.19 - lr: 0.000006 - momentum: 0.000000
204
+ 2023-10-14 20:12:28,049 ----------------------------------------------------------------------------------------------------
205
+ 2023-10-14 20:12:28,050 EPOCH 9 done: loss 0.0067 - lr: 0.000006
206
+ 2023-10-14 20:12:33,774 DEV : loss 0.40529340505599976 - f1-score (micro avg) 0.6425
207
+ 2023-10-14 20:12:33,810 ----------------------------------------------------------------------------------------------------
208
+ 2023-10-14 20:12:44,875 epoch 10 - iter 180/1809 - loss 0.00252138 - time (sec): 11.06 - samples/sec: 3459.07 - lr: 0.000005 - momentum: 0.000000
209
+ 2023-10-14 20:12:56,013 epoch 10 - iter 360/1809 - loss 0.00344089 - time (sec): 22.20 - samples/sec: 3465.51 - lr: 0.000004 - momentum: 0.000000
210
+ 2023-10-14 20:13:07,057 epoch 10 - iter 540/1809 - loss 0.00356348 - time (sec): 33.25 - samples/sec: 3425.09 - lr: 0.000004 - momentum: 0.000000
211
+ 2023-10-14 20:13:17,866 epoch 10 - iter 720/1809 - loss 0.00377280 - time (sec): 44.05 - samples/sec: 3445.07 - lr: 0.000003 - momentum: 0.000000
212
+ 2023-10-14 20:13:29,123 epoch 10 - iter 900/1809 - loss 0.00362108 - time (sec): 55.31 - samples/sec: 3436.69 - lr: 0.000003 - momentum: 0.000000
213
+ 2023-10-14 20:13:40,147 epoch 10 - iter 1080/1809 - loss 0.00375605 - time (sec): 66.34 - samples/sec: 3438.83 - lr: 0.000002 - momentum: 0.000000
214
+ 2023-10-14 20:13:50,911 epoch 10 - iter 1260/1809 - loss 0.00378037 - time (sec): 77.10 - samples/sec: 3430.31 - lr: 0.000002 - momentum: 0.000000
215
+ 2023-10-14 20:14:02,469 epoch 10 - iter 1440/1809 - loss 0.00397690 - time (sec): 88.66 - samples/sec: 3384.85 - lr: 0.000001 - momentum: 0.000000
216
+ 2023-10-14 20:14:13,575 epoch 10 - iter 1620/1809 - loss 0.00377560 - time (sec): 99.76 - samples/sec: 3395.83 - lr: 0.000001 - momentum: 0.000000
217
+ 2023-10-14 20:14:25,207 epoch 10 - iter 1800/1809 - loss 0.00397882 - time (sec): 111.40 - samples/sec: 3395.03 - lr: 0.000000 - momentum: 0.000000
218
+ 2023-10-14 20:14:25,715 ----------------------------------------------------------------------------------------------------
219
+ 2023-10-14 20:14:25,716 EPOCH 10 done: loss 0.0040 - lr: 0.000000
220
+ 2023-10-14 20:14:31,605 DEV : loss 0.41697490215301514 - f1-score (micro avg) 0.642
221
+ 2023-10-14 20:14:32,074 ----------------------------------------------------------------------------------------------------
222
+ 2023-10-14 20:14:32,076 Loading model from best epoch ...
223
+ 2023-10-14 20:14:33,753 SequenceTagger predicts: Dictionary with 13 tags: O, S-loc, B-loc, E-loc, I-loc, S-pers, B-pers, E-pers, I-pers, S-org, B-org, E-org, I-org
224
+ 2023-10-14 20:14:42,798
225
+ Results:
226
+ - F-score (micro) 0.6298
227
+ - F-score (macro) 0.4749
228
+ - Accuracy 0.4735
229
+
230
+ By class:
231
+ precision recall f1-score support
232
+
233
+ loc 0.6396 0.7327 0.6830 591
234
+ pers 0.5278 0.7451 0.6179 357
235
+ org 0.2059 0.0886 0.1239 79
236
+
237
+ micro avg 0.5811 0.6874 0.6298 1027
238
+ macro avg 0.4577 0.5221 0.4749 1027
239
+ weighted avg 0.5674 0.6874 0.6173 1027
240
+
241
+ 2023-10-14 20:14:42,799 ----------------------------------------------------------------------------------------------------