Hugging Face's logo

stefan-it
/
hmbench-letemps-fr-hmbert-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4

Token Classification
Flair
PyTorch
French
sequence-tagger-model
Model card Files Community
1
stefan-it commited on
Commit
6c5de33
·
1 Parent(s): 9c9927b

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 +240 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:26651454e7ab194684b6480edfe8a2ee7368c472d4c617c31695aa20cf43a864
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 00:56:24 0.0000 0.2506 0.1209 0.4869 0.7643 0.5948 0.4307
3
+ 2 00:58:23 0.0000 0.0843 0.1270 0.5262 0.8032 0.6359 0.4746
4
+ 3 01:00:23 0.0000 0.0604 0.1630 0.5248 0.7872 0.6297 0.4703
5
+ 4 01:02:21 0.0000 0.0471 0.2472 0.5364 0.7849 0.6373 0.4754
6
+ 5 01:04:19 0.0000 0.0334 0.2819 0.5281 0.7860 0.6317 0.4731
7
+ 6 01:06:17 0.0000 0.0233 0.3204 0.5577 0.7243 0.6302 0.4665
8
+ 7 01:08:15 0.0000 0.0172 0.3245 0.5541 0.7563 0.6396 0.4804
9
+ 8 01:10:13 0.0000 0.0126 0.3657 0.5463 0.7620 0.6364 0.4757
10
+ 9 01:12:11 0.0000 0.0077 0.3877 0.5453 0.7712 0.6389 0.4773
11
+ 10 01:14:06 0.0000 0.0053 0.3943 0.5533 0.7780 0.6467 0.4854
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,240 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-15 00:54:28,384 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-15 00:54:28,385 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
+ 2023-10-15 00:54:28,385 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-15 00:54:28,386 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
+ 2023-10-15 00:54:28,386 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-15 00:54:28,386 Train: 14465 sentences
55
+ 2023-10-15 00:54:28,386 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-15 00:54:28,386 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-15 00:54:28,386 Training Params:
58
+ 2023-10-15 00:54:28,386 - learning_rate: "5e-05"
59
+ 2023-10-15 00:54:28,386 - mini_batch_size: "8"
60
+ 2023-10-15 00:54:28,386 - max_epochs: "10"
61
+ 2023-10-15 00:54:28,386 - shuffle: "True"
62
+ 2023-10-15 00:54:28,386 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-15 00:54:28,386 Plugins:
64
+ 2023-10-15 00:54:28,386 - LinearScheduler | warmup_fraction: '0.1'
65
+ 2023-10-15 00:54:28,386 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-15 00:54:28,386 Final evaluation on model from best epoch (best-model.pt)
67
+ 2023-10-15 00:54:28,386 - metric: "('micro avg', 'f1-score')"
68
+ 2023-10-15 00:54:28,386 ----------------------------------------------------------------------------------------------------
69
+ 2023-10-15 00:54:28,386 Computation:
70
+ 2023-10-15 00:54:28,386 - compute on device: cuda:0
71
+ 2023-10-15 00:54:28,386 - embedding storage: none
72
+ 2023-10-15 00:54:28,386 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-15 00:54:28,386 Model training base path: "hmbench-letemps/fr-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4"
74
+ 2023-10-15 00:54:28,386 ----------------------------------------------------------------------------------------------------
75
+ 2023-10-15 00:54:28,386 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-15 00:54:40,113 epoch 1 - iter 180/1809 - loss 1.36932055 - time (sec): 11.73 - samples/sec: 3221.87 - lr: 0.000005 - momentum: 0.000000
77
+ 2023-10-15 00:54:50,988 epoch 1 - iter 360/1809 - loss 0.78656800 - time (sec): 22.60 - samples/sec: 3318.33 - lr: 0.000010 - momentum: 0.000000
78
+ 2023-10-15 00:55:02,032 epoch 1 - iter 540/1809 - loss 0.57710329 - time (sec): 33.64 - samples/sec: 3360.36 - lr: 0.000015 - momentum: 0.000000
79
+ 2023-10-15 00:55:12,671 epoch 1 - iter 720/1809 - loss 0.46762299 - time (sec): 44.28 - samples/sec: 3393.68 - lr: 0.000020 - momentum: 0.000000
80
+ 2023-10-15 00:55:23,778 epoch 1 - iter 900/1809 - loss 0.39831031 - time (sec): 55.39 - samples/sec: 3406.54 - lr: 0.000025 - momentum: 0.000000
81
+ 2023-10-15 00:55:34,661 epoch 1 - iter 1080/1809 - loss 0.35140217 - time (sec): 66.27 - samples/sec: 3414.05 - lr: 0.000030 - momentum: 0.000000
82
+ 2023-10-15 00:55:45,480 epoch 1 - iter 1260/1809 - loss 0.31666739 - time (sec): 77.09 - samples/sec: 3414.18 - lr: 0.000035 - momentum: 0.000000
83
+ 2023-10-15 00:55:56,429 epoch 1 - iter 1440/1809 - loss 0.28905303 - time (sec): 88.04 - samples/sec: 3419.36 - lr: 0.000040 - momentum: 0.000000
84
+ 2023-10-15 00:56:07,587 epoch 1 - iter 1620/1809 - loss 0.26797933 - time (sec): 99.20 - samples/sec: 3426.93 - lr: 0.000045 - momentum: 0.000000
85
+ 2023-10-15 00:56:19,166 epoch 1 - iter 1800/1809 - loss 0.25117700 - time (sec): 110.78 - samples/sec: 3415.31 - lr: 0.000050 - momentum: 0.000000
86
+ 2023-10-15 00:56:19,662 ----------------------------------------------------------------------------------------------------
87
+ 2023-10-15 00:56:19,662 EPOCH 1 done: loss 0.2506 - lr: 0.000050
88
+ 2023-10-15 00:56:24,421 DEV : loss 0.12091385573148727 - f1-score (micro avg) 0.5948
89
+ 2023-10-15 00:56:24,450 saving best model
90
+ 2023-10-15 00:56:24,830 ----------------------------------------------------------------------------------------------------
91
+ 2023-10-15 00:56:35,904 epoch 2 - iter 180/1809 - loss 0.09488290 - time (sec): 11.07 - samples/sec: 3502.43 - lr: 0.000049 - momentum: 0.000000
92
+ 2023-10-15 00:56:47,030 epoch 2 - iter 360/1809 - loss 0.08892161 - time (sec): 22.20 - samples/sec: 3436.57 - lr: 0.000049 - momentum: 0.000000
93
+ 2023-10-15 00:56:57,888 epoch 2 - iter 540/1809 - loss 0.08874244 - time (sec): 33.06 - samples/sec: 3441.96 - lr: 0.000048 - momentum: 0.000000
94
+ 2023-10-15 00:57:08,872 epoch 2 - iter 720/1809 - loss 0.08742615 - time (sec): 44.04 - samples/sec: 3468.08 - lr: 0.000048 - momentum: 0.000000
95
+ 2023-10-15 00:57:19,849 epoch 2 - iter 900/1809 - loss 0.08807139 - time (sec): 55.02 - samples/sec: 3462.99 - lr: 0.000047 - momentum: 0.000000
96
+ 2023-10-15 00:57:31,096 epoch 2 - iter 1080/1809 - loss 0.08743360 - time (sec): 66.26 - samples/sec: 3469.58 - lr: 0.000047 - momentum: 0.000000
97
+ 2023-10-15 00:57:42,000 epoch 2 - iter 1260/1809 - loss 0.08731323 - time (sec): 77.17 - samples/sec: 3458.92 - lr: 0.000046 - momentum: 0.000000
98
+ 2023-10-15 00:57:52,947 epoch 2 - iter 1440/1809 - loss 0.08517886 - time (sec): 88.12 - samples/sec: 3442.76 - lr: 0.000046 - momentum: 0.000000
99
+ 2023-10-15 00:58:05,025 epoch 2 - iter 1620/1809 - loss 0.08456187 - time (sec): 100.19 - samples/sec: 3407.26 - lr: 0.000045 - momentum: 0.000000
100
+ 2023-10-15 00:58:16,180 epoch 2 - iter 1800/1809 - loss 0.08446233 - time (sec): 111.35 - samples/sec: 3397.65 - lr: 0.000044 - momentum: 0.000000
101
+ 2023-10-15 00:58:16,714 ----------------------------------------------------------------------------------------------------
102
+ 2023-10-15 00:58:16,715 EPOCH 2 done: loss 0.0843 - lr: 0.000044
103
+ 2023-10-15 00:58:23,259 DEV : loss 0.12702764570713043 - f1-score (micro avg) 0.6359
104
+ 2023-10-15 00:58:23,291 saving best model
105
+ 2023-10-15 00:58:23,759 ----------------------------------------------------------------------------------------------------
106
+ 2023-10-15 00:58:34,801 epoch 3 - iter 180/1809 - loss 0.05664383 - time (sec): 11.04 - samples/sec: 3424.57 - lr: 0.000044 - momentum: 0.000000
107
+ 2023-10-15 00:58:46,143 epoch 3 - iter 360/1809 - loss 0.05872794 - time (sec): 22.38 - samples/sec: 3432.01 - lr: 0.000043 - momentum: 0.000000
108
+ 2023-10-15 00:58:57,084 epoch 3 - iter 540/1809 - loss 0.05857268 - time (sec): 33.32 - samples/sec: 3403.64 - lr: 0.000043 - momentum: 0.000000
109
+ 2023-10-15 00:59:08,293 epoch 3 - iter 720/1809 - loss 0.06057457 - time (sec): 44.53 - samples/sec: 3387.71 - lr: 0.000042 - momentum: 0.000000
110
+ 2023-10-15 00:59:19,331 epoch 3 - iter 900/1809 - loss 0.06151060 - time (sec): 55.57 - samples/sec: 3367.52 - lr: 0.000042 - momentum: 0.000000
111
+ 2023-10-15 00:59:30,825 epoch 3 - iter 1080/1809 - loss 0.06087473 - time (sec): 67.06 - samples/sec: 3361.04 - lr: 0.000041 - momentum: 0.000000
112
+ 2023-10-15 00:59:42,483 epoch 3 - iter 1260/1809 - loss 0.06048810 - time (sec): 78.72 - samples/sec: 3344.19 - lr: 0.000041 - momentum: 0.000000
113
+ 2023-10-15 00:59:53,394 epoch 3 - iter 1440/1809 - loss 0.06164535 - time (sec): 89.63 - samples/sec: 3362.29 - lr: 0.000040 - momentum: 0.000000
114
+ 2023-10-15 01:00:04,335 epoch 3 - iter 1620/1809 - loss 0.06135065 - time (sec): 100.57 - samples/sec: 3367.20 - lr: 0.000039 - momentum: 0.000000
115
+ 2023-10-15 01:00:16,005 epoch 3 - iter 1800/1809 - loss 0.06014152 - time (sec): 112.24 - samples/sec: 3372.10 - lr: 0.000039 - momentum: 0.000000
116
+ 2023-10-15 01:00:16,523 ----------------------------------------------------------------------------------------------------
117
+ 2023-10-15 01:00:16,523 EPOCH 3 done: loss 0.0604 - lr: 0.000039
118
+ 2023-10-15 01:00:23,799 DEV : loss 0.16295944154262543 - f1-score (micro avg) 0.6297
119
+ 2023-10-15 01:00:23,837 ----------------------------------------------------------------------------------------------------
120
+ 2023-10-15 01:00:35,272 epoch 4 - iter 180/1809 - loss 0.03961425 - time (sec): 11.43 - samples/sec: 3311.94 - lr: 0.000038 - momentum: 0.000000
121
+ 2023-10-15 01:00:46,336 epoch 4 - iter 360/1809 - loss 0.04234972 - time (sec): 22.50 - samples/sec: 3319.62 - lr: 0.000038 - momentum: 0.000000
122
+ 2023-10-15 01:00:57,554 epoch 4 - iter 540/1809 - loss 0.04351680 - time (sec): 33.72 - samples/sec: 3401.86 - lr: 0.000037 - momentum: 0.000000
123
+ 2023-10-15 01:01:08,577 epoch 4 - iter 720/1809 - loss 0.04425001 - time (sec): 44.74 - samples/sec: 3398.31 - lr: 0.000037 - momentum: 0.000000
124
+ 2023-10-15 01:01:19,570 epoch 4 - iter 900/1809 - loss 0.04646110 - time (sec): 55.73 - samples/sec: 3402.61 - lr: 0.000036 - momentum: 0.000000
125
+ 2023-10-15 01:01:30,811 epoch 4 - iter 1080/1809 - loss 0.04600080 - time (sec): 66.97 - samples/sec: 3392.58 - lr: 0.000036 - momentum: 0.000000
126
+ 2023-10-15 01:01:42,161 epoch 4 - iter 1260/1809 - loss 0.04633692 - time (sec): 78.32 - samples/sec: 3386.88 - lr: 0.000035 - momentum: 0.000000
127
+ 2023-10-15 01:01:53,180 epoch 4 - iter 1440/1809 - loss 0.04642934 - time (sec): 89.34 - samples/sec: 3387.53 - lr: 0.000034 - momentum: 0.000000
128
+ 2023-10-15 01:02:04,071 epoch 4 - iter 1620/1809 - loss 0.04694786 - time (sec): 100.23 - samples/sec: 3394.43 - lr: 0.000034 - momentum: 0.000000
129
+ 2023-10-15 01:02:15,553 epoch 4 - iter 1800/1809 - loss 0.04701010 - time (sec): 111.71 - samples/sec: 3385.85 - lr: 0.000033 - momentum: 0.000000
130
+ 2023-10-15 01:02:16,080 ----------------------------------------------------------------------------------------------------
131
+ 2023-10-15 01:02:16,080 EPOCH 4 done: loss 0.0471 - lr: 0.000033
132
+ 2023-10-15 01:02:21,725 DEV : loss 0.24718838930130005 - f1-score (micro avg) 0.6373
133
+ 2023-10-15 01:02:21,763 saving best model
134
+ 2023-10-15 01:02:22,289 ----------------------------------------------------------------------------------------------------
135
+ 2023-10-15 01:02:33,602 epoch 5 - iter 180/1809 - loss 0.04139222 - time (sec): 11.31 - samples/sec: 3281.09 - lr: 0.000033 - momentum: 0.000000
136
+ 2023-10-15 01:02:44,892 epoch 5 - iter 360/1809 - loss 0.03425512 - time (sec): 22.60 - samples/sec: 3361.65 - lr: 0.000032 - momentum: 0.000000
137
+ 2023-10-15 01:02:56,018 epoch 5 - iter 540/1809 - loss 0.03244197 - time (sec): 33.72 - samples/sec: 3392.18 - lr: 0.000032 - momentum: 0.000000
138
+ 2023-10-15 01:03:06,923 epoch 5 - iter 720/1809 - loss 0.03138031 - time (sec): 44.63 - samples/sec: 3400.16 - lr: 0.000031 - momentum: 0.000000
139
+ 2023-10-15 01:03:19,372 epoch 5 - iter 900/1809 - loss 0.03282818 - time (sec): 57.08 - samples/sec: 3327.57 - lr: 0.000031 - momentum: 0.000000
140
+ 2023-10-15 01:03:30,561 epoch 5 - iter 1080/1809 - loss 0.03451375 - time (sec): 68.26 - samples/sec: 3346.02 - lr: 0.000030 - momentum: 0.000000
141
+ 2023-10-15 01:03:41,400 epoch 5 - iter 1260/1809 - loss 0.03406145 - time (sec): 79.10 - samples/sec: 3384.90 - lr: 0.000029 - momentum: 0.000000
142
+ 2023-10-15 01:03:52,273 epoch 5 - iter 1440/1809 - loss 0.03339172 - time (sec): 89.98 - samples/sec: 3392.53 - lr: 0.000029 - momentum: 0.000000
143
+ 2023-10-15 01:04:02,972 epoch 5 - iter 1620/1809 - loss 0.03362623 - time (sec): 100.68 - samples/sec: 3398.84 - lr: 0.000028 - momentum: 0.000000
144
+ 2023-10-15 01:04:13,377 epoch 5 - iter 1800/1809 - loss 0.03349583 - time (sec): 111.08 - samples/sec: 3402.69 - lr: 0.000028 - momentum: 0.000000
145
+ 2023-10-15 01:04:13,958 ----------------------------------------------------------------------------------------------------
146
+ 2023-10-15 01:04:13,958 EPOCH 5 done: loss 0.0334 - lr: 0.000028
147
+ 2023-10-15 01:04:19,608 DEV : loss 0.28187352418899536 - f1-score (micro avg) 0.6317
148
+ 2023-10-15 01:04:19,639 ----------------------------------------------------------------------------------------------------
149
+ 2023-10-15 01:04:30,695 epoch 6 - iter 180/1809 - loss 0.02173032 - time (sec): 11.05 - samples/sec: 3382.71 - lr: 0.000027 - momentum: 0.000000
150
+ 2023-10-15 01:04:41,645 epoch 6 - iter 360/1809 - loss 0.02768280 - time (sec): 22.00 - samples/sec: 3446.91 - lr: 0.000027 - momentum: 0.000000
151
+ 2023-10-15 01:04:52,882 epoch 6 - iter 540/1809 - loss 0.02661026 - time (sec): 33.24 - samples/sec: 3421.56 - lr: 0.000026 - momentum: 0.000000
152
+ 2023-10-15 01:05:04,092 epoch 6 - iter 720/1809 - loss 0.02450565 - time (sec): 44.45 - samples/sec: 3428.34 - lr: 0.000026 - momentum: 0.000000
153
+ 2023-10-15 01:05:14,986 epoch 6 - iter 900/1809 - loss 0.02344113 - time (sec): 55.35 - samples/sec: 3425.30 - lr: 0.000025 - momentum: 0.000000
154
+ 2023-10-15 01:05:25,899 epoch 6 - iter 1080/1809 - loss 0.02273246 - time (sec): 66.26 - samples/sec: 3429.11 - lr: 0.000024 - momentum: 0.000000
155
+ 2023-10-15 01:05:36,612 epoch 6 - iter 1260/1809 - loss 0.02264429 - time (sec): 76.97 - samples/sec: 3434.75 - lr: 0.000024 - momentum: 0.000000
156
+ 2023-10-15 01:05:47,521 epoch 6 - iter 1440/1809 - loss 0.02292782 - time (sec): 87.88 - samples/sec: 3437.93 - lr: 0.000023 - momentum: 0.000000
157
+ 2023-10-15 01:05:58,443 epoch 6 - iter 1620/1809 - loss 0.02269623 - time (sec): 98.80 - samples/sec: 3446.68 - lr: 0.000023 - momentum: 0.000000
158
+ 2023-10-15 01:06:09,409 epoch 6 - iter 1800/1809 - loss 0.02324916 - time (sec): 109.77 - samples/sec: 3444.92 - lr: 0.000022 - momentum: 0.000000
159
+ 2023-10-15 01:06:09,911 ----------------------------------------------------------------------------------------------------
160
+ 2023-10-15 01:06:09,911 EPOCH 6 done: loss 0.0233 - lr: 0.000022
161
+ 2023-10-15 01:06:17,183 DEV : loss 0.320418119430542 - f1-score (micro avg) 0.6302
162
+ 2023-10-15 01:06:17,221 ----------------------------------------------------------------------------------------------------
163
+ 2023-10-15 01:06:28,192 epoch 7 - iter 180/1809 - loss 0.02087467 - time (sec): 10.97 - samples/sec: 3370.81 - lr: 0.000022 - momentum: 0.000000
164
+ 2023-10-15 01:06:39,389 epoch 7 - iter 360/1809 - loss 0.01917004 - time (sec): 22.17 - samples/sec: 3315.62 - lr: 0.000021 - momentum: 0.000000
165
+ 2023-10-15 01:06:50,537 epoch 7 - iter 540/1809 - loss 0.01840608 - time (sec): 33.31 - samples/sec: 3351.73 - lr: 0.000021 - momentum: 0.000000
166
+ 2023-10-15 01:07:01,437 epoch 7 - iter 720/1809 - loss 0.01729318 - time (sec): 44.21 - samples/sec: 3365.69 - lr: 0.000020 - momentum: 0.000000
167
+ 2023-10-15 01:07:12,404 epoch 7 - iter 900/1809 - loss 0.01787055 - time (sec): 55.18 - samples/sec: 3387.47 - lr: 0.000019 - momentum: 0.000000
168
+ 2023-10-15 01:07:23,323 epoch 7 - iter 1080/1809 - loss 0.01780583 - time (sec): 66.10 - samples/sec: 3405.28 - lr: 0.000019 - momentum: 0.000000
169
+ 2023-10-15 01:07:34,647 epoch 7 - iter 1260/1809 - loss 0.01795576 - time (sec): 77.42 - samples/sec: 3411.06 - lr: 0.000018 - momentum: 0.000000
170
+ 2023-10-15 01:07:45,565 epoch 7 - iter 1440/1809 - loss 0.01738303 - time (sec): 88.34 - samples/sec: 3410.11 - lr: 0.000018 - momentum: 0.000000
171
+ 2023-10-15 01:07:56,827 epoch 7 - iter 1620/1809 - loss 0.01737434 - time (sec): 99.60 - samples/sec: 3426.76 - lr: 0.000017 - momentum: 0.000000
172
+ 2023-10-15 01:08:07,817 epoch 7 - iter 1800/1809 - loss 0.01726721 - time (sec): 110.59 - samples/sec: 3420.15 - lr: 0.000017 - momentum: 0.000000
173
+ 2023-10-15 01:08:08,321 ----------------------------------------------------------------------------------------------------
174
+ 2023-10-15 01:08:08,321 EPOCH 7 done: loss 0.0172 - lr: 0.000017
175
+ 2023-10-15 01:08:14,981 DEV : loss 0.32448261976242065 - f1-score (micro avg) 0.6396
176
+ 2023-10-15 01:08:15,030 saving best model
177
+ 2023-10-15 01:08:15,537 ----------------------------------------------------------------------------------------------------
178
+ 2023-10-15 01:08:26,680 epoch 8 - iter 180/1809 - loss 0.00879191 - time (sec): 11.14 - samples/sec: 3328.67 - lr: 0.000016 - momentum: 0.000000
179
+ 2023-10-15 01:08:37,699 epoch 8 - iter 360/1809 - loss 0.00863859 - time (sec): 22.16 - samples/sec: 3403.22 - lr: 0.000016 - momentum: 0.000000
180
+ 2023-10-15 01:08:48,540 epoch 8 - iter 540/1809 - loss 0.00936001 - time (sec): 33.00 - samples/sec: 3408.03 - lr: 0.000015 - momentum: 0.000000
181
+ 2023-10-15 01:08:59,444 epoch 8 - iter 720/1809 - loss 0.01103862 - time (sec): 43.90 - samples/sec: 3454.19 - lr: 0.000014 - momentum: 0.000000
182
+ 2023-10-15 01:09:10,253 epoch 8 - iter 900/1809 - loss 0.01061716 - time (sec): 54.71 - samples/sec: 3445.10 - lr: 0.000014 - momentum: 0.000000
183
+ 2023-10-15 01:09:21,080 epoch 8 - iter 1080/1809 - loss 0.01176131 - time (sec): 65.54 - samples/sec: 3452.23 - lr: 0.000013 - momentum: 0.000000
184
+ 2023-10-15 01:09:32,223 epoch 8 - iter 1260/1809 - loss 0.01247786 - time (sec): 76.68 - samples/sec: 3435.24 - lr: 0.000013 - momentum: 0.000000
185
+ 2023-10-15 01:09:43,301 epoch 8 - iter 1440/1809 - loss 0.01260504 - time (sec): 87.76 - samples/sec: 3445.49 - lr: 0.000012 - momentum: 0.000000
186
+ 2023-10-15 01:09:54,201 epoch 8 - iter 1620/1809 - loss 0.01224412 - time (sec): 98.66 - samples/sec: 3445.31 - lr: 0.000012 - momentum: 0.000000
187
+ 2023-10-15 01:10:05,425 epoch 8 - iter 1800/1809 - loss 0.01259242 - time (sec): 109.88 - samples/sec: 3438.46 - lr: 0.000011 - momentum: 0.000000
188
+ 2023-10-15 01:10:06,017 ----------------------------------------------------------------------------------------------------
189
+ 2023-10-15 01:10:06,017 EPOCH 8 done: loss 0.0126 - lr: 0.000011
190
+ 2023-10-15 01:10:13,253 DEV : loss 0.3656839430332184 - f1-score (micro avg) 0.6364
191
+ 2023-10-15 01:10:13,303 ----------------------------------------------------------------------------------------------------
192
+ 2023-10-15 01:10:24,414 epoch 9 - iter 180/1809 - loss 0.00881178 - time (sec): 11.11 - samples/sec: 3404.28 - lr: 0.000011 - momentum: 0.000000
193
+ 2023-10-15 01:10:35,674 epoch 9 - iter 360/1809 - loss 0.00668887 - time (sec): 22.37 - samples/sec: 3402.69 - lr: 0.000010 - momentum: 0.000000
194
+ 2023-10-15 01:10:47,053 epoch 9 - iter 540/1809 - loss 0.00681184 - time (sec): 33.75 - samples/sec: 3376.53 - lr: 0.000009 - momentum: 0.000000
195
+ 2023-10-15 01:10:57,765 epoch 9 - iter 720/1809 - loss 0.00818791 - time (sec): 44.46 - samples/sec: 3395.24 - lr: 0.000009 - momentum: 0.000000
196
+ 2023-10-15 01:11:09,154 epoch 9 - iter 900/1809 - loss 0.00797673 - time (sec): 55.85 - samples/sec: 3392.20 - lr: 0.000008 - momentum: 0.000000
197
+ 2023-10-15 01:11:20,175 epoch 9 - iter 1080/1809 - loss 0.00746067 - time (sec): 66.87 - samples/sec: 3398.87 - lr: 0.000008 - momentum: 0.000000
198
+ 2023-10-15 01:11:31,090 epoch 9 - iter 1260/1809 - loss 0.00715338 - time (sec): 77.79 - samples/sec: 3400.87 - lr: 0.000007 - momentum: 0.000000
199
+ 2023-10-15 01:11:42,392 epoch 9 - iter 1440/1809 - loss 0.00709472 - time (sec): 89.09 - samples/sec: 3402.95 - lr: 0.000007 - momentum: 0.000000
200
+ 2023-10-15 01:11:53,443 epoch 9 - iter 1620/1809 - loss 0.00765987 - time (sec): 100.14 - samples/sec: 3403.53 - lr: 0.000006 - momentum: 0.000000
201
+ 2023-10-15 01:12:04,239 epoch 9 - iter 1800/1809 - loss 0.00773264 - time (sec): 110.93 - samples/sec: 3408.14 - lr: 0.000006 - momentum: 0.000000
202
+ 2023-10-15 01:12:04,763 ----------------------------------------------------------------------------------------------------
203
+ 2023-10-15 01:12:04,763 EPOCH 9 done: loss 0.0077 - lr: 0.000006
204
+ 2023-10-15 01:12:11,370 DEV : loss 0.38767531514167786 - f1-score (micro avg) 0.6389
205
+ 2023-10-15 01:12:11,406 ----------------------------------------------------------------------------------------------------
206
+ 2023-10-15 01:12:22,343 epoch 10 - iter 180/1809 - loss 0.00415207 - time (sec): 10.94 - samples/sec: 3527.98 - lr: 0.000005 - momentum: 0.000000
207
+ 2023-10-15 01:12:33,418 epoch 10 - iter 360/1809 - loss 0.00612190 - time (sec): 22.01 - samples/sec: 3473.52 - lr: 0.000004 - momentum: 0.000000
208
+ 2023-10-15 01:12:44,320 epoch 10 - iter 540/1809 - loss 0.00583415 - time (sec): 32.91 - samples/sec: 3445.48 - lr: 0.000004 - momentum: 0.000000
209
+ 2023-10-15 01:12:55,272 epoch 10 - iter 720/1809 - loss 0.00534481 - time (sec): 43.87 - samples/sec: 3471.74 - lr: 0.000003 - momentum: 0.000000
210
+ 2023-10-15 01:13:06,141 epoch 10 - iter 900/1809 - loss 0.00597649 - time (sec): 54.73 - samples/sec: 3464.86 - lr: 0.000003 - momentum: 0.000000
211
+ 2023-10-15 01:13:16,696 epoch 10 - iter 1080/1809 - loss 0.00547680 - time (sec): 65.29 - samples/sec: 3466.99 - lr: 0.000002 - momentum: 0.000000
212
+ 2023-10-15 01:13:27,949 epoch 10 - iter 1260/1809 - loss 0.00530282 - time (sec): 76.54 - samples/sec: 3470.09 - lr: 0.000002 - momentum: 0.000000
213
+ 2023-10-15 01:13:38,673 epoch 10 - iter 1440/1809 - loss 0.00538807 - time (sec): 87.27 - samples/sec: 3474.27 - lr: 0.000001 - momentum: 0.000000
214
+ 2023-10-15 01:13:49,520 epoch 10 - iter 1620/1809 - loss 0.00507514 - time (sec): 98.11 - samples/sec: 3473.47 - lr: 0.000001 - momentum: 0.000000
215
+ 2023-10-15 01:14:00,390 epoch 10 - iter 1800/1809 - loss 0.00530648 - time (sec): 108.98 - samples/sec: 3465.34 - lr: 0.000000 - momentum: 0.000000
216
+ 2023-10-15 01:14:01,003 ----------------------------------------------------------------------------------------------------
217
+ 2023-10-15 01:14:01,003 EPOCH 10 done: loss 0.0053 - lr: 0.000000
218
+ 2023-10-15 01:14:06,689 DEV : loss 0.39427119493484497 - f1-score (micro avg) 0.6467
219
+ 2023-10-15 01:14:06,725 saving best model
220
+ 2023-10-15 01:14:07,568 ----------------------------------------------------------------------------------------------------
221
+ 2023-10-15 01:14:07,569 Loading model from best epoch ...
222
+ 2023-10-15 01:14:08,941 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
223
+ 2023-10-15 01:14:16,629
224
+ Results:
225
+ - F-score (micro) 0.6507
226
+ - F-score (macro) 0.5152
227
+ - Accuracy 0.4973
228
+
229
+ By class:
230
+ precision recall f1-score support
231
+
232
+ loc 0.6339 0.7648 0.6933 591
233
+ pers 0.5823 0.7535 0.6569 357
234
+ org 0.2407 0.1646 0.1955 79
235
+
236
+ micro avg 0.5972 0.7147 0.6507 1027
237
+ macro avg 0.4856 0.5610 0.5152 1027
238
+ weighted avg 0.5857 0.7147 0.6423 1027
239
+
240
+ 2023-10-15 01:14:16,629 ----------------------------------------------------------------------------------------------------