Hugging Face's logo

stefan-it
/
hmbench-icdar-fr-hmteams-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-3

Token Classification
Flair
PyTorch
TensorBoard
French
sequence-tagger-model
Model card Files Metrics Community
stefan-it commited on
Commit
86f1d6b
•
1 Parent(s): aa734d0

Upload folder using huggingface_hub

Browse files
Files changed (6) hide show
  1. best-model.pt +3 -0
  2. dev.tsv +0 -0
  3. loss.tsv +11 -0
  4. runs/events.out.tfevents.1697545813.bce904bcef33.2023.8 +3 -0
  5. test.tsv +0 -0
  6. training.log +238 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d60f84e5c08e83b0085a2ed97517ee7e2af8337aa2511de575edbfea170a828c
3
+ size 440941957
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 12:31:52 0.0000 0.4277 0.0889 0.7024 0.7262 0.7141 0.5676
3
+ 2 12:33:30 0.0000 0.1185 0.1045 0.6800 0.7670 0.7209 0.5810
4
+ 3 12:35:08 0.0000 0.0882 0.1184 0.7198 0.7557 0.7373 0.6034
5
+ 4 12:36:46 0.0000 0.0650 0.1447 0.7240 0.7568 0.7400 0.6065
6
+ 5 12:38:24 0.0000 0.0490 0.1686 0.7516 0.7941 0.7723 0.6434
7
+ 6 12:40:01 0.0000 0.0370 0.1863 0.7519 0.7783 0.7649 0.6353
8
+ 7 12:41:37 0.0000 0.0270 0.2106 0.7444 0.7873 0.7653 0.6374
9
+ 8 12:43:13 0.0000 0.0181 0.2295 0.7401 0.7862 0.7625 0.6324
10
+ 9 12:44:48 0.0000 0.0120 0.2399 0.7428 0.7907 0.7660 0.6395
11
+ 10 12:46:25 0.0000 0.0089 0.2463 0.7538 0.7862 0.7697 0.6435
runs/events.out.tfevents.1697545813.bce904bcef33.2023.8 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bf88c7ce1985ab9970dce4b9a326db640a40fb275994c132a0c267addbb5bd81
3
+ size 1108164
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,238 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-17 12:30:13,946 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-17 12:30:13,947 Model: "SequenceTagger(
3
+ (embeddings): TransformerWordEmbeddings(
4
+ (model): ElectraModel(
5
+ (embeddings): ElectraEmbeddings(
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): ElectraEncoder(
13
+ (layer): ModuleList(
14
+ (0-11): 12 x ElectraLayer(
15
+ (attention): ElectraAttention(
16
+ (self): ElectraSelfAttention(
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): ElectraSelfOutput(
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): ElectraIntermediate(
29
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
30
+ (intermediate_act_fn): GELUActivation()
31
+ )
32
+ (output): ElectraOutput(
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
+ )
41
+ )
42
+ (locked_dropout): LockedDropout(p=0.5)
43
+ (linear): Linear(in_features=768, out_features=13, bias=True)
44
+ (loss_function): CrossEntropyLoss()
45
+ )"
46
+ 2023-10-17 12:30:13,947 ----------------------------------------------------------------------------------------------------
47
+ 2023-10-17 12:30:13,947 MultiCorpus: 7936 train + 992 dev + 992 test sentences
48
+ - NER_ICDAR_EUROPEANA Corpus: 7936 train + 992 dev + 992 test sentences - /root/.flair/datasets/ner_icdar_europeana/fr
49
+ 2023-10-17 12:30:13,947 ----------------------------------------------------------------------------------------------------
50
+ 2023-10-17 12:30:13,947 Train: 7936 sentences
51
+ 2023-10-17 12:30:13,947 (train_with_dev=False, train_with_test=False)
52
+ 2023-10-17 12:30:13,947 ----------------------------------------------------------------------------------------------------
53
+ 2023-10-17 12:30:13,948 Training Params:
54
+ 2023-10-17 12:30:13,948 - learning_rate: "3e-05"
55
+ 2023-10-17 12:30:13,948 - mini_batch_size: "4"
56
+ 2023-10-17 12:30:13,948 - max_epochs: "10"
57
+ 2023-10-17 12:30:13,948 - shuffle: "True"
58
+ 2023-10-17 12:30:13,948 ----------------------------------------------------------------------------------------------------
59
+ 2023-10-17 12:30:13,948 Plugins:
60
+ 2023-10-17 12:30:13,948 - TensorboardLogger
61
+ 2023-10-17 12:30:13,948 - LinearScheduler | warmup_fraction: '0.1'
62
+ 2023-10-17 12:30:13,948 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-17 12:30:13,948 Final evaluation on model from best epoch (best-model.pt)
64
+ 2023-10-17 12:30:13,948 - metric: "('micro avg', 'f1-score')"
65
+ 2023-10-17 12:30:13,948 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-17 12:30:13,948 Computation:
67
+ 2023-10-17 12:30:13,948 - compute on device: cuda:0
68
+ 2023-10-17 12:30:13,948 - embedding storage: none
69
+ 2023-10-17 12:30:13,948 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-17 12:30:13,948 Model training base path: "hmbench-icdar/fr-hmteams/teams-base-historic-multilingual-discriminator-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-3"
71
+ 2023-10-17 12:30:13,948 ----------------------------------------------------------------------------------------------------
72
+ 2023-10-17 12:30:13,948 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-17 12:30:13,948 Logging anything other than scalars to TensorBoard is currently not supported.
74
+ 2023-10-17 12:30:23,424 epoch 1 - iter 198/1984 - loss 2.38234386 - time (sec): 9.47 - samples/sec: 1802.33 - lr: 0.000003 - momentum: 0.000000
75
+ 2023-10-17 12:30:32,782 epoch 1 - iter 396/1984 - loss 1.42791489 - time (sec): 18.83 - samples/sec: 1754.21 - lr: 0.000006 - momentum: 0.000000
76
+ 2023-10-17 12:30:42,378 epoch 1 - iter 594/1984 - loss 1.04332496 - time (sec): 28.43 - samples/sec: 1760.25 - lr: 0.000009 - momentum: 0.000000
77
+ 2023-10-17 12:30:51,670 epoch 1 - iter 792/1984 - loss 0.84433293 - time (sec): 37.72 - samples/sec: 1746.56 - lr: 0.000012 - momentum: 0.000000
78
+ 2023-10-17 12:31:01,111 epoch 1 - iter 990/1984 - loss 0.70291310 - time (sec): 47.16 - samples/sec: 1760.29 - lr: 0.000015 - momentum: 0.000000
79
+ 2023-10-17 12:31:10,437 epoch 1 - iter 1188/1984 - loss 0.60723553 - time (sec): 56.49 - samples/sec: 1779.05 - lr: 0.000018 - momentum: 0.000000
80
+ 2023-10-17 12:31:19,931 epoch 1 - iter 1386/1984 - loss 0.53958743 - time (sec): 65.98 - samples/sec: 1783.04 - lr: 0.000021 - momentum: 0.000000
81
+ 2023-10-17 12:31:29,162 epoch 1 - iter 1584/1984 - loss 0.49563480 - time (sec): 75.21 - samples/sec: 1767.27 - lr: 0.000024 - momentum: 0.000000
82
+ 2023-10-17 12:31:38,675 epoch 1 - iter 1782/1984 - loss 0.46032639 - time (sec): 84.73 - samples/sec: 1747.00 - lr: 0.000027 - momentum: 0.000000
83
+ 2023-10-17 12:31:48,550 epoch 1 - iter 1980/1984 - loss 0.42801054 - time (sec): 94.60 - samples/sec: 1730.52 - lr: 0.000030 - momentum: 0.000000
84
+ 2023-10-17 12:31:48,732 ----------------------------------------------------------------------------------------------------
85
+ 2023-10-17 12:31:48,733 EPOCH 1 done: loss 0.4277 - lr: 0.000030
86
+ 2023-10-17 12:31:52,355 DEV : loss 0.08888775110244751 - f1-score (micro avg) 0.7141
87
+ 2023-10-17 12:31:52,388 saving best model
88
+ 2023-10-17 12:31:52,812 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-17 12:32:02,080 epoch 2 - iter 198/1984 - loss 0.12840545 - time (sec): 9.27 - samples/sec: 1627.82 - lr: 0.000030 - momentum: 0.000000
90
+ 2023-10-17 12:32:11,544 epoch 2 - iter 396/1984 - loss 0.12995343 - time (sec): 18.73 - samples/sec: 1685.97 - lr: 0.000029 - momentum: 0.000000
91
+ 2023-10-17 12:32:21,075 epoch 2 - iter 594/1984 - loss 0.13177287 - time (sec): 28.26 - samples/sec: 1661.53 - lr: 0.000029 - momentum: 0.000000
92
+ 2023-10-17 12:32:30,451 epoch 2 - iter 792/1984 - loss 0.12718319 - time (sec): 37.64 - samples/sec: 1686.76 - lr: 0.000029 - momentum: 0.000000
93
+ 2023-10-17 12:32:39,752 epoch 2 - iter 990/1984 - loss 0.12715837 - time (sec): 46.94 - samples/sec: 1710.38 - lr: 0.000028 - momentum: 0.000000
94
+ 2023-10-17 12:32:48,979 epoch 2 - iter 1188/1984 - loss 0.12525218 - time (sec): 56.17 - samples/sec: 1721.81 - lr: 0.000028 - momentum: 0.000000
95
+ 2023-10-17 12:32:58,260 epoch 2 - iter 1386/1984 - loss 0.12111745 - time (sec): 65.45 - samples/sec: 1721.22 - lr: 0.000028 - momentum: 0.000000
96
+ 2023-10-17 12:33:07,791 epoch 2 - iter 1584/1984 - loss 0.12050247 - time (sec): 74.98 - samples/sec: 1725.69 - lr: 0.000027 - momentum: 0.000000
97
+ 2023-10-17 12:33:17,055 epoch 2 - iter 1782/1984 - loss 0.11986072 - time (sec): 84.24 - samples/sec: 1742.62 - lr: 0.000027 - momentum: 0.000000
98
+ 2023-10-17 12:33:26,450 epoch 2 - iter 1980/1984 - loss 0.11870868 - time (sec): 93.64 - samples/sec: 1748.11 - lr: 0.000027 - momentum: 0.000000
99
+ 2023-10-17 12:33:26,633 ----------------------------------------------------------------------------------------------------
100
+ 2023-10-17 12:33:26,633 EPOCH 2 done: loss 0.1185 - lr: 0.000027
101
+ 2023-10-17 12:33:30,335 DEV : loss 0.10451896488666534 - f1-score (micro avg) 0.7209
102
+ 2023-10-17 12:33:30,357 saving best model
103
+ 2023-10-17 12:33:31,401 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-17 12:33:40,719 epoch 3 - iter 198/1984 - loss 0.08984418 - time (sec): 9.31 - samples/sec: 1711.83 - lr: 0.000026 - momentum: 0.000000
105
+ 2023-10-17 12:33:50,267 epoch 3 - iter 396/1984 - loss 0.08580776 - time (sec): 18.86 - samples/sec: 1733.27 - lr: 0.000026 - momentum: 0.000000
106
+ 2023-10-17 12:33:59,560 epoch 3 - iter 594/1984 - loss 0.08686906 - time (sec): 28.15 - samples/sec: 1740.78 - lr: 0.000026 - momentum: 0.000000
107
+ 2023-10-17 12:34:08,951 epoch 3 - iter 792/1984 - loss 0.08819915 - time (sec): 37.54 - samples/sec: 1738.69 - lr: 0.000025 - momentum: 0.000000
108
+ 2023-10-17 12:34:18,364 epoch 3 - iter 990/1984 - loss 0.08990454 - time (sec): 46.95 - samples/sec: 1721.05 - lr: 0.000025 - momentum: 0.000000
109
+ 2023-10-17 12:34:27,571 epoch 3 - iter 1188/1984 - loss 0.08992539 - time (sec): 56.16 - samples/sec: 1725.40 - lr: 0.000025 - momentum: 0.000000
110
+ 2023-10-17 12:34:36,738 epoch 3 - iter 1386/1984 - loss 0.08943909 - time (sec): 65.33 - samples/sec: 1739.46 - lr: 0.000024 - momentum: 0.000000
111
+ 2023-10-17 12:34:46,042 epoch 3 - iter 1584/1984 - loss 0.08839958 - time (sec): 74.63 - samples/sec: 1763.54 - lr: 0.000024 - momentum: 0.000000
112
+ 2023-10-17 12:34:55,179 epoch 3 - iter 1782/1984 - loss 0.08821887 - time (sec): 83.77 - samples/sec: 1763.87 - lr: 0.000024 - momentum: 0.000000
113
+ 2023-10-17 12:35:04,417 epoch 3 - iter 1980/1984 - loss 0.08832888 - time (sec): 93.01 - samples/sec: 1759.91 - lr: 0.000023 - momentum: 0.000000
114
+ 2023-10-17 12:35:04,605 ----------------------------------------------------------------------------------------------------
115
+ 2023-10-17 12:35:04,606 EPOCH 3 done: loss 0.0882 - lr: 0.000023
116
+ 2023-10-17 12:35:08,254 DEV : loss 0.11844022572040558 - f1-score (micro avg) 0.7373
117
+ 2023-10-17 12:35:08,282 saving best model
118
+ 2023-10-17 12:35:08,809 ----------------------------------------------------------------------------------------------------
119
+ 2023-10-17 12:35:18,197 epoch 4 - iter 198/1984 - loss 0.07187323 - time (sec): 9.39 - samples/sec: 1824.51 - lr: 0.000023 - momentum: 0.000000
120
+ 2023-10-17 12:35:27,464 epoch 4 - iter 396/1984 - loss 0.06422557 - time (sec): 18.65 - samples/sec: 1765.13 - lr: 0.000023 - momentum: 0.000000
121
+ 2023-10-17 12:35:36,725 epoch 4 - iter 594/1984 - loss 0.05702430 - time (sec): 27.91 - samples/sec: 1754.53 - lr: 0.000022 - momentum: 0.000000
122
+ 2023-10-17 12:35:46,077 epoch 4 - iter 792/1984 - loss 0.05974918 - time (sec): 37.27 - samples/sec: 1745.22 - lr: 0.000022 - momentum: 0.000000
123
+ 2023-10-17 12:35:55,423 epoch 4 - iter 990/1984 - loss 0.05976304 - time (sec): 46.61 - samples/sec: 1756.14 - lr: 0.000022 - momentum: 0.000000
124
+ 2023-10-17 12:36:04,760 epoch 4 - iter 1188/1984 - loss 0.06152848 - time (sec): 55.95 - samples/sec: 1743.51 - lr: 0.000021 - momentum: 0.000000
125
+ 2023-10-17 12:36:14,344 epoch 4 - iter 1386/1984 - loss 0.06363359 - time (sec): 65.53 - samples/sec: 1746.73 - lr: 0.000021 - momentum: 0.000000
126
+ 2023-10-17 12:36:23,697 epoch 4 - iter 1584/1984 - loss 0.06505279 - time (sec): 74.89 - samples/sec: 1749.25 - lr: 0.000021 - momentum: 0.000000
127
+ 2023-10-17 12:36:33,075 epoch 4 - iter 1782/1984 - loss 0.06419124 - time (sec): 84.26 - samples/sec: 1749.67 - lr: 0.000020 - momentum: 0.000000
128
+ 2023-10-17 12:36:42,532 epoch 4 - iter 1980/1984 - loss 0.06501747 - time (sec): 93.72 - samples/sec: 1746.19 - lr: 0.000020 - momentum: 0.000000
129
+ 2023-10-17 12:36:42,710 ----------------------------------------------------------------------------------------------------
130
+ 2023-10-17 12:36:42,711 EPOCH 4 done: loss 0.0650 - lr: 0.000020
131
+ 2023-10-17 12:36:46,392 DEV : loss 0.1447092890739441 - f1-score (micro avg) 0.74
132
+ 2023-10-17 12:36:46,416 saving best model
133
+ 2023-10-17 12:36:47,006 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-17 12:36:56,365 epoch 5 - iter 198/1984 - loss 0.04826648 - time (sec): 9.36 - samples/sec: 1762.56 - lr: 0.000020 - momentum: 0.000000
135
+ 2023-10-17 12:37:05,582 epoch 5 - iter 396/1984 - loss 0.04460051 - time (sec): 18.57 - samples/sec: 1778.58 - lr: 0.000019 - momentum: 0.000000
136
+ 2023-10-17 12:37:14,761 epoch 5 - iter 594/1984 - loss 0.04859344 - time (sec): 27.75 - samples/sec: 1754.84 - lr: 0.000019 - momentum: 0.000000
137
+ 2023-10-17 12:37:24,133 epoch 5 - iter 792/1984 - loss 0.04917495 - time (sec): 37.13 - samples/sec: 1747.55 - lr: 0.000019 - momentum: 0.000000
138
+ 2023-10-17 12:37:33,311 epoch 5 - iter 990/1984 - loss 0.05024337 - time (sec): 46.30 - samples/sec: 1758.51 - lr: 0.000018 - momentum: 0.000000
139
+ 2023-10-17 12:37:42,390 epoch 5 - iter 1188/1984 - loss 0.04949250 - time (sec): 55.38 - samples/sec: 1746.11 - lr: 0.000018 - momentum: 0.000000
140
+ 2023-10-17 12:37:51,559 epoch 5 - iter 1386/1984 - loss 0.04899176 - time (sec): 64.55 - samples/sec: 1763.07 - lr: 0.000018 - momentum: 0.000000
141
+ 2023-10-17 12:38:01,063 epoch 5 - iter 1584/1984 - loss 0.04885371 - time (sec): 74.06 - samples/sec: 1757.50 - lr: 0.000017 - momentum: 0.000000
142
+ 2023-10-17 12:38:11,039 epoch 5 - iter 1782/1984 - loss 0.04868603 - time (sec): 84.03 - samples/sec: 1746.55 - lr: 0.000017 - momentum: 0.000000
143
+ 2023-10-17 12:38:20,399 epoch 5 - iter 1980/1984 - loss 0.04905152 - time (sec): 93.39 - samples/sec: 1752.32 - lr: 0.000017 - momentum: 0.000000
144
+ 2023-10-17 12:38:20,600 ----------------------------------------------------------------------------------------------------
145
+ 2023-10-17 12:38:20,600 EPOCH 5 done: loss 0.0490 - lr: 0.000017
146
+ 2023-10-17 12:38:24,317 DEV : loss 0.16858862340450287 - f1-score (micro avg) 0.7723
147
+ 2023-10-17 12:38:24,353 saving best model
148
+ 2023-10-17 12:38:24,882 ----------------------------------------------------------------------------------------------------
149
+ 2023-10-17 12:38:34,941 epoch 6 - iter 198/1984 - loss 0.03580324 - time (sec): 10.05 - samples/sec: 1626.98 - lr: 0.000016 - momentum: 0.000000
150
+ 2023-10-17 12:38:44,376 epoch 6 - iter 396/1984 - loss 0.03921053 - time (sec): 19.49 - samples/sec: 1675.47 - lr: 0.000016 - momentum: 0.000000
151
+ 2023-10-17 12:38:53,527 epoch 6 - iter 594/1984 - loss 0.04029851 - time (sec): 28.64 - samples/sec: 1719.10 - lr: 0.000016 - momentum: 0.000000
152
+ 2023-10-17 12:39:02,801 epoch 6 - iter 792/1984 - loss 0.03891771 - time (sec): 37.91 - samples/sec: 1733.83 - lr: 0.000015 - momentum: 0.000000
153
+ 2023-10-17 12:39:12,060 epoch 6 - iter 990/1984 - loss 0.03830666 - time (sec): 47.17 - samples/sec: 1727.45 - lr: 0.000015 - momentum: 0.000000
154
+ 2023-10-17 12:39:21,102 epoch 6 - iter 1188/1984 - loss 0.03802511 - time (sec): 56.21 - samples/sec: 1716.50 - lr: 0.000015 - momentum: 0.000000
155
+ 2023-10-17 12:39:30,203 epoch 6 - iter 1386/1984 - loss 0.03724256 - time (sec): 65.32 - samples/sec: 1728.66 - lr: 0.000014 - momentum: 0.000000
156
+ 2023-10-17 12:39:39,417 epoch 6 - iter 1584/1984 - loss 0.03704625 - time (sec): 74.53 - samples/sec: 1749.30 - lr: 0.000014 - momentum: 0.000000
157
+ 2023-10-17 12:39:48,403 epoch 6 - iter 1782/1984 - loss 0.03731915 - time (sec): 83.52 - samples/sec: 1751.47 - lr: 0.000014 - momentum: 0.000000
158
+ 2023-10-17 12:39:57,620 epoch 6 - iter 1980/1984 - loss 0.03699281 - time (sec): 92.73 - samples/sec: 1764.63 - lr: 0.000013 - momentum: 0.000000
159
+ 2023-10-17 12:39:57,804 ----------------------------------------------------------------------------------------------------
160
+ 2023-10-17 12:39:57,804 EPOCH 6 done: loss 0.0370 - lr: 0.000013
161
+ 2023-10-17 12:40:01,835 DEV : loss 0.18626417219638824 - f1-score (micro avg) 0.7649
162
+ 2023-10-17 12:40:01,856 ----------------------------------------------------------------------------------------------------
163
+ 2023-10-17 12:40:10,895 epoch 7 - iter 198/1984 - loss 0.02448812 - time (sec): 9.04 - samples/sec: 1728.08 - lr: 0.000013 - momentum: 0.000000
164
+ 2023-10-17 12:40:20,200 epoch 7 - iter 396/1984 - loss 0.02596145 - time (sec): 18.34 - samples/sec: 1808.38 - lr: 0.000013 - momentum: 0.000000
165
+ 2023-10-17 12:40:29,294 epoch 7 - iter 594/1984 - loss 0.02560325 - time (sec): 27.44 - samples/sec: 1795.83 - lr: 0.000012 - momentum: 0.000000
166
+ 2023-10-17 12:40:38,357 epoch 7 - iter 792/1984 - loss 0.02967938 - time (sec): 36.50 - samples/sec: 1784.59 - lr: 0.000012 - momentum: 0.000000
167
+ 2023-10-17 12:40:47,456 epoch 7 - iter 990/1984 - loss 0.02859136 - time (sec): 45.60 - samples/sec: 1786.09 - lr: 0.000012 - momentum: 0.000000
168
+ 2023-10-17 12:40:56,662 epoch 7 - iter 1188/1984 - loss 0.02775822 - time (sec): 54.81 - samples/sec: 1792.79 - lr: 0.000011 - momentum: 0.000000
169
+ 2023-10-17 12:41:05,904 epoch 7 - iter 1386/1984 - loss 0.02726516 - time (sec): 64.05 - samples/sec: 1799.31 - lr: 0.000011 - momentum: 0.000000
170
+ 2023-10-17 12:41:15,602 epoch 7 - iter 1584/1984 - loss 0.02669701 - time (sec): 73.75 - samples/sec: 1781.74 - lr: 0.000011 - momentum: 0.000000
171
+ 2023-10-17 12:41:24,753 epoch 7 - iter 1782/1984 - loss 0.02701107 - time (sec): 82.90 - samples/sec: 1779.19 - lr: 0.000010 - momentum: 0.000000
172
+ 2023-10-17 12:41:33,839 epoch 7 - iter 1980/1984 - loss 0.02701648 - time (sec): 91.98 - samples/sec: 1779.84 - lr: 0.000010 - momentum: 0.000000
173
+ 2023-10-17 12:41:34,032 ----------------------------------------------------------------------------------------------------
174
+ 2023-10-17 12:41:34,033 EPOCH 7 done: loss 0.0270 - lr: 0.000010
175
+ 2023-10-17 12:41:37,645 DEV : loss 0.21058925986289978 - f1-score (micro avg) 0.7653
176
+ 2023-10-17 12:41:37,668 ----------------------------------------------------------------------------------------------------
177
+ 2023-10-17 12:41:46,762 epoch 8 - iter 198/1984 - loss 0.01705926 - time (sec): 9.09 - samples/sec: 1768.31 - lr: 0.000010 - momentum: 0.000000
178
+ 2023-10-17 12:41:56,004 epoch 8 - iter 396/1984 - loss 0.01786307 - time (sec): 18.33 - samples/sec: 1751.47 - lr: 0.000009 - momentum: 0.000000
179
+ 2023-10-17 12:42:05,265 epoch 8 - iter 594/1984 - loss 0.01741473 - time (sec): 27.60 - samples/sec: 1751.47 - lr: 0.000009 - momentum: 0.000000
180
+ 2023-10-17 12:42:14,328 epoch 8 - iter 792/1984 - loss 0.01765238 - time (sec): 36.66 - samples/sec: 1770.48 - lr: 0.000009 - momentum: 0.000000
181
+ 2023-10-17 12:42:23,346 epoch 8 - iter 990/1984 - loss 0.01703398 - time (sec): 45.68 - samples/sec: 1791.83 - lr: 0.000008 - momentum: 0.000000
182
+ 2023-10-17 12:42:32,394 epoch 8 - iter 1188/1984 - loss 0.01847245 - time (sec): 54.72 - samples/sec: 1777.38 - lr: 0.000008 - momentum: 0.000000
183
+ 2023-10-17 12:42:41,750 epoch 8 - iter 1386/1984 - loss 0.01810602 - time (sec): 64.08 - samples/sec: 1784.70 - lr: 0.000008 - momentum: 0.000000
184
+ 2023-10-17 12:42:50,814 epoch 8 - iter 1584/1984 - loss 0.01757736 - time (sec): 73.14 - samples/sec: 1797.17 - lr: 0.000007 - momentum: 0.000000
185
+ 2023-10-17 12:43:00,069 epoch 8 - iter 1782/1984 - loss 0.01826162 - time (sec): 82.40 - samples/sec: 1793.46 - lr: 0.000007 - momentum: 0.000000
186
+ 2023-10-17 12:43:09,366 epoch 8 - iter 1980/1984 - loss 0.01810935 - time (sec): 91.70 - samples/sec: 1784.56 - lr: 0.000007 - momentum: 0.000000
187
+ 2023-10-17 12:43:09,566 ----------------------------------------------------------------------------------------------------
188
+ 2023-10-17 12:43:09,566 EPOCH 8 done: loss 0.0181 - lr: 0.000007
189
+ 2023-10-17 12:43:13,069 DEV : loss 0.2294856458902359 - f1-score (micro avg) 0.7625
190
+ 2023-10-17 12:43:13,092 ----------------------------------------------------------------------------------------------------
191
+ 2023-10-17 12:43:22,433 epoch 9 - iter 198/1984 - loss 0.01158194 - time (sec): 9.34 - samples/sec: 1781.47 - lr: 0.000006 - momentum: 0.000000
192
+ 2023-10-17 12:43:31,554 epoch 9 - iter 396/1984 - loss 0.01046138 - time (sec): 18.46 - samples/sec: 1815.26 - lr: 0.000006 - momentum: 0.000000
193
+ 2023-10-17 12:43:40,615 epoch 9 - iter 594/1984 - loss 0.01195194 - time (sec): 27.52 - samples/sec: 1778.57 - lr: 0.000006 - momentum: 0.000000
194
+ 2023-10-17 12:43:49,584 epoch 9 - iter 792/1984 - loss 0.01270015 - time (sec): 36.49 - samples/sec: 1793.86 - lr: 0.000005 - momentum: 0.000000
195
+ 2023-10-17 12:43:58,665 epoch 9 - iter 990/1984 - loss 0.01275360 - time (sec): 45.57 - samples/sec: 1798.68 - lr: 0.000005 - momentum: 0.000000
196
+ 2023-10-17 12:44:08,177 epoch 9 - iter 1188/1984 - loss 0.01293569 - time (sec): 55.08 - samples/sec: 1798.77 - lr: 0.000005 - momentum: 0.000000
197
+ 2023-10-17 12:44:17,253 epoch 9 - iter 1386/1984 - loss 0.01193801 - time (sec): 64.16 - samples/sec: 1785.93 - lr: 0.000004 - momentum: 0.000000
198
+ 2023-10-17 12:44:26,657 epoch 9 - iter 1584/1984 - loss 0.01171246 - time (sec): 73.56 - samples/sec: 1779.80 - lr: 0.000004 - momentum: 0.000000
199
+ 2023-10-17 12:44:35,793 epoch 9 - iter 1782/1984 - loss 0.01204442 - time (sec): 82.70 - samples/sec: 1778.13 - lr: 0.000004 - momentum: 0.000000
200
+ 2023-10-17 12:44:44,963 epoch 9 - iter 1980/1984 - loss 0.01199424 - time (sec): 91.87 - samples/sec: 1779.72 - lr: 0.000003 - momentum: 0.000000
201
+ 2023-10-17 12:44:45,158 ----------------------------------------------------------------------------------------------------
202
+ 2023-10-17 12:44:45,158 EPOCH 9 done: loss 0.0120 - lr: 0.000003
203
+ 2023-10-17 12:44:48,773 DEV : loss 0.239894837141037 - f1-score (micro avg) 0.766
204
+ 2023-10-17 12:44:48,802 ----------------------------------------------------------------------------------------------------
205
+ 2023-10-17 12:44:58,606 epoch 10 - iter 198/1984 - loss 0.00778125 - time (sec): 9.80 - samples/sec: 1640.50 - lr: 0.000003 - momentum: 0.000000
206
+ 2023-10-17 12:45:07,977 epoch 10 - iter 396/1984 - loss 0.01047784 - time (sec): 19.17 - samples/sec: 1692.10 - lr: 0.000003 - momentum: 0.000000
207
+ 2023-10-17 12:45:17,019 epoch 10 - iter 594/1984 - loss 0.00869626 - time (sec): 28.21 - samples/sec: 1768.46 - lr: 0.000002 - momentum: 0.000000
208
+ 2023-10-17 12:45:26,416 epoch 10 - iter 792/1984 - loss 0.00853870 - time (sec): 37.61 - samples/sec: 1758.41 - lr: 0.000002 - momentum: 0.000000
209
+ 2023-10-17 12:45:35,450 epoch 10 - iter 990/1984 - loss 0.00825831 - time (sec): 46.65 - samples/sec: 1761.77 - lr: 0.000002 - momentum: 0.000000
210
+ 2023-10-17 12:45:44,585 epoch 10 - iter 1188/1984 - loss 0.00873510 - time (sec): 55.78 - samples/sec: 1765.22 - lr: 0.000001 - momentum: 0.000000
211
+ 2023-10-17 12:45:53,971 epoch 10 - iter 1386/1984 - loss 0.00826154 - time (sec): 65.17 - samples/sec: 1758.30 - lr: 0.000001 - momentum: 0.000000
212
+ 2023-10-17 12:46:03,208 epoch 10 - iter 1584/1984 - loss 0.00813699 - time (sec): 74.40 - samples/sec: 1771.11 - lr: 0.000001 - momentum: 0.000000
213
+ 2023-10-17 12:46:12,401 epoch 10 - iter 1782/1984 - loss 0.00820321 - time (sec): 83.60 - samples/sec: 1769.26 - lr: 0.000000 - momentum: 0.000000
214
+ 2023-10-17 12:46:21,338 epoch 10 - iter 1980/1984 - loss 0.00880791 - time (sec): 92.53 - samples/sec: 1769.43 - lr: 0.000000 - momentum: 0.000000
215
+ 2023-10-17 12:46:21,524 ----------------------------------------------------------------------------------------------------
216
+ 2023-10-17 12:46:21,524 EPOCH 10 done: loss 0.0089 - lr: 0.000000
217
+ 2023-10-17 12:46:25,165 DEV : loss 0.24632978439331055 - f1-score (micro avg) 0.7697
218
+ 2023-10-17 12:46:25,612 ----------------------------------------------------------------------------------------------------
219
+ 2023-10-17 12:46:25,614 Loading model from best epoch ...
220
+ 2023-10-17 12:46:27,982 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
221
+ 2023-10-17 12:46:31,129
222
+ Results:
223
+ - F-score (micro) 0.788
224
+ - F-score (macro) 0.7077
225
+ - Accuracy 0.6683
226
+
227
+ By class:
228
+ precision recall f1-score support
229
+
230
+ LOC 0.8242 0.8733 0.8480 655
231
+ PER 0.7200 0.8072 0.7611 223
232
+ ORG 0.5246 0.5039 0.5141 127
233
+
234
+ micro avg 0.7655 0.8119 0.7880 1005
235
+ macro avg 0.6896 0.7281 0.7077 1005
236
+ weighted avg 0.7632 0.8119 0.7865 1005
237
+
238
+ 2023-10-17 12:46:31,129 ----------------------------------------------------------------------------------------------------