Hugging Face's logo

hmteams
/
flair-hipe-2022-newseye-fr

Token Classification
Flair
PyTorch
TensorBoard
French
sequence-tagger-model
Model card Files Metrics Community
stefan-it commited on
Commit
8838757
·
1 Parent(s): 7080714

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.1697555360.0468bd9609d6.7281.12 +3 -0
  5. test.tsv +0 -0
  6. training.log +239 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:24cde505a28fb52b5204a1b83a949766e925dc1a373bd43817cbc813aeab49bd
3
+ size 440954373
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 15:10:33 0.0000 0.6293 0.1184 0.6949 0.7592 0.7256 0.5899
3
+ 2 15:11:49 0.0000 0.1053 0.1073 0.7566 0.8245 0.7891 0.6645
4
+ 3 15:13:04 0.0000 0.0670 0.1304 0.7654 0.8299 0.7963 0.6778
5
+ 4 15:14:19 0.0000 0.0472 0.1481 0.8029 0.8422 0.8220 0.7140
6
+ 5 15:15:35 0.0000 0.0349 0.1596 0.7847 0.8231 0.8035 0.6875
7
+ 6 15:16:50 0.0000 0.0280 0.1735 0.7995 0.8245 0.8118 0.7014
8
+ 7 15:18:05 0.0000 0.0211 0.1963 0.8264 0.8354 0.8309 0.7275
9
+ 8 15:19:21 0.0000 0.0152 0.2025 0.8132 0.8408 0.8268 0.7220
10
+ 9 15:20:35 0.0000 0.0117 0.2101 0.8138 0.8327 0.8231 0.7166
11
+ 10 15:21:53 0.0000 0.0085 0.2074 0.8172 0.8395 0.8282 0.7225
runs/events.out.tfevents.1697555360.0468bd9609d6.7281.12 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f891b38973ade8a321a080970c7641f91161b7d55cd43cf0d5d6124f069b75ba
3
+ size 502461
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,239 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-17 15:09:20,533 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-17 15:09:20,535 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=17, bias=True)
44
+ (loss_function): CrossEntropyLoss()
45
+ )"
46
+ 2023-10-17 15:09:20,535 ----------------------------------------------------------------------------------------------------
47
+ 2023-10-17 15:09:20,535 MultiCorpus: 7142 train + 698 dev + 2570 test sentences
48
+ - NER_HIPE_2022 Corpus: 7142 train + 698 dev + 2570 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fr/with_doc_seperator
49
+ 2023-10-17 15:09:20,535 ----------------------------------------------------------------------------------------------------
50
+ 2023-10-17 15:09:20,535 Train: 7142 sentences
51
+ 2023-10-17 15:09:20,535 (train_with_dev=False, train_with_test=False)
52
+ 2023-10-17 15:09:20,535 ----------------------------------------------------------------------------------------------------
53
+ 2023-10-17 15:09:20,535 Training Params:
54
+ 2023-10-17 15:09:20,535 - learning_rate: "3e-05"
55
+ 2023-10-17 15:09:20,535 - mini_batch_size: "8"
56
+ 2023-10-17 15:09:20,535 - max_epochs: "10"
57
+ 2023-10-17 15:09:20,535 - shuffle: "True"
58
+ 2023-10-17 15:09:20,535 ----------------------------------------------------------------------------------------------------
59
+ 2023-10-17 15:09:20,536 Plugins:
60
+ 2023-10-17 15:09:20,536 - TensorboardLogger
61
+ 2023-10-17 15:09:20,536 - LinearScheduler | warmup_fraction: '0.1'
62
+ 2023-10-17 15:09:20,536 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-17 15:09:20,536 Final evaluation on model from best epoch (best-model.pt)
64
+ 2023-10-17 15:09:20,536 - metric: "('micro avg', 'f1-score')"
65
+ 2023-10-17 15:09:20,536 ----------------------------------------------------------------------------------------------------
66
+ 2023-10-17 15:09:20,536 Computation:
67
+ 2023-10-17 15:09:20,536 - compute on device: cuda:0
68
+ 2023-10-17 15:09:20,536 - embedding storage: none
69
+ 2023-10-17 15:09:20,536 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-17 15:09:20,536 Model training base path: "hmbench-newseye/fr-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4"
71
+ 2023-10-17 15:09:20,536 ----------------------------------------------------------------------------------------------------
72
+ 2023-10-17 15:09:20,536 ----------------------------------------------------------------------------------------------------
73
+ 2023-10-17 15:09:20,536 Logging anything other than scalars to TensorBoard is currently not supported.
74
+ 2023-10-17 15:09:27,546 epoch 1 - iter 89/893 - loss 3.17468277 - time (sec): 7.01 - samples/sec: 3610.80 - lr: 0.000003 - momentum: 0.000000
75
+ 2023-10-17 15:09:34,586 epoch 1 - iter 178/893 - loss 2.10197739 - time (sec): 14.05 - samples/sec: 3591.62 - lr: 0.000006 - momentum: 0.000000
76
+ 2023-10-17 15:09:41,428 epoch 1 - iter 267/893 - loss 1.58385290 - time (sec): 20.89 - samples/sec: 3551.93 - lr: 0.000009 - momentum: 0.000000
77
+ 2023-10-17 15:09:48,238 epoch 1 - iter 356/893 - loss 1.29295235 - time (sec): 27.70 - samples/sec: 3536.72 - lr: 0.000012 - momentum: 0.000000
78
+ 2023-10-17 15:09:54,809 epoch 1 - iter 445/893 - loss 1.10083788 - time (sec): 34.27 - samples/sec: 3532.97 - lr: 0.000015 - momentum: 0.000000
79
+ 2023-10-17 15:10:01,485 epoch 1 - iter 534/893 - loss 0.95582211 - time (sec): 40.95 - samples/sec: 3558.63 - lr: 0.000018 - momentum: 0.000000
80
+ 2023-10-17 15:10:09,028 epoch 1 - iter 623/893 - loss 0.84747117 - time (sec): 48.49 - samples/sec: 3514.68 - lr: 0.000021 - momentum: 0.000000
81
+ 2023-10-17 15:10:16,180 epoch 1 - iter 712/893 - loss 0.75381137 - time (sec): 55.64 - samples/sec: 3538.81 - lr: 0.000024 - momentum: 0.000000
82
+ 2023-10-17 15:10:23,433 epoch 1 - iter 801/893 - loss 0.68340637 - time (sec): 62.90 - samples/sec: 3547.95 - lr: 0.000027 - momentum: 0.000000
83
+ 2023-10-17 15:10:30,490 epoch 1 - iter 890/893 - loss 0.63034171 - time (sec): 69.95 - samples/sec: 3548.03 - lr: 0.000030 - momentum: 0.000000
84
+ 2023-10-17 15:10:30,661 ----------------------------------------------------------------------------------------------------
85
+ 2023-10-17 15:10:30,661 EPOCH 1 done: loss 0.6293 - lr: 0.000030
86
+ 2023-10-17 15:10:33,451 DEV : loss 0.11842236667871475 - f1-score (micro avg) 0.7256
87
+ 2023-10-17 15:10:33,469 saving best model
88
+ 2023-10-17 15:10:33,816 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-17 15:10:41,206 epoch 2 - iter 89/893 - loss 0.12232714 - time (sec): 7.39 - samples/sec: 3748.60 - lr: 0.000030 - momentum: 0.000000
90
+ 2023-10-17 15:10:48,052 epoch 2 - iter 178/893 - loss 0.11680102 - time (sec): 14.23 - samples/sec: 3631.43 - lr: 0.000029 - momentum: 0.000000
91
+ 2023-10-17 15:10:54,953 epoch 2 - iter 267/893 - loss 0.11328328 - time (sec): 21.14 - samples/sec: 3618.39 - lr: 0.000029 - momentum: 0.000000
92
+ 2023-10-17 15:11:01,904 epoch 2 - iter 356/893 - loss 0.11241774 - time (sec): 28.09 - samples/sec: 3575.71 - lr: 0.000029 - momentum: 0.000000
93
+ 2023-10-17 15:11:08,351 epoch 2 - iter 445/893 - loss 0.11060263 - time (sec): 34.53 - samples/sec: 3583.85 - lr: 0.000028 - momentum: 0.000000
94
+ 2023-10-17 15:11:15,113 epoch 2 - iter 534/893 - loss 0.11047951 - time (sec): 41.30 - samples/sec: 3576.47 - lr: 0.000028 - momentum: 0.000000
95
+ 2023-10-17 15:11:22,245 epoch 2 - iter 623/893 - loss 0.11086841 - time (sec): 48.43 - samples/sec: 3549.93 - lr: 0.000028 - momentum: 0.000000
96
+ 2023-10-17 15:11:29,735 epoch 2 - iter 712/893 - loss 0.10835554 - time (sec): 55.92 - samples/sec: 3542.40 - lr: 0.000027 - momentum: 0.000000
97
+ 2023-10-17 15:11:36,607 epoch 2 - iter 801/893 - loss 0.10647861 - time (sec): 62.79 - samples/sec: 3542.10 - lr: 0.000027 - momentum: 0.000000
98
+ 2023-10-17 15:11:44,089 epoch 2 - iter 890/893 - loss 0.10520436 - time (sec): 70.27 - samples/sec: 3533.00 - lr: 0.000027 - momentum: 0.000000
99
+ 2023-10-17 15:11:44,268 ----------------------------------------------------------------------------------------------------
100
+ 2023-10-17 15:11:44,268 EPOCH 2 done: loss 0.1053 - lr: 0.000027
101
+ 2023-10-17 15:11:49,317 DEV : loss 0.10727142542600632 - f1-score (micro avg) 0.7891
102
+ 2023-10-17 15:11:49,336 saving best model
103
+ 2023-10-17 15:11:49,791 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-17 15:11:56,578 epoch 3 - iter 89/893 - loss 0.07446046 - time (sec): 6.79 - samples/sec: 3584.31 - lr: 0.000026 - momentum: 0.000000
105
+ 2023-10-17 15:12:03,419 epoch 3 - iter 178/893 - loss 0.07112677 - time (sec): 13.63 - samples/sec: 3654.95 - lr: 0.000026 - momentum: 0.000000
106
+ 2023-10-17 15:12:11,006 epoch 3 - iter 267/893 - loss 0.06742536 - time (sec): 21.21 - samples/sec: 3644.25 - lr: 0.000026 - momentum: 0.000000
107
+ 2023-10-17 15:12:17,869 epoch 3 - iter 356/893 - loss 0.06600880 - time (sec): 28.08 - samples/sec: 3630.84 - lr: 0.000025 - momentum: 0.000000
108
+ 2023-10-17 15:12:24,959 epoch 3 - iter 445/893 - loss 0.06825375 - time (sec): 35.17 - samples/sec: 3645.98 - lr: 0.000025 - momentum: 0.000000
109
+ 2023-10-17 15:12:31,281 epoch 3 - iter 534/893 - loss 0.06844461 - time (sec): 41.49 - samples/sec: 3627.52 - lr: 0.000025 - momentum: 0.000000
110
+ 2023-10-17 15:12:37,870 epoch 3 - iter 623/893 - loss 0.06742613 - time (sec): 48.08 - samples/sec: 3612.75 - lr: 0.000024 - momentum: 0.000000
111
+ 2023-10-17 15:12:45,061 epoch 3 - iter 712/893 - loss 0.06721754 - time (sec): 55.27 - samples/sec: 3605.20 - lr: 0.000024 - momentum: 0.000000
112
+ 2023-10-17 15:12:52,659 epoch 3 - iter 801/893 - loss 0.06717275 - time (sec): 62.87 - samples/sec: 3579.35 - lr: 0.000024 - momentum: 0.000000
113
+ 2023-10-17 15:12:59,217 epoch 3 - iter 890/893 - loss 0.06707278 - time (sec): 69.42 - samples/sec: 3571.54 - lr: 0.000023 - momentum: 0.000000
114
+ 2023-10-17 15:12:59,452 ----------------------------------------------------------------------------------------------------
115
+ 2023-10-17 15:12:59,452 EPOCH 3 done: loss 0.0670 - lr: 0.000023
116
+ 2023-10-17 15:13:04,383 DEV : loss 0.1304038017988205 - f1-score (micro avg) 0.7963
117
+ 2023-10-17 15:13:04,400 saving best model
118
+ 2023-10-17 15:13:04,853 ----------------------------------------------------------------------------------------------------
119
+ 2023-10-17 15:13:12,174 epoch 4 - iter 89/893 - loss 0.04868423 - time (sec): 7.32 - samples/sec: 3502.35 - lr: 0.000023 - momentum: 0.000000
120
+ 2023-10-17 15:13:19,287 epoch 4 - iter 178/893 - loss 0.04241356 - time (sec): 14.43 - samples/sec: 3520.55 - lr: 0.000023 - momentum: 0.000000
121
+ 2023-10-17 15:13:26,239 epoch 4 - iter 267/893 - loss 0.04403277 - time (sec): 21.38 - samples/sec: 3544.00 - lr: 0.000022 - momentum: 0.000000
122
+ 2023-10-17 15:13:32,878 epoch 4 - iter 356/893 - loss 0.04443889 - time (sec): 28.02 - samples/sec: 3549.20 - lr: 0.000022 - momentum: 0.000000
123
+ 2023-10-17 15:13:39,966 epoch 4 - iter 445/893 - loss 0.04684561 - time (sec): 35.11 - samples/sec: 3518.63 - lr: 0.000022 - momentum: 0.000000
124
+ 2023-10-17 15:13:46,763 epoch 4 - iter 534/893 - loss 0.04726451 - time (sec): 41.91 - samples/sec: 3522.77 - lr: 0.000021 - momentum: 0.000000
125
+ 2023-10-17 15:13:54,015 epoch 4 - iter 623/893 - loss 0.04633020 - time (sec): 49.16 - samples/sec: 3525.55 - lr: 0.000021 - momentum: 0.000000
126
+ 2023-10-17 15:14:01,146 epoch 4 - iter 712/893 - loss 0.04752612 - time (sec): 56.29 - samples/sec: 3522.62 - lr: 0.000021 - momentum: 0.000000
127
+ 2023-10-17 15:14:08,348 epoch 4 - iter 801/893 - loss 0.04742352 - time (sec): 63.49 - samples/sec: 3521.82 - lr: 0.000020 - momentum: 0.000000
128
+ 2023-10-17 15:14:15,272 epoch 4 - iter 890/893 - loss 0.04712026 - time (sec): 70.42 - samples/sec: 3519.42 - lr: 0.000020 - momentum: 0.000000
129
+ 2023-10-17 15:14:15,544 ----------------------------------------------------------------------------------------------------
130
+ 2023-10-17 15:14:15,544 EPOCH 4 done: loss 0.0472 - lr: 0.000020
131
+ 2023-10-17 15:14:19,791 DEV : loss 0.1480644792318344 - f1-score (micro avg) 0.822
132
+ 2023-10-17 15:14:19,808 saving best model
133
+ 2023-10-17 15:14:20,262 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-17 15:14:27,481 epoch 5 - iter 89/893 - loss 0.02697342 - time (sec): 7.21 - samples/sec: 3458.11 - lr: 0.000020 - momentum: 0.000000
135
+ 2023-10-17 15:14:34,191 epoch 5 - iter 178/893 - loss 0.02931270 - time (sec): 13.92 - samples/sec: 3520.13 - lr: 0.000019 - momentum: 0.000000
136
+ 2023-10-17 15:14:41,111 epoch 5 - iter 267/893 - loss 0.03360478 - time (sec): 20.84 - samples/sec: 3527.20 - lr: 0.000019 - momentum: 0.000000
137
+ 2023-10-17 15:14:47,741 epoch 5 - iter 356/893 - loss 0.03435094 - time (sec): 27.48 - samples/sec: 3527.12 - lr: 0.000019 - momentum: 0.000000
138
+ 2023-10-17 15:14:55,013 epoch 5 - iter 445/893 - loss 0.03355795 - time (sec): 34.75 - samples/sec: 3491.76 - lr: 0.000018 - momentum: 0.000000
139
+ 2023-10-17 15:15:02,081 epoch 5 - iter 534/893 - loss 0.03507889 - time (sec): 41.81 - samples/sec: 3504.28 - lr: 0.000018 - momentum: 0.000000
140
+ 2023-10-17 15:15:09,134 epoch 5 - iter 623/893 - loss 0.03463081 - time (sec): 48.87 - samples/sec: 3519.94 - lr: 0.000018 - momentum: 0.000000
141
+ 2023-10-17 15:15:16,208 epoch 5 - iter 712/893 - loss 0.03474055 - time (sec): 55.94 - samples/sec: 3524.52 - lr: 0.000017 - momentum: 0.000000
142
+ 2023-10-17 15:15:23,557 epoch 5 - iter 801/893 - loss 0.03527026 - time (sec): 63.29 - samples/sec: 3527.17 - lr: 0.000017 - momentum: 0.000000
143
+ 2023-10-17 15:15:30,470 epoch 5 - iter 890/893 - loss 0.03498960 - time (sec): 70.20 - samples/sec: 3534.99 - lr: 0.000017 - momentum: 0.000000
144
+ 2023-10-17 15:15:30,640 ----------------------------------------------------------------------------------------------------
145
+ 2023-10-17 15:15:30,640 EPOCH 5 done: loss 0.0349 - lr: 0.000017
146
+ 2023-10-17 15:15:35,393 DEV : loss 0.15961149334907532 - f1-score (micro avg) 0.8035
147
+ 2023-10-17 15:15:35,410 ----------------------------------------------------------------------------------------------------
148
+ 2023-10-17 15:15:42,408 epoch 6 - iter 89/893 - loss 0.02420361 - time (sec): 7.00 - samples/sec: 3548.22 - lr: 0.000016 - momentum: 0.000000
149
+ 2023-10-17 15:15:48,903 epoch 6 - iter 178/893 - loss 0.02406819 - time (sec): 13.49 - samples/sec: 3543.80 - lr: 0.000016 - momentum: 0.000000
150
+ 2023-10-17 15:15:56,114 epoch 6 - iter 267/893 - loss 0.02420778 - time (sec): 20.70 - samples/sec: 3530.79 - lr: 0.000016 - momentum: 0.000000
151
+ 2023-10-17 15:16:03,630 epoch 6 - iter 356/893 - loss 0.02473679 - time (sec): 28.22 - samples/sec: 3491.53 - lr: 0.000015 - momentum: 0.000000
152
+ 2023-10-17 15:16:10,515 epoch 6 - iter 445/893 - loss 0.02601161 - time (sec): 35.10 - samples/sec: 3511.49 - lr: 0.000015 - momentum: 0.000000
153
+ 2023-10-17 15:16:17,643 epoch 6 - iter 534/893 - loss 0.02577712 - time (sec): 42.23 - samples/sec: 3540.32 - lr: 0.000015 - momentum: 0.000000
154
+ 2023-10-17 15:16:24,783 epoch 6 - iter 623/893 - loss 0.02608412 - time (sec): 49.37 - samples/sec: 3539.28 - lr: 0.000014 - momentum: 0.000000
155
+ 2023-10-17 15:16:31,686 epoch 6 - iter 712/893 - loss 0.02665696 - time (sec): 56.27 - samples/sec: 3545.51 - lr: 0.000014 - momentum: 0.000000
156
+ 2023-10-17 15:16:38,520 epoch 6 - iter 801/893 - loss 0.02745268 - time (sec): 63.11 - samples/sec: 3553.56 - lr: 0.000014 - momentum: 0.000000
157
+ 2023-10-17 15:16:45,563 epoch 6 - iter 890/893 - loss 0.02802573 - time (sec): 70.15 - samples/sec: 3535.40 - lr: 0.000013 - momentum: 0.000000
158
+ 2023-10-17 15:16:45,762 ----------------------------------------------------------------------------------------------------
159
+ 2023-10-17 15:16:45,762 EPOCH 6 done: loss 0.0280 - lr: 0.000013
160
+ 2023-10-17 15:16:50,006 DEV : loss 0.17348243296146393 - f1-score (micro avg) 0.8118
161
+ 2023-10-17 15:16:50,025 ----------------------------------------------------------------------------------------------------
162
+ 2023-10-17 15:16:57,708 epoch 7 - iter 89/893 - loss 0.01778190 - time (sec): 7.68 - samples/sec: 3402.72 - lr: 0.000013 - momentum: 0.000000
163
+ 2023-10-17 15:17:04,684 epoch 7 - iter 178/893 - loss 0.01956122 - time (sec): 14.66 - samples/sec: 3453.95 - lr: 0.000013 - momentum: 0.000000
164
+ 2023-10-17 15:17:11,587 epoch 7 - iter 267/893 - loss 0.01856255 - time (sec): 21.56 - samples/sec: 3439.09 - lr: 0.000012 - momentum: 0.000000
165
+ 2023-10-17 15:17:18,586 epoch 7 - iter 356/893 - loss 0.01988639 - time (sec): 28.56 - samples/sec: 3483.76 - lr: 0.000012 - momentum: 0.000000
166
+ 2023-10-17 15:17:25,653 epoch 7 - iter 445/893 - loss 0.01938792 - time (sec): 35.63 - samples/sec: 3491.34 - lr: 0.000012 - momentum: 0.000000
167
+ 2023-10-17 15:17:32,453 epoch 7 - iter 534/893 - loss 0.02095563 - time (sec): 42.43 - samples/sec: 3513.52 - lr: 0.000011 - momentum: 0.000000
168
+ 2023-10-17 15:17:39,259 epoch 7 - iter 623/893 - loss 0.02165179 - time (sec): 49.23 - samples/sec: 3520.41 - lr: 0.000011 - momentum: 0.000000
169
+ 2023-10-17 15:17:46,264 epoch 7 - iter 712/893 - loss 0.02144328 - time (sec): 56.24 - samples/sec: 3506.66 - lr: 0.000011 - momentum: 0.000000
170
+ 2023-10-17 15:17:53,665 epoch 7 - iter 801/893 - loss 0.02112030 - time (sec): 63.64 - samples/sec: 3505.50 - lr: 0.000010 - momentum: 0.000000
171
+ 2023-10-17 15:18:00,568 epoch 7 - iter 890/893 - loss 0.02096110 - time (sec): 70.54 - samples/sec: 3519.21 - lr: 0.000010 - momentum: 0.000000
172
+ 2023-10-17 15:18:00,782 ----------------------------------------------------------------------------------------------------
173
+ 2023-10-17 15:18:00,782 EPOCH 7 done: loss 0.0211 - lr: 0.000010
174
+ 2023-10-17 15:18:05,006 DEV : loss 0.19632981717586517 - f1-score (micro avg) 0.8309
175
+ 2023-10-17 15:18:05,022 saving best model
176
+ 2023-10-17 15:18:05,530 ----------------------------------------------------------------------------------------------------
177
+ 2023-10-17 15:18:12,423 epoch 8 - iter 89/893 - loss 0.01429665 - time (sec): 6.89 - samples/sec: 3489.66 - lr: 0.000010 - momentum: 0.000000
178
+ 2023-10-17 15:18:19,109 epoch 8 - iter 178/893 - loss 0.01707880 - time (sec): 13.58 - samples/sec: 3533.97 - lr: 0.000009 - momentum: 0.000000
179
+ 2023-10-17 15:18:26,443 epoch 8 - iter 267/893 - loss 0.01664126 - time (sec): 20.91 - samples/sec: 3503.31 - lr: 0.000009 - momentum: 0.000000
180
+ 2023-10-17 15:18:33,523 epoch 8 - iter 356/893 - loss 0.01557798 - time (sec): 27.99 - samples/sec: 3547.51 - lr: 0.000009 - momentum: 0.000000
181
+ 2023-10-17 15:18:41,081 epoch 8 - iter 445/893 - loss 0.01559462 - time (sec): 35.55 - samples/sec: 3562.24 - lr: 0.000008 - momentum: 0.000000
182
+ 2023-10-17 15:18:48,096 epoch 8 - iter 534/893 - loss 0.01483953 - time (sec): 42.56 - samples/sec: 3576.21 - lr: 0.000008 - momentum: 0.000000
183
+ 2023-10-17 15:18:55,230 epoch 8 - iter 623/893 - loss 0.01549873 - time (sec): 49.70 - samples/sec: 3559.73 - lr: 0.000008 - momentum: 0.000000
184
+ 2023-10-17 15:19:02,083 epoch 8 - iter 712/893 - loss 0.01562235 - time (sec): 56.55 - samples/sec: 3547.03 - lr: 0.000007 - momentum: 0.000000
185
+ 2023-10-17 15:19:08,926 epoch 8 - iter 801/893 - loss 0.01515266 - time (sec): 63.39 - samples/sec: 3543.06 - lr: 0.000007 - momentum: 0.000000
186
+ 2023-10-17 15:19:15,547 epoch 8 - iter 890/893 - loss 0.01524696 - time (sec): 70.01 - samples/sec: 3538.74 - lr: 0.000007 - momentum: 0.000000
187
+ 2023-10-17 15:19:15,834 ----------------------------------------------------------------------------------------------------
188
+ 2023-10-17 15:19:15,834 EPOCH 8 done: loss 0.0152 - lr: 0.000007
189
+ 2023-10-17 15:19:21,293 DEV : loss 0.20254144072532654 - f1-score (micro avg) 0.8268
190
+ 2023-10-17 15:19:21,322 ----------------------------------------------------------------------------------------------------
191
+ 2023-10-17 15:19:28,171 epoch 9 - iter 89/893 - loss 0.01120597 - time (sec): 6.85 - samples/sec: 3518.61 - lr: 0.000006 - momentum: 0.000000
192
+ 2023-10-17 15:19:34,750 epoch 9 - iter 178/893 - loss 0.01273785 - time (sec): 13.43 - samples/sec: 3581.27 - lr: 0.000006 - momentum: 0.000000
193
+ 2023-10-17 15:19:41,616 epoch 9 - iter 267/893 - loss 0.01274289 - time (sec): 20.29 - samples/sec: 3581.05 - lr: 0.000006 - momentum: 0.000000
194
+ 2023-10-17 15:19:48,297 epoch 9 - iter 356/893 - loss 0.01222354 - time (sec): 26.97 - samples/sec: 3590.32 - lr: 0.000005 - momentum: 0.000000
195
+ 2023-10-17 15:19:55,627 epoch 9 - iter 445/893 - loss 0.01162771 - time (sec): 34.30 - samples/sec: 3585.72 - lr: 0.000005 - momentum: 0.000000
196
+ 2023-10-17 15:20:02,456 epoch 9 - iter 534/893 - loss 0.01219240 - time (sec): 41.13 - samples/sec: 3617.29 - lr: 0.000005 - momentum: 0.000000
197
+ 2023-10-17 15:20:09,469 epoch 9 - iter 623/893 - loss 0.01234197 - time (sec): 48.15 - samples/sec: 3586.34 - lr: 0.000004 - momentum: 0.000000
198
+ 2023-10-17 15:20:16,331 epoch 9 - iter 712/893 - loss 0.01245901 - time (sec): 55.01 - samples/sec: 3593.61 - lr: 0.000004 - momentum: 0.000000
199
+ 2023-10-17 15:20:23,343 epoch 9 - iter 801/893 - loss 0.01235256 - time (sec): 62.02 - samples/sec: 3585.82 - lr: 0.000004 - momentum: 0.000000
200
+ 2023-10-17 15:20:31,178 epoch 9 - iter 890/893 - loss 0.01174338 - time (sec): 69.85 - samples/sec: 3547.85 - lr: 0.000003 - momentum: 0.000000
201
+ 2023-10-17 15:20:31,426 ----------------------------------------------------------------------------------------------------
202
+ 2023-10-17 15:20:31,426 EPOCH 9 done: loss 0.0117 - lr: 0.000003
203
+ 2023-10-17 15:20:35,866 DEV : loss 0.21009869873523712 - f1-score (micro avg) 0.8231
204
+ 2023-10-17 15:20:35,892 ----------------------------------------------------------------------------------------------------
205
+ 2023-10-17 15:20:44,383 epoch 10 - iter 89/893 - loss 0.01168285 - time (sec): 8.49 - samples/sec: 2852.25 - lr: 0.000003 - momentum: 0.000000
206
+ 2023-10-17 15:20:51,070 epoch 10 - iter 178/893 - loss 0.00956698 - time (sec): 15.18 - samples/sec: 3187.54 - lr: 0.000003 - momentum: 0.000000
207
+ 2023-10-17 15:20:58,430 epoch 10 - iter 267/893 - loss 0.00810400 - time (sec): 22.54 - samples/sec: 3286.46 - lr: 0.000002 - momentum: 0.000000
208
+ 2023-10-17 15:21:05,344 epoch 10 - iter 356/893 - loss 0.00938662 - time (sec): 29.45 - samples/sec: 3290.52 - lr: 0.000002 - momentum: 0.000000
209
+ 2023-10-17 15:21:12,268 epoch 10 - iter 445/893 - loss 0.00895064 - time (sec): 36.37 - samples/sec: 3304.60 - lr: 0.000002 - momentum: 0.000000
210
+ 2023-10-17 15:21:19,480 epoch 10 - iter 534/893 - loss 0.00957070 - time (sec): 43.58 - samples/sec: 3335.18 - lr: 0.000001 - momentum: 0.000000
211
+ 2023-10-17 15:21:26,763 epoch 10 - iter 623/893 - loss 0.00910774 - time (sec): 50.87 - samples/sec: 3355.39 - lr: 0.000001 - momentum: 0.000000
212
+ 2023-10-17 15:21:33,995 epoch 10 - iter 712/893 - loss 0.00887038 - time (sec): 58.10 - samples/sec: 3359.76 - lr: 0.000001 - momentum: 0.000000
213
+ 2023-10-17 15:21:41,182 epoch 10 - iter 801/893 - loss 0.00857775 - time (sec): 65.29 - samples/sec: 3378.07 - lr: 0.000000 - momentum: 0.000000
214
+ 2023-10-17 15:21:48,658 epoch 10 - iter 890/893 - loss 0.00850610 - time (sec): 72.76 - samples/sec: 3408.45 - lr: 0.000000 - momentum: 0.000000
215
+ 2023-10-17 15:21:48,903 ----------------------------------------------------------------------------------------------------
216
+ 2023-10-17 15:21:48,903 EPOCH 10 done: loss 0.0085 - lr: 0.000000
217
+ 2023-10-17 15:21:53,236 DEV : loss 0.207401305437088 - f1-score (micro avg) 0.8282
218
+ 2023-10-17 15:21:53,620 ----------------------------------------------------------------------------------------------------
219
+ 2023-10-17 15:21:53,622 Loading model from best epoch ...
220
+ 2023-10-17 15:21:54,979 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
221
+ 2023-10-17 15:22:05,756
222
+ Results:
223
+ - F-score (micro) 0.7185
224
+ - F-score (macro) 0.6326
225
+ - Accuracy 0.5744
226
+
227
+ By class:
228
+ precision recall f1-score support
229
+
230
+ LOC 0.7239 0.7397 0.7317 1095
231
+ PER 0.7950 0.7816 0.7882 1012
232
+ ORG 0.4939 0.5630 0.5262 357
233
+ HumanProd 0.3710 0.6970 0.4842 33
234
+
235
+ micro avg 0.7065 0.7309 0.7185 2497
236
+ macro avg 0.5959 0.6953 0.6326 2497
237
+ weighted avg 0.7151 0.7309 0.7220 2497
238
+
239
+ 2023-10-17 15:22:05,756 ----------------------------------------------------------------------------------------------------