Hugging Face's logo

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

Token Classification
Flair
PyTorch
TensorBoard
French
sequence-tagger-model
Model card Files Metrics Community
stefan-it commited on
Commit
a520b0e
•
1 Parent(s): 0167170

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.1697748481.46dc0c540dd0.4731.15 +3 -0
  5. test.tsv +0 -0
  6. training.log +245 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:39a64a7db6071a0e3865f02d8627b684b2398d800833f2e1e1f8c57edf60d4a8
3
+ size 19048098
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 20:48:25 0.0000 1.2221 0.3181 0.2899 0.0939 0.1418 0.0779
3
+ 2 20:48:50 0.0000 0.4144 0.2430 0.3774 0.4272 0.4008 0.2691
4
+ 3 20:49:16 0.0000 0.3337 0.2154 0.4227 0.4503 0.4361 0.2974
5
+ 4 20:49:41 0.0000 0.2910 0.2039 0.4476 0.4707 0.4589 0.3148
6
+ 5 20:50:06 0.0000 0.2639 0.1942 0.4859 0.5143 0.4997 0.3510
7
+ 6 20:50:31 0.0000 0.2434 0.1893 0.4937 0.5320 0.5121 0.3637
8
+ 7 20:50:57 0.0000 0.2274 0.1883 0.5204 0.5374 0.5288 0.3787
9
+ 8 20:51:22 0.0000 0.2197 0.1870 0.4884 0.5714 0.5266 0.3774
10
+ 9 20:51:47 0.0000 0.2128 0.1849 0.4899 0.5605 0.5228 0.3739
11
+ 10 20:52:13 0.0000 0.2059 0.1856 0.5000 0.5578 0.5273 0.3786
runs/events.out.tfevents.1697748481.46dc0c540dd0.4731.15 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:46026d7e1bf0d0c198652af8b75ad2a560498afaaa81ad972538848a15c311a5
3
+ size 502461
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,245 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-19 20:48:01,185 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-19 20:48:01,185 Model: "SequenceTagger(
3
+ (embeddings): TransformerWordEmbeddings(
4
+ (model): BertModel(
5
+ (embeddings): BertEmbeddings(
6
+ (word_embeddings): Embedding(32001, 128)
7
+ (position_embeddings): Embedding(512, 128)
8
+ (token_type_embeddings): Embedding(2, 128)
9
+ (LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
10
+ (dropout): Dropout(p=0.1, inplace=False)
11
+ )
12
+ (encoder): BertEncoder(
13
+ (layer): ModuleList(
14
+ (0-1): 2 x BertLayer(
15
+ (attention): BertAttention(
16
+ (self): BertSelfAttention(
17
+ (query): Linear(in_features=128, out_features=128, bias=True)
18
+ (key): Linear(in_features=128, out_features=128, bias=True)
19
+ (value): Linear(in_features=128, out_features=128, bias=True)
20
+ (dropout): Dropout(p=0.1, inplace=False)
21
+ )
22
+ (output): BertSelfOutput(
23
+ (dense): Linear(in_features=128, out_features=128, bias=True)
24
+ (LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
25
+ (dropout): Dropout(p=0.1, inplace=False)
26
+ )
27
+ )
28
+ (intermediate): BertIntermediate(
29
+ (dense): Linear(in_features=128, out_features=512, bias=True)
30
+ (intermediate_act_fn): GELUActivation()
31
+ )
32
+ (output): BertOutput(
33
+ (dense): Linear(in_features=512, out_features=128, bias=True)
34
+ (LayerNorm): LayerNorm((128,), 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=128, out_features=128, bias=True)
42
+ (activation): Tanh()
43
+ )
44
+ )
45
+ )
46
+ (locked_dropout): LockedDropout(p=0.5)
47
+ (linear): Linear(in_features=128, out_features=17, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-19 20:48:01,185 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-19 20:48:01,185 MultiCorpus: 7142 train + 698 dev + 2570 test sentences
52
+ - NER_HIPE_2022 Corpus: 7142 train + 698 dev + 2570 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fr/with_doc_seperator
53
+ 2023-10-19 20:48:01,185 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-19 20:48:01,185 Train: 7142 sentences
55
+ 2023-10-19 20:48:01,185 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-19 20:48:01,185 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-19 20:48:01,185 Training Params:
58
+ 2023-10-19 20:48:01,186 - learning_rate: "5e-05"
59
+ 2023-10-19 20:48:01,186 - mini_batch_size: "8"
60
+ 2023-10-19 20:48:01,186 - max_epochs: "10"
61
+ 2023-10-19 20:48:01,186 - shuffle: "True"
62
+ 2023-10-19 20:48:01,186 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-19 20:48:01,186 Plugins:
64
+ 2023-10-19 20:48:01,186 - TensorboardLogger
65
+ 2023-10-19 20:48:01,186 - LinearScheduler | warmup_fraction: '0.1'
66
+ 2023-10-19 20:48:01,186 ----------------------------------------------------------------------------------------------------
67
+ 2023-10-19 20:48:01,186 Final evaluation on model from best epoch (best-model.pt)
68
+ 2023-10-19 20:48:01,186 - metric: "('micro avg', 'f1-score')"
69
+ 2023-10-19 20:48:01,186 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-19 20:48:01,186 Computation:
71
+ 2023-10-19 20:48:01,186 - compute on device: cuda:0
72
+ 2023-10-19 20:48:01,186 - embedding storage: none
73
+ 2023-10-19 20:48:01,186 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-19 20:48:01,186 Model training base path: "hmbench-newseye/fr-dbmdz/bert-tiny-historic-multilingual-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4"
75
+ 2023-10-19 20:48:01,186 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-19 20:48:01,186 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-19 20:48:01,186 Logging anything other than scalars to TensorBoard is currently not supported.
78
+ 2023-10-19 20:48:03,348 epoch 1 - iter 89/893 - loss 3.31788487 - time (sec): 2.16 - samples/sec: 11311.02 - lr: 0.000005 - momentum: 0.000000
79
+ 2023-10-19 20:48:05,631 epoch 1 - iter 178/893 - loss 3.00220485 - time (sec): 4.44 - samples/sec: 11232.29 - lr: 0.000010 - momentum: 0.000000
80
+ 2023-10-19 20:48:07,923 epoch 1 - iter 267/893 - loss 2.52761114 - time (sec): 6.74 - samples/sec: 11190.98 - lr: 0.000015 - momentum: 0.000000
81
+ 2023-10-19 20:48:10,295 epoch 1 - iter 356/893 - loss 2.08600669 - time (sec): 9.11 - samples/sec: 11235.03 - lr: 0.000020 - momentum: 0.000000
82
+ 2023-10-19 20:48:12,581 epoch 1 - iter 445/893 - loss 1.83262530 - time (sec): 11.39 - samples/sec: 11102.08 - lr: 0.000025 - momentum: 0.000000
83
+ 2023-10-19 20:48:15,291 epoch 1 - iter 534/893 - loss 1.65539925 - time (sec): 14.10 - samples/sec: 10637.43 - lr: 0.000030 - momentum: 0.000000
84
+ 2023-10-19 20:48:17,545 epoch 1 - iter 623/893 - loss 1.51778252 - time (sec): 16.36 - samples/sec: 10597.71 - lr: 0.000035 - momentum: 0.000000
85
+ 2023-10-19 20:48:19,817 epoch 1 - iter 712/893 - loss 1.39687716 - time (sec): 18.63 - samples/sec: 10637.02 - lr: 0.000040 - momentum: 0.000000
86
+ 2023-10-19 20:48:22,116 epoch 1 - iter 801/893 - loss 1.30006566 - time (sec): 20.93 - samples/sec: 10682.81 - lr: 0.000045 - momentum: 0.000000
87
+ 2023-10-19 20:48:24,252 epoch 1 - iter 890/893 - loss 1.22292912 - time (sec): 23.07 - samples/sec: 10766.58 - lr: 0.000050 - momentum: 0.000000
88
+ 2023-10-19 20:48:24,309 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-19 20:48:24,309 EPOCH 1 done: loss 1.2221 - lr: 0.000050
90
+ 2023-10-19 20:48:25,279 DEV : loss 0.31805744767189026 - f1-score (micro avg) 0.1418
91
+ 2023-10-19 20:48:25,292 saving best model
92
+ 2023-10-19 20:48:25,326 ----------------------------------------------------------------------------------------------------
93
+ 2023-10-19 20:48:27,490 epoch 2 - iter 89/893 - loss 0.44839489 - time (sec): 2.16 - samples/sec: 12092.79 - lr: 0.000049 - momentum: 0.000000
94
+ 2023-10-19 20:48:29,745 epoch 2 - iter 178/893 - loss 0.45190939 - time (sec): 4.42 - samples/sec: 11426.89 - lr: 0.000049 - momentum: 0.000000
95
+ 2023-10-19 20:48:32,002 epoch 2 - iter 267/893 - loss 0.43792015 - time (sec): 6.68 - samples/sec: 11102.76 - lr: 0.000048 - momentum: 0.000000
96
+ 2023-10-19 20:48:34,385 epoch 2 - iter 356/893 - loss 0.43913817 - time (sec): 9.06 - samples/sec: 11081.95 - lr: 0.000048 - momentum: 0.000000
97
+ 2023-10-19 20:48:36,597 epoch 2 - iter 445/893 - loss 0.42993763 - time (sec): 11.27 - samples/sec: 11119.56 - lr: 0.000047 - momentum: 0.000000
98
+ 2023-10-19 20:48:38,813 epoch 2 - iter 534/893 - loss 0.43160330 - time (sec): 13.49 - samples/sec: 11151.24 - lr: 0.000047 - momentum: 0.000000
99
+ 2023-10-19 20:48:41,016 epoch 2 - iter 623/893 - loss 0.42662350 - time (sec): 15.69 - samples/sec: 11100.26 - lr: 0.000046 - momentum: 0.000000
100
+ 2023-10-19 20:48:43,209 epoch 2 - iter 712/893 - loss 0.42179532 - time (sec): 17.88 - samples/sec: 11134.96 - lr: 0.000046 - momentum: 0.000000
101
+ 2023-10-19 20:48:45,471 epoch 2 - iter 801/893 - loss 0.41961257 - time (sec): 20.14 - samples/sec: 11127.37 - lr: 0.000045 - momentum: 0.000000
102
+ 2023-10-19 20:48:47,717 epoch 2 - iter 890/893 - loss 0.41430294 - time (sec): 22.39 - samples/sec: 11086.08 - lr: 0.000044 - momentum: 0.000000
103
+ 2023-10-19 20:48:47,788 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-19 20:48:47,789 EPOCH 2 done: loss 0.4144 - lr: 0.000044
105
+ 2023-10-19 20:48:50,606 DEV : loss 0.2429792881011963 - f1-score (micro avg) 0.4008
106
+ 2023-10-19 20:48:50,619 saving best model
107
+ 2023-10-19 20:48:50,656 ----------------------------------------------------------------------------------------------------
108
+ 2023-10-19 20:48:52,917 epoch 3 - iter 89/893 - loss 0.34817659 - time (sec): 2.26 - samples/sec: 10227.98 - lr: 0.000044 - momentum: 0.000000
109
+ 2023-10-19 20:48:55,197 epoch 3 - iter 178/893 - loss 0.34419809 - time (sec): 4.54 - samples/sec: 10597.89 - lr: 0.000043 - momentum: 0.000000
110
+ 2023-10-19 20:48:57,484 epoch 3 - iter 267/893 - loss 0.34570190 - time (sec): 6.83 - samples/sec: 10723.08 - lr: 0.000043 - momentum: 0.000000
111
+ 2023-10-19 20:48:59,738 epoch 3 - iter 356/893 - loss 0.35673527 - time (sec): 9.08 - samples/sec: 10912.76 - lr: 0.000042 - momentum: 0.000000
112
+ 2023-10-19 20:49:01,958 epoch 3 - iter 445/893 - loss 0.35787478 - time (sec): 11.30 - samples/sec: 10929.46 - lr: 0.000042 - momentum: 0.000000
113
+ 2023-10-19 20:49:04,293 epoch 3 - iter 534/893 - loss 0.34975742 - time (sec): 13.64 - samples/sec: 10906.45 - lr: 0.000041 - momentum: 0.000000
114
+ 2023-10-19 20:49:06,578 epoch 3 - iter 623/893 - loss 0.34623751 - time (sec): 15.92 - samples/sec: 10910.39 - lr: 0.000041 - momentum: 0.000000
115
+ 2023-10-19 20:49:08,859 epoch 3 - iter 712/893 - loss 0.34105492 - time (sec): 18.20 - samples/sec: 10933.97 - lr: 0.000040 - momentum: 0.000000
116
+ 2023-10-19 20:49:11,148 epoch 3 - iter 801/893 - loss 0.33634117 - time (sec): 20.49 - samples/sec: 10927.23 - lr: 0.000039 - momentum: 0.000000
117
+ 2023-10-19 20:49:13,428 epoch 3 - iter 890/893 - loss 0.33396920 - time (sec): 22.77 - samples/sec: 10869.31 - lr: 0.000039 - momentum: 0.000000
118
+ 2023-10-19 20:49:13,514 ----------------------------------------------------------------------------------------------------
119
+ 2023-10-19 20:49:13,514 EPOCH 3 done: loss 0.3337 - lr: 0.000039
120
+ 2023-10-19 20:49:16,351 DEV : loss 0.21542218327522278 - f1-score (micro avg) 0.4361
121
+ 2023-10-19 20:49:16,366 saving best model
122
+ 2023-10-19 20:49:16,401 ----------------------------------------------------------------------------------------------------
123
+ 2023-10-19 20:49:18,738 epoch 4 - iter 89/893 - loss 0.30607610 - time (sec): 2.34 - samples/sec: 10539.75 - lr: 0.000038 - momentum: 0.000000
124
+ 2023-10-19 20:49:21,031 epoch 4 - iter 178/893 - loss 0.30905617 - time (sec): 4.63 - samples/sec: 10680.88 - lr: 0.000038 - momentum: 0.000000
125
+ 2023-10-19 20:49:23,318 epoch 4 - iter 267/893 - loss 0.29857101 - time (sec): 6.92 - samples/sec: 10720.72 - lr: 0.000037 - momentum: 0.000000
126
+ 2023-10-19 20:49:25,577 epoch 4 - iter 356/893 - loss 0.29476408 - time (sec): 9.17 - samples/sec: 10783.77 - lr: 0.000037 - momentum: 0.000000
127
+ 2023-10-19 20:49:27,831 epoch 4 - iter 445/893 - loss 0.29327210 - time (sec): 11.43 - samples/sec: 10831.79 - lr: 0.000036 - momentum: 0.000000
128
+ 2023-10-19 20:49:30,154 epoch 4 - iter 534/893 - loss 0.29218114 - time (sec): 13.75 - samples/sec: 10873.04 - lr: 0.000036 - momentum: 0.000000
129
+ 2023-10-19 20:49:32,370 epoch 4 - iter 623/893 - loss 0.29353192 - time (sec): 15.97 - samples/sec: 10834.37 - lr: 0.000035 - momentum: 0.000000
130
+ 2023-10-19 20:49:34,613 epoch 4 - iter 712/893 - loss 0.29422867 - time (sec): 18.21 - samples/sec: 10899.70 - lr: 0.000034 - momentum: 0.000000
131
+ 2023-10-19 20:49:36,847 epoch 4 - iter 801/893 - loss 0.29355665 - time (sec): 20.45 - samples/sec: 10841.59 - lr: 0.000034 - momentum: 0.000000
132
+ 2023-10-19 20:49:39,153 epoch 4 - iter 890/893 - loss 0.29116940 - time (sec): 22.75 - samples/sec: 10896.21 - lr: 0.000033 - momentum: 0.000000
133
+ 2023-10-19 20:49:39,226 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-19 20:49:39,227 EPOCH 4 done: loss 0.2910 - lr: 0.000033
135
+ 2023-10-19 20:49:41,610 DEV : loss 0.20385973155498505 - f1-score (micro avg) 0.4589
136
+ 2023-10-19 20:49:41,624 saving best model
137
+ 2023-10-19 20:49:41,659 ----------------------------------------------------------------------------------------------------
138
+ 2023-10-19 20:49:43,712 epoch 5 - iter 89/893 - loss 0.25219515 - time (sec): 2.05 - samples/sec: 11876.44 - lr: 0.000033 - momentum: 0.000000
139
+ 2023-10-19 20:49:45,968 epoch 5 - iter 178/893 - loss 0.26468865 - time (sec): 4.31 - samples/sec: 11290.43 - lr: 0.000032 - momentum: 0.000000
140
+ 2023-10-19 20:49:48,261 epoch 5 - iter 267/893 - loss 0.27013532 - time (sec): 6.60 - samples/sec: 11053.64 - lr: 0.000032 - momentum: 0.000000
141
+ 2023-10-19 20:49:50,577 epoch 5 - iter 356/893 - loss 0.26947573 - time (sec): 8.92 - samples/sec: 11028.19 - lr: 0.000031 - momentum: 0.000000
142
+ 2023-10-19 20:49:52,776 epoch 5 - iter 445/893 - loss 0.27194910 - time (sec): 11.12 - samples/sec: 11145.21 - lr: 0.000031 - momentum: 0.000000
143
+ 2023-10-19 20:49:54,655 epoch 5 - iter 534/893 - loss 0.26870618 - time (sec): 13.00 - samples/sec: 11451.82 - lr: 0.000030 - momentum: 0.000000
144
+ 2023-10-19 20:49:56,476 epoch 5 - iter 623/893 - loss 0.26893292 - time (sec): 14.82 - samples/sec: 11703.51 - lr: 0.000029 - momentum: 0.000000
145
+ 2023-10-19 20:49:58,730 epoch 5 - iter 712/893 - loss 0.26943529 - time (sec): 17.07 - samples/sec: 11640.91 - lr: 0.000029 - momentum: 0.000000
146
+ 2023-10-19 20:50:01,091 epoch 5 - iter 801/893 - loss 0.26558780 - time (sec): 19.43 - samples/sec: 11484.67 - lr: 0.000028 - momentum: 0.000000
147
+ 2023-10-19 20:50:03,363 epoch 5 - iter 890/893 - loss 0.26382300 - time (sec): 21.70 - samples/sec: 11422.87 - lr: 0.000028 - momentum: 0.000000
148
+ 2023-10-19 20:50:03,435 ----------------------------------------------------------------------------------------------------
149
+ 2023-10-19 20:50:03,435 EPOCH 5 done: loss 0.2639 - lr: 0.000028
150
+ 2023-10-19 20:50:06,328 DEV : loss 0.19422198832035065 - f1-score (micro avg) 0.4997
151
+ 2023-10-19 20:50:06,349 saving best model
152
+ 2023-10-19 20:50:06,385 ----------------------------------------------------------------------------------------------------
153
+ 2023-10-19 20:50:08,645 epoch 6 - iter 89/893 - loss 0.24073724 - time (sec): 2.26 - samples/sec: 11025.22 - lr: 0.000027 - momentum: 0.000000
154
+ 2023-10-19 20:50:10,868 epoch 6 - iter 178/893 - loss 0.23869287 - time (sec): 4.48 - samples/sec: 10742.28 - lr: 0.000027 - momentum: 0.000000
155
+ 2023-10-19 20:50:13,108 epoch 6 - iter 267/893 - loss 0.23653702 - time (sec): 6.72 - samples/sec: 10700.65 - lr: 0.000026 - momentum: 0.000000
156
+ 2023-10-19 20:50:15,366 epoch 6 - iter 356/893 - loss 0.23682304 - time (sec): 8.98 - samples/sec: 10740.98 - lr: 0.000026 - momentum: 0.000000
157
+ 2023-10-19 20:50:17,617 epoch 6 - iter 445/893 - loss 0.23901517 - time (sec): 11.23 - samples/sec: 10742.46 - lr: 0.000025 - momentum: 0.000000
158
+ 2023-10-19 20:50:19,888 epoch 6 - iter 534/893 - loss 0.24103990 - time (sec): 13.50 - samples/sec: 10774.77 - lr: 0.000024 - momentum: 0.000000
159
+ 2023-10-19 20:50:22,139 epoch 6 - iter 623/893 - loss 0.24250770 - time (sec): 15.75 - samples/sec: 10852.32 - lr: 0.000024 - momentum: 0.000000
160
+ 2023-10-19 20:50:24,407 epoch 6 - iter 712/893 - loss 0.24499793 - time (sec): 18.02 - samples/sec: 10936.70 - lr: 0.000023 - momentum: 0.000000
161
+ 2023-10-19 20:50:26,664 epoch 6 - iter 801/893 - loss 0.24412070 - time (sec): 20.28 - samples/sec: 10977.66 - lr: 0.000023 - momentum: 0.000000
162
+ 2023-10-19 20:50:28,983 epoch 6 - iter 890/893 - loss 0.24378175 - time (sec): 22.60 - samples/sec: 10963.58 - lr: 0.000022 - momentum: 0.000000
163
+ 2023-10-19 20:50:29,062 ----------------------------------------------------------------------------------------------------
164
+ 2023-10-19 20:50:29,062 EPOCH 6 done: loss 0.2434 - lr: 0.000022
165
+ 2023-10-19 20:50:31,413 DEV : loss 0.1892632693052292 - f1-score (micro avg) 0.5121
166
+ 2023-10-19 20:50:31,426 saving best model
167
+ 2023-10-19 20:50:31,462 ----------------------------------------------------------------------------------------------------
168
+ 2023-10-19 20:50:34,181 epoch 7 - iter 89/893 - loss 0.21102632 - time (sec): 2.72 - samples/sec: 8398.67 - lr: 0.000022 - momentum: 0.000000
169
+ 2023-10-19 20:50:36,427 epoch 7 - iter 178/893 - loss 0.22949609 - time (sec): 4.96 - samples/sec: 9490.59 - lr: 0.000021 - momentum: 0.000000
170
+ 2023-10-19 20:50:38,659 epoch 7 - iter 267/893 - loss 0.22714217 - time (sec): 7.20 - samples/sec: 10067.28 - lr: 0.000021 - momentum: 0.000000
171
+ 2023-10-19 20:50:40,892 epoch 7 - iter 356/893 - loss 0.22411998 - time (sec): 9.43 - samples/sec: 10212.94 - lr: 0.000020 - momentum: 0.000000
172
+ 2023-10-19 20:50:43,256 epoch 7 - iter 445/893 - loss 0.22268983 - time (sec): 11.79 - samples/sec: 10343.03 - lr: 0.000019 - momentum: 0.000000
173
+ 2023-10-19 20:50:45,358 epoch 7 - iter 534/893 - loss 0.22316819 - time (sec): 13.90 - samples/sec: 10467.56 - lr: 0.000019 - momentum: 0.000000
174
+ 2023-10-19 20:50:47,671 epoch 7 - iter 623/893 - loss 0.22564912 - time (sec): 16.21 - samples/sec: 10384.21 - lr: 0.000018 - momentum: 0.000000
175
+ 2023-10-19 20:50:50,031 epoch 7 - iter 712/893 - loss 0.22635014 - time (sec): 18.57 - samples/sec: 10611.04 - lr: 0.000018 - momentum: 0.000000
176
+ 2023-10-19 20:50:52,321 epoch 7 - iter 801/893 - loss 0.22691152 - time (sec): 20.86 - samples/sec: 10735.21 - lr: 0.000017 - momentum: 0.000000
177
+ 2023-10-19 20:50:54,568 epoch 7 - iter 890/893 - loss 0.22770015 - time (sec): 23.10 - samples/sec: 10734.74 - lr: 0.000017 - momentum: 0.000000
178
+ 2023-10-19 20:50:54,639 ----------------------------------------------------------------------------------------------------
179
+ 2023-10-19 20:50:54,639 EPOCH 7 done: loss 0.2274 - lr: 0.000017
180
+ 2023-10-19 20:50:56,997 DEV : loss 0.18830116093158722 - f1-score (micro avg) 0.5288
181
+ 2023-10-19 20:50:57,012 saving best model
182
+ 2023-10-19 20:50:57,045 ----------------------------------------------------------------------------------------------------
183
+ 2023-10-19 20:50:59,273 epoch 8 - iter 89/893 - loss 0.19916108 - time (sec): 2.23 - samples/sec: 10591.83 - lr: 0.000016 - momentum: 0.000000
184
+ 2023-10-19 20:51:01,665 epoch 8 - iter 178/893 - loss 0.21119586 - time (sec): 4.62 - samples/sec: 10765.32 - lr: 0.000016 - momentum: 0.000000
185
+ 2023-10-19 20:51:03,984 epoch 8 - iter 267/893 - loss 0.21979280 - time (sec): 6.94 - samples/sec: 10695.84 - lr: 0.000015 - momentum: 0.000000
186
+ 2023-10-19 20:51:06,299 epoch 8 - iter 356/893 - loss 0.21652746 - time (sec): 9.25 - samples/sec: 10710.65 - lr: 0.000014 - momentum: 0.000000
187
+ 2023-10-19 20:51:08,560 epoch 8 - iter 445/893 - loss 0.22339324 - time (sec): 11.51 - samples/sec: 10656.55 - lr: 0.000014 - momentum: 0.000000
188
+ 2023-10-19 20:51:10,778 epoch 8 - iter 534/893 - loss 0.22339473 - time (sec): 13.73 - samples/sec: 10739.10 - lr: 0.000013 - momentum: 0.000000
189
+ 2023-10-19 20:51:13,062 epoch 8 - iter 623/893 - loss 0.22183591 - time (sec): 16.02 - samples/sec: 10697.70 - lr: 0.000013 - momentum: 0.000000
190
+ 2023-10-19 20:51:15,299 epoch 8 - iter 712/893 - loss 0.21960297 - time (sec): 18.25 - samples/sec: 10714.14 - lr: 0.000012 - momentum: 0.000000
191
+ 2023-10-19 20:51:17,631 epoch 8 - iter 801/893 - loss 0.22015520 - time (sec): 20.59 - samples/sec: 10823.07 - lr: 0.000012 - momentum: 0.000000
192
+ 2023-10-19 20:51:19,875 epoch 8 - iter 890/893 - loss 0.21912504 - time (sec): 22.83 - samples/sec: 10865.83 - lr: 0.000011 - momentum: 0.000000
193
+ 2023-10-19 20:51:19,943 ----------------------------------------------------------------------------------------------------
194
+ 2023-10-19 20:51:19,943 EPOCH 8 done: loss 0.2197 - lr: 0.000011
195
+ 2023-10-19 20:51:22,761 DEV : loss 0.18704333901405334 - f1-score (micro avg) 0.5266
196
+ 2023-10-19 20:51:22,774 ----------------------------------------------------------------------------------------------------
197
+ 2023-10-19 20:51:25,008 epoch 9 - iter 89/893 - loss 0.22409608 - time (sec): 2.23 - samples/sec: 10905.75 - lr: 0.000011 - momentum: 0.000000
198
+ 2023-10-19 20:51:27,275 epoch 9 - iter 178/893 - loss 0.21545283 - time (sec): 4.50 - samples/sec: 10899.19 - lr: 0.000010 - momentum: 0.000000
199
+ 2023-10-19 20:51:29,461 epoch 9 - iter 267/893 - loss 0.22001584 - time (sec): 6.69 - samples/sec: 10951.22 - lr: 0.000009 - momentum: 0.000000
200
+ 2023-10-19 20:51:31,753 epoch 9 - iter 356/893 - loss 0.22377309 - time (sec): 8.98 - samples/sec: 10998.28 - lr: 0.000009 - momentum: 0.000000
201
+ 2023-10-19 20:51:34,036 epoch 9 - iter 445/893 - loss 0.22189321 - time (sec): 11.26 - samples/sec: 11162.57 - lr: 0.000008 - momentum: 0.000000
202
+ 2023-10-19 20:51:36,280 epoch 9 - iter 534/893 - loss 0.21704728 - time (sec): 13.50 - samples/sec: 11064.48 - lr: 0.000008 - momentum: 0.000000
203
+ 2023-10-19 20:51:38,531 epoch 9 - iter 623/893 - loss 0.21551115 - time (sec): 15.76 - samples/sec: 11080.72 - lr: 0.000007 - momentum: 0.000000
204
+ 2023-10-19 20:51:40,762 epoch 9 - iter 712/893 - loss 0.21465534 - time (sec): 17.99 - samples/sec: 11037.93 - lr: 0.000007 - momentum: 0.000000
205
+ 2023-10-19 20:51:43,131 epoch 9 - iter 801/893 - loss 0.21346279 - time (sec): 20.36 - samples/sec: 10987.62 - lr: 0.000006 - momentum: 0.000000
206
+ 2023-10-19 20:51:45,480 epoch 9 - iter 890/893 - loss 0.21290729 - time (sec): 22.70 - samples/sec: 10917.64 - lr: 0.000006 - momentum: 0.000000
207
+ 2023-10-19 20:51:45,552 ----------------------------------------------------------------------------------------------------
208
+ 2023-10-19 20:51:45,552 EPOCH 9 done: loss 0.2128 - lr: 0.000006
209
+ 2023-10-19 20:51:47,908 DEV : loss 0.18491902947425842 - f1-score (micro avg) 0.5228
210
+ 2023-10-19 20:51:47,922 ----------------------------------------------------------------------------------------------------
211
+ 2023-10-19 20:51:50,065 epoch 10 - iter 89/893 - loss 0.19783421 - time (sec): 2.14 - samples/sec: 12237.43 - lr: 0.000005 - momentum: 0.000000
212
+ 2023-10-19 20:51:52,362 epoch 10 - iter 178/893 - loss 0.19939993 - time (sec): 4.44 - samples/sec: 11666.48 - lr: 0.000004 - momentum: 0.000000
213
+ 2023-10-19 20:51:55,147 epoch 10 - iter 267/893 - loss 0.20297035 - time (sec): 7.22 - samples/sec: 10795.24 - lr: 0.000004 - momentum: 0.000000
214
+ 2023-10-19 20:51:57,423 epoch 10 - iter 356/893 - loss 0.20510717 - time (sec): 9.50 - samples/sec: 10783.91 - lr: 0.000003 - momentum: 0.000000
215
+ 2023-10-19 20:51:59,726 epoch 10 - iter 445/893 - loss 0.20467266 - time (sec): 11.80 - samples/sec: 10781.26 - lr: 0.000003 - momentum: 0.000000
216
+ 2023-10-19 20:52:01,947 epoch 10 - iter 534/893 - loss 0.20537293 - time (sec): 14.02 - samples/sec: 10806.69 - lr: 0.000002 - momentum: 0.000000
217
+ 2023-10-19 20:52:04,321 epoch 10 - iter 623/893 - loss 0.20318246 - time (sec): 16.40 - samples/sec: 10731.05 - lr: 0.000002 - momentum: 0.000000
218
+ 2023-10-19 20:52:06,590 epoch 10 - iter 712/893 - loss 0.20334880 - time (sec): 18.67 - samples/sec: 10705.27 - lr: 0.000001 - momentum: 0.000000
219
+ 2023-10-19 20:52:08,869 epoch 10 - iter 801/893 - loss 0.20499649 - time (sec): 20.95 - samples/sec: 10675.00 - lr: 0.000001 - momentum: 0.000000
220
+ 2023-10-19 20:52:11,144 epoch 10 - iter 890/893 - loss 0.20573355 - time (sec): 23.22 - samples/sec: 10664.10 - lr: 0.000000 - momentum: 0.000000
221
+ 2023-10-19 20:52:11,215 ----------------------------------------------------------------------------------------------------
222
+ 2023-10-19 20:52:11,215 EPOCH 10 done: loss 0.2059 - lr: 0.000000
223
+ 2023-10-19 20:52:13,598 DEV : loss 0.18561328947544098 - f1-score (micro avg) 0.5273
224
+ 2023-10-19 20:52:13,641 ----------------------------------------------------------------------------------------------------
225
+ 2023-10-19 20:52:13,642 Loading model from best epoch ...
226
+ 2023-10-19 20:52:13,719 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
227
+ 2023-10-19 20:52:18,262
228
+ Results:
229
+ - F-score (micro) 0.4184
230
+ - F-score (macro) 0.256
231
+ - Accuracy 0.2729
232
+
233
+ By class:
234
+ precision recall f1-score support
235
+
236
+ LOC 0.4267 0.4968 0.4591 1095
237
+ PER 0.4579 0.4733 0.4655 1012
238
+ ORG 0.1359 0.0784 0.0995 357
239
+ HumanProd 0.0000 0.0000 0.0000 33
240
+
241
+ micro avg 0.4159 0.4209 0.4184 2497
242
+ macro avg 0.2551 0.2621 0.2560 2497
243
+ weighted avg 0.3921 0.4209 0.4042 2497
244
+
245
+ 2023-10-19 20:52:18,262 ----------------------------------------------------------------------------------------------------