20230823184639 / README.md
dkqjrm's picture
update model card README.md
7d92de4
|
raw
history blame
5.07 kB
metadata
license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - super_glue
metrics:
  - accuracy
model-index:
  - name: '20230823184639'
    results: []

20230823184639

This model is a fine-tuned version of bert-large-cased on the super_glue dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1007
  • Accuracy: 0.7184

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 0.001
  • train_batch_size: 8
  • eval_batch_size: 8
  • seed: 11
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 60.0

Training results

Training Loss Epoch Step Validation Loss Accuracy
No log 1.0 312 0.2166 0.5307
0.2471 2.0 624 0.1849 0.5199
0.2471 3.0 936 0.2081 0.4729
0.2218 4.0 1248 0.1789 0.4910
0.221 5.0 1560 0.2006 0.4946
0.221 6.0 1872 0.1834 0.5632
0.2009 7.0 2184 0.1840 0.5523
0.2009 8.0 2496 0.1722 0.5415
0.1974 9.0 2808 0.1734 0.5668
0.1963 10.0 3120 0.1574 0.6245
0.1963 11.0 3432 0.2281 0.4982
0.1897 12.0 3744 0.1829 0.4982
0.1851 13.0 4056 0.1629 0.5379
0.1851 14.0 4368 0.1433 0.6498
0.1835 15.0 4680 0.1490 0.6426
0.1835 16.0 4992 0.1646 0.5812
0.1745 17.0 5304 0.1594 0.6390
0.1679 18.0 5616 0.1566 0.6462
0.1679 19.0 5928 0.1295 0.6895
0.1727 20.0 6240 0.1444 0.6354
0.1636 21.0 6552 0.1444 0.6282
0.1636 22.0 6864 0.1249 0.6823
0.1611 23.0 7176 0.1404 0.6606
0.1611 24.0 7488 0.1167 0.6859
0.1533 25.0 7800 0.1138 0.6895
0.1565 26.0 8112 0.1148 0.7148
0.1565 27.0 8424 0.1320 0.6462
0.1477 28.0 8736 0.1445 0.6643
0.152 29.0 9048 0.1106 0.6823
0.152 30.0 9360 0.1403 0.6823
0.1478 31.0 9672 0.1240 0.7076
0.1478 32.0 9984 0.1246 0.6823
0.1419 33.0 10296 0.1076 0.7184
0.1434 34.0 10608 0.1068 0.6931
0.1434 35.0 10920 0.1166 0.6968
0.1381 36.0 11232 0.1059 0.7004
0.1371 37.0 11544 0.1225 0.7040
0.1371 38.0 11856 0.1140 0.7076
0.1354 39.0 12168 0.1131 0.7256
0.1354 40.0 12480 0.1074 0.7148
0.1341 41.0 12792 0.1068 0.7329
0.1316 42.0 13104 0.1084 0.7004
0.1316 43.0 13416 0.1018 0.7148
0.1318 44.0 13728 0.1160 0.7292
0.1295 45.0 14040 0.1051 0.7148
0.1295 46.0 14352 0.1078 0.7076
0.128 47.0 14664 0.1059 0.7004
0.128 48.0 14976 0.1035 0.7256
0.1268 49.0 15288 0.1030 0.7004
0.1264 50.0 15600 0.1016 0.7148
0.1264 51.0 15912 0.1022 0.7004
0.1266 52.0 16224 0.1027 0.7040
0.1235 53.0 16536 0.1037 0.7112
0.1235 54.0 16848 0.1083 0.7184
0.121 55.0 17160 0.1008 0.7076
0.121 56.0 17472 0.1017 0.7184
0.1215 57.0 17784 0.1001 0.7148
0.1239 58.0 18096 0.1004 0.7148
0.1239 59.0 18408 0.1005 0.7184
0.1193 60.0 18720 0.1007 0.7184

Framework versions

  • Transformers 4.26.1
  • Pytorch 2.0.1+cu118
  • Datasets 2.12.0
  • Tokenizers 0.13.3