toy / README.md
datarpit's picture
update model card README.md
9ffd6ea
metadata
license: apache-2.0
tags:
  - generated_from_trainer
model-index:
  - name: toy
    results: []

toy

This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.2124

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: 2e-05
  • train_batch_size: 8
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 50

Training results

Training Loss Epoch Step Validation Loss
0.4798 1.0 231 0.2252
0.3378 2.0 462 0.1777
0.1024 3.0 693 0.1586
0.0736 4.0 924 0.1664
0.1237 5.0 1155 0.1692
0.1049 6.0 1386 0.1818
0.0239 7.0 1617 0.2127
0.0036 8.0 1848 0.1888
0.0051 9.0 2079 0.2061
0.0003 10.0 2310 0.1905
0.0005 11.0 2541 0.2011
0.0003 12.0 2772 0.1928
0.0029 13.0 3003 0.2563
0.0002 14.0 3234 0.2076
0.0002 15.0 3465 0.1980
0.0001 16.0 3696 0.2013
0.0001 17.0 3927 0.2089
0.0001 18.0 4158 0.1984
0.0001 19.0 4389 0.2017
0.0001 20.0 4620 0.2013
0.0001 21.0 4851 0.2142
0.0001 22.0 5082 0.1943
0.0001 23.0 5313 0.2003
0.0 24.0 5544 0.2015
0.0001 25.0 5775 0.2031
0.0002 26.0 6006 0.2600
0.0022 27.0 6237 0.2269
0.0 28.0 6468 0.2125
0.0 29.0 6699 0.2172
0.0 30.0 6930 0.2185
0.0 31.0 7161 0.2004
0.0 32.0 7392 0.2077
0.0 33.0 7623 0.2333
0.0003 34.0 7854 0.2102
0.0 35.0 8085 0.2095
0.0 36.0 8316 0.2030
0.0 37.0 8547 0.2038
0.0 38.0 8778 0.2062
0.0 39.0 9009 0.2080
0.0 40.0 9240 0.2083
0.0 41.0 9471 0.2063
0.0 42.0 9702 0.2146
0.0 43.0 9933 0.2168
0.0 44.0 10164 0.2112
0.0 45.0 10395 0.2109
0.0 46.0 10626 0.2116
0.0 47.0 10857 0.2122
0.0 48.0 11088 0.2122
0.0 49.0 11319 0.2124
0.0 50.0 11550 0.2124

Framework versions

  • Transformers 4.17.0
  • Pytorch 1.10.0
  • Datasets 1.18.3
  • Tokenizers 0.11.6