ludziej's picture
End of training
a0dda80 verified
|
raw
history blame
3.71 kB
metadata
tags:
  - generated_from_trainer
model-index:
  - name: calculator_model_test
    results: []

calculator_model_test

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

  • Loss: 0.0408

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: 512
  • eval_batch_size: 512
  • 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
3.1232 1.0 5 2.3752
2.1776 2.0 10 1.8879
1.7349 3.0 15 1.4779
1.3722 4.0 20 1.2288
1.1464 5.0 25 1.0354
0.995 6.0 30 0.9079
0.8838 7.0 35 0.8040
0.7827 8.0 40 0.7075
0.7079 9.0 45 0.6453
0.6523 10.0 50 0.5931
0.6042 11.0 55 0.5485
0.5548 12.0 60 0.5170
0.5339 13.0 65 0.5695
0.5492 14.0 70 0.4823
0.4951 15.0 75 0.4626
0.464 16.0 80 0.4308
0.4377 17.0 85 0.3924
0.4059 18.0 90 0.3690
0.3782 19.0 95 0.3322
0.3458 20.0 100 0.3135
0.3307 21.0 105 0.2936
0.2999 22.0 110 0.2577
0.2739 23.0 115 0.2444
0.2461 24.0 120 0.2236
0.2264 25.0 125 0.1957
0.2046 26.0 130 0.1637
0.1819 27.0 135 0.1415
0.16 28.0 140 0.1238
0.1454 29.0 145 0.1092
0.1297 30.0 150 0.0997
0.1188 31.0 155 0.0876
0.1105 32.0 160 0.0897
0.1033 33.0 165 0.0779
0.0941 34.0 170 0.0702
0.0897 35.0 175 0.0656
0.085 36.0 180 0.0634
0.0796 37.0 185 0.0596
0.0768 38.0 190 0.0560
0.0728 39.0 195 0.0556
0.0702 40.0 200 0.0516
0.0662 41.0 205 0.0493
0.063 42.0 210 0.0472
0.0613 43.0 215 0.0459
0.0617 44.0 220 0.0449
0.0579 45.0 225 0.0425
0.0556 46.0 230 0.0428
0.0555 47.0 235 0.0413
0.0536 48.0 240 0.0409
0.053 49.0 245 0.0409
0.0532 50.0 250 0.0408

Framework versions

  • Transformers 4.37.2
  • Pytorch 2.1.0+cu121
  • Datasets 2.17.1
  • Tokenizers 0.15.2