ludziej's picture
End of training
d847a74 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: 0.5850

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
3.396 1.0 6 2.7406
2.3939 2.0 12 1.9948
1.7939 3.0 18 1.6558
1.5985 4.0 24 1.6015
1.6024 5.0 30 1.5540
1.5383 6.0 36 1.5356
1.4729 7.0 42 1.5437
1.4981 8.0 48 1.4425
1.3622 9.0 54 1.4418
1.3585 10.0 60 1.3501
1.2883 11.0 66 1.2932
1.2361 12.0 72 1.3210
1.2274 13.0 78 1.2883
1.2667 14.0 84 1.2169
1.1679 15.0 90 1.1638
1.1237 16.0 96 1.1617
1.144 17.0 102 1.0919
1.0771 18.0 108 1.0928
1.059 19.0 114 1.0184
1.0024 20.0 120 0.9676
0.9949 21.0 126 0.9555
0.986 22.0 132 1.0286
1.0278 23.0 138 0.9337
0.9413 24.0 144 0.9162
0.912 25.0 150 0.9396
0.9621 26.0 156 0.8786
0.9102 27.0 162 0.8947
0.9384 28.0 168 0.8820
0.868 29.0 174 0.8468
0.8512 30.0 180 0.8478
0.8302 31.0 186 0.8099
0.8327 32.0 192 0.7622
0.8193 33.0 198 0.7596
0.7834 34.0 204 0.7329
0.7608 35.0 210 0.7229
0.7686 36.0 216 0.7257
0.7262 37.0 222 0.6940
0.7342 38.0 228 0.7117
0.7209 39.0 234 0.6646
0.7099 40.0 240 0.6530
0.6908 41.0 246 0.6397
0.6895 42.0 252 0.6356
0.6761 43.0 258 0.6202
0.668 44.0 264 0.6174
0.6735 45.0 270 0.6074
0.6502 46.0 276 0.6019
0.6427 47.0 282 0.5953
0.6326 48.0 288 0.5904
0.6467 49.0 294 0.5872
0.6307 50.0 300 0.5850

Framework versions

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