gossminn's picture
update model card README.md
2328436
metadata
license: mit
tags:
  - generated_from_trainer
model-index:
  - name: predict-perception-bertino-focus-object
    results: []

predict-perception-bertino-focus-object

This model is a fine-tuned version of indigo-ai/BERTino on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.2766
  • R2: 0.5460

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: 0.0001
  • train_batch_size: 20
  • eval_batch_size: 8
  • seed: 1996
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 47

Training results

Training Loss Epoch Step Validation Loss R2
0.4798 1.0 14 0.4519 0.2581
0.2481 2.0 28 0.3042 0.5007
0.12 3.0 42 0.3746 0.3851
0.0969 4.0 56 0.3186 0.4770
0.0907 5.0 70 0.3727 0.3882
0.0673 6.0 84 0.2847 0.5327
0.0457 7.0 98 0.3141 0.4844
0.0431 8.0 112 0.3369 0.4470
0.028 9.0 126 0.3039 0.5012
0.0244 10.0 140 0.2964 0.5135
0.0201 11.0 154 0.3072 0.4958
0.0153 12.0 168 0.3049 0.4995
0.0155 13.0 182 0.2924 0.5201
0.015 14.0 196 0.2585 0.5757
0.0181 15.0 210 0.3258 0.4652
0.0136 16.0 224 0.3142 0.4842
0.0105 17.0 238 0.2536 0.5837
0.0104 18.0 252 0.2407 0.6050
0.0107 19.0 266 0.2727 0.5524
0.0084 20.0 280 0.3117 0.4883
0.0102 21.0 294 0.2999 0.5078
0.0074 22.0 308 0.3018 0.5047
0.0068 23.0 322 0.2826 0.5361
0.0054 24.0 336 0.2804 0.5398
0.0044 25.0 350 0.2912 0.5220
0.0048 26.0 364 0.2813 0.5382
0.005 27.0 378 0.2933 0.5186
0.0046 28.0 392 0.2820 0.5371
0.004 29.0 406 0.2717 0.5541
0.0054 30.0 420 0.2717 0.5540
0.0042 31.0 434 0.2699 0.5570
0.0033 32.0 448 0.2630 0.5684
0.0038 33.0 462 0.2578 0.5767
0.0032 34.0 476 0.2687 0.5589
0.004 35.0 490 0.2737 0.5507
0.0031 36.0 504 0.2753 0.5481
0.0037 37.0 518 0.2819 0.5373
0.0034 38.0 532 0.2759 0.5471
0.0034 39.0 546 0.2835 0.5347
0.0029 40.0 560 0.2814 0.5381
0.0033 41.0 574 0.2801 0.5403
0.0025 42.0 588 0.2759 0.5472
0.0029 43.0 602 0.2790 0.5421
0.0028 44.0 616 0.2801 0.5401
0.003 45.0 630 0.2772 0.5451
0.0028 46.0 644 0.2764 0.5463
0.0026 47.0 658 0.2766 0.5460

Framework versions

  • Transformers 4.16.2
  • Pytorch 1.10.2+cu113
  • Datasets 1.18.3
  • Tokenizers 0.11.0