gossminn's picture
update model card README.md
8b04135
metadata
license: mit
tags:
  - generated_from_trainer
model-index:
  - name: predict-perception-bertino-focus-victim
    results: []

predict-perception-bertino-focus-victim

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: 0.2497
  • R2: 0.6131

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
0.5438 1.0 14 0.4405 0.3175
0.2336 2.0 28 0.2070 0.6792
0.0986 3.0 42 0.2868 0.5555
0.0907 4.0 56 0.2916 0.5481
0.0652 5.0 70 0.2187 0.6611
0.0591 6.0 84 0.2320 0.6406
0.0478 7.0 98 0.2501 0.6125
0.0347 8.0 112 0.2425 0.6243
0.021 9.0 126 0.2670 0.5863
0.0214 10.0 140 0.2853 0.5580
0.0172 11.0 154 0.2726 0.5776
0.0177 12.0 168 0.2629 0.5927
0.0152 13.0 182 0.2396 0.6287
0.012 14.0 196 0.2574 0.6012
0.0119 15.0 210 0.2396 0.6288
0.0128 16.0 224 0.2517 0.6100
0.0109 17.0 238 0.2509 0.6112
0.008 18.0 252 0.2522 0.6092
0.0101 19.0 266 0.2503 0.6121
0.0075 20.0 280 0.2527 0.6084
0.0082 21.0 294 0.2544 0.6058
0.0061 22.0 308 0.2510 0.6111
0.006 23.0 322 0.2402 0.6279
0.005 24.0 336 0.2539 0.6066
0.0058 25.0 350 0.2438 0.6222
0.0051 26.0 364 0.2439 0.6221
0.006 27.0 378 0.2442 0.6216
0.0061 28.0 392 0.2416 0.6257
0.0053 29.0 406 0.2519 0.6097
0.0045 30.0 420 0.2526 0.6085
0.0034 31.0 434 0.2578 0.6006
0.0039 32.0 448 0.2557 0.6038
0.0043 33.0 462 0.2538 0.6068
0.0041 34.0 476 0.2535 0.6072
0.0042 35.0 490 0.2560 0.6033
0.0037 36.0 504 0.2576 0.6009
0.0036 37.0 518 0.2634 0.5919
0.0037 38.0 532 0.2582 0.5999
0.0038 39.0 546 0.2552 0.6045
0.0034 40.0 560 0.2563 0.6028
0.0033 41.0 574 0.2510 0.6110
0.0029 42.0 588 0.2515 0.6103
0.0033 43.0 602 0.2525 0.6088
0.0028 44.0 616 0.2522 0.6093
0.0028 45.0 630 0.2526 0.6085
0.0027 46.0 644 0.2494 0.6136
0.0024 47.0 658 0.2497 0.6131

Framework versions

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