xlm-roberta-base-kin-finetuned-RC

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

  • Loss: 0.2068
  • F1: 0.5673
  • Roc Auc: 0.7554
  • Accuracy: 0.65

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: 32
  • eval_batch_size: 32
  • seed: 42
  • optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_steps: 100
  • num_epochs: 20

Training results

Training Loss Epoch Step Validation Loss F1 Roc Auc Accuracy
0.3372 1.0 123 0.3504 0.0 0.5 0.2959
0.3383 2.0 246 0.3199 0.0 0.5 0.2959
0.3208 3.0 369 0.2941 0.1868 0.5728 0.3480
0.2913 4.0 492 0.2852 0.2785 0.6222 0.3980
0.2613 5.0 615 0.2767 0.4047 0.6659 0.4531
0.2449 6.0 738 0.2649 0.4176 0.6791 0.4816
0.2226 7.0 861 0.2636 0.4390 0.6889 0.4939
0.2054 8.0 984 0.2459 0.4594 0.7017 0.5449
0.1972 9.0 1107 0.2435 0.4724 0.7094 0.5592
0.1667 10.0 1230 0.2430 0.4667 0.7068 0.5541
0.1566 11.0 1353 0.2261 0.4980 0.7253 0.6031
0.1404 12.0 1476 0.2213 0.4952 0.7210 0.6
0.149 13.0 1599 0.2167 0.5080 0.7317 0.6265
0.1228 14.0 1722 0.2144 0.5381 0.7443 0.6316
0.1327 15.0 1845 0.2115 0.5507 0.7487 0.6439
0.1236 16.0 1968 0.2092 0.5568 0.7512 0.65
0.118 17.0 2091 0.2075 0.5539 0.7503 0.6480
0.1201 18.0 2214 0.2072 0.5657 0.7535 0.6490
0.1068 19.0 2337 0.2068 0.5673 0.7554 0.65
0.1035 20.0 2460 0.2067 0.5672 0.7552 0.6520

Framework versions

  • Transformers 4.47.0
  • Pytorch 2.5.1+cu121
  • Datasets 3.2.0
  • Tokenizers 0.21.0
Downloads last month
9
Safetensors
Model size
278M params
Tensor type
F32
·
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.

Model tree for Kuongan/xlm-roberta-base-kin-finetuned-RC

Finetuned
(2758)
this model