qfrodicio's picture
Training complete
fe1eb96 verified
metadata
base_model: MMG/mlm-spanish-roberta-base
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - precision
  - recall
  - f1
model-index:
  - name: roberta-finetuned-intention-prediction-es
    results: []

roberta-finetuned-intention-prediction-es

This model is a fine-tuned version of MMG/mlm-spanish-roberta-base on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 1.9097
  • Accuracy: 0.6918
  • Precision: 0.6953
  • Recall: 0.6918
  • F1: 0.6848

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: 16
  • eval_batch_size: 16
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 20

Training results

Training Loss Epoch Step Validation Loss Accuracy Precision Recall F1
2.2985 1.0 102 1.7435 0.4970 0.4378 0.4970 0.4215
1.3399 2.0 204 1.4205 0.5828 0.5872 0.5828 0.5624
0.8893 3.0 306 1.2699 0.6393 0.6276 0.6393 0.6192
0.5691 4.0 408 1.3327 0.6515 0.6604 0.6515 0.6417
0.3837 5.0 510 1.3836 0.6592 0.6710 0.6592 0.6528
0.2543 6.0 612 1.4253 0.6641 0.6703 0.6641 0.6528
0.1669 7.0 714 1.5317 0.6650 0.6795 0.6650 0.6546
0.1139 8.0 816 1.5939 0.6725 0.6754 0.6725 0.6615
0.0805 9.0 918 1.6987 0.6594 0.6696 0.6594 0.6518
0.0578 10.0 1020 1.6960 0.6793 0.6782 0.6793 0.6690
0.0374 11.0 1122 1.7590 0.6824 0.6877 0.6824 0.6729
0.03 12.0 1224 1.7425 0.6842 0.6859 0.6842 0.6785
0.0183 13.0 1326 1.8165 0.6830 0.6846 0.6830 0.6774
0.0152 14.0 1428 1.8348 0.6866 0.6927 0.6866 0.6799
0.0109 15.0 1530 1.8562 0.6940 0.6967 0.6940 0.6855
0.0097 16.0 1632 1.8766 0.6889 0.6947 0.6889 0.6833
0.0073 17.0 1734 1.8745 0.6920 0.6948 0.6920 0.6851
0.0062 18.0 1836 1.8944 0.6895 0.6919 0.6895 0.6825
0.0057 19.0 1938 1.9103 0.6936 0.6984 0.6936 0.6867
0.0052 20.0 2040 1.9097 0.6918 0.6953 0.6918 0.6848

Framework versions

  • Transformers 4.35.2
  • Pytorch 2.1.0+cu121
  • Datasets 2.16.1
  • Tokenizers 0.15.0