Pablo94's picture
End of training
4655901 verified
metadata
base_model: finiteautomata/beto-sentiment-analysis
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - precision
  - recall
model-index:
  - name: beto-sentiment-analysis-finetuned-detests-wandb24
    results: []

beto-sentiment-analysis-finetuned-detests-wandb24

This model is a fine-tuned version of finiteautomata/beto-sentiment-analysis on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.6204
  • Accuracy: 0.8674
  • F1-score: 0.7993
  • Precision: 0.8225
  • Recall: 0.7822
  • Auc: 0.7822

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: 5e-05
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 4

Training results

Training Loss Epoch Step Validation Loss Accuracy F1-score Precision Recall Auc
0.393 1.0 77 0.3365 0.8592 0.7633 0.8424 0.7287 0.7287
0.1947 2.0 154 0.3843 0.8396 0.7845 0.7716 0.8023 0.8023
0.0597 3.0 231 0.5486 0.8740 0.8046 0.8398 0.7814 0.7814
0.0028 4.0 308 0.6204 0.8674 0.7993 0.8225 0.7822 0.7822

Framework versions

  • Transformers 4.37.2
  • Pytorch 2.1.0+cu121
  • Datasets 2.17.0
  • Tokenizers 0.15.1