MatteoFasulo's picture
End of training
28a7edf verified
metadata
library_name: transformers
base_model: cardiffnlp/twitter-roberta-base-hate
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - precision
  - recall
model-index:
  - name: twitter-roberta-base-hate_69
    results: []

twitter-roberta-base-hate_69

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

  • Loss: 0.3841
  • F1-score: 0.8519
  • Accuracy: 0.8531
  • Precision: 0.8508
  • Recall: 0.8544

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-06
  • train_batch_size: 16
  • eval_batch_size: 16
  • seed: 69
  • optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: linear
  • num_epochs: 6

Training results

Training Loss Epoch Step Validation Loss F1-score Accuracy Precision Recall
No log 1.0 180 0.3803 0.8198 0.8217 0.8189 0.8212
No log 2.0 360 0.3648 0.8346 0.8357 0.8338 0.8379
0.3622 3.0 540 0.3787 0.8529 0.8531 0.8557 0.8603
0.3622 4.0 720 0.3777 0.8408 0.8427 0.8401 0.8417
0.3622 5.0 900 0.3768 0.8554 0.8566 0.8542 0.8575
0.2436 6.0 1080 0.3841 0.8519 0.8531 0.8508 0.8544

Framework versions

  • Transformers 4.47.1
  • Pytorch 2.5.1+cu124
  • Datasets 3.2.0
  • Tokenizers 0.21.0