metadata
library_name: transformers
license: mit
base_model: cardiffnlp/twitter-roberta-large-hate-latest
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: twitter-roberta-large-hate-latest-offensive-eval-kn
results: []
twitter-roberta-large-hate-latest-offensive-eval-kn
This model is a fine-tuned version of cardiffnlp/twitter-roberta-large-hate-latest on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.7861
- Accuracy: 0.7427
- Precision: 0.4604
- Recall: 0.3928
- F1: 0.3901
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: 1e-05
- train_batch_size: 16
- eval_batch_size: 128
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 0.8805 | 0.9968 | 157 | 0.8450 | 0.7339 | 0.3531 | 0.3488 | 0.3407 |
| 0.7579 | 2.0 | 315 | 0.7816 | 0.7607 | 0.5057 | 0.4154 | 0.4200 |
| 0.7177 | 2.9968 | 472 | 0.7848 | 0.7571 | 0.4702 | 0.4043 | 0.4209 |
| 0.6914 | 4.0 | 630 | 0.8011 | 0.7446 | 0.4242 | 0.4029 | 0.4077 |
| 0.5218 | 4.9841 | 785 | 0.8166 | 0.7429 | 0.4122 | 0.4146 | 0.4115 |
Framework versions
- Transformers 4.45.1
- Pytorch 2.4.0
- Datasets 3.0.1
- Tokenizers 0.20.0