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-hinglish-binary
results: []
twitter-roberta-large-hate-latest-hinglish-binary
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: 0.8288
- Accuracy: 0.6779
- Precision: 0.6455
- Recall: 0.6188
- F1: 0.6218
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: 128
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 0.6392 | 0.9709 | 25 | 0.6656 | 0.6376 | 0.3188 | 0.5 | 0.3894 |
| 0.63 | 1.9806 | 51 | 0.7095 | 0.6403 | 0.8197 | 0.5038 | 0.3975 |
| 0.6117 | 2.9903 | 77 | 0.6043 | 0.6921 | 0.6720 | 0.6174 | 0.6176 |
| 0.5974 | 4.0 | 103 | 0.6540 | 0.6730 | 0.6787 | 0.5667 | 0.5369 |
| 0.5512 | 4.9709 | 128 | 0.6216 | 0.6948 | 0.6745 | 0.6228 | 0.6243 |
| 0.5162 | 5.9806 | 154 | 0.6164 | 0.7057 | 0.6873 | 0.6394 | 0.6438 |
| 0.4525 | 6.9903 | 180 | 0.6715 | 0.6948 | 0.6757 | 0.6211 | 0.6221 |
| 0.4207 | 8.0 | 206 | 0.7581 | 0.7084 | 0.6844 | 0.6562 | 0.6620 |
| 0.3381 | 8.9709 | 231 | 0.7439 | 0.6812 | 0.6560 | 0.6575 | 0.6567 |
| 0.3016 | 9.7087 | 250 | 0.7481 | 0.7030 | 0.6762 | 0.6649 | 0.6688 |
Framework versions
- Transformers 4.45.1
- Pytorch 2.4.0
- Datasets 3.0.1
- Tokenizers 0.20.0