metadata
license: mit
base_model: roberta-large
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: roberta-large-hate-offensive-normal-speech-lr-2e-05
results: []
roberta-large-hate-offensive-normal-speech-lr-2e-05
This model is a fine-tuned version of roberta-large on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0293
- Accuracy: 0.9837
- Weighted f1: 0.9837
- Weighted recall: 0.9837
- Weighted precision: 0.9839
- Micro f1: 0.9837
- Micro recall: 0.9837
- Micro precision: 0.9837
- Macro f1: 0.9832
- Macro recall: 0.9821
- Macro precision: 0.9845
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: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Weighted f1 | Weighted recall | Weighted precision | Micro f1 | Micro recall | Micro precision | Macro f1 | Macro recall | Macro precision |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.5253 | 1.0 | 153 | 0.1270 | 0.9642 | 0.9647 | 0.9642 | 0.9681 | 0.9642 | 0.9642 | 0.9642 | 0.9633 | 0.9662 | 0.9633 |
| 0.0921 | 2.0 | 306 | 0.0878 | 0.9805 | 0.9805 | 0.9805 | 0.9807 | 0.9805 | 0.9805 | 0.9805 | 0.9803 | 0.9791 | 0.9818 |
| 0.0413 | 3.0 | 459 | 0.0590 | 0.9870 | 0.9870 | 0.9870 | 0.9875 | 0.9870 | 0.9870 | 0.9870 | 0.9860 | 0.9869 | 0.9857 |
| 0.0261 | 4.0 | 612 | 0.0523 | 0.9902 | 0.9902 | 0.9902 | 0.9904 | 0.9902 | 0.9902 | 0.9902 | 0.9896 | 0.9896 | 0.9900 |
| 0.012 | 5.0 | 765 | 0.0293 | 0.9837 | 0.9837 | 0.9837 | 0.9839 | 0.9837 | 0.9837 | 0.9837 | 0.9832 | 0.9821 | 0.9845 |
Framework versions
- Transformers 4.34.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.14.6.dev0
- Tokenizers 0.13.3