bert-base-uncased-Abusive_Or_Threatening_Speech
This model is a fine-tuned version of bert-base-uncased. It achieves the following results on the evaluation set:
- Loss: 0.0874
- Accuracy: 0.9720
- F1: 0.7590
- Recall: 0.8406
- Precision: 0.6918
Model description
For more information on how it was created, check out the following link: https://github.com/DunnBC22/NLP_Projects/blob/main/Binary%20Classification/Malignant%20Comments/Malignant%20Comments%20-%20BERT-Base.ipynb
Intended uses & limitations
This model is intended to demonstrate my ability to solve a complex problem using technology.
Training and evaluation data
Dataset Source: https://www.kaggle.com/datasets/surekharamireddy/malignant-comment-classification
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 2
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Recall | Precision |
---|---|---|---|---|---|---|---|
0.1008 | 1.0 | 1531 | 0.0874 | 0.9720 | 0.7590 | 0.8406 | 0.6918 |
0.0673 | 2.0 | 3062 | 0.0981 | 0.9719 | 0.7591 | 0.8450 | 0.6891 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0
- Datasets 2.8.0
- Tokenizers 0.12.1