|
|
--- |
|
|
license: apache-2.0 |
|
|
language: |
|
|
- en |
|
|
metrics: |
|
|
- accuracy |
|
|
- f1 |
|
|
- precision |
|
|
- recall |
|
|
base_model: FacebookAI/roberta-base |
|
|
pipeline_tag: text-classification |
|
|
library_name: transformers |
|
|
tags: |
|
|
- roberta |
|
|
- sentiment-analysis |
|
|
- transformers |
|
|
- text-classification |
|
|
- custom-dataset |
|
|
eval_results: |
|
|
eval_accuracy: 0.91 |
|
|
eval_f1: 0.90 |
|
|
eval_precision: 0.92 |
|
|
eval_recall: 0.89 |
|
|
--- |
|
|
|
|
|
# 🚀 Sentiment-RoBERTa-Base |
|
|
|
|
|
A fine-tuned [RoBERTa-base](https://huggingface.co/roberta-base) model for **binary sentiment classification** (positive/negative). |
|
|
Trained on a custom dataset across multiple sources including tweets, social comments, and headlines to handle **real-world tone detection**. |
|
|
|
|
|
✅ Use this model to build sentiment-aware applications, feedback classifiers, social media monitoring tools, or LLM content filters. |
|
|
|
|
|
## 🧠 Model Details |
|
|
|
|
|
| Property | Value | |
|
|
|-----------------------|---------------------------| |
|
|
| Base Model | `roberta-base` | |
|
|
| Fine-tuned Tasks | Binary Sentiment Analysis | |
|
|
| Classes | `0 = Negative`, `1 = Positive` | |
|
|
| Language | English (`en`) | |
|
|
| Dataset | Custom multi-source | |
|
|
| Framework | 🤗 Transformers | |
|
|
| Model Size | ~125M parameters | |
|
|
|
|
|
## 📊 Evaluation (on 20% held-out test set) |
|
|
|
|
|
| Metric | Score | |
|
|
|--------------|-------| |
|
|
| Accuracy | 91% | |
|
|
| F1 Score | 90% | |
|
|
| Precision | 92% | |
|
|
| Recall | 89% | |
|
|
|
|
|
## ⚙️ Quick Start |
|
|
|
|
|
### 💡 Install Required Libraries |
|
|
|
|
|
```bash |
|
|
pip install transformers torch |
