roberta-base-finetuned-yelp-polarity
| Property | Value |
|---|---|
| Author | VictorSanh |
| Base Model | RoBERTa-base |
| Task | Binary Sentiment Classification |
| Accuracy | 98.08% |
| Model URL | Hugging Face |
What is roberta-base-finetuned-yelp-polarity?
This is a specialized sentiment analysis model built upon the RoBERTa-base architecture, specifically fine-tuned for analyzing Yelp reviews. The model performs binary sentiment classification, categorizing reviews as either positive or negative with remarkable accuracy of 98.08% on the test set.
Implementation Details
The model was trained using carefully selected hyperparameters optimized for sentiment classification. The training process involved 2 epochs with a learning rate of 1e-05, using AdamW optimizer with epsilon value of 1e-08. The training utilized a batch size of 32 with gradient accumulation steps of 1 and implemented 3500 warmup steps.
- Weight decay: 0.0
- Maximum gradient norm: 1.0
- Training epochs: 2.0
- Batch size: 32 per device
Core Capabilities
- Binary sentiment classification of text
- Specialized for Yelp review analysis
- High accuracy performance (98.08%)
- Efficient processing with optimized parameters
Frequently Asked Questions
Q: What makes this model unique?
This model combines the robust architecture of RoBERTa-base with specialized training for Yelp review sentiment analysis, achieving exceptional accuracy while maintaining efficient processing capabilities.
Q: What are the recommended use cases?
The model is ideal for applications requiring binary sentiment analysis of customer reviews, particularly in the context of business reviews similar to Yelp's format. It can be effectively used for customer feedback analysis, market research, and automated review classification systems.
