| Property | Value |
|---|---|
| License | Apache 2.0 |
| Base Model | DistilRoBERTa-base |
| F1 Score | 0.8835 |
| Training Dataset | Financial Phrasebank + Kaggle Covid-19 Dataset |
What is distilroberta-finetuned-financial-text-classification?
This model is a specialized financial sentiment analysis tool built on the DistilRoBERTa architecture, fine-tuned on a comprehensive dataset of 4,840 financial news items. It's particularly notable for including Covid-19 impact data, making it relevant for modern financial analysis scenarios.
Implementation Details
The model was trained using carefully optimized hyperparameters, including a learning rate of 2e-05, batch size of 64, and Adam optimizer. The training process spanned 10 epochs with native AMP mixed precision training, achieving a final validation loss of 0.4463.
- Balanced weight adjustment for underrepresented classes
- Native AMP mixed precision training
- Linear learning rate scheduler
- Comprehensive Covid-19 impact integration
Core Capabilities
- Financial sentiment classification (negative, neutral, positive)
- Covid-19 impact analysis on market sentiment
- High-accuracy financial text analysis (89% accuracy)
- Efficient processing with DistilRoBERTa architecture
Frequently Asked Questions
Q: What makes this model unique?
The model's uniqueness lies in its integration of Covid-19 related financial sentiment data, making it more relevant for contemporary market analysis. Additionally, its weighted training approach ensures balanced performance across all sentiment categories.
Q: What are the recommended use cases?
The model is ideal for analyzing financial news articles, market reports, and economic updates. It's particularly useful for automated sentiment analysis in financial markets, especially when Covid-19 impact assessment is crucial.