rubert-base-cased-sentiment
| Property | Value |
|---|---|
| Parameter Count | 178M |
| Downloads | 170,968 |
| Language | Russian |
| Task | Sentiment Analysis |
What is rubert-base-cased-sentiment?
This is a specialized Russian language sentiment analysis model based on DeepPavlov's rubert-base-cased-conversational architecture. The model has been trained on a comprehensive dataset of 351,797 texts and is designed to classify Russian text into three sentiment categories: neutral, positive, and negative.
Implementation Details
The model utilizes the BERT architecture with PyTorch framework and includes safetensors support. It processes input texts up to 512 tokens and returns probability distributions across the three sentiment classes.
- Built on DeepPavlov's conversational BERT base
- Trained on multiple Russian sentiment datasets including RuTweetCorp, RuReviews, and RuSentiment
- Supports batch processing with padding and truncation
Core Capabilities
- Three-way sentiment classification (Neutral, Positive, Negative)
- Handles various types of Russian text input
- Optimized for social media and review content
- Supports inference endpoints for production deployment
Frequently Asked Questions
Q: What makes this model unique?
This model stands out for its comprehensive training on diverse Russian language datasets, including social media posts, product reviews, and medical facility reviews, making it robust for real-world applications.
Q: What are the recommended use cases?
The model is particularly well-suited for analyzing customer feedback, social media monitoring, and automated sentiment analysis of Russian language content in business and research contexts.
