cls_sentimento_sebrae
| Property | Value |
|---|---|
| Parameter Count | 109M |
| Model Type | Text Classification (BERT-based) |
| Language | Portuguese |
| CO2 Emissions | 0.6308g |
| Accuracy | 96.5% |
What is cls_sentimento_sebrae?
cls_sentimento_sebrae is a specialized Portuguese language sentiment analysis model trained on Sebrae RS's internal dataset. It's designed to classify text into three sentiment categories: Negative, Neutral, and Positive. With over 400,000 downloads, this model demonstrates exceptional performance with a 96.5% accuracy rate and impressive F1 scores across multiple metrics.
Implementation Details
Built on the Transformers architecture using PyTorch and Safetensors, this model processes Portuguese text input through a BERT-based classification system. It's optimized for production use with integrated inference endpoints and shows remarkable efficiency with only 0.6308g of CO2 emissions during training.
- Architecture: BERT-based transformer model
- Training Metrics: 0.143 Loss, 0.935 Macro F1
- Precision/Recall: 0.938 Macro Precision, 0.933 Macro Recall
Core Capabilities
- Three-class sentiment classification (Negative, Neutral, Positive)
- Specialized for Portuguese language text
- Production-ready with inference endpoints
- Environmentally conscious with low carbon footprint
Frequently Asked Questions
Q: What makes this model unique?
The model's specialization in Portuguese language sentiment analysis, combined with its high accuracy and low environmental impact, makes it particularly valuable for Brazilian business applications, especially given its training on Sebrae RS data.
Q: What are the recommended use cases?
The model is ideal for sentiment analysis of Portuguese language customer feedback, reviews, and social media comments, particularly in Brazilian business contexts. Its high accuracy makes it suitable for production environments requiring reliable sentiment classification.
