| Property | Value |
|---|---|
| License | MIT |
| Framework | PyTorch 1.7.1 |
| Accuracy | 93.23% |
| Base Model | RoBERTa-base |
What is roberta-base-sst2?
roberta-base-sst2 is a fine-tuned version of the RoBERTa base model specifically optimized for sentiment analysis using the SST2 (Stanford Sentiment Treebank) dataset. This model demonstrates impressive performance with 93.23% accuracy on the evaluation set and a loss of 0.1952.
Implementation Details
The model was trained using a carefully curated set of hyperparameters including a learning rate of 2e-05, batch sizes of 16 for training and 8 for evaluation, and the Adam optimizer. The training process spanned 10 epochs with a linear learning rate scheduler and 6% warmup ratio.
- Built on Transformers 4.21.3 framework
- Implements PyTorch 1.7.1
- Uses Datasets 1.18.3 and Tokenizers 0.11.6
- Trained with seed 42 for reproducibility
Core Capabilities
- High-accuracy sentiment classification (93.23%)
- Optimized for English language text
- Suitable for production deployment via inference endpoints
- Efficient text classification with transformer architecture
Frequently Asked Questions
Q: What makes this model unique?
This model stands out for its high accuracy in sentiment analysis tasks, achieved through careful fine-tuning of the RoBERTa architecture on the SST2 dataset. The training process showed consistent improvement, reaching optimal performance with minimal overfitting.
Q: What are the recommended use cases?
The model is ideal for sentiment analysis tasks in English text, particularly for binary classification scenarios. It's well-suited for production environments needing reliable sentiment detection with high accuracy.