| Property | Value |
|---|---|
| License | Apache 2.0 |
| Author | Apple |
| Task | Text Classification (SST-2) |
| Framework | PyTorch / Core ML |
What is ane-distilbert-base-uncased-finetuned-sst-2-english?
This model is a specialized version of DistilBERT optimized specifically for the Apple Neural Engine (ANE), designed to perform efficient sentiment classification using the SST-2 dataset. It represents a significant advancement in deploying transformer models on Apple devices, combining the efficiency of DistilBERT with hardware-specific optimizations.
Implementation Details
The model is available in two formats: PyTorch and Core ML (DistilBERT_fp16.mlpackage). While the PyTorch version serves as a reference implementation, the Core ML version is specifically optimized for hardware acceleration on Apple devices via the Neural Engine.
- Supports max sequence length of 128 tokens
- Includes both attention mask and input ID processing
- Optimized for fp16 precision on Apple Neural Engine
- Implements the complete DistilBERT architecture with SST-2 fine-tuning
Core Capabilities
- Efficient sentiment analysis on Apple devices
- Hardware-accelerated inference through ANE
- Seamless integration with both PyTorch and Core ML workflows
- Support for batch processing and dynamic input lengths
Frequently Asked Questions
Q: What makes this model unique?
This model stands out due to its specific optimization for the Apple Neural Engine, enabling efficient inference on Apple devices while maintaining the accuracy of the original DistilBERT model for sentiment analysis.
Q: What are the recommended use cases?
The model is ideal for iOS and macOS applications requiring sentiment analysis, particularly when hardware acceleration and efficient processing are priorities. It's especially suitable for on-device inference in production environments.