Skip to content

RusTorch v0.6.19: CoreML Integration with Jupyter Support

Choose a tag to compare

View all tags
@JunSuzukiJapan JunSuzukiJapan released this 20 Sep 04:30
· 48 commits to main since this release
v0.6.19
ec4ff7d
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

🚀 RusTorch v0.6.19 - CoreML Integration Release

🎯 Major Features

  • CoreML Integration: Full Apple Neural Engine support for Apple Silicon devices
  • Jupyter Support: Both Rust and Python kernels with CoreML acceleration
  • Python Bindings: Complete PyO3-based Python API with automatic differentiation
  • Hybrid Execution: Intelligent device fallback (CoreML → Metal → CPU)

📊 New Capabilities

  • Smart device selection with performance optimization
  • Multi-language documentation (10 languages: EN, DE, ES, FR, IT, JA, KO, PT, RU, ZH)
  • Enhanced error handling with unified RusTorchError system
  • Thread-safe tensor operations with Arc<RwLock> patterns

🔧 Technical Improvements

  • Automatic differentiation with Variable wrapper types
  • Efficient tensor operations using ndarray
  • Feature-gated compilation for optimal builds
  • Comprehensive test coverage including platform-specific tests

🌐 Platform Support

  • macOS: CoreML + Metal acceleration
  • Windows/Linux: CUDA + OpenCL support
  • WebAssembly: Browser-based tensor operations
  • Cross-platform compatibility maintained

📚 Documentation & Examples

  • Interactive Jupyter notebooks for all supported languages
  • CoreML performance comparison demos
  • Neural network training examples
  • Device management tutorials

🛠️ Development Experience

  • Improved Python interoperability
  • Enhanced error messages and debugging
  • Streamlined build process
  • Better development tooling integration

This release represents a significant milestone in bringing PyTorch-like functionality to Rust with native Apple Silicon acceleration.

Assets 2
Loading