Melvin Alvarez
6.8 KiB
6.8 KiB
OptimizR v1.0.0 Release Notes
Release Date: February 16, 2026
Status: ✅ Stable Release
🎉 First Stable Release
OptimizR v1.0.0 marks the first production-ready stable release with a commitment to semantic versioning going forward. The API is now stable and breaking changes will only occur in major version bumps.
📦 Distribution
crates.io (Rust)
cargo add optimizr
PyPI (Python)
pip install optimizr
🆕 What's New in v1.0.0
Publication & Distribution
- ✅ Published to crates.io - Available in Rust package registry
- ✅ Published to PyPI - Available via pip install
- ✅ Stable API - Semantic versioning from v1.0.0 forward
- ✅ Production Ready - Comprehensive testing and validation
Documentation
- 📚 ReadTheDocs - Full documentation at https://optimiz-r.readthedocs.io
- 📖 Getting Started Guide - Quick start for new users
- 📝 API Reference - Complete function and class documentation
- 🎓 Tutorials - Step-by-step guides for all algorithms
- 🔬 Theory & Math - Mathematical foundations and references
Build System Improvements
- 🏗️ Fixed Cargo.toml - Removed python-bindings from default features
- Resolves linker errors when using as Rust library
- Python bindings now opt-in feature (automatically enabled by maturin)
- 🐍 Maturin Configuration - Explicit python-bindings feature in pyproject.toml
- Ensures correct PyO3 extension builds for PyPI
- Fixes cross-platform compatibility
Metadata Updates
- 👥 Authors: HFThot Research Lab admin@hfthot-lab.eu
- 🔗 Repository: https://github.com/ThotDjehuty/optimiz-r
- 📚 Documentation: https://optimiz-r.readthedocs.io
🚀 Features (Stable)
Optimization Algorithms
- ✅ Differential Evolution - 5 strategies (rand/1, best/1, current-to-best/1, rand/2, best/2)
- ✅ Adaptive jDE - Self-tuning mutation factor and crossover rate
- ✅ Grid Search - Exhaustive parameter space exploration
Hidden Markov Models
- ✅ Baum-Welch Training - EM algorithm for parameter learning
- ✅ Viterbi Decoding - Most likely state sequence
- ✅ Gaussian Emissions - Continuous observation models
MCMC Sampling
- ✅ Metropolis-Hastings - Bayesian parameter estimation
- ✅ Adaptive Proposals - Gaussian random walk
- ✅ Convergence Diagnostics - Acceptance rate tracking
Mean Field Games (v0.3.0+)
- ✅ 1D MFG Solver - Large population dynamics
- ✅ HJB-Fokker-Planck Coupling - Fixed-point iteration
- ✅ Agent Population Dynamics - Spatial-temporal evolution
Mathematical Toolkit
- ✅ Numerical Differentiation - gradient(), hessian(), jacobian()
- ✅ Statistics - mean(), variance(), skewness(), kurtosis()
- ✅ Linear Algebra - norms, normalization, trace, outer product
- ✅ Information Theory - mutual_information(), shannon_entropy()
⚡ Performance
- 50-100× faster than pure Python implementations
- 95% memory reduction vs NumPy/SciPy
- Parallel-ready with Rayon infrastructure
- Production-tested on multi-dimensional problems
📊 Benchmarks
Differential Evolution (Rosenbrock 10D)
- OptimizR (Rust): 0.12s
- SciPy (Python): 8.9s
- Speedup: 74×
HMM Training (1000 observations, 3 states)
- OptimizR (Rust): 0.03s
- hmmlearn (Python): 2.4s
- Speedup: 80×
Mean Field Games (100×100 grid)
- OptimizR (Rust): 0.4s
- Pure Python: 45s
- Speedup: 112×
🔧 Breaking Changes from v0.3.0
Cargo Feature Flags
# OLD (v0.3.0):
[features]
default = ["python-bindings"] # Always included
# NEW (v1.0.0):
[features]
default = [] # No default features
python-bindings = ["pyo3", "numpy"] # Opt-in
Impact:
- Rust-only users: No breaking changes (python-bindings not needed)
- Python users: No impact (maturin automatically enables python-bindings)
If you're using OptimizR as a Rust library and explicitly depend on Python bindings:
# Update your Cargo.toml:
[dependencies]
optimizr = { version = "1.0", features = ["python-bindings"] }
📝 Migration Guide
From v0.3.0 to v1.0.0
For Rust Users: No code changes required. If you were using python-bindings explicitly, add it to features list.
For Python Users:
# Upgrade via pip
pip install --upgrade optimizr
# Or specify version
pip install optimizr==1.0.0
API Compatibility:
✅ All Python APIs remain unchanged
✅ All Rust APIs remain unchanged
✅ Function signatures are identical
✅ Return types are identical
✅ No deprecations or removals
🐛 Bug Fixes
- Fixed linking errors when using OptimizR as Rust-only library
- Fixed PyInit__core symbol warning in maturin builds
- Resolved flate2 yanked dependency warning
📚 Documentation
New Documentation
- Complete ReadTheDocs site: https://optimiz-r.readthedocs.io
- Getting Started guide
- Installation instructions for all platforms
- Tutorial notebooks (7 validated examples)
- API reference with examples
- Theory and mathematical background
Validated Tutorial Notebooks
- ✅ Hidden Markov Models - Regime detection
- ✅ MCMC Sampling - Bayesian inference
- ✅ Differential Evolution - Global optimization
- ✅ Optimal Control - HJB solver (theory)
- ✅ Real-World Applications - Complete workflows
- ✅ Performance Benchmarks - Rust vs Python
- ✅ Mean Field Games - Population dynamics
🔮 Roadmap
v1.1.0 (Q2 2026)
- Additional DE variants (JADE, SHADE, L-SHADE)
- Particle Swarm Optimization (PSO)
- CMA-ES algorithm
- More HMM emission distributions
v1.2.0 (Q3 2026)
- GPU acceleration via CUDA/ROCm
- Additional language bindings (R, Julia, JavaScript)
- Distributed computing support
- Advanced parallel strategies
v2.0.0 (2027)
- Neural Evolution Strategies (NES)
- Multi-objective optimization
- Constraint handling methods
- Advanced uncertainty quantification
🙏 Acknowledgments
Built with:
- Rust - Systems programming language
- PyO3 - Rust bindings for Python
- Maturin - Build and publish Rust crates as Python packages
- NumPy - Numerical computing in Python
Inspired by:
- scipy.optimize
- scikit-learn
- hmmlearn
- emcee
📞 Support & Community
- Issues: GitHub Issues
- Discussions: GitHub Discussions
- Email: contact@hfthot-lab.eu
📄 License
MIT License - see LICENSE file for details.
OptimizR v1.0.0 - Fast optimization for data science and machine learning 🚀
Thank you to all contributors and early adopters who helped make this release possible!