Mini-ML: High-Performance Vector & Matrix Library in C for Python

Mini-ML is a lightweight C backend library integrated with Python for fast vector and matrix operations, designed as a foundation for building machine learning algorithms. It leverages low-level C optimizations and Python wrappers to give you performance comparable to NumPy while remaining easy to use.

---

🔹 Features

Core Data Structures

• Vector
  • Creation, deletion, printing
  • Basic operations: add, subtract, scalar multiply/divide
  • Dot product (optimized with loop unrolling and SIMD-ready)
  • Magnitude (Euclidean norm)
• Matrix
  • Creation, deletion, printing
  • Basic operations: add, subtract, scalar multiply/divide
  • Transpose
  • Matrix multiplication (naive and optimized)
  • Compatible with Python lists

Python Integration

• Full Python wrappers for all vectors and matrices
• Returns Python-accessible capsules for memory-safe operations
• Easy to call from Python ML code

Optimizations

• Loop unrolling for dot product
• SIMD-ready structure for future AVX/SSE optimization
• Minimal memory overhead, contiguous allocations
• Benchmarkable against NumPy for performance comparison

---

Benchmarking

Vector Dot Product

Matrix Multiplication

🔹 Installation

git clone https://github.com/yourusername/mini-ml.git
python3 setup.py build_ext --inplace