search

ashvardanian / SimSIMD

Up to 200x Faster Inner Products and Vector Similarity — for Python, JavaScript, Rust, C, and Swift, supporting f64, f32, f16 real & complex, i8, and binary vectors using SIMD for both x86 AVX2 & AVX-512 and Arm NEON & SVE 📐
https://ashvardanian.com/posts/simsimd-faster-scipy/
Apache License 2.0
797 stars 42 forks source link

Complex vectors for Fourier representations, Image processing, DSP, and Quantum states #73

Closed ashvardanian closed 3 months ago

ashvardanian commented 5 months ago

Vectors of complex numbers are common in different scientific applications, and aren't fully supported by NumPy, which only handles double-precision variants.

ashvardanian commented 4 months ago

The implementation of complex inner products is ready and will be shipped in the next major release, as it breaks the API, introducing new interfaces. Optimized Haversine distances may be included there as well.