PyHDC

PyHDC is a Python library for Hyperdimensional Computing (HDC) and Vector Symbolic Architectures (VSA). It provides a clean, composable API for building and querying hypervectors, which are high-dimensional representations that encode meaning as geometry.

import pyhdc

enc = pyhdc.MAP_C(dimension=10_000)
color = enc.generate()   # a random 10,000-D hypervector
shape = enc.generate()
obj   = color.bind(shape)          # association: "colored shape"
print(obj.unbind(shape).similarity(color))   # ~= 1.0

PyHDC supports 14 encoding schemes (MAP, HRR, FHRR, BSC, BSDC, VTB, MBAT, etc.), 7 random generator families for reproducible experiments, and a dual NumPy / PyTorch backend with GPU support.

---