We will move the discussion of binary hypervector implementation to this issues such that #25 can focus on the complex representation and varying bit-widths.
hv = torchhd.random_hv(10, 1000, dtype=torch.bool) # creates a boolean tensor
torchhd.bind(hv[0], hv[1]) # performs XOR
torchhd.bundle(hv[0], hv[1], hv[2]) # needs three inputs to calculated the majority function
We will move the discussion of binary hypervector implementation to this issues such that #25 can focus on the complex representation and varying bit-widths.