Speedups using numPy and PyTorch

wmkirby1 / schur-transform

Computes an efficient quantum algorithm for the quantum Schur transform on n qubits.

3 stars 3 forks source link

Speedups using numPy and PyTorch #5

Closed erabinov closed 7 months ago

erabinov commented 10 months ago

As discussed, here are some suggested speedups for the algorithms in this repository.

All operators are now represented as numPy arrays. This should be backwards-compatible with any code that uses the old version, as long as that code uses only indexing operations on Python lists.

Most of the speedups occur because of numPy's more efficient implementations of array methods. Matrices are internally converted into sparse formats when this helps efficiency of matrix operations.

wmkirby1 commented 7 months ago

Just reviewed, looks great, and I'm sorry this took me so long to get to! Thanks Eugene!

erabinov commented 7 months ago

Glad to have been of help!