0

I have some code that is beautifully expressed using vectors of 256-bit unsigned integers.

Currently I use normal Python to loop over the vector, but I would like to speed this up.

TensorFlow, NumPy and SciPy does not seem to support uint256.

Is there a numerical package in which I can use uint256 in vectors? Does this package support arbitrarily long uints?

Ole Tange
  • 31,768
  • 5
  • 86
  • 104
  • Apparently `sympy` might contain something. But my guess is that for non-standard integer types, you have to roll your own code (or use a C++ library) if you want anything performant. – Mateen Ulhaq Sep 14 '19 at 06:04
  • 1
    Actually, with `numpy`, you can use Python integers (which have no bound on size): https://stackoverflow.com/a/37272717/365102. No idea how fast that is. – Mateen Ulhaq Sep 14 '19 at 06:06

0 Answers0