Python module for Linear Algebra computations using CUDA. The module relies on Cython and NumPy/C API.
Converting a NumPy array to a GPU-array:
import numpy as np, mathmodule as mm
numpy_array = np.random.random(5)
gpu_array = mm.PyVector(numpy_array)
And GPU-array to NumPy array:
numpy_array = gpu_array.toNumPy()