def test_gradient_descent(self):
A = random_spd(5)
b = np.random.randn(5)
expected = Cholesky(A).solve(b)
actual = GradientDescent(1000000).solve(A, b)
self.assertTrue(np.allclose(expected, actual))
def test_conjugate_gradient(self):
max_iters = 100_000
A = random_spd(5)
b = np.random.randn(5)
expected = Cholesky(A).solve(b)
actual = ConjugateGradient(max_iters).solve(A, b)
self.assertTrue(np.allclose(expected, actual))
"""QR algorithm vs projected iteration.
"""
import time
import numpy as np
import numpy.linalg as LA
from linalg.eigen import multi
from linalg import utils
if __name__ == "__main__":
M = utils.random_spd(10)
tic = time.time()
eigvals, eigvecs = multi.qr_algorithm(M, hess=False)
toc = time.time()
time_qr = toc - tic
print("QR: {}s".format(time_qr))
tic = time.time()
eigvals, eigvecs = multi.projected_iteration(M, len(M))
toc = time.time()
time_proj = toc - tic
print("Projected Iteration: {}s".format(time_proj))