Example of using patch_NLM with a solver about proximal HOT 8 CLOSED

>>> prob.solve(verbose=True, max_iters=2000)
Estimated params [sigma = 1.000 | tau = 1.000 | theta = 1.000 | L_est = 1.0000]
Prox NLM params are: [sigma =nan prior=0 sigma_scale=6.0]
OpenCV Error: Assertion failed (almost_dist2weight_[0] == fixed_point_mult_) in FastNlMeansDenoisingInvoker, file -------src-dir-------/opencv-2.4.10/modules/photo/src/fast_nlmeans_denoising_invoker.hpp, line 150
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/github/ProxImaL/proximal/algorithms/problem.py", line 136, in solve
    *args, **kwargs)
  File "/github/ProxImaL/proximal/algorithms/pock_chambolle.py", line 128, in solve
    lin_solver=lin_solver, options=lin_solver_options).flatten()
  File "/github/ProxImaL/proximal/prox_fns/prox_fn.py", line 84, in prox
    xhat = self._prox(rho_hat, v, *args, **kwargs)
  File "/github/ProxImaL/proximal/prox_fns/patch_NLM.py", line 96, in _prox
    self.searchWindowSizeNLM)
cv2.error: -------src-dir-------/opencv-2.4.10/modules/photo/src/fast_nlmeans_denoising_invoker.hpp:150: error: (-215) almost_dist2weight_[0] == fixed_point_mult_ in function FastNlMeansDenoisingInvoker

import numpy as np
import odl
import proximal

# Create space, a square from [0, 0] to [100, 100] with (100 x 100) points
space = odl.uniform_discr([0, 0], [100, 100], [100, 100])

# Create odl operator for laplacian
laplacian = odl.Laplacian(space)

# Create right hand side
rhs = laplacian(odl.phantom.shepp_logan(space, modified=True))
rhs += odl.phantom.white_noise(space) * np.std(rhs) * 0.0  # disable noise
rhs.show('noisy data')

# Convert laplacian to cvx operator
norm_bound = odl.power_method_opnorm(laplacian, 10, rhs)
cvx_laplacian = odl.operator.oputils.as_cvx_operator(laplacian,
                                                     norm_bound=norm_bound)

# Convert to array
rhs_arr = rhs.asarray()

# Set up optimization problem
x = proximal.Variable(space.shape)
funcs = [10 * proximal.sum_squares(cvx_laplacian(x) - rhs_arr),
         proximal.norm1(proximal.grad(x))]
prob = proximal.Problem(funcs)

# Solve the problem
prob.solve(verbose=True, solver='pc', max_iters=1000)

# Convert back to odl and display result
result_odl = space.element(x.value)
result_odl.show('result')

Estimated params [sigma = 1.000 | tau = 1.000 | theta = 1.000 | L_est = 1.0000]
iter 1: ||r||_2 = nan, eps_pri = nan, ||s||_2 = nan, eps_dual = nan
iter 2: ||r||_2 = nan, eps_pri = nan, ||s||_2 = nan, eps_dual = nan
iter 3: ||r||_2 = nan, eps_pri = nan, ||s||_2 = nan, eps_dual = nan
iter 4: ||r||_2 = nan, eps_pri = nan, ||s||_2 = nan, eps_dual = nan
iter 5: ||r||_2 = nan, eps_pri = nan, ||s||_2 = nan, eps_dual = nan
iter 6: ||r||_2 = nan, eps_pri = nan, ||s||_2 = nan, eps_dual = nan

Estimated params [sigma = 1.000 | tau = 1.000 | theta = 1.000 | L_est = 1.0000]
iter 1: ||r||_2 = 10.380, eps_pri = 0.188, ||s||_2 = 0.030, eps_dual = 0.105
iter 2: ||r||_2 = 8.084, eps_pri = 0.188, ||s||_2 = 0.014, eps_dual = 0.105
iter 3: ||r||_2 = 7.428, eps_pri = 0.188, ||s||_2 = 0.010, eps_dual = 0.105
iter 4: ||r||_2 = 7.460, eps_pri = 0.188, ||s||_2 = 0.009, eps_dual = 0.104
iter 5: ||r||_2 = 7.519, eps_pri = 0.189, ||s||_2 = 0.009, eps_dual = 0.103
iter 6: ||r||_2 = 7.450, eps_pri = 0.189, ||s||_2 = 0.009, eps_dual = 0.103

Estimated params [sigma = 1.000 | tau = 1.000 | theta = 1.000 | L_est = 1.0000]
iter 1: ||r||_2 = 10.397, eps_pri = 0.188, ||s||_2 = 0.025, eps_dual = 0.105
iter 2: ||r||_2 = 8.072, eps_pri = 0.188, ||s||_2 = 0.013, eps_dual = 0.105
iter 3: ||r||_2 = 7.353, eps_pri = 0.188, ||s||_2 = 0.010, eps_dual = 0.105
iter 4: ||r||_2 = 7.365, eps_pri = 0.188, ||s||_2 = 0.009, eps_dual = 0.104
iter 5: ||r||_2 = 7.445, eps_pri = 0.189, ||s||_2 = 0.009, eps_dual = 0.103
iter 6: ||r||_2 = 7.397, eps_pri = 0.189, ||s||_2 = 0.009, eps_dual = 0.103