def test_11(self):
N = 63
M = 4
Nd = 8
D = np.random.randn(Nd, Nd, M)
X0 = np.zeros((N, N, M))
xr = np.random.randn(N, N, M)
xp = np.abs(xr) > 3
X0[xp] = np.random.randn(X0[xp].size)
S = np.sum(ifftn(
fftn(D, (N, N), (0, 1)) * fftn(X0, None, (0, 1)), None,
(0, 1)).real,
axis=2)
lmbda = 1e-2
L = 1e3
opt = cbpdn.ConvBPDN.Options({
'Verbose': False,
'MaxMainIter': 2000,
'RelStopTol': 1e-9,
'L': L,
'Backtrack': BacktrackStandard()
})
b = cbpdn.ConvBPDN(D, S, lmbda, opt)
b.solve()
X1 = b.X.squeeze()
assert sl.rrs(X0, X1) < 5e-4
Sr = b.reconstruct().squeeze()
assert sl.rrs(S, Sr) < 2e-4
def test_13(self):
N = 64
M = 4
Nd = 8
D0 = cr.normalise(cr.zeromean(np.random.randn(Nd, Nd, M), (Nd, Nd, M),
dimN=2),
dimN=2)
X = np.zeros((N, N, M))
xr = np.random.randn(N, N, M)
xp = np.abs(xr) > 3
X[xp] = np.random.randn(X[xp].size)
S = np.sum(ifftn(
fftn(D0, (N, N), (0, 1)) * fftn(X, None, (0, 1)), None,
(0, 1)).real,
axis=2)
L = 0.5
opt = ccmod.ConvCnstrMOD.Options({
'Verbose': False,
'MaxMainIter': 3000,
'ZeroMean': True,
'RelStopTol': 0.,
'L': L,
'Backtrack': BacktrackStandard()
})
Xr = X.reshape(X.shape[0:2] + (
1,
1,
) + X.shape[2:])
Sr = S.reshape(S.shape + (1, ))
c = ccmod.ConvCnstrMOD(Xr, Sr, D0.shape, opt)
c.solve()
D1 = cr.bcrop(c.X, D0.shape).squeeze()
assert rrs(D0, D1) < 1e-4
assert np.array(c.getitstat().Rsdl)[-1] < 1e-5
def test_08(self):
N = 8
M = 16
D = np.random.randn(N, M)
s = np.random.randn(N, 1)
dt = np.float64
opt = bpdn.BPDN.Options({
'Verbose': False,
'MaxMainIter': 20,
'Backtrack': BacktrackStandard(gamma_u=1.1),
'DataType': dt
})
b = bpdn.BPDN(D, s, lmbda=1.0, opt=opt)
b.solve()
assert b.X.dtype == dt
assert b.Y.dtype == dt
def test_05(self):
N = 8
M = 8
D = np.random.randn(N, M)
s = np.random.randn(N, 1)
try:
opt = bpdn.BPDN.Options({
'FastSolve': True,
'Verbose': False,
'MaxMainIter': 10,
'Backtrack': BacktrackStandard()
})
b = bpdn.BPDN(D, s, lmbda=1.0, opt=opt)
b.solve()
except Exception as e:
print(e)
assert 0
def test_03(self):
N = 8
M = 16
D = np.random.randn(N, M)
s = np.random.randn(N, 1)
try:
opt = bpdn.BPDN.Options({
'Verbose': False,
'MaxMainIter': 80,
'Backtrack': BacktrackStandard(),
'RelStopTol': 1e-4
})
b = bpdn.BPDN(D, s, lmbda=1.0, opt=opt)
b.solve()
except Exception as e:
print(e)
assert 0
assert np.array(b.getitstat().Rsdl).min() < 2e-4
def test_05(self):
N = 16
M = 4
Nd = 8
X = np.random.randn(N, N, 1, 1, M)
S = np.random.randn(N, N, 1)
dt = np.float64
opt = ccmod.ConvCnstrMOD.Options({
'Verbose': False,
'MaxMainIter': 20,
'Backtrack': BacktrackStandard(),
'DataType': dt
})
c = ccmod.ConvCnstrMOD(X, S, (Nd, Nd, M), opt=opt)
c.solve()
assert c.X.dtype == dt
assert c.Xf.dtype == complex_dtype(dt)
assert c.Yf.dtype == complex_dtype(dt)
def test_06(self):
N = 16
Nd = 5
K = 2
M = 4
D = np.random.randn(Nd, Nd, M)
s = np.random.randn(N, N, K)
dt = np.float32
opt = cbpdn.ConvBPDN.Options({
'Verbose': False,
'MaxMainIter': 20,
'Backtrack': BacktrackStandard(),
'DataType': dt
})
lmbda = 1e-1
b = cbpdn.ConvBPDN(D, s, lmbda, opt=opt)
b.solve()
assert b.X.dtype == dt
assert b.Xf.dtype == fft.complex_dtype(dt)
assert b.Yf.dtype == fft.complex_dtype(dt)
"""
D = util.convdicts()['RGB:8x8x3x64']
plot.imview(util.tiledict(D), fgsz=(7, 7))
"""
Set :class:`.pgm.cbpdn.ConvBPDN` solver options. Note the possibility of changing parameters in the backtracking algorithm.
"""
lmbda = 1e-1
L = 1e1
opt = cbpdn.ConvBPDN.Options({
'Verbose': True,
'MaxMainIter': 250,
'RelStopTol': 8e-5,
'L': L,
'Backtrack': BacktrackStandard(maxiter=15)
})
"""
Initialise and run CSC solver.
"""
b = cbpdn.ConvBPDN(D, sh, lmbda, opt)
X = b.solve()
print("ConvBPDN solve time: %.2fs" % b.timer.elapsed('solve'))
"""
Reconstruct image from sparse representation.
"""
shr = b.reconstruct().squeeze()
imgr = sl + shr
print("Reconstruction PSNR: %.2fdB\n" % sm.psnr(img, imgr))
si = np.random.permutation(list(range(0, M - 1)))
x0[si[0:L]] = np.random.randn(L, 1)
# Construct reference and noisy signal
s0 = D.dot(x0)
s = s0 + sigma * np.random.randn(N, 1)
"""
Set regularisation parameter and options for BPDN solver with standard PGM backtracking.
"""
lmbda = 2.98e1
L_sc = 9.e2
opt = bpdn.BPDN.Options({
'Verbose': True,
'MaxMainIter': 50,
'Backtrack': BacktrackStandard(),
'L': L_sc
})
"""
Initialise and run BPDN object
"""
b1 = bpdn.BPDN(D, s, lmbda, opt)
x1 = b1.solve()
print("BPDN standard PGM backtracking solve time: %.2fs" %
b1.timer.elapsed('solve'))
"""
Set options for BPDN solver with robust PGM backtracking.
"""
D0 = np.random.randn(16, 16, 3, 96)
"""
Set regularization parameter and options for dictionary learning solver. Note the multi-scale dictionary filter sizes. Also note the possibility of changing parameters in the backtracking algorithm.
"""
lmbda = 0.2
L_sc = 36.0
L_du = 50.0
dsz = ((8, 8, 3, 32), (12, 12, 3, 32), (16, 16, 3, 32))
opt = cbpdndl.ConvBPDNDictLearn.Options(
{
'Verbose': True,
'MaxMainIter': 200,
'DictSize': dsz,
'CBPDN': {
'Backtrack': BacktrackStandard(gamma_u=1.1),
'L': L_sc
},
'CCMOD': {
'Backtrack': BacktrackStandard(),
'L': L_du
}
},
xmethod='pgm',
dmethod='pgm')
"""
Create solver object and solve.
"""
d = cbpdndl.ConvBPDNDictLearn(D0, sh, lmbda, opt, xmethod='pgm', dmethod='pgm')
D1 = d.solve()
