def output_Image(flag, Img, file, path):
if flag == 0:
fig1 = plot.figure()
plot.subplot(1, 1, 1)
plot.imview(util.tiledict(Img[0]),
title='Initial Dctionary(Layer1)',
fig=fig1)
fig1.savefig(path + '\\' + file + '(Layer1).png')
fig2 = plot.figure()
plot.subplot(1, 1, 1)
plot.imview(util.tiledict(Img[1]),
title='Initial Dctionary(Layer2)',
fig=fig2)
fig2.savefig(path + '\\' + file + '(Layer2).png')
elif flag == 1:
fig3 = plot.figure()
for i in range(len(Img)):
plot.subplot(3, 3, i + 1)
plot.imview(util.tiledict(Img[i]), title='', fig=fig3)
fig3.savefig(path + '\\' + file + '.png')
elif flag == 2:
fig4 = plot.figure()
for i in range(len(Img)):
plot.subplot(3, 3, i + 1)
plot.imview(util.tiledict(Img[i]), title='', fig=fig4)
fig4.savefig(path + '\\' + file + '.png')
elif flag == 3:
fig_t = plot.figure()
for i in range(9):
plot.subplot(1, 1, 1)
plot.imview(util.tiledict(Img), title='', fig=fig_t)
fig_t.savefig(path + '\\' + file + '.png')
def plot_solver_stats(pe):
"""Plot iteration statistics for a :class:`.cdl.PSFEstimator` solve.
Parameters
----------
pe : :class:`.cdl.PSFEstimator`
Solver object for which solver iteration stats are to be plotted
Returns
-------
:class:`matplotlib.figure.Figure` object
Figure object for this figure
"""
itsx = pe.slvX.getitstat()
itsg = pe.slvG.getitstat()
fig = plot.figure(figsize=(20, 5))
fig.suptitle('Solver iteration statistics')
plot.subplot(1, 3, 1)
N = min(len(itsx.ObjFun), len(itsg.DFid))
fnvl = np.vstack(
(itsx.ObjFun[-N:], itsg.ObjFun[-N:], itsx.DFid[-N:], itsg.DFid[-N:])).T
lgnd = ('X Func.', 'G Func,', 'X D. Fid.', 'G D. Fid.')
plot.plot(fnvl,
ptyp='semilogy',
xlbl='Iterations',
ylbl='Functional',
lgnd=lgnd,
fig=fig)
plot.subplot(1, 3, 2)
plot.plot(np.vstack((itsx.PrimalRsdl, itsx.DualRsdl)).T,
ptyp='semilogy',
xlbl='Iterations',
ylbl='X Residual',
lgnd=['Primal', 'Dual'],
fig=fig)
plot.subplot(1, 3, 3)
if hasattr(itsg, 'PrimalRsdl'):
plot.plot(np.vstack((itsg.PrimalRsdl, itsg.DualRsdl)).T,
ptyp='semilogy',
xlbl='Iterations',
ylbl='G Residual',
lgnd=['Primal', 'Dual'],
fig=fig)
else:
plot.plot(itsg.Rsdl,
ptyp='semilogy',
xlbl='Iterations',
ylbl='G Residual',
fig=fig)
fig.show()
return fig
b = tvl2.TVL2Denoise(imgn, lmbda, opt)
imgr = b.solve()
"""
Display solve time and denoising performance.
"""
print("TVL2Denoise solve time: %5.2f s" % b.timer.elapsed('solve'))
print("Noisy image PSNR: %5.2f dB" % metric.psnr(img, imgn))
print("Denoised image PSNR: %5.2f dB" % metric.psnr(img, imgr))
"""
Display reference, corrupted, and denoised images.
"""
fig = plot.figure(figsize=(20, 5))
plot.subplot(1, 3, 1)
plot.imview(img, fgrf=fig, title='Reference')
plot.subplot(1, 3, 2)
plot.imview(imgn, fgrf=fig, title='Corrupted')
plot.subplot(1, 3, 3)
plot.imview(imgr, fgrf=fig, title=r'Restored ($\ell_2$-TV)')
fig.show()
"""
Get iterations statistics from solver object and plot functional value, ADMM primary and dual residuals, and automatically adjusted ADMM penalty parameter against the iteration number.
"""
its = b.getitstat()
fig = plot.figure(figsize=(20, 5))
plot.subplot(1, 3, 1)
plot.plot(its.ObjFun, fgrf=fig, xlbl='Iterations', ylbl='Functional')
plot.subplot(1, 3, 2)
"""
Plot comparison of reference and recovered representations.
"""
plot.plot(np.hstack((x0, x)), title='Sparse representation',
lgnd=['Reference', 'Reconstructed'])
"""
Plot lmbda error curve, functional value, residuals, and rho
"""
its = b.getitstat()
fig = plot.figure(figsize=(15, 10))
plot.subplot(2, 2, 1)
plot.plot(fvmx, x=lrng, ptyp='semilogx', xlbl='$\lambda$',
ylbl='Error', fig=fig)
plot.subplot(2, 2, 2)
plot.plot(its.ObjFun, xlbl='Iterations', ylbl='Functional', fig=fig)
plot.subplot(2, 2, 3)
plot.plot(np.vstack((its.PrimalRsdl, its.DualRsdl)).T,
ptyp='semilogy', xlbl='Iterations', ylbl='Residual',
lgnd=['Primal', 'Dual'], fig=fig)
plot.subplot(2, 2, 4)
plot.plot(its.Rho, xlbl='Iterations', ylbl='Penalty Parameter', fig=fig)
fig.show()
# Wait for enter on keyboard
for it in range(iter):
img_index = np.random.randint(0, sh.shape[-1])
d.solve(sh[..., [img_index]])
d.display_end()
D1 = d.getdict()
print("OnlineConvBPDNDictLearn solve time: %.2fs" % d.timer.elapsed('solve'))
"""
Display initial and final dictionaries.
"""
D1 = D1.squeeze()
fig = plot.figure(figsize=(14, 7))
plot.subplot(1, 2, 1)
plot.imview(util.tiledict(D0), title='D0', fig=fig)
plot.subplot(1, 2, 2)
plot.imview(util.tiledict(D1), title='D1', fig=fig)
fig.show()
"""
Get iterations statistics from solver object and plot functional value.
"""
its = d.getitstat()
fig = plot.figure(figsize=(7, 7))
plot.plot(np.vstack((its.DeltaD, its.Eta)).T, xlbl='Iterations',
lgnd=('Delta D', 'Eta'), fig=fig)
fig.show()
'rho': 50.0 * lmbda + 0.5
},
'CCMOD': {
'ZeroMean': True
}
})
Wr = np.reshape(W, W.shape[0:3] + (W.shape[3], 1))
d2 = cbpdndl.ConvBPDNMaskDcplDictLearn(D0, shpw, lmbda, Wr, opt2)
D2 = d2.solve()
# Reconstruct from ConvBPDNMaskDcplDictLearn solution
sr2 = d2.reconstruct()[:-Npr, :-Npc].squeeze() + sl
# Compare dictionaries
fig1 = plot.figure(1, figsize=(14, 7))
plot.subplot(1, 2, 1)
plot.imview(util.tiledict(D1.squeeze()),
fgrf=fig1,
title='Without Mask Decoupling')
plot.subplot(1, 2, 2)
plot.imview(util.tiledict(D2.squeeze()),
fgrf=fig1,
title='With Mask Decoupling')
fig1.show()
# Display reference and test images (with unmasked lowpass component)
fig2 = plot.figure(2, figsize=(14, 14))
plot.subplot(2, 2, 1)
plot.imview(S[..., 0], fgrf=fig2, title='Reference')
plot.subplot(2, 2, 2)
plot.imview(shpw[:-Npr, :-Npc, :, 0] + sl[..., 0], fgrf=fig2, title='Test')
"""
Reconstruct image from sparse representation.
"""
shr = b.reconstruct().squeeze()
imgr = sl + shr
print("Reconstruction PSNR: %.2fdB\n" % sm.psnr(img, imgr))
"""
Display low pass component and sum of absolute values of coefficient maps of highpass component.
"""
fig = plot.figure(figsize=(14, 7))
plot.subplot(1, 2, 1)
plot.imview(sl, title='Lowpass component', fig=fig)
plot.subplot(1, 2, 2)
plot.imview(np.sum(abs(X), axis=b.cri.axisM).squeeze(), cmap=plot.cm.Blues,
title='Sparse representation', fig=fig)
fig.show()
"""
Display original and reconstructed images.
"""
fig = plot.figure(figsize=(14, 7))
plot.subplot(1, 2, 1)
plot.imview(img, title='Original', fig=fig)
plot.subplot(1, 2, 2)
a = cbpdn.ConvBPDN(D, pad(imgnh), lmbda, opt1, dimK=0)
b = cbpdntv.ConvBPDNScalarTV(D, pad(imgnh), lmbda, mu, opt2)
c = cbpdntv.ConvBPDNVectorTV(D, pad(imgnh), lmbda, mu, opt3)
X1 = a.solve()
X2 = b.solve()
X3 = c.solve()
imgdp1 = a.reconstruct().squeeze()
imgd1 = np.clip(crop(imgdp1) + imgnl, 0, 1)
imgdp2 = b.reconstruct().squeeze()
imgd2 = np.clip(crop(imgdp2) + imgnl, 0, 1)
imgdp3 = c.reconstruct().squeeze()
imgd3 = np.clip(crop(imgdp3) + imgnl, 0, 1)
print("ConvBPDN solve time: %5.2f s" % b.timer.elapsed('solve'))
print("Noisy image PSNR: %5.2f dB" % sm.psnr(img, imgn))
print("bpdn PSNR: %5.2f dB" % sm.psnr(img, imgd1))
print("scalar image PSNR: %5.2f dB" % sm.psnr(img, imgd2))
print("vector image PSNR: %5.2f dB" % sm.psnr(img, imgd3))
fig = plot.figure(figsize=(21, 21))
plot.subplot(2, 3, 1)
plot.imview(img, title='Reference', fig=fig)
plot.subplot(2, 3, 2)
plot.imview(imgn, title='Noisy', fig=fig)
plot.subplot(2, 3, 3)
plot.imview(imgd1, title='Conv bpdn', fig=fig)
plot.subplot(2, 3, 4)
plot.imview(imgd2, title='scalar TV', fig=fig)
plot.subplot(2, 3, 5)
plot.imview(imgd3, title='vector TV', fig=fig)
fig.show()
dimK=0)
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))
"""
Display low pass component and sum of absolute values of coefficient maps of highpass component.
"""
fig = plot.figure(figsize=(14, 7))
plot.subplot(1, 2, 1)
plot.imview(sl, title='Lowpass component', fig=fig)
plot.subplot(1, 2, 2)
plot.imview(np.sum(abs(X), axis=b.cri.axisM).squeeze(),
cmap=plot.cm.Blues,
title='Sparse representation',
fig=fig)
fig.show()
"""
Show activation of grouped elements column-wise for first four groups of both schemes. As mu is lowered, the vertical pairs should look more and more similar. You will likely need to zoom in to see the activations clearly. In general, you should observe that the second-scheme pairs should have more similar-looking activations than first-scheme pairs, proportional to mu of course.
"""
fig = plot.figure(figsize=(14, 10))
for i in range(4):
plot.subplot(3, 4, i + 1)
plot.imview(abs(X[:, :, :, :, i]).squeeze(),
"""
Report performances of different methods of solving the same problem.
"""
print("Corrupted image PSNR: %5.2f dB" % metric.psnr(img, imgw))
print("Serial Reconstruction PSNR: %5.2f dB" % metric.psnr(img, imgr))
print("Parallel Reconstruction PSNR: %5.2f dB\n" % metric.psnr(img, imgr_par))
"""
Display reference, test, and reconstructed images
"""
fig = plot.figure(figsize=(14, 14))
plot.subplot(2, 2, 1)
plot.imview(img, fig=fig, title='Reference Image')
plot.subplot(2, 2, 2)
plot.imview(imgw, fig=fig, title=('Corrupted Image PSNR: %5.2f dB' %
metric.psnr(img, imgw)))
plot.subplot(2, 2, 3)
plot.imview(imgr, fig=fig, title=('Serial reconstruction PSNR: %5.2f dB' %
metric.psnr(img, imgr)))
plot.subplot(2, 2, 4)
plot.imview(imgr_par, fig=fig, title=('Parallel reconstruction PSNR: %5.2f dB' %
metric.psnr(img, imgr_par)))
fig.show()
"""
Display lowpass component and sparse representation
dimK=0)
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))
"""
Display low pass component and sum of absolute values of coefficient maps of highpass component.
"""
fig = plot.figure(figsize=(14, 7))
plot.subplot(1, 2, 1)
plot.imview(sl, title='Lowpass component', fig=fig)
plot.subplot(1, 2, 2)
plot.imview(np.sum(abs(X), axis=b.cri.axisM).squeeze(),
cmap=plot.cm.Blues,
title='Sparse representation',
fig=fig)
fig.show()
"""
Show activation of grouped elements column-wise for first four groups. As mu is lowered, the vertical pairs should look more and more similar. You will likely need to zoom in to see the activations clearly.
"""
fig = plot.figure(figsize=(14, 7))
for i in range(4):
plot.subplot(2, 4, i + 1)
plot.imview(abs(X[:, :, :, :, i]).squeeze(),
Load example piano piece.
"""
audio, fs = librosa.load(os.path.join('data',
'MAPS_MUS-deb_clai_ENSTDkCl_excerpt.wav'), 16000)
"""
Load dictionary and display it. Grouped elements are displayed on the same axis.
"""
M = 4
D = np.load(os.path.join('data', 'pianodict.npz'))['elems']
fig = plot.figure(figsize=(10, 16))
for i in range(24):
plot.subplot(8, 3, i + 1, title=f'{librosa.midi_to_note(i+60)}',
ylim=[-6, 6])
plot.gca().get_xaxis().set_visible(False)
plot.gca().get_yaxis().set_visible(False)
plot.gca().set_frame_on(False)
for k in range(M):
plot.plot(D[M*(i+1)-k-1], fig=fig)
fig.show()
"""
Set :class:`.admm.cbpdn.ConvBPDNInhib` solver options.
"""
lmbda = 1e-1
mu = 1e1
gamma = 1e0
