plt.imshow(gam_gaus_pp[..., 0], cmap=plt.cm.hot)
plt.title('Gamma-Gaussian mixture,\n first component posterior proba.')
plt.colorbar()
plt.subplot(3, 3, 5)
plt.imshow(gam_gaus_pp[..., 1], cmap=plt.cm.hot)
plt.title('Gamma-Gaussian mixture,\n second component posterior proba.')
plt.colorbar()
plt.subplot(3, 3, 6)
plt.imshow(gam_gaus_pp[..., 2], cmap=plt.cm.hot)
plt.title('Gamma-Gaussian mixture,\n third component posterior proba.')
plt.colorbar()
###############################################################################
# fit Beta's histogram with a mixture of Gaussians
alpha = 0.01
gaus_mix_pp = en.three_classes_GMM_fit(Beta, None, alpha, prior_strength=100)
gaus_mix_pp = np.reshape(gaus_mix_pp, (shape[0], shape[1], 3))
plt.figure(fig.number)
plt.subplot(3, 3, 7)
plt.imshow(gaus_mix_pp[..., 0], cmap=plt.cm.hot)
plt.title('Gaussian mixture,\n first component posterior proba.')
plt.colorbar()
plt.subplot(3, 3, 8)
plt.imshow(gaus_mix_pp[..., 1], cmap=plt.cm.hot)
plt.title('Gaussian mixture,\n second component posterior proba.')
plt.colorbar()
plt.subplot(3, 3, 9)
plt.imshow(gaus_mix_pp[..., 2], cmap=plt.cm.hot)
plt.title('Gamma-Gaussian mixture,\n third component posterior proba.')
# read the functional image
rbeta = load(input_image)
beta = rbeta.get_data()
beta = beta[mask > 0]
mf = plt.figure(figsize=(13, 5))
a1 = plt.subplot(1, 3, 1)
a2 = plt.subplot(1, 3, 2)
a3 = plt.subplot(1, 3, 3)
# fit beta's histogram with a Gamma-Gaussian mixture
bfm = np.array([2.5, 3.0, 3.5, 4.0, 4.5])
bfp = en.gamma_gaussian_fit(beta, bfm, verbose=1, mpaxes=a1)
# fit beta's histogram with a mixture of Gaussians
alpha = 0.01
pstrength = 100
bfq = en.three_classes_GMM_fit(beta, bfm, alpha, pstrength,
verbose=1, mpaxes=a2)
# fit the null mode of beta with the robust method
efdr = en.NormalEmpiricalNull(beta)
efdr.learn()
efdr.plot(bar=0, mpaxes=a3)
a1.set_title('Fit of the density with \n a Gamma-Gaussian mixture')
a2.set_title('Fit of the density with \n a mixture of Gaussians')
a3.set_title('Robust fit of the density \n with a single Gaussian')
plt.show()
plt.imshow(gam_gaus_pp[..., 0], cmap=plt.cm.hot)
plt.title('Gamma-Gaussian mixture,\n first component posterior proba.')
plt.colorbar()
plt.subplot(3, 3, 5)
plt.imshow(gam_gaus_pp[..., 1], cmap=plt.cm.hot)
plt.title('Gamma-Gaussian mixture,\n second component posterior proba.')
plt.colorbar()
plt.subplot(3, 3, 6)
plt.imshow(gam_gaus_pp[..., 2], cmap=plt.cm.hot)
plt.title('Gamma-Gaussian mixture,\n third component posterior proba.')
plt.colorbar()
###############################################################################
# fit Beta's histogram with a mixture of Gaussians
alpha = 0.01
gaus_mix_pp = en.three_classes_GMM_fit(Beta, None, alpha, prior_strength=100)
gaus_mix_pp = np.reshape(gaus_mix_pp, (shape[0], shape[1], 3))
plt.figure(fig.number)
plt.subplot(3, 3, 7)
plt.imshow(gaus_mix_pp[..., 0], cmap=plt.cm.hot)
plt.title('Gaussian mixture,\n first component posterior proba.')
plt.colorbar()
plt.subplot(3, 3, 8)
plt.imshow(gaus_mix_pp[..., 1], cmap=plt.cm.hot)
plt.title('Gaussian mixture,\n second component posterior proba.')
plt.colorbar()
plt.subplot(3, 3, 9)
plt.imshow(gaus_mix_pp[..., 2], cmap=plt.cm.hot)
plt.title('Gamma-Gaussian mixture,\n third component posterior proba.')
