nreg = len(names)
ROI_tc = mroi.get_roi_feature('signal')
glm.fit(ROI_tc.T, X, method=method, model=model)
mp.figure()
mp.subplot(1, 2, 1)
b1 = mp.bar(np.arange(nreg-1), glm.beta[:-1,0], width=.4, color='blue',
label='r1')
b2 = mp.bar(np.arange(nreg-1)+0.3, glm.beta[:-1,1], width=.4, color='red',
label='r2')
mp.xticks(np.arange(nreg-1), names[:-1])
mp.legend()
mp.title('parameters estimates for the roi time courses')
bx = mp.subplot(1, 2 ,2)
mroi.plot_discrete_feature('contrast', bx)
########################################
# fitted and adjusted response
########################################
res = ROI_tc -np.dot(glm.beta.T, X.T)
proj = np.eye(nreg)
proj[2:] = 0
fit = np.dot(np.dot(glm.beta.T,proj),X.T)
# plot it
mp.figure()
for k in range(mroi.k):
mp.subplot(mroi.k, 1, k+1)
