def test_permutation_test(): """Test one way permutation test""" covset = generate_cov(10, 30) labels = np.array([0, 1]).repeat(5) # base p = PermutationTest(10) p.test(covset, labels) # fit perm p = PermutationTest(10, fit_perms=True) p.test(covset, labels) # unique perms p = PermutationTest(1000) p.test(covset, labels) p.summary() p.plot(nbins=2)
def test_permutation_test(): """Test one way permutation test""" covset = generate_cov(10, 30) labels = np.array([0, 1]).repeat(5) p = PermutationTest(10) p.test(covset, labels) p.summary()
covest = Covariances() Fs = 160 window = 2 * Fs Nwindow = 20 Ns = epochs_data.shape[2] step = int((Ns - window) / Nwindow) time_bins = range(0, Ns - window, step) pv = [] Fv = [] # For each frequency bin, estimate the stats for t in time_bins: covmats = covest.fit_transform(epochs_data[:, ::1, t:(t + window)]) p_test = PermutationTest(5000) p, F = p_test.test(covmats, labels) print p_test.summary() pv.append(p) Fv.append(F[0]) time = np.array(time_bins) / float(Fs) + tmin plot(time, Fv, lw=2) plt.xlabel('Time') plt.ylabel('F-value') significant = np.array(pv) < 0.001 plot(time, significant, 'r', lw=2) plt.legend(['F-value', 'p<0.001']) plt.grid() plt.show()