diff --git a/doc/plot/plot_ISV.py b/doc/plot/plot_ISV.py index 91b9ff61fdee5cc709c3797f3d5d753ddaf6780d..fc3f9c173916b7e545669ee30b014cfcf626435e 100644 --- a/doc/plot/plot_ISV.py +++ b/doc/plot/plot_ISV.py @@ -57,6 +57,10 @@ ubm = bob.learn.em.GMMMachine(n_gaussians).fit(X) gmm_stats = [ubm.acc_statistics(x[np.newaxis]) for x in X] isv_machine = bob.learn.em.ISVMachine(ubm, r_U).fit(gmm_stats, y) +# gmm_stats = [ubm.acc_statistics(x) for x in [setosa, versicolor, virginica]] +# isv_machine = bob.learn.em.ISVMachine(ubm, r_U).fit(gmm_stats, [0, 1, 2]) + + # isvbase = isv_train([setosa, versicolor, virginica], ubm) # Variability direction diff --git a/doc/plot/plot_JFA.py b/doc/plot/plot_JFA.py index f77e6779c9f6f9b52ac78e72d76b498b8ec61d92..cae3e317df04b45f0f2890c96046194988bbc003 100644 --- a/doc/plot/plot_JFA.py +++ b/doc/plot/plot_JFA.py @@ -58,6 +58,9 @@ ubm = bob.learn.em.GMMMachine(n_gaussians).fit(X) gmm_stats = [ubm.acc_statistics(x[np.newaxis]) for x in X] jfa_machine = bob.learn.em.JFAMachine(ubm, r_U, r_V).fit(gmm_stats, y) +# gmm_stats = [ubm.acc_statistics(x) for x in [setosa, versicolor, virginica]] +# jfa_machine = bob.learn.em.JFAMachine(ubm, r_U, r_V).fit(gmm_stats, [0, 1, 2]) + # Variability direction U u0 = jfa_machine.U[0:2, 0] / np.linalg.norm(jfa_machine.U[0:2, 0])