def test_euclidean_10000(): # test for predict using euclidean distance m1 = APM(n_macrostates=2, metric='euclidean', lag_time=10) m2 = APM(n_macrostates=2, metric='euclidean', lag_time=10) data = rs.randn(10000, 2) labels1 = m1.fit_predict([data]) labels2 = m2.fit([data]).MacroAssignments_ eq(labels1[0], labels2[0])
def test_rmsd(): # test for predict using euclidean distance m1 = APM(n_macrostates=4, metric='rmsd', lag_time=1) m2 = APM(n_macrostates=4, metric='rmsd', lag_time=1) labels1 = m1.fit_predict([trj]) labels2 = m2.fit([trj]).MacroAssignments_ eq(labels1[0], labels2[0])
def test_euclidean_10000(): # test for predict using euclidean distance m1 = APM(n_macrostates=2, metric='euclidean', lag_time=10) m2 = APM(n_macrostates=2, metric='euclidean', lag_time=10) data = np.random.randn(10000, 2) labels1 = m1.fit_predict([data]) labels2 = m2.fit([data]).MacroAssignments_ eq(labels1[0], labels2[0])
def test_rmsd(): # test for predict using rmsd m1 = APM(n_macrostates=4, metric='rmsd', lag_time=1, random_state=rs) m2 = APM(n_macrostates=4, metric='rmsd', lag_time=1, random_state=rs) labels1 = m1.fit_predict([trj]) labels2 = m2.fit([trj]).MacroAssignments_ eq(labels1[0], labels2[0])
def test_euclidean(): # test for predict using euclidean distance data = rs.randn(100, 2) m1 = APM(n_macrostates=2, metric='euclidean', lag_time=1, random_state=rs) m2 = APM(n_macrostates=2, metric='euclidean', lag_time=1, random_state=rs) labels1 = m1.fit_predict([data]) labels2 = m2.fit([data]).MacroAssignments_ eq(labels1[0], labels2[0])
def test_shapes(): # make sure all the shapes are correct of the fit parameters m = APM(n_macrostates=3, metric='euclidean', lag_time=1) m.fit([rs.randn(100, 2)]) assert isinstance(m.labels_, list) eq(m.labels_[0].shape, (100, ))
def test_shapes(): # make sure all the shapes are correct of the fit parameters m = APM(n_macrostates=3, metric='euclidean', lag_time=1) m.fit([np.random.randn(100, 2)]) assert isinstance(m.labels_, list) eq(m.labels_[0].shape, (100,))