def test_nodeargs():
skip_if_no_external('mdp', min_version='2.4')
ds = normal_feature_dataset(perlabel=10, nlabels=2, nfeatures=4)
for svd_val in [True, False]:
pcm = PCAMapper(alg='PCA', svd=svd_val)
assert_equal(pcm.node.svd, svd_val)
pcm.train(ds)
assert_equal(pcm.node.svd, svd_val)
for output_dim in [0.5, 0.95, 0.99, 10, 50, 100]:
pcm = PCAMapper(alg='PCA', output_dim=output_dim)
for i in range(2): # so we also test on trained one
if isinstance(output_dim, float):
assert_equal(pcm.node.desired_variance, output_dim)
else:
assert_equal(pcm.node.output_dim, output_dim)
pcm.train(ds)
if isinstance(output_dim, float):
assert_not_equal(pcm.node.output_dim, output_dim)
# some dimensions are chosen
assert_true(pcm.node.output_dim > 0)
def test_pcamapper():
# data: 40 sample feature line in 20d space (40x20; samples x features)
ndlin = Dataset(np.concatenate([np.arange(40)
for i in range(20)]).reshape(20,-1).T)
pm = PCAMapper()
# train PCA
assert_raises(mdp.NodeException, pm.train, ndlin)
ndlin.samples = ndlin.samples.astype('float')
ndlin_noise = ndlin.copy()
ndlin_noise.samples += np.random.random(size=ndlin.samples.shape)
# we have no variance for more than one PCA component, hence just one
# actual non-zero eigenvalue
assert_raises(mdp.NodeException, pm.train, ndlin)
pm.train(ndlin_noise)
assert_equal(pm.proj.shape, (20, 20))
# now project data into PCA space
p = pm.forward(ndlin.samples)
assert_equal(p.shape, (40, 20))
# check that the mapped data can be fully recovered by 'reverse()'
assert_array_almost_equal(pm.reverse(p), ndlin)
