def test_construct_from_graph(self):
"""
Check viability and correctness of LSTs constructed from a similarity
graph.
"""
tree = dcl.construct_tree_from_graph(self.knn_graph, self.density, prune_threshold=self.gamma)
self._check_tree_viability(tree)
self._check_tree_correctness(tree)
def test_construct_from_graph(self):
"""
Check viability and correctness of LSTs constructed from a similarity
graph.
"""
tree = dcl.construct_tree_from_graph(self.knn_graph,
self.density,
prune_threshold=self.gamma)
self._check_tree_viability(tree)
self._check_tree_correctness(tree)
## Construct a similarity graph and density estimate by hand. For this demo we
# use the k-nearest neighbors similarity graph and density estimator that
# DeBaCl uses by default, but this is the place to plug in custom similarity
# graphs and density estimators. In particular, this functionality means the
# user is not limited to tabular data or pre-set distance functions. If the
# user can make a similarity graph and estimate a density (or pseudo-density),
# then a level set tree can be estimated with DeBaCl.
k = int(0.05 * n)
knn_graph, radii = dcl.utils.knn_graph(X, k, method='kd-tree')
density = dcl.utils.knn_density(radii, n, p, k)
## Build the level set tree.
gamma = int(0.1 * n)
tree = dcl.construct_tree_from_graph(knn_graph,
density,
prune_threshold=gamma,
verbose=True)
print tree
## Save the tree to disk and load it back (just for demo purposes).
tree.save('crater_tree.lst')
tree2 = dcl.load_tree('crater_tree.lst')
print tree2
## Get leaf clusters and leaf node indices.
leaves = tree.get_leaf_nodes()
labels = tree.get_clusters(method='leaf')
## Plot the tree, coloring the nodes used to cluster.
fig = tree.plot(form='mass', color_nodes=leaves, colormap=plt.cm.Spectral)[0]
## Construct a similarity graph and density estimate by hand. For this demo we
# use the k-nearest neighbors similarity graph and density estimator that
# DeBaCl uses by default, but this is the place to plug in custom similarity
# graphs and density estimators. In particular, this functionality means the
# user is not limited to tabular data or pre-set distance functions. If the
# user can make a similarity graph and estimate a density (or pseudo-density),
# then a level set tree can be estimated with DeBaCl.
k = int(0.05 * n)
knn_graph, radii = dcl.utils.knn_graph(X, k, method='kd-tree')
density = dcl.utils.knn_density(radii, n, p, k)
## Build the level set tree.
gamma = int(0.1 * n)
tree = dcl.construct_tree_from_graph(knn_graph, density, prune_threshold=gamma,
verbose=True)
print tree
## Save the tree to disk and load it back (just for demo purposes).
tree.save('crater_tree.lst')
tree2 = dcl.load_tree('crater_tree.lst')
print tree2
## Get leaf clusters and leaf node indices.
leaves = tree.get_leaf_nodes()
labels = tree.get_clusters(method='leaf')