Python scipy.cluster.hierarchy.to_tree() Examples
The following are 11
code examples of scipy.cluster.hierarchy.to_tree().
You can vote up the ones you like or vote down the ones you don't like,
and go to the original project or source file by following the links above each example.
You may also want to check out all available functions/classes of the module
scipy.cluster.hierarchy
, or try the search function
.
.
Example #1
| Source File: corClust.py From KitNET-py with MIT License | 5 votes |
|
def cluster(self,maxClust):
D = self.corrDist()
Z = linkage(D[np.triu_indices(self.n, 1)]) # create a linkage matrix based on the distance matrix
if maxClust < 1:
maxClust = 1
if maxClust > self.n:
maxClust = self.n
map = self.__breakClust__(to_tree(Z),maxClust)
return map
# a recursive helper function which breaks down the dendrogram branches until all clusters have no more than maxClust elements
Example #2
| Source File: clustering.py From anvio with GNU General Public License v3.0 | 5 votes |
|
def get_clustering_as_tree(vectors, linkage=constants.linkage_method_default, distance=constants.distance_metric_default, progress=progress):
is_distance_and_linkage_compatible(distance, linkage)
progress.update('Clustering data with "%s" linkage using "%s" distance' % (linkage, distance))
linkage = hierarchy.linkage(vectors, metric=distance, method=linkage)
progress.update('Recovering the tree from the clustering result')
tree = hierarchy.to_tree(linkage, rd=False)
return tree
Example #3
| Source File: corClust.py From Kitsune-py with MIT License | 5 votes |
|
def cluster(self,maxClust):
D = self.corrDist()
Z = linkage(D[np.triu_indices(self.n, 1)]) # create a linkage matrix based on the distance matrix
if maxClust < 1:
maxClust = 1
if maxClust > self.n:
maxClust = self.n
map = self.__breakClust__(to_tree(Z),maxClust)
return map
# a recursive helper function which breaks down the dendrogram branches until all clusters have no more than maxClust elements
Example #4
| Source File: sdm.py From scedar with MIT License | 5 votes |
|
def hct_from_lkg(hac_z):
return HClustTree(sch.to_tree(hac_z))
Example #5
| Source File: test_mirac.py From scedar with MIT License | 5 votes |
|
def test_mirac_wrong_args(self):
x = np.zeros((10, 10))
# wrong min_cl_n
with pytest.raises(ValueError) as excinfo:
cluster.MIRAC(x, metric='euclidean', min_cl_n=-0.1)
with pytest.raises(ValueError) as excinfo:
cluster.MIRAC(x, metric='euclidean', min_cl_n=-0.1)
# wrong cl_mdl_scale_factor
with pytest.raises(ValueError) as excinfo:
cluster.MIRAC(x, metric='euclidean', cl_mdl_scale_factor=-0.1)
# wrong encode type
with pytest.raises(ValueError) as excinfo:
cluster.MIRAC(x, metric='euclidean', encode_type='1')
with pytest.raises(ValueError) as excinfo:
cluster.MIRAC(x, metric='euclidean', encode_type=1)
with pytest.raises(ValueError) as excinfo:
cluster.MIRAC(x, metric='euclidean', dim_reduct_method='NONN')
# hac tree n_leaves different from n_samples
z = sch.linkage([[0], [5], [6], [8], [9], [12]],
method='single', optimal_ordering=True)
hct = eda.HClustTree(sch.to_tree(z))
with pytest.raises(ValueError) as excinfo:
cluster.MIRAC(x, metric='euclidean', hac_tree=hct)
# no specific purpose. Just to exaust the coverage
Example #6
| Source File: test_dense_sdm.py From scedar with MIT License | 5 votes |
|
def test_bi_partition_min_no_spl(self):
# ____|____ 6
# | ___|____ 5
# | | __|___ 4
# | | | |
# 3 2 1 0
z = sch.linkage([[0, 0], [1, 1], [3, 3], [6, 6]],
metric='euclidean', method='complete',
optimal_ordering=True)
hct = eda.HClustTree(sch.to_tree(z))
assert hct.leaf_ids() == [3, 2, 1, 0]
labs, sids, lst, rst = hct.bi_partition(
soft_min_subtree_size=2, return_subtrees=True)
assert labs == [0, 0, 1, 1]
assert sids == [3, 2, 1, 0]
# hct should be changed accordingly
assert hct.leaf_ids() == [3, 2, 1, 0]
assert hct.left_leaf_ids() == [3, 2]
assert hct.right_leaf_ids() == [1, 0]
# subtrees
assert lst.leaf_ids() == [3, 2]
assert rst.leaf_ids() == [1, 0]
# prev
assert lst._prev is hct
assert rst._prev is hct
# ids
assert lst._node.id == 5
assert lst._node.left.id == 3
assert lst._node.right.id == 2
# ids
assert rst._node.id == 4
assert rst._node.left.id == 1
assert rst._node.right.id == 0
Example #7
| Source File: test_dense_sdm.py From scedar with MIT License | 5 votes |
|
def test_bi_partition_min_no_spl_lr_rev(self):
# left right reversed
# ____|____ 6
# | ___|____ 5
# | | __|___ 4
# | | | |
# 3 2 1 0
z = sch.linkage([[0, 0], [1, 1], [3, 3], [6, 6]],
metric='euclidean', method='complete',
optimal_ordering=True)
root = sch.to_tree(z)
# reverse left right subtree
root_left = root.left
root.left = root.right
root.right = root_left
hct = eda.HClustTree(root)
assert hct.leaf_ids() == [2, 1, 0, 3]
labs, sids, lst, rst = hct.bi_partition(
soft_min_subtree_size=2, return_subtrees=True)
assert labs == [0, 0, 1, 1]
assert sids == [2, 1, 0, 3]
# hct should be changed accordingly
assert hct.leaf_ids() == [2, 1, 0, 3]
assert hct.left_leaf_ids() == [2, 1]
assert hct.right_leaf_ids() == [0, 3]
# subtrees
assert lst.leaf_ids() == [2, 1]
assert rst.leaf_ids() == [0, 3]
# prev
assert lst._prev is hct
assert rst._prev is hct
assert hct._left is lst._node
assert hct._right is rst._node
# ids
assert rst._node.id == 4
assert rst._node.left.id == 0
assert rst._node.right.id == 3
# ids
assert lst._node.id == 5
assert lst._node.left.id == 2
assert lst._node.right.id == 1
Example #8
| Source File: test_dense_sdm.py From scedar with MIT License | 5 votes |
|
def test_bi_partition_min_spl(self):
# _____|_____
# | ____|____
# | __|__ __|__
# | | | | |
# 4 3 2 1 0
z = sch.linkage([[0, 0], [1, 1], [3, 3], [4, 4], [10, 10]],
metric='euclidean', method='complete',
optimal_ordering=True)
hct = eda.HClustTree(sch.to_tree(z))
assert hct.leaf_ids() == [4, 3, 2, 1, 0]
assert hct.left_leaf_ids() == [4]
assert hct.right().left().leaf_ids() == [3, 2]
assert hct.right().right().leaf_ids() == [1, 0]
labs, sids, lst, rst = hct.bi_partition(
soft_min_subtree_size=2, return_subtrees=True)
assert labs == [0, 0, 0, 1, 1]
assert sids == [4, 3, 2, 1, 0]
# hct should be changed accordingly
assert hct.leaf_ids() == [4, 3, 2, 1, 0]
assert hct.left_leaf_ids() == [4, 3, 2]
assert hct.right_leaf_ids() == [1, 0]
# left
assert lst._prev is hct
assert lst._node.left.left.id == 4
assert lst._node.left.right.id == 3
assert lst._node.right.id == 2
# right
assert rst._prev is hct
assert rst._node.left.id == 1
assert rst._node.right.id == 0
Example #9
| Source File: test_dense_sdm.py From scedar with MIT License | 5 votes |
|
def test_bi_partition_min_switch_spl(self):
# _______|________
# | _____|_____
# | ____|____ |
# | __|__ __|__ |
# | | | | | |
# 0 1 2 3 4 5
# round 1: ( ((0, (1, 2)), (3, 4)), (5) )
# round 2: ( (0, (1, 2), (3, (4, 5)) )
z = sch.linkage([[0], [5], [6], [8], [9], [12]],
method='single', optimal_ordering=True)
root = sch.to_tree(z)
assert root.left.id == 0
assert root.right.right.id == 5
assert root.right.left.left.left.id == 1
assert root.right.left.left.right.id == 2
assert root.right.left.right.left.id == 3
assert root.right.left.right.right.id == 4
hct = eda.HClustTree(root)
labs, sids, lst, rst = hct.bi_partition(
soft_min_subtree_size=3, return_subtrees=True)
assert labs == [0, 0, 0, 1, 1, 1]
assert sids == [0, 1, 2, 3, 4, 5]
# lst
assert hct._left is lst._node
assert lst._prev is hct
assert lst.left_leaf_ids() == [0]
assert lst.right_leaf_ids() == [1, 2]
# rst
assert hct._right is rst._node
assert rst._prev is hct
assert rst.left_leaf_ids() == [3]
assert rst.right_leaf_ids() == [4, 5]
Example #10
| Source File: hierarchical_clustering.py From CompareM with GNU General Public License v3.0 | 4 votes |
|
def run(self, pairwise_value_file,
method,
similarity,
max_sim_value,
name_col1,
name_col2,
value_col,
output_tree):
"""Perform hierarchical clustering on pairwise value files.
Parameters
----------
pairwise_value_file : str
File with pairwise similarity or dissimilarity values.
method : str
Clustering method to use.
similarity : boolean
Flag indicating file contain similarity values.
max_sim_value : float
Maximum value of similarity scores.
name_col1 : int
Index of first column with genome names.
name_col2 : int
Index of second column with genome names.
value_col : int
Index of column with similarity or dissimilarity values.
"""
diss_vector, genome_labels = self._parse_data(pairwise_value_file,
name_col1,
name_col2,
value_col,
similarity,
max_sim_value)
clusters = hierarchy.linkage(diss_vector, method=method)
tree = hierarchy.to_tree(clusters)
newick_str = self._save_newick(tree, "", tree.dist, genome_labels)
fout = open(output_tree, 'w')
fout.write(newick_str + '\n')
fout.close()
Example #11
| Source File: hierarchical_clustering.py From SqueezeMeta with GNU General Public License v3.0 | 4 votes |
|
def run(self, pairwise_value_file,
method,
similarity,
max_sim_value,
name_col1,
name_col2,
value_col,
output_tree):
"""Perform hierarchical clustering on pairwise value files.
Parameters
----------
pairwise_value_file : str
File with pairwise similarity or dissimilarity values.
method : str
Clustering method to use.
similarity : boolean
Flag indicating file contain similarity values.
max_sim_value : float
Maximum value of similarity scores.
name_col1 : int
Index of first column with genome names.
name_col2 : int
Index of second column with genome names.
value_col : int
Index of column with similarity or dissimilarity values.
"""
diss_vector, genome_labels = self._parse_data(pairwise_value_file,
name_col1,
name_col2,
value_col,
similarity,
max_sim_value)
clusters = hierarchy.linkage(diss_vector, method=method)
tree = hierarchy.to_tree(clusters)
newick_str = self._save_newick(tree, "", tree.dist, genome_labels)
fout = open(output_tree, 'w')
fout.write(newick_str + '\n')
fout.close()
