def test_euclidean_cols(self):
assert_almost_equal = np.testing.assert_almost_equal
data = self.cont_data
dist = distance.Euclidean(data, axis=0, normalize=False)
assert_almost_equal(
dist,
[[0, 8.062257748, 4.242640687], [8.062257748, 0, 5.196152423],
[4.242640687, 5.196152423, 0]])
with data.unlocked():
data.X[1, 1] = np.nan
dist = distance.Euclidean(data, axis=0, normalize=False)
assert_almost_equal(
dist,
[[0, 6.218252702, 4.242640687], [6.218252702, 0, 2.581988897],
[4.242640687, 2.581988897, 0]])
with data.unlocked():
data.X[1, 0] = np.nan
dist = distance.Euclidean(data, axis=0, normalize=False)
assert_almost_equal(
dist,
[[0, 6.218252702, 5.830951895], [6.218252702, 0, 2.581988897],
[5.830951895, 2.581988897, 0]])
def test_euclidean_cols_normalized(self):
assert_almost_equal = np.testing.assert_almost_equal
data = self.cont_data
dist = distance.Euclidean(data, axis=0, normalize=True)
assert_almost_equal(
dist,
[[0, 2.455273959, 0.649839392],
[2.455273959, 0, 2.473176308],
[0.649839392, 2.473176308, 0]])
with data.unlocked():
data.X[1, 1] = np.nan
dist = distance.Euclidean(data, axis=0, normalize=True)
assert_almost_equal(
dist,
[[0, 2, 0.649839392],
[2, 0, 1.704275472],
[0.649839392, 1.704275472, 0]])
with data.unlocked():
data.X[1, 0] = np.nan
dist = distance.Euclidean(data, axis=0, normalize=True)
assert_almost_equal(
dist,
[[0, 2, 1.450046001],
[2, 0, 1.704275472],
[1.450046001, 1.704275472, 0]])
def test_euclidean_cont(self):
assert_almost_equal = np.testing.assert_almost_equal
data = self.cont_data
dist = distance.Euclidean(data, axis=1, normalize=False)
assert_almost_equal(
dist,
np.sqrt(
np.array([[0, 12, 5, 38], [12, 0, 21, 82], [5, 21, 0, 41],
[38, 82, 41, 0]])))
with data.unlocked():
data.X[1, 0] = np.nan
dist = distance.Euclidean(data, axis=1, normalize=False)
assert_almost_equal(dist, [[0, 4.472135955, 2.236067977, 6.164414003],
[4.472135955, 0, 5.385164807, 6.480740698],
[2.236067977, 5.385164807, 0, 6.403124237],
[6.164414003, 6.480740698, 6.403124237, 0]])
with data.unlocked():
data.X[0, 0] = np.nan
dist = distance.Euclidean(data, axis=1, normalize=False)
assert_almost_equal(dist, [[0, 5.099019514, 4.795831523, 4.472135955],
[5.099019514, 0, 5.916079783, 6],
[4.795831523, 5.916079783, 0, 6.403124237],
[4.472135955, 6, 6.403124237, 0]])
def test_two_tables(self):
assert_almost_equal = np.testing.assert_almost_equal
dist = distance.Euclidean(self.cont_data,
self.cont_data2,
normalize=True)
assert_almost_equal(
dist,
[
[1.17040218, 0.47809144],
[2.78516478, 1.96961039],
[1.28668394, 0.79282497],
[1.27179413, 1.54919334],
],
)
model = distance.Euclidean(normalize=True).fit(self.cont_data)
dist = model(self.cont_data, self.cont_data2)
assert_almost_equal(
dist,
[
[1.17040218, 0.47809144],
[2.78516478, 1.96961039],
[1.28668394, 0.79282497],
[1.27179413, 1.54919334],
],
)
dist = model(self.cont_data2)
assert_almost_equal(dist, [[0, 0.827119692], [0.827119692, 0]])
def test_euclidean_cont_normalized(self):
assert_almost_equal = np.testing.assert_almost_equal
data = self.cont_data
model = distance.Euclidean(axis=1, normalize=True).fit(data)
assert_almost_equal(model.means, [2, 2.75, 1.5])
assert_almost_equal(model.vars, [9, 2.1875, 1.25])
assert_almost_equal(model.dist_missing2_cont, [1, 1, 1])
dist = model(data)
assert_almost_equal(
dist,
[
[0, 1.654239383, 1.146423008, 1.621286967],
[1.654239383, 0, 2.068662631, 3.035242727],
[1.146423008, 2.068662631, 0, 1.956673562],
[1.621286967, 3.035242727, 1.956673562, 0],
],
)
dist = distance.Euclidean(data, axis=1, normalize=True)
assert_almost_equal(
dist,
[
[0, 1.654239383, 1.146423008, 1.621286967],
[1.654239383, 0, 2.068662631, 3.035242727],
[1.146423008, 2.068662631, 0, 1.956673562],
[1.621286967, 3.035242727, 1.956673562, 0],
],
)
data.X[1, 0] = np.nan
model = distance.Euclidean(axis=1, normalize=True).fit(data)
assert_almost_equal(model.means, [3, 2.75, 1.5])
assert_almost_equal(model.vars, [8, 2.1875, 1.25])
dist = model(data)
assert_almost_equal(
dist,
[
[0, 1.806733438, 1.146423008, 1.696635326],
[1.806733438, 0, 2.192519751, 2.675283697],
[1.146423008, 2.192519751, 0, 2.019547333],
[1.696635326, 2.675283697, 2.019547333, 0],
],
)
data.X[0, 0] = np.nan
model = distance.Euclidean(axis=1, normalize=True).fit(data)
assert_almost_equal(model.means, [4, 2.75, 1.5])
assert_almost_equal(model.vars, [9, 2.1875, 1.25])
dist = model(data)
assert_almost_equal(
dist,
[
[0, 1.874642823, 1.521277659, 1.276154939],
[1.874642823, 0, 2.248809209, 2.580143961],
[1.521277659, 2.248809209, 0, 1.956673562],
[1.276154939, 2.580143961, 1.956673562, 0],
],
)
def test_euclidean_disc(self):
assert_almost_equal = np.testing.assert_almost_equal
data = self.disc_data
model = distance.Euclidean().fit(data)
assert_almost_equal(model.dist_missing_disc,
[[1 / 3, 2 / 3, 1, 1], [2 / 3, 2 / 3, 1, 2 / 3],
[2 / 3, 1 / 3, 1, 1]])
assert_almost_equal(model.dist_missing2_disc,
[1 - 5 / 9, 1 - 3 / 9, 1 - 5 / 9])
dist = model(data)
assert_almost_equal(
dist, np.sqrt(np.array([[0, 2, 3], [2, 0, 2], [3, 2, 0]])))
with data.unlocked():
data.X[1, 0] = np.nan
model = distance.Euclidean().fit(data)
assert_almost_equal(model.dist_missing_disc,
[[1 / 2, 1 / 2, 1, 1], [2 / 3, 2 / 3, 1, 2 / 3],
[2 / 3, 1 / 3, 1, 1]])
assert_almost_equal(model.dist_missing2_disc,
[1 - 2 / 4, 1 - 3 / 9, 1 - 5 / 9])
with data.unlocked():
dist = model(data)
assert_almost_equal(
dist, np.sqrt(np.array([[0, 2.5, 3], [2.5, 0, 1.5], [3, 1.5, 0]])))
with data.unlocked():
data.X[0, 0] = np.nan
model = distance.Euclidean().fit(data)
assert_almost_equal(
model.dist_missing_disc,
[[1, 0, 1, 1], [2 / 3, 2 / 3, 1, 2 / 3], [2 / 3, 1 / 3, 1, 1]])
assert_almost_equal(model.dist_missing2_disc,
[1 - 1, 1 - 3 / 9, 1 - 5 / 9])
dist = model(data)
assert_almost_equal(
dist, np.sqrt(np.array([[0, 2, 2], [2, 0, 1], [2, 1, 0]])))
data = self.disc_data4
with data.unlocked():
data.X[:2, 0] = np.nan
model = distance.Euclidean().fit(data)
assert_almost_equal(model.dist_missing_disc,
[[1 / 2, 1 / 2, 1, 1], [3 / 4, 2 / 4, 1, 3 / 4],
[3 / 4, 1 / 4, 1, 1]])
assert_almost_equal(model.dist_missing2_disc,
[1 - 2 / 4, 1 - 6 / 16, 1 - 10 / 16])
dist = model(data)
assert_almost_equal(
dist,
np.sqrt(
np.array([[0, 2.5, 2.5, 2.5], [2.5, 0, 0.5, 1.5],
[2.5, 0.5, 0, 2], [2.5, 1.5, 2, 0]])))
def test_euclidean_mixed(self):
assert_almost_equal = np.testing.assert_almost_equal
data = self.mixed_data
model = distance.Euclidean(axis=1, normalize=True).fit(data)
assert_almost_equal(model.means, [1 / 3, 3, 1])
assert_almost_equal(model.vars, [8 / 9, 8 / 3, 2 / 3])
assert_almost_equal(
model.dist_missing_disc,
[[1 / 3, 2 / 3, 1, 1], [2 / 3, 2 / 3, 1, 2 / 3],
[2 / 3, 1 / 3, 1, 1]],
)
assert_almost_equal(model.dist_missing2_cont, [1, 1, 1])
assert_almost_equal(model.dist_missing2_disc,
[1 - 5 / 9, 1 - 3 / 9, 1 - 5 / 9])
dist = model(data)
assert_almost_equal(
dist,
[
[0, 2.828427125, 2.121320344],
[2.828427125, 0, 2.828427125],
[2.121320344, 2.828427125, 0],
],
)
def build_linkage_matrix(topic_table):
#print(topic_table.domain)
x = data.Table(topic_table)
print(x.X)
dist_matrix = distance.Euclidean(x.X)
#d = Orange.misc.distmatrix.__new__(dist_matrix)
print(dist_matrix)
#linkage = hierarchical.dist_matrix_linkage(dist_matrix,linkage = hierarchical.AVERAGE
linkage = scipy.cluster.hierarchy.linkage(dist_matrix,
method=hierarchical.AVERAGE)
print((linkage))
return linkage
def test_main():
from PyQt4.QtGui import QApplication
import sip
import Orange.distance as distance
app = QApplication([])
w = OWHierarchicalClustering()
data = Orange.data.Table("iris.tab")
matrix = distance.Euclidean(data)
w.set_distances(matrix)
w.handleNewSignals()
w.show()
w.raise_()
rval = app.exec_()
w.onDeleteWidget()
sip.delete(w)
del w
app.processEvents()
return rval
f1 = 'Hierarchical.csv'
if os.path.exists(f1):
os.remove(f1)
f2 = 'KMeans.csv'
if os.path.exists(f2):
os.remove(f2)
data = Orange.data.Table('output.tab')
#matrix = Orange.misc.SymMatrix(len(data))
numDocs = len(data)
print "Count of documents in Reuters dataset: " + str(numDocs) + "\n"
print "1. Constructing Distance Matrices\n"
starter = time.time()
constructorEuclidean = distance.Euclidean()
EuclideanDistanceMat = distance.distance_matrix(
data, distance_constructor=constructorEuclidean)
euclidean_hierarchical_clustering = clustering.hierarchical.HierarchicalClustering(
)
euclidean_hierarchical_clustering.linkage = clustering.hierarchical.AVERAGE
euclideanRoot = euclidean_hierarchical_clustering(EuclideanDistanceMat)
ender = time.time()
timer = ender - starter
starter1 = time.time()
constructorManhattan = distance.Manhattan()
ManhattanDistanceMat = distance.distance_matrix(
data, distance_constructor=constructorManhattan)
manhattan_hierarchical_clustering = clustering.hierarchical.HierarchicalClustering(
)
def paint(self, painter, *args):
if self._line is None:
self.boundingRect()
painter.save()
painter.setPen(self.pen())
painter.drawLine(self._line)
painter.restore()
def clusters_at_height(root, height):
"""Return a list of clusters by cutting the clustering at `height`.
"""
lower = set()
cluster_list = []
for cl in preorder(root):
if cl in lower:
continue
if cl.value.height < height:
cluster_list.append(cl)
lower.update(preorder(cl))
return cluster_list
if __name__ == "__main__": # pragma: no cover
from Orange import distance
data = Orange.data.Table("iris")
matrix = distance.Euclidean(distance._preprocess(data))
WidgetPreview(OWHierarchicalClustering).run(matrix)
import Orange.data
import Orange.misc
from Orange.widgets import widget, gui, settings
from Orange import distance
_METRICS = [
("Euclidean", distance.Euclidean()),
("Manhattan", distance.Manhattan()),
("Cosine", distance.Cosine()),
("Jaccard", distance.Jaccard()),
("Mahalanobis", distance.Mahalanobis()),
("Spearman", distance.SpearmanR()),
("Spearman absolute", distance.SpearmanRAbsolute()),
("Pearson", distance.PearsonR()),
("Pearson absolute", distance.PearsonRAbsolute()),
]
class OWDistances(widget.OWWidget):
name = "Distances"
description = "Compute a matrix of pairwise distances."
icon = "icons/Distance.svg"
inputs = [("Data", Orange.data.Table, "set_data")]
outputs = [("Distances", Orange.misc.DistMatrix)]
axis = settings.Setting(0)
metric_idx = settings.Setting(0)
autocommit = settings.Setting(False)
want_main_area = False
