def _nearest(self, point):
"""
Returns the cluster nearest to point
This method uses the distance method of each Cluster to compute the distance
between point and the cluster centroid. It returns the Cluster that is closest.
Ties are broken in favor of clusters occurring earlier self._clusters.
Parameter point: The point to compare.
Precondition: point is a list of numbers (int or float), with the same dimension
as the dataset.
"""
assert a6checks.is_point(point)
assert len(point) == self._dataset.getDimension()
list_of_clusters = self.getClusters()
nearest_cluster = None
nearest_distance = None
for i in range(len(list_of_clusters)):
if (i == 0):
nearest_cluster = list_of_clusters[i]
nearest_distance = list_of_clusters[i].distance(point)
if (list_of_clusters[i].distance(point) < nearest_distance):
nearest_cluster = list_of_clusters[i]
nearest_distance = list_of_clusters[i].distance(point)
return nearest_cluster
def distance(self, point):
"""
Returns the euclidean distance from point to this cluster's centroid.
Parameter point: The point to be measured
Precondition: point is a list of numbers (int or float), with the same dimension
as the centroid.
"""
assert a6checks.is_point(point) and len(point)==len(self._centroid)
return numpy.sqrt(sum([(b-a)*(b-a) for a,b in zip(point, self._centroid)]))
def addPoint(self, point):
"""
Adds a COPY of point at the end of _contents.
This method does not add the point directly. It adds a copy of the point.
Precondition: point is a list of numbers (int or float), len(point) = _dimension."""
assert a6checks.is_point(point) and len(point) == self._dimension
apoint = []
for value in point:
apoint.append(value)
self._contents.append(apoint)
def addPoint(self,point):
"""
Adds a COPY of point at the end of _contents.
This method does not add the point directly. It adds a copy of the point.
Precondition: point is a list of numbers (int or float), len(point) = _dimension."""
assert a6checks.is_point(point) == True
assert len(point) == self._dimension
pointcopy = []
for x in range(0, len(point)):
pointcopy.append(point[x])
self._contents.append(pointcopy)
def addPoint(self, point):
"""
Adds a COPY of point at the end of _contents.
This method does not add the point directly. It adds a copy of the point.
Precondition: point is a list of numbers (int or float), len(point) = _dimension."""
#Assert Precondition
assert a6checks.is_point(point) == True
copy_point = point[:]
return self._contents.append(copy_point)
def _nearest(self, point):
"""
Returns the cluster nearest to point
Parameter point: The point to compare.
Precondition: point is a list of numbers (int or float), with the same dimension
as the dataset.
"""
assert a6checks.is_point(point) == True
assert len(point) == self._dataset.getDimension()
mindist = self._clusters[0].distance(point)
nearestcluster = self._clusters[0]
for newclusterobject in self._clusters:
if newclusterobject.distance(point) < mindist:
mindist = newclusterobject.distance(point)
nearestcluster = newclusterobject
return nearestcluster
def distance(self, point):
"""
Returns the euclidean distance from point to this cluster's centroid.
Parameter point: The point to be measured
Precondition: point is a list of numbers (int or float), with the same dimension
as the centroid.
"""
assert a6checks.is_point(point) == True
assert len(point) == len(self._centroid)
pointresult = 0
pointdim = len(point)
total = 0
centroidind = self.getCentroid()
for x in range(pointdim):
pointresult = point[x] - centroidind[x]
square = (pointresult)**2
total = total + square
root = math.sqrt(total)
return root
def __init__(self, dset, centroid):
"""
Initializes a new empty cluster whose centroid is a copy of <centroid>
Parameter dset: the dataset
Precondition: dset is an instance of Dataset
Parameter centroid: the cluster centroid
Precondition: centroid is a list of ds.getDimension() numbers
"""
assert isinstance(dset, a6dataset.Dataset)
self._dataset= dset
self._indices= []
result=[]
for apoint in centroid:
result.append(apoint)
self._centroid= result
if self._centroid != []:
assert a6checks.is_point(centroid) and len(centroid)==self._dataset.getDimension()
if self._indices != []:
assert 0 <= min(self.indices) and max(self._indices) <= self._dataset.getSize()