def square_euclidean_dist(v1,v2):
if (v1 == v2).all():
return 0.0
valids = 0
distance= 0.0
for i in xrange(len(v1)):
if numpy.isfinite(v1[i]) and numpy.isfinite(v2[i]):
valids += 1
d = v1[i]-v2[i]
distance += d*d
if valids==0:
raise ValueError, "Cannot calculate values"
return distance/valids
def square_euclidean_dist(v1, v2):
if (v1 == v2).all():
return 0.0
valids = 0
distance = 0.0
for i in xrange(len(v1)):
if numpy.isfinite(v1[i]) and numpy.isfinite(v2[i]):
valids += 1
d = v1[i] - v2[i]
distance += d * d
if valids == 0:
raise ValueError, "Cannot calculate values"
return distance / valids
def safe_mean(values):
""" Returns mean value discarding non finite values """
valid_values = []
for v in values:
if numpy.isfinite(v):
valid_values.append(v)
return numpy.mean(valid_values), numpy.std(valid_values)
def safe_mean(values):
""" Returns mean value discarding non finite values """
valid_values = []
for v in values:
if numpy.isfinite(v):
valid_values.append(v)
return numpy.mean(valid_values), numpy.std(valid_values)
def safe_mean_vector(vectors):
""" Returns mean profile discarding non finite values """
# if only one vector, avg = itself
if len(vectors) == 1:
return vectors[0], numpy.zeros(len(vectors[0]))
# Takes the vector length form the first item
length = len(vectors[0])
safe_mean = []
safe_std = []
for pos in xrange(length):
pos_mean = []
for v in vectors:
if numpy.isfinite(v[pos]):
pos_mean.append(v[pos])
safe_mean.append(numpy.mean(pos_mean))
safe_std.append(numpy.std(pos_mean))
return safe_mean, safe_std
def safe_mean_vector(vectors):
""" Returns mean profile discarding non finite values """
# if only one vector, avg = itself
if len(vectors)==1:
return vectors[0], numpy.zeros(len(vectors[0]))
# Takes the vector length form the first item
length = len(vectors[0])
safe_mean = []
safe_std = []
for pos in xrange(length):
pos_mean = []
for v in vectors:
if numpy.isfinite(v[pos]):
pos_mean.append(v[pos])
safe_mean.append(numpy.mean(pos_mean))
safe_std.append(numpy.std(pos_mean))
return safe_mean, safe_std
def link_to_arraytable(self, arraytbl):
""" Allows to link a given arraytable object to the tree
structure under this node. Row names in the arraytable object
are expected to match leaf names.
Returns a list of nodes for with profiles could not been found
in arraytable.
"""
# Initialize tree with array data
if type(arraytbl) == ArrayTable:
array = arraytbl
else:
array = ArrayTable(arraytbl)
missing_leaves = []
matrix_values = [i for r in xrange(len(array.matrix))\
for i in array.matrix[r] if numpy.isfinite(i)]
array._matrix_min = min(matrix_values)
array._matrix_max = max(matrix_values)
for n in self.traverse():
n.arraytable = array
if n.is_leaf() and n.name in array.rowNames:
n._profile = array.get_row_vector(n.name)
elif n.is_leaf():
n._profile = [numpy.nan]*len(array.colNames)
missing_leaves.append(n)
if len(missing_leaves)>0:
print >>stderr, \
"""[%d] leaf names could not be mapped to the matrix rows.""" %\
len(missing_leaves)
self.arraytable = array
