def getrow(self, i):
"""Returns a copy of row i of the matrix, as a (1 x n) sparse
matrix (row vector).
"""
# Spmatrix subclasses should override this method for efficiency.
# Pre-multiply by a (1 x m) row vector 'a' containing all zeros
# except for a_i = 1
from csr import csr_matrix
m = self.shape[0]
a = csr_matrix((1, m), dtype=self.dtype)
a[0, i] = 1
return a * self
def tocsr(self):
M, N = self.shape
indptr = np.empty(M + 1, dtype=np.intc)
indices = np.empty(self.nnz, dtype=np.intc)
data = np.empty(self.nnz, dtype=upcast(self.dtype))
csc_tocsr(M, N, self.indptr, self.indices, self.data, indptr, indices, data)
from csr import csr_matrix
A = csr_matrix((data, indices, indptr), shape=self.shape)
A.has_sorted_indices = True
return A
def tocsr(self):
M, N = self.shape
indptr = np.empty(M + 1, dtype=np.intc)
indices = np.empty(self.nnz, dtype=np.intc)
data = np.empty(self.nnz, dtype=upcast(self.dtype))
csc_tocsr(M, N, \
self.indptr, self.indices, self.data, \
indptr, indices, data)
from csr import csr_matrix
A = csr_matrix((data, indices, indptr), shape=self.shape)
A.has_sorted_indices = True
return A
def estimate_blocksize(A,efficiency=0.7):
"""Attempt to determine the blocksize of a sparse matrix
Returns a blocksize=(r,c) such that
- A.nnz / A.tobsr( (r,c) ).nnz > efficiency
"""
if not (isspmatrix_csr(A) or isspmatrix_csc(A)):
A = csr_matrix(A)
if A.nnz == 0:
return (1,1)
if not 0 < efficiency < 1.0:
raise ValueError,'efficiency must satisfy 0.0 < efficiency < 1.0'
high_efficiency = (1.0 + efficiency) / 2.0
nnz = float(A.nnz)
M,N = A.shape
if M % 2 == 0 and N % 2 == 0:
e22 = nnz / ( 4 * count_blocks(A,(2,2)) )
else:
e22 = 0.0
if M % 3 == 0 and N % 3 == 0:
e33 = nnz / ( 9 * count_blocks(A,(3,3)) )
else:
e33 = 0.0
if e22 > high_efficiency and e33 > high_efficiency:
e66 = nnz / ( 36 * count_blocks(A,(6,6)) )
if e66 > efficiency:
return (6,6)
else:
return (3,3)
else:
if M % 4 == 0 and N % 4 == 0:
e44 = nnz / ( 16 * count_blocks(A,(4,4)) )
else:
e44 = 0.0
if e44 > efficiency:
return (4,4)
elif e33 > efficiency:
return (3,3)
elif e22 > efficiency:
return (2,2)
else:
return (1,1)
def tocsr(self):
"""Return a copy of this matrix in Compressed Sparse Row format
Duplicate entries will be summed together.
Example
-------
>>> from numpy import array
>>> from scipy.sparse import coo_matrix
>>> row = array([0,0,1,3,1,0,0])
>>> col = array([0,2,1,3,1,0,0])
>>> data = array([1,1,1,1,1,1,1])
>>> A = coo_matrix( (data,(row,col)), shape=(4,4)).tocsr()
>>> A.todense()
matrix([[3, 0, 1, 0],
[0, 2, 0, 0],
[0, 0, 0, 0],
[0, 0, 0, 1]])
"""
from csr import csr_matrix
if self.nnz == 0:
return csr_matrix(self.shape, dtype=self.dtype)
else:
M,N = self.shape
indptr = np.empty(M + 1, dtype=np.intc)
indices = np.empty(self.nnz, dtype=np.intc)
data = np.empty(self.nnz, dtype=upcast(self.dtype))
coo_tocsr(M, N, self.nnz, \
self.row, self.col, self.data, \
indptr, indices, data)
A = csr_matrix((data, indices, indptr), shape=self.shape)
A.sum_duplicates()
return A
def tocsr(self):
"""Return a copy of this matrix in Compressed Sparse Row format
Duplicate entries will be summed together.
Examples
--------
>>> from numpy import array
>>> from scipy.sparse import coo_matrix
>>> row = array([0,0,1,3,1,0,0])
>>> col = array([0,2,1,3,1,0,0])
>>> data = array([1,1,1,1,1,1,1])
>>> A = coo_matrix( (data,(row,col)), shape=(4,4)).tocsr()
>>> A.todense()
matrix([[3, 0, 1, 0],
[0, 2, 0, 0],
[0, 0, 0, 0],
[0, 0, 0, 1]])
"""
from csr import csr_matrix
if self.nnz == 0:
return csr_matrix(self.shape, dtype=self.dtype)
else:
M, N = self.shape
indptr = np.empty(M + 1, dtype=np.intc)
indices = np.empty(self.nnz, dtype=np.intc)
data = np.empty(self.nnz, dtype=upcast(self.dtype))
coo_tocsr(M, N, self.nnz, \
self.row, self.col, self.data, \
indptr, indices, data)
A = csr_matrix((data, indices, indptr), shape=self.shape)
A.sum_duplicates()
return A
def estimate_blocksize(A, efficiency=0.7):
"""Attempt to determine the blocksize of a sparse matrix
Returns a blocksize=(r,c) such that
- A.nnz / A.tobsr( (r,c) ).nnz > efficiency
"""
if not (isspmatrix_csr(A) or isspmatrix_csc(A)):
A = csr_matrix(A)
if A.nnz == 0:
return (1, 1)
if not 0 < efficiency < 1.0:
raise ValueError('efficiency must satisfy 0.0 < efficiency < 1.0')
high_efficiency = (1.0 + efficiency) / 2.0
nnz = float(A.nnz)
M, N = A.shape
if M % 2 == 0 and N % 2 == 0:
e22 = nnz / (4 * count_blocks(A, (2, 2)))
else:
e22 = 0.0
if M % 3 == 0 and N % 3 == 0:
e33 = nnz / (9 * count_blocks(A, (3, 3)))
else:
e33 = 0.0
if e22 > high_efficiency and e33 > high_efficiency:
e66 = nnz / (36 * count_blocks(A, (6, 6)))
if e66 > efficiency:
return (6, 6)
else:
return (3, 3)
else:
if M % 4 == 0 and N % 4 == 0:
e44 = nnz / (16 * count_blocks(A, (4, 4)))
else:
e44 = 0.0
if e44 > efficiency:
return (4, 4)
elif e33 > efficiency:
return (3, 3)
elif e22 > efficiency:
return (2, 2)
else:
return (1, 1)
def getrow(self, i):
"""Returns a copy of row i of the matrix, as a (1 x n) sparse
matrix (row vector).
"""
# Spmatrix subclasses should override this method for efficiency.
# Pre-multiply by a (1 x m) row vector 'a' containing all zeros
# except for a_i = 1
from csr import csr_matrix
m = self.shape[0]
if i < 0:
i += m
if i < 0 or i >= m:
raise IndexError("index out of bounds")
row_selector = csr_matrix(([1], [[0], [i]]), shape=(1,m), dtype=self.dtype)
return row_selector * self
def count_blocks(A,blocksize):
"""For a given blocksize=(r,c) count the number of occupied
blocks in a sparse matrix A
"""
r,c = blocksize
if r < 1 or c < 1:
raise ValueError,'r and c must be positive'
if isspmatrix_csr(A):
M,N = A.shape
return csr_count_blocks(M,N,r,c,A.indptr,A.indices)
elif isspmatrix_csc(A):
return count_blocks(A.T,(c,r))
else:
return count_blocks(csr_matrix(A),blocksize)
def count_blocks(A, blocksize):
"""For a given blocksize=(r,c) count the number of occupied
blocks in a sparse matrix A
"""
r, c = blocksize
if r < 1 or c < 1:
raise ValueError('r and c must be positive')
if isspmatrix_csr(A):
M, N = A.shape
return csr_count_blocks(M, N, r, c, A.indptr, A.indices)
elif isspmatrix_csc(A):
return count_blocks(A.T, (c, r))
else:
return count_blocks(csr_matrix(A), blocksize)
def getrow(self, i):
"""Returns a copy of row i of the matrix, as a (1 x n) sparse
matrix (row vector).
"""
# Spmatrix subclasses should override this method for efficiency.
# Pre-multiply by a (1 x m) row vector 'a' containing all zeros
# except for a_i = 1
from csr import csr_matrix
m = self.shape[0]
if i < 0:
i += m
if i < 0 or i >= m:
raise IndexError("index out of bounds")
row_selector = csr_matrix(([1], [[0], [i]]), shape=(1,m), dtype=self.dtype)
return row_selector * self
def __init__(self, arg1, shape=None, dtype=None, copy=False):
spmatrix.__init__(self)
self.dtype = getdtype(dtype, arg1, default=float)
# First get the shape
if isspmatrix(arg1):
if isspmatrix_lil(arg1) and copy:
A = arg1.copy()
else:
A = arg1.tolil()
if dtype is not None:
A = A.astype(dtype)
self.shape = A.shape
self.dtype = A.dtype
self.rows = A.rows
self.data = A.data
elif isinstance(arg1,tuple):
if isshape(arg1):
if shape is not None:
raise ValueError('invalid use of shape parameter')
M, N = arg1
self.shape = (M,N)
self.rows = np.empty((M,), dtype=object)
self.data = np.empty((M,), dtype=object)
for i in range(M):
self.rows[i] = []
self.data[i] = []
else:
raise TypeError('unrecognized lil_matrix constructor usage')
else:
#assume A is dense
try:
A = np.asmatrix(arg1)
except TypeError:
raise TypeError('unsupported matrix type')
else:
from csr import csr_matrix
A = csr_matrix(A, dtype=dtype).tolil()
self.shape = A.shape
self.dtype = A.dtype
self.rows = A.rows
self.data = A.data
def __init__(self, arg1, shape=None, dtype=None, copy=False):
spmatrix.__init__(self)
self.dtype = getdtype(dtype, arg1, default=float)
# First get the shape
if isspmatrix(arg1):
if isspmatrix_lil(arg1) and copy:
A = arg1.copy()
else:
A = arg1.tolil()
if dtype is not None:
A = A.astype(dtype)
self.shape = A.shape
self.dtype = A.dtype
self.rows = A.rows
self.data = A.data
elif isinstance(arg1, tuple):
if isshape(arg1):
if shape is not None:
raise ValueError('invalid use of shape parameter')
M, N = arg1
self.shape = (M, N)
self.rows = np.empty((M, ), dtype=object)
self.data = np.empty((M, ), dtype=object)
for i in range(M):
self.rows[i] = []
self.data[i] = []
else:
raise TypeError('unrecognized lil_matrix constructor usage')
else:
#assume A is dense
try:
A = np.asmatrix(arg1)
except TypeError:
raise TypeError('unsupported matrix type')
else:
from csr import csr_matrix
A = csr_matrix(A, dtype=dtype).tolil()
self.shape = A.shape
self.dtype = A.dtype
self.rows = A.rows
self.data = A.data
def eliminate_zeros(self):
R,C = self.blocksize
M,N = self.shape
mask = (self.data != 0).reshape(-1,R*C).sum(axis=1) #nonzero blocks
nonzero_blocks = mask.nonzero()[0]
if len(nonzero_blocks) == 0:
return #nothing to do
self.data[:len(nonzero_blocks)] = self.data[nonzero_blocks]
from csr import csr_matrix
# modifies self.indptr and self.indices *in place*
proxy = csr_matrix((mask,self.indices,self.indptr),shape=(M//R,N//C))
proxy.eliminate_zeros()
self.prune()
def eliminate_zeros(self):
R,C = self.blocksize
M,N = self.shape
mask = (self.data != 0).reshape(-1,R*C).sum(axis=1) #nonzero blocks
nonzero_blocks = mask.nonzero()[0]
if len(nonzero_blocks) == 0:
return #nothing to do
self.data[:len(nonzero_blocks)] = self.data[nonzero_blocks]
from csr import csr_matrix
# modifies self.indptr and self.indices *in place*
proxy = csr_matrix((mask,self.indices,self.indptr),shape=(M/R,N/C))
proxy.eliminate_zeros()
self.prune()
def tocsr(self):
""" Return Compressed Sparse Row format arrays for this matrix.
"""
indptr = np.asarray([len(x) for x in self.rows], dtype=np.intc)
indptr = np.concatenate( (np.array([0], dtype=np.intc), np.cumsum(indptr)) )
nnz = indptr[-1]
indices = []
for x in self.rows:
indices.extend(x)
indices = np.asarray(indices, dtype=np.intc)
data = []
for x in self.data:
data.extend(x)
data = np.asarray(data, dtype=self.dtype)
from csr import csr_matrix
return csr_matrix((data, indices, indptr), shape=self.shape)
def tocsr(self):
""" Return Compressed Sparse Row format arrays for this matrix.
"""
indptr = np.asarray([len(x) for x in self.rows], dtype=np.intc)
indptr = np.concatenate( (np.array([0], dtype=np.intc), np.cumsum(indptr)) )
nnz = indptr[-1]
indices = []
for x in self.rows:
indices.extend(x)
indices = np.asarray(indices, dtype=np.intc)
data = []
for x in self.data:
data.extend(x)
data = np.asarray(data, dtype=self.dtype)
from csr import csr_matrix
return csr_matrix((data, indices, indptr), shape=self.shape)
def cs_graph_components(x):
"""
Determine connected compoments of a graph stored as a compressed
sparse row or column matrix. For speed reasons, the symmetry of the
matrix x is not checked. A nonzero at index `(i, j)` means that node
`i` is connected to node `j` by an edge. The number of rows/columns
of the matrix thus corresponds to the number of nodes in the graph.
Parameters
-----------
x: ndarray-like, 2 dimensions, or sparse matrix
The adjacency matrix of the graph. Only the upper triangular part
is used.
Returns
--------
n_comp: int
The number of connected components.
label: ndarray (ints, 1 dimension):
The label array of each connected component (-2 is used to
indicate empty rows in the matrix: 0 everywhere, including
diagonal). This array has the length of the number of nodes,
i.e. one label for each node of the graph. Nodes having the same
label belong to the same connected component.
Notes
------
The matrix is assumed to be symmetric and the upper triangular part
of the matrix is used. The matrix is converted to a CSR matrix unless
it is already a CSR.
Example
-------
>>> from scipy.sparse import cs_graph_components
>>> import numpy as np
>>> D = np.eye(4)
>>> D[0,1] = D[1,0] = 1
>>> cs_graph_components(D)
(3, array([0, 0, 1, 2]))
>>> from scipy.sparse import dok_matrix
>>> cs_graph_components(dok_matrix(D))
(3, array([0, 0, 1, 2]))
"""
try:
shape = x.shape
except AttributeError:
raise ValueError(_msg0)
if not ((len(x.shape) == 2) and (x.shape[0] == x.shape[1])):
raise ValueError(_msg1 % x.shape)
if isspmatrix(x):
x = x.tocsr()
else:
x = csr_matrix(x)
label = np.empty((shape[0], ), dtype=x.indptr.dtype)
n_comp = _cs_graph_components(shape[0], x.indptr, x.indices, label)
return n_comp, label
def transpose(self, copy=False):
from csr import csr_matrix
M, N = self.shape
return csr_matrix((self.data, self.indices, self.indptr), (N, M), copy=copy)
def cs_graph_components(x):
"""
Determine connected compoments of a graph stored as a compressed
sparse row or column matrix. For speed reasons, the symmetry of the
matrix x is not checked. A nonzero at index `(i, j)` means that node
`i` is connected to node `j` by an edge. The number of rows/columns
of the matrix thus corresponds to the number of nodes in the graph.
Parameters
-----------
x: ndarray-like, 2 dimensions, or sparse matrix
The adjacency matrix of the graph. Only the upper triangular part
is used.
Returns
--------
n_comp: int
The number of connected components.
label: ndarray (ints, 1 dimension):
The label array of each connected component (-2 is used to
indicate empty rows in the matrix: 0 everywhere, including
diagonal). This array has the length of the number of nodes,
i.e. one label for each node of the graph. Nodes having the same
label belong to the same connected component.
Notes
------
The matrix is assumed to be symmetric and the upper triangular part
of the matrix is used. The matrix is converted to a CSR matrix unless
it is already a CSR.
Example
-------
>>> from scipy.sparse import cs_graph_components
>>> import numpy as np
>>> D = np.eye(4)
>>> D[0,1] = D[1,0] = 1
>>> cs_graph_components(D)
(3, array([0, 0, 1, 2]))
>>> from scipy.sparse import dok_matrix
>>> cs_graph_components(dok_matrix(D))
(3, array([0, 0, 1, 2]))
"""
try:
shape = x.shape
except AttributeError:
raise ValueError(_msg0)
if not ((len(x.shape) == 2) and (x.shape[0] == x.shape[1])):
raise ValueError(_msg1 % x.shape)
if isspmatrix(x):
x = x.tocsr()
else:
x = csr_matrix(x)
label = np.empty((shape[0],), dtype=x.indptr.dtype)
n_comp = _cs_graph_components(shape[0], x.indptr, x.indices, label)
return n_comp, label
def transpose(self, copy=False):
from csr import csr_matrix
M, N = self.shape
return csr_matrix((self.data, self.indices, self.indptr), (N, M),
copy=copy)
def kron(A, B, format=None):
"""kronecker product of sparse matrices A and B
Parameters
----------
A : sparse or dense matrix
first matrix of the product
B : sparse or dense matrix
second matrix of the product
format : string
format of the result (e.g. "csr")
Returns
-------
kronecker product in a sparse matrix format
Examples
--------
>>> A = csr_matrix(array([[0,2],[5,0]]))
>>> B = csr_matrix(array([[1,2],[3,4]]))
>>> kron(A,B).todense()
matrix([[ 0, 0, 2, 4],
[ 0, 0, 6, 8],
[ 5, 10, 0, 0],
[15, 20, 0, 0]])
>>> kron(A,[[1,2],[3,4]]).todense()
matrix([[ 0, 0, 2, 4],
[ 0, 0, 6, 8],
[ 5, 10, 0, 0],
[15, 20, 0, 0]])
"""
B = coo_matrix(B)
if (format is None or format == "bsr") and 2*B.nnz >= B.shape[0] * B.shape[1]:
#B is fairly dense, use BSR
A = csr_matrix(A,copy=True)
output_shape = (A.shape[0]*B.shape[0], A.shape[1]*B.shape[1])
if A.nnz == 0 or B.nnz == 0:
# kronecker product is the zero matrix
return coo_matrix( output_shape )
B = B.toarray()
data = A.data.repeat(B.size).reshape(-1,B.shape[0],B.shape[1])
data = data * B
return bsr_matrix((data,A.indices,A.indptr), shape=output_shape)
else:
#use COO
A = coo_matrix(A)
output_shape = (A.shape[0]*B.shape[0], A.shape[1]*B.shape[1])
if A.nnz == 0 or B.nnz == 0:
# kronecker product is the zero matrix
return coo_matrix( output_shape )
# expand entries of a into blocks
row = A.row.repeat(B.nnz)
col = A.col.repeat(B.nnz)
data = A.data.repeat(B.nnz)
row *= B.shape[0]
col *= B.shape[1]
# increment block indices
row,col = row.reshape(-1,B.nnz),col.reshape(-1,B.nnz)
row += B.row
col += B.col
row,col = row.reshape(-1),col.reshape(-1)
# compute block entries
data = data.reshape(-1,B.nnz) * B.data
data = data.reshape(-1)
return coo_matrix((data,(row,col)), shape=output_shape).asformat(format)
def kron(A, B, format=None):
"""kronecker product of sparse matrices A and B
Parameters
----------
A : sparse or dense matrix
first matrix of the product
B : sparse or dense matrix
second matrix of the product
format : string
format of the result (e.g. "csr")
Returns
-------
kronecker product in a sparse matrix format
Examples
--------
>>> A = csr_matrix(array([[0,2],[5,0]]))
>>> B = csr_matrix(array([[1,2],[3,4]]))
>>> kron(A,B).todense()
matrix([[ 0, 0, 2, 4],
[ 0, 0, 6, 8],
[ 5, 10, 0, 0],
[15, 20, 0, 0]])
>>> kron(A,[[1,2],[3,4]]).todense()
matrix([[ 0, 0, 2, 4],
[ 0, 0, 6, 8],
[ 5, 10, 0, 0],
[15, 20, 0, 0]])
"""
B = coo_matrix(B)
if (format is None
or format == "bsr") and 2 * B.nnz >= B.shape[0] * B.shape[1]:
#B is fairly dense, use BSR
A = csr_matrix(A, copy=True)
output_shape = (A.shape[0] * B.shape[0], A.shape[1] * B.shape[1])
if A.nnz == 0 or B.nnz == 0:
# kronecker product is the zero matrix
return coo_matrix(output_shape)
B = B.toarray()
data = A.data.repeat(B.size).reshape(-1, B.shape[0], B.shape[1])
data = data * B
return bsr_matrix((data, A.indices, A.indptr), shape=output_shape)
else:
#use COO
A = coo_matrix(A)
output_shape = (A.shape[0] * B.shape[0], A.shape[1] * B.shape[1])
if A.nnz == 0 or B.nnz == 0:
# kronecker product is the zero matrix
return coo_matrix(output_shape)
# expand entries of a into blocks
row = A.row.repeat(B.nnz)
col = A.col.repeat(B.nnz)
data = A.data.repeat(B.nnz)
row *= B.shape[0]
col *= B.shape[1]
# increment block indices
row, col = row.reshape(-1, B.nnz), col.reshape(-1, B.nnz)
row += B.row
col += B.col
row, col = row.reshape(-1), col.reshape(-1)
# compute block entries
data = data.reshape(-1, B.nnz) * B.data
data = data.reshape(-1)
return coo_matrix((data, (row, col)),
shape=output_shape).asformat(format)
