def syr2(X, u, v, alpha=1.0, beta=1.0, reordered=False):
r"""
Computes the projected rank 2 update of a cspmatrix X
.. math::
X := \alpha*P(u v^T + v u^T) + \beta X.
"""
assert X.is_factor is False, "cspmatrix factor object"
symb = X.symb
n = symb.n
snptr = symb.snptr
snode = symb.snode
blkval = X.blkval
blkptr = symb.blkptr
relptr = symb.relptr
snrowidx = symb.snrowidx
sncolptr = symb.sncolptr
if symb.p is not None and reordered is False:
up = u[symb.p]
vp = v[symb.p]
else:
up = u
vp = v
for k in range(symb.Nsn):
nn = snptr[k + 1] - snptr[k]
na = relptr[k + 1] - relptr[k]
nj = na + nn
for i in range(nn):
blas.scal(beta, blkval, n=nj - i, offset=blkptr[k] + (nj + 1) * i)
uk = up[snrowidx[sncolptr[k]:sncolptr[k + 1]]]
vk = vp[snrowidx[sncolptr[k]:sncolptr[k + 1]]]
blas.syr2(uk, vk, blkval, n=nn, offsetA=blkptr[k], ldA=nj, alpha=alpha)
blas.ger(uk,
vk,
blkval,
m=na,
n=nn,
offsetx=nn,
offsetA=blkptr[k] + nn,
ldA=nj,
alpha=alpha)
blas.ger(vk,
uk,
blkval,
m=na,
n=nn,
offsetx=nn,
offsetA=blkptr[k] + nn,
ldA=nj,
alpha=alpha)
return
def syr2(X, u, v, alpha = 1.0, beta = 1.0, reordered=False):
r"""
Computes the projected rank 2 update of a cspmatrix X
.. math::
X := \alpha*P(u v^T + v u^T) + \beta X.
"""
assert X.is_factor is False, "cspmatrix factor object"
symb = X.symb
n = symb.n
snptr = symb.snptr
snode = symb.snode
blkval = X.blkval
blkptr = symb.blkptr
relptr = symb.relptr
snrowidx = symb.snrowidx
sncolptr = symb.sncolptr
if symb.p is not None and reordered is False:
up = u[symb.p]
vp = v[symb.p]
else:
up = u
vp = v
for k in range(symb.Nsn):
nn = snptr[k+1]-snptr[k]
na = relptr[k+1]-relptr[k]
nj = na + nn
for i in range(nn): blas.scal(beta, blkval, n = nj-i, offset = blkptr[k]+(nj+1)*i)
uk = up[snrowidx[sncolptr[k]:sncolptr[k+1]]]
vk = vp[snrowidx[sncolptr[k]:sncolptr[k+1]]]
blas.syr2(uk, vk, blkval, n = nn, offsetA = blkptr[k], ldA = nj, alpha = alpha)
blas.ger(uk, vk, blkval, m = na, n = nn, offsetx = nn, offsetA = blkptr[k]+nn, ldA = nj, alpha = alpha)
blas.ger(vk, uk, blkval, m = na, n = nn, offsetx = nn, offsetA = blkptr[k]+nn, ldA = nj, alpha = alpha)
return
def kernel_matrix(X,
kernel,
sigma=1.0,
theta=1.0,
degree=1,
V=None,
width=None):
"""
Computes the kernel matrix or a partial kernel matrix.
Input arguments.
X is an N x n matrix.
kernel is a string with values 'linear', 'rfb', 'poly', or 'tanh'.
'linear': k(u,v) = u'*v/sigma.
'rbf': k(u,v) = exp(-||u - v||^2 / (2*sigma)).
'poly': k(u,v) = (u'*v/sigma)**degree.
'tanh': k(u,v) = tanh(u'*v/sigma - theta). kernel is a
sigma and theta are positive numbers.
degree is a positive integer.
V is an N x N sparse matrix (default is None).
width is a positive integer (default is None).
Output.
Q, an N x N matrix or sparse matrix.
If V is a sparse matrix, a partial kernel matrix with the sparsity
pattern V is returned.
If width is specified and V = 'band', a partial kernel matrix
with band sparsity is returned (width is the half-bandwidth).
a, an N x 1 matrix with the products <xi,xi>/sigma.
"""
N, n = X.size
#### dense (full) kernel matrix
if V is None:
if verbose: print("building kernel matrix ..")
# Qij = xi'*xj / sigma
Q = matrix(0.0, (N, N))
blas.syrk(X, Q, alpha=1.0 / sigma)
a = Q[::N + 1] # ai = ||xi||**2 / sigma
if kernel == 'linear':
pass
elif kernel == 'rbf':
# Qij := Qij - 0.5 * ( ai + aj )
# = -||xi - xj||^2 / (2*sigma)
ones = matrix(1.0, (N, 1))
blas.syr2(a, ones, Q, alpha=-0.5)
Q = exp(Q)
elif kernel == 'tanh':
Q = exp(Q - theta)
Q = div(Q - Q**-1, Q + Q**-1)
elif kernel == 'poly':
Q = Q**degree
else:
raise ValueError('invalid kernel type')
#### general sparse partial kernel matrix
elif type(V) is cvxopt.base.spmatrix:
if verbose: print("building projected kernel matrix ...")
Q = +V
base.syrk(X, Q, partial=True, alpha=1.0 / sigma)
# ai = ||xi||**2 / sigma
a = matrix(Q[::N + 1], (N, 1))
if kernel == 'linear':
pass
elif kernel == 'rbf':
ones = matrix(1.0, (N, 1))
# Qij := Qij - 0.5 * ( ai + aj )
# = -||xi - xj||^2 / (2*sigma)
p = chompack.maxcardsearch(V)
symb = chompack.symbolic(Q, p)
Qc = chompack.cspmatrix(symb) + Q
chompack.syr2(Qc, a, ones, alpha=-0.5)
Q = Qc.spmatrix(reordered=False)
Q.V = exp(Q.V)
elif kernel == 'tanh':
v = +Q.V
v = exp(v - theta)
v = div(v - v**-1, v + v**-1)
Q.V = v
elif kernel == 'poly':
Q.V = Q.V**degree
else:
raise ValueError('invalid kernel type')
#### banded partial kernel matrix
elif V == 'band' and width is not None:
# Lower triangular part of band matrix with bandwidth 2*w+1.
if verbose: print("building projected kernel matrix ...")
I = [i for k in range(N) for i in range(k, min(width + k + 1, N))]
J = [k for k in range(N) for i in range(min(width + 1, N - k))]
V = matrix(0.0, (len(I), 1))
oy = 0
for k in range(N): # V[:,k] = Xtrain[k:k+w, :] * Xtrain[k,:].T
m = min(width + 1, N - k)
blas.gemv(X,
X,
V,
m=m,
ldA=N,
incx=N,
offsetA=k,
offsetx=k,
offsety=oy)
oy += m
blas.scal(1.0 / sigma, V)
# ai = ||xi||**2 / sigma
a = matrix(V[[i for i in range(len(I)) if I[i] == J[i]]], (N, 1))
if kernel == 'linear':
Q = spmatrix(V, I, J, (N, N))
elif kernel == 'rbf':
Q = spmatrix(V, I, J, (N, N))
ones = matrix(1.0, (N, 1))
# Qij := Qij - 0.5 * ( ai + aj )
# = -||xi - xj||^2 / (2*sigma)
symb = chompack.symbolic(Q)
Qc = chompack.cspmatrix(symb) + Q
chompack.syr2(Qc, a, ones, alpha=-0.5)
Q = Qc.spmatrix(reordered=False)
Q.V = exp(Q.V)
elif kernel == 'tanh':
V = exp(V - theta)
V = div(V - V**-1, V + V**-1)
Q = spmatrix(V, I, J, (N, N))
elif kernel == 'poly':
Q = spmatrix(V**degree, I, J, (N, N))
else:
raise ValueError('invalid kernel type')
else:
raise TypeError('invalid type V')
return Q, a
def edmcompletion(A, reordered = True, **kwargs):
"""
Euclidean distance matrix completion. The routine takes an EDM-completable
cspmatrix :math:`A` and returns a dense EDM :math:`X`
that satisfies
.. math::
P( X ) = A
:param A: :py:class:`cspmatrix`
:param reordered: boolean
"""
assert isinstance(A, cspmatrix) and A.is_factor is False, "A must be a cspmatrix"
tol = kwargs.get('tol',1e-15)
X = matrix(A.spmatrix(reordered = True, symmetric = True))
symb = A.symb
n = symb.n
snptr = symb.snptr
sncolptr = symb.sncolptr
snrowidx = symb.snrowidx
# visit supernodes in reverse (descending) order
for k in range(symb.Nsn-1,-1,-1):
nn = snptr[k+1]-snptr[k]
beta = snrowidx[sncolptr[k]:sncolptr[k+1]]
nj = len(beta)
if nj-nn == 0: continue
alpha = beta[nn:]
nu = beta[:nn]
eta = matrix([matrix(range(beta[kk]+1,beta[kk+1])) for kk in range(nj-1)] + [matrix(range(beta[-1]+1,n))])
ne = len(eta)
# Compute Yaa, Yan, Yea, Ynn, Yee
Yaa = -0.5*X[alpha,alpha] - 0.5*X[alpha[0],alpha[0]]
blas.syr2(X[alpha,alpha[0]], matrix(1.0,(nj-nn,1)), Yaa, alpha = 0.5)
Ynn = -0.5*X[nu,nu] - 0.5*X[alpha[0],alpha[0]]
blas.syr2(X[nu,alpha[0]], matrix(1.0,(nn,1)), Ynn, alpha = 0.5)
Yee = -0.5*X[eta,eta] - 0.5*X[alpha[0],alpha[0]]
blas.syr2(X[eta,alpha[0]], matrix(1.0,(ne,1)), Yee, alpha = 0.5)
Yan = -0.5*X[alpha,nu] - 0.5*X[alpha[0],alpha[0]]
Yan += 0.5*matrix(1.0,(nj-nn,1))*X[alpha[0],nu]
Yan += 0.5*X[alpha,alpha[0]]*matrix(1.0,(1,nn))
Yea = -0.5*X[eta,alpha] - 0.5*X[alpha[0],alpha[0]]
Yea += 0.5*matrix(1.0,(ne,1))*X[alpha[0],alpha]
Yea += 0.5*X[eta,alpha[0]]*matrix(1.0,(1,nj-nn))
# EVD: Yaa = Z*diag(w)*Z.T
w = matrix(0.0,(Yaa.size[0],1))
Z = matrix(0.0,Yaa.size)
lapack.syevr(Yaa, w, jobz='V', range='A', uplo='L', Z=Z)
# Pseudo-inverse: Yp = pinv(Yaa)
lambda_max = max(w)
Yp = Z*spmatrix([1.0/wi if wi > lambda_max*tol else 0.0 for wi in w],range(len(w)),range(len(w)))*Z.T
# Compute update
tmp = -2.0*Yea*Yp*Yan + matrix(1.0,(ne,1))*Ynn[::nn+1].T + Yee[::ne+1]*matrix(1.0,(1,nn))
X[eta,nu] = tmp
X[nu,eta] = tmp.T
if reordered:
return X
else:
return X[symb.ip,symb.ip]
