def reconstructLowRank(U, s, V, k):
"""
Take the SVD of a low rank matrix and partially compute it with at most
k values. If k is an array of values [0, U.shape[0]*V.shape[0]] then these
indices are used for reconstruction.
"""
(m, n) = (U.shape[0], V.shape[0])
if type(k) == numpy.ndarray:
inds = k
inds = numpy.unique(inds)
rowInds, colInds = numpy.unravel_index(inds, (m, n))
elif type(k) == tuple:
rowInds, colInds = k
else:
inds = numpy.random.randint(0, n * m, k)
inds = numpy.unique(inds)
rowInds, colInds = numpy.unravel_index(inds, (m, n))
U = numpy.ascontiguousarray(U)
V = numpy.ascontiguousarray(V)
X = SparseUtilsCython.partialReconstructPQ((rowInds, colInds), U * s,
V)
return X
def reconstructLowRankPQ(P, Q, inds):
"""
Given an array of unique indices inds in [0, U.shape[0]*V.shape[0]-1],
partially reconstruct $P*Q^T$. The returned matrix is a scipy csc_matrix.
"""
(m, n) = (P.shape[0], Q.shape[0])
if type(inds) == tuple:
rowInds, colInds = inds
rowInds = numpy.array(rowInds, numpy.int)
colInds = numpy.array(colInds, numpy.int)
else:
rowInds, colInds = numpy.unravel_index(inds, (m, n))
X = SparseUtilsCython.partialReconstructPQ((rowInds, colInds), P, Q)
return X
def reconstructLowRankPQ(P, Q, inds):
"""
Given an array of unique indices inds in [0, U.shape[0]*V.shape[0]-1],
partially reconstruct $P*Q^T$. The returned matrix is a scipy csc_matrix.
"""
(m, n) = (P.shape[0], Q.shape[0])
if type(inds) == tuple:
rowInds, colInds = inds
rowInds = numpy.array(rowInds, numpy.int)
colInds = numpy.array(colInds, numpy.int)
else:
rowInds, colInds = numpy.unravel_index(inds, (m, n))
X = SparseUtilsCython.partialReconstructPQ((rowInds, colInds), P, Q)
return X
def reconstructLowRank(U, s, V, k):
"""
Take the SVD of a low rank matrix and partially compute it with at most
k values. If k is an array of values [0, U.shape[0]*V.shape[0]] then these
indices are used for reconstruction.
"""
(m, n) = (U.shape[0], V.shape[0])
if type(k) == numpy.ndarray:
inds = k
inds = numpy.unique(inds)
rowInds, colInds = numpy.unravel_index(inds, (m, n))
elif type(k) == tuple:
rowInds, colInds = k
else:
inds = numpy.random.randint(0, n*m, k)
inds = numpy.unique(inds)
rowInds, colInds = numpy.unravel_index(inds, (m, n))
U = numpy.ascontiguousarray(U)
V = numpy.ascontiguousarray(V)
X = SparseUtilsCython.partialReconstructPQ((rowInds, colInds), U*s, V)
return X
def next(self):
X = self.XIterator.next()
logging.debug("Learning on matrix with shape: " +
str(X.shape) + " and " + str(X.nnz) +
" non-zeros")
if self.iterativeSoftImpute.weighted:
#Compute row and col probabilities
up, vp = SparseUtils.nonzeroRowColsProbs(X)
nzuInds = up == 0
nzvInds = vp == 0
u = numpy.sqrt(1 / (up + numpy.array(nzuInds, numpy.int)))
v = numpy.sqrt(1 / (vp + numpy.array(nzvInds, numpy.int)))
u[nzuInds] = 0
v[nzvInds] = 0
if self.rhos != None:
self.iterativeSoftImpute.setRho(self.rhos.next())
if not scipy.sparse.isspmatrix_csc(X):
raise ValueError("X must be a csc_matrix not " +
str(type(X)))
#Figure out what lambda should be
#PROPACK has problems with convergence
Y = scipy.sparse.csc_matrix(X, dtype=numpy.float)
U, s, V = ExpSU.SparseUtils.svdArpack(Y, 1, kmax=20)
del Y
#U, s, V = SparseUtils.svdPropack(X, 1, kmax=20)
maxS = s[0]
logging.debug("Largest singular value : " + str(maxS))
(n, m) = X.shape
if self.j == 0:
self.oldU = numpy.zeros((n, 1))
self.oldS = numpy.zeros(1)
self.oldV = numpy.zeros((m, 1))
else:
oldN = self.oldU.shape[0]
oldM = self.oldV.shape[0]
if self.iterativeSoftImpute.updateAlg == "initial":
if n > oldN:
self.oldU = Util.extendArray(
self.oldU, (n, self.oldU.shape[1]))
elif n < oldN:
self.oldU = self.oldU[0:n, :]
if m > oldM:
self.oldV = Util.extendArray(
self.oldV, (m, self.oldV.shape[1]))
elif m < oldN:
self.oldV = self.oldV[0:m, :]
elif self.iterativeSoftImpute.updateAlg == "zero":
self.oldU = numpy.zeros((n, 1))
self.oldS = numpy.zeros(1)
self.oldV = numpy.zeros((m, 1))
else:
raise ValueError("Unknown SVD update algorithm: " +
self.updateAlg)
rowInds, colInds = X.nonzero()
gamma = self.iterativeSoftImpute.eps + 1
i = 0
self.iterativeSoftImpute.measures = numpy.zeros(
(self.iterativeSoftImpute.maxIterations, 4))
while gamma > self.iterativeSoftImpute.eps:
if i == self.iterativeSoftImpute.maxIterations:
logging.debug("Maximum number of iterations reached")
break
ZOmega = SparseUtilsCython.partialReconstructPQ(
(rowInds, colInds), self.oldU * self.oldS, self.oldV)
Y = X - ZOmega
#Y = Y.tocsc()
#del ZOmega
Y = csarray(Y, storagetype="row")
gc.collect()
#os.system('taskset -p 0xffffffff %d' % os.getpid())
if self.iterativeSoftImpute.svdAlg == "propack":
L = LinOperatorUtils.sparseLowRankOp(Y,
self.oldU,
self.oldS,
self.oldV,
parallel=False)
newU, newS, newV = SparseUtils.svdPropack(
L,
k=self.iterativeSoftImpute.k,
kmax=self.iterativeSoftImpute.kmax)
elif self.iterativeSoftImpute.svdAlg == "arpack":
L = LinOperatorUtils.sparseLowRankOp(Y,
self.oldU,
self.oldS,
self.oldV,
parallel=False)
newU, newS, newV = SparseUtils.svdArpack(
L,
k=self.iterativeSoftImpute.k,
kmax=self.iterativeSoftImpute.kmax)
elif self.iterativeSoftImpute.svdAlg == "svdUpdate":
newU, newS, newV = SVDUpdate.addSparseProjected(
self.oldU, self.oldS, self.oldV, Y,
self.iterativeSoftImpute.k)
elif self.iterativeSoftImpute.svdAlg == "rsvd":
L = LinOperatorUtils.sparseLowRankOp(Y,
self.oldU,
self.oldS,
self.oldV,
parallel=True)
newU, newS, newV = RandomisedSVD.svd(
L,
self.iterativeSoftImpute.k,
p=self.iterativeSoftImpute.p,
q=self.iterativeSoftImpute.q)
elif self.iterativeSoftImpute.svdAlg == "rsvdUpdate":
L = LinOperatorUtils.sparseLowRankOp(Y,
self.oldU,
self.oldS,
self.oldV,
parallel=True)
if self.j == 0:
newU, newS, newV = RandomisedSVD.svd(
L,
self.iterativeSoftImpute.k,
p=self.iterativeSoftImpute.p,
q=self.iterativeSoftImpute.q)
else:
newU, newS, newV = RandomisedSVD.svd(
L,
self.iterativeSoftImpute.k,
p=self.iterativeSoftImpute.p,
q=self.iterativeSoftImpute.qu,
omega=self.oldV)
elif self.iterativeSoftImpute.svdAlg == "rsvdUpdate2":
if self.j == 0:
L = LinOperatorUtils.sparseLowRankOp(Y,
self.oldU,
self.oldS,
self.oldV,
parallel=True)
newU, newS, newV = RandomisedSVD.svd(
L,
self.iterativeSoftImpute.k,
p=self.iterativeSoftImpute.p,
q=self.iterativeSoftImpute.q)
else:
#Need linear operator which is U s V
L = LinOperatorUtils.lowRankOp(
self.oldU, self.oldS, self.oldV)
Y = GeneralLinearOperator.asLinearOperator(
Y, parallel=True)
newU, newS, newV = RandomisedSVD.updateSvd(
L,
self.oldU,
self.oldS,
self.oldV,
Y,
self.iterativeSoftImpute.k,
p=self.iterativeSoftImpute.p)
else:
raise ValueError("Unknown SVD algorithm: " +
self.iterativeSoftImpute.svdAlg)
if self.iterativeSoftImpute.weighted and i == 0:
delta = numpy.diag((u * newU.T).dot(newU))
pi = numpy.diag((v * newV.T).dot(newV))
lmbda = (maxS / numpy.max(
delta * pi)) * self.iterativeSoftImpute.rho
lmbdav = lmbda * delta * pi
elif not self.iterativeSoftImpute.weighted:
lmbda = maxS * self.iterativeSoftImpute.rho
if i == 0:
logging.debug("lambda: " + str(lmbda))
lmbdav = lmbda
newS = newS - lmbdav
#Soft threshold
newS = numpy.clip(newS, 0, numpy.max(newS))
normOldZ = (self.oldS**2).sum()
normNewZmOldZ = (self.oldS**2).sum() + (
newS**2).sum() - 2 * numpy.trace(
(self.oldV.T.dot(newV * newS)).dot(
newU.T.dot(self.oldU * self.oldS)))
#We can get newZ == oldZ in which case we break
if normNewZmOldZ < self.tol:
gamma = 0
elif abs(normOldZ) < self.tol:
gamma = self.iterativeSoftImpute.eps + 1
else:
gamma = normNewZmOldZ / normOldZ
if self.iterativeSoftImpute.verbose:
theta1 = (
self.iterativeSoftImpute.k -
numpy.linalg.norm(self.oldU.T.dot(newU), 'fro')**
2) / self.iterativeSoftImpute.k
theta2 = (
self.iterativeSoftImpute.k -
numpy.linalg.norm(self.oldV.T.dot(newV), 'fro')**
2) / self.iterativeSoftImpute.k
thetaS = numpy.linalg.norm(
newS - self.oldS)**2 / numpy.linalg.norm(newS)**2
self.iterativeSoftImpute.measures[i, :] = numpy.array(
[gamma, theta1, theta2, thetaS])
self.oldU = newU.copy()
self.oldS = newS.copy()
self.oldV = newV.copy()
logging.debug("Iteration " + str(i) + " gamma=" +
str(gamma))
i += 1
if self.iterativeSoftImpute.postProcess:
#Add the mean vectors
previousS = newS
newU = numpy.c_[newU, numpy.array(X.mean(1)).ravel()]
newV = numpy.c_[newV, numpy.array(X.mean(0)).ravel()]
newS = self.iterativeSoftImpute.unshrink(X, newU, newV)
#Note that this increases the rank of U and V by 1
#print("Difference in s after postprocessing: " + str(numpy.linalg.norm(previousS - newS[0:-1])))
logging.debug("Difference in s after postprocessing: " +
str(numpy.linalg.norm(previousS -
newS[0:-1])))
logging.debug("Number of iterations for rho=" +
str(self.iterativeSoftImpute.rho) + ": " +
str(i))
self.j += 1
return (newU, newS, newV)
def learnModel(self, X, fullMatrices=True):
"""
Learn the matrix completion using a sparse matrix X. This is the simple
version of the soft impute algorithm in which we store the entire
matrices, newZ and oldZ.
"""
if not scipy.sparse.isspmatrix_csc(X):
raise ValueError("Input matrix must be csc_matrix")
(n, m) = X.shape
oldU = numpy.zeros((n, 1))
oldS = numpy.zeros(1)
oldV = numpy.zeros((m, 1))
omega = X.nonzero()
tol = 10**-6
rowInds = numpy.array(omega[0], numpy.int)
colInds = numpy.array(omega[1], numpy.int)
ZList = []
for rho in self.rhos:
gamma = self.eps + 1
i = 0
Y = scipy.sparse.csc_matrix(X, dtype=numpy.float)
U, s, V = ExpSU.SparseUtils.svdArpack(Y, 1, kmax=20)
lmbda = rho*numpy.max(s)
while gamma > self.eps:
ZOmega = SparseUtilsCython.partialReconstructPQ((rowInds, colInds), oldU*oldS, oldV)
Y = X - ZOmega
Y = Y.tocsc()
newU, newS, newV = ExpSU.SparseUtils.svdSparseLowRank(Y, oldU, oldS, oldV)
#Soft threshold
newS = newS - lmbda
newS = numpy.clip(newS, 0, numpy.max(newS))
normOldZ = (oldS**2).sum()
normNewZmOldZ = (oldS**2).sum() + (newS**2).sum() - 2*numpy.trace((oldV.T.dot(newV*newS)).dot(newU.T.dot(oldU*oldS)))
#We can get newZ == oldZ in which case we break
if normNewZmOldZ < tol:
gamma = 0
elif abs(normOldZ) < tol:
gamma = self.eps + 1
else:
gamma = normNewZmOldZ/normOldZ
oldU = newU.copy()
oldS = newS.copy()
oldV = newV.copy()
logging.debug("Iteration " + str(i) + " gamma="+str(gamma))
i += 1
logging.debug("Number of iterations for lambda="+str(rho) + ": " + str(i))
if fullMatrices:
newZ = scipy.sparse.lil_matrix((newU*newS).dot(newV.T))
ZList.append(newZ)
else:
ZList.append((newU,newS,newV))
if self.rhos.shape[0] != 1:
return ZList
else:
return ZList[0]
def next(self):
X = self.XIterator.next()
logging.debug("Learning on matrix with shape: " + str(X.shape) + " and " + str(X.nnz) + " non-zeros")
if self.iterativeSoftImpute.weighted:
#Compute row and col probabilities
up, vp = SparseUtils.nonzeroRowColsProbs(X)
nzuInds = up==0
nzvInds = vp==0
u = numpy.sqrt(1/(up + numpy.array(nzuInds, numpy.int)))
v = numpy.sqrt(1/(vp + numpy.array(nzvInds, numpy.int)))
u[nzuInds] = 0
v[nzvInds] = 0
if self.rhos != None:
self.iterativeSoftImpute.setRho(self.rhos.next())
if not scipy.sparse.isspmatrix_csc(X):
raise ValueError("X must be a csc_matrix not " + str(type(X)))
#Figure out what lambda should be
#PROPACK has problems with convergence
Y = scipy.sparse.csc_matrix(X, dtype=numpy.float)
U, s, V = ExpSU.SparseUtils.svdArpack(Y, 1, kmax=20)
del Y
#U, s, V = SparseUtils.svdPropack(X, 1, kmax=20)
maxS = s[0]
logging.debug("Largest singular value : " + str(maxS))
(n, m) = X.shape
if self.j == 0:
self.oldU = numpy.zeros((n, 1))
self.oldS = numpy.zeros(1)
self.oldV = numpy.zeros((m, 1))
else:
oldN = self.oldU.shape[0]
oldM = self.oldV.shape[0]
if self.iterativeSoftImpute.updateAlg == "initial":
if n > oldN:
self.oldU = Util.extendArray(self.oldU, (n, self.oldU.shape[1]))
elif n < oldN:
self.oldU = self.oldU[0:n, :]
if m > oldM:
self.oldV = Util.extendArray(self.oldV, (m, self.oldV.shape[1]))
elif m < oldN:
self.oldV = self.oldV[0:m, :]
elif self.iterativeSoftImpute.updateAlg == "zero":
self.oldU = numpy.zeros((n, 1))
self.oldS = numpy.zeros(1)
self.oldV = numpy.zeros((m, 1))
else:
raise ValueError("Unknown SVD update algorithm: " + self.updateAlg)
rowInds, colInds = X.nonzero()
gamma = self.iterativeSoftImpute.eps + 1
i = 0
self.iterativeSoftImpute.measures = numpy.zeros((self.iterativeSoftImpute.maxIterations, 4))
while gamma > self.iterativeSoftImpute.eps:
if i == self.iterativeSoftImpute.maxIterations:
logging.debug("Maximum number of iterations reached")
break
ZOmega = SparseUtilsCython.partialReconstructPQ((rowInds, colInds), self.oldU*self.oldS, self.oldV)
Y = X - ZOmega
#Y = Y.tocsc()
#del ZOmega
Y = csarray(Y, storagetype="row")
gc.collect()
#os.system('taskset -p 0xffffffff %d' % os.getpid())
if self.iterativeSoftImpute.svdAlg=="propack":
L = LinOperatorUtils.sparseLowRankOp(Y, self.oldU, self.oldS, self.oldV, parallel=False)
newU, newS, newV = SparseUtils.svdPropack(L, k=self.iterativeSoftImpute.k, kmax=self.iterativeSoftImpute.kmax)
elif self.iterativeSoftImpute.svdAlg=="arpack":
L = LinOperatorUtils.sparseLowRankOp(Y, self.oldU, self.oldS, self.oldV, parallel=False)
newU, newS, newV = SparseUtils.svdArpack(L, k=self.iterativeSoftImpute.k, kmax=self.iterativeSoftImpute.kmax)
elif self.iterativeSoftImpute.svdAlg=="svdUpdate":
newU, newS, newV = SVDUpdate.addSparseProjected(self.oldU, self.oldS, self.oldV, Y, self.iterativeSoftImpute.k)
elif self.iterativeSoftImpute.svdAlg=="rsvd":
L = LinOperatorUtils.sparseLowRankOp(Y, self.oldU, self.oldS, self.oldV, parallel=True)
newU, newS, newV = RandomisedSVD.svd(L, self.iterativeSoftImpute.k, p=self.iterativeSoftImpute.p, q=self.iterativeSoftImpute.q)
elif self.iterativeSoftImpute.svdAlg=="rsvdUpdate":
L = LinOperatorUtils.sparseLowRankOp(Y, self.oldU, self.oldS, self.oldV, parallel=True)
if self.j == 0:
newU, newS, newV = RandomisedSVD.svd(L, self.iterativeSoftImpute.k, p=self.iterativeSoftImpute.p, q=self.iterativeSoftImpute.q)
else:
newU, newS, newV = RandomisedSVD.svd(L, self.iterativeSoftImpute.k, p=self.iterativeSoftImpute.p, q=self.iterativeSoftImpute.qu, omega=self.oldV)
elif self.iterativeSoftImpute.svdAlg=="rsvdUpdate2":
if self.j == 0:
L = LinOperatorUtils.sparseLowRankOp(Y, self.oldU, self.oldS, self.oldV, parallel=True)
newU, newS, newV = RandomisedSVD.svd(L, self.iterativeSoftImpute.k, p=self.iterativeSoftImpute.p, q=self.iterativeSoftImpute.q)
else:
#Need linear operator which is U s V
L = LinOperatorUtils.lowRankOp(self.oldU, self.oldS, self.oldV)
Y = GeneralLinearOperator.asLinearOperator(Y, parallel=True)
newU, newS, newV = RandomisedSVD.updateSvd(L, self.oldU, self.oldS, self.oldV, Y, self.iterativeSoftImpute.k, p=self.iterativeSoftImpute.p)
else:
raise ValueError("Unknown SVD algorithm: " + self.iterativeSoftImpute.svdAlg)
if self.iterativeSoftImpute.weighted and i==0:
delta = numpy.diag((u*newU.T).dot(newU))
pi = numpy.diag((v*newV.T).dot(newV))
lmbda = (maxS/numpy.max(delta*pi))*self.iterativeSoftImpute.rho
lmbdav = lmbda*delta*pi
elif not self.iterativeSoftImpute.weighted:
lmbda = maxS*self.iterativeSoftImpute.rho
if i==0:
logging.debug("lambda: " + str(lmbda))
lmbdav = lmbda
newS = newS - lmbdav
#Soft threshold
newS = numpy.clip(newS, 0, numpy.max(newS))
normOldZ = (self.oldS**2).sum()
normNewZmOldZ = (self.oldS**2).sum() + (newS**2).sum() - 2*numpy.trace((self.oldV.T.dot(newV*newS)).dot(newU.T.dot(self.oldU*self.oldS)))
#We can get newZ == oldZ in which case we break
if normNewZmOldZ < self.tol:
gamma = 0
elif abs(normOldZ) < self.tol:
gamma = self.iterativeSoftImpute.eps + 1
else:
gamma = normNewZmOldZ/normOldZ
if self.iterativeSoftImpute.verbose:
theta1 = (self.iterativeSoftImpute.k - numpy.linalg.norm(self.oldU.T.dot(newU), 'fro')**2)/self.iterativeSoftImpute.k
theta2 = (self.iterativeSoftImpute.k - numpy.linalg.norm(self.oldV.T.dot(newV), 'fro')**2)/self.iterativeSoftImpute.k
thetaS = numpy.linalg.norm(newS - self.oldS)**2/numpy.linalg.norm(newS)**2
self.iterativeSoftImpute.measures[i, :] = numpy.array([gamma, theta1, theta2, thetaS])
self.oldU = newU.copy()
self.oldS = newS.copy()
self.oldV = newV.copy()
logging.debug("Iteration " + str(i) + " gamma="+str(gamma))
i += 1
if self.iterativeSoftImpute.postProcess:
#Add the mean vectors
previousS = newS
newU = numpy.c_[newU, numpy.array(X.mean(1)).ravel()]
newV = numpy.c_[newV, numpy.array(X.mean(0)).ravel()]
newS = self.iterativeSoftImpute.unshrink(X, newU, newV)
#Note that this increases the rank of U and V by 1
#print("Difference in s after postprocessing: " + str(numpy.linalg.norm(previousS - newS[0:-1])))
logging.debug("Difference in s after postprocessing: " + str(numpy.linalg.norm(previousS - newS[0:-1])))
logging.debug("Number of iterations for rho="+str(self.iterativeSoftImpute.rho) + ": " + str(i))
self.j += 1
return (newU, newS, newV)
