本文整理汇总了Python中scipy.linalg.special_matrices.triu方法的典型用法代码示例。如果您正苦于以下问题：Python special_matrices.triu方法的具体用法？Python special_matrices.triu怎么用？Python special_matrices.triu使用的例子？那么恭喜您, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在类scipy.linalg.special_matrices的用法示例。
在下文中一共展示了special_matrices.triu方法的3个代码示例，这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞，您的评价将有助于系统推荐出更棒的Python代码示例。
示例1: triu_indices
# 需要导入模块: from scipy.linalg import special_matrices [as 别名]
# 或者: from scipy.linalg.special_matrices import triu [as 别名]
def triu_indices(n, k=0):
m = numpy.ones((n, n), int)
a = triu(m, k)
return numpy.where(a != 0) 示例2: qr_destroy
# 需要导入模块: from scipy.linalg import special_matrices [as 别名]
# 或者: from scipy.linalg.special_matrices import triu [as 别名]
def qr_destroy(la):
"""
Return QR decomposition of `la[0]`. Content of `la` gets destroyed in the process.
Using this function should be less memory intense than calling `scipy.linalg.qr(la[0])`,
because the memory used in `la[0]` is reclaimed earlier.
"""
a = numpy.asfortranarray(la[0])
del la[0], la # now `a` is the only reference to the input matrix
m, n = a.shape
# perform q, r = QR(a); code hacked out of scipy.linalg.qr
logger.debug("computing QR of %s dense matrix" % str(a.shape))
geqrf, = get_lapack_funcs(('geqrf',), (a,))
qr, tau, work, info = geqrf(a, lwork=-1, overwrite_a=True)
qr, tau, work, info = geqrf(a, lwork=work[0], overwrite_a=True)
del a # free up mem
assert info >= 0
r = triu(qr[:n, :n])
if m < n: # rare case, #features < #topics
qr = qr[:, :m] # retains fortran order
gorgqr, = get_lapack_funcs(('orgqr',), (qr,))
q, work, info = gorgqr(qr, tau, lwork=-1, overwrite_a=True)
q, work, info = gorgqr(qr, tau, lwork=work[0], overwrite_a=True)
assert info >= 0, "qr failed"
assert q.flags.f_contiguous
return q, r 示例3: qr_destroy
# 需要导入模块: from scipy.linalg import special_matrices [as 别名]
# 或者: from scipy.linalg.special_matrices import triu [as 别名]
def qr_destroy(la):
"""
Return QR decomposition of `la[0]`. Content of `la` gets destroyed in the process.
Using this function should be less memory intense than calling `scipy.linalg.qr(la[0])`,
because the memory used in `la[0]` is reclaimed earlier.
"""
a = numpy.asfortranarray(la[0])
del la[0], la # now `a` is the only reference to the input matrix
m, n = a.shape
# perform q, r = QR(a); code hacked out of scipy.linalg.qr
logger.debug("computing QR of %s dense matrix" % str(a.shape))
geqrf, = get_lapack_funcs(('geqrf',), (a,))
qr, tau, work, info = geqrf(a, lwork=-1, overwrite_a=True)
qr, tau, work, info = geqrf(a, lwork=work[0], overwrite_a=True)
del a # free up mem
assert info >= 0
r = triu(qr[:n, :n])
if m < n: # rare case, #features < #topics
qr = qr[:, :m] # retains fortran order
gorgqr, = get_lapack_funcs(('orgqr',), (qr,))
q, work, info = gorgqr(qr, tau, lwork=-1, overwrite_a=True)
q, work, info = gorgqr(qr, tau, lwork=work[0], overwrite_a=True)
assert info >= 0, "qr failed"
assert q.flags.f_contiguous
return q, r