1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
|
"""
Wrappers to LAPACK library
==========================
NOTE: this module is deprecated -- use scipy.linalg.lapack instead!
flapack -- wrappers for Fortran [*] LAPACK routines
clapack -- wrappers for ATLAS LAPACK routines
calc_lwork -- calculate optimal lwork parameters
get_lapack_funcs -- query for wrapper functions.
[*] If ATLAS libraries are available then Fortran routines
actually use ATLAS routines and should perform equally
well to ATLAS routines.
Module flapack
++++++++++++++
In the following all function names are shown without
type prefix (s,d,c,z). Optimal values for lwork can
be computed using calc_lwork module.
Linear Equations
----------------
Drivers::
lu,piv,x,info = gesv(a,b,overwrite_a=0,overwrite_b=0)
lub,piv,x,info = gbsv(kl,ku,ab,b,overwrite_ab=0,overwrite_b=0)
c,x,info = posv(a,b,lower=0,overwrite_a=0,overwrite_b=0)
Computational routines::
lu,piv,info = getrf(a,overwrite_a=0)
x,info = getrs(lu,piv,b,trans=0,overwrite_b=0)
inv_a,info = getri(lu,piv,lwork=min_lwork,overwrite_lu=0)
c,info = potrf(a,lower=0,clean=1,overwrite_a=0)
x,info = potrs(c,b,lower=0,overwrite_b=0)
inv_a,info = potri(c,lower=0,overwrite_c=0)
inv_c,info = trtri(c,lower=0,unitdiag=0,overwrite_c=0)
Linear Least Squares (LLS) Problems
-----------------------------------
Drivers::
v,x,s,rank,info = gelss(a,b,cond=-1.0,lwork=min_lwork,overwrite_a=0,overwrite_b=0)
Computational routines::
qr,tau,info = geqrf(a,lwork=min_lwork,overwrite_a=0)
q,info = orgqr|ungqr(qr,tau,lwork=min_lwork,overwrite_qr=0,overwrite_tau=1)
Generalized Linear Least Squares (LSE and GLM) Problems
-------------------------------------------------------
Standard Eigenvalue and Singular Value Problems
-----------------------------------------------
Drivers::
w,v,info = syev|heev(a,compute_v=1,lower=0,lwork=min_lwork,overwrite_a=0)
w,v,info = syevd|heevd(a,compute_v=1,lower=0,lwork=min_lwork,overwrite_a=0)
w,v,info = syevr|heevr(a,compute_v=1,lower=0,vrange=,irange=,atol=-1.0,lwork=min_lwork,overwrite_a=0)
t,sdim,(wr,wi|w),vs,info = gees(select,a,compute_v=1,sort_t=0,lwork=min_lwork,select_extra_args=(),overwrite_a=0)
wr,(wi,vl|w),vr,info = geev(a,compute_vl=1,compute_vr=1,lwork=min_lwork,overwrite_a=0)
u,s,vt,info = gesdd(a,compute_uv=1,lwork=min_lwork,overwrite_a=0)
Computational routines::
ht,tau,info = gehrd(a,lo=0,hi=n-1,lwork=min_lwork,overwrite_a=0)
ba,lo,hi,pivscale,info = gebal(a,scale=0,permute=0,overwrite_a=0)
Generalized Eigenvalue and Singular Value Problems
--------------------------------------------------
Drivers::
w,v,info = sygv|hegv(a,b,itype=1,compute_v=1,lower=0,lwork=min_lwork,overwrite_a=0,overwrite_b=0)
w,v,info = sygvd|hegvd(a,b,itype=1,compute_v=1,lower=0,lwork=min_lwork,overwrite_a=0,overwrite_b=0)
(alphar,alphai|alpha),beta,vl,vr,info = ggev(a,b,compute_vl=1,compute_vr=1,lwork=min_lwork,overwrite_a=0,overwrite_b=0)
Auxiliary routines
------------------
a,info = lauum(c,lower=0,overwrite_c=0)
a = laswp(a,piv,k1=0,k2=len(piv)-1,off=0,inc=1,overwrite_a=0)
Module clapack
++++++++++++++
Linear Equations
----------------
Drivers::
lu,piv,x,info = gesv(a,b,rowmajor=1,overwrite_a=0,overwrite_b=0)
c,x,info = posv(a,b,lower=0,rowmajor=1,overwrite_a=0,overwrite_b=0)
Computational routines::
lu,piv,info = getrf(a,rowmajor=1,overwrite_a=0)
x,info = getrs(lu,piv,b,trans=0,rowmajor=1,overwrite_b=0)
inv_a,info = getri(lu,piv,rowmajor=1,overwrite_lu=0)
c,info = potrf(a,lower=0,clean=1,rowmajor=1,overwrite_a=0)
x,info = potrs(c,b,lower=0,rowmajor=1,overwrite_b=0)
inv_a,info = potri(c,lower=0,rowmajor=1,overwrite_c=0)
inv_c,info = trtri(c,lower=0,unitdiag=0,rowmajor=1,overwrite_c=0)
Auxiliary routines
------------------
a,info = lauum(c,lower=0,rowmajor=1,overwrite_c=0)
Module calc_lwork
+++++++++++++++++
Optimal lwork is maxwrk. Default is minwrk.
minwrk,maxwrk = gehrd(prefix,n,lo=0,hi=n-1)
minwrk,maxwrk = gesdd(prefix,m,n,compute_uv=1)
minwrk,maxwrk = gelss(prefix,m,n,nrhs)
minwrk,maxwrk = getri(prefix,n)
minwrk,maxwrk = geev(prefix,n,compute_vl=1,compute_vr=1)
minwrk,maxwrk = heev(prefix,n,lower=0)
minwrk,maxwrk = syev(prefix,n,lower=0)
minwrk,maxwrk = gees(prefix,n,compute_v=1)
minwrk,maxwrk = geqrf(prefix,m,n)
minwrk,maxwrk = gqr(prefix,m,n)
"""
from __future__ import division, print_function, absolute_import
__all__ = ['get_lapack_funcs','calc_lwork','flapack','clapack']
from numpy import deprecate
from . import calc_lwork
# The following ensures that possibly missing flavor (C or Fortran) is
# replaced with the available one. If none is available, exception
# is raised at the first attempt to use the resources.
@deprecate(old_name="scipy.lib.lapack", new_name="scipy.linalg.lapack")
def _deprecated():
pass
try:
_deprecated()
except DeprecationWarning as e:
# don't fail import if DeprecationWarnings raise error -- works around
# the situation with Numpy's test framework
pass
from . import flapack
from . import clapack
_use_force_clapack = 1
if hasattr(clapack,'empty_module'):
clapack = flapack
_use_force_clapack = 0
elif hasattr(flapack,'empty_module'):
flapack = clapack
_type_conv = {'f':'s', 'd':'d', 'F':'c', 'D':'z'} # 'd' will be default for 'i',..
_inv_type_conv = {'s':'f','d':'d','c':'F','z':'D'}
@deprecate
def get_lapack_funcs(names,arrays=(),debug=0,force_clapack=1):
"""Return available LAPACK function objects with names.
arrays are used to determine the optimal prefix of
LAPACK routines.
If force_clapack is True then available Atlas routine
is returned for column major storaged arrays with
rowmajor argument set to False.
"""
force_clapack = 0 # XXX: Don't set it true! The feature is unreliable
# and may cause incorrect results.
# See test_basic.test_solve.check_20Feb04_bug.
ordering = []
for i in range(len(arrays)):
t = arrays[i].dtype.char
if t not in _type_conv:
t = 'd'
ordering.append((t,i))
if ordering:
ordering.sort()
required_prefix = _type_conv[ordering[0][0]]
else:
required_prefix = 'd'
dtypechar = _inv_type_conv[required_prefix]
# Default lookup:
if ordering and arrays[ordering[0][1]].flags['FORTRAN']:
# prefer Fortran code for leading array with column major order
m1,m2 = flapack,clapack
else:
# in all other cases, C code is preferred
m1,m2 = clapack,flapack
if not _use_force_clapack:
force_clapack = 0
funcs = []
m1_name = m1.__name__.split('.')[-1]
m2_name = m2.__name__.split('.')[-1]
for name in names:
func_name = required_prefix + name
func = getattr(m1,func_name,None)
if func is None:
func = getattr(m2,func_name)
func.module_name = m2_name
else:
func.module_name = m1_name
if force_clapack and m1 is flapack:
func2 = getattr(m2,func_name,None)
if func2 is not None:
import new
func_code = None # defined in exec
exec(_colmajor_func_template % {'func_name':func_name})
func = new.function(func_code,{'clapack_func':func2},func_name)
func.module_name = m2_name
func.__doc__ = func2.__doc__
func.prefix = required_prefix
func.dtypechar = dtypechar
funcs.append(func)
return tuple(funcs)
_colmajor_func_template = '''\
def %(func_name)s(*args,**kws):
if "rowmajor" not in kws:
kws["rowmajor"] = 0
return clapack_func(*args,**kws)
func_code = %(func_name)s.func_code
'''
from numpy.testing import Tester
test = Tester().test
|
