]>
gitweb.michael.orlitzky.com - sage.d.git/blob - mjo/eja/eja_utils.py
1 from sage
.functions
.other
import sqrt
2 from sage
.matrix
.constructor
import matrix
3 from sage
.modules
.free_module_element
import vector
7 Scale the vector, matrix, or cartesian-product-of-those-things
10 This works around the inability to scale certain elements of
11 Cartesian product spaces, as reported in
13 https://trac.sagemath.org/ticket/31435
17 This will do the wrong thing if you feed it a tuple or list.
21 sage: from mjo.eja.eja_utils import _scale
25 sage: v = vector(QQ, (1,2,3))
28 sage: m = matrix(QQ, [[1,2],[3,4]])
29 sage: M = cartesian_product([m.parent(), m.parent()])
30 sage: _scale(M((m,m)), 2)
36 if hasattr(x
, 'cartesian_factors'):
38 return P(tuple( _scale(x_i
, alpha
)
39 for x_i
in x
.cartesian_factors() ))
46 Flatten a vector, matrix, or cartesian product of those things
49 If the entries of the matrix themselves belong to a real vector
50 space (such as the complex numbers which can be thought of as
51 pairs of real numbers), they will also be expanded in vector form
52 and flattened into the list.
56 sage: from mjo.eja.eja_utils import _all2list
57 sage: from mjo.octonions import Octonions, OctonionMatrixAlgebra
61 sage: V1 = VectorSpace(QQ,2)
62 sage: V2 = MatrixSpace(QQ,2)
66 sage: y2 = V2([0,1,1,0])
67 sage: _all2list((x1,y1))
69 sage: _all2list((x2,y2))
71 sage: M = cartesian_product([V1,V2])
72 sage: _all2list(M((x1,y1)))
74 sage: _all2list(M((x2,y2)))
79 sage: _all2list(Octonions().one())
80 [1, 0, 0, 0, 0, 0, 0, 0]
81 sage: _all2list(OctonionMatrixAlgebra(1).one())
82 [1, 0, 0, 0, 0, 0, 0, 0]
85 if hasattr(x
, 'to_vector'):
86 # This works on matrices of e.g. octonions directly, without
87 # first needing to convert them to a list of octonions and
88 # then recursing down into the list. It also avoids the wonky
89 # list(x) when x is an element of a CFM. I don't know what it
90 # returns but it aint the coordinates. This will fall through
91 # to the iterable case the next time around.
92 return _all2list(x
.to_vector())
96 except TypeError: # x is not iterable
100 # Avoid the retardation of list(QQ(1)) == [1].
103 return sum(list( map(_all2list
, xl
) ), [])
108 return vector(m
.base_ring(), m
.list())
111 return matrix(v
.base_ring(), sqrt(v
.degree()), v
.list())
113 def gram_schmidt(v
, inner_product
=None):
115 Perform Gram-Schmidt on the list ``v`` which are assumed to be
116 vectors over the same base ring. Returns a list of orthonormalized
117 vectors over the smallest extention ring containing the necessary
122 sage: from mjo.eja.eja_utils import gram_schmidt
126 The usual inner-product and norm are default::
128 sage: v1 = vector(QQ,(1,2,3))
129 sage: v2 = vector(QQ,(1,-1,6))
130 sage: v3 = vector(QQ,(2,1,-1))
132 sage: u = gram_schmidt(v)
133 sage: all( u_i.inner_product(u_i).sqrt() == 1 for u_i in u )
135 sage: bool(u[0].inner_product(u[1]) == 0)
137 sage: bool(u[0].inner_product(u[2]) == 0)
139 sage: bool(u[1].inner_product(u[2]) == 0)
143 But if you supply a custom inner product, the result is
144 orthonormal with respect to that (and not the usual inner
147 sage: v1 = vector(QQ,(1,2,3))
148 sage: v2 = vector(QQ,(1,-1,6))
149 sage: v3 = vector(QQ,(2,1,-1))
151 sage: B = matrix(QQ, [ [6, 4, 2],
154 sage: ip = lambda x,y: (B*x).inner_product(y)
155 sage: norm = lambda x: ip(x,x)
156 sage: u = gram_schmidt(v,ip)
157 sage: all( norm(u_i) == 1 for u_i in u )
159 sage: ip(u[0],u[1]).is_zero()
161 sage: ip(u[0],u[2]).is_zero()
163 sage: ip(u[1],u[2]).is_zero()
166 This Gram-Schmidt routine can be used on matrices as well, so long
167 as an appropriate inner-product is provided::
169 sage: E11 = matrix(QQ, [ [1,0],
171 sage: E12 = matrix(QQ, [ [0,1],
173 sage: E22 = matrix(QQ, [ [0,0],
175 sage: I = matrix.identity(QQ,2)
176 sage: trace_ip = lambda X,Y: (X*Y).trace()
177 sage: gram_schmidt([E11,E12,I,E22], inner_product=trace_ip)
179 [1 0] [ 0 1/2*sqrt(2)] [0 0]
180 [0 0], [1/2*sqrt(2) 0], [0 1]
183 It even works on Cartesian product spaces whose factors are vector
186 sage: V1 = VectorSpace(AA,2)
187 sage: V2 = MatrixSpace(AA,2)
188 sage: M = cartesian_product([V1,V2])
190 sage: x2 = V1([1,-1])
192 sage: y2 = V2([0,1,1,0])
193 sage: z1 = M((x1,y1))
194 sage: z2 = M((x2,y2))
196 ....: return a[0].inner_product(b[0]) + (a[1]*b[1]).trace()
197 sage: U = gram_schmidt([z1,z2], inner_product=ip)
207 Ensure that zero vectors don't get in the way::
209 sage: v1 = vector(QQ,(1,2,3))
210 sage: v2 = vector(QQ,(1,-1,6))
211 sage: v3 = vector(QQ,(0,0,0))
213 sage: len(gram_schmidt(v)) == 2
216 if inner_product
is None:
217 inner_product
= lambda x
,y
: x
.inner_product(y
)
218 norm
= lambda x
: inner_product(x
,x
).sqrt()
220 v
= list(v
) # make a copy, don't clobber the input
222 # Drop all zero vectors before we start.
223 v
= [ v_i
for v_i
in v
if not v_i
.is_zero() ]
231 # Our "zero" needs to belong to the right space for sum() to work.
232 zero
= v
[0].parent().zero()
235 if hasattr(v
[0], 'cartesian_factors'):
236 # Only use the slow implementation if necessary.
240 return sc(x
, (inner_product(x
,y
)/inner_product(x
,x
)))
242 # First orthogonalize...
243 for i
in range(1,len(v
)):
244 # Earlier vectors can be made into zero so we have to ignore them.
245 v
[i
] -= sum( (proj(v
[j
],v
[i
])
247 if not v
[j
].is_zero() ),
250 # And now drop all zero vectors again if they were "orthogonalized out."
251 v
= [ v_i
for v_i
in v
if not v_i
.is_zero() ]
253 # Just normalize. If the algebra is missing the roots, we can't add
254 # them here because then our subalgebra would have a bigger field
255 # than the superalgebra.
256 for i
in range(len(v
)):
257 v
[i
] = sc(v
[i
], ~
norm(v
[i
]))