/
orthogonal_eval.pyx
269 lines (209 loc) · 6.5 KB
/
orthogonal_eval.pyx
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
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
"""
Evaluate orthogonal polynomial values using recurrence relations
or by calling special functions.
References
----------
.. [AMS55] Abramowitz & Stegun, Section 22.5.
.. [MH] Mason & Handscombe, Chebyshev Polynomials, CRC Press (2003).
"""
#
# Copyright (C) 2009 Pauli Virtanen
# Distributed under the same license as Scipy.
#
#------------------------------------------------------------------------------
# Direct evaluation of polynomials
#------------------------------------------------------------------------------
cdef extern from "math.h":
double sqrt(double x) nogil
cdef double eval_poly_chebyt(long k, double x) nogil:
# Use Chebyshev T recurrence directly, see [MH]
cdef long m
cdef double b2, b1, b0
b2 = 0
b1 = -1
b0 = 0
x = 2*x
for m in range(k+1):
b2 = b1
b1 = b0
b0 = x*b1 - b2
return (b0 - b2)/2.0
#------------------------------------------------------------------------------
# Ufunc boilerplate
#------------------------------------------------------------------------------
cdef extern from "numpy/arrayobject.h":
void import_array()
ctypedef int npy_intp
cdef enum NPY_TYPES:
NPY_LONG
NPY_DOUBLE
cdef extern from "numpy/ufuncobject.h":
void import_ufunc()
ctypedef void (*PyUFuncGenericFunction)(char**, npy_intp*, npy_intp*, void*)
object PyUFunc_FromFuncAndData(PyUFuncGenericFunction* func, void** data,
char* types, int ntypes, int nin, int nout,
int identity, char* name, char* doc, int c)
cdef void _loop_id_d(char **args, npy_intp *dimensions, npy_intp *steps,
void *func) nogil:
cdef int i
cdef double x
cdef char *ip1=args[0], *ip2=args[1], *op=args[2]
for i in range(0, dimensions[0]):
(<double*>op)[0] = (<double(*)(long,double) nogil>func)(
(<long*>ip1)[0], (<double*>ip2)[0])
ip1 += steps[0]; ip2 += steps[1]; op += steps[2]
cdef char _id_d_types[3]
cdef PyUFuncGenericFunction _id_d_funcs[1]
_id_d_types[0] = NPY_LONG
_id_d_types[1] = NPY_DOUBLE
_id_d_types[2] = NPY_DOUBLE
_id_d_funcs[0] = _loop_id_d
import_array()
import_ufunc()
#--
cdef void *chebyt_data[1]
chebyt_data[0] = <void*>eval_poly_chebyt
_eval_chebyt = PyUFunc_FromFuncAndData(_id_d_funcs, chebyt_data,
_id_d_types, 1, 2, 1, 0, "", "", 0)
#------------------------------------------------------------------------------
# Actual evaluation functions
#------------------------------------------------------------------------------
import numpy as np
from scipy.special._cephes import gamma, hyp2f1, hyp1f1, gammaln
from numpy import exp
def binom(n, k):
"""
binom(n, k)
Binomial coefficient
"""
return np.exp(gammaln(1+n) - gammaln(1+k) - gammaln(1+n-k))
def eval_jacobi(n, alpha, beta, x, out=None):
"""
eval_jacobi(n, alpha, beta, x, out=None)
Evaluate Jacobi polynomial at a point.
"""
d = binom(n+alpha, n)
a = -n
b = n + alpha + beta + 1
c = alpha + 1
g = (1-x)/2.0
return hyp2f1(a, b, c, g) * d
def eval_sh_jacobi(n, p, q, x, out=None):
"""
eval_sh_jacobi(n, p, q, x, out=None)
Evaluate shifted Jacobi polynomial at a point.
"""
factor = np.exp(gammaln(1+n) + gammaln(n+p) - gammaln(2*n+p))
return factor * eval_jacobi(n, p-q, q-1, 2*x-1)
def eval_gegenbauer(n, alpha, x, out=None):
"""
eval_gegenbauer(n, alpha, x, out=None)
Evaluate Gegenbauer polynomial at a point.
"""
d = gamma(n+2*alpha)/gamma(1+n)/gamma(2*alpha)
a = -n
b = n + 2*alpha
c = alpha + 0.5
g = (1-x)/2.0
return hyp2f1(a, b, c, g) * d
def eval_chebyt(n, x, out=None):
"""
eval_chebyt(n, x, out=None)
Evaluate Chebyshev T polynomial at a point.
This routine is numerically stable for `x` in ``[-1, 1]`` at least
up to order ``10000``.
"""
return _eval_chebyt(n, x, out)
def eval_chebyu(n, x, out=None):
"""
eval_chebyu(n, x, out=None)
Evaluate Chebyshev U polynomial at a point.
"""
d = n+1
a = -n
b = n+2
c = 1.5
g = (1-x)/2.0
return hyp2f1(a, b, c, g) * d
def eval_chebys(n, x, out=None):
"""
eval_chebys(n, x, out=None)
Evaluate Chebyshev S polynomial at a point.
"""
return eval_chebyu(n, x/2, out=out)
def eval_chebyc(n, x, out=None):
"""
eval_chebyc(n, x, out=None)
Evaluate Chebyshev C polynomial at a point.
"""
return 2*eval_chebyt(n, x/2.0, out)
def eval_sh_chebyt(n, x, out=None):
"""
eval_sh_chebyt(n, x, out=None)
Evaluate shifted Chebyshev T polynomial at a point.
"""
return eval_chebyt(n, 2*x-1, out=out)
def eval_sh_chebyu(n, x, out=None):
"""
eval_sh_chebyu(n, x, out=None)
Evaluate shifted Chebyshev U polynomial at a point.
"""
return eval_chebyu(n, 2*x-1, out=out)
def eval_legendre(n, x, out=None):
"""
eval_legendre(n, x, out=None)
Evaluate Legendre polynomial at a point.
"""
d = 1
a = -n
b = n+1
c = 1
g = (1-x)/2.0
return hyp2f1(a, b, c, g) * d
def eval_sh_legendre(n, x, out=None):
"""
eval_sh_legendre(n, x, out=None)
Evaluate shifted Legendre polynomial at a point.
"""
return eval_legendre(n, 2*x-1, out=out)
def eval_genlaguerre(n, alpha, x, out=None):
"""
eval_genlaguerre(n, alpha, x, out=None)
Evaluate generalized Laguerre polynomial at a point.
"""
d = binom(n+alpha, n)
a = -n
b = alpha + 1
g = x
return hyp1f1(a, b, g) * d
def eval_laguerre(n, x, out=None):
"""
eval_laguerre(n, x, out=None)
Evaluate Laguerre polynomial at a point.
"""
return eval_genlaguerre(n, 0., x, out=out)
def eval_hermite(n, x, out=None):
"""
eval_hermite(n, x, out=None)
Evaluate Hermite polynomial at a point.
"""
n, x = np.broadcast_arrays(n, x)
n, x = np.atleast_1d(n, x)
if out is None:
out = np.zeros_like(0*n + 0*x)
if (n % 1 != 0).any():
raise ValueError("Order must be integer")
even = (n % 2 == 0)
m = n[even]/2
out[even] = ((-1)**m * 2**(2*m) * gamma(1+m)
* eval_genlaguerre(m, -0.5, x[even]**2))
m = (n[~even]-1)/2
out[~even] = ((-1)**m * 2**(2*m+1) * gamma(1+m)
* x[~even] * eval_genlaguerre(m, 0.5, x[~even]**2))
return out
def eval_hermitenorm(n, x, out=None):
"""
eval_hermitenorm(n, x, out=None)
Evaluate normalized Hermite polynomial at a point.
"""
return eval_hermite(n, x/sqrt(2)) * 2**(-n/2.0)