Python | Numpy np.lagpow()方法
原文:https://www . geesforgeks . org/python-numpy-NP-lag pow-method/
**np.lagpow()**
方法用于将拉盖尔级数提升到给定的幂。它将拉盖尔系列 c
提升为功率。
语法:
np.lagpow(c, pow, maxpower=16)
参数: c:【array _ like】从低到高排序的拉盖尔级数系数一维数组。 功率:【整数】该系列将要提升到的功率。 最大功率:【整数,可选】最大允许功率。默认值为 16。回归:【ndarray】拉盖尔系异能。
代码#1 :
# Python program explaining
# numpy.lagpow() method
# importing numpy as np
# and numpy.polynomial.laguerre module as geek
import numpy as np
import numpy.polynomial.laguerre as geek
# Input laguerre series coefficients
s = (2, 4, 8)
# using np.lagpow() method
res = geek.lagpow(s, pow = 3)
# Resulting laguerre series coefficient
print (res)
Output:
[ 3176\. -25680\. 100512\. -206592\. 246528\. -161280\. 46080.]
代码#2 :
# Python program explaining
# numpy.lagpow() method
# importing numpy as np
# and numpy.polynomial.laguerre module as geek
import numpy as np
import numpy.polynomial.laguerre as geek
# Laguerre series coefficients
s = (1, 2, 3, 4, 5)
# using np.lagpow() method
res = geek.lagpow(s, pow = 4)
# Resulting laguerre series
print (res)
Output:
[ 1.88268525e+08 -4.12583040e+09 4.24806624e+10 -2.72789533e+11
1.22285197e+12 -4.05794076e+12 1.03122236e+13 -2.04726003e+13
3.20914270e+13 -3.98546302e+13 3.90872742e+13 -2.99578594e+13
1.75922875e+13 -7.65283919e+12 2.32607876e+12 -4.41441000e+11
3.94143750e+10]