如何反转 scipy.interpolate.interp1d 行为

Question

From the documentation of scipy.interpolate.interp1d:

x and y are arrays of values used to approximate some function f: y = f(x). This class returns a function whose call method uses interpolation to find the value of new points.

Parameters

x(N,) array_like: A 1-D array of real values.

y(…,N,…) array_like: A ND array of real values. The length of y along the interpolation axis must be equal to the length of x.

So basically, it assumes that I have some x for which I have calculated multiple y=f(x) that I want to interpolate. But my case is exactly opposite: I have a single y array that I want to interpolate for many different x arrays. In other words, I have the situation:

import numpy as np
from scipy.interpolate import interp1d

n_to_interpolate = 10

x = np.cumsum(np.random.rand(n_to_interpolate,100),axis=1)
y = np.linspace(0,1,100)

interp1d(x,y)

But this gives an error because the shapes do not match. Is there a way to achieve what I want?

Answer 1

You'd need to loop over the 1d slices of the x array.

If you only need a linear interpolation, np.interp is faster then interp1d