Hey there, let's break down why this error is happening and fix it right away.

The core problem here is your s(t) function uses math.cos—a function from Python's built-in math module that only works with single scalar values. But you're passing it n1T, which is a full numpy array of time points. The math module doesn't support vectorized operations on arrays, so it throws that error when you feed it anything larger than a single number.

Here are the best solutions, ordered by how recommended they are for your signal processing use case:

1. Switch to Numpy's Trigonometric Functions (Highly Recommended) #

Numpy is built specifically for vectorized operations—its functions can handle entire arrays seamlessly. Since you already have numpy imported (from your from numpy import arange... line), just swap out math.cos and math.pi with their numpy equivalents:

import numpy as np  # Your existing imports already cover numpy, but adding this explicitly is safe

def s(t):
    return np.cos(2 * np.pi * 10 * t)

s1_n = s(n1T)  # This will work perfectly with your array now

This is the optimal approach because numpy's functions are optimized for speed, which is critical for signal processing tasks that often deal with large datasets.

2. Vectorize Your Existing Function (Quick Test Fix) #

If you don't want to rewrite your function, you can use np.vectorize to wrap it into a version that accepts arrays. Note this is just a convenience wrapper and won't be as fast as using numpy's native functions:

import numpy as np
from math import cos, pi

def s(t):
    return cos(2 * pi * 10 * t)

# Convert the scalar-only function to handle arrays
s_vectorized = np.vectorize(s)
s1_n = s_vectorized(n1T)

3. Iterate Over Array Elements (Not Recommended) #

You could manually loop through each element in n1T with a list comprehension, but this is inefficient for large arrays and defeats the purpose of using numpy:

s1_n = np.array([s(t) for t in n1T])

Stick with option 1 for the cleanest, most performant code in your signal processing workflow.

内容的提问来源于stack exchange，提问作者Sabet