Python numpy统计量计算函数单元测试失败求助

Let's break down what's going wrong here and fix it step by step.

The Root Cause of the TypeError #

Your calculate function is returning the result of multiple print() calls, and each print() returns None. So when you do return print(mean), print(variance), ..., you're actually returning a tuple of 6 None values. The unit tests expect a single dictionary containing all the statistics, so when assertAlmostEqual tries to subtract this tuple from the expected dict, it throws the TypeError you're seeing.

Modified Working Code #

Here's the fixed version of your function that will pass all the tests:

import numpy as np

def calculate(arr):
    if len(arr) != 9:
        raise ValueError('List must contain nine numbers.')
    
    arr = np.array(arr)
    new_arr = arr.reshape((3, 3))
    
    # Build a single stats dictionary matching the test's expected structure
    stats = {
        'mean': [np.mean(new_arr, 0).tolist(), np.mean(new_arr, 1).tolist(), np.mean(arr).tolist()],
        'variance': [np.var(new_arr, 0).tolist(), np.var(new_arr, 1).tolist(), np.var(arr).tolist()],
        'standard deviation': [np.std(new_arr, 0).tolist(), np.std(new_arr, 1).tolist(), np.std(arr).tolist()],
        'max': [np.max(new_arr, 0).tolist(), np.max(new_arr, 1).tolist(), np.max(arr).tolist()],
        'min': [np.min(new_arr, 0).tolist(), np.min(new_arr, 1).tolist(), np.min(arr).tolist()],
        'sum': [np.sum(new_arr, 0).tolist(), np.sum(new_arr, 1).tolist(), np.sum(arr).tolist()]
    }
    
    # Optional: print for debugging, but don't return the print result
    print(stats)
    return stats

# Test the function directly if needed
calculate([0, 1, 2, 3, 4, 5, 6, 7, 8])

Key Changes Made #

• Replaced separate small dictionaries with one unified stats dictionary that exactly matches the structure the unit tests expect.
• Returned the stats dictionary instead of a tuple of print() results (since print() always returns None).
• Kept the print statement as an optional debugging tool without affecting the function's return value.

Recommended Learning Resources #

• NumPy Statistics: Master axis-based calculations and core stats functions with NumPy's official Statistics Routines documentation.
• Python Function Returns: Brush up on return value basics—remember print() doesn't return the value it displays, it returns None. This is covered in Python's Function Definitions tutorial.
• Unit Testing: Learn how assertAlmostEqual works (it compares numeric values or structures containing them) and other unittest methods via the official unittest documentation.

内容的提问来源于stack exchange，提问作者Doryk