Function that only accepts list as input and returning dictionary of statistic measures as output

Question

I need some help with the following problem:

I've got most of the code right, just missing something with the input filtering and 3rd test input (nested list input/output). First two outputting as they should.

Function Specs:

list as input. If the function does not receive a list, it must raise a TypeError. If the length of the list is smaller than or equal to 1, raise a ValueError. Function must return a dict, whose keys represent the sample statistics of the input with the following format:

If the input list contains multiple rows, the function should report on the mean/std/max/etc of each row in the form of a list.

All numerical values should be rounded to 2 decimal places

HINT: Use numpy functions with the axis keyword. When doing an unbiased estimate, try googling the ddof keyword. The axis keyword should be either 0, 1, or -1. One of these will work better than the other.

Expected Outputs:

stats([1, 1, 1, 1]) == {'mean': 1, 'std': 0, 'min': 1, 'median', 1, 'max': 1}
stats([1, 2, 2, 3, 4]) == {'mean': 2.4, 'std': 1.14, 'min': 1, 'median': 2.0, 'max': 4}
stats([[1, 2], [3, 4]]) == {
    'mean': [1.5, 3.5],
    'std': [0.71, 0.71],
    'min': [1, 3],
    'median': [1.5, 3.5],
    'max': [2, 4]
}

Error code received:

PASSED     question_5_mean
PASSED     question_5_mean
FAILED     question_5_mean
           Inputs: [[[1, 2], [3, 4]]]
           ValueError: The truth value of an array with more than one element is ambiguous. Use a.any() or a.all()
PASSED     question_5_mean
PASSED     question_5_ValueError
PASSED     question_5_ValueError
FAILED     question_5_TypeError
           Inputs: ['string input']
           def test_question_5_TypeError(test_input):
FAILED     question_5_TypeError
           Inputs: [{'dict': 1, 'input': 2}]
           def test_question_5_TypeError(test_input):

Answer 1

Try this,

def stats(lst):
  if type(lst) != list:
    raise TypeError("Please enter a list.")
  elif len(lst) <= 1:
    raise ValueError("Please enter a list with more than 1 elements.")

  arr = np.array(lst)
  statistics = lambda row:(row.mean(), np.median(row), row.std(), row.max(), row.min())
  data = [[] for i in range(5)]
  
  for row in range(len(arr)):
    for i, item in enumerate(statistics(arr[row, :])):
      data[i].append(item)

  return dict(zip(("mean", "median", "std", "max", "min"), data))

stats([[1,2],[3,4]])

Output -

{'max': [2, 4],
 'mean': [1.5, 3.5],
 'median': [1.5, 3.5],
 'min': [1, 3],
 'std': [0.5, 0.5]}