How to Check if a NumPy Array Is Empty in Python (size vs len)

To check if a NumPy array is empty, test arr.size == 0. size is the total number of elements, so it is 0 only when the array holds nothing, whatever its shape. Don’t use if not arr: (it raises an error for arrays with more than one element), and be careful with len(arr), which only counts the first dimension. For a plain Python list, if not my_list: is the usual check.

Every example was run with Python 3.12.5 and NumPy 2.5.3; the output shown is from those runs.

import numpy as np

arr = np.array([])

if arr.size == 0:
    print("The array is empty")

Output:

The array is empty

size vs len() vs shape

import numpy as np

arrays = {
    "np.array([])": np.array([]),
    "np.array([5])": np.array([5]),
    "np.array([0, 0])": np.array([0, 0]),
    "np.zeros((0, 3))": np.zeros((0, 3)),
    "np.zeros((3, 0))": np.zeros((3, 0)),
    "np.zeros((2, 3))": np.zeros((2, 3)),
}
for label, a in arrays.items():
    print(f"{label:18} shape={str(a.shape):8} size={a.size:<2} len={len(a):<2} empty={a.size == 0}")

Output:

np.array([])       shape=(0,)     size=0  len=0  empty=True
np.array([5])      shape=(1,)     size=1  len=1  empty=False
np.array([0, 0])   shape=(2,)     size=2  len=2  empty=False
np.zeros((0, 3))   shape=(0, 3)   size=0  len=0  empty=True
np.zeros((3, 0))   shape=(3, 0)   size=0  len=3  empty=True
np.zeros((2, 3))   shape=(2, 3)   size=6  len=2  empty=False

The 3×0 array is the trap: it has 3 rows but no elements, so len() says 3 while size correctly says 0:

import numpy as np

a = np.zeros((3, 0))       # 3 rows, 0 columns: no elements at all
print("len(a):", len(a))   # counts rows only
print("a.size:", a.size)   # counts every element

Output:

len(a): 3
a.size: 0

Check	Empty 1-D array	3×0 array	Recommended?
arr.size == 0	True	True	Yes
len(arr) == 0	True	False	Only for 1-D arrays
arr.shape[0] == 0	True	False	Same problem as len()
not arr	Error	Error	No

Why if not arr doesn’t work

import numpy as np

arr = np.array([1, 2, 3])
if not arr:
    print("empty")

Output:

ValueError: The truth value of an array with more than one element is ambiguous. Use a.any() or a.all()

An array with several elements has no single True/False value, so NumPy refuses to guess. Recent NumPy versions also refuse an empty array:

import numpy as np

arr = np.array([])
print(bool(arr))

Output:

ValueError: The truth value of an empty array is ambiguous. Use `array.size > 0` to check that an array is not empty.

A one-element array is the confusing exception: it takes the truth value of that one element, so np.array([0]) is falsy even though it is not empty:

import numpy as np

print(bool(np.array([0])), bool(np.array([7])))   # one element: its own truth value

Output:

False True

There is no np.isempty()

Many people search for numpy isempty, but the function doesn’t exist:

import numpy as np

print(np.isempty(np.array([])))

Output:

AttributeError: module 'numpy' has no attribute 'isempty'. Did you mean: 'empty'?

Use arr.size == 0 (or np.size(arr) == 0) instead. size is documented in the NumPy reference for ndarray.size.

Check if a NumPy array is not empty

import numpy as np

scores = np.array([88, 92, 79])

if scores.size > 0:
    print("scores is not empty; mean =", scores.mean())

Output:

scores is not empty; mean = 86.33333333333333

Don’t confuse “not empty” with “has non-zero values”. any() answers the second question:

import numpy as np

zeros = np.zeros(3)
print("size:", zeros.size, "-> empty?", zeros.size == 0)
print("any():", zeros.any(), "-> any non-zero values?")

Output:

size: 3 -> empty? False
any(): False -> any non-zero values?

np.empty() does not create an empty array

np.empty(3) is a fast way to allocate 3 elements without setting their values. Only a size of 0 gives an empty array (more in NumPy empty array):

import numpy as np

a = np.empty(3)          # allocates 3 elements; does NOT make an empty array
print("np.empty(3) size:", a.size)

b = np.empty(0)          # this one really is empty
print("np.empty(0) size:", b.size, "| np.array([]) size:", np.array([]).size)

Output:

np.empty(3) size: 3
np.empty(0) size: 0 | np.array([]) size: 0

Practical examples

Filtering can return an empty array

import numpy as np

temps = np.array([21.5, 23.0, 19.8])
hot = temps[temps > 30]           # filtering can return an empty array

if hot.size == 0:
    print("No day was above 30 degrees")
else:
    print("Hot days:", hot)

Output:

No day was above 30 degrees

Many reductions fail on an empty array, so check before calling them:

import numpy as np

hot = np.array([])
print(hot.max())

Calling max() on an empty array:

ValueError: zero-size array to reduction operation maximum which has no identity

import numpy as np

def safe_max(a, default=None):
    return a.max() if a.size else default

print(safe_max(np.array([3, 9, 4])), safe_max(np.array([])))

Output:

9 None

Loop while an array is not empty

import numpy as np

queue = np.array([10, 20, 30])
while queue.size:                 # loop until the array is empty
    first, queue = queue[0], queue[1:]
    print("processing", first, "- left:", queue.size)

Output:

processing 10 - left: 2
processing 20 - left: 1
processing 30 - left: 0

An array full of NaN is not empty

import numpy as np

a = np.array([np.nan, np.nan])
print("empty:", a.size == 0)
print("all values missing:", np.isnan(a).all())
print("after dropping NaN, empty:", a[~np.isnan(a)].size == 0)

Output:

empty: False
all values missing: True
after dropping NaN, empty: True

Check if a Python list or array is empty

Python’s built-in list is often called an array. For lists, an empty list is falsy, so if not numbers: works (more in check if a list is empty in Python):

numbers = []                      # a Python list
if not numbers:
    print("list is empty")

numbers.append(5)
if numbers:
    print("list is not empty:", numbers)

Output:

list is empty
list is not empty: [5]

The standard-library array module behaves the same way:

from array import array

a = array("i")                    # array module: typed array of ints
print("empty:", not a, "| len:", len(a))
a.append(7)
print("empty:", not a, "| len:", len(a))

Output:

empty: True | len: 0
empty: False | len: 1

One check for any kind of array

np.size() works on lists, tuples and NumPy arrays alike:

from array import array
import numpy as np

def is_empty(x):
    """Works for NumPy arrays, lists, tuples and array.array."""
    return np.size(x) == 0

for x in [np.array([]), np.zeros((3, 0)), [], [1, 2], (), (0,), array("i"), array("i", [7])]:
    print(f"{x!r:40} -> {is_empty(x)}")

Output:

array([], dtype=float64)                 -> True
array([], shape=(3, 0), dtype=float64)   -> True
[]                                       -> True
[1, 2]                                   -> False
()                                       -> True
(0,)                                     -> False
array('i')                               -> True
array('i', [7])                          -> False

Keep exploring NumPy and empty checks:

• Create an empty NumPy array with np.empty()
• Create an empty array in Python
• Check if a list is empty in Python
• Check if a variable is empty or None

Frequently asked questions