NumPy ndarray.compress() Method

NumPy ndarray.compress()

The numpy.ndarray.compress() method is used to filter elements of an array along a specified axis based on a boolean condition. It returns a new array containing only the elements where the condition is True.

Syntax

ndarray.compress(condition, axis=None, out=None)

Parameters

Parameter	Type	Description
condition	1D array of bool	A boolean array determining which elements to keep. The length must match the size of the specified axis.
axis	int, optional	The axis along which filtering is applied. If None, the condition is applied to the flattened array.
out	ndarray, optional	Optional output array to store results. Must have a compatible shape.

Return Value

Returns a new NumPy array containing only elements where the condition is True. If an axis is specified, elements are selected along that axis.

---

Examples

1. Filtering Elements from a 1D Array

In this example, we filter values from a one-dimensional array based on a boolean condition.

import numpy as np

# Create a 1D array
arr = np.array([10, 20, 30, 40, 50])

# Define a condition: Keep elements greater than 25
condition = arr > 25

# Apply compress function
result = arr.compress(condition)

# Print the result
print(result)  # Output: [30 40 50]

Output:

[30 40 50]

Since the condition is applied element-wise, only elements greater than 25 are included in the output.

2. Filtering Elements Along a Specific Axis

Here, we apply compress() along an axis in a 2D array.

import numpy as np

# Create a 2D array
arr = np.array([[1, 2, 3], 
                [4, 5, 6], 
                [7, 8, 9]])

# Define a boolean condition for rows
condition = np.array([True, False, True])

# Apply compress along axis 0 (row-wise)
result = arr.compress(condition, axis=0)

# Print the result
print(result)

Output:

[[1 2 3]
 [7 8 9]]

Since condition = [True, False, True], the function keeps only the first and third rows, discarding the second row.

3. Filtering Columns Using the axis Parameter

We use compress() to filter specific columns from a 2D array.

import numpy as np

# Create a 2D array
arr = np.array([[10, 20, 30], 
                [40, 50, 60], 
                [70, 80, 90]])

# Define a boolean condition for columns
condition = np.array([False, True, True])

# Apply compress along axis 1 (column-wise)
result = arr.compress(condition, axis=1)

# Print the result
print(result)

Output:

[[20 30]
 [50 60]
 [80 90]]

Here, we filter columns based on condition = [False, True, True], keeping only the second and third columns.

4. Using compress() on a Flattened Array

If no axis is specified, the method operates on the flattened array.

import numpy as np

# Create a 2D array
arr = np.array([[1, 2, 3], 
                [4, 5, 6], 
                [7, 8, 9]])

# Define a condition on the flattened array
condition = arr.flatten() % 2 == 0  # Keep only even numbers

# Apply compress without specifying an axis
result = arr.compress(condition)

# Print the result
print(result)

Output:

[2 4 6 8]

Since compress() is applied without an axis, the entire array is flattened first, then filtered to include only even numbers.