Squeezing Dimensions¶

The np.squeeze function removes all size-1 dimensions from an array.

np.squeeze Function¶

Removes axes of length one from the array shape.

1. Basic Usage¶

import numpy as np

def main():
    x = np.random.normal(size=(1, 3, 1, 2, 1, 5, 1))
    y = np.squeeze(x)
    print(f"{x.shape = }")
    print(f"{y.shape = }")

if __name__ == "__main__":
    main()

Output:

x.shape = (1, 3, 1, 2, 1, 5, 1)
y.shape = (3, 2, 5)

2. All Size-1 Removed¶

Every dimension with size 1 is eliminated from the result.

Selective Squeeze¶

Remove only specific size-1 axes using the axis parameter.

1. Single Axis¶

import numpy as np

def main():
    x = np.zeros((1, 3, 1, 4))
    y = np.squeeze(x, axis=0)
    print(f"{x.shape = }")
    print(f"{y.shape = }")

if __name__ == "__main__":
    main()

Output:

x.shape = (1, 3, 1, 4)
y.shape = (3, 1, 4)

2. Multiple Axes¶

import numpy as np

def main():
    x = np.zeros((1, 3, 1, 4))
    y = np.squeeze(x, axis=(0, 2))
    print(f"{x.shape = }")
    print(f"{y.shape = }")

if __name__ == "__main__":
    main()

Output:

x.shape = (1, 3, 1, 4)
y.shape = (3, 4)

Error Handling¶

Squeezing non-singleton dimensions raises an error.

1. ValueError Example¶

import numpy as np

def main():
    x = np.zeros((1, 3, 1, 4))
    try:
        y = np.squeeze(x, axis=1)
    except ValueError as e:
        print(f"Error: {e}")

if __name__ == "__main__":
    main()

2. Safe Practice¶

Only squeeze axes you know have size 1.

Common Use Cases¶

Squeeze is frequently needed after certain operations.

1. After Slicing¶

Slicing with a single index reduces dimensionality; squeeze cleans up.

2. Model Outputs¶

Neural network outputs often have extra batch or channel dimensions.

3. Broadcasting Result¶

Broadcasting may introduce size-1 dimensions that need removal.