np.expand_dims() and np.squeeze() in NumPy - Cheat Sheet & Quick Revision

Recall & Review

beginner

What does np.expand_dims() do to an array?

np.expand_dims() adds a new axis (dimension) to an array at the specified position. This increases the number of dimensions by one.

beginner

What is the purpose of np.squeeze()?

np.squeeze() removes axes of length 1 from an array, reducing its number of dimensions.

beginner

How would you add a new axis as the first dimension to a 1D array a using np.expand_dims()?

Use np.expand_dims(a, axis=0). This changes shape from (n,) to (1, n).

intermediate

If an array has shape (1, 5, 1), what will be its shape after applying np.squeeze()?

The shape will be (5,) because all axes with length 1 are removed.

intermediate

Can np.squeeze() remove a specific axis only?

Yes, by specifying the axis parameter, np.squeeze() removes the axis only if its length is 1. Otherwise, it raises an error.

What does np.expand_dims(arr, axis=1) do to a 1D array arr of shape (4,)?

A Adds a new axis at position 1, resulting in shape (4, 1)

B Adds a new axis at position 0, resulting in shape (1, 4)

C Removes all axes of length 1

D Does nothing

Adding a new axis at position 1 changes shape from (4,) to (4, 1).

What will be the shape after np.squeeze() is applied to an array of shape (1, 3, 1, 5)?

A (1, 3, 1, 5)

B (3, 5)

C (3, 1, 5)

D (1, 3, 5)

All axes with length 1 are removed, so (1, 3, 1, 5) becomes (3, 5).

Which function would you use to add a dimension to a 2D array to make it 3D?

A np.expand_dims()

B np.squeeze()

C np.reshape()

D np.flatten()

np.expand_dims() adds a new axis, increasing dimensions.

What happens if you try np.squeeze() on an axis that is not length 1 when specifying the axis parameter?

A It removes the axis anyway

B It adds a new axis

C It raises an error

D It ignores that axis and removes others

np.squeeze() raises an error if the specified axis is not length 1.

If you have a 1D array with shape (10,), how can you convert it to shape (10, 1) using np.expand_dims()?

A np.expand_dims(arr, axis=0)

B np.squeeze(arr)

C np.reshape(arr, (10, 1))

D np.expand_dims(arr, axis=1)

Adding a new axis at position 1 changes shape from (10,) to (10, 1).

Explain in your own words what np.expand_dims() and np.squeeze() do to an array's shape.

Think about adding or removing dimensions that have size 1.

Describe a real-life example where you might need to use np.expand_dims() or np.squeeze() when working with data.

Consider how images or time series data might need reshaping.