Python

Shape Manipulation

Learn how to transform array dimensions, flatten multi-dimensional data, and combine arrays using stacking.

By TechCoder TeamLast updated: 2026-06-02
In a Nutshell

Learn how to transform array dimensions, flatten multi-dimensional data, and combine arrays using stacking. This hands-on tutorial focuses on practical implementation of shape manipulation concepts.

Module 5: Shape Manipulation

Data doesn't always come in the shape you need. Reshaping and stacks are essential when preparing data for machine learning models or visualization.

Lesson 11: Reshaping Arrays

Reshaping allows you to change the number of rows and columns without changing the data itself.

Rules for Reshaping:

  • The total number of elements must remain the same.
  • You can use -1 for one dimension, and NumPy will calculate it for you automatically.

Flattening vs Raveling:

  • .flatten(): Returns a 1D copy of the array.
  • .ravel(): Returns a 1D view (faster, but affects original).
PYTHON PLAYGROUND
⏳ Loading editor…

Lesson 12: Joining & Splitting

Joining Arrays

Combining arrays is often done by "stacking".

  • np.concatenate((a, b)): Joins arrays along an existing axis.
  • np.vstack((a, b)): Vertical stack (row-wise).
  • np.hstack((a, b)): Horizontal stack (column-wise).

Splitting Arrays

Opposite of joining.

  • np.split(): Splits an array into multiple sub-arrays.
  • np.vsplit(), np.hsplit(): Specialized vertical and horizontal splits.
PYTHON PLAYGROUND
⏳ Loading editor…

Practice: Matrix Reshaping

Challenge: Create a target matrix of 100 elements (using np.arange(100)). Reshape it into a 10x10 matrix. Then, extract the first 5 rows and stack them horizontally with the last 5 rows.

Quiz

Question 1 of 5

What does the -1 parameter do in reshape(5, -1)?

It reverses the array
It sets the dimension to the total number of elements
It tells NumPy to automatically calculate that dimension
It creates a negative index

Key Takeaways

✅ Reshaping is only possible if the element count matches.
✅ .ravel() is faster than .flatten() because it doesn't copy data.
✅ Use vstack and hstack for intuitive array combining.