np.concatenate() for joining arrays in NumPy

Introduction

We use np.concatenate() to join two or more arrays into one. It helps combine data from different sources into a single array.

You have separate lists of numbers and want to analyze them together.

You collected data in parts and want to merge them for a report.

You want to add new rows or columns to an existing dataset.

You want to combine results from different experiments into one array.

Syntax

NumPy

import numpy as np

# Define arrays to join
array1 = np.array([...])
array2 = np.array([...])

# Join arrays along an axis
joined_array = np.concatenate((array1, array2), axis=0)

The arrays must have the same shape except in the dimension you join on.

The axis parameter decides if you join rows (axis=0) or columns (axis=1).

Examples

Joins two 1D arrays into one longer array.

NumPy

import numpy as np

# Example 1: Join 1D arrays
array1 = np.array([1, 2, 3])
array2 = np.array([4, 5])
result = np.concatenate((array1, array2))
print(result)

Joins arrays by adding rows. The number of columns must match.

NumPy

import numpy as np

# Example 2: Join 2D arrays along rows (axis=0)
array1 = np.array([[1, 2], [3, 4]])
array2 = np.array([[5, 6]])
result = np.concatenate((array1, array2), axis=0)
print(result)

Joins arrays by adding columns. The number of rows must match.

NumPy

import numpy as np

# Example 3: Join 2D arrays along columns (axis=1)
array1 = np.array([[1, 2], [3, 4]])
array2 = np.array([[5], [6]])
result = np.concatenate((array1, array2), axis=1)
print(result)

Joining with an empty array returns the original array unchanged.

NumPy

import numpy as np

# Example 4: Edge case - empty array
array1 = np.array([1, 2, 3])
array2 = np.array([])
result = np.concatenate((array1, array2))
print(result)

Sample Program

This program creates two arrays and joins them by adding rows. It prints the arrays before and after joining so you can see the change.

NumPy

import numpy as np

# Create two 2D arrays
array1 = np.array([[10, 20], [30, 40]])
array2 = np.array([[50, 60]])

print("Array 1 before joining:")
print(array1)

print("Array 2 before joining:")
print(array2)

# Join arrays along rows (axis=0)
joined_array = np.concatenate((array1, array2), axis=0)

print("\nJoined array after np.concatenate along axis=0:")
print(joined_array)

Important Notes

Time complexity: Joining arrays takes time proportional to the total size of the output array.

Space complexity: A new array is created, so space grows with the combined size.

Common mistake: Trying to join arrays with mismatched shapes on non-joining axes causes errors.

Use np.concatenate() when you want to join arrays along an existing axis. For adding new dimensions, consider np.stack().

Summary

np.concatenate() joins arrays along a chosen axis.

Arrays must match in shape except on the joining axis.

It creates a new combined array without changing the originals.

Practice

1. What does the function np.concatenate() do in NumPy?

easy

A. Splits an array into multiple smaller arrays

B. Joins two or more arrays along an existing axis

C. Sorts the elements of an array in ascending order

D. Creates a new array filled with zeros

Solution

1. Step 1: Understand the purpose of np.concatenate()

   It is used to join arrays along a specified axis, combining them into one larger array.

2. Step 2: Compare with other array functions

   Splitting, sorting, or creating zero arrays are different functions like np.split(), np.sort(), and np.zeros().

3. Final Answer:

   Joins two or more arrays along an existing axis -> Option B

4. Quick Check:

   np.concatenate() = join arrays [OK]

Hint: Remember: concatenate means join arrays along axis [OK]

Common Mistakes:

• Confusing concatenate with split or sort functions
• Thinking it modifies original arrays
• Assuming it creates arrays from scratch

2. Which of the following is the correct syntax to join two 1D arrays a and b using np.concatenate()?

easy

A. np.concatenate([a, b])

B. np.concatenate(a, axis=1)

C. np.concatenate(a + b)

D. np.concatenate(a, b)

Solution

1. Step 1: Check the function signature

   np.concatenate() expects a single argument: a sequence (like a list or tuple) of arrays to join.

2. Step 2: Analyze each option

   np.concatenate([a, b]) correctly passes a list of arrays [a, b]. np.concatenate(a, b) passes two separate arguments, which is invalid. np.concatenate(a + b) tries to add arrays before concatenation, which is not the function's purpose. np.concatenate(a, axis=1) passes axis=1 for 1D arrays, which is invalid.

3. Final Answer:

   np.concatenate([a, b]) -> Option A

4. Quick Check:

   Pass arrays as a list to concatenate() [OK]

Hint: Always pass arrays inside a list or tuple to concatenate [OK]

Common Mistakes:

• Passing arrays as separate arguments instead of a list
• Using plus (+) operator instead of concatenate
• Setting wrong axis for 1D arrays

3. What is the output of the following code?

import numpy as np
x = np.array([[1, 2], [3, 4]])
y = np.array([[5, 6]])
result = np.concatenate((x, y), axis=0)
print(result)

medium

A. [[1 2]
[3 4]
[5 6]]

B. [[1 2 5 6]
[3 4 5 6]]

C. [[1 2]
[3 4]]

D. Error: shapes do not match

Solution

1. Step 1: Check shapes of arrays

   x has shape (2, 2), y has shape (1, 2). They match in columns (axis=1) but differ in rows (axis=0).

2. Step 2: Concatenate along axis=0 (rows)

   Joining along rows stacks y below x, resulting in a (3, 2) array with rows from x then y.

3. Final Answer:

   [[1 2] [3 4] [5 6]] -> Option A

4. Quick Check:

   Concatenate rows stacks arrays vertically [OK]

Hint: Concatenate axis=0 stacks rows; axis=1 stacks columns [OK]

Common Mistakes:

• Confusing axis=0 and axis=1
• Expecting horizontal stacking with axis=0
• Ignoring shape compatibility

4. What is wrong with this code snippet?

import numpy as np
x = np.array([[1, 2], [3, 4]])
y = np.array([5, 6, 7])
result = np.concatenate((x, y), axis=0)
print(result)

medium

A. Error because y must be reshaped to 3D

B. No error, output is concatenated array

C. Error because axis=0 is invalid for 2D arrays

D. Error because arrays have incompatible shapes along axis=0

Solution

1. Step 1: Check shapes of arrays

   x shape is (2, 2), y shape is (3,). They differ in both dimensions.

2. Step 2: Understand concatenate axis rules

   Arrays must match in all dimensions except the concatenation axis. Here, axis=0 means rows, but y has no matching columns.

3. Final Answer:

   Error because arrays have incompatible shapes along axis=0 -> Option D

4. Quick Check:

   Shapes must match except on concat axis [OK]

Hint: Check array shapes match except on concat axis [OK]

Common Mistakes:

• Ignoring shape mismatch errors
• Assuming 1D array can concatenate with 2D without reshape
• Misunderstanding axis parameter

5. You have three arrays:

a = np.array([[1, 2], [3, 4]])
b = np.array([[5, 6]])
c = np.array([[7, 8]])

You want to join them into one array with shape (4, 2). Which code correctly does this?

hard

A. np.concatenate((a, b), axis=0) + c

B. np.concatenate([a, b, c], axis=1)

C. np.concatenate((a, b, c), axis=0)

D. np.concatenate((a, b, c), axis=2)

Solution

1. Step 1: Check shapes and desired output

   a shape is (2, 2), b and c are (1, 2). Joining along rows (axis=0) stacks them vertically to get (4, 2).

2. Step 2: Analyze options

   np.concatenate((a, b, c), axis=0) correctly concatenates all three arrays along axis=0. np.concatenate([a, b, c], axis=1) tries axis=1 which would require matching rows, which is false. np.concatenate((a, b), axis=0) + c adds arrays after concatenation, which is invalid. np.concatenate((a, b, c), axis=2) uses axis=2 which is invalid for 2D arrays.

3. Final Answer:

   np.concatenate((a, b, c), axis=0) -> Option C

4. Quick Check:

   Stack arrays vertically with axis=0 [OK]

Hint: Use axis=0 to stack arrays vertically when shapes match columns [OK]

Common Mistakes:

• Using wrong axis for concatenation
• Trying to add arrays instead of concatenate
• Using invalid axis number for 2D arrays