How to Convert a Dictionary into a NumPy Array?

Python's NumPy library provides powerful tools for working with structured data. When dealing with dictionary data, you may need to convert it to NumPy arrays for mathematical operations and data analysis.

This tutorial will show you how to convert both simple and nested dictionaries into NumPy arrays using various methods.

Converting a Simple Dictionary

Let's start by converting a basic dictionary with key-value pairs −

import numpy as np

my_dict = {'a': 1, 'b': 2, 'c': 3, 'd': 4}

# Convert dictionary items to NumPy array
my_array = np.array(list(my_dict.items()))
print("Dictionary items as array:")
print(my_array)
print(f"Array shape: {my_array.shape}")
print(f"Data type: {my_array.dtype}")

Dictionary items as array:
[['a' '1']
 ['b' '2']
 ['c' '3']
 ['d' '4']]
Array shape: (4, 2)
Data type: <U21

Converting Only Values

If you only need the values from the dictionary −

import numpy as np

my_dict = {'a': 10, 'b': 20, 'c': 30, 'd': 40}

# Convert only values to NumPy array
values_array = np.array(list(my_dict.values()))
print("Values as array:")
print(values_array)

# Convert only keys to NumPy array  
keys_array = np.array(list(my_dict.keys()))
print("Keys as array:")
print(keys_array)

Values as array:
[10 20 30 40]
Keys as array:
['a' 'b' 'c' 'd']

Converting Nested Dictionaries

For nested dictionaries, we can extract values at different levels −

import numpy as np

nested_dict = {
    "row1": [1, 2, 3],
    "row2": [4, 5, 6], 
    "row3": [7, 8, 9]
}

# Convert nested dictionary values to 2D array
nested_array = np.array(list(nested_dict.values()))
print("Nested dictionary as 2D array:")
print(nested_array)
print(f"Shape: {nested_array.shape}")

Nested dictionary as 2D array:
[[1 2 3]
 [4 5 6]
 [7 8 9]]
Shape: (3, 3)

Complex Nested Structure

import numpy as np

complex_dict = {
    "students": {"math": [85, 90, 78], "science": [92, 88, 95]},
    "teachers": {"math": [1, 2], "science": [3, 4]}
}

# Extract specific nested values
math_scores = np.array(complex_dict["students"]["math"])
science_scores = np.array(complex_dict["students"]["science"])

print("Math scores:", math_scores)
print("Science scores:", science_scores)

# Combine into 2D array
all_scores = np.array([math_scores, science_scores])
print("Combined scores:")
print(all_scores)

Math scores: [85 90 78]
Science scores: [92 88 95]
Combined scores:
[[85 90 78]
 [92 88 95]]

Method Comparison

Accessing Array Elements

Once converted, you can access elements using standard NumPy indexing −

import numpy as np

data_dict = {'x': [1, 2], 'y': [3, 4], 'z': [5, 6]}
data_array = np.array(list(data_dict.values()))

print("Full array:")
print(data_array)

print("\nFirst row:", data_array[0])
print("Element at [1,1]:", data_array[1, 1])
print("Last column:", data_array[:, -1])

Full array:
[[1 2]
 [3 4]
 [5 6]]

First row: [1 2]
Element at [1,1]: 4
Last column: [2 4 6]

Conclusion

Converting dictionaries to NumPy arrays enables efficient mathematical operations and data manipulation. Use dict.values() for numerical data and dict.items() when you need both keys and values preserved in the array structure.