Why dtypes matter for performance in NumPy - See It in Action

Why dtypes matter for performance

📖 Scenario: Imagine you are working with a large list of numbers representing daily temperatures in Celsius for a city over many years. You want to analyze this data quickly and efficiently.

🎯 Goal: You will create a NumPy array with a specific data type, then change its data type to see how it affects memory usage and performance when calculating the average temperature.

📋 What You'll Learn

Create a NumPy array with a specific dtype

Create a variable to hold a different dtype

Convert the array to the new dtype

Calculate and print the average temperature

💡 Why This Matters

🌍 Real World

Data scientists often work with large datasets where choosing the right data type can save memory and speed up calculations.

💼 Career

Understanding data types and their impact on performance is important for optimizing data processing and analysis tasks in real jobs.

1

Create a NumPy array with dtype float64

Import NumPy as np. Create a NumPy array called temps with these exact values: [23.5, 25.0, 22.1, 24.3, 26.7] and set its data type to float64.

NumPy

# Your code here

Need a hint?

Use np.array() with the dtype parameter set to np.float64.

2

Create a variable for the new dtype float32

Create a variable called new_dtype and set it to np.float32.

NumPy

import numpy as np

temps = np.array([23.5, 25.0, 22.1, 24.3, 26.7], dtype=np.float64)
# Your code here

Need a hint?

Just assign np.float32 to the variable new_dtype.

3

Convert the array to the new dtype

Create a new NumPy array called temps_float32 by converting temps to the data type stored in new_dtype using the astype() method.

NumPy

import numpy as np

temps = np.array([23.5, 25.0, 22.1, 24.3, 26.7], dtype=np.float64)
new_dtype = np.float32
# Your code here

Need a hint?

Use temps.astype(new_dtype) to convert the array.

4

Calculate and print the average temperature

Calculate the average temperature of temps_float32 using the mean() method and print the result with the exact text: Average temperature: <value> where <value> is the calculated average.

NumPy

import numpy as np

temps = np.array([23.5, 25.0, 22.1, 24.3, 26.7], dtype=np.float64)
new_dtype = np.float32
temps_float32 = temps.astype(new_dtype)
# Your code here

Need a hint?

Use temps_float32.mean() to get the average and print it with the exact text format.