NumPy Practical Examples: Useful Techniques :

Reader
November 20, 2024
NumPy Practical Examples: Useful Techniques
by:
blow post content copied from  Real Python
click here to view original post

The NumPy library is a Python library used for scientific computing. It provides you with a multidimensional array object for storing and analyzing data in a wide variety of ways. In this tutorial, you’ll see examples of some features NumPy provides that aren’t always highlighted in other tutorials. You’ll also get the chance to practice your new skills with various exercises.

In this tutorial, you’ll learn how to:

  • Create multidimensional arrays from data stored in files
  • Identify and remove duplicate data from a NumPy array
  • Use structured NumPy arrays to reconcile the differences between datasets
  • Analyze and chart specific parts of hierarchical data
  • Create vectorized versions of your own functions

If you’re new to NumPy, it’s a good idea to familiarize yourself with the basics of data science in Python before you start. Also, you’ll be using Matplotlib in this tutorial to create charts. While it’s not essential, getting acquainted with Matplotlib beforehand might be beneficial.

Take the Quiz: Test your knowledge with our interactive “NumPy Practical Examples: Useful Techniques” quiz. You’ll receive a score upon completion to help you track your learning progress:

NumPy Practical Examples: Useful Techniques

Interactive Quiz

NumPy Practical Examples: Useful Techniques

This quiz will challenge your knowledge of working with NumPy arrays. You won't find all the answers in the tutorial, so you'll need to do some extra investigating. By finding all the answers, you're sure to learn some interesting things along the way.

Setting Up Your Working Environment

Before you can get started with this tutorial, you’ll need to do some initial setup. In addition to NumPy, you’ll need to install the Matplotlib library, which you’ll use to chart your data. You’ll also be using Python’s pathlib library to access your computer’s file system, but there’s no need to install pathlib because it’s part of Python’s standard library.

You might consider using a virtual environment to make sure your tutorial’s setup doesn’t interfere with anything in your existing Python environment.

Using a Jupyter Notebook within JupyterLab to run your code instead of a Python REPL is another useful option. It allows you to experiment and document your findings, as well as quickly view and edit files. The downloadable version of the code and exercise solutions are presented in Jupyter Notebook format.

The commands for setting things up on the common platforms are shown below:

Fire up a Windows PowerShell(Admin) or Terminal(Admin) prompt, depending on the version of Windows that you’re using. Now type in the following commands:

Windows PowerShell 
PS> python -m venv venv\
PS> venv\Scripts\activate
(venv) PS> python -m pip install numpy matplotlib jupyterlab
(venv) PS> jupyter lab

Here you create a virtual environment named venv\, which you then activate. If the activation is successful, then the virtual environment’s name will precede your Powershell prompt. Next, you install numpy and matplotlib into this virtual environment, followed by the optional jupyterlab. Finally, you start JupyterLab.

Fire up a terminal and type in the following commands:

Shell 
$ python -m venv venv/
$ source venv/bin/activate
(venv) $ python -m pip install numpy matplotlib jupyterlab
(venv) $ jupyter lab

Here you create a virtual environment named venv/, which you then activate. If the activation is successful, then the virtual environment’s name will precede your command prompt. Next, you install numpy and matplotlib into this virtual environment, followed by the optional jupyterlab. Finally, you start JupyterLab.

You’ll notice that your prompt is preceded by (venv). This means that anything you do from this point forward will stay in this environment and remain separate from other Python work you have elsewhere.

Now that you have everything set up, it’s time to begin the main part of your learning journey.

NumPy Example 1: Creating Multidimensional Arrays From Files

When you create a NumPy array, you create a highly-optimized data structure. One of the reasons for this is that a NumPy array stores all of its elements in a contiguous area of memory. This memory management technique means that the data is stored in the same memory region, making access times fast. This is, of course, highly desirable, but an issue occurs when you need to expand your array.

Suppose you need to import multiple files into a multidimensional array. You could read them into separate arrays and then combine them using np.concatenate(). However, this would create a copy of your original array before expanding the copy with the additional data. The copying is necessary to ensure the updated array will still exist contiguously in memory since the original array may have had non-related content adjacent to it.

Constantly copying arrays each time you add new data from a file can make processing slow and is wasteful of your system’s memory. The problem becomes worse the more data you add to your array. Although this copying process is built into NumPy, you can minimize its effects with these two steps:

  1. When setting up your initial array, determine how large it needs to be before populating it. You may even consider over-estimating its size to support any future data additions. Once you know these sizes, you can create your array upfront.

  2. The second step is to populate it with the source data. This data will be slotted into your existing array without any need for it to be expanded.

Next, you’ll explore how to populate a three-dimensional NumPy array.

Populating Arrays With File Data

In this first example, you’ll use the data from three files to populate a three-dimensional array. The content of each file is shown below, and you’ll also find these files in the downloadable materials:

The first file has two rows and three columns with the following content:

CSV file1.csv 
1.1, 1.2, 1.3
1.4, 1.5, 1.6

Read the full article at https://realpython.com/numpy-example/ »

[ Improve Your Python With 🐍 Python Tricks 💌 – Get a short & sweet Python Trick delivered to your inbox every couple of days. >> Click here to learn more and see examples ]


November 20, 2024 at 07:30PM
Click here for more details...

=============================
The original post is available in Real Python by
this post has been published as it is through automation. Automation script brings all the top bloggers post under a single umbrella.
The purpose of this blog, Follow the top Salesforce bloggers and collect all blogs in a single place through automation.
============================

Salesforce