Join our book community on Discord

https://packt.link/EarlyAccessCommunity

Taipy is an open-source Python library that allows anyone working with data (data scientists, data and business intelligence (BI) analysts, etc.…) to create complete data applications. A complete data application is a tool that handles every step of data work—from collection to analysis—in one place, so users can easily manage data and generate insights. Data professionals use the Python ecosystem to build models, conduct analysis, and generate visualizations. Taipy solves the last-mile problem: deploying these solutions into production environments.In this chapter, you'll discover Taipy's main uses and the reasoning behind the library's creation. Then, you’ll install Taipy and some development tools to start your journey. Together, we will create our first application using two types of syntax allowed by Taipy: the Markdown API and the Python API (our book focuses on...

Technical requirements

You can find all the code provided in this chapter in the GitHub repository for this book at https://github.com/PacktPublishing/Getting-Started-with-Taipy/tree/main/chapter%201.To install Taipy, you need at least Python version 3.9. We recommend working with newer versions, such as 3.12 or above.

Meeting Taipy – a data application library

Taipy is an open-source Python library created by Avaiga, a company founded by Vincent Gosselin and Albert Antoine. Both Vincent and Albert transitioned from Java to Python as they saw it become the most relevant language in data-related operations (data analysis, data science, and even data engineering).Indeed, Python offers a convenient way for data professionals to test and experiment with models, analyze data, create interesting visuals, and interact with data in enterprise structures (retrieving from databases, inserting into databases, carrying out data transformations, and so on). Here are some popular Python libraries for these tasks:

• Machine learning (ML): scikit-learn, spaCy, and Natural Language Toolkit (NLTK)
• Data manipulation: NumPy, pandas, Polars, and PySpark (Spark API for Python)
• Data visualization: Plotly, Matplotlib, Altair, Seaborn, and Bokeh
• Artificial intelligence (AI): Python APIs for almost all large language models...

The main components of Taipy

As a project, Taipy is primarily a Python library, but it also includes other significant components:

• Taipy Studio, the Visual Studio Code (VS Code) extension described earlier.
• Taipy CLI, a command-line interface (CLI). It's also coded in Python, and you install it with the other components when you install Taipy with pip. Technically, it's part of the Python library, but you won't use it as such.

The project has a free and open-source software (OSS) part, and an Enterprise Edition, with a paid subscription. Taipy Enterprise has its own classes, but we won't be covering those in this book (we might discuss them briefly when appropriate). An exception to this is Taipy Designer, a drag-and-drop application builder that you can test for free; we talk about it in Chapter 15. As an OSS library, Taipy has two main sublibraries:

• A GUI generation component (the taipy.gui sublibrary)
• Taipy for Scenario Management (Orchestrator(), Cycle(), and...

Installing Taipy and creating your setup

Before installing Taipy to create new projects, as with any other Python library, you can create a virtual environment. You can create isolated Python environments for each project with virtual environments, preventing dependency conflicts. This makes your projects more organized and reproducible. You can also create environments for collaboration, so several users use the same versions of the libraries, and you can also reproduce your virtual environment to ensure proper deployment of your applications (with a requirements.txt file). Here’s a way to create a virtual environment:

$ python -m venv name-of-your-environment

Then, activate it with the following command:

$ .\name-of-your-environment\Scripts\activate

For macOS and Linux, use the following:

$ source name-of-your-environment/bin/activate

If you use Conda, you can create it using this command:

$ conda create --name name-of-your-environment

Then, activate it using the following...

Creating your first Taipy application

It's time to create your first, minimal application. This section has two goals:

• Make sure that your setup works fine
• Use the big elements of Taipy in a Hello World! app and see their differences

Creating GUIs with the Python API

To test whether your Taipy installation works well, you can try the following code. Write it in a file called main.py:

import taipy.gui.builder as tgb
from taipy.gui import Gui
with tgb.Page() as hello_earth_python:
    tgb.text("# Hello Earth!", mode="md")
page_to_run = Gui(page=hello_earth_python)
if __name__ == "__main__":
    page_to_run.run(use_reloader=True, dark_mode=False)

You should see something like this:

In this code, we imported the Gui class, which "holds" the Taipy application. It takes a Page element as an argument, which can be of different types, as we saw in the previous section. In this case, we...

Using the Taipy CLI

Taipy's documentation of the CLI tool ( https://docs.taipy.io/en/release-4.0/userman/ecosystem/cli/) is straightforward. You can use the CLI tool from any terminal, such as the one in VS Code. Let's go over the most important uses of the Taipy CLI.

$ taipy create
  [1/6] Application root folder name (taipy_application): my_first_app_from_cli
  [2/6] Application main Python file (main.py):
  [3/6]...

Why is Taipy different?

Besides Taipy, the Python ecosystem offers solutions to create applications, but none address the issue that Taipy solves: creating production-ready data applications.For instance, compare the following with Taipy:

• Python has some GUI application frameworks, such as tkinter. These are mainly used for simple or experimental GUI apps and are not ideal for production environments, which are typically browser-based. Tkinter also has a difficult syntax.
• You can find web application frameworks for Python users, such as Django or Flask. While these frameworks are powerful tools for web developers, they may not be the most straightforward option for data scientists looking to quickly build data applications. In fact, Taipy leverages Flask.
• Plotly Dash is a great data visualization tool. It allows the creation of complete dashboards for data analysis and can be production-ready. Dash has a harder syntax, and it doesn't offer the same flexibility to compare scenarios...

Summary

In this chapter, you discovered Taipy: an open-source Python library that allows data scientists and data analysts to share their models and visuals in production environments. It also makes it easy to shift between development and production versions of your applications.After completing this chapter, your setup should be complete. You ran your first applications using different APIs for both Taipy for GUI and Scenario Management, and you used Taipy Studio and the CLI tool.Now that everything is working and you know the project's structure, you can dive into creating compelling user interfaces with Taipy GUI in the next chapter!

Questions

1. What problems does Taipy solve?
2. What are the main components (or tools) of Taipy? What do you use them for?
3. What is the difference between Taipy GUI and Taipy Scenario Management?
4. How do you create a scenario with Taipy?
5. How do you declare a multi-page application? Tip: create a multi-page GUI application with the Taipy CLI and see how it's coded in the main file.

Answers

1. Taipy turns data models and datasets into fully functional applications that non-technical users can interact with. While many Python tools help data professionals with tasks such as modeling, visualization, and analysis, they cannot often deploy these solutions in production environments for end users.
2. The main components of Taipy are Taipy GUI and Taipy Scenario Management:
   • Taipy Scenario Management is used for managing and automating data workflows. It helps you handle complex data pipelines, schedule tasks, and manage dependencies between different steps in your data processing. It is essential for organizing and running the backend processes of your data applications.
   • Taipy GUI is used for creating the frontend, or user interface, of your data applications. It allows you to build interactive dashboards, forms, and visualizations that users can interact with.

Next to these two main components, we also have the following:

• Taipy CLI helps you set up and manage your Taipy...