Fabric offers a variety of powerful and unique features that significantly improve the development and execution of data-centric tasks. In this section, we will reveal a few of its unique features, giving you an idea of Microsoft Fabric Notebook’s huge potential. The notable capabilities encompass advanced functionalities such as seamless integration with Apache Spark, rich visualizations, interactive web-based coding environment, support for multiple programming languages, collaborative features, and streamlined workflows for machine learning experimentation and job development. Here we are going to introduce some of their unique features, you can experiment more from the How to use Microsoft Fabric notebooks and Accelerate data prep with Data Wrangler in Microsoft Fabric
Use multiple languages
In a notebook, it is possible to incorporate multiple programming languages by specifying the language magic command at the beginning of a cell. Additionally, the language of a cell can be switched using the language picker. For example…
i. Cell run with spark’s language
ii. Cell run with HTML language
Or you can select the language by default from the notebook’s Languages. Then all cells will run with the selected language.
Drag and drop to insert snippets
The use of drag-and-drop functionality within the notebook is a convenient way to read data from the Lakehouse Explorer. This feature allows users to effortlessly import data by dragging and dropping files of various formats, including text files, tables, images, and more. Subsequently, the notebook automatically generates a corresponding code snippet that facilitates data preview and manipulation operations.
The Microsoft Fabric notebook’s built-in tool allows users to view a comprehensive list of variables within the current session, including their names, types, lengths, and values. As variables are defined in code cells, they are automatically displayed in the variables explorer. Moreover, by clicking on any column header, users can conveniently sort the variables within the table based on their respective properties. To access the variables explorer, users can navigate to the “View” tab on the notebook ribbon and select the “Variables” button.
The display function
display(df) enables the transformation of SQL query results and Apache Spark data frames into visually appealing data representations. It can be applied to data frames created in PySpark and Scala, providing a convenient way to generate rich data visualizations for enhanced analysis and presentation purposes.
Within the notebook ribbon, under the “Data” tab, users can utilize the Data Wrangler dropdown prompt to access and browse the active Pandas DataFrames that are available for editing. By selecting the desired “DataFrame” from the dropdown menu, users can seamlessly open it in Data Wrangler, facilitating efficient data exploration and manipulation tasks.
- Data Wrangler generates a descriptive overview of the displayed DataFrame in the Summary panel, providing information on dimensions, missing values, and more. Selecting a column in the grid prompts the Summary panel to update with column-specific descriptive statistics, while quick insights are available in the column headers.
- The Operations panel in Data Wrangler contains a searchable list of data-cleaning steps. Users can easily access this panel to explore and utilize various operations for cleaning and transforming their data.
- When a specific operation is selected in Data Wrangler, the resulting changes are automatically previewed in the display grid, accompanied by the corresponding code displayed in the panel below. Users can choose to apply the previewed code by selecting “Apply” or “Discard” the changes.
- The toolbar above the Data Wrangler display grid offers functionalities to save the generated code. Users can copy the code to the clipboard or export it as a function to the notebook, closing Data Wrangler and appending the function to a code cell. Additionally, the updated DataFrame in the Data Wrangler display grid can be downloaded as a CSV file.
Check out our other blog Mastering Data Science with Microsoft Fabric: A Tutorial for Beginners to leverage the power of the Microsoft Fabric Notebook for Machine Learning experiments.