Click the name of the data file or click the Stat icon to open the Data Wrangling window.
The Data Wrangling window opens. It shows the Data at a glance [#1] submenu that lists all features in the data file, the Analyze Features [#2] submenu that is used to analyze the features of the data, and the Data Wrangling [#3] submenu that is used to transform, enrich, and clean the data.
Data at a Glance
The Columns [#5] shows the data type as “ColumnName - DataType”.
Click the Arrows [#6] icon to move to the next page if there are more columns in the data file than what is visible.
Quick statistics [#7] of the selected feature are seen on the bottom left. See autogenerated charts to the right.
Use the Edit icon (#8) to customize a chat to your liking.
Note: The display of the chart depends on the data. As an example, a map cannot be displayed if there is no location information in the data.
Analyze Features
Click the Analyze Features submenu to start.
Info: Analysis for any data type follows the same method.
Select one feature for the X-Axis [#1] by clicking the radio button. Select one or more features for the Y-Axis [#2] by using the checkboxes. Click Update Chart [#3] to see a chart for a visual comparison of the features.
For images, the columns are the classes [#1]. These are automatically identified.
Select an image .zip file [#2] following the folder path and double-click it. The images in the folder will be listed [#4]. Double-click an image to view [#5]. Open what you require, compare, and delete what you don't want.
Data Wrangling
Data Wrangling is used to transform, enrich, sample and apply wrangling techniques to data.
Under "Columns" [#A], see the list of features in the data file and their data types. Select a column [#1] to see the unique values in the selected column under Value(count) [#2]. The number of times that value occurs is shown in the bracket.
Info: Only 10 values are shown. Click on navigation pages on the top to see the next ten and repeat this process.
The Raw data visualization [#3] chart shows the distribution of the unique values and the number of times they occur. Hover the mouse over a bar to see a unique value and the number of times that value occurs.
The wrangling operations [#B] shows the algorithms and functions [#4] that can be applied to the selected feature [#2].
Completed wrangling step [#6] is a list of all completed operations on that data. The new wrangling step [#7] is a list of selected operations yet to be applied and completed.
Set parameters for a wrangling operation [#5], click Apply [5a] to move the selected wrangling operation to the new wrangling step [#7] to be applied.
Click Apply [#7a] to move the selected wrangling step to the Completed wrangling step [#6].
Select an operation [#4] to be applied to the columns selected in [#1]. The operations have been grouped into categories. A list of the operations and how they work on the different categories of data are explained in Data Wrangling Algorithms and Functions.
For every operation [#4] selected, read the auto-generated command [#B_i and [B_ii]. If needed, set parameters [#5] based on the operation. Click Add to step [#5a] and the wrangling operation will be recorded and shown under New wrangling step [#7].
Select and add as many operations to see them in [#7].
Click Preview Data to see how the data looks before clicking Save as and naming the new wrangled data file.
Idea: Save the Wrangling Steps [#C_i] as Recipes [#C] for repeated use on other data.