Exploratory Data Analysis User Interface, also known as EDA UI, is a User Interface (UI) generated on teradataml DataFrame where you can run additional functions on teradataml DataFrame without adding any additional code.
Once the teradataml DataFrame is returned to Jupyter cell, along with the result, additional UI will be given to users to perform the following functionalities on the resultant teradataml DataFrame.
- Generates column statistics for the resultant teradataml DataFrame.
- Generates python types and teradata types for resultant teradataml DataFrame.
- Generates the summary for every teradataml DataFrame Column which has details about count of NULLs, not NULLs etc.
- Generates Categorical columns summary.
- Generates details about Futile columns.
- Can plot the data and generates the corresponding image.
- Can run teradataml analytic function or a series of analytic functions.
- Can persist the teradataml DataFrame to Vantage.
Important notes regarding EDA UI:
- EDA UI is available only for Jupyter notebooks.
- EDA UI won't be displayed if a teradataml DataFrame is printed. It will be displayed only when teradataml DataFrame is returned to a Jupyter notebook cell.
- EDA UI is displayed only with a valid Jupyter session. Hence, a downloaded Jupyter notebook will not display EDA UI. However, if you start executing the downloaded notebook, it will display the EDA UI again for that session.
- Exploratory Data Analysis UI (EDA UI) works based on ipywidgets. In some cases, ipywidgets do not work on a Jupyter notebook. If this occurs, Teradata recommends selecting a kernel again.
The following image shows the EDA UI for teradataml DataFrame.