We need a simple class that can do all the basic stuff that one does manually in the Data Exploration and Preprocessing Phase. The key functionalities should include:
- First and last rows
- Data Types
- Missing values
- Duplicates
- Correlations / Heatmap (lower triangular matrix)
- Descriptive Statistics
- Basic Visualizations
It might be useful to create a metaclass independant of the data type. This class could define methods that sub-classes need to implement.
Dependencies
Before the analysis class can be implemented, the general structure / interface of the report package is required.
We need a simple class that can do all the basic stuff that one does manually in the Data Exploration and Preprocessing Phase. The key functionalities should include:
It might be useful to create a metaclass independant of the data type. This class could define methods that sub-classes need to implement.
Dependencies
Before the analysis class can be implemented, the general structure / interface of the report package is required.