Analyzer

This module serves as the main interface of a html-like page that collates all available features. Each feature component is presented as a tab which allow user to navigate to feature component of their choice. The content of each feature component differs from one component to another serving different task and displayed differently depending if the analysis is regression or classification.

class rarity.GapAnalyzer(data_loader: Union[rarity.data_loader.data_loader.CSVDataLoader, rarity.data_loader.data_loader.DataframeLoader], user_defined_title: Optional[str] = None, user_defined_port=8000)

GapAnalyzer is the main class object collating all developed feature components for Rarity. Auto-generated error analysis supports single model and bimodal (max at comparison of 2 models side by side) on tasks such as regression, binary classification and multiclass classification

Parameters

• data_loader (CSVDataLoader or DataframeLoader) – This is the class object from data_loader compiling xfeatures, yTrue, yPredict via either CSVDataLoader (for both offline and inline analysis) or DataframeLoader (for inline analysis)

• user_defined_title (str) – Title of analysis, text field defined by user

• user_defined_port (int) – Port defined by user to spin up Rarity’s gap analysis. If not mentioned, defaults to 8000.

Important Attributes:

analysis_type (str):

Analysis type defined by user. Corresponding feature components will be auto-populated based on the specified analysis type. Supported analysis types : Regression, Binary Classification, Multiclass Classification

_layout() → dash_bootstrap_components._components.Container.Container

The main app layout of Rarity

run() → dash.dash.Dash

Spin up Rarity web application built with dash components