Metadata-Version: 2.1
Name: erexplain
Version: 0.1.2
Summary: library to explain models for Entity Resolution
Home-page: https://github.com/vincenzomartello/ERExplain.git
Author: Vincenzo Martello
Author-email: v.martello@libero.it
License: UNKNOWN
Download-URL: https://github.com/vincenzomartello/ERExplain/archive/0.1.2.tar.gz
Description: # ExplainER
        
        [![Build Status](https://travis-ci.org/joemccann/dillinger.svg?branch=master)](https://travis-ci.org/joemccann/dillinger)
        
        ERExplain is a package to explain any Entity Resolution model, using a Black-Box approach
        
        
        # Key Features
        The explanations can be provided into two different ways:
          - **Attribute-Level**: the system shows to the user a histogram reporting the importance of the attributes with respect to the prediction. The importance is a number in [0,1]
          - **Value-level**: the system shows to the user, for each chosen attribute set/single attribute A, the most influential values for A, with respect to the prediction
        ### Tech
        
        Dillinger uses some well-know libraries:
        
        * [Pandas](https://pandas.pydata.org/): To create perturbation
        * [Numpy](https://numpy.org/)
        * [MLXtend](http://rasbt.github.io/mlxtend/): To create explanations at value level
        
        ### Installation
        
        To install the package simply run
        
        ```sh
        $ pip install erexplain
        ```
        
        
        
        
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Operating System :: OS Independent
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.6
Description-Content-Type: text/markdown
