ai-app / shapash

🔅 Shapash makes Machine Learning models transparent and understandable by everyone

🎉 What's new ?

---

⚠️ Starting from Shapash v2.0.0: '.compile()' parameters must be provided in the SmartExplainer init: ⚠️

For clearer initiation, method's parameters must be provided at the init: xpl = SmartExplainer(model, backend, preprocessing, postprocessing, features_groups) instead of xpl.compile(x, model, backend, preprocessing, postprocessing, features_groups)

---

Version	New Feature	Description	Tutorial
2.0.x	Refactoring Shapash	Refactoring attributes of compile methods and init. Refactoring implementation for new backends
1.7.x	Variabilize Colors	Giving possibility to have your own colour palette for outputs adapted to your design
1.6.x	Explainability Quality Metrics article	To help increase confidence in explainability methods, you can evaluate the relevance of your explainability using 3 metrics: Stability, Consistency and Compacity
1.5.x	ACV Backend	A new way of estimating Shapley values using ACV. More info about ACV here.
1.4.x	Groups of features demo	You can now regroup features that share common properties together. This option can be useful if your model has a lot of features.
1.3.x	Shapash Report demo	A standalone HTML report that constitutes a basis of an audit document.

🔍 Overview

Shapash is a Python library which aims to make machine learning interpretable and understandable by everyone. It provides several types of visualization that display explicit labels that everyone can understand.

Data Scientists can understand their models easily and share their results. End users can understand the decision proposed by a model using a summary of the most influential criteria.

Shapash also contributes to data science auditing by displaying usefull information about any model and data in a unique report.

• Readthedocs:
• Presentation video for french speakers
• Medium:
  • Understand your model with Shapash - Towards AI
  • Model auditability - Towards DS
  • Group of features - Towards AI
  • Building confidence on explainability - Towards DS

🤝 Contributors

🏆 Awards

🔥 Features

• Display clear and understandable results: plots and outputs use explicit labels for each feature and its values

• Allow Data Scientists to quickly understand their models by using a webapp to easily navigate between global and local explainability, and understand how the different features contribute: Live Demo Shapash-Monitor

• Summarize and export the local explanation

Shapash proposes a short and clear local explanation. It allows each user, whatever their Data background, to understand a local prediction of a supervised model thanks to a summarized and explicit explanation

• Evaluate the quality of your explainability using different metrics

• Easily share and discuss results with non-Data users

• Deploy interpretability part of your project: From model training to deployment (API or Batch Mode)

• Contribute to the auditability of your model by generating a standalone HTML report of your projects. Report Example

We hope that this report will bring a valuable support to auditing models and data related to a better AI governance. Data Scientists can now deliver to anyone who is interested in their project a document that freezes different aspects of their work as a basis of an audit report. This document can be easily shared across teams (internal audit, DPO, risk, compliance...).

⚙️ How Shapash works

Shapash is an overlay package for libraries dedicated to the interpretability of models. It uses Shap or Lime backend to compute contributions. Shapash builds on the different steps necessary to build a machine learning model to make the results understandable

Shapash works for Regression, Binary Classification or Multiclass problem.
It is compatible with many models: Catboost, Xgboost, LightGBM, Sklearn Ensemble, Linear models, SVM.
Shapash can use category-encoders object, sklearn ColumnTransformer or simply features dictionary.

• Category_encoder: OneHotEncoder, OrdinalEncoder, BaseNEncoder, BinaryEncoder, TargetEncoder
• Sklearn ColumnTransformer: OneHotEncoder, OrdinalEncoder, StandardScaler, QuantileTransformer, PowerTransformer

🛠 Installation

Shapash is intended to work with Python versions 3.6 to 3.9. Installation can be done with pip:

pip install shapash

In order to generate the Shapash Report some extra requirements are needed. You can install these using the following command :

pip install shapash[report]

If you encounter compatibility issues you may check the corresponding section in the Shapash documentation here.

🕐 Quickstart

The 4 steps to display results:

• Step 1: Declare SmartExplainer Object

  You can declare features dict here to specify the labels to display

from shapash import SmartExplainer
xpl = SmartExplainer(
  features_dict=house_dict,  # Optional parameter
  model=regressor,
  preprocessing=encoder, # Optional: compile step can use inverse_transform method
  postprocessing=postprocess # Optional: see tutorial postprocessing
)

• Step 2: Compile Model, Dataset, Encoders, ...

  There are 2 mandatory parameters in compile method: Model and Dataset

xpl.compile(
    x=Xtest,    
    y_pred=y_pred, # Optional    
)

• Step 3: Display output

  There are several outputs and plots available. for example, you can launch the web app:

app = xpl.run_app()

Live Demo Shapash-Monitor

• Step 4: Generate the Shapash Report

  This step allows to generate a standalone html report of your project using the different splits of your dataset and also the metrics you used:

xpl.generate_report(
    output_file='path/to/output/report.html',
    project_info_file='path/to/project_info.yml',
    x_train=Xtrain,
    y_train=ytrain,
    y_test=ytest,
    title_story="House prices report",
    title_description="""This document is a data science report of the kaggle house prices tutorial project.
        It was generated using the Shapash library.""",
    metrics=[{‘name’: ‘MSE’, ‘path’: ‘sklearn.metrics.mean_squared_error’}]
)

Report Example

• Step 5: From training to deployment : SmartPredictor Object

  Shapash provides a SmartPredictor object to deploy the summary of local explanation for the operational needs. It is an object dedicated to deployment, lighter than SmartExplainer with additional consistency checks. SmartPredictor can be used with an API or in batch mode. It provides predictions, detailed or summarized local explainability using appropriate wording.

predictor = xpl.to_smartpredictor()

See the tutorial part to know how to use the SmartPredictor object

📖 Tutorials

This github repository offers many tutorials to allow you to easily get started with Shapash.

Overview

• Launch the webapp with a concrete use case
• Jupyter Overviews - The main outputs and methods available with the SmartExplainer object
• Shapash in production: From model training to deployment (API or Batch Mode)
• Use groups of features
• Deploy local explainability in production with SmartPredictor

Charts and plots

• Shapash Features Importance
• Contribution plot to understand how one feature affects a prediction
• Summarize, display and export local contribution using filter and local_plot method
• Contributions Comparing plot to understand why predictions on several individuals are different
• Visualize interactions between couple of variables
• Customize colors in Webapp, plots and report

Different ways to use Encoders and Dictionaries

• Use Category_Encoder & inverse transformation
• Use ColumnTransformers
• Use Simple Python Dictionnaries

Displaying data with postprocessing

Using postprocessing parameter in compile method

Using different backends

• Compute Shapley Contributions using Shap
• Use Lime to compute local explanation, Summarize-it with Shapash
• Use ACV backend to compute Active Shapley Values and SDP global importance
• Compile faster Lime and consistency of contributions

Evaluating the quality of your explainability

• Building confidence on explainability methods using Stability, Consistency and Compacity metrics

Generate a report of your project

• Generate a standalone HTML report of your project with generate_report