OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Evaluation & Observability

Shapash

Shapash is an open-source Python library for interpreting machine-learning models through visualizations, an interactive web app and standalone reports. It lets users navigate between local and global explanations, examine feature interactions and share results with technical and non-technical audiences.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Shapash — OpenViST project profile in AI Evaluation & Observability.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Shapash after you have a model and data to explain, when visual exploration or shareable model-explanation outputs are needed.

Who it is for

  • Data scientists explaining models to colleagues
  • Machine-learning teams examining local and global model behavior
  • Teams preparing model-explanation views or reports for review

Documented uses

  • Inspect explanations for individual predictions and overall model behavior.
  • Explore interactions between model features in a web app.
  • Share model-explanation results with non-data specialists.
  • Generate a standalone HTML report using dataset splits and selected metrics.
Before you choose

Requirements & limits

  • The documented workflow requires an existing model and dataset; Shapash is not a model-training system.
  • The project specifies Python 3.11–3.14 and pip installation.
  • Report generation requires the report installation extra.
  • The README lists CatBoost, XGBoost, LightGBM, scikit-learn ensembles, linear models and SVM as supported model families and points to integration guidance for others.
  • The repository lists an Apache-2.0 license.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Launch the Shapash web app from a compiled explainer

Display model-explanation results in the Shapash web app after creating and compiling a SmartExplainer.

Before you start

  • A Python environment with Shapash installed.
  • A SmartExplainer initialized with a model and compiled with a dataset.

Steps

  1. Create a SmartExplainer with the model, optionally setting feature labels, preprocessing, or postprocessing.

  2. Compile it with a dataset using xpl.compile(x=xtest); optional parameters include predictions, target values, and additional data.

  3. Launch the web app with app = xpl.run_app().

Expected outcome

The app is launched to explore Shapash results, including local and global explainability and feature interactions.

Guide limitations

  • This workflow assumes that the model and dataset are available and that the explainer has been compiled.
  • The source describes launching the app but does not specify hosting, access control, or sharing configuration.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
10 October 2026
Original website title
MAIF/shapash · Official homepage title · retained in its source language
How we review sourcesSuggest a correction