eli5
Debug machine learning classifiers and explain their predictions
Decision gist · record as of 2026-08-14
Yes, with conditions. Install if you need model explainability for scikit-learn or gradient-boosted models in a research or production debugging context. The low install friction and permissive license make it a low-risk addition. However, the aging maintenance status (481 days since last release) means you should verify that the framework versions you use are compatible; test thoroughly before relying on it in critical pipelines.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires scikit-learn for most use cases; optional dependencies (Keras, XGBoost, LightGBM, CatBoost) needed for framework-specific explanations.
- Low friction: pure Python wheel with seven runtime dependencies (attrs, jinja2, numpy, scipy, scikit-learn, graphviz, tabulate).
- Maintenance status is aging—last release 481 days ago—but the repository remains active and supports current Python versions (3.9–3.13).
License · maintenance · safety
MIT license (permissive) — MIT license (permissive): you can use, modify, and distribute eli5 freely in commercial and private projects with minimal restrictions, provided you include the license notice.
last release 2025-04-20 (481 days) · last repo commit 2025-04-20 · 331 stars
0 known vulnerabilities (OSV.dev, 2026-08-14) · 137,234 downloads/mo, #11,374 on PyPI
Alternatives
Verify before relying
pip install eli5
import eli5
from scikit-learn import datasets
from sklearn.linear_model import LogisticRegression
X, y = datasets.load_iris(return_X_y=True)
clf = LogisticRegression().fit(X, y)
eli5.explain_prediction(clf, X[0])- Whether OpenAI client support (mentioned in changelog 0.15.0) is fully functional or experimental.
- Current state of Keras 3.x and TensorFlow 2.x compatibility beyond the 0.16.0 changelog note.
- Whether all black-box inspection features (TextExplainer, permutation importance) are equally mature or if some remain experimental.
What it is and what it does
ELI5 is a Python package for debugging and explaining machine learning model predictions. It provides native support for scikit-learn, XGBoost, LightGBM, CatBoost, Keras, and other frameworks, allowing you to inspect model weights, visualize feature importances, print decision trees as text or SVG, and highlight text features used in predictions. For black-box models, it implements LIME-based text explanation and permutation importance methods.
Explanations can be rendered as text for console output, HTML for notebooks and dashboards, pandas DataFrames for further processing, or JSON for custom client-side rendering. The package is built on attrs, jinja2, numpy, scipy, scikit-learn, graphviz, and tabulate, making it straightforward to integrate into existing ML workflows.
Use it for
- Debug linear classifiers and regressors by examining learned weights and prediction contributions.
- Visualize feature importances and decision paths for tree-based ensembles.
- Explain text classifier predictions by highlighting which words or features influenced the decision.
- Inspect black-box model behavior using LIME for arbitrary classifiers or permutation importance.
- Generate Grad-CAM visualizations for Keras image classifiers to understand prediction drivers.
- Undo hashing in pipelines containing HashingVectorizer to recover interpretable feature names.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes, with conditions.
Install if you need model explainability for scikit-learn or gradient-boosted models in a research or production debugging context. The low install friction and permissive license make it a low-risk addition. However, the aging maintenance status (481 days since last release) means you should verify that the framework versions you use are compatible; test thoroughly before relying on it in critical pipelines.
Install
eli5 on PyPI
Before you install
Low friction: pure Python wheel with seven runtime dependencies (attrs, jinja2, numpy, scipy, scikit-learn, graphviz, tabulate). Maintenance status is aging—last release 481 days ago—but the repository remains active and supports current Python versions (3.9–3.13).
Requires scikit-learn for most use cases; optional dependencies (Keras, XGBoost, LightGBM, CatBoost) needed for framework-specific explanations.
License in practice
MIT license (permissive): you can use, modify, and distribute eli5 freely in commercial and private projects with minimal restrictions, provided you include the license notice.
Quickstart
pip install eli5
import eli5
from scikit-learn import datasets
from sklearn.linear_model import LogisticRegression
X, y = datasets.load_iris(return_X_y=True)
clf = LogisticRegression().fit(X, y)
eli5.explain_prediction(clf, X[0])Verify before relying
- Whether OpenAI client support (mentioned in changelog 0.15.0) is fully functional or experimental.
- Current state of Keras 3.x and TensorFlow 2.x compatibility beyond the 0.16.0 changelog note.
- Whether all black-box inspection features (TextExplainer, permutation importance) are equally mature or if some remain experimental.
Package facts
| License | MIT license permissive |
| Python support | Supports the current Python release >=3.9 |
| Install friction | Low. Pure-Python wheel |
| Runtime dependencies | 7 packagesattrsjinja2numpyscipyscikit-learngraphviztabulate |
| Maintenance | Aging 481 days since the last release |
| Last repo commit | |
| First released | |
| Downloads | 137,234 / month, #11,374 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
| Classifiers | Development Status :: 4 - BetaIntended Audience :: DevelopersLicense :: OSI Approved :: MIT LicenseOperating System :: OS IndependentProgramming Language :: PythonProgramming Language :: Python :: 3Programming Language :: Python :: 3.10Programming Language :: Python :: 3.11Programming Language :: Python :: 3.12Programming Language :: Python :: 3.13Programming Language :: Python :: 3.9 |
Evidence: eli5-0.16.0-py2.py3-none-any.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “machine learning model explainability”
- eli5Explains predictions and weights of machine learning classifiers from…
- interpretInterpretML provides interpretable machine learning models and…
- shapSHAP computes Shapley values to explain individual predictions and…
Give your agent the search over MCP, or paste the wish link into any chat.
More Artificial Intelligence packages
LiteLLM provides a unified Python interface to call 100+ LLM providers (OpenAI, Anthropic, Gemini, Bedrock, Azure, and others) using OpenAI-compatible API format, available as both a Python SDK and a self-hosted AI Gateway proxy server.
Install it if you need to work with multiple LLM providers or want to centralize LLM routing in your organization.
Client library and CLI tool for downloading, uploading, and managing models, datasets, and repositories on the Hugging Face Hub platform.
Install it if you work with Hugging Face Hub models or datasets.
LangChain provides a framework for building agents and LLM-powered applications by composing language models, tools, and memory through a unified API that abstracts over multiple model providers.
hf-xet provides chunk-based deduplication and efficient file transfer for the Hugging Face Hub, enabling faster uploads and downloads of large files with local disk caching.
Tokenizers converts raw text into token sequences for NLP models, with support for training custom vocabularies and using pre-built tokenizers (BPE, WordPiece) optimized for speed via Rust.
Transformers provides a unified framework for loading, fine-tuning, and running state-of-the-art pretrained models across text, vision, audio, video, and multimodal tasks using PyTorch, JAX, or TensorFlow.
Install it if you need to run or train any transformer-based model for NLP, vision, audio, or multimodal tasks.
See also lime · shap · interpret-core · treeinterpreter · trove-classifiers · interpret · scikit-multilearn · dtreeviz · pytabkit · betacal