3 个仓库
Exact attribution methods for decision trees and ensemble models like Random Forests or XGBoost.
Distinct from Array-Based Tree Modeling: Focuses on model explanation rather than memory representations of tree data structures
Explore 3 awesome GitHub repositories matching education & learning resources · Tree Model Interpretability. Refine with filters or upvote what's useful.
SHAP is a machine learning explainer that uses a game-theoretic framework to estimate the contribution of each feature to a model prediction. It provides a set of tools for quantifying how individual input features push a specific output away from a baseline value. The project includes specialized explainers for different architectures, including high-speed implementations for decision trees and ensemble models, linearization algorithms for deep learning networks, and covariance integration for linear models. It also features a model-agnostic interpretability tool that uses a kernel method to
Computes exact Shapley values for trees and ensembles to uncover complex risk factors and feature contributions.
该项目是一个全面的教育资源和技术手册,专注于可解释机器学习和可解释 AI(XAI)。它作为一本教科书和参考资料,用于实现使复杂的机器学习模型对人类透明且易于理解的技术。 该资源提供了关于构建本质上透明的模型(如决策树和稀疏线性模型)以及将事后解释方法应用于黑盒系统的指导。它详细介绍了量化特征重要性、为单个预测生成理由以及使用代理模型近似复杂决策过程的具体方法。 内容涵盖了广泛的分析功能,包括全局和局部特征影响分析、计算机视觉可解释性以及使用 Shapley 值等博弈论贡献。它还通过可解释性评估、识别模型捷径的调试工作流以及透明算法结构的设计来解决模型评估问题。 该项目以 Jupyter Notebooks 集合的形式实现。
Provides exact attribution methods for decision trees and ensemble models like Random Forests or XGBoost.
EconML 是一个 Python 因果推理库,旨在结合机器学习和计量经济学来估计异质处理效应。它作为计算条件平均处理效应的工具包,以确定特定干预措施如何影响个人或子群体。 该项目提供了一个用于双重机器学习和正交机器学习的框架,以从高维混杂因素中分离因果信号。它包括针对因果森林和工具变量学习者的专门实现,即使在存在未观察到的混杂因素的情况下,也允许恢复因果关系。 该库涵盖了广泛的功能,包括通过反驳测试和校准曲线进行的因果模型验证、个性化处理策略的构建以及动态处理机制的分析。它还支持用于不确定性量化的统计推理,以及使用基于树的模型和 Shapley 值对效应异质性的解释。 该项目主要通过 Jupyter Notebooks 实现和演示。
Analyzes complex treatment effect models using tree-based interpreters to uncover the characteristics driving outcomes.