Hub
WorkflowWorkflow

Compute Local Model-agnostic Explanations (LIMEs)

LIME Machine learning interpretability Mli Explanation Instance-level explanation
+7
James Tsai profile image

Last edited: 

Drag & drop
Like
Download workflow
Copy short link
Workflow preview
這是使用LIME計算說明的示例。 選擇了XGBoost模型，但是可以使用任何模型及其Learner和Predictor節點集。 -閱讀有關葡萄酒的數據集 -在訓練和測試中對數據進行分區 -選擇一些測試集實例行進行解釋 -在輸入表中為每個實例創建本地樣本（LIME Loop Start） -使用預測變量節點和訓練有素的模型對樣本進行評分 -計算LIME，即通過使用樣本訓練局部GLM並提取權重來進行局部模型不可知的解釋。 -在組合視圖中可視化它們（右鍵單擊>打開視圖）

Used extensions & nodes

Created with KNIME Analytics Platform version 4.1.3
  • KNIME Core Trusted extension

    KNIME AG, Zurich, Switzerland

    Version 4.1.3

    knime
  • KNIME Data Generation Trusted extension

    KNIME AG, Zurich, Switzerland

    Version 4.1.0

    knime
  • KNIME H2O Machine Learning Integration Trusted extension

    KNIME AG, Zurich, Switzerland

    Version 4.1.1

    knime
  • KNIME JavaScript Views Trusted extension

    KNIME AG, Zurich, Switzerland

    Version 4.1.2

    knime
  • KNIME Machine Learning Interpretability Extension Trusted extension

    KNIME AG, Zurich, Switzerland

    Version 4.1.0

    knime
  • KNIME Math Expression (JEP) Trusted extension

    KNIME AG, Zurich, Switzerland

    Version 4.1.0

    knime
  • KNIME Plotly Trusted extension

    KNIME AG, Zurich, Switzerland

    Version 4.1.3

    knime
  • KNIME Quick Forms Trusted extension

    KNIME AG, Zurich, Switzerland

    Version 4.1.3

    knime
  • KNIME XGBoost Integration Trusted extension

    KNIME AG, Zurich, Switzerland

    Version 4.1.0

    knime
Loading deployments
Loading ad hoc executions
Discussion