Interpretable Machine Learning Applications: Part 2

von
Coursera Project Network
In diesem angeleitetes Projekt werden Sie:

Apply Local Interpretable Model-agnostic Explanations (LIME) as a machine learning interpretation

Explain individual predictions being made by a trained machine learning model.

Add aspects for individual predictions in your Machine Learning applications.

Clock90-120 minutes
BeginnerAnfänger
CloudKein Download erforderlich
VideoVideo auf geteiltem Bildschirm
Comment DotsEnglisch
LaptopNur Desktop

By the end of this project, you will be able to develop intepretable machine learning applications explaining individual predictions rather than explaining the behavior of the prediction model as a whole. This will be done via the well known Local Interpretable Model-agnostic Explanations (LIME) as a machine learning interpretation and explanation model. In particular, in this project, you will learn how to go beyond the development and use of machine learning (ML) models, such as regression classifiers, in that we add on explainability and interpretation aspects for individual predictions. In this sense, the project will boost your career as a ML developer and modeler in that you will be able to explain and justify the behaviour of your ML model. The project will also benefit your career as a decision-maker in an executive position interested in deploying trusted and accountable ML applications. This guided project is primarily targeting data scientists and machine learning modelers, who wish to enhance their machine learning application development with explanation components for predictions being made. The guided project is also targeting executive planners within business companies and public organizations interested in using machine learning applications for automating, or informing, human decision making, not as a ‘black box’, but also gaining some insight into the behavior of a machine learning classifier. Note: This guided project based course works best for learners who are based in the North America region. We’re currently working on providing the same experience in other regions.

Kompetenzen, die Sie erwerben werden

  • Machine Learning Regression Classifiers
  • Programming in Python
  • Performance analysis of prediction models
  • Interpretable and Explainable Models

Schritt für Schritt lernen

In einem Video, das auf einer Hälfte Ihres Arbeitsbereichs abgespielt wird, führt Sie Ihr Dozent durch diese Schritte:

  1. Explore and understand the features and values from the available data about red wine quality

  2. Transform the available data into a classification dataset and problem

  3. Prepare the data for training and validation purposes

  4. Train, validate, estimate, and contrast the performance of three regression classifiers: Decision Tree, Random Forest, AdaBoost

  5. Prepare and train the “explainer” in terms of the LIME library

  6. Display and interpret explanations of individual predictions made by the three classifiers

Ablauf angeleiteter Projekte

Ihr Arbeitsbereich ist ein Cloud-Desktop direkt in Ihrem Browser, kein Download erforderlich

Ihr Dozent leitet Sie in einem Video mit geteiltem Bildschirm Schritt für Schritt an.

Häufig gestellte Fragen

Häufig gestellte Fragen

Haben Sie weitere Fragen? Besuchen Sie das Learner Help Center.