Interpretable machine learning with Python
Find out what you will learn throughout the course (if the video does not show, try allowing cookies in your browser).
What you'll learn
👉 Explain intrinsically interpretable models.
👉 Explain black box models with post-hoc methods.
👉 Make global and local explanations.
👉 Feature permutation importance.
👉 Partial dependency plots, ALE plots and ICE plots.
👉 LIME, Shapley values and SHAP.
👉 Surrogate models and perturbation explainability methods.
👉 Explain with Python open source libraries.
What you'll get
Lifetime access
Instructor support
Certificate of completion
💬 English subtitles
Instructor
Soledad Galli, PhD
Sole is a lead data scientist, instructor, and developer of open source software. She created and maintains the Python library Feature-engine, which allows us to impute data, encode categorical variables, transform, create, and select features. Sole is also the author of the"Python Feature Engineering Cookbook," published by Packt.
More about Sole on LinkedIn.
30 days money back guarantee
If you're disappointed for whatever reason, you'll get a full refund.
So you can buy with confidence.
Course description
Welcome to the most comprehensive online course on Interpretable Machine Learning!
In this course, you will learn methods and tools to explain interpretable and black-box algorithms, as well as the trade-offs and risks of using each method and how to implement them in Python.
What is interpretability in machine learning?
Interpretability in machine learning refers to our ability to understand and explain how machine learning algorithms make predictions. It involves unraveling the inner workings of machine learning models to gain insights into their decision-making processes, or using alternative post-hoc methods to understand the output of more complex models.
Interpretable machine learning enables us to understand why a model produced a particular prediction, providing transparency and accountability in AI systems.
In the context of machine learning, interpretability helps answer questions such as:
- How does the model use input features to make predictions?
- What are the most important features driving the model's decisions?
- Are there any biases or unintended consequences in the model's decision-making process?
- Can we identify and rectify any errors or inconsistencies in the model's behavior?
Why is interpretable machine learning important?
Interpretability has become an indispensable aspect of modern machine learning and artificial intelligence. Interpretability empowers organizations to deploy models they can trust and understand.
Interpretations of machine learning models offer valuable insights that can be utilized in various ways. For example:
- Model Debugging and Improvement: Interpretations can help identify and rectify issues in the model's behavior.
- Model Transparency and Explainability: Interpretations provide transparency by explaining the underlying factors driving the model's decisions.
- Feature Importance Analysis: Interpretations allow us to identify the most influential features in the model's decision-making process, and hence we can deploy simpler, faster, and more interpretable models.
- Insights into Complex Relationships: Interpretations provide insights into the complex relationships between the variables in our data.
As the demand for explainable AI continues to rise across various industries, mastering interpretability techniques has become crucial for data scientists, researchers, and data professionals.
What will you learn?
In this course, you will embark on a transformative learning journey that will equip you with the knowledge and skills to unravel the mysteries of interpretable and black-box machine learning models. Whether you are working with linear models, decision trees, neural networks, or deep learning algorithms, this course will empower you to make sense of their inner workings and provide meaningful insights.
Throughout the course, we will cover a wide range of topics, including interpretability methods, explainable ML algorithms, visualization techniques, feature importance analysis, and model-agnostic approaches. You will learn to leverage popular Python libraries such as LIME and SHAP to interpret and explain complex machine learning models effectively.
Here are some key concepts and skills you will gain:
- Understand the importance of interpretability in machine learning and artificial intelligence.
- Learn techniques to explain black-box models, including neural networks and deep learning algorithms.
- Gain insights into the interpretation of regression and classification models.
- Explore interpretability techniques for tabular data and image-based datasets.
- Master the use of LIME, SHAP, and other tools for model interpretation and visualization.
- Use counterfactual explanations, saliency maps, and surrogate models to explain black-box algorithms.
- Apply interpretability techniques to real-world scenarios in healthcare, finance, and more.
By the end of this course, you will be equipped with the ability to unravel the complexities of machine learning models, understand their decision-making processes, and communicate their insights effectively.
Whether you are a data scientist, researcher, or professional looking to enhance your understanding of machine learning interpretability, this course is designed to empower you with the skills necessary to excel in your field.
Course prerequisites
To get the most out of this course, you need to have:
- Basic knowledge of machine learning algorithms and Python programming
- Familiarity with machine learning models for regression and classification, including logistic and linear regression, random forest classifiers, and gradient boosting machines.
- Familiarity with model performance metrics like ROC-AUC, MSE, and accuracy.
Who is this course for?
This course is designed for professionals and students seeking a deeper understanding of interpretability techniques. It is suitable for data scientists, researchers, and professionals in computer science, data science, and related fields, who want to improve their skills and advance their careers.
To wrap-up
This comprehensive machine learning interpretability course contains over 50 lectures spread across 10 hours of in-demand video, more than 10 quizzes and assessments, demonstrations using real-world use cases, and all topics include hands-on Python code examples in Jupyter notebooks that you can use for reference, practice, and reuse in your own projects.
Join us on this enlightening journey as we unlock the power of interpretable machine learning together. Enroll now and take a significant step towards mastering the art of machine learning interpretability.
Enroll today and unlock the secrets behind machine learning interpretability!
Course Curriculum
- Linear regression (15:10)
- Linear regression model - demo (8:52)
- Interpreting the coefficients (12:43)
- Interpreting the coefficients - sklearn (10:50)
- Interpreting the coefficients- stastsmodels (7:22)
- Local explainability (7:13)
- Local explainability - demo (5:53)
- Linear model assumptions (5:15)
- Effect of multicolinearity - demo (13:58)
- Sparse models with Lasso (4:52)
- Lasso: optimizing for interpretability - demo (13:35)
- Advantages and limitations of linear models (3:02)
- Additional reading resources
- Quiz
- Exercise (4:10)
- Added Treat: A Movie We Recommend 🍿
- Logistic regression model (6:11)
- Fitting the model (9:27)
- Assessing the model (15:39)
- Calculating the statistics - demo (8:49)
- Logistic regression - demo (10:31)
- Assessing the coefficients (6:24)
- Assessing the coefficients - empirically (8:21)
- Assessing the coefficients - bootstrapping (4:56)
- Interpreting the coefficients (9:04)
- Global explanations - demo (8:00)
- Local explanations - demo (4:54)
- Considerations (10:39)
- Additional resources
- Quiz
- Exercise (15:15)
- Extra Treat: Our Reading Suggestion 📕
Frequently Asked Questions
When does the course begin and end?
You can start taking the course from the moment you enroll. The course is self-paced, so you can watch the tutorials and apply what you learn whenever you find it most convenient.
For how long can I access the course?
The course has lifetime access. This means that once you enroll, you will have unlimited access to the course for as long as you like.
What if I don't like the course?
There is a 30-day money back guarantee. If you don't find the course useful, contact us within the first 30 days of purchase and you will get a full refund.
Will I get a certificate?
Yes, you'll get a certificate of completion after completing all lectures, quizzes and assignments.