Practical machine learning in Python
Instructor. Dimitri Yatsenko, Ph.D.
Monday 5.30 - 6.45 PM
An overview of the key concepts of machine learning through practical examples and applications. Programming projects will be used for learning techniques, for interpreting results and understanding scaling up from thousands of records to millions/billions
All assignments will be in Python 3.6 or newer with libraries for data analytics:
By the end of the course, students will be expected to become skilled in applying these libraries to solving new problems.
Python has become a leading language for data analysis and machine learning thanks to the rich ecosystem of best-in-class open-source libraries developed by the community. Python Developers Survey 2019
Textbook:
-
Aurélien Géron, Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow, 2nd Edition. Publisher: O'Reilley Media, Inc. ISBN: 9781492032649.
-
Companion Repo https://github.com/ageron/handson-ml2
These above are practical resources emphasizing techniques and examples.
For a more fundamental treatment of machine learning fundamentals, I recommend reading
- Stuart Russell, Peter Norvig "Artificial Intelligence: A Modern Approach", Pearson; 4th Edition (May 8, 2020). ISBN-10: 0134610997
10 Programming Assignments
| grade | percent |
|---|---|
| A | >=94% |
| A- | >=90% |
| B+ | >=86% |
| B | >=82% |
| B- | >=78% |
| C+ | >=74% |
| C | >=70% |
| C- | >=66% |
| D+ | >=62% |
| D | >=58% |
| F | >=0% |
(Tentative, subject to change)
- Aug 24. Artificial Intelligence and Machine Learning. Applications and Examples. Configure development environment.
- Aug 31. Operations on numerical arrays, example data sets, visualization. Project 1 Due Sep 2
- Sep 7 - Labor Day
- Sep 14. Classification. ROC Analysis.
- Sep 21. Regression.
- Sep 28. Regression.
- Oct 5. SVM
- Oct 12 - Fall Break
- Oct 19. Model selection, feature selection, cross-validation. Regression project Due Oct 29
- Oct 26. Random forests.
- Nov 2. Regularization. Dimensionality reduction.
- Nov 9. Unsupervised learning
- Nov 16. Neural networks and deep learning.
- Nov 23. Gradient descent, backpropagation
- Nov 30. Keras, tensorflow, or pytorch
- Dec 7 - Review
- Dec 14 - Final