|
Prerequisites
This class assumes that you have strong general programming strength (critical thinking, problem solving, etc.) but not super strong actual programming fundamentals (the fundamentals of programming that are necessary will be recapped in the first few weeks).
Class Description
Hopefully in this class we’ll be able to explore a few different ways to tackle data problems and learn machine learning and deep learning tools and concepts. We’ll hopefully go through a review of general python programming and bash scripting, then dive into simplistic machine learning models,more advanced ones, then deep learning models and frameworks. We will also look at data manipulation, some common strategies, and determine what strategy is the most effective. We will look into results analysis as well as performance optimization after, looking at how we can marginally improve results. Finally, if time permits, we will look at very advanced and complicated deep learning strategies (data augmentation, etc.) This class will be taught in Python.
Class SetUp
Before the first class we’ll send an email about setup (I like python to be setup in a specific way - it makes it easier to ensure that packages are in the right places, there isn’t ambiguity, etc.) I will assume that students will either have Windows (Windows 10) or a Mac. If you have Windows that isn’t Windows 10.
Topics
-
Introduction, Math Basics, and Python Basics
-
Python Data Science Packages
-
Linear Regression, Simple Machine Learning
-
SVM, Random Forest, and Scikit-Learn
-
Simple Neural Networks and Keras
-
Other Neural Networks with Keras
-
Image Classification
-
Tensorflow and other Deep Learning Frameworks
-
Results Analysis and Calculation
-
Unsupervised Learning, k-Nearest Neighbors
-
Performance Optimization
-
Transfer Learning, other Advanced Topics
|
