Machine learning goes beyond normal coding, which requires step-by-step instructions, by using algorithms that can independently learn patterns and make decisions. This skillset is in high demand, as machine learning algorithms now run the majority of trading on Wall Street and the product recommendations at big companies like Amazon, Spotify, and Netflix.
This course will begin with linear and logistic regression, the most time-tested and reliable tools for approaching a machine learning problem. The course will then progress to algorithms with a very different theoretical basis, such as k-nearest neighbors, decision trees, and random forest. This will bring important statistical concepts to the forefront, such as bias, variance and overfitting. You’ll also learn how to measure the accuracy of your models, as well as tips for choosing effective features and algorithms.
The course will be focused on the practical skills needed to solve real-world problems with machine learning. The mathematical foundations for each machine learning algorithm will be explained visually, but there will not be a formal math component. Entering students are expected to be comfortable with writing Python programs, as well as the Numpy and Pandas libraries.
Register for a Class
We provide computers to use during class. Choose a Mac or a PC during the checkout process.
This course is offered at our design school, Noble Desktop, in SoHo. View upcoming dates and register for this course directly on Noble Desktop’s website.Register at nobledesktop.com
What You'll Learn
- How to clean and balance your data using the Pandas library
- Applying machine learning algorithms such as logistic regression and random forest using the scikit-learn library
- Choosing good features to use as input for your algorithms
- Properly splitting data into training, test and cross-validation sets
- Important theoretical concepts like overfitting, variance and bias
- Evaluating the performance of your machine learning models
Full Course SyllabusDownload PDF Outline
Basic Regression Analysis
- Linear Regression
- Mean squared error
- Training set vs Test set
- Cross validation
Advanced Regression Analysis
- Multi-linear regression
- Feature engineering
- Regression vs Classification
- Logistic Regression
- Sigmoid function
- K-nearest neighbors
- Model-based vs memory-based
- Parametric vs non-parametric
- Evaluating performance
- Decision tree
- Bias-variance tradeoff
- Random forest
- Ensemble methods