Introduction to Machine Learning with TensorFlow

Welcome to the Machine Learning Engineer Nanodegree program! Learn about the program structure and the projects you'll work on in this program.

What to do if you have questions about your account or general questions about the program.

In this lesson, get your computer set up with Python 3 using Anaconda, as well as setting up a text editor.

Before diving into the many algorithms of machine learning, it is important to take a step back and understand the big picture associated with the entire field.

Linear regression is one of the most fundamental algorithms in machine learning. In this lesson, learn how linear regression works!

The perceptron algorithm is an algorithm for classifying data. It is the building block of neural networks.

Decision trees are a structure for decision-making where each decision leads to a set of consequences or additional decisions.

Naive Bayesian Algorithms are powerful tools for creating classifiers for incoming labeled data. Specifically Naive Bayes is frequently used with text data and classification problems.

Support vector machines are a common method used for classification problems. They have been proven effective using what is known as the 'kernel' trick!

Bagging and boosting are two common ensemble methods for combining simple algorithms to make more advanced models that work better than the simple algorithms would on their own.

Learn the main metrics to evaluate models, such as accuracy, precision, recall, and more!

Learn the main types of errors that can occur during training, and several methods to deal with them and optimize your machine learning models.

You've covered a wide variety of methods for performing supervised learning -- now it's time to put those into action!

Meet your instructors, get a short overview of what you'll be learning, check your prerequisites, and learn how to use the workspaces and notebooks found throughout the lessons.

In this lesson, Luis will give you solid foundations on deep learning and neural networks. You'll also implement gradient descent and backpropagation in Python right here in the classroom.

Mat will introduce you to a different error function and guide you through implementing gradient descent using numpy matrix multiplication.

Now that you know what neural networks are, in this lesson you will learn several techniques to improve their training.

Learn how to use TensorFlow for building deep learning models.

In this project, you'll build a Python application that can train an image classifier on a dataset, then predict new images using the trained model.

Clustering is one of the most common methods of unsupervised learning. Here, we'll discuss the K-means clustering algorithm.

We continue to look at clustering methods. Here, we'll discuss hierarchical clustering and density-based clustering (DBSCAN).

In this lesson, we discuss Gaussian mixture model clustering. We then talk about the cluster analysis process and how to validate clustering results.

Often we need to reduce a large number of features in our data to a smaller, more relevant set. Principal Component Analysis, or PCA, is a method of feature extraction and dimensionality reduction.

In this lesson, we will look at two other methods for feature extraction and dimensionality reduction: Random Projection and Independent Component Analysis (ICA).

In this project, you'll apply your unsupervised learning skills to two demographics datasets, to identify segments and clusters in the population, and see how customers of a company map to them.