Training, Evaluating, and Tuning Deep Neural Network Models with TensorFlow-Slim

This course builds on the training in Marvin Bertin's "Introduction to TensorFlow-Slim", which covered the basic concepts and uses of the TensorFlow-Slim (TF-Slim) API. In a series of lessons designed for learners with basic machine learning knowledge and some previous TensorFlow experience, you'll explore many of TF-Slim's most advanced features; using them to build and train sophisticated deep learning models.

As you work through the examples, you'll come to appreciate TF-Slim's primary benefit: Its ability to enable the work of machine learning while avoiding code complexity, a significant problem in the world of increasingly deep neural networks.

• Learn to construct and customize losses functions for regression, classification, and multi-task problems
• Discover how to combine various metrics and use them to measure model performance
• Understand how to automate training and evaluation routines
• Learn how to train and evaluate a convolutional neural network model
• See how you can improve model performance by using fine-tuning on pre-trained models
• Gain experience using transfer learning for new predictive tasks

Marvin Bertin is a data scientist with Driver, a San Francisco based biotech startup. Before that, he worked as a deep learning researcher for the AI company Skymind. Marvin holds degrees in Data Science and Mechanical Engineering, has authored a number of courses on deep learning, and is a speaker at machine learning and deep learning conferences.