Hands-on deep learning with TensorFlow 2.0 and Azure

Maxim Lukiyanov (Microsoft), Vaidyaraman Sambasivam (Microsoft), Mehrnoosh Sameki (MERS) (Microsoft), Santhosh Pillai (Microsoft)

9:00am—5:00pm Monday, October 28—Tuesday, October 29

Location: Room 212

Participants should plan to attend both days of training course. Note: to attend training courses, you must be registered for a Platinum or Training pass; does not include access to tutorials on Tuesday.

Maxim Lukiyanov, Vaidyaraman Sambasivam, Mehrnoosh Samekihow, and Santhosh Pillai demonstrate how AzureML helps data scientists be more productive when working through developing TensorFlow models for production. You'll see the whole model development lifecycle from training to deployment and ML ops to model interpretability.

What you'll learn, and how you can apply it

Learn to train TensorFlow models efficiently and productively using AzureML, perform, monitor, and manage large-scale hyperparameter sweeps and distributed training runs
Understand how MLOps helps bring machine learning workflows to production by augmenting existing DevOps (CI/CD) practices to account for the complexity of model training, validation, deployment, and monitoring
Discover how to explain your TensorFlow model predictions at training time locally and remotely, and visualize predictions and explanations to review training results and register the best model and its corresponding explainer
Be able to convert your models to ONNX format for framework interoperability and accelerated inferencing on a variety of devices and hardware, including CPU, GPU, field-programmable gate array (FPGA), and edge devices

Who is this presentation for?

You're a data scientist, software developer, or data engineer.

Level

Intermediate

Prerequisites:

Experience coding in Python
A basic understanding of machine learning, deep learning topics, and TensorFlow

Hardware and/or installation requirements:

A laptop with an up-to-date version of Edge Chromium or Chrome installed

Outline

Day 1

Training TensorFlow 2.0 models efficiently using Azure Machine Learning
MLOps: Automated, repeatable model training deployment using Azure Machine Learning

Day 2

Explain TensorFlow model predictions with the Azure Machine Learning interpretability toolkit
Deploying TensorFlow and ONNX models in the cloud and the edge using the Azure Machine Learning

About your instructors

Maxim Lukiyanov is a principal program manager on the Azure Machine Learning team at Microsoft. He works on large-scale deep learning training.

Vaidyaraman Sambasivam is a principal program manager with Azure AI – Machine Learning team at Microsoft. He is responsible for building a highly secure, scalable and performant ML inference platform enabling users to run inferences on their ML models in different ways by effectively managing their performance/cost tradeoff without adding complexity.

Mehrnoosh Sameki is a technical program manager at Microsoft, responsible for leading the product efforts on machine learning interpretability within the Azure Machine Learning platform. Previously, she was a data scientist at Rue Gilt Groupe, incorporating data science and machine learning in the retail space to drive revenue and enhance customers’ personalized shopping experiences. She earned her PhD degree in computer science at Boston University.

Santhosh Pillai is a principal program manager with the Azure machine learning team at Microsoft. Santhosh is responsible for data scientists’ experimentation experience with Azure machine learning service, specifically its highly optimized ML workflow orchestration engine, AzureML Pipelines, that can stitch together multistep workflows across heterogeneous computes. He’s been working on the machine learning platform (infrastructure, SDK, and graph authoring UX) for Microsoft and its customers over the last several years.