Brought to you by NumFOCUS Foundation and O’Reilly Media
The official Jupyter Conference
Aug 21-22, 2018: Training
Aug 22-24, 2018: Tutorials & Conference
New York, NY

Explore the AWS machine learning platform using Amazon SageMaker (Day 2)

Wenming Ye (Amazon Web Services), Miro Enev (NVIDIA)
9:00am–5:00pm Wednesday, August 22, 2018
Location: Concourse E Level: Intermediate

Outline

Day 1

  • Introduction to the Amazon data and ML platform
  • Hands-on exercises: Set up an AWS Account; run examples from Amazon ML API services (Transcribe, Translate, Lex, Polly, Comprehend, Rekognition, and more); build an end-to-end ML application with the API services inside SageMaker’s Jupyter notebook
  • Introduction to deep learning and SageMaker
  • Hands-on exercises: Use SageMaker to build, train, and deploy a machine learning model to the cloud; modify your model to retrain with a larger dataset with a GPU

Day 2

  • Running common deep learning frameworks like PyTorch, MxNet, TensorFlow, and Caffe 2 on AWS
  • Hands-on exercises: Register your device; create a DeepLens project using the object detection template; deploy the model to the device; check the output on the IoT console and on the video stream
  • Architecture group discussion
  • Hands-on exercises: Modify your model to focus on new objects using SageMaker; deploy the new model to the device; build an end-to-end application using Lambda, SNS, and other AWS services
Photo of Wenming Ye

Wenming Ye

Amazon Web Services

Wenming Ye is an AI/ML solutions architect at Amazon Web Services, helping researchers and enterprise customers to use cloud-based machine learning services to rapidly scale their innovations. Previously, Wenming had a diverse R&D experience at Microsoft Research, SQL engineering team, and successful startups.

Photo of Miro Enev

Miro Enev

NVIDIA

Miro Enev is a senior solutions architect at NVIDIA, specializing in advancing data science and machine intelligence while respecting human values. He supports the Pacific Northwest teams engaged with cloud, industrial, and retail clients while participating in research in deep reinforcement learning and edge-to-cloud AI. Miro holds a PhD from the University of Washington’s computer science and engineering department, where his thesis was on machine learning applications for information privacy in emerging sensor contexts. He studied cognitive science and computer science as an undergraduate at the University of California, Berkeley.