Browsi: Using ML to Predict Ad Viewability with Amazon SageMaker

Опубликовано 31 октября 2018, 17:52

On the next This Is My Architecture amzn.to/2R8HxB1, learn how Browsi is using Amazon SageMaker with TensorFlow to train models that predict ad viewability in real-time. We'll explore how they collect data with ECS running on EC2 Spot Instances, store the data in Kinesis and S3, process the data with Spark running on EMR, train machine learning models using SageMaker, expose scalable prediction endpoints on ECS, and calculate statistics with Kinesis Analytics.

Host: Matt Yanchyshyn, Global Tech Lead, AWS
Customer: Shauli Mizrachy, VP R&D, Browsi

Subscribe:
More AWS videos bit.ly/2O3zS75
More AWS events videos bit.ly/316g9t4

#AWS