Skip to content

1 : Tensorflow + Kubernetes + ElasticSearch + Hadoop + Spark + Beam

Photo of Yaz
Hosted By
Yaz
1 : Tensorflow + Kubernetes + ElasticSearch + Hadoop + Spark + Beam

Details

General Monthly Meetup of GDG Cloud London for September 2016.

You must have an RSVP for entry and is first come first served.

Talks are 15 minutes with 5 minutes for questions.

Pizza and Drinks and chance to network following talks.

"TensorFlow and Deep Learning - Without a PhD"
Presented by Yaz from GDG Cloud (http://gdgcloud.com)

Image recognition is a problem that clearly illustrates the advantages of deep learning over traditional programming approaches. In this deep dive, how to quickly get set up with TensorFlow on Ubuntu using containers will be shown. To be even more efficient, what is becoming known as transfer learning will be demonstrated. An existing image recognition model will be used rather than the time consuming approach of building one from scratch. Subsequently, this classifier model will be trained with an image dataset. And finally, the retrained model will be live tested with new external images.

"Containing Data with Kubernetes"
Presented by Christian Simon from JetStack.io (http://www.jetstack.io/)

In Kubernetes 1.3, an alpha ‘PetSet’ feature was added in order to better support the deployment of distributed databases natively in a Kubernetes cluster. We are going take a look how to use PetSets, Dynamic Volume Provisioning and Init Containers for spinning up an ElasticSearch cluster and show that with the latest release running stafeful applications in Kubernetes became a lot easier.

"Hadoop and Spark with Dataproc"
Presented by Grace Mollison from Google (https://google.com)

In this talk Grace Mollison a Software Engineer at Google is going to be talking about Google Dataproc, a managed Hadoop & Spark on the Google Cloud Platform ( https://cloud.google.com/dataproc/ ).

"Apache Beam and Dataflow"
Presented by Mete Atamel from Google (https://google.com)

In this session, Mete, a Developer Advocate at Google, will take a look at the history of massive-scale data processing and how it has evolved over the years. We will learn about Apache Beam (incubating) an open source programming model unifying batch and stream processing and see how Apache Beam pipelines can be executed in Google Cloud Dataflow, a fully-managed cloud service for batch and stream data processing.

Photo of GDG Cloud London group
GDG Cloud London
See more events