What is Machine Learning? How do I build an end-to-end Machine Learning on TensorFlow and Hopsworks?
In this workshop we will go through an example of how you can produce a ML-result from beginning to end.
Bring a laptop with you!
This is the second ML-workshop out of two. The content of this workshop has been choosen by the participants of the first workshop (that was held on the 2th May). You do not need to have participated in the first workshop to participate in this second one.
Dr. Jim Dowling from Logical Clocks AB will guide you through the workhop. Jim speaks both Swedish and English.
Registration by the reply alternative on this page.
The event is free of charge (with financing from Region Norrbotten och Tillväxtverket). Warmly welcome!
Contacts of the host:
Petra Pihl, RISE[masked]
Content of the first workshop:
- What is Machine Learning? How do I build an end-to-end Machine Learning on TensorFlow and Hopsworks
Data may be the new oil, but refined data is the fuel for AI. Machine learning (ML) systems are only as good as the data they are trained on and getting the data in the right format at the right time is a challenge. In this workshop, we will start with an introduction to ML, the opportunities it presents, and the challenges to becoming an AI-enabled company. We will then proceed to build an end-to-end ML pipeline using TensorFlow. To this end, we will then introduce the open-source Hopsworks platform, an end-to-end platform for ML on Big Data. SICS ICE hosts a managed version of the Hopsworks platform at www.hops.site, and our hands-on workshop will use GPUs/compute/storage on the SICS ICE platform. We will walk you through the Hopsworks platform, writing an end-to-end deep learning application in only Python. Bring a laptop for the hands-on part, and it helps if you have written Python code before (although it is not a requirement). We will work with TensorFlow, and show scale out our application horizontally with Spark in Hopsworks. We will also walk through an ML pipelines that is orchestrated by Airflow.