With the growing amount of data generated in virtually every sector, handling and interpreting large data sets is increasingly being seen as a value proposition for a considerable number of industries. Part statistician, part mathematician, and part computer scientist, the role of the Data Scientist within these industries is becoming increasingly prominent, however few tools exist to handle data of very large magnitude. At RTDS Inc., our proprietary software (SymetryML TM) leverages Apache Spark to quickly process large datasets, generating GPU-implementable predictive models with zero rescan, thus requiring little to no latency in model updating. In this hands-on tutorial, we will cover the basic use of SymetryML TM through both the web user interface and the Python API, and provide an overview of our current efforts in applying this technology to suit specialized problems in health, finance, and the internet of things.

The workshop will examine the following items:

1. Intro To RTDS and Set Up
2. How SymetryML Works
Hardware acceleration (GPU, Spark)
What makes it real time
How is SymetryML distributed
3. Using SymetryML For Finance with Web UI demo (GPU)
4. Using SymetryML For Bioinformatics and Healthcare with iPython demo

Requirements

Please bring a laptop with a modern browser and python installed.

Speakers

Anatoly Likhatchev, MSc has close to 20 years of experience in consulting and business development in areas of finance, information technology, biotechnology, automated trading systems, e-commerce, human resources, and manufacturing. He has extensive knowledge in technical trading on the stock markets and devoted his Masters Thesis (McGill University) to developing automated trading systems using neural nets and genetic algorithms. Anatoly’s work at RTDS Inc. has centered around the development of a financial modeling environment, with extensive work also building predictive models from intraday trading data.

Babak Afshin-Pour, PhD received a BSc degree in biomedical engineering and MSc and PhD degrees in electrical engineering from the University of Tehran. After the PhD, Babak was awarded a three-year post-doctoral fellowship at the Rotman Research Institute, University of Toronto. Following the fellowship, Babak became a scientific associate in the same institute. His interests are in the areas of big data analysis using Apache Spark, advanced medical signal and image processing, evaluation and optimizing of fMRI analysis techniques, graph theoretical network analysis, and analysis of multi-site neuroimaging data. At RTDS Inc., Babak has been integral in linking SymetryML with the Apache Spark framework.

Gabriel Musso, PhD obtained his BSc and MSc degrees in the life sciences before switching to computational biology for his PhD. Working at the Center for Cellular and Biomolecular Research in Toronto, Gabe’s doctoral work focused on using interaction networks to identify patterns of genome evolution. Gabe then spent 4 years as a research fellow at Harvard Medical School, working mainly on using machine learning to predict the effects of genetic mutation and to perform in silico drug screening. Gabe now works on the biology and health vertical at RTDS, using SymetryML to develop models that integrate biology, chemistry, genetics, and imaging data to diagnose disease.

Scheduled as part of