Welcome to source{d} bi-weekly, a newsletter with the latest news, resources and events related to Code as Data and Machine Learning on Code. Sign up for source{d} bi-weekly newsletter.

Introducing source{d} Engine v0.11

We are excited to introduce the source{d} Engine v0.11, the first release of our new periodic product release cycle. This means you can expect more consistent releases for source{d} products from now on. Read the blog to hear about all the new features, integrations and bug fixes. Learn More.

source{d} News

Online meetup recap: Predicting the future of Kubernetes from its code [Blog]
by DevRel Team

During our last online meetup we demonstrated our ability to extract insights from the Kubernetes codebase by using the source{d} Engine. This meetup explains how you can perform this analysis yourself, how to explain past events which left traces in the source code and look at the project and contribution trends to predict the future of the Kubernetes project.

MSR Interview #6: Waldemar Hummer [Blog]
by Victor Coisne

This article is the sixth episode of our MSR Interview blog series. This week, we’re publishing the interview with Waldemar Hummer who's a researcher at IBM Research AI, T. J. Watson lab in New York.

Source{d} Named a “Pioneer” Vendor in Analyzing Source Code [Blog]
by DevRel Team

Today we are excited to announce that analyst research group, Ovum, has released a new “On the Radar” report recognized source{d} as a “pioneer” in analyzing source code! On the Radar is a series of research notes by Ovum about vendors bringing innovative ideas, products, or business models to their market, so we are so humbled to be featured.

Machine Learning on Code [Slides]
by Michael Fromberger

Learn how source{d} tools can ingest all of the world’s public git repositories turning code into ASTs ready for machine learning and other analyses, all exposed through a flexible and friendly API.

Community News

CommunityBridge gives better visibility into open source code [Article]
by Christian Hargrave

The Linux Foundation has launched CommunityBridge, a platform that aims to empower developers and the individuals and organizations who support them, to advance sustainability, security, and diversity in open source technology. With the help of source{d}, the Linux Foundation will be able to provide the Open Source community with greater visibility into each project codebases making it possible to better understand people, process, and software portfolio.

Generative Code Modeling with Graphs [Research Paper]
by Marc Brockschmidt  Miltiadis Allamanis Alexander L. Gaunt and Oleksandr Polozov

Generative models for source code are an interesting structured prediction problem, requiring to reason about both hard syntactic and semantic constraints as well as about natural, likely programs. In this paper, the authors present a novel model for this problem that uses a graph to represent the intermediate state of the generated output.

Light on Math ML: Attention in Deep Networks with Keras [Article]
by Thushan Ganegedara

With the unveiling of TensorFlow 2.0 it is hard to ignore the conspicuous attention (no pun intended!) given to Keras. There was greater focus on advocating Keras for implementing deep networks. This article explains how Keras in TensorFlow 2.0 will come with three powerful APIs for implementing deep networks.

Synthetic Datasets for Neural Program Synthesis [Research Paper]
by Richard Shin, Neel Kant, Kavi Gupta, Chris Bender, Brandon Trabucco, etc

The goal of program synthesis is to automatically generate programs in a particular language from corresponding specifications, e.g. input-output behavior. In this paper, the authors propose a new methodology for controlling and evaluating the bias of synthetic data distributions over both programs and specifications.


April 3rd: Code as Data workshop: Get insights from your codebase (San Francisco, CA)

April 5th: source{d} reading club (Online)

April 12th: GothamGo Conference 2019 (New York City, USA)

April 18th: source{d} talk at AI Paris meetup (Paris, France)

Featured Community Member

Waldemar Hummer currently works as a researcher at IBM Research AI, T. J. Watson lab in New York. Check out his website to see his impressive list of papers and projects. Make sure to follow Jürgen on twitter @w_hummer to stay up to date with his latest publications and projects