We introduce DeepForge, a gateway to deep learning for the scientific community. DeepForge is designed to lower the barrier to entry and facilitate the rapid development of deep learning models. Utilizing a cloud-based infrastructure, DeepForge facilitates rapid development by promoting reproducibility, collaboration and remote execution of machine learning pipelines. This represents an interdisciplinary approach to facilitating deep learning development as it leverages the strengths of Model Integrated Computing to provide a powerful hybrid textual-visual programming platform for the scientific community. In the demonstration, we present the core concepts of DeepForge and an image classification example.
Our goal is to create a cloud-ready repository of open-source coastal modeling tools which enable scientists and engineers to use high performance computers to study a variety of physical and ecological processes. The system we are building leverages Jupyter notebooks, Docker, Singularity, and the Agave Platform to create a platform for running jobs and analyzing data in a way that is (1) intuitive, (2) repeatable, (3) and collaborative.
The paper describes the methodology to integrate the four technologies into the system that serves the coastal resilience research collaboratory. Two commonly used open-source numerical models are used to demonstrate the utilities of the system. Simulation results of ocean waves generated by Hurricane Issac (2012) and coastal wave evolution are presented as an illustration. The sustainability of the collaboratory is discussed in detail.
Rion Dooley, PhD is the Director of Platform Services and Solutions at Data Machines Corp. He came to DMC from Texas Advanced Computing Center where he spent the last decade integrating new and emerging technologies to build solutions that make it easier to conduct open, digital science... Read More →
In recent years, the explosion of social media platforms and the public collection of social data has brought forth a growing desire and need for research capabilities in the realm of social media and social data analytics. Research on this scale, however, requires a high level of computational and data-science expertise, limiting the researchers who are capable of undertaking social media data-driven research to those with significant computational expertise or those who have access to such experts as part of their research team. The Social Media Macroscope (SMM) is a science gateway with the goal of removing that limitation and making social media data, analytics, and visualization tools accessible to researchers and students of all levels of expertise. The SMM provides a single point of access to a suite of intuitive web interfaces for performing social media data collection, analysis, and visualization via for open-source and commercial tools. Within the SMM social scientists are able to process and store large datasets and collaborate with other researchers by sharing ideas, data, and methods. This document functions as a brief primer on the initial build of the SMM. As a clarifying note, the SMM is currently in a proof-of-concept stage.
