Through a new partnership between the UC San Diego Library, Halıcıoğlu Data Science Institute (HDSI), and Lawrence Livermore National Laboratory (LLNL) Data Science Institute, Library patrons can now access and analyze two new “real world” data sets from LLNL.
The Lawrence Livermore National Lab Open Data Initiative collection shares some of LLNL’s challenging and unique data sets, which range in complexity from large-scale, domain-specific simulated data to more well-understood material. The two data sets that have been made are available so far are “Cars Overhead with Context (COWC),” which contains annotated images, and “The JAG inertial confinement fusion simulation data set for multi-modal scientific deep learning,” which contains pre-trained machine learning models. These data sets can be used to test hardware solutions for scalable machine learning platforms and are publicly available for download and use. Additional data sets planned for inclusion will also be available in the Open Data Initiative collection in the coming months.
This partnership was formed in an effort to increase access to data sets of broad interest to the larger community of data science researchers, including UC San Diego students honing their skills in data science methodologies.
To view the data sets in the Library’s Digital Collections, please visit this page.