-
Notifications
You must be signed in to change notification settings - Fork 0
/
params.json
1 lines (1 loc) · 1.17 KB
/
params.json
1
{"name":"Thunder","tagline":"Neural data analysis with the Spark cluster computing framework","body":"Thunder\r\n=======\r\n\r\nLibrary for neural data analysis with the Spark cluster computing framework\r\n\r\n## About\r\n\r\nSpark is a powerful new framework for cluster computing, particularly well suited to iterative computations; see the [project webpage](http://spark-project.org/documentation.html). Thunder is a family of analyses for finding structure in high-dimensional spatiotemporal neural imaging data (e.g. calcium) implemented in Spark. \r\n\r\nIt includes low-level utilities for data loading, saving, signal processing, and shared algorithms (regression, factorization, etc.), and high-level functions that can be scripted to easily combine analyses. The standard package is written in Python with Pyspark, making extensive use of scipy and numpy. A subset of functions, including prototypes of real-time analyses, are available for Scala., because they use functionality not yet available in Pyspark. We plan to port all functionality to Scala in the future.\r\n","google":"UA-46696594-1","note":"Don't delete this file! It's used internally to help with page regeneration."}