R/pittsburgh.R

In gasper: Graph Signal Processing

#' Pittsburgh Census Tracts Network.
#'
#' A dataset representing the graph structure of 402 census tracts of Allegheny County, PA.
#'
#' @format A list with 7 elements:
#' \itemize{
#'   \item \code{A} A dense matrix capturing the connections between spatially adjacent census tracts in Allegheny County.
#'   \item \code{sA} A sparse matrix representation of the census tracts' adjacency matrix.
#'   \item \code{xy} A matrix indicating the spatial location of each census tract.
#'   \item \code{f} Artificial signal with inhomogeneous smoothness across nodes and two sharp peaks near the center. This signal is formed using a mixture of five Gaussians in the underlying spatial coordinates.
#'   \item \code{y} Noisy version of the signal.
#'   \item \code{f1} Synthetic signal, possibly generated with specific smoothness parameters.
#'   \item \code{geo} Geometry data for the census tract nodes.
#' }
#'
#' @source Data and associated materials were sourced from codes provided by Yu-Xiang Wang (UC Santa Barbara) and are associated with the referenced paper.
#'
#' @references
#' Wang, Y. X., Sharpnack, J., Smola, A. J., & Tibshirani, R. J. (2016). Trend Filtering on Graphs. Journal of Machine Learning Research, 17, 1-41.
#'
"pittsburgh"