#' Dijkstra's Algorithm
#'
#' Calculates the shortest path from the initial node to every other node in the graph.
#'
#' @param graph A \code{data.frame} with three variables (\code{v1}, \code{v2} and \code{w}).
#' @param init_node A numeric scalar.
#' @return A \code{vector} containing the shortest path to every other node from the starting node as a vector.
#' @references \url{https://en.wikipedia.org/wiki/Dijkstra\%27s_algorithm}
#' @export
dijkstra <- function(graph, init_node){
stopifnot(is.data.frame(graph), length(names(graph))==3, names(graph)==c("v1","v2","w"), is.numeric(init_node), length(init_node)==1)
Q = unique(graph$v1) # Create a vector containing the set of nodes
stopifnot(init_node %in% graph$v1) # stops if init_node is not a node in graph
distance = as.list(rep(Inf, length(Q))) # Create a distance list with Inf as the elements
names(distance) = Q # Name every element of the distance list with its respective node
distance[init_node] = 0 # Change the distance value of initial node to 0
while (length(Q)!=0){
u = as.numeric(names(which.min(unlist(distance[Q])))) # Pick a node from the list distance with minimum distance value that's in Q
Q = Q[-which(Q==u)] # Remove the node u from Q
neighbor = intersect(Q, graph[which(graph$v1==u), ][,2]) # Create a vector of nodes that neighbor u
for (i in neighbor){
alt <- distance[u] + graph[which(graph$v1==u & graph$v2==i), ][3] # Calculate the distance from init_node to i (neighboring nodes)
if (alt < distance[i]){
distance[i] = alt # Change the value in the distance list if it's smaller than the current value
}
}
}
names(distance) = NULL # Remove the list name
return(unlist(distance)) # Return the vector of distance values
}
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.