Type:	Package
Title:	Interface for 'GraphFrames'
Version:	0.1.2
Maintainer:	Kevin Kuo <kevin.kuo@rstudio.com>
Description:	A 'sparklyr' https://spark.rstudio.com/ extension that provides an R interface for 'GraphFrames' https://graphframes.github.io/. 'GraphFrames' is a package for 'Apache Spark' that provides a DataFrame-based API for working with graphs. Functionality includes motif finding and common graph algorithms, such as PageRank and Breadth-first search.
URL:	https://github.com/rstudio/graphframes
BugReports:	https://github.com/rstudio/graphframes/issues
License:	Apache License 2.0 | file LICENSE
Encoding:	UTF-8
LazyData:	true
RoxygenNote:	6.1.0
Imports:	sparklyr, tibble, forge
Suggests:	testthat, covr, dplyr
NeedsCompilation:	no
Packaged:	2018-10-30 19:01:05 UTC; kevinykuo
Author:	Kevin Kuo [aut, cre]
Repository:	CRAN
Date/Publication:	2018-10-30 19:20:03 UTC

Breadth-first search (BFS)

Description

Breadth-first search (BFS)

Usage

gf_bfs(x, from_expr, to_expr, max_path_length = 10, edge_filter = NULL,
  ...)

Arguments

Examples

## Not run: 
g <- gf_friends(sc)
gf_bfs(g, from_expr = "name = 'Esther'", to_expr = "age < 32")

## End(Not run)

Cache the GraphFrame

Description

Cache the GraphFrame

Usage

gf_cache(x)

Arguments

Chain graph

Description

Returns a chain graph of the given size with Long ID type. The vertex IDs are 0, 1, ..., n-1, and the edges are (0, 1), (1, 2), ...., (n-2, n-1).

Usage

gf_chain(sc, n)

Arguments

Examples

## Not run: 
gf_chain(sc, 5)

## End(Not run)

Connected components

Description

Computes the connected component membership of each vertex and returns a DataFrame of vertex information with each vertex assigned a component ID.

Usage

gf_connected_components(x, broadcast_threshold = 1000000L,
  algorithm = c("graphframes", "graphx"), checkpoint_interval = 2L,
  ...)

Arguments

Examples

## Not run: 
# checkpoint directory is required for gf_connected_components()
spark_set_checkpoint_dir(sc, tempdir())
g <- gf_friends(sc)
gf_connected_components(g)

## End(Not run)

Degrees of vertices

Description

Degrees of vertices

Usage

gf_degrees(x)

Arguments

Edges column names

Description

Edges column names

Usage

gf_edge_columns(x)

Arguments

Extract edges DataFrame

Description

Extract edges DataFrame

Usage

gf_edges(x)

Arguments

Motif finding: Searching the graph for structural patterns

Description

Motif finding uses a simple Domain-Specific Language (DSL) for expressing structural queries. For example, gf_find(g, "(a)-[e]->(b); (b)-[e2]->(a)") will search for pairs of vertices a,b connected by edges in both directions. It will return a DataFrame of all such structures in the graph, with columns for each of the named elements (vertices or edges) in the motif. In this case, the returned columns will be in order of the pattern: "a, e, b, e2."

Usage

gf_find(x, pattern)

Arguments

Examples

## Not run: 
gf_friends(sc) %>%
  gf_find("(a)-[e]->(b); (b)-[e2]->(a)")

## End(Not run)

Graph of friends in a social network.

Description

Graph of friends in a social network.

Usage

gf_friends(sc)

Arguments

Examples

## Not run: 
library(sparklyr)
sc <- spark_connect(master = "local")
gf_friends(sc)

## End(Not run)

Create a new GraphFrame

Description

Create a new GraphFrame

Usage

gf_graphframe(vertices = NULL, edges)

Arguments

Examples

## Not run: 
library(sparklyr)
sc <- spark_connect(master = "local", version = "2.3.0")
v_tbl <- sdf_copy_to(
  sc, data.frame(id = 1:3, name = LETTERS[1:3])
)
e_tbl <- sdf_copy_to(
  sc, data.frame(src = c(1, 2, 2), dst = c(2, 1, 3),
                 action = c("love", "hate", "follow"))
)
gf_graphframe(v_tbl, e_tbl)
gf_graphframe(edges = e_tbl)

## End(Not run)

Generate a grid Ising model with random parameters

Description

Generate a grid Ising model with random parameters

Usage

gf_grid_ising_model(sc, n, v_std = 1, e_std = 1)

Arguments

Details

This method generates a grid Ising model with random parameters. Ising models are probabilistic graphical models over binary variables xi. Each binary variable xi corresponds to one vertex, and it may take values -1 or +1. The probability distribution P(X) (over all xi) is parameterized by vertex factors ai and edge factors bij:

P(X) = (1/Z) * exp[ \sum_i a_i x_i + \sum_{ij} b_{ij} x_i x_j ]

Value

GraphFrame. Vertices have columns "id" and "a". Edges have columns "src", "dst", and "b". Edges are directed, but they should be treated as undirected in any algorithms run on this model. Vertex IDs are of the form "i,j". E.g., vertex "1,3" is in the second row and fourth column of the grid.

Examples

## Not run: 
gf_grid_ising_model(sc, 5)

## End(Not run)

In-degrees of vertices

Description

In-degrees of vertices

Usage

gf_in_degrees(x)

Arguments

Label propagation algorithm (LPA)

Description

Run static Label Propagation for detecting communities in networks. Each node in the network is initially assigned to its own community. At every iteration, nodes send their community affiliation to all neighbors and update their state to the mode community affiliation of incoming messages. LPA is a standard community detection algorithm for graphs. It is very inexpensive computationally, although (1) convergence is not guaranteed and (2) one can end up with trivial solutions (all nodes are identified into a single community).

Usage

gf_lpa(x, max_iter, ...)

Arguments

Examples

## Not run: 
g <- gf_friends(sc)
gf_lpa(g, max_iter = 5)

## End(Not run)

Out-degrees of vertices

Description

Out-degrees of vertices

Usage

gf_out_degrees(x)

Arguments

PageRank

Description

PageRank

Usage

gf_pagerank(x, tol = NULL, reset_probability = 0.15, max_iter = NULL,
  source_id = NULL, ...)

Arguments

Examples

## Not run: 
g <- gf_friends(sc)
gf_pagerank(g, reset_probability = 0.15, tol = 0.01)

## End(Not run)

Persist the GraphFrame

Description

Persist the GraphFrame

Usage

gf_persist(x, storage_level = "MEMORY_AND_DISK")

Arguments

Register a GraphFrame object

Description

Register a GraphFrame object

Usage

gf_register(x)

Arguments

Strongly connected components

Description

Compute the strongly connected component (SCC) of each vertex and return a DataFrame with each vertex assigned to the SCC containing that vertex.

Usage

gf_scc(x, max_iter, ...)

Arguments

Examples

## Not run: 
g <- gf_friends(sc)
gf_scc(g, max_iter = 10)

## End(Not run)

Shortest paths

Description

Computes shortest paths from every vertex to the given set of landmark vertices. Note that this takes edge direction into account.

Usage

gf_shortest_paths(x, landmarks, ...)

Arguments

Examples

## Not run: 
g <- gf_friends(sc)
gf_shortest_paths(g, landmarks = c("a", "d"))

## End(Not run)

Generate a star graph

Description

Returns a star graph with Long ID type, consisting of a central element indexed 0 (the root) and the n other leaf vertices 1, 2, ..., n.

Usage

gf_star(sc, n)

Arguments

Examples

## Not run: 
gf_star(sc, 5)

## End(Not run)

Computes the number of triangles passing through each vertex.

Description

This algorithm ignores edge direction; i.e., all edges are treated as undirected. In a multigraph, duplicate edges will be counted only once.

Usage

gf_triangle_count(x, ...)

Arguments

Examples

## Not run: 
g <- gf_friends(sc)
gf_triangle_count(g)

## End(Not run)

Triplets of graph

Description

Triplets of graph

Usage

gf_triplets(x)

Arguments

Generate two blobs

Description

Two densely connected blobs (vertices 0->n-1 and n->2n-1) connected by a single edge (0->n).

Usage

gf_two_blobs(sc, blob_size)

Arguments

Examples

## Not run: 
gf_two_blobs(sc, 3)

## End(Not run)

Unpersist the GraphFrame

Description

Unpersist the GraphFrame

Usage

gf_unpersist(x, blocking = FALSE)

Arguments

Vertices column names

Description

Vertices column names

Usage

gf_vertex_columns(x)

Arguments

Extract vertices DataFrame

Description

Extract vertices DataFrame

Usage

gf_vertices(x)

Arguments

Retrieve a GraphFrame

Description

Retrieve a GraphFrame

Usage

spark_graphframe(x, ...)

spark_graphframe(x, ...)

Arguments