Skip to contents

Stream exact k-nearest neighbours into destination big.matrix objects

Source: R/knn_interface.R
knn_stream_bigmatrix.Rd

Stream exact k-nearest neighbours into destination big.matrix objects

Usage

knn_stream_bigmatrix(
  x,
  query = NULL,
  xpIndex,
  xpDistance = NULL,
  k = 10L,
  metric = "euclidean",
  block_size = knn_default_block_size(),
  plan = NULL,
  exclude_self = is.null(query)
)

Arguments

x

A bigmemory::big.matrix, an external pointer referencing a big.matrix, or a prepared reference returned by knn_prepare_bigmatrix().

query

Optional query source. Supply NULL for self-search, another big.matrix or external pointer for streamed queries, or a dense numeric matrix.

xpIndex

A writable bigmemory::big.matrix or external pointer that receives the 1-based neighbour indices.

xpDistance

Optional writable bigmemory::big.matrix or external pointer that receives the neighbour distances.

k

Number of neighbours to return.

metric

Distance metric. Supported values are "euclidean", "sqeuclidean", and "cosine".

block_size

Number of rows to process per query and reference block.

plan

Optional execution plan returned by knn_plan_bigmatrix().

exclude_self

Logical flag controlling whether a query row may return itself as a neighbour when query references the same matrix as x.

Value

A list with components index, distance, k, metric, n_ref, n_query, exact, and backend. The index and distance entries reference the supplied destination objects.