Aggregate data frame by partitions — get_by

Convenience function to aggregate the data frame according to the partitioning. Can specify a aggregation function like in aggregate.

get_by_partition(
  df,
  partitioning,
  FUN,
  columns = NULL,
  simplify = TRUE,
  drop = TRUE,
  BPPARAM = BiocParallel::SerialParam()
)

Arguments

df	Data frame that shall be aggregated.
partitioning	Nested list, specifying which `df` rows belong to one partition.
FUN	Aggregation function. Can be a base function like `unique`, `length`, etc., or a custom function.
columns	Optional: Only aggregate the specified columns of `df`. Defaults to all columns.
simplify	a logical indicating whether results should be simplified to a vector or matrix if possible.
drop	a logical indicating whether to drop unused combinations of grouping values. The non-default case `drop=FALSE` has been amended for R 3.5.0 to drop unused combinations.
BPPARAM	If multicore processing should be used, specify a `BiocParallelParam` object here. Among others, can be `SerialParam()` (default) for standard non-multicore processing or `MulticoreParam('number_cores')` for multicore processing. See `BiocParallel` for more information.

Data frame with Group column that specifies the partition and one column per specified column with aggregated values.