Files | Enumerations | Functions

Files

file  reduction.hpp
 

Enumerations

enum  cudf::scan_type : bool { INCLUSIVE, EXCLUSIVE }
 Enum to describe scan operation type.
 

Functions

std::unique_ptr< scalarcudf::reduce (column_view const &col, std::unique_ptr< reduce_aggregation > const &agg, data_type output_dtype, rmm::mr::device_memory_resource *mr=rmm::mr::get_current_device_resource())
 Computes the reduction of the values in all rows of a column. More...
 
std::unique_ptr< columncudf::segmented_reduce (column_view const &segmented_values, device_span< size_type const > offsets, segmented_reduce_aggregation const &agg, data_type output_dtype, null_policy null_handling, rmm::mr::device_memory_resource *mr=rmm::mr::get_current_device_resource())
 Compute reduction of each segment in the input column. More...
 
std::unique_ptr< columncudf::scan (const column_view &input, std::unique_ptr< scan_aggregation > const &agg, scan_type inclusive, null_policy null_handling=null_policy::EXCLUDE, rmm::mr::device_memory_resource *mr=rmm::mr::get_current_device_resource())
 Computes the scan of a column. More...
 
std::pair< std::unique_ptr< scalar >, std::unique_ptr< scalar > > cudf::minmax (column_view const &col, rmm::mr::device_memory_resource *mr=rmm::mr::get_current_device_resource())
 Determines the minimum and maximum values of a column. More...
 

Detailed Description

Function Documentation

◆ minmax()

std::pair<std::unique_ptr<scalar>, std::unique_ptr<scalar> > cudf::minmax ( column_view const &  col,
rmm::mr::device_memory_resource mr = rmm::mr::get_current_device_resource() 
)

Determines the minimum and maximum values of a column.

Parameters
colcolumn to compute minmax
mrDevice memory resource used to allocate the returned column's device memory
Returns
A std::pair of scalars with the first scalar being the minimum value and the second scalar being the maximum value of the input column.

◆ reduce()

std::unique_ptr<scalar> cudf::reduce ( column_view const &  col,
std::unique_ptr< reduce_aggregation > const &  agg,
data_type  output_dtype,
rmm::mr::device_memory_resource mr = rmm::mr::get_current_device_resource() 
)

Computes the reduction of the values in all rows of a column.

This function does not detect overflows in reductions. Using a higher precision data_type may prevent overflow. Only min and max ops are supported for reduction of non-arithmetic types (timestamp, string...). The null values are skipped for the operation. If the column is empty, the member is_valid() of the output scalar will contain false.

Exceptions
cudf::logic_errorif reduction is called for non-arithmetic output type and operator other than min and max.
cudf::logic_errorif input column data type is not convertible to output data type.
cudf::logic_errorif min or max reduction is called and the output type does not match the input column data type.
cudf::logic_errorif any or all reduction is called and the output type is not bool8.
cudf::logic_errorif mean, var, or std reduction is called and the output type is not floating point.

If the input column has arithmetic type, output_dtype can be any arithmetic type. If the input column has non-arithmetic type, e.g. timestamp or string, the same output type must be specified.

If the reduction fails, the member is_valid of the output scalar will contain false.

Parameters
colInput column view
aggAggregation operator applied by the reduction
output_dtypeThe computation and output precision.
mrDevice memory resource used to allocate the returned scalar's device memory
Returns
Output scalar with reduce result.

◆ scan()

std::unique_ptr<column> cudf::scan ( const column_view input,
std::unique_ptr< scan_aggregation > const &  agg,
scan_type  inclusive,
null_policy  null_handling = null_policy::EXCLUDE,
rmm::mr::device_memory_resource mr = rmm::mr::get_current_device_resource() 
)

Computes the scan of a column.

The null values are skipped for the operation, and if an input element at i is null, then the output element at i will also be null.

Exceptions
cudf::logic_errorif column datatype is not numeric type.
Parameters
[in]inputThe input column view for the scan
[in]aggunique_ptr to aggregation operator applied by the scan
[in]inclusiveThe flag for applying an inclusive scan if scan_type::INCLUSIVE, an exclusive scan if scan_type::EXCLUSIVE.
[in]null_handlingExclude null values when computing the result if null_policy::EXCLUDE. Include nulls if null_policy::INCLUDE. Any operation with a null results in a null.
[in]mrDevice memory resource used to allocate the returned scalar's device memory
Returns
unique pointer to new output column

◆ segmented_reduce()

std::unique_ptr<column> cudf::segmented_reduce ( column_view const &  segmented_values,
device_span< size_type const >  offsets,
segmented_reduce_aggregation const &  agg,
data_type  output_dtype,
null_policy  null_handling,
rmm::mr::device_memory_resource mr = rmm::mr::get_current_device_resource() 
)

Compute reduction of each segment in the input column.

This function does not detect overflows in reductions. When given integral and floating point inputs, their values are promoted to int64_t and double respectively to compute, and casted to output_dtype before returning.

Null values are treated as identities during reduction.

If the segment is empty, the row corresponding to the result of the segment is null.

If any index in offsets is out of bound of segmented_values , the behavior is undefined.

Note
If the input column has arithmetic type, output_dtype can be any arithmetic type. If the input column has non-arithmetic type, e.g. timestamp, the same output type must be specified.
If input is not empty, the result is always nullable.
Exceptions
cudf::logic_errorif reduction is called for non-arithmetic output type and operator other than min and max.
cudf::logic_errorif input column data type is not convertible to output data type.
cudf::logic_errorif min or max reduction is called and the output type does not match the input column data type.
cudf::logic_errorif any or all reduction is called and the output type is not bool8.
Parameters
segmented_valuesColumn view of segmented inputs.
offsetsEach segment's offset of segmented_values. A list of offsets with size num_segments + 1. The size of ith segment is offsets[i+1] - offsets[i].
aggAggregation operator applied by the reduction.
output_dtypeThe output precision.
null_handlingIf INCLUDE, the reduction is valid if all elements in a segment are valid, otherwise null. If EXCLUDE, the reduction is valid if any element in the segment is valid, otherwise null.
mrDevice memory resource used to allocate the returned scalar's device memory
Returns
Output column with results of segmented reduction.