Ctrl+K

Copy Slice#

group copy_slice

Functions

std::vector<column_view> slice(column_view const &input, host_span<size_type const> indices, rmm::cuda_stream_view stream = cudf::get_default_stream())#

Slices a column_view into a set of column_views according to a set of indices.

The returned views of input are constructed from an even number indices where the ith returned column_view views the elements in input indicated by the range [indices[2*i], indices[(2*i)+1]).

For all i it is expected indices[i] <= input.size() For all i%2==0, it is expected that indices[i] <= indices[i+1]

input:   {10, 12, 14, 16, 18, 20, 22, 24, 26, 28}
indices: {1, 3, 5, 9, 2, 4, 8, 8}
output:  {{12, 14}, {20, 22, 24, 26}, {14, 16}, {}}

Note

It is the caller’s responsibility to ensure that the returned views do not outlive the viewed device memory.

Throws:

  • cudf::logic_error – if indices size is not even.

  • cudf::logic_error – When the values in the pair are strictly decreasing.

  • cudf::logic_error – When any of the values in the pair don’t belong to the range [0, input.size()).

Parameters:

  • input – View of column to slice

  • indices – Indices used to take slices of input

  • stream – CUDA stream used for device memory operations and kernel launches

Returns:

Vector of views of input indicated by the ranges in indices

std::vector<column_view> slice(column_view const &input, std::initializer_list<size_type> indices, rmm::cuda_stream_view stream = cudf::get_default_stream())#

Slices a column_view into a set of column_views according to a set of indices.

The returned views of input are constructed from an even number indices where the ith returned column_view views the elements in input indicated by the range [indices[2*i], indices[(2*i)+1]).

For all i it is expected indices[i] <= input.size() For all i%2==0, it is expected that indices[i] <= indices[i+1]

input:   {10, 12, 14, 16, 18, 20, 22, 24, 26, 28}
indices: {1, 3, 5, 9, 2, 4, 8, 8}
output:  {{12, 14}, {20, 22, 24, 26}, {14, 16}, {}}

Note

It is the caller’s responsibility to ensure that the returned views do not outlive the viewed device memory.

Throws:

  • cudf::logic_error – if indices size is not even.

  • cudf::logic_error – When the values in the pair are strictly decreasing.

  • cudf::logic_error – When any of the values in the pair don’t belong to the range [0, input.size()).

Parameters:

  • input – View of column to slice

  • indices – Indices used to take slices of input

  • stream – CUDA stream used for device memory operations and kernel launches

Returns:

Vector of views of input indicated by the ranges in indices

std::vector<table_view> slice(table_view const &input, host_span<size_type const> indices, rmm::cuda_stream_view stream = cudf::get_default_stream())#

Slices a table_view into a set of table_views according to a set of indices.

The returned views of input are constructed from an even number indices where the ith returned table_view views the elements in input indicated by the range [indices[2*i], indices[(2*i)+1]).

For all i it is expected indices[i] <= input.size() For all i%2==0, it is expected that indices[i] <= indices[i+1]

input:   [{10, 12, 14, 16, 18, 20, 22, 24, 26, 28},
          {50, 52, 54, 56, 58, 60, 62, 64, 66, 68}]
indices: {1, 3, 5, 9, 2, 4, 8, 8}
output:  [{{12, 14}, {20, 22, 24, 26}, {14, 16}, {}},
          {{52, 54}, {60, 22, 24, 26}, {14, 16}, {}}]

Note

It is the caller’s responsibility to ensure that the returned views do not outlive the viewed device memory.

Throws:

  • cudf::logic_error – if indices size is not even.

  • cudf::logic_error – When the values in the pair are strictly decreasing.

  • cudf::logic_error – When any of the values in the pair don’t belong to the range [0, input.size()).

Parameters:

  • input – View of table to slice

  • indices – Indices used to take slices of input

  • stream – CUDA stream used for device memory operations and kernel launches

Returns:

Vector of views of input indicated by the ranges in indices

std::vector<table_view> slice(table_view const &input, std::initializer_list<size_type> indices, rmm::cuda_stream_view stream = cudf::get_default_stream())#

Slices a table_view into a set of table_views according to a set of indices.

The returned views of input are constructed from an even number indices where the ith returned table_view views the elements in input indicated by the range [indices[2*i], indices[(2*i)+1]).

For all i it is expected indices[i] <= input.size() For all i%2==0, it is expected that indices[i] <= indices[i+1]

input:   [{10, 12, 14, 16, 18, 20, 22, 24, 26, 28},
          {50, 52, 54, 56, 58, 60, 62, 64, 66, 68}]
indices: {1, 3, 5, 9, 2, 4, 8, 8}
output:  [{{12, 14}, {20, 22, 24, 26}, {14, 16}, {}},
          {{52, 54}, {60, 22, 24, 26}, {14, 16}, {}}]

Note

It is the caller’s responsibility to ensure that the returned views do not outlive the viewed device memory.

Throws:

  • cudf::logic_error – if indices size is not even.

  • cudf::logic_error – When the values in the pair are strictly decreasing.

  • cudf::logic_error – When any of the values in the pair don’t belong to the range [0, input.size()).

Parameters:

  • input – View of table to slice

  • indices – Indices used to take slices of input

  • stream – CUDA stream used for device memory operations and kernel launches

Returns:

Vector of views of input indicated by the ranges in indices

Show Source