Shuffle¶

Overview¶

A primitive to shuffle tensor data along an axis. More…

// structs

struct dnnl_shuffle_desc_t;
struct dnnl::shuffle_backward;
struct dnnl::shuffle_forward;

// global functions

dnnl_status_t DNNL_API dnnl_shuffle_forward_desc_init(
    dnnl_shuffle_desc_t* shuffle_desc,
    dnnl_prop_kind_t prop_kind,
    const dnnl_memory_desc_t* data_desc,
    int axis,
    dnnl_dim_t group_size
    );

dnnl_status_t DNNL_API dnnl_shuffle_backward_desc_init(
    dnnl_shuffle_desc_t* shuffle_desc,
    const dnnl_memory_desc_t* diff_data_desc,
    int axis,
    dnnl_dim_t group_size
    );

Detailed Documentation¶

A primitive to shuffle tensor data along an axis.

Global Functions¶

dnnl_status_t DNNL_API dnnl_shuffle_forward_desc_init(
    dnnl_shuffle_desc_t* shuffle_desc,
    dnnl_prop_kind_t prop_kind,
    const dnnl_memory_desc_t* data_desc,
    int axis,
    dnnl_dim_t group_size
    )

Initializes a descriptor for shuffle forward propagation primitive.

Parameters:

shuffle_desc	Output descriptor for a shuffle primitive.
prop_kind	Propagation kind. Possible values are dnnl_forward_training and dnnl_forward_inference.
data_desc	Source and destination memory descriptor.
axis	The axis along which the data is shuffled.
group_size	Shuffle group size.

Returns:

dnnl_success on success and a status describing the error otherwise.

dnnl_status_t DNNL_API dnnl_shuffle_backward_desc_init(
    dnnl_shuffle_desc_t* shuffle_desc,
    const dnnl_memory_desc_t* diff_data_desc,
    int axis,
    dnnl_dim_t group_size
    )

Initializes a descriptor for shuffle backward propagation primitive.