org.bytedeco.dnnl

Class post_ops

java.lang.Object

org.bytedeco.javacpp.Pointer

org.bytedeco.dnnl.dnnl_post_ops_handle

org.bytedeco.dnnl.post_ops

All Implemented Interfaces:

AutoCloseable

@Namespace(value="dnnl")
 @Properties(inherit=dnnl.class)
public class post_ops
extends dnnl_post_ops_handle

\endcond

Post-ops. Post-ops are computations executed after the main primitive computations and are attached to the primitive via primitive attributes.

See Also:

dev_guide_attributes_post_ops

Nested Class Summary

Nested classes/interfaces inherited from class org.bytedeco.javacpp.Pointer

Pointer.CustomDeallocator, Pointer.Deallocator, Pointer.NativeDeallocator, Pointer.ReferenceCounter

Field Summary

Fields inherited from class org.bytedeco.javacpp.Pointer

address, capacity, limit, position

Constructor Summary

Constructors

Constructor and Description
post_ops()
post_ops(dnnl_post_ops t)
post_ops(dnnl_post_ops t, boolean weak)
post_ops(long size) Native array allocator.
post_ops(Pointer p) Pointer cast constructor.
post_ops(post_ops arg0)

Method Summary

Modifier and Type	Method and Description
void	append_binary(dnnl.algorithm aalgorithm, memory.desc src1_desc) Appends a binary post-op.
void	append_binary(int aalgorithm, memory.desc src1_desc)
void	append_dw_k3s1p1(memory.data_type weights_data_type, memory.data_type bias_data_type, memory.data_type dst_data_type, int mask, float[] scales)
void	append_dw_k3s1p1(memory.data_type weights_data_type, memory.data_type bias_data_type, memory.data_type dst_data_type, int mask, FloatBuffer scales)
void	append_dw_k3s1p1(memory.data_type weights_data_type, memory.data_type bias_data_type, memory.data_type dst_data_type, int mask, FloatPointer scales) Appends a depthwise post-op convolution with stride 1.
void	append_dw_k3s2p1(memory.data_type weights_data_type, memory.data_type bias_data_type, memory.data_type dst_data_type, int mask, float[] scales)
void	append_dw_k3s2p1(memory.data_type weights_data_type, memory.data_type bias_data_type, memory.data_type dst_data_type, int mask, FloatBuffer scales)
void	append_dw_k3s2p1(memory.data_type weights_data_type, memory.data_type bias_data_type, memory.data_type dst_data_type, int mask, FloatPointer scales) Appends a depthwise post-op convolution with stride 2.
void	append_eltwise(float scale, dnnl.algorithm aalgorithm, float alpha, float beta) Appends an elementwise post-op.
void	append_eltwise(float scale, int aalgorithm, float alpha, float beta)
void	append_prelu(int mask) Appends a prelu forward post-op.
void	append_sum()
void	append_sum(float scale, int zero_point)
void	append_sum(float scale, int zero_point, memory.data_type data_type) Appends an accumulation (sum) post-op.
void	append_sum(float scale, memory.data_type data_type) Appends an accumulation (sum) post-op.
void	get_params_binary(int index, int[] aalgorithm, memory.desc src1_desc)
void	get_params_binary(int index, IntBuffer aalgorithm, memory.desc src1_desc)
void	get_params_binary(int index, IntPointer aalgorithm, memory.desc src1_desc) Returns the parameters of a binary post-op.
void	get_params_dw_k3s1p1(int index, memory.data_type weights_data_type, memory.data_type bias_data_type, memory.data_type dst_data_type, int[] mask, float[] scales)
void	get_params_dw_k3s1p1(int index, memory.data_type weights_data_type, memory.data_type bias_data_type, memory.data_type dst_data_type, IntBuffer mask, FloatBuffer scales)
void	get_params_dw_k3s1p1(int index, memory.data_type weights_data_type, memory.data_type bias_data_type, memory.data_type dst_data_type, IntPointer mask, FloatPointer scales) Returns the parameters of an depthwise post-op with stride 1.
void	get_params_dw_k3s2p1(int index, memory.data_type weights_data_type, memory.data_type bias_data_type, memory.data_type dst_data_type, int[] mask, float[] scales)
void	get_params_dw_k3s2p1(int index, memory.data_type weights_data_type, memory.data_type bias_data_type, memory.data_type dst_data_type, IntBuffer mask, FloatBuffer scales)
void	get_params_dw_k3s2p1(int index, memory.data_type weights_data_type, memory.data_type bias_data_type, memory.data_type dst_data_type, IntPointer mask, FloatPointer scales) Returns the parameters of an depthwise post-op with stride 2.
void	get_params_eltwise(int index, float[] scale, int[] aalgorithm, float[] alpha, float[] beta)
void	get_params_eltwise(int index, FloatBuffer scale, IntBuffer aalgorithm, FloatBuffer alpha, FloatBuffer beta)
void	get_params_eltwise(int index, FloatPointer scale, IntPointer aalgorithm, FloatPointer alpha, FloatPointer beta) Returns parameters of an elementwise post-op.
void	get_params_prelu(int index, int[] mask)
void	get_params_prelu(int index, IntBuffer mask)
void	get_params_prelu(int index, IntPointer mask) Returns the parameters of a prelu post-op.
void	get_params_sum(int index, float[] scale)
void	get_params_sum(int index, float[] scale, int[] zero_point, memory.data_type data_type)
void	get_params_sum(int index, float[] scale, memory.data_type data_type)
void	get_params_sum(int index, FloatBuffer scale)
void	get_params_sum(int index, FloatBuffer scale, IntBuffer zero_point, memory.data_type data_type)
void	get_params_sum(int index, FloatBuffer scale, memory.data_type data_type)
void	get_params_sum(int index, FloatPointer scale) Returns the parameters of an accumulation (sum) post-op.
void	get_params_sum(int index, FloatPointer scale, IntPointer zero_point, memory.data_type data_type) Returns the parameters of an accumulation (sum) post-op.
void	get_params_sum(int index, FloatPointer scale, memory.data_type data_type) Returns the parameters of an accumulation (sum) post-op.
post_ops	getPointer(long i)
primitive.kind	kind(int index) Returns the primitive kind of post-op at entry with a certain index.
int	len() Returns the number of post-ops entries.
post_ops	position(long position)

Methods inherited from class org.bytedeco.dnnl.dnnl_post_ops_handle

asBoolean, asDnnl_post_ops, equals, get, get, notEquals, put, reset, reset

Methods inherited from class org.bytedeco.javacpp.Pointer

address, asBuffer, asByteBuffer, availablePhysicalBytes, calloc, capacity, capacity, close, deallocate, deallocate, deallocateReferences, deallocator, deallocator, equals, fill, formatBytes, free, getPointer, getPointer, getPointer, hashCode, interruptDeallocatorThread, isNull, isNull, limit, limit, malloc, maxBytes, maxPhysicalBytes, memchr, memcmp, memcpy, memmove, memset, offsetAddress, offsetof, offsetof, parseBytes, physicalBytes, position, put, realloc, referenceCount, releaseReference, retainReference, setNull, sizeof, sizeof, toString, totalBytes, totalCount, totalPhysicalBytes, withDeallocator, zero

Methods inherited from class java.lang.Object

clone, finalize, getClass, notify, notifyAll, wait, wait, wait

Constructor Detail

post_ops

public post_ops()

post_ops

public post_ops(@Const @ByRef
                post_ops arg0)

post_ops

public post_ops(dnnl_post_ops t,
                @Cast(value="bool")
                boolean weak)

post_ops

public post_ops(dnnl_post_ops t)

post_ops

public post_ops(Pointer p)

Pointer cast constructor. Invokes Pointer(Pointer).

post_ops

public post_ops(long size)

Native array allocator. Access with Pointer.position(long).

Method Detail

position

public post_ops position(long position)

Overrides:

position in class dnnl_post_ops_handle

getPointer

public post_ops getPointer(long i)

Overrides:

getPointer in class dnnl_post_ops_handle

len

public int len()

Returns the number of post-ops entries.

kind

public primitive.kind kind(int index)

Returns the primitive kind of post-op at entry with a certain index.

Parameters:

index - Index of the post-op to return the kind for.

Returns:

Primitive kind of the post-op at the specified index.

append_sum

public void append_sum(float scale,
                       memory.data_type data_type)

Appends an accumulation (sum) post-op. Prior to accumulating the result, the previous value would be multiplied by a scaling factor \p scale. The kind of this post-op is #dnnl::primitive::kind::sum. This feature may improve performance for cases like residual learning blocks, where the result of convolution is accumulated to the previously computed activations. The parameter \p scale may be used for the integer-based computations when the result and previous activations have different logical scaling factors. In the simplest case when the accumulation is the only post-op, the computations will be dst[:] := scale * dst[:] + op(...) instead of dst[:] := op(...). If \p data_type is specified, the original dst tensor will be reinterpreted as a tensor with the provided data type. Because it is a reinterpretation, data_type and dst data type should have the same size. As a result, computations will be dst[:] <- scale * as_data_type(dst[:]) + op(...) instead of dst[:] <- op(...). \note This post-op executes in-place and does not change the destination layout.

Parameters:

scale - Scaling factor.

data_type - Data type.

append_sum

public void append_sum()

append_sum

public void append_sum(float scale,
                       int zero_point,
                       memory.data_type data_type)

Appends an accumulation (sum) post-op. Prior to accumulating the result, the previous value will be will be reduced by zero point \p zero_point and multiplied by a scaling factor \p scale. The kind of this post-op is #dnnl::primitive::kind::sum. This feature may improve performance for cases like dequantize the asymmetrically quantized sum's src1 tensor to f32 domain before performing the sum operation by subtracting \p zero_point before the scaling. In the simplest case when the accumulation is the only post-op, the computations will be dst[:] := scale * (dst[:] - zero_point) + op(...) instead of dst[:] := op(...). If \p data_type is specified, the original dst tensor will be reinterpreted as a tensor with the provided data type. Because it is a reinterpretation, data_type and dst data type should have the same size. As a result, computations will be dst[:] <- scale * (as_data_type(dst[:]) - zero_point) + op(...) instead of dst[:] <- op(...). \note This post-op executes in-place and does not change the destination layout.

Parameters:

scale - Scaling factor.

zero_point - Zero point.

data_type - Data type.

append_sum

public void append_sum(float scale,
                       int zero_point)

get_params_sum

public void get_params_sum(int index,
                           @ByRef
                           FloatPointer scale)

Returns the parameters of an accumulation (sum) post-op.

Parameters:

index - Index of the sum post-op.

scale - Scaling factor of the sum post-op.

get_params_sum

public void get_params_sum(int index,
                           @ByRef
                           FloatBuffer scale)

get_params_sum

public void get_params_sum(int index,
                           @ByRef
                           float[] scale)

get_params_sum

public void get_params_sum(int index,
                           @ByRef
                           FloatPointer scale,
                           memory.data_type data_type)

Returns the parameters of an accumulation (sum) post-op.

Parameters:

index - Index of the sum post-op.

scale - Scaling factor of the sum post-op.

data_type - Data type of the sum post-op.

get_params_sum

public void get_params_sum(int index,
                           @ByRef
                           FloatBuffer scale,
                           memory.data_type data_type)

get_params_sum

public void get_params_sum(int index,
                           @ByRef
                           float[] scale,
                           memory.data_type data_type)

get_params_sum

public void get_params_sum(int index,
                           @ByRef
                           FloatPointer scale,
                           @ByRef
                           IntPointer zero_point,
                           memory.data_type data_type)

Returns the parameters of an accumulation (sum) post-op.

Parameters:

index - Index of the sum post-op.

scale - Scaling factor of the sum post-op.

zero_point - Single scalar int32_t value of zeropoint.

data_type - Data type of the sum post-op.

get_params_sum

public void get_params_sum(int index,
                           @ByRef
                           FloatBuffer scale,
                           @ByRef
                           IntBuffer zero_point,
                           memory.data_type data_type)

get_params_sum

public void get_params_sum(int index,
                           @ByRef
                           float[] scale,
                           @ByRef
                           int[] zero_point,
                           memory.data_type data_type)

append_eltwise

public void append_eltwise(float scale,
                           dnnl.algorithm aalgorithm,
                           float alpha,
                           float beta)

Appends an elementwise post-op. The kind of this post-op is #dnnl::primitive::kind::eltwise. In the simplest case when the elementwise is the only post-op, the computations would be dst[:] := scale * eltwise_op (op(...)) instead of dst[:] <- op(...), where eltwise_op is configured with the given parameters.

Parameters:

scale - Scaling factor.

aalgorithm - Elementwise algorithm.

alpha - Alpha parameter for the elementwise algorithm.

beta - Beta parameter for the elementwise algorithm.

append_eltwise

public void append_eltwise(float scale,
                           @Cast(value="dnnl::algorithm")
                           int aalgorithm,
                           float alpha,
                           float beta)

get_params_eltwise

public void get_params_eltwise(int index,
                               @ByRef
                               FloatPointer scale,
                               @ByRef @Cast(value="dnnl::algorithm*")
                               IntPointer aalgorithm,
                               @ByRef
                               FloatPointer alpha,
                               @ByRef
                               FloatPointer beta)

Returns parameters of an elementwise post-op.

Parameters:

index - Index of the post-op.

scale - Output scaling factor.

aalgorithm - Output elementwise algorithm kind.

alpha - Output alpha parameter for the elementwise algorithm.

beta - Output beta parameter for the elementwise algorithm.

get_params_eltwise

public void get_params_eltwise(int index,
                               @ByRef
                               FloatBuffer scale,
                               @ByRef @Cast(value="dnnl::algorithm*")
                               IntBuffer aalgorithm,
                               @ByRef
                               FloatBuffer alpha,
                               @ByRef
                               FloatBuffer beta)

get_params_eltwise

public void get_params_eltwise(int index,
                               @ByRef
                               float[] scale,
                               @ByRef @Cast(value="dnnl::algorithm*")
                               int[] aalgorithm,
                               @ByRef
                               float[] alpha,
                               @ByRef
                               float[] beta)

append_dw_k3s1p1

public void append_dw_k3s1p1(memory.data_type weights_data_type,
                             memory.data_type bias_data_type,
                             memory.data_type dst_data_type,
                             int mask,
                             @StdVector
                             FloatPointer scales)

Appends a depthwise post-op convolution with stride 1. This post-op can only be fused with a 2D 1x1 convolution (convolution with weights spatial dimension equal to 1 i.e., kh=kw=1). The kind of this post-op is #dnnl_convolution. The number of outputs for primitive remain same as before. The output size remain same as the original primitive due to stride=1. The Post-op can be defined as: dst[:] <- scales * (conv_dw(conv_1x1)) See \ref dev_guide_attributes_post_ops_depthwise and \ref dev_guide_attributes_post_ops_depthwise_fusion for more info.

Parameters:

weights_data_type - Weights data type of depthwise post-op

bias_data_type - Bias data type of depthwise post-op

dst_data_type - Output data type of depthwise post-op

mask - Output scaling factors correspondence mask that defines the correspondence between the output tensor dimensions and the \p scales array. The set i-th bit indicates that a dedicated output scaling factor is used for each index along that dimension. The mask value of 0 implies a common scaling factor for the whole output tensor.

scales - Output pointer to a constant array of float scaling factors.

append_dw_k3s1p1

public void append_dw_k3s1p1(memory.data_type weights_data_type,
                             memory.data_type bias_data_type,
                             memory.data_type dst_data_type,
                             int mask,
                             @StdVector
                             FloatBuffer scales)

append_dw_k3s1p1

public void append_dw_k3s1p1(memory.data_type weights_data_type,
                             memory.data_type bias_data_type,
                             memory.data_type dst_data_type,
                             int mask,
                             @StdVector
                             float[] scales)

get_params_dw_k3s1p1

public void get_params_dw_k3s1p1(int index,
                                 memory.data_type weights_data_type,
                                 memory.data_type bias_data_type,
                                 memory.data_type dst_data_type,
                                 @ByRef
                                 IntPointer mask,
                                 @StdVector
                                 FloatPointer scales)

Returns the parameters of an depthwise post-op with stride 1.

Parameters:

index - Index of the elementwise post-op.

weights_data_type - Weights data type of depthwise post-op

bias_data_type - Bias data type of depthwise post-op

dst_data_type - Output data type of depthwise post-op

mask - Output scaling factors correspondence mask that defines the correspondence between the output tensor dimensions and the \p scales array. The set i-th bit indicates that a dedicated output scaling factor is used for each index along that dimension. The mask value of 0 implies a common scaling factor for the whole output tensor.

scales - Output pointer to a constant array of float scaling factors.

get_params_dw_k3s1p1

public void get_params_dw_k3s1p1(int index,
                                 memory.data_type weights_data_type,
                                 memory.data_type bias_data_type,
                                 memory.data_type dst_data_type,
                                 @ByRef
                                 IntBuffer mask,
                                 @StdVector
                                 FloatBuffer scales)

get_params_dw_k3s1p1

public void get_params_dw_k3s1p1(int index,
                                 memory.data_type weights_data_type,
                                 memory.data_type bias_data_type,
                                 memory.data_type dst_data_type,
                                 @ByRef
                                 int[] mask,
                                 @StdVector
                                 float[] scales)

append_dw_k3s2p1

public void append_dw_k3s2p1(memory.data_type weights_data_type,
                             memory.data_type bias_data_type,
                             memory.data_type dst_data_type,
                             int mask,
                             @StdVector
                             FloatPointer scales)

Appends a depthwise post-op convolution with stride 2. This post-op can only be fused with a 2D 1x1 convolution (convolution with weights spatial dimension equal to 1 i.e., kh=kw=1). The kind of this post-op is #dnnl_convolution. The number of outputs for primitive remain same as before. The output spatial size can be derived as below: output_height = ceil(output_height_1x1_convolution, stride) output_width = ceil(output_width_1x1_convolution, stride) The Post-op can be defined as: dst[:] <- scales * (conv_dw(conv_1x1)) See \ref dev_guide_attributes_post_ops_depthwise and \ref dev_guide_attributes_post_ops_depthwise_fusion for more info.

Parameters:

weights_data_type - Weights data type of depthwise post-op

bias_data_type - Bias data type of depthwise post-op

dst_data_type - Output data type of depthwise post-op

mask - Output scaling factors correspondence mask that defines the correspondence between the output tensor dimensions and the \p scales array. The set i-th bit indicates that a dedicated output scaling factor is used for each index along that dimension. The mask value of 0 implies a common scaling factor for the whole output tensor.

scales - Output pointer to a constant array of float scaling factors.

append_dw_k3s2p1

public void append_dw_k3s2p1(memory.data_type weights_data_type,
                             memory.data_type bias_data_type,
                             memory.data_type dst_data_type,
                             int mask,
                             @StdVector
                             FloatBuffer scales)

append_dw_k3s2p1

public void append_dw_k3s2p1(memory.data_type weights_data_type,
                             memory.data_type bias_data_type,
                             memory.data_type dst_data_type,
                             int mask,
                             @StdVector
                             float[] scales)

get_params_dw_k3s2p1

public void get_params_dw_k3s2p1(int index,
                                 memory.data_type weights_data_type,
                                 memory.data_type bias_data_type,
                                 memory.data_type dst_data_type,
                                 @ByRef
                                 IntPointer mask,
                                 @StdVector
                                 FloatPointer scales)

Returns the parameters of an depthwise post-op with stride 2.

Parameters:

index - Index of the elementwise post-op.

weights_data_type - Weights data type of depthwise post-op

bias_data_type - Bias data type of depthwise post-op

dst_data_type - Output data type of depthwise post-op

mask - Output scaling factors correspondence mask that defines the correspondence between the output tensor dimensions and the \p scales array. The set i-th bit indicates that a dedicated output scaling factor is used for each index along that dimension. The mask value of 0 implies a common scaling factor for the whole output tensor.

scales - Output pointer to a constant array of float scaling factors.

get_params_dw_k3s2p1

public void get_params_dw_k3s2p1(int index,
                                 memory.data_type weights_data_type,
                                 memory.data_type bias_data_type,
                                 memory.data_type dst_data_type,
                                 @ByRef
                                 IntBuffer mask,
                                 @StdVector
                                 FloatBuffer scales)

get_params_dw_k3s2p1

public void get_params_dw_k3s2p1(int index,
                                 memory.data_type weights_data_type,
                                 memory.data_type bias_data_type,
                                 memory.data_type dst_data_type,
                                 @ByRef
                                 int[] mask,
                                 @StdVector
                                 float[] scales)

append_binary

public void append_binary(dnnl.algorithm aalgorithm,
                          @Const @ByRef
                          memory.desc src1_desc)

Appends a binary post-op. The kind of this post operation is #dnnl_binary. In the simplest case when the binary is the only post operation, the computations would be: dst[:] <- binary_op (dst[:], another_input[:]) where binary_op is configured with the given parameters. binary_op supports broadcast semantics for a second operand.

Parameters:

aalgorithm - Binary algorithm for the post-op.

src1_desc - Memory descriptor of a second operand.

append_binary

public void append_binary(@Cast(value="dnnl::algorithm")
                          int aalgorithm,
                          @Const @ByRef
                          memory.desc src1_desc)

get_params_binary

public void get_params_binary(int index,
                              @ByRef @Cast(value="dnnl::algorithm*")
                              IntPointer aalgorithm,
                              @ByRef
                              memory.desc src1_desc)

Returns the parameters of a binary post-op.

Parameters:

index - Index of the binary post-op.

aalgorithm - Output binary algorithm kind.

src1_desc - Output memory descriptor of a second operand.

get_params_binary

public void get_params_binary(int index,
                              @ByRef @Cast(value="dnnl::algorithm*")
                              IntBuffer aalgorithm,
                              @ByRef
                              memory.desc src1_desc)

get_params_binary

public void get_params_binary(int index,
                              @ByRef @Cast(value="dnnl::algorithm*")
                              int[] aalgorithm,
                              @ByRef
                              memory.desc src1_desc)

append_prelu

public void append_prelu(int mask)

Appends a prelu forward post-op. The kind of this post-op is #dnnl::primitive::kind::prelu. The post-op can be defined as: dst[:] <- prelu(dst[:], weights[:]) prelu: dst[:] <- dst[:] if dst[:] > 0 dst[:] <- dst[:] * weights[:] if dst[:] <= 0 Example usage:

      int mb = 32, oc = 32,
          oh = 14, ow = 14; // convolution output params
      // unique weights per output channel
      vector weights = { ... };
      int oc_dim = 1; // mb_dim = 0, channel_dim = 1, height_dim = 2, ...
 
      // construct a convolution descriptor
      dnnl::convolution::desc conv_d;
 
      dnnl::primitive_attr attr;
      attr.append_prelu(1 << oc_dim);
 
      dnnl::primitive_desc conv_pd(conv_d, attr, engine);
      memory prelu_weights({{1}, dt::f32, {1}}, eng, weights.data());
 
      std::unordered_map conv_args;
 
      conv_args.insert(
       {DNNL_ARG_ATTR_MULTIPLE_POST_OP(0) | DNNL_ARG_WEIGHTS, prelu_weights})

Parameters:

mask - Defines the correspondence between the output tensor dimensions and the prelu weights tensor. The set i-th bit indicates that a dedicated weights value is used for each index along that dimension. Set the mask to 0 to use a common weights value for the whole output tensor.

get_params_prelu

public void get_params_prelu(int index,
                             @ByRef
                             IntPointer mask)

Returns the parameters of a prelu post-op.

Parameters:

index - Index of the prelu post-op.

maks - Weights mask of prelu post-op.

get_params_prelu

public void get_params_prelu(int index,
                             @ByRef
                             IntBuffer mask)

get_params_prelu

public void get_params_prelu(int index,
                             @ByRef
                             int[] mask)