%% -*-Fundamental-*- %define-kinds canonical ndk hal_1.0 hal_1.1 hal_1.2 hal_1.3 aidl %kind ndk %define ANN ANEURALNETWORKS_ %define Ann ANeuralNetworks %define DeclareOperation ANEURALNETWORKS_%{1} = %{2} %define DeclareOperation_1.2 ANEURALNETWORKS_%{1} = %{2} %define DeclareOperation_1.3 ANEURALNETWORKS_%{1} = %{2} %define DeclareOperation_fl6 ANEURALNETWORKS_%{1} = %{2} %define DeclareOperation_fl7 ANEURALNETWORKS_%{1} = %{2} %define FusedActivationFunc FuseCode %define DeclareFusedActivationFunc ANEURALNETWORKS_FUSED_%{1} = %{2} %define DeclareExecutionPreference ANEURALNETWORKS_PREFER_%{1} = %{2} %define DeclareDeviceType ANEURALNETWORKS_DEVICE_%{1} = %{2} %define OperandType OperandCode %define OperandTypeLinkPfx ANEURALNETWORKS_ %define OperationTypeLinkPfx ANEURALNETWORKS_ %define runtime_or_driver runtime %define NNAPILevel3 NNAPI feature level 3 %define NNAPILevel4 NNAPI feature level 4 %define NNAPILevel6 NNAPI feature level 6 %define NNAPILevel7 NNAPI feature level 7 %define BeforeNNAPILevel3For Before NNAPI feature level 3, for %define or_1.2 or {@link ANEURALNETWORKS_%{1}} %define NDK_if_specified (if specified) %define otherOperandParameters other operand parameters %section AVAIL1 * * Available since NNAPI feature level 1. %/section %section AVAIL1Short * * Available since NNAPI feature level 1. %/section %section AVAIL2 * * Available since NNAPI feature level 2. %/section %section AVAIL3 * * Available since NNAPI feature level 3. %/section %section AVAIL4 * * Available since NNAPI feature level 4. %/section %section AVAIL6 * * Available since NNAPI feature level 6. %/section %section AVAIL7 * * Available since NNAPI feature level 7. %/section %section OutputState * * Important: As of NNAPI feature level 3, there is no way to get the output state tensors out * and NNAPI does not maintain internal states. This operator does not support the usage pattern * in which multiple cells are chained and state tensors are propagated. %/section %section PaddingCodeValues * {@link PaddingCode} values. %/section %/kind %kind aidl canonical hal* %define ANN %define Ann %define FusedActivationFunc FusedActivationFunc %define DeclareFusedActivationFunc %{1} = %{2} %define DeclareExecutionPreference %{1} = %{2} %define DeclareDeviceType %{1} = %{2} %define OperandType OperandType %define OperandTypeLinkPfx OperandType:: %define OperationTypeLinkPfx OperationType:: %define runtime_or_driver driver %define NNAPILevel3 HAL version 1.2 %define NNAPILevel4 HAL version 1.3 %define NNAPILevel6 NNAPI feature level 6 %define NNAPILevel7 NNAPI feature level 7 %define NDK_if_specified %define otherOperandParameters extraParams %section AVAIL1 %/section %section AVAIL1Short %/section %section AVAIL2 %/section %section AVAIL3 %/section %section AVAIL4 %/section %section AVAIL6 %/section %section AVAIL7 %/section %section PaddingCodeValues * following values: {0 (NONE), 1 (SAME), 2 (VALID)}. %/section %section OutputState %/section %/kind %kind hal_1.0 hal_1.1 %define DeclareOperation %{1} = %{2} %define BeforeNNAPILevel3For For %define or_1.2 %section NHWC_NCHW * Supported tensor rank: 4, with "NHWC" (i.e., Num_samples, Height, Width, * and Channels) data layout. %/section %section GenericZero %/section %section ZeroBatchesNNAPILevel3 %/section %define DeclareOperation_1.2 @@@NOT_DEFINED@@@ %define DeclareOperation_1.3 @@@NOT_DEFINED@@@ %define DeclareOperation_fl6 @@@NOT_DEFINED@@@ %define DeclareOperation_fl7 @@@NOT_DEFINED@@@ %/kind %kind aidl canonical hal_1.2 hal_1.3 %define BeforeNNAPILevel3For Before HAL version 1.2, for %define or_1.2 or {@link OperandType::%{1}} %/kind %kind hal_1.2 %define DeclareOperation %{1} = @1.1::OperationType:%{1} %define DeclareOperation_1.2 %{1} = %{2} %define DeclareOperation_1.3 @@@NOT_DEFINED@@@ %define DeclareOperation_fl6 @@@NOT_DEFINED@@@ %define DeclareOperation_fl7 @@@NOT_DEFINED@@@ %/kind %kind hal_1.3 %define DeclareOperation %{1} = @1.2::OperationType:%{1} %define DeclareOperation_1.2 %{1} = @1.2::OperationType:%{1} %define DeclareOperation_1.3 %{1} = %{2} %define DeclareOperation_fl6 @@@NOT_DEFINED@@@ %define DeclareOperation_fl7 @@@NOT_DEFINED@@@ %/kind %kind aidl %define DeclareOperation %{1} = %{2} %define DeclareOperation_1.2 %{1} = %{2} %define DeclareOperation_1.3 %{1} = %{2} %define DeclareOperation_fl6 %{1} = %{2} %define DeclareOperation_fl7 %{1} = %{2} %define DeclareEnumValue %{1} = %{2} %define OperandLifeTime OperandLifeTime %define :: :: %define vec std::vector %define string std::string %define init_bool = false %define init_float = 0.0f %define init_int = 0 %define init_pod {} %define Dimensions Dimensions %define concat_or_skip_first %{2} %/kind %kind canonical %define DeclareOperation %{1} = %{2} %define DeclareOperation_1.2 %{1} = %{2} %define DeclareOperation_1.3 %{1} = %{2} %define DeclareOperation_fl6 %{1} = %{2} %define DeclareOperation_fl7 %{1} = %{2} %define DeclareEnumValue %{1} = %{2} %define OperandLifeTime Operand::LifeTime %define :: :: %define vec std::vector %define string std::string %define init_bool = false %define init_float = 0.0f %define init_int = 0 %define init_pod {} %define Dimensions Dimensions %define concat_or_skip_first %{2} %/kind %kind hal* %define DeclareEnumValue %{1} %define OperandLifeTime OperandLifeTime %define :: . %define vec vec %define string string %define init_bool %define init_float %define init_int %define init_pod %define Dimensions vec %define concat_or_skip_first %{1}%{2} %/kind %kind ndk %define DeclareEnumValue @@@NOT_DEFINED@@@ %define OperandLifeTime @@@NOT_DEFINED@@@ %define :: @@@NOT_DEFINED@@@ %define vec @@@NOT_DEFINED@@@ %define string @@@NOT_DEFINED@@@ %define init_bool @@@NOT_DEFINED@@@ %define init_float @@@NOT_DEFINED@@@ %define init_int @@@NOT_DEFINED@@@ %define init_pod @@@NOT_DEFINED@@@ %define Dimensions @@@NOT_DEFINED@@@ %define concat_or_skip_first @@@NOT_DEFINED@@@ %/kind %kind aidl canonical ndk hal_1.2 hal_1.3 %section NHWC_NCHW * Supported tensor rank: 4, with "NHWC" or "NCHW" data layout. * With the default data layout NHWC, the data is stored in the order of: * [batch, height, width, channels]. Alternatively, the data layout could * be NCHW, the data storage order of: [batch, channels, height, width]. * NCHW is supported since %{NNAPILevel3}. %/section %section GenericZero * Since %{NNAPILevel3}, generic zero-sized input tensor is supported. Zero * dimension is only compatible with 0 or 1. The size of the output * dimension is zero if either of corresponding input dimension is zero. * %/section %section ZeroBatchesNNAPILevel3 * Since %{NNAPILevel3}, zero batches is supported for this tensor. %/section %/kind %kind aidl canonical ndk hal_1.3 %define AndQuant8Signed %/kind %kind hal_1.0 hal_1.1 hal_1.2 %define AndQuant8Signed %/kind %kind ndk hal_1.0 hal_1.1 hal_1.2 %define model_or_subgraph model %define MODEL_or_SUBGRAPH MODEL %define the_model_or_a_subgraph the model %/kind %kind aidl canonical hal_1.3+ %define model_or_subgraph subgraph %define MODEL_or_SUBGRAPH SUBGRAPH %define the_model_or_a_subgraph a subgraph %/kind %% Declaring enums that work across all kinds: %% %% %{enum X underlying_hal_type} { %% %{DeclareX ...}, %% ... %% }%{ndk_enum_name X}; %% %% Note that %{ndk_enum_name X} can be omitted for non-NDK enums because the %% macro definition is empty for all other kinds. %kind aidl %define enum enum %{1} %define ndk_enum_name %define DeclarePriority %{1} = %{2} %/kind %kind canonical %define enum enum class %{1} %define ndk_enum_name %define DeclarePriority %{1} = %{2} %/kind %kind ndk %define enum typedef enum %define ndk_enum_name %{1} %define DeclarePriority ANEURALNETWORKS_PRIORITY_%{1} = %{3} %/kind %kind hal* %define enum enum %{1} : %{2} %define ndk_enum_name %define DeclarePriority %{1} %/kind %section OEMDeprecationAndOperandTypeRangeMaxComment /* * DEPRECATED. Since HAL version 1.2, extensions are the preferred * alternative to OEM operation and data types. * * OEM specific scalar value. * OEM = 10000, */ /* * DEPRECATED. Since HAL version 1.2, extensions are the preferred * alternative to OEM operation and data types. * * A tensor of OEM specific values. * TENSOR_OEM_BYTE = 10001, */ /* ADDING A NEW FUNDAMENTAL TYPE REQUIRES UPDATING THE VALUE OF * OperandTypeRange::FUNDAMENTAL_MAX. */ /* ADDING A NEW OEM TYPE REQUIRES UPDATING THE VALUE OF * OperandTypeRange::OEM_MAX. */ %/section %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% HAL OperandType for 1.0 %% NDK OperandCode for API 27 %section canonical_empty_line %kind canonical %/kind %/section %section Operand_1.0_Comment /** * Operand types. * * The type of an operand in a model. * * Types prefaced with %{ANN}TENSOR_* must be used for tensor data (i.e., tensors * with at least one dimension). Types not prefaced by %{ANN}TENSOR_* represent * scalar values and must have no dimensions. * * Although we define many types, most operators accept just a few * types. Most used are {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32}, * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, * and {@link %{OperandTypeLinkPfx}INT32}. %insert AVAIL1Short */ %/section %section Operand_1.0 /** A 32 bit floating point scalar value. */ %{ANN}FLOAT32 = 0, %insert canonical_empty_line /** A signed 32 bit integer scalar value. */ %{ANN}INT32 = 1, %insert canonical_empty_line /** An unsigned 32 bit integer scalar value. */ %{ANN}UINT32 = 2, %insert canonical_empty_line /** A tensor of 32 bit floating point values. */ %{ANN}TENSOR_FLOAT32 = 3, %insert canonical_empty_line /** A tensor of 32 bit integer values. */ %{ANN}TENSOR_INT32 = 4, %insert canonical_empty_line /** * A tensor of 8 bit unsigned integers that represent real numbers. * * Attached to this tensor are two numbers that can be used to convert the * 8 bit integer to the real value and vice versa. These two numbers are: * - scale: a 32 bit floating point value greater than zero. * - zeroPoint: a 32 bit integer, in range [0, 255]. * * The formula is: * real_value = (integer_value - zeroPoint) * scale. */ %{ANN}TENSOR_QUANT8_ASYMM = 5, %/section %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% HAL OperationType for 1.0 %% NDK OperationCode for API 27 %section Operation_1.0_Comment /** * Operation types. * * The type of an operation in a model. %insert AVAIL1Short */ %/section %section Operation_1.0 /** * Adds two tensors, element-wise. * * Takes two input tensors of identical {@link %{OperandType}} and compatible * dimensions. The output is the sum of both input tensors, optionally * modified by an activation function. * * Two dimensions are compatible when: * 1. they are equal, or * 2. one of them is 1 * * The size of the output is the maximum size along each dimension of the * input operands. It starts with the trailing dimensions, and works its * way forward. * * Example: * * input1.dimension = {4, 1, 2} * input2.dimension = {5, 4, 3, 1} * output.dimension = {5, 4, 3, 2} * %insert GenericZero * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: up to 4 * * Inputs: * * 0: A tensor. * * 1: A tensor of the same {@link %{OperandType}}, and compatible dimensions * as input0. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scales and zeroPoint can be different from input0 scale and zeroPoint. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scales and zeroPoint can be different from input0 scale and zeroPoint. %/kind * * 2: An {@link %{OperandTypeLinkPfx}INT32} scalar, and has to be one of the * {@link %{FusedActivationFunc}} values. Specifies the activation to * invoke on the result. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_INT32} tensor, * the {@link %{FusedActivationFunc}} must be "NONE". %/kind * * Outputs: * * 0: The sum, a tensor of the same {@link %{OperandType}} as input0. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint can be different from inputs' scale and zeroPoint. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint can be different from inputs' scale and zeroPoint. %/kind %insert AVAIL1 */ %{DeclareOperation ADD 0}, /** * Performs a 2-D average pooling operation. * * The output dimensions are functions of the filter dimensions, stride, and * padding. * * The values in the output tensor are computed as: * * output[b, i, j, channel] = * sum_{di, dj}( * input[b, strides[1] * i + di, strides[2] * j + dj, channel] * ) / sum(1) * * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * %insert NHWC_NCHW * * Both explicit padding and implicit padding are supported. * * Inputs (explicit padding): * * 0: A 4-D tensor, of shape [batches, height, width, depth], specifying * the input. %insert ZeroBatchesNNAPILevel3 * * 1: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the left, in the ‘width’ dimension. * * 2: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the right, in the ‘width’ dimension. * * 3: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the top, in the ‘height’ dimension. * * 4: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the bottom, in the ‘height’ dimension. * * 5: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘width’ dimension. * * 6: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘height’ dimension. * * 7: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the filter * width. * * 8: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the filter * height. * * 9: An {@link %{OperandTypeLinkPfx}INT32} scalar, and has to be one of the * {@link %{FusedActivationFunc}} values. Specifies the activation to * invoke on the result. %kind aidl canonical ndk hal_1.2+ * * 10: An optional {@link %{OperandTypeLinkPfx}BOOL} scalar, default to false. * Set to true to specify NCHW data layout for input0 and output0. * Available since %{NNAPILevel3}. %/kind * * Inputs (implicit padding): * * 0: A 4-D tensor, of shape [batches, height, width, depth], specifying * the input. %insert ZeroBatchesNNAPILevel3 * * 1: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the implicit * padding scheme, has to be one of the %insert PaddingCodeValues * * 2: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘width’ dimension. * * 3: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘height’ dimension. * * 4: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the filter * width. * * 5: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the filter * height. * * 6: An {@link %{OperandTypeLinkPfx}INT32} scalar, and has to be one of the * {@link %{FusedActivationFunc}} values. Specifies the activation to * invoke on the result. %kind aidl canonical ndk hal_1.2+ * * 7: An optional {@link %{OperandTypeLinkPfx}BOOL} scalar, default to false. * Set to true to specify NCHW data layout for input0 and output0. * Available since %{NNAPILevel3}. %/kind * * Outputs: * * 0: The output 4-D tensor, of shape * [batches, out_height, out_width, depth]. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint must be the same as input0. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint must be the same as input0. %/kind %insert AVAIL1 */ %{DeclareOperation AVERAGE_POOL_2D 1}, /** * Concatenates the input tensors along the given dimension. * * The input tensors must have identical {@link %{OperandType}} and the same * dimensions except the dimension along the concatenation axis. * * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.2+ * (full support since %{NNAPILevel3}, see the input section) %/kind %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: up to 4 * * Inputs: * * 0 ~ n-1: The list of n input tensors, of shape * [D0, D1, ..., Daxis(i), ..., Dm]. %kind aidl canonical ndk hal_1.2+ * Before %{NNAPILevel3}, all input tensors of %else * All input tensors of %/kind * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * must have the same scale and zeroPoint as the output tensor. %kind aidl canonical ndk hal_1.3+ * Input tensors of * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} * are allowed to have different scale and zeroPoint. %/kind %kind aidl canonical ndk hal_1.2+ * Since %{NNAPILevel3}, zero-sized tensors are supported. %/kind * * n: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the * concatenation axis. * * Outputs: * * 0: The output, a tensor of the same {@link %{OperandType}} as the input * tensors. The output shape is [D0, D1, ..., sum(Daxis(i)), ..., Dm]. %kind aidl canonical ndk hal_1.2+ * Since %{NNAPILevel3}, for a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint values can be different from * input tensors. Before %{NNAPILevel3} they have to be the same as for the * input tensors. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, the scale and zeroPoint * values must be the same as the input tensors'. %/kind %kind aidl canonical hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint values can be different from input tensors. %/kind %insert AVAIL1 */ %{DeclareOperation CONCATENATION 2}, /** * Performs a 2-D convolution operation. * * The CONV_2D op sweeps a 2-D filter that can mix channels together over a * batch of images, applying the filter to each window of each image of the * appropriate size. * * The output dimensions are functions of the filter dimensions, stride, and * padding. * * The values in the output tensor are computed as: * * output[b, i, j, channel] = * sum_{di, dj, k} ( * input[b, strides[1] * i + di, strides[2] * j + dj, k] * * filter[channel, di, dj, k] * ) + bias[channel] * * Supported tensor {@link %{OperandType}} configurations: * * 32 bit floating point: * * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} for input, filter, output, and bias. * * * Quantized: * * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} for input, filter, and output. * * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} for bias (with scale set to * * * input.scale * filter.scale). * %kind aidl canonical ndk hal_1.2+ * Available since %{NNAPILevel3}: * * 16 bit floating point: * * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} for input, filter, output, and bias. * * * Quantized with symmetric per channel quantization for the filter: * * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} for input, and output. * * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM_PER_CHANNEL} for filter. * * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} for bias (scale set to 0.0, * * * each value scaling is separate and equal to input.scale * filter.scales[channel]). * %/kind %kind aidl ndk hal_1.3+ * Available since %{NNAPILevel4}: * * Quantized signed (since %{NNAPILevel4}): * * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} for input, filter, and output. * * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} for bias (with scale set to * * * input.scale * filter.scale). * * * Quantized signed with filter symmetric per channel quantization * (since %{NNAPILevel4}): * * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} for input, and output. * * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM_PER_CHANNEL} for filter. * * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} for bias (scale set to 0.0, * * * each value scaling is separate and equal to input.scale * filter.scales[channel]). * %/kind %insert NHWC_NCHW * * Both explicit padding and implicit padding are supported. * * Inputs (explicit padding): * * 0: A 4-D tensor, of shape [batches, height, width, depth_in], * specifying the input. %insert ZeroBatchesNNAPILevel3 * * 1: A 4-D tensor, of shape * [depth_out, filter_height, filter_width, depth_in], specifying the * filter. %kind aidl canonical ndk hal_1.2+ * For tensor of type {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM_PER_CHANNEL} * the channel dimension (%{Ann}SymmPerChannelQuantParams::channelDim) * must be set to 0. %/kind * * 2: A 1-D tensor, of shape [depth_out], specifying the bias. For input * tensor of type {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * %{or_1.2 TENSOR_FLOAT16} the bias must be of the same type. %kind aidl canonical ndk hal_1.3+ * For filter tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * and {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED}, %else * For filter tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, %/kind * the bias should be of {@link %{OperandTypeLinkPfx}TENSOR_INT32}, with zeroPoint * of 0 and bias_scale == input_scale * filter_scale. %kind aidl canonical ndk hal_1.2+ * For filter tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM_PER_CHANNEL}, * the bias should be of {@link %{OperandTypeLinkPfx}TENSOR_INT32}, with zeroPoint of 0 * and bias_scale of 0. The actual scale of each value 'i' is equal to * bias_scale[i] = input_scale * filter_scale[i]. %/kind * * 3: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the left, in the ‘width’ dimension. * * 4: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the right, in the ‘width’ dimension. * * 5: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the top, in the ‘height’ dimension. * * 6: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the bottom, in the ‘height’ dimension. * * 7: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘width’ dimension. * * 8: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘height’ dimension. * * 9: An {@link %{OperandTypeLinkPfx}INT32} scalar, and has to be one of the * {@link %{FusedActivationFunc}} values. Specifies the activation to * invoke on the result. %kind aidl canonical ndk hal_1.2+ * * 10: An optional {@link %{OperandTypeLinkPfx}BOOL} scalar, default to false. * Set to true to specify NCHW data layout for input0 and output0. * Available since %{NNAPILevel3}. * * 11: An optional {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the dilation * factor for width. Defaults to 1. If set to k > 1, there will be k-1 skipped * cells between each filter element on width dimension. If this input is set, * input 12 (dilation factor for height) must be specified as well. * Available since %{NNAPILevel3}. * * 12: An optional {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the dilation * factor for height. Defaults to 1. If set to k > 1, there will be k-1 skipped * cells between each filter element on height dimension. If this input is set, * input 11 (dilation factor for width) must be specified as well. * Available since %{NNAPILevel3}. %/kind * * Inputs (implicit padding): * * 0: A 4-D tensor, of shape [batches, height, width, depth_in], * specifying the input. %insert ZeroBatchesNNAPILevel3 * * 1: A 4-D tensor, of shape * [depth_out, filter_height, filter_width, depth_in], specifying the * filter. %kind aidl canonical ndk hal_1.2+ * For tensor of type {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM_PER_CHANNEL} * the channel dimension (%{Ann}SymmPerChannelQuantParams::channelDim) * must be set to 0. %/kind * * 2: A 1-D tensor, of shape [depth_out], specifying the bias. For input * tensor of type {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * %{or_1.2 TENSOR_FLOAT16} the bias must be of the same * type. %kind aidl canonical ndk hal_1.3+ * For filter tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * and {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED}, %else * For filter tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, %/kind * the bias should be of {@link %{OperandTypeLinkPfx}TENSOR_INT32}, with zeroPoint * of 0 and bias_scale == input_scale * filter_scale. %kind aidl canonical ndk hal_1.2+ * For filter tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM_PER_CHANNEL}, * the bias should be of {@link %{OperandTypeLinkPfx}TENSOR_INT32}, with zeroPoint of 0 * and bias_scale of 0. The actual scale of each value 'i' is equal to * bias_scale[i] = input_scale * filter_scale[i]. %/kind * * 3: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the implicit * padding scheme, has to be one of the %insert PaddingCodeValues * * 4: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘width’ dimension. * * 5: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘height’ dimension. * * 6: An {@link %{OperandTypeLinkPfx}INT32} scalar, and has to be one of the * {@link %{FusedActivationFunc}} values. Specifies the activation to * invoke on the result. %kind aidl canonical ndk hal_1.2+ * * 7: An optional {@link %{OperandTypeLinkPfx}BOOL} scalar, default to false. * Set to true to specify NCHW data layout for input0 and output0. * Available since %{NNAPILevel3}. * * 8: An optional {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the dilation * factor for width. Defaults to 1. If set to k > 1, there will be k-1 skipped * cells between each filter element on width dimension. If this input is set, * input 9 (dilation factor for height) must be specified as well. * Available since %{NNAPILevel3}. * * 9: An optional {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the dilation * factor for height. Defaults to 1. If set to k > 1, there will be k-1 skipped * cells between each filter element on height dimension. If this input is set, * input 8 (dilation factor for width) must be specified as well. * Available since %{NNAPILevel3}. %/kind * * Outputs: * * 0: The output 4-D tensor, of shape * [batches, out_height, out_width, depth_out]. * %{BeforeNNAPILevel3For} output tensor of * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, the following condition must * be satisfied: output_scale > input_scale * filter_scale %insert AVAIL1 */ %{DeclareOperation CONV_2D 3}, /** * Performs a depthwise 2-D convolution operation. * * Given an input tensor of shape [batches, height, width, depth_in] and a * filter tensor of shape [1, filter_height, filter_width, depth_out] * containing depth_out convolutional filters of depth 1, DEPTHWISE_CONV * applies a different filter to each input channel (expanding from 1 * channel to channel_multiplier channels for each), then concatenates the * results together. * * The output has depth_out = depth_in * depth_multiplier channels. * The output dimensions are functions of the filter dimensions, stride, and * padding. * * The values in the output tensor are computed as: * * output[b, i, j, k * channel_multiplier + q] = * sum_{di, dj} ( * input[b, strides[1] * i + di, strides[2] * j + dj, k] * * filter[1, di, dj, k * channel_multiplier + q] * ) + bias[k * channel_multiplier + q] * * Supported tensor {@link %{OperandType}} configurations: * * 32 bit floating point: * * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} for input, filter, output, and bias. * * * Quantized: * * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} for input, filter, and output. * * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} for bias (with scale set to * * * input.scale * filter.scale). * %kind aidl canonical ndk hal_1.2+ * Available since %{NNAPILevel3}: * * 16 bit floating point: * * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} for input, filter, output, and bias. * * * Quantized with symmetric per channel quantization for the filter: * * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} for input, and output. * * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM_PER_CHANNEL} for filter. * * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} for bias (scale set to 0.0, * * * each value scaling is separate and equal to input.scale * filter.scales[channel]). * %/kind %kind aidl canonical ndk hal_1.3+ * Available since %{NNAPILevel4}: * * Quantized signed (since %{NNAPILevel4}): * * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} for input, filter, and output. * * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} for bias (with scale set to * * * input.scale * filter.scale). * * * Quantized signed with filter symmetric per channel quantization * (since %{NNAPILevel4}): * * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} for input, and output. * * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM_PER_CHANNEL} for filter. * * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} for bias (scale set to 0.0, * * * each value scaling is separate and equal to input.scale * filter.scales[channel]). * %/kind %insert NHWC_NCHW * * Both explicit padding and implicit padding are supported. * * Inputs (explicit padding): * * 0: A 4-D tensor, of shape [batches, height, width, depth_in], * specifying the input. * * 1: A 4-D tensor, of shape [1, filter_height, filter_width, depth_out], * specifying the filter. %kind aidl canonical ndk hal_1.2+ * For tensor of type {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM_PER_CHANNEL} * the channel dimension (%{Ann}SymmPerChannelQuantParams::channelDim) * must be set to 3. %/kind * * 2: A 1-D tensor, of shape [depth_out], specifying the bias. For input * tensor of type {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * %{or_1.2 TENSOR_FLOAT16} the bias must be of the same type. %kind aidl canonical ndk hal_1.3+ * For filter tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * and {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED}, %else * For filter tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, %/kind * the bias should be of {@link %{OperandTypeLinkPfx}TENSOR_INT32}, with zeroPoint * of 0 and bias_scale == input_scale * filter_scale. %kind aidl canonical ndk hal_1.2+ * For filter tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM_PER_CHANNEL}, * the bias should be of {@link %{OperandTypeLinkPfx}TENSOR_INT32}, with zeroPoint of 0 * and bias_scale of 0. The actual scale of each value 'i' is equal to * bias_scale[i] = input_scale * filter_scale[i]. %/kind * * 3: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the left, in the ‘width’ dimension. * * 4: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the right, in the ‘width’ dimension. * * 5: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the top, in the ‘height’ dimension. * * 6: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the bottom, in the ‘height’ dimension. * * 7: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘width’ dimension. * * 8: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘height’ dimension. * * 9: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the depthwise * multiplier. * * 10: An {@link %{OperandTypeLinkPfx}INT32} scalar, and has to be one of the * {@link %{FusedActivationFunc}} values. Specifies the activation to * invoke on the result. %kind aidl canonical ndk hal_1.2+ * * 11: An optional {@link %{OperandTypeLinkPfx}BOOL} scalar, default to false. * Set to true to specify NCHW data layout for input0 and output0. * Available since %{NNAPILevel3}. * * 12: An optional {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the dilation * factor for width. Defaults to 1. If set to k > 1, there will be k-1 skipped * cells between each filter element on width dimension. If this input is set, * input 13 (dilation factor for height) must be specified as well. * Available since %{NNAPILevel3}. * * 13: An optional {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the dilation * factor for height. Defaults to 1. If set to k > 1, there will be k-1 skipped * cells between each filter element on height dimension. If this input is set, * input 12 (dilation factor for width) must be specified as well. * Available since %{NNAPILevel3}. %/kind * * Inputs (implicit padding): * * 0: A 4-D tensor, of shape [batches, height, width, depth_in], * specifying the input. * * 1: A 4-D tensor, of shape [1, filter_height, filter_width, depth_out], * specifying the filter. * * 2: A 1-D tensor, of shape [depth_out], specifying the bias. For input * tensor of type {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * %{or_1.2 TENSOR_FLOAT16} the bias must be of the same type. %kind aidl canonical ndk hal_1.3+ * For filter tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * and {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED}, %else * For filter tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, %/kind * the bias should be of {@link %{OperandTypeLinkPfx}TENSOR_INT32}, with zeroPoint * of 0 and bias_scale == input_scale * filter_scale. %kind aidl canonical ndk hal_1.2+ * For filter tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM_PER_CHANNEL}, * the bias should be of {@link %{OperandTypeLinkPfx}TENSOR_INT32}, with zeroPoint of 0 * and bias_scale of 0. The actual scale of each value 'i' is equal to * bias_scale[i] = input_scale * filter_scale[i]. %/kind * * 3: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the implicit * padding scheme, has to be one of the %insert PaddingCodeValues * * 4: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘width’ dimension. * * 5: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘height’ dimension. * * 6: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the depthwise * multiplier. * * 7: An {@link %{OperandTypeLinkPfx}INT32} scalar, and has to be one of the * {@link %{FusedActivationFunc}} values. Specifies the activation to * invoke on the result. %kind aidl canonical ndk hal_1.2+ * * 8: An optional {@link %{OperandTypeLinkPfx}BOOL} scalar, default to false. * Set to true to specify NCHW data layout for input0 and output0. * Available since %{NNAPILevel3}. * * 9: An optional {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the dilation * factor for width. Defaults to 1. If set to k > 1, there will be k-1 skipped * cells between each filter element on width dimension. If this input is set, * input 10 (dilation factor for height) must be specified as well. * Available since %{NNAPILevel3}. * * 10: An optional {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the dilation * factor for height. Defaults to 1. If set to k > 1, there will be k-1 skipped * cells between each filter element on height dimension. If this input is set, * input 9 (dilation factor for width) must be specified as well. * Available since %{NNAPILevel3}. %/kind * * Outputs: * * 0: The output 4-D tensor, of shape * [batches, out_height, out_width, depth_out]. %{BeforeNNAPILevel3For} * output tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, * the following condition must be satisfied: * output_scale > input_scale * filter_scale %insert AVAIL1 */ %{DeclareOperation DEPTHWISE_CONV_2D 4}, /** * Rearranges data from depth into blocks of spatial data. * * More specifically, this op outputs a copy of the input tensor where * values from the depth dimension are moved in spatial blocks to the height * and width dimensions. The value block_size indicates the input block size * and how the data is moved. * * Chunks of data of size block_size * block_size from depth are rearranged * into non-overlapping blocks of size block_size x block_size. * * The width of the output tensor is input_depth * block_size, whereas the * height is input_height * block_size. The depth of the input tensor must * be divisible by block_size * block_size * * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * %insert NHWC_NCHW * * Inputs: * * 0: A 4-D tensor, of shape [batches, height, width, depth_in], * specifying the input. * * 1: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the block_size. * block_size must be >=1 and block_size * block_size must be a divisor * of the input depth. %kind aidl canonical ndk hal_1.2+ * * 2: An optional {@link %{OperandTypeLinkPfx}BOOL} scalar, default to false. * Set to true to specify NCHW data layout for input0 and output0. * Available since %{NNAPILevel3}. %/kind * * Outputs: * * 0: The output 4-D tensor, of shape [batch, height*block_size, * width*block_size, depth/(block_size*block_size)]. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint must be the same as input0. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint must be the same as input0. %/kind %insert AVAIL1 */ %{DeclareOperation DEPTH_TO_SPACE 5}, /** * Dequantizes the input tensor. * * The formula is: * * output = (input - zeroPoint) * scale. * * Supported input tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM} (since %{NNAPILevel3}) * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM_PER_CHANNEL} (since %{NNAPILevel3}) %/kind %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported output tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32}. * * Supported tensor rank: up to 4 * * Inputs: * * 0: A tensor. %kind aidl canonical ndk hal_1.2+ * Since %{NNAPILevel3}, this tensor may be zero-sized. %/kind * * Outputs: * * 0: A tensor with the same shape as input0. %insert AVAIL1 */ %{DeclareOperation DEQUANTIZE 6}, /** * Looks up sub-tensors in the input tensor. * * This operator takes for input a tensor of values (Values) and * a one-dimensional tensor of selection indices (Lookups). * The output tensor is the concatenation of sub-tensors of Values as * selected by Lookups. * * Think of Values as being sliced along its first dimension: * The entries in Lookups select which slices are concatenated together * to create the output tensor. * * For example, if Values has shape of [40, 200, 300] and * Lookups has shape of [3], all three values found in Lookups are * expected to be between 0 and 39. The resulting tensor must * have shape of [3, 200, 300]. * * If a value in Lookups is out of bounds, the operation must fail * and an error must be reported. * * Supported value tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel4}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} (since %{NNAPILevel3}) * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} (since %{NNAPILevel3}) %/kind %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported value tensor rank: from 2 * * Inputs: * * 0: Lookups. A 1-D tensor of {@link %{OperandTypeLinkPfx}TENSOR_INT32}. * The values are indices into the first dimension of Values. * * 1: Values. An n-D tensor, where n >= 2, from which sub-tensors are * extracted. * * Output: * * 0: A n-D tensor with the same rank and shape as the Values * tensor, except for the first dimension which has the same size * as Lookups' only dimension. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, %/kind * the scale and zeroPoint must be the same as input1. %insert AVAIL1 */ %{DeclareOperation EMBEDDING_LOOKUP 7}, /** * Computes element-wise floor() on the input tensor. * * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * Supported tensor rank: up to 4 * * Inputs: * * 0: A tensor. * * Outputs: * * 0: The output tensor, of the same {@link %{OperandType}} and dimensions as * the input tensor. %insert AVAIL1 */ %{DeclareOperation FLOOR 8}, /** * Denotes a fully (densely) connected layer, which connects all elements * in the input tensor with each element in the output tensor. * * This layer implements the operation: * * outputs = activation(inputs * weights’ + bias) * * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: up to 4. * * Inputs: * * 0: A tensor of at least rank 2, specifying the input. If rank is * greater than 2, then it gets flattened to a 2-D Tensor. The * (flattened) 2-D Tensor is reshaped (if necessary) to * [batch_size, input_size], where "input_size" corresponds to the * number of inputs to the layer, matching the second dimension of * weights, and "batch_size" is calculated by dividing the number of * elements by "input_size". %kind aidl canonical ndk hal_1.2+ * Since %{NNAPILevel3}, zero batch_size is supported for this tensor. %/kind * * 1: A 2-D tensor, specifying the weights, of shape * [num_units, input_size], where "num_units" corresponds to the number * of output nodes. * * 2: A 1-D tensor, of shape [num_units], specifying the bias. For input * tensor of {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32}, the bias should * also be of {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32}. %kind aidl canonical ndk hal_1.3+ * For input tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * and {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED}, %else * For input tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, %/kind * the bias should be of {@link %{OperandTypeLinkPfx}TENSOR_INT32}, * with zeroPoint of 0 and bias_scale == input_scale * filter_scale. * * 3: An {@link %{OperandTypeLinkPfx}INT32} scalar, and has to be one of the * {@link %{FusedActivationFunc}} values. Specifies the activation to * invoke on the result. * * Outputs: * * 0: The output tensor, of shape [batch_size, num_units]. %{BeforeNNAPILevel3For} * output tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, the following * condition must be satisfied: output_scale > input_scale * filter_scale. %insert AVAIL1 */ %{DeclareOperation FULLY_CONNECTED 9}, /** * Looks up sub-tensors in the input tensor using a key-value map. * * This operator takes for input a tensor of values (Values), * a one-dimensional tensor of selection values (Lookups) and * a one-dimensional tensor that maps these values to Values * indexes. The output tensor is the concatenation of sub-tensors of * Values as selected by Lookups via Keys. * * Think of Values as being sliced along its outer-most dimension. * The output is a concatenation of selected slices, with one slice * for each entry of Lookups. The slice selected is the one at the * same index as the Maps entry that matches the value in Lookups. * * For a hit, the corresponding sub-tensor of Values is included * in the Output tensor. For a miss, the corresponding sub-tensor in * Output must have zero values. * * For example, if Values has shape of [40, 200, 300], * Keys should have a shape of [40]. If Lookups tensor has shape * of [3], three slices are being concatenated, so the resulting tensor * must have the shape of [3, 200, 300]. If the first entry in Lookups * has the value 123456, that value must be located in Keys tensor. * If the sixth entry of Keys contains 123456, the sixth slice of Values * must be selected. If no entry in Keys has 123456, a slice of zeroes * must be concatenated. * * Supported value tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * * Supported value tensor rank: from 2 * * Inputs: * * 0: Lookups. A 1-D {@link %{OperandTypeLinkPfx}TENSOR_INT32} tensor with * shape [ k ]. * * 1: Keys. A 1-D {@link %{OperandTypeLinkPfx}TENSOR_INT32} tensor with shape * [ n ]; Keys and Values pair represent a map, i.e., the ith element * in Keys (Keys[i]) is the key to select the ith sub-tensor in Values * (Values[i]), where 0 <= i <= n-1. Keys tensor *MUST* be sorted in * ascending order. * * 2: Values. A tensor with shape of [ n, … ]; i.e., the first dimension * must be n. * * Outputs: * * 0: Output. A tensor with shape [ k …]. * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint must be the same as input2. * * 1: Hits. A boolean tensor with shape [ k ] indicates whether the lookup * hits (True) or not (False). * Stored as {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} with offset 0 * and scale 1.0f. * A non-zero byte represents True, a hit. A zero indicates otherwise. %insert AVAIL1 */ %{DeclareOperation HASHTABLE_LOOKUP 10}, /** * Applies L2 normalization along the axis dimension. * * The values in the output tensor are computed as: * * output[batch, row, col, channel] = * input[batch, row, col, channel] / * sqrt(sum_{c} pow(input[batch, row, col, c], 2)) * %kind aidl canonical ndk hal_1.2+ * By default the axis dimension is the last dimension of the input tensor. * %/kind * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} (since %{NNAPILevel3}) %/kind %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * %kind aidl canonical ndk hal_1.2+ * Supported tensor rank: up to 4 * Tensors with rank less than 4 are only supported since %{NNAPILevel3}. %else * Supported tensor rank: 4, with "NHWC" data layout (i.e., Num_samples, * Height, Width, and Channels). %/kind * * Inputs: %kind aidl canonical ndk hal_1.2+ * * 0: An n-D tensor, specifying the tensor to be normalized. %else * * 0: A 4-D tensor, specifying the tensor to be normalized. %/kind %kind aidl canonical ndk hal_1.2+ * * 1: An optional {@link %{OperandTypeLinkPfx}INT32} scalar, default to -1, * specifying the dimension normalization would be performed on. * Negative index is used to specify axis from the end (e.g. -1 for * the last axis). Must be in the range [-n, n). * Available since %{NNAPILevel3}. %/kind * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} and same shape as input0. %kind aidl canonical ndk hal_1.2+ * For {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, * the scale must be 1.f / 128 and the zeroPoint must be 128. %/kind %kind aidl canonical ndk hal_1.3+ * For {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED}, * the scale must be 1.f / 128 and the zeroPoint must be 0. * * NOTE: Before %{NNAPILevel4}, if the elements along an axis are all zeros, * the result is undefined. Since %{NNAPILevel4}, if the elements along an axis * are all zeros, the result is logical zero. %/kind %insert AVAIL1 */ %{DeclareOperation L2_NORMALIZATION 11}, /** * Performs an 2-D L2 pooling operation. * * The output dimensions are functions of the filter dimensions, stride, and * padding. * * The values in the output tensor are computed as: * * output[b, i, j, c] = * sqrt(sum_{di, dj} pow(input[b, strides[1] * i + di, strides[2] * j + dj, c], 2) / * sum(1)) * * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * %insert NHWC_NCHW * * Both explicit padding and implicit padding are supported. * * Inputs (explicit padding): * * 0: A 4-D tensor, of shape [batches, height, width, depth], specifying * the input. %insert ZeroBatchesNNAPILevel3 * * 1: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the left, in the ‘width’ dimension. * * 2: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the right, in the ‘width’ dimension. * * 3: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the top, in the ‘height’ dimension. * * 4: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the bottom, in the ‘height’ dimension. * * 5: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘width’ dimension. * * 6: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘height’ dimension. * * 7: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the filter * width. * * 8: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the filter * height. * * 9: An {@link %{OperandTypeLinkPfx}INT32} scalar, and has to be one of the * {@link %{FusedActivationFunc}} values. Specifies the activation to * invoke on the result. %kind aidl canonical ndk hal_1.2+ * * 10: An optional {@link %{OperandTypeLinkPfx}BOOL} scalar, default to false. * Set to true to specify NCHW data layout for input0 and output0. * Available since %{NNAPILevel3}. %/kind * * Inputs (implicit padding): * * 0: A 4-D tensor, of shape [batches, height, width, depth], specifying * the input. %insert ZeroBatchesNNAPILevel3 * * 1: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the implicit * padding scheme, has to be one of the %insert PaddingCodeValues * * 2: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘width’ dimension. * * 3: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘height’ dimension. * * 4: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the filter * width. * * 5: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the filter * height. * * 6: An {@link %{OperandTypeLinkPfx}INT32} scalar, and has to be one of the * {@link %{FusedActivationFunc}} values. Specifies the activation to * invoke on the result. %kind aidl canonical ndk hal_1.2+ * * 7: An optional {@link %{OperandTypeLinkPfx}BOOL} scalar, default to false. * Set to true to specify NCHW data layout for input0 and output0. * Available since %{NNAPILevel3}. %/kind * * Outputs: * * 0: The output 4-D tensor, of shape * [batches, out_height, out_width, depth]. %insert AVAIL1 */ %{DeclareOperation L2_POOL_2D 12}, /** * Applies Local Response Normalization along the depth dimension. * * The 4-D input tensor is treated as a 3-D array of 1-D vectors (along the * last dimension), and each vector is normalized independently. Within a * given vector, each component is divided by the weighted, squared sum of * inputs within depth_radius. * * The output is calculated using this formula: * * sqr_sum[a, b, c, d] = sum( * pow(input[a, b, c, d - depth_radius : d + depth_radius + 1], 2)) * output = input / pow((bias + alpha * sqr_sum), beta) * %kind aidl canonical ndk hal_1.2+ * For input tensor with rank less than 4, independently normalizes each * 1-D slice along specified dimension. * %/kind * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * %kind aidl canonical ndk hal_1.2+ * Supported tensor rank: up to 4 * Tensors with rank less than 4 are only supported since %{NNAPILevel3}. %else * Supported tensor rank: 4, with "NHWC" data layout. %/kind * * Inputs: * * 0: A 4-D tensor, of shape [batches, height, width, depth], specifying * the input. * * 1: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the radius of * the normalization window. * * 2: A scalar, specifying the bias, must not be zero. %kind aidl canonical ndk hal_1.2+ * For input tensor of {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16}, the bias * value must be of {@link %{OperandTypeLinkPfx}FLOAT16}. %/kind * For input tensor of {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32}, the bias * value must be of {@link %{OperandTypeLinkPfx}FLOAT32}. * * 3: A scalar, specifying the scale factor, alpha. %kind aidl canonical ndk hal_1.2+ * For input tensor of {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16}, the * alpha value must be of {@link %{OperandTypeLinkPfx}FLOAT16}. %/kind * For input tensor of {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32}, the * alpha value must be of {@link %{OperandTypeLinkPfx}FLOAT32}. * * 4: A scalar, specifying the exponent, beta. %kind aidl canonical ndk hal_1.2+ * For input tensor of {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16}, the beta * value must be of {@link %{OperandTypeLinkPfx}FLOAT16}. %/kind * For input tensor of {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32}, the beta * value must be of {@link %{OperandTypeLinkPfx}FLOAT32}. %kind aidl canonical ndk hal_1.2+ * * 5: An optional {@link %{OperandTypeLinkPfx}INT32} scalar, default to -1, * specifying the dimension normalization would be performed on. * Negative index is used to specify axis from the end (e.g. -1 for * the last axis). Must be in the range [-n, n). * Available since %{NNAPILevel3}. %/kind * * Outputs: * * 0: The output tensor of same shape as input0. %insert AVAIL1 */ %{DeclareOperation LOCAL_RESPONSE_NORMALIZATION 13}, /** * Computes sigmoid activation on the input tensor element-wise. * * The output is calculated using this formula: * * output = 1 / (1 + exp(-input)) * * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: up to 4. * * Inputs: * * 0: A tensor, specifying the input. %kind aidl canonical ndk hal_1.2+ * Since %{NNAPILevel3}, this tensor may be zero-sized. %/kind * * Outputs: * * 0: The output tensor of same shape as input0. * For {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, * the scale must be 1.f / 256 and the zeroPoint must be 0. %kind aidl canonical ndk hal_1.3+ * For {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED}, * the scale must be 1.f / 256 and the zeroPoint must be -128. %/kind %insert AVAIL1 */ %{DeclareOperation LOGISTIC 14}, /** * Projects an input to a bit vector via locality senstive hashing. * * Supported input tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * * Supported input tensor rank: from 1 * * Inputs: * * 0: Hash functions. Dim.size == 2, DataType: Float. * Tensor[0].Dim[0]: Number of hash functions. * Tensor[0].Dim[1]: Number of projected output bits generated by each * hash function. * If the projection type is Sparse: * Tensor[0].Dim[1] + ceil(log2(Tensor[0].Dim[0])) <= 32 * * * 1: Input. Dim.size >= 1, no restriction on DataType. * * 2: Weight. Optional. Dim.size == 1, DataType: Float. * If not set, each input element is considered to have the same weight * of 1.0. * Tensor[1].Dim[0] == Tensor[2].Dim[0] * * 3: Type: * Sparse: %kind aidl canonical ndk hal_1.2+ * Value LSHProjectionType_SPARSE(=3) (since %{NNAPILevel3}). %else * Value LSHProjectionType_SPARSE(=1). %/kind * Computed bit vector is considered to be sparse. * Each output element is an int32 made up of multiple bits * computed from hash functions. * %kind aidl canonical ndk hal_1.2+ * NOTE: To avoid collisions across hash functions, an offset value * of k * (1 << Tensor[0].Dim[1]) will be added to each signature, * where k is the index of the hash function. * * Value LSHProjectionType_SPARSE_DEPRECATED(=1). * Legacy behavior that does not include the offset value. * %/kind * Dense: * Value LSHProjectionType_DENSE(=2). * Computed bit vector is considered to be dense. Each output * element represents a bit and can take the value of either * 0 or 1. * * Outputs: * * 0: If the projection type is Sparse: * Output.Dim == { Tensor[0].Dim[0] } * A tensor of int32 that represents hash signatures. * * If the projection type is Dense: * Output.Dim == { Tensor[0].Dim[0] * Tensor[0].Dim[1] } * A flattened tensor that represents projected bit vectors. %insert AVAIL1 %kind aidl canonical ndk hal_1.2+ * The offset value for sparse projections was added in %{NNAPILevel3}. %/kind */ %{DeclareOperation LSH_PROJECTION 15}, /** * Performs a single time step in a Long Short-Term Memory (LSTM) layer * * The LSTM operation is described by the following equations. * * \f{eqnarray*}{ * i_t =& \sigma(W_{xi}x_t+W_{hi}h_{t-1}+W_{ci}C_{t-1}+b_i) & \\ * f_t =& \sigma(W_{xf}x_t+W_{hf}h_{t-1}+W_{cf}C_{t-1}+b_f) & \\ * C_t =& clip(f_t \odot C_{t-1} + i_t \odot * g(W_{xc}x_t+W_{hc}h_{t-1}+b_c),\ t_{cell}) & \\ * o_t =& \sigma(W_{xo}x_t+W_{ho}h_{t-1}+W_{co}C_t+b_o) & \\ * & & \\ * & clip(W_{proj}(o_t \odot g(C_t))+b_{proj},\ t_{proj}) * & if\ there\ is\ a\ projection; \\ * h_t =& & \\ * & o_t \odot g(C_t) & otherwise. \\ * \f} * Where: * * \f$x_t\f$ is the input, * * \f$i_t\f$ is the input gate, * * \f$f_t\f$ is the forget gate, * * \f$C_t\f$ is the cell state, * * \f$o_t\f$ is the output, * * \f$h_t\f$ is the output state, * * \f$\sigma\f$ is the logistic sigmoid function, * * \f$g\f$ is the cell input and cell output activation function, usually * \f$tahn\f$, * * \f$W_{xi}\f$ is the input-to-input weight matrix, * * \f$W_{hi}\f$ is the recurrent to input weight matrix, * * \f$W_{ci}\f$ is the cell-to-input weight matrix, * * \f$b_i\f$ is the input gate bias, * * \f$W_{xf}\f$ is the input-to-forget weight matrix, * * \f$W_{hf}\f$ is the recurrent-to-forget weight matrix, * * \f$W_{cf}\f$ is the cell-to-forget weight matrix, * * \f$b_f\f$ is the forget gate bias, * * \f$W_{xc}\f$ is the input-to-cell weight matrix, * * \f$W_{hc}\f$ is the recurrent-to-cell weight matrix, * * \f$b_c\f$ is the cell bias, * * \f$W_{xo}\f$ is the input-to-output weight matrix, * * \f$W_{ho}\f$ is the recurrent-to-output weight matrix, * * \f$W_{co}\f$ is the cell-to-output weight matrix, * * \f$b_o\f$ is the output gate bias, * * \f$W_{proj}\f$ is the projection weight matrix, * * \f$b_{proj}\f$ is the projection bias, * * \f$t_{cell}\f$ is the threshold for clipping the cell state, and * * \f$t_{proj}\f$ is the threshold for clipping the projected output. * * \f$\odot\f$ is the * * Hadamard product that takes two matrices and produces another * matrix, each element of which is the product of the corresponding * elements of the input matrices. * %kind aidl canonical ndk hal_1.2+ * Since %{NNAPILevel3} LSTM supports layer normalization. * In case layer normalization is used, the inputs to internal activation * functions (sigmoid and \f$g\f$) are normalized, rescaled and recentered * following an approach from section 3.1 from * https://arxiv.org/pdf/1607.06450.pdf * %/kind * The operation has the following independently optional inputs: * * The cell-to-input weights (\f$W_{ci}\f$), cell-to-forget weights * (\f$W_{cf}\f$) and cell-to-output weights (\f$W_{co}\f$) either all * have values or neither of them have values (i.e., all set to null). If * they have values, the peephole optimization is used. * * The input-to-input weights (\f$W_{xi}\f$), recurrent-to-input weights * (\f$W_{hi}\f$) and input gate bias (\f$b_i\f$) either all have values, * or none of them have values. If they have no values, coupling of input * and forget gates (CIFG) is used, in which case the input gate * (\f$i_t\f$) is calculated using the following equation instead. * \f{eqnarray*}{ * i_t = 1 - f_t * \f} * In case peephole optimization is used and CIFG is not used * cell-to-input (\f$W_{ci}\f$) weights must be present. Otherwise, the * cell-to-input weights must have no value. * * The projection weights (\f$W_{proj}\f$) is required only for the * recurrent projection layer, and should otherwise have no value. * * The projection bias (\f$b_{proj}\f$) may (but not required to) have a * value if the recurrent projection layer exists, and should otherwise * have no value. %kind aidl canonical ndk hal_1.2+ * * (%{NNAPILevel3} or later) The four layer normalization weights either all have * values or none of them have values. Additionally, if CIFG is used, * input layer normalization weights tensor is omitted and the other layer * normalization weights either all have values or none of them have * values. Layer normalization is used when the values of all the layer * normalization weights are present. %/kind * * References: * * The default non-peephole non-CIFG implementation is based on: * http://www.bioinf.jku.at/publications/older/2604.pdf * S. Hochreiter and J. Schmidhuber. "Long Short-Term Memory". Neural * Computation, 9(8):1735-1780, 1997. * * The peephole implementation and projection layer is based on: * https://research.google.com/pubs/archive/43905.pdf * Hasim Sak, Andrew Senior, and Francoise Beaufays. "Long short-term memory * recurrent neural network architectures for large scale acoustic * modeling." INTERSPEECH, 2014. * (However, the concept of peephole optimization was introduced in work * prior to this paper.) * * The coupling of input and forget gate (CIFG) is based on: * http://arxiv.org/pdf/1503.04069.pdf * Greff et al. "LSTM: A Search Space Odyssey" * %kind aidl canonical ndk hal_1.2+ * The layer normalization is based on: * https://arxiv.org/pdf/1607.06450.pdf * Jimmy Ba et al. "Layer Normalization" * %/kind * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * All input and output tensors must be of the same type. * * Inputs: * * 0: The input (\f$x_t\f$). * A 2-D tensor of shape [batch_size, input_size], where “batch_size” * corresponds to the batching dimension, and “input_size” is the size * of the input. * * 1: The input-to-input weights (\f$W_{xi}\f$). Optional. * A 2-D tensor of shape [num_units, input_size], where “num_units” * corresponds to the number of cell units. * * 2: The input-to-forget weights (\f$W_{xf}\f$). * A 2-D tensor of shape [num_units, input_size]. * * 3: The input-to-cell weights (\f$W_{xc}\f$). * A 2-D tensor of shape [num_units, input_size]. * * 4: The input-to-output weights (\f$W_{xo}\f$). * A 2-D tensor of shape [num_units, input_size]. * * 5: The recurrent-to-input weights (\f$W_{hi}\f$). Optional. * A 2-D tensor of shape [num_units, output_size], where “output_size” * corresponds to either the number of cell units (i.e., “num_units”), * or the second dimension of the “projection_weights”, if defined. * * 6: The recurrent-to-forget weights (\f$W_{hf}\f$). * A 2-D tensor of shape [num_units, output_size]. * * 7: The recurrent-to-cell weights (\f$W_{hc}\f$). * A 2-D tensor of shape [num_units, output_size]. * * 8: The recurrent-to-output weights (\f$W_{ho}\f$). * A 2-D tensor of shape [num_units, output_size]. * * 9: The cell-to-input weights (\f$W_{ci}\f$). Optional. * A 1-D tensor of shape [num_units]. * * 10:The cell-to-forget weights (\f$W_{cf}\f$). Optional. * A 1-D tensor of shape [num_units]. * * 11:The cell-to-output weights (\f$W_{co}\f$). Optional. * A 1-D tensor of shape [num_units]. * * 12:The input gate bias (\f$b_i\f$). Optional. * A 1-D tensor of shape [num_units]. * * 13:The forget gate bias (\f$b_f\f$). * A 1-D tensor of shape [num_units]. * * 14:The cell bias (\f$b_c\f$). * A 1-D tensor of shape [num_units]. * * 15:The output gate bias (\f$b_o\f$). * A 1-D tensor of shape [num_units]. * * 16:The projection weights (\f$W_{proj}\f$). Optional. * A 2-D tensor of shape [output_size, num_units]. * * 17:The projection bias (\f$b_{proj}\f$). Optional. * A 1-D tensor of shape [output_size]. * * 18:The output state (in) (\f$h_{t-1}\f$). * A 2-D tensor of shape [batch_size, output_size]. * * 19:The cell state (in) (\f$C_{t-1}\f$). * A 2-D tensor of shape [batch_size, num_units]. * * 20:The activation function (\f$g\f$). * A value indicating the activation function: * * * 21:The clipping threshold (\f$t_{cell}\f$) for the cell state, such * that values are bound within [-cell_clip, cell_clip]. If set to 0.0 * then clipping is disabled. %kind aidl canonical ndk hal_1.2+ * Until %{NNAPILevel3} this scalar must be of type {@link * %{OperandTypeLinkPfx}FLOAT32}. Since %{NNAPILevel3}, if all the input * tensors have type {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32}, this * scalar must be of the type {@link %{OperandTypeLinkPfx}FLOAT32}, * otherwise if all the input tensors have the type {@link * %{OperandTypeLinkPfx}TENSOR_FLOAT16}, this scalar must be of type {@link * %{OperandTypeLinkPfx}FLOAT16}. %/kind * * 22:The clipping threshold (\f$t_{proj}\f$) for the output from the * projection layer, such that values are bound within * [-proj_clip, proj_clip]. If set to 0.0 then clipping is disabled. %kind aidl canonical ndk hal_1.2+ * Until %{NNAPILevel3} this scalar must be of type {@link * %{OperandTypeLinkPfx}FLOAT32}. Since %{NNAPILevel3}, if all the input * tensors have type {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32}, this * scalar must be of the type {@link %{OperandTypeLinkPfx}FLOAT32}, * otherwise if all the input tensors have the type {@link * %{OperandTypeLinkPfx}TENSOR_FLOAT16}, this scalar must be of type {@link * %{OperandTypeLinkPfx}FLOAT16}. * Since %{NNAPILevel3} there are additional inputs to this op: * * 23:The input layer normalization weights. * A 1-D tensor of shape [num_units]. Used to rescale normalized inputs * to activation at input gate. * * 24:The forget layer normalization weights. * A 1-D tensor of shape [num_units]. Used to rescale normalized inputs * to activation at forget gate. * * 25:The cell layer normalization weights. * A 1-D tensor of shape [num_units]. Used to rescale normalized inputs * to activation at cell gate. * * 26:The output layer normalization weights. * A 1-D tensor of shape [num_units]. Used to rescale normalized inputs * to activation at output gate. %/kind * * Outputs: * * 0: The scratch buffer. * A 2-D tensor of shape [batch_size, num_units * 3] with CIFG, or * [batch_size, num_units * 4] without CIFG. * * 1: The output state (out) (\f$h_t\f$). * A 2-D tensor of shape [batch_size, output_size]. * * 2: The cell state (out) (\f$C_t\f$). * A 2-D tensor of shape [batch_size, num_units]. * * 3: The output (\f$o_t\f$). * A 2-D tensor of shape [batch_size, output_size]. This is effectively * the same as the current “output state (out)” value. %insert AVAIL1 */ %{DeclareOperation LSTM 16}, /** * Performs an 2-D max pooling operation. * * The output dimensions are functions of the filter dimensions, stride, and * padding. * * The values in the output tensor are computed as: * * output[b, i, j, channel] = * max_{di, dj} ( * input[b, strides[1] * i + di, strides[2] * j + dj, channel] * ) * * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * %insert NHWC_NCHW * * Both explicit padding and implicit padding are supported. * * Inputs (explicit padding): * * 0: A 4-D tensor, of shape [batches, height, width, depth], specifying * the input. %insert ZeroBatchesNNAPILevel3 * * 1: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the left, in the ‘width’ dimension. * * 2: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the right, in the ‘width’ dimension. * * 3: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the top, in the ‘height’ dimension. * * 4: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the bottom, in the ‘height’ dimension. * * 5: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘width’ dimension. * * 6: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘height’ dimension. * * 7: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the filter * width. * * 8: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the filter * height. * * 9: An {@link %{OperandTypeLinkPfx}INT32} scalar, and has to be one of the * {@link %{FusedActivationFunc}} values. Specifies the activation to * invoke on the result. %kind aidl canonical ndk hal_1.2+ * * 10: An optional {@link %{OperandTypeLinkPfx}BOOL} scalar, default to false. * Set to true to specify NCHW data layout for input0 and output0. * Available since %{NNAPILevel3}. %/kind * * Inputs (implicit padding): * * 0: A 4-D tensor, of shape [batches, height, width, depth], specifying * the input. %insert ZeroBatchesNNAPILevel3 * * 1: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the implicit * padding scheme, has to be one of the %insert PaddingCodeValues * * 2: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘width’ dimension. * * 3: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘height’ dimension. * * 4: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the filter * width. * * 5: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the filter * height. * * 6: An {@link %{OperandTypeLinkPfx}INT32} scalar, and has to be one of the * {@link %{FusedActivationFunc}} values. Specifies the activation to * invoke on the result. %kind aidl canonical ndk hal_1.2+ * * 7: An optional {@link %{OperandTypeLinkPfx}BOOL} scalar, default to false. * Set to true to specify NCHW data layout for input0 and output0. * Available since %{NNAPILevel3}. %/kind * * Outputs: * * 0: The output 4-D tensor, of shape * [batches, out_height, out_width, depth]. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint must be the same as input0. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint must be the same as input0. %/kind %insert AVAIL1 */ %{DeclareOperation MAX_POOL_2D 17}, /** * Multiplies two tensors, element-wise. * * Takes two input tensors of identical {@link %{OperandType}} and compatible * dimensions. The output is the product of both input tensors, optionally * modified by an activation function. * * Two dimensions are compatible when: * 1. they are equal, or * 2. one of them is 1 * * The size of the resulting output is the maximum size along each dimension * of the input operands. It starts with the trailing dimensions, and works * its way forward. * %insert GenericZero * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: up to 4 * * Inputs: * * 0: A tensor. * * 1: A tensor of the same {@link %{OperandType}}, and compatible dimensions * as input0. * * 2: An {@link %{OperandTypeLinkPfx}INT32} scalar, and has to be one of the * {@link %{FusedActivationFunc}} values. Specifies the activation to * invoke on the result. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_INT32} tensor, * the {@link %{FusedActivationFunc}} must be "NONE". %/kind * * Outputs: * * 0: The product, a tensor of the same {@link %{OperandType}} as input0. %kind aidl canonical ndk hal_1.3+ * For output tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * and {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED}, * the following condition must be satisfied: * output_scale > input1_scale * input2_scale. %else * For output tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, * the following condition must be satisfied: * output_scale > input1_scale * input2_scale. %/kind %insert AVAIL1 */ %{DeclareOperation MUL 18}, /** * Computes rectified linear activation on the input tensor element-wise. * * The output is calculated using this formula: * * output = max(0, input) * * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: up to 4. * * Inputs: * * 0: A tensor, specifying the input. %kind aidl canonical ndk hal_1.2+ * Since %{NNAPILevel3}, this tensor may be zero-sized. %/kind * * Outputs: * * 0: The output tensor of same shape as input0. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint must be the same as input0. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint must be the same as input0. %/kind %insert AVAIL1 */ %{DeclareOperation RELU 19}, /** * Computes rectified linear 1 activation on the input tensor element-wise. * * The output is calculated using this formula: * * output = min(1.f, max(-1.f, input)) * * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: up to 4. * * Inputs: * * 0: A tensor, specifying the input. %kind aidl canonical ndk hal_1.2+ * Since %{NNAPILevel3}, this tensor may be zero-sized. %/kind * * Outputs: * * 0: The output tensor of the same shape as input0. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint must be the same as input0. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint must be the same as input0. %/kind %insert AVAIL1 */ %{DeclareOperation RELU1 20}, /** * Computes rectified linear 6 activation on the input tensor element-wise. * * The output is calculated using this formula: * * output = min(6, max(0, input)) * * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: up to 4. * * Inputs: * * 0: A tensor, specifying the input. %kind aidl canonical ndk hal_1.2+ * Since %{NNAPILevel3}, this tensor may be zero-sized. %/kind * * Outputs: * * 0: The output tensor of same shape as input0. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint must be the same as input0. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint must be the same as input0. %/kind %insert AVAIL1 */ %{DeclareOperation RELU6 21}, /** * Reshapes a tensor. * * Given tensor, this operation returns a tensor that has the same values as * tensor, but with a newly specified shape. * * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind %kind aidl canonical ndk * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} (since %{NNAPILevel6}) %/kind * * Supported tensor rank: up to 4. * * Inputs: * * 0: A tensor, specifying the tensor to be reshaped. * * 1: A 1-D tensor of {@link %{OperandTypeLinkPfx}TENSOR_INT32}, defining the * shape of the output tensor. The number of elements implied by shape * must be the same as the number of elements in the input tensor. * * If one component of shape is the special value -1, the size of that * dimension is computed so that the total size remains constant. In * particular, a shape of [-1] flattens into 1-D. At most one component * of shape can be -1. * * Outputs: * * 0: The output tensor, of shape specified by the input shape. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint must be the same as input0. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint must be the same as input0. %/kind %insert AVAIL1 */ %{DeclareOperation RESHAPE 22}, /** * Resizes images to given size using the bilinear interpretation. * * Resized images must be distorted if their output aspect ratio is not the * same as input aspect ratio. The corner pixels of output may not be the * same as corner pixels of input. * * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} (since %{NNAPILevel3}) %/kind %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * %insert NHWC_NCHW * %kind aidl canonical ndk hal_1.2+ * Both resizing by shape and resizing by scale are supported. * %/kind * Inputs (resizing by shape): * * 0: A 4-D tensor, of shape [batches, height, width, depth], specifying * the input. %insert ZeroBatchesNNAPILevel3 * * 1: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the output * width of the output tensor. * * 2: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the output * height of the output tensor. %kind aidl canonical ndk hal_1.2+ * * 3: An optional {@link %{OperandTypeLinkPfx}BOOL} scalar, default to false. * Set to true to specify NCHW data layout for input0 and output0. * Available since %{NNAPILevel3}. %/kind %kind aidl canonical ndk hal_1.3+ * * 4: Align corners. An optional {@link %{OperandTypeLinkPfx}BOOL} * scalar, default to false. If True, the centers of the 4 corner * pixels of the input and output tensors are aligned, preserving the * values at the corner pixels. * Available since %{NNAPILevel4}. * * 5: Half pixel centers. An optional {@link %{OperandTypeLinkPfx}BOOL} * scalar, default to false. If True, the pixel centers are assumed to * be at (0.5, 0.5). This is the default behavior of image.resize in * TF 2.0. If this parameter is True, then align_corners parameter * must be False. * Available since %{NNAPILevel4}. %/kind %kind aidl canonical ndk hal_1.2+ * * Inputs (resizing by scale, since %{NNAPILevel3}): * * 0: A 4-D tensor, of shape [batches, height, width, depth], specifying * the input. Zero batches is supported for this tensor. * * 1: A scalar, specifying width_scale, the scaling factor of the width * dimension from the input tensor to the output tensor. The output * width is calculated as new_width = floor(width * width_scale). * The scalar must be of {@link %{OperandTypeLinkPfx}FLOAT16} if input0 is * of {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} and of * {@link %{OperandTypeLinkPfx}FLOAT32} otherwise. * * 2: A scalar, specifying height_scale, the scaling factor of the height * dimension from the input tensor to the output tensor. The output * height is calculated as new_height = floor(height * height_scale). * The scalar must be of {@link %{OperandTypeLinkPfx}FLOAT16} if input0 is * of {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} and of * {@link %{OperandTypeLinkPfx}FLOAT32} otherwise. * * 3: An optional {@link %{OperandTypeLinkPfx}BOOL} scalar, default to false. * Set to true to specify NCHW data layout for input0 and output0. %/kind %kind aidl canonical ndk hal_1.3+ * * 4: Align corners. An optional {@link %{OperandTypeLinkPfx}BOOL} * scalar, default to false. If True, the centers of the 4 corner * pixels of the input and output tensors are aligned, preserving the * values at the corner pixels. * Available since %{NNAPILevel4}. * * 5: Half pixel centers. An optional {@link %{OperandTypeLinkPfx}BOOL} * scalar, default to false. If True, the pixel centers are assumed to * be at (0.5, 0.5). This is the default behavior of image.resize in * TF 2.0. If this parameter is True, then align_corners parameter * must be False. * Available since %{NNAPILevel4}. %/kind * * Outputs: * * 0: The output 4-D tensor, of shape * [batches, new_height, new_width, depth]. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint must be the same as input0. %/kind %kind canonical ndk hal_1.2 * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint must be the same as input0. %/kind %insert AVAIL1 */ %{DeclareOperation RESIZE_BILINEAR 23}, /** * A basic recurrent neural network layer. * * This layer implements the operation: * outputs = state = activation(inputs * input_weights + * state * recurrent_weights + bias) * * Where: * * “input_weights” is a weight matrix that multiplies the inputs; * * “recurrent_weights” is a weight matrix that multiplies the current * “state” which itself is the output from the previous time step * computation; * * “bias” is a bias vector (added to each output vector in the batch); * * “activation” is the function passed as the “fused_activation_function” * argument (if not “NONE”). * * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * The input tensors must all be the same type. * * Inputs: * * 0: input. * A 2-D tensor of shape [batch_size, input_size], where “batch_size” * corresponds to the batching dimension, and “input_size” is the size * of the input. * * 1: weights. * A 2-D tensor of shape [num_units, input_size], where “num_units” * corresponds to the number of units. * * 2: recurrent_weights. * A 2-D tensor of shape [num_units, num_units], with columns * corresponding to the weights from each unit. * * 3: bias. * A 1-D tensor of shape [num_units]. * * 4: hidden state (in). * A 2-D tensor of shape [batch_size, num_units]. * * 5: fused_activation_function. * An optional {@link %{FusedActivationFunc}} value indicating the * activation function. If “NONE” is specified then it results in a * linear activation. * * Outputs: * * 0: hidden state (out). * A 2-D tensor of shape [batch_size, num_units]. * * * 1: output. * A 2-D tensor of shape [batch_size, num_units]. This is effectively * the same as the current state value. %insert AVAIL1 */ %{DeclareOperation RNN 24}, /** * Computes the softmax activation on the input tensor element-wise, per * batch, by normalizing the input vector so the maximum coefficient is * zero. * * The output is calculated using this formula: * * output[batch, i] = * exp((input[batch, i] - max(input[batch, :])) * beta) / * sum_{k}{exp((input[batch, k] - max(input[batch, :])) * beta)} * * For input tensor with rank other than 2, the activation will be applied * independently on each 1-D slice along specified dimension. * * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * %kind hal_1.0 hal_1.1 * Supported tensor rank: 2 or 4. %/kind %kind aidl canonical ndk hal_1.2+ * Supported tensor rank: up to 4. * Tensors with rank other than 2 or 4 are only supported since %{NNAPILevel3}. %/kind * * Inputs: * * 0: A 2-D or 4-D tensor, specifying the tensor to be reshaped. %kind aidl canonical ndk hal_1.2+ * Since %{NNAPILevel3}, this tensor may be zero-sized. %/kind %kind aidl canonical ndk hal_1.3+ * * 1: A scalar, specifying the positive scaling factor for the exponent, * beta. If input0 is of {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32}, * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} or * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED}, the scalar * must be of {@link %{OperandTypeLinkPfx}FLOAT32}. %else * * 1: A scalar, specifying the positive scaling factor for the exponent, * beta. If input0 is of {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} or * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, the scalar must be of * {@link %{OperandTypeLinkPfx}FLOAT32}. %/kind %kind aidl canonical ndk hal_1.2+ * If input0 is of {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16}, then the * scalar must be of {@link %{OperandTypeLinkPfx}FLOAT16}. %/kind %kind aidl canonical ndk hal_1.2+ * * 2: An optional {@link %{OperandTypeLinkPfx}INT32} scalar, default to -1, * specifying the dimension the activation would be performed on. * Negative index is used to specify axis from the end (e.g. -1 for * the last axis). Must be in the range [-n, n). * Available since %{NNAPILevel3}. %/kind * * Outputs: * * 0: The output tensor of same shape as input0. * For {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, * the scale must be 1.f / 256 and the zeroPoint must be 0. %kind aidl canonical ndk hal_1.3+ * For {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED}, * the scale must be 1.f / 256 and the zeroPoint must be -128. %/kind %insert AVAIL1 */ %{DeclareOperation SOFTMAX 25}, /** * Rearranges blocks of spatial data, into depth. * * More specifically, this op outputs a copy of the input tensor where * values from the height and width dimensions are moved to the depth * dimension. The value block_size indicates the input block size and how * the data is moved. * * Chunks of data of size block_size * block_size from depth are rearranged * into non-overlapping blocks of size block_size x block_size. * * The depth of the output tensor is input_depth * block_size * block_size. * The input tensor's height and width must be divisible by block_size. * * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * %insert NHWC_NCHW * * Inputs: * * 0: A 4-D tensor, of shape [batches, height, width, depth_in], * specifying the input. * * 1: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the block_size. * block_size must be >=1 and block_size must be a divisor of both the * input height and width. %kind aidl canonical ndk hal_1.2+ * * 2: An optional {@link %{OperandTypeLinkPfx}BOOL} scalar, default to false. * Set to true to specify NCHW data layout for input0 and output0. * Available since %{NNAPILevel3}. %/kind * * Outputs: * * 0: The output 4-D tensor, of shape [batches, height/block_size, * width/block_size, depth_in*block_size*block_size]. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint must be the same as input0. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint must be the same as input0. %/kind %insert AVAIL1 */ %{DeclareOperation SPACE_TO_DEPTH 26}, /** * SVDF op is a kind of stateful layer derived from the notion that a * densely connected layer that's processing a sequence of input frames can * be approximated by using a singular value decomposition of each of its * nodes. The implementation is based on: * * https://research.google.com/pubs/archive/43813.pdf * * P. Nakkiran, R. Alvarez, R. Prabhavalkar, C. Parada. * “Compressing Deep Neural Networks using a Rank-Constrained Topology”. * INTERSPEECH, 2015. * * It processes the incoming input using a 2-stage filtering mechanism: * * stage 1 performs filtering on the "features" dimension, whose outputs * get pushed into a memory of fixed-size memory_size. * * stage 2 performs filtering on the "time" dimension of the memory_size * memoized outputs of stage 1. * * Specifically, for rank 1, this layer implements the operation: * * memory = push(conv1d(inputs, weights_feature, feature_dim, * "%{ANN}PADDING_VALID")); * outputs = activation(memory * weights_time + bias); * * Where: * * “weights_feature” is a weights matrix that processes the inputs (by * convolving the input with every “feature filter”), and whose outputs * get pushed, stacked in order, into the fixed-size “memory” (the oldest * entry gets dropped); * * “weights_time” is a weights matrix that processes the “memory” (by a * batched matrix multiplication on the num_units); * * “bias” is an optional bias vector (added to each output vector in the * batch); and * * “activation” is the function passed as the “fused_activation_function” * argument (if not “NONE”). * * Each rank adds a dimension to the weights matrices by means of stacking * the filters. * * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * All input tensors must be the same type. * * Inputs: * * 0: input. * A 2-D tensor of shape [batch_size, input_size], where “batch_size” * corresponds to the batching dimension, and “input_size” is the size * of the input. * * 1: weights_feature. * A 2-D tensor of shape [num_units, input_size], where “num_units” * corresponds to the number of units. * * 2: weights_time. * A 2-D tensor of shape [num_units, memory_size], where “memory_size” * corresponds to the fixed-size of the memory. * * 3: bias. * An optional 1-D tensor of shape [num_units]. * * 4: state (in). * A 2-D tensor of shape [batch_size, (memory_size - 1) * num_units * rank]. * * 5: rank. * The rank of the SVD approximation. * * 6: fused_activation_function. * An optional {@link %{FusedActivationFunc}} value indicating the * activation function. If “NONE” is specified then it results in a * linear activation. * * Outputs: * * 0: state (out). * A 2-D tensor of the same {@link %{OperandType}} as the inputs, with shape * [batch_size, (memory_size - 1) * num_units * rank]. * * 1: output. * A 2-D tensor of the same {@link %{OperandType}} as the inputs, with shape * [batch_size, num_units]. %insert AVAIL1 */ %{DeclareOperation SVDF 27}, /** * Computes hyperbolic tangent of input tensor element-wise. * * The output is calculated using this formula: * * output = tanh(input) * * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} (since %{NNAPILevel3}) %/kind %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: up to 4. * * Inputs: * * 0: A tensor, specifying the input. %kind aidl canonical ndk hal_1.2+ * Since %{NNAPILevel3}, this tensor may be zero-sized. %/kind * * Outputs: * * 0: The output tensor of same shape as input0. %kind aidl canonical ndk hal_1.2+ * For {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, * the scale must be 1.f / 128 and the zeroPoint must be 128. %/kind %kind aidl canonical ndk hal_1.3+ * For {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED}, * the scale must be 1.f / 128 and the zeroPoint must be 0. %/kind %insert AVAIL1 */ %{DeclareOperation TANH 28}, %/section %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% HAL OperationType for 1.1 %% NDK OperationCode for API 28 %section Operation_1.1 /** * BatchToSpace for N-dimensional tensors. * * This operation reshapes the batch dimension (dimension 0) into M + 1 * dimensions of shape block_shape + [batch], interleaves these blocks back * into the grid defined by the spatial dimensions [1, ..., M], to obtain a * result with the same rank as the input. * * This is the reverse of SpaceToBatch. * * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * %insert NHWC_NCHW * * Inputs: * * 0: An n-D tensor, specifying the tensor to be reshaped * * 1: A 1-D Tensor of {@link %{OperandTypeLinkPfx}TENSOR_INT32}, the block * sizes for each spatial dimension of the input tensor. All values * must be >= 1. %kind aidl canonical ndk hal_1.2+ * * 2: An optional {@link %{OperandTypeLinkPfx}BOOL} scalar, default to false. * Set to true to specify NCHW data layout for input0 and output0. * Available since API level 29. %/kind * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} as input0. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint must be the same as input0. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint must be the same as input0. %/kind %insert AVAIL2 */ %{DeclareOperation BATCH_TO_SPACE_ND 29}, /** * Element-wise division of two tensors. * * Takes two input tensors of identical {@link %{OperandType}} and compatible * dimensions. The output is the result of dividing the first input tensor * by the second, optionally modified by an activation function. * %kind aidl canonical ndk hal_1.3+ * For inputs of {@link %{OperandTypeLinkPfx}TENSOR_INT32}, performs * "floor division" ("//" in Python). For example, * 5 // 2 = 2 * -5 // 2 = -3 * %/kind * Two dimensions are compatible when: * 1. they are equal, or * 2. one of them is 1 * * The size of the output is the maximum size along each dimension of the * input operands. It starts with the trailing dimensions, and works its way * forward. * * Example: * input1.dimension = {4, 1, 2} * input2.dimension = {5, 4, 3, 1} * output.dimension = {5, 4, 3, 2} * %insert GenericZero * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: up to 4 * * Inputs: * * 0: An n-D tensor, specifying the first input. * * 1: A tensor of the same {@link %{OperandType}}, and compatible dimensions * as input0. * * 2: An {@link %{OperandTypeLinkPfx}INT32} scalar, and has to be one of the * {@link %{FusedActivationFunc}} values. Specifies the activation to * invoke on the result. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_INT32} tensor, * the {@link %{FusedActivationFunc}} must be "NONE". %/kind * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} as input0. %insert AVAIL2 */ %{DeclareOperation DIV 30}, /** * Computes the mean of elements across dimensions of a tensor. * * Reduces the input tensor along the given dimensions to reduce. Unless * keep_dims is true, the rank of the tensor is reduced by 1 for each entry * in axis. If keep_dims is true, the reduced dimensions are retained with * length 1. * * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: up to 4 * * Inputs: * * 0: A tensor, specifying the input. * * 1: A 1-D Tensor of {@link %{OperandTypeLinkPfx}TENSOR_INT32}. The dimensions * to reduce. Must be in the range * [-rank(input_tensor), rank(input_tensor)). * * NOTE: When the operation was introduced, the documentation * incorrectly stated that if dimensions were empty, the operation * would reduce across all dimensions. This behavior was never * implemented. * * * 2: An {@link %{OperandTypeLinkPfx}INT32} scalar, keep_dims. If positive, * retains reduced dimensions with length 1. * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} as input0. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint must be the same as input0. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint must be the same as input0. %/kind * If all dimensions are reduced and keep_dims is false, the output * shape is [1]. %insert AVAIL2 */ %{DeclareOperation MEAN 31}, /** * Pads a tensor. * * This operation pads a tensor according to the specified paddings. * * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind %kind aidl canonical ndk hal_1.2+ * (full support since %{NNAPILevel3}, see the output section) %else * (the pad value is undefined) %/kind * * Supported tensor rank: up to 4 * * Inputs: * * 0: An n-D tensor, specifying the tensor to be padded. * * 1: A 2-D Tensor of {@link %{OperandTypeLinkPfx}TENSOR_INT32}, the paddings * for each spatial dimension of the input tensor. The shape of the * tensor must be {rank(input0), 2}. * padding[i, 0] specifies the number of elements to be padded in the * front of dimension i. * padding[i, 1] specifies the number of elements to be padded after the * end of dimension i. * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} as input0. The * output tensor has the same rank as input0, and each * dimension of the output tensor has the same size as the * corresponding dimension of the input tensor plus the size * of the padding: * output0.dimension[i] = * padding[i, 0] + input0.dimension[i] + padding[i, 1] %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint must be the same as input0. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint must be the same as input0. %/kind %kind aidl canonical ndk hal_1.2+ * * NOTE: Before %{NNAPILevel3}, the pad value for * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} is undefined. * Since %{NNAPILevel3}, the pad value is always the logical zero. %/kind %insert AVAIL2 */ %{DeclareOperation PAD 32}, /** * SpaceToBatch for N-Dimensional tensors. * * This operation divides "spatial" dimensions [1, ..., M] of the input into * a grid of blocks of shape block_shape, and interleaves these blocks with * the "batch" dimension (0) such that in the output, the spatial dimensions * [1, ..., M] correspond to the position within the grid, and the batch * dimension combines both the position within a spatial block and the * original batch position. Prior to division into blocks, the spatial * dimensions of the input are optionally zero padded according to paddings. * * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind %kind aidl canonical ndk hal_1.2+ * (full support since %{NNAPILevel3}, see the output section) %else * (the pad value is undefined) %/kind * %insert NHWC_NCHW * * Inputs: * * 0: An n-D tensor, specifying the input. * * 1: A 1-D Tensor of {@link %{OperandTypeLinkPfx}TENSOR_INT32}, the block * sizes for each spatial dimension of the input tensor. All values * must be >= 1. * * 2: A 2-D Tensor of {@link %{OperandTypeLinkPfx}TENSOR_INT32}, the paddings * for each spatial dimension of the input tensor. All values must be * >= 0. The shape of the tensor must be {M, 2}, where M is the number * of spatial dimensions. * padding[i, 0] specifies the number of element to be padded in the * front of dimension i. * padding[i, 1] specifies the number of element to be padded after the * end of dimension i. %kind aidl canonical ndk hal_1.2+ * * 3: An optional {@link %{OperandTypeLinkPfx}BOOL} scalar, default to false. * Set to true to specify NCHW data layout for input0 and output0. * Available since %{NNAPILevel3}. %/kind * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} as input0. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint must be the same as input0. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint must be the same as input0. %/kind %kind aidl canonical ndk hal_1.2+ * * NOTE: Before %{NNAPILevel3}, the pad value for * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} is undefined. * Since %{NNAPILevel3}, the pad value is always the logical zero. %/kind %insert AVAIL2 */ %{DeclareOperation SPACE_TO_BATCH_ND 33}, /** * Removes dimensions of size 1 from the shape of a tensor. * * Given a tensor input, this operation returns a tensor of the same * {@link %{OperandType}} with all dimensions of size 1 removed. If you don't * want to remove all size 1 dimensions, you can remove specific size 1 * dimensions by specifying the axes (input1). * * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: up to 4 * * Inputs: * * 0: An n-D tensor, the tensor to be squeezed. * * 1: An optional 1-D tensor of {@link %{OperandTypeLinkPfx}TENSOR_INT32}. The * dimensions to squeeze. If specified only squeezes the dimensions * listed. Otherwise, squeezes all dimensions. The dimension index * starts at 0. An error must be reported if squeezing a dimension that * is not 1. * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} as input0. Contains the * same data as input, but has one or more dimensions of size 1 * removed. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint must be the same as input0. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint must be the same as input0. %/kind * If all input dimensions are equal to 1 and are to be squeezed, the * output shape is [1]. %insert AVAIL2 */ %{DeclareOperation SQUEEZE 34}, /** * Extracts a strided slice of a tensor. * * Roughly speaking, this op extracts a slice of size (end - begin) / stride * from the given input tensor. Starting at the location specified by begin * the slice continues by adding stride to the index until all dimensions * are not less than end. Note that a stride can be negative, which causes a * reverse slice. * * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: up to 4 * * Inputs: * * 0: An n-D tensor, specifying the tensor to be sliced. * * 1: begin, a 1-D tensor of {@link %{OperandTypeLinkPfx}TENSOR_INT32}. The * starts of the dimensions of the input tensor to be sliced. The * length must be of rank(input0). * * 2: end, a 1-D tensor of {@link %{OperandTypeLinkPfx}TENSOR_INT32}. The * ends of the dimensions of the input tensor to be sliced. The length * must be of rank(input0). * * 3: strides, a 1-D tensor of {@link %{OperandTypeLinkPfx}TENSOR_INT32}. The * strides of the dimensions of the input tensor to be sliced. The * length must be of rank(input0). The entries must be non-zero. * * 4: begin_mask, an {@link %{OperandTypeLinkPfx}INT32} scalar. If the ith bit * of begin_mask is set, begin[i] is ignored and the fullest possible * range in that dimension is used instead. * * 5: end_mask, an {@link %{OperandTypeLinkPfx}INT32} scalar. If the ith bit of * end_mask is set, end[i] is ignored and the fullest possible range in * that dimension is used instead. * * 6: shrink_axis_mask, an {@link %{OperandTypeLinkPfx}INT32} scalar. If the * ith bit of shrink_axis_mask is set, the ith dimension specification * shrinks the dimensionality by 1, taking on the value at index * begin[i]. In this case, the ith specification must define a * slice of size 1, e.g. begin[i] = x, end[i] = x + 1. * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} as input0 and rank (n - k), * where k is the number of bits set in shrink_axis_mask. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint must be the same as input0. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint must be the same as input0. %/kind * If shrink_axis_mask is true for all input dimensions, the output * shape is [1]. %insert AVAIL2 */ %{DeclareOperation STRIDED_SLICE 35}, /** * Element-wise subtraction of two tensors. * * Takes two input tensors of identical {@link %{OperandType}} and compatible * dimensions. The output is the result of subtracting the second input * tensor from the first one, optionally modified by an activation function. * * Two dimensions are compatible when: * 1. they are equal, or * 2. one of them is 1 * * The size of the output is the maximum size along each dimension of the * input operands. It starts with the trailing dimensions, and works its way * forward. * * Example: * input1.dimension = {4, 1, 2} * input2.dimension = {5, 4, 3, 1} * output.dimension = {5, 4, 3, 2} * %insert GenericZero * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} (since %{NNAPILevel3}) %/kind %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: up to 4 * * Inputs: * * 0: An n-D tensor, specifying the first input. * * 1: A tensor of the same {@link %{OperandType}}, and compatible dimensions * as input0. * * 2: An {@link %{OperandTypeLinkPfx}INT32} scalar, and has to be one of the * {@link %{FusedActivationFunc}} values. Specifies the activation to * invoke on the result. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_INT32} tensor, * the {@link %{FusedActivationFunc}} must be "NONE". %/kind * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} as input0. %kind hal_1.2 * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint can be different from inputs' scale and zeroPoint. %/kind %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint can be different from inputs' scale and zeroPoint. %/kind %insert AVAIL2 */ %{DeclareOperation SUB 36}, /** * Transposes the input tensor, permuting the dimensions according to the * perm tensor. * * The returned tensor's dimension i corresponds to the input dimension * perm[i]. If perm is not given, it is set to (n-1...0), where n is the * rank of the input tensor. Hence by default, this operation performs a * regular matrix transpose on 2-D input Tensors. * * Supported tensor {@link %{OperandType}}: %kind aidl canonical ndk hal_1.2+ * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} (since %{NNAPILevel3}) %/kind * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: up to 4 * * Inputs: * * 0: An n-D tensor, specifying the tensor to be transposed. %kind aidl canonical ndk hal_1.2+ * Since %{NNAPILevel3}, this tensor may be zero-sized. %/kind * * 1: An optional 1-D Tensor of {@link %{OperandTypeLinkPfx}TENSOR_INT32}, * the permutation of the dimensions of the input tensor. * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} as input0. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint must be the same as input0. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint must be the same as input0. %/kind %insert AVAIL2 */ %{DeclareOperation TRANSPOSE 37}, %/section %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% HAL OperandType for 1.2 %% NDK OperandCode for API 29 %section Operand_1.2 /** * An 8 bit boolean scalar value. * * Values of this operand type are either true or false. A zero value * represents false; any other value represents true. %insert AVAIL3 */ %{ANN}BOOL = 6, %insert canonical_empty_line /** * A tensor of 16 bit signed integers that represent real numbers. * * Attached to this tensor is a number representing real value scale that is * used to convert the 16 bit number to a real value in the following way: * realValue = integerValue * scale. * * scale is a 32 bit floating point with value greater than zero. %insert AVAIL3 */ %{ANN}TENSOR_QUANT16_SYMM = 7, %insert canonical_empty_line /** * A tensor of IEEE 754 16 bit floating point values. %insert AVAIL3 */ %{ANN}TENSOR_FLOAT16 = 8, %insert canonical_empty_line /** * A tensor of 8 bit boolean values. * * Values of this operand type are either true or false. A zero value * represents false; any other value represents true. %insert AVAIL3 */ %{ANN}TENSOR_BOOL8 = 9, %insert canonical_empty_line /** * An IEEE 754 16 bit floating point scalar value. %insert AVAIL3 */ %{ANN}FLOAT16 = 10, %insert canonical_empty_line /** * A tensor of 8 bit signed integers that represent real numbers. * * This tensor is associated with additional fields that can * be used to convert the 8 bit signed integer to the real value and vice versa. * These fields are: * - channelDim: a 32 bit unsigned integer indicating channel dimension. * - scales: an array of positive 32 bit floating point values. * The size of the scales array must be equal to dimensions[channelDim]. * %kind ndk * {@link ANeuralNetworksModel_setOperandSymmPerChannelQuantParams} must be used * to set the parameters for an Operand of this type. * %/kind %kind aidl canonical hal_1.2+ * {@link %{Ann}SymmPerChannelQuantParams} must hold the parameters for an Operand of this type. %/kind * The channel dimension of this tensor must not be unknown (dimensions[channelDim] != 0). * * The formula is: * realValue[..., C, ...] = * integerValue[..., C, ...] * scales[C] * where C is an index in the Channel dimension. %insert AVAIL3 */ %{ANN}TENSOR_QUANT8_SYMM_PER_CHANNEL = 11, %insert canonical_empty_line /** * A tensor of 16 bit unsigned integers that represent real numbers. * * Attached to this tensor are two numbers that can be used to convert the * 16 bit integer to the real value and vice versa. These two numbers are: * - scale: a 32 bit floating point value greater than zero. * - zeroPoint: a 32 bit integer, in range [0, 65535]. * * The formula is: * real_value = (integer_value - zeroPoint) * scale. %insert AVAIL3 */ %{ANN}TENSOR_QUANT16_ASYMM = 12, %insert canonical_empty_line /** * A tensor of 8 bit signed integers that represent real numbers. * * Attached to this tensor is a number representing real value scale that is * used to convert the 8 bit number to a real value in the following way: * realValue = integerValue * scale. * * scale is a 32 bit floating point with value greater than zero. %insert AVAIL3 */ %{ANN}TENSOR_QUANT8_SYMM = 13, %/section %section Operand_1.2_MAX FUNDAMENTAL_MAX = 13, %/section %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% HAL OperationType for 1.2 %% NDK OperationCode for API 29 %section Operation_1.2 /** * Computes the absolute value of a tensor, element-wise. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: from 1. * * Inputs: * * 0: A tensor. * * Outputs: * * 0: The output tensor of same shape as input0. %insert AVAIL3 */ %{DeclareOperation_1.2 ABS 38}, /** * Returns the index of the largest element along an axis. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: from 1 * * Inputs: * * 0: An n-D tensor specifying the input. Must be non-empty. * * 1: An {@link %{OperandTypeLinkPfx}INT32} scalar specifying the axis to * reduce across. Negative index is used to specify axis from the * end (e.g. -1 for the last axis). Must be in the range [-n, n). * * Outputs: * * 0: An (n - 1)-D {@link %{OperandTypeLinkPfx}TENSOR_INT32} tensor. * If input is 1-dimensional, the output shape is [1]. %insert AVAIL3 */ // There is no underscore in ARG_MAX to avoid name conflict with // the macro defined in libc/kernel/uapi/linux/limits.h. %{DeclareOperation_1.2 ARGMAX 39}, /** * Returns the index of the smallest element along an axis. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: from 1 * * Inputs: * * 0: An n-D tensor specifying the input. Must be non-empty. * * 1: An {@link %{OperandTypeLinkPfx}INT32} scalar specifying the axis to * reduce across. Negative index is used to specify axis from the * end (e.g. -1 for the last axis). Must be in the range [-n, n). * * Outputs: * * 0: An (n - 1)-D {@link %{OperandTypeLinkPfx}TENSOR_INT32} tensor. * If input is 1-dimensional, the output shape is [1]. %insert AVAIL3 */ %kind aidl %{DeclareOperation_1.2 ARGMIN 40}, // See ARGMAX for naming discussion. %else %{DeclareOperation_1.2 ARGMIN 40}, // See ARGMAX for naming discussion. %/kind /** * Transform axis-aligned bounding box proposals using bounding box deltas. * * Given the positions of bounding box proposals and the corresponding * bounding box deltas for each class, return the refined bounding box * regions. The resulting bounding boxes are cliped against the edges of * the image. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_ASYMM} * * Inputs: * * 0: A 2-D Tensor of shape [num_rois, 4], specifying the locations of the * bounding box proposals, each line with format [x1, y1, x2, y2]. * For tensor of type {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_ASYMM}, * the zeroPoint must be 0 and the scale must be 0.125. Zero num_rois * is supported for this tensor. * * 1: A 2-D Tensor of shape [num_rois, num_classes * 4], specifying the * bounding box delta for each region of interest and each class. The * bounding box deltas are organized in the following order * [dx, dy, dw, dh], where dx and dy is the relative correction factor * for the center position of the bounding box with respect to the width * and height, dw and dh is the log-scale relative correction factor * for the width and height. For input0 of type * {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_ASYMM}, this tensor should be %kind aidl canonical ndk hal_1.3+ * of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} or * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED}. Zero num_rois is %else * of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}. Zero num_rois is %/kind * supported for this tensor. * * 2: An 1-D {@link %{OperandTypeLinkPfx}TENSOR_INT32} tensor, of shape * [num_rois], specifying the batch index of each box. Boxes with * the same batch index are grouped together. Zero num_rois is * supported for this tensor. * * 3: A 2-D Tensor of shape [batches, 2], specifying the information of * each image in the batch, each line with format * [image_height, image_width]. * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} as input0, with shape * [num_rois, num_classes * 4], specifying the coordinates of each * output bounding box for each class, with format [x1, y1, x2, y2]. * For type of {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_ASYMM}, the * scale must be 0.125 and the zero point must be 0. %insert AVAIL3 */ %{DeclareOperation_1.2 AXIS_ALIGNED_BBOX_TRANSFORM 41}, /** * A recurrent neural network layer that applies an LSTM cell to a * sequence of inputs in forward and backward directions. * * The op supports cross-linking via an auxiliary input. Regular cell feeds * one input into the two RNN cells in the following way: * * INPUT (INPUT_REVERSED) * | | * --------------------- * | FW_LSTM BW_LSTM | * --------------------- * | | * FW_OUT BW_OUT * * An op with cross-linking takes two inputs and feeds them into the RNN * cells in the following way: * * AUX_INPUT (AUX_INPUT_REVERSED) * | | * INPUT | (INPUT_R'D.)| * | | | | * ----------------------- * | \ / \ / | * | FW_LSTM BW_LSTM | * ----------------------- * | | * FW_OUT BW_OUT * * The cross-linking mode is enabled iff auxiliary input and auxiliary * weights are present. While stacking this op on top of itself, this * allows to connect both forward and backward outputs from previous cell * to the next cell's input. * %kind aidl canonical ndk hal_1.3+ * Since %{NNAPILevel4} parallel linking mode is supported. The mode is * enabled if auxiliary input is present but auxiliary weights are omitted. * In this case, the cell feeds inputs into the RNN in the following way: * * INPUT (AUX_INPUT_REVERSED) * | | * --------------------- * | FW_LSTM BW_LSTM | * --------------------- * | | * FW_OUT BW_OUT * * While stacking this op on top of itself, this allows to connect both * forward and backward outputs from previous cell to the next cell's * corresponding inputs. * %/kind * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * Supported tensor rank: 3, either time-major or batch-major. * * All input and output tensors must be of the same type. * * Inputs: * * 0: The input. * A 3-D tensor of shape: * If time-major: [max_time, batch_size, input_size] * If batch-major: [batch_size, max_time, input_size] * where "max_time" is the number of timesteps (sequence length), * "batch_size" corresponds to the batching dimension, and * "input_size" is the size of the input. * * 1: The forward input-to-input weights. Optional. * A 2-D tensor of shape [fw_num_units, input_size], where “fw_num_units” * corresponds to the number of forward cell units. * * 2: The forward input-to-forget weights. * A 2-D tensor of shape [fw_num_units, input_size]. * * 3: The forward input-to-cell weights. * A 2-D tensor of shape [fw_num_units, input_size]. * * 4: The forward input-to-output weights. * A 2-D tensor of shape [fw_num_units, input_size]. * * 5: The forward recurrent-to-input weights. Optional. * A 2-D tensor of shape [fw_num_units, fw_output_size], where “fw_output_size” * corresponds to either the number of cell units (i.e., fw_num_units), * or the second dimension of the “fw_projection_weights”, if defined. * * 6: The forward recurrent-to-forget weights. * A 2-D tensor of shape [fw_num_units, fw_output_size]. * * 7: The forward recurrent-to-cell weights. * A 2-D tensor of shape [fw_num_units, fw_output_size]. * * 8: The forward recurrent-to-output weights. * A 2-D tensor of shape [fw_num_units, fw_output_size]. * * 9: The forward cell-to-input weights. Optional. * A 1-D tensor of shape [fw_num_units]. * * 10: The forward cell-to-forget weights. Optional. * A 1-D tensor of shape [fw_num_units]. * * 11: The forward cell-to-output weights. Optional. * A 1-D tensor of shape [fw_num_units]. * * 12: The forward input gate bias. Optional. * A 1-D tensor of shape [fw_num_units]. * * 13: The forward forget gate bias. * A 1-D tensor of shape [fw_num_units]. * * 14: The forward cell gate bias. * A 1-D tensor of shape [fw_num_units]. * * 15: The forward output gate bias. * A 1-D tensor of shape [fw_num_units]. * * 16: The forward projection weights. Optional. * A 2-D tensor of shape [fw_output_size, fw_num_units]. * * 17: The forward projection bias. Optional. * A 1-D tensor of shape [fw_output_size]. * * 18: The backward input-to-input weights. Optional. * A 2-D tensor of shape [bw_num_units, input_size], where “bw_num_units” * corresponds to the number of backward cell units. * * 19: The backward input-to-forget weights. * A 2-D tensor of shape [bw_num_units, input_size]. * * 20: The backward input-to-cell weights. * A 2-D tensor of shape [bw_num_units, input_size]. * * 21: The backward input-to-output weights. * A 2-D tensor of shape [bw_num_units, input_size]. * * 22: The backward recurrent-to-input weights. Optional. * A 2-D tensor of shape [bw_num_units, bw_output_size], where “bw_output_size” * corresponds to either the number of cell units (i.e., “bw_num_units”), * or the second dimension of the “bw_projection_weights”, if defined. * * 23: The backward recurrent-to-forget weights. * A 2-D tensor of shape [bw_num_units, bw_output_size]. * * 24: The backward recurrent-to-cell weights. * A 2-D tensor of shape [bw_num_units, bw_output_size]. * * 25: The backward recurrent-to-output weights. * A 2-D tensor of shape [bw_num_units, bw_output_size]. * * 26: The backward cell-to-input weights. Optional. * A 1-D tensor of shape [bw_num_units]. * * 27: The backward cell-to-forget weights. Optional. * A 1-D tensor of shape [bw_num_units]. * * 28: The backward cell-to-output weights. Optional. * A 1-D tensor of shape [bw_num_units]. * * 29: The backward input gate bias. Optional. * A 1-D tensor of shape [bw_num_units]. * * 30: The backward forget gate bias. * A 1-D tensor of shape [bw_num_units]. * * 31: The backward cell gate bias. * A 1-D tensor of shape [bw_num_units]. * * 32: The backward output gate bias. * A 1-D tensor of shape [bw_num_units]. * * 33: The backward projection weights. Optional. * A 2-D tensor of shape [bw_output_size, bw_num_units]. * * 34: The backward projection bias. Optional. * A 1-D tensor of shape [bw_output_size]. * * 35: The forward input activation state. * A 2-D tensor of shape [batch_size, bw_output_size]. * * 36: The forward input cell state. * A 2-D tensor of shape [batch_size, bw_num_units]. * * 37: The backward input activation state. * A 2-D tensor of shape [batch_size, bw_output_size]. * * 38: The backward input cell state. * A 2-D tensor of shape [batch_size, bw_num_units]. %kind aidl canonical ndk hal_1.3+ * * 39: The auxiliary input. Optional. * A 3-D tensor of shape [max_time, batch_size, aux_input_size], * where “batch_size” corresponds to the batching dimension, and * “aux_input_size” is the size of the auxiliary input. Optional. See * the docs above for the usage modes explanation. * * 40: The forward auxiliary input-to-input weights. * Optional. See the docs above for the usage modes explanation. * A 2-D tensor of shape [fw_num_units, aux_input_size]. * * 41: The forward auxiliary input-to-forget weights. * Optional. See the docs above for the usage modes explanation. * A 2-D tensor of shape [fw_num_units, aux_input_size]. * * 42: The forward auxiliary input-to-cell weights. * Optional. See the docs above for the usage modes explanation. * A 2-D tensor of shape [fw_num_units, aux_input_size]. * * 43: The forward auxiliary input-to-output weights. * Optional. See the docs above for the usage modes explanation. * A 2-D tensor of shape [fw_num_units, aux_input_size]. * * 44: The backward auxiliary input-to-input weights. * Optional. See the docs above for the usage modes explanation. * A 2-D tensor of shape [bw_num_units, aux_input_size]. * * 45: The backward auxiliary input-to-forget weights. * Optional. See the docs above for the usage modes explanation. * A 2-D tensor of shape [bw_num_units, aux_input_size]. * * 46: The backward auxiliary input-to-cell weights. * Optional. See the docs above for the usage modes explanation. * A 2-D tensor of shape [bw_num_units, aux_input_size]. * * 47: The backward auxiliary input-to-output weights. * Optional. See the docs above for the usage modes explanation. * A 2-D tensor of shape [bw_num_units, aux_input_size]. %else * * 39: The auxiliary input. Optional. * A 3-D tensor of shape [max_time, batch_size, input_size], where “batch_size” * corresponds to the batching dimension, and “input_size” is the size * of the input. * * 40: The forward auxiliary input-to-input weights. Optional. * A 2-D tensor of shape [fw_num_units, input_size]. * * 41: The forward auxiliary input-to-forget weights. Optional. * A 2-D tensor of shape [fw_num_units, input_size]. * * 42: The forward auxiliary input-to-cell weights. Optional. * A 2-D tensor of shape [fw_num_units, input_size]. * * 43: The forward auxiliary input-to-output weights. Optional. * A 2-D tensor of shape [fw_num_units, input_size]. * * 44: The backward auxiliary input-to-input weights. Optional. * A 2-D tensor of shape [bw_num_units, input_size]. * * 45: The backward auxiliary input-to-forget weights. Optional. * A 2-D tensor of shape [bw_num_units, input_size]. * * 46: The backward auxiliary input-to-cell weights. Optional. * A 2-D tensor of shape [bw_num_units, input_size]. * * 47: The backward auxiliary input-to-output weights. Optional. * A 2-D tensor of shape [bw_num_units, input_size]. %/kind * * 48: The activation function. * A value indicating the activation function: * * * 49: The clipping threshold for the cell state, such * that values are bound within [-cell_clip, cell_clip]. If set to 0.0 * then clipping is disabled. * If all the input tensors have type {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32}, * this scalar must be of the type {@link %{OperandTypeLinkPfx}FLOAT32}, * otherwise if all the input tensors have the type * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16}, this scalar must be * of type {@link %{OperandTypeLinkPfx}FLOAT16}. * * 50: The clipping threshold for the output from the * projection layer, such that values are bound within * [-proj_clip, proj_clip]. If set to 0.0 then clipping is disabled. * If all the input tensors have type {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32}, * this scalar must be of the type {@link %{OperandTypeLinkPfx}FLOAT32}, * otherwise if all the input tensors have the type * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16}, this scalar must be * of type {@link %{OperandTypeLinkPfx}FLOAT16}. * * 51: merge_outputs * An {@link %{OperandTypeLinkPfx}BOOL} scalar specifying if the outputs * from forward and backward cells should be merged. * * 52: time_major * An {@link %{OperandTypeLinkPfx}BOOL} scalar specifying the shape format * of input and output tensors. * * 53: The forward input layer normalization weights. Optional. * A 1-D tensor of shape [fw_num_units]. Used to rescale normalized inputs * to activation at input gate. * * 54: The forward forget layer normalization weights. Optional. * A 1-D tensor of shape [fw_num_units]. Used to rescale normalized inputs * to activation at forget gate. * * 55: The forward cell layer normalization weights. Optional. * A 1-D tensor of shape [fw_num_units]. Used to rescale normalized inputs * to activation at cell gate. * * 56: The forward output layer normalization weights. Optional. * A 1-D tensor of shape [fw_num_units]. Used to rescale normalized inputs * to activation at output gate. * * 57: The backward input layer normalization weights. Optional. * A 1-D tensor of shape [bw_num_units]. Used to rescale normalized inputs * to activation at input gate. * * 58: The backward forget layer normalization weights. Optional. * A 1-D tensor of shape [bw_num_units]. Used to rescale normalized inputs * to activation at forget gate. * * 59: The backward cell layer normalization weights. Optional. * A 1-D tensor of shape [bw_num_units]. Used to rescale normalized inputs * to activation at cell gate. * * 60: The backward output layer normalization weights. Optional. * A 1-D tensor of shape [bw_num_units]. Used to rescale normalized inputs * to activation at output gate. * * Outputs: * * 0: The forward output. * A 3-D tensor of shape: * If time-major and not merge_outputs: * [max_time, batch_size, fw_output_size] * If time-major and merge_outputs: * [max_time, batch_size, fw_output_size + bw_output_size] * If batch-major and not merge_outputs: * [batch_size, max_time, fw_output_size] * If batch-major and merge_outputs: * [batch_size, max_time, fw_output_size + bw_output_size] * * 1: The backward output. Unused if merge_outputs is true. * A 3-D tensor of shape: * If time-major: [max_time, batch_size, bw_output_size] * If batch-major: [batch_size, max_time, bw_output_size] %kind aidl canonical ndk hal_1.3+ * * 2: The forward activation state output. * A 2-D tensor of shape [batch_size, fw_output_size] containing an * activation state from the last time step in the sequence. This * output is optional and can be omitted. If this output is present * then outputs 3-5 must be present as well. * Available since %{NNAPILevel4}. * * 3: The forward cell state output. * A tensor of shape [batch_size, fw_cell_size] containing a cell state * from the last time step in the sequence. This output is optional * and can be omitted. If this output is present * then outputs 2, 4, 5 must be present as well. * Available since %{NNAPILevel4}. * * 4: The backward activation state output. * A 2-D tensor of shape [batch_size, bw_output_size] containing an * activation state from the last time step in the sequence. This * output is optional and can be omitted. If this output is present * then outputs 2, 3, 5 must be present as well. * Available since %{NNAPILevel4}. * * 5: The backward cell state output. * A tensor of shape [batch_size, bw_cell_size] containing a cell state * from the last time step in the sequence. This output is optional * and can be omitted. If this output is present * then outputs 2-4 must be present as well. * Available since %{NNAPILevel4}. %/kind %insert AVAIL3 %insert OutputState */ %{DeclareOperation_1.2 BIDIRECTIONAL_SEQUENCE_LSTM 42}, /** * A recurrent neural network layer that applies a basic RNN cell to a * sequence of inputs in forward and backward directions. * * This Op unrolls the input along the sequence dimension, and implements * the following operation for each element in the sequence s = * 1...sequence_length: * fw_outputs[s] = fw_state = activation(inputs[s] * fw_input_weights’ + * fw_state * fw_recurrent_weights’ + fw_bias) * * And for each element in sequence t = sequence_length : 1 * bw_outputs[t] = bw_state = activation(inputs[t] * bw_input_weights’ + * bw_state * bw_recurrent_weights’ + bw_bias) * * Where: * * “{fw,bw}_input_weights” is a weight matrix that multiplies the inputs; * * “{fw,bw}_recurrent_weights” is a weight matrix that multiplies the * current “state” which itself is the output from the previous time step * computation; * * “{fw,bw}_bias” is a bias vector (added to each output vector in the * batch); * * “activation” is the function passed as the “fused_activation_function” * argument (if not “NONE”). * * The op supports cross-linking via an auxiliary input. Regular cell feeds * one input into the two RNN cells in the following way: * * INPUT (INPUT_REVERSED) * | | * --------------------- * | FW_RNN BW_RNN | * --------------------- * | | * FW_OUT BW_OUT * * An op with cross-linking takes two inputs and feeds them into the RNN * cells in the following way: * * AUX_INPUT (AUX_INPUT_REVERSED) * | | * INPUT | (INPUT_R'D.)| * | | | | * ----------------------- * | \ / \ / | * | FW_RNN BW_RNN | * ----------------------- * | | * FW_OUT BW_OUT * * The cross-linking mode is enabled iff auxiliary input and auxiliary * weights are present. While stacking this op on top of itself, this * allows to connect both forward and backward outputs from previous cell * to the next cell's input. * %kind aidl canonical ndk hal_1.3+ * Since %{NNAPILevel4} parallel linking mode is supported. The mode is * enabled if auxiliary input is present but auxiliary weights are omitted. * In this case, the cell feeds inputs into the RNN in the following way: * * INPUT (AUX_INPUT_REVERSED) * | | * --------------------- * | FW_RNN BW_RNN | * --------------------- * | | * FW_OUT BW_OUT * * While stacking this op on top of itself, this allows to connect both * forward and backward outputs from previous cell to the next cell's * corresponding inputs. * %/kind * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * The input tensors must all be the same type. * * Inputs: * * 0: input. * A 3-D tensor. The shape is defined by the input 6 (timeMajor). If * it is set to true, then the input has a shape [maxTime, batchSize, * inputSize], otherwise the input has a shape [batchSize, maxTime, * inputSize]. * * 1: fwWeights. * A 2-D tensor of shape [fwNumUnits, inputSize]. * * 2: fwRecurrentWeights. * A 2-D tensor of shape [fwNumUnits, fwNumUnits]. * * 3: fwBias. * A 1-D tensor of shape [fwNumUnits]. * * 4: fwHiddenState. * A 2-D tensor of shape [batchSize, fwNumUnits]. Specifies a hidden * state input for the first time step of the computation. * * 5: bwWeights. * A 2-D tensor of shape [bwNumUnits, inputSize]. * * 6: bwRecurrentWeights. * A 2-D tensor of shape [bwNumUnits, bwNumUnits]. * * 7: bwBias. * A 1-D tensor of shape [bwNumUnits]. * * 8: bwHiddenState * A 2-D tensor of shape [batchSize, bwNumUnits]. Specifies a hidden * state input for the first time step of the computation. %kind aidl canonical ndk hal_1.3+ * * 9: auxInput. * A 3-D tensor. The shape is defined by the input 6 (timeMajor). If * it is set to true, then the input has a shape [maxTime, batchSize, * auxInputSize], otherwise the input has a shape [batchSize, maxTime, * auxInputSize]. Can be omitted. See the docs above for the usage * modes explanation. * * 10:fwAuxWeights. * A 2-D tensor of shape [fwNumUnits, auxInputSize]. Can be omitted. * See the docs above for the usage modes explanation. * * 11:bwAuxWeights. * A 2-D tensor of shape [bwNumUnits, auxInputSize]. Can be omitted. * See the docs above for the usage modes explanation. %else * * 9: auxInput. * A 3-D tensor. The shape is the same as of the input 0. * * 10:fwAuxWeights. * A 2-D tensor of shape [fwNumUnits, inputSize]. * * 11:bwAuxWeights. * A 2-D tensor of shape [bwNumUnits, inputSize]. %/kind * * 12:fusedActivationFunction. * A {@link %{FusedActivationFunc}} value indicating the activation function. If * “NONE” is specified then it results in a linear activation. * * 13:timeMajor * An {@link %{OperandTypeLinkPfx}BOOL} scalar specifying the shape format * of input and output tensors. * * 14:mergeOutputs * An {@link %{OperandTypeLinkPfx}BOOL} scalar specifying if the outputs * from forward and backward cells are separate (if set to false) or * concatenated (if set to true). * Outputs: * * 0: fwOutput. * A 3-D tensor. The first two dimensions of the shape are defined by * the input 6 (timeMajor) and the third dimension is defined by the * input 14 (mergeOutputs). If timeMajor is set to true, then the first * two dimensions are [maxTime, batchSize], otherwise they are set to * [batchSize, maxTime]. If mergeOutputs is set to true, then the third * dimension is equal to (fwNumUnits + bwNumUnits), otherwise it is set * to fwNumUnits. * * 1: bwOutput. * A 3-D tensor. If the input 14 (mergeOutputs) is set to true, then * this tensor is not produced. The shape is defined by the input 6 * (timeMajor). If it is set to true, then the shape is set to * [maxTime, batchSize, bwNumUnits], otherwise the shape is set to * [batchSize, maxTime, bwNumUnits]. %kind aidl canonical ndk hal_1.3+ * * 2: The forward hidden state output. * A 2-D tensor of shape [batchSize, fwNumUnits] containing a hidden * state from the last time step in the sequence. This output is * optional and can be omitted. If this output is present then output * 3 must be present as well. * Available since %{NNAPILevel4}. * * 3: The backward hidden state output. * A 2-D tensor of shape [batchSize, bwNumUnits] containing a hidden * state from the last time step in the sequence. This output is * optional and can be omitted. If this output is present then output * 2 must be present as well. * Available since %{NNAPILevel4}. %/kind %insert AVAIL3 %insert OutputState */ %{DeclareOperation_1.2 BIDIRECTIONAL_SEQUENCE_RNN 43}, /** * Greedily selects a subset of bounding boxes in descending order of score. * * This op applies NMS algorithm to each class. In each loop of execution, * the box with maximum score gets selected and removed from the pending set. * The scores of the rest of boxes are lowered according to the * intersection-over-union (IOU) overlapping with the previously selected * boxes and a specified NMS kernel method. Any boxes with score less * than a threshold are removed from the pending set. * * Three NMS kernels are supported: * * Hard: score_new = score_old * (1 if IoU < threshold else 0) * * Linear: score_new = score_old * (1 if IoU < threshold else 1 - IoU) * * Gaussian: score_new = score_old * exp(- IoU^2 / sigma) * * Axis-aligned bounding boxes are represented by its upper-left corner * coordinate (x1,y1) and lower-right corner coordinate (x2,y2). A valid * bounding box should satisfy x1 <= x2 and y1 <= y2. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Inputs: * * 0: A 2-D Tensor of shape [num_rois, num_classes], specifying the score * of each bounding box proposal. The boxes are grouped by batches in the * first dimension. Zero num_rois is supported for this tensor. * * 1: A 2-D Tensor specifying the bounding boxes of shape * [num_rois, num_classes * 4], organized in the order [x1, y1, x2, y2]. * The boxes are grouped by batches in the first dimension. The sequential * order of the boxes corresponds with input0. For input0 of type * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, this tensor should be of * {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_ASYMM}, with zeroPoint of 0 and * scale of 0.125. %kind aidl canonical ndk hal_1.3+ * For input0 of type {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED}, * this tensor should be of {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_ASYMM}, * with zeroPoint of -128 and scale of 0.125. %/kind * Zero num_rois is supported for this tensor. * * 2: A 1-D {@link %{OperandTypeLinkPfx}TENSOR_INT32} tensor, of shape * [num_rois], specifying the batch index of each box. Boxes with * the same batch index are grouped together. * * 3: An {@link %{OperandTypeLinkPfx}FLOAT32} scalar, score_threshold. Boxes * with scores lower than the threshold are filtered before sending * to the NMS algorithm. * * 4: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the maximum * number of selected bounding boxes for each image. Set to a negative * value for unlimited number of output bounding boxes. * * 5: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the NMS * kernel method, options are 0:hard, 1:linear, 2:gaussian. * * 6: An {@link %{OperandTypeLinkPfx}FLOAT32} scalar, specifying the IoU * threshold in hard and linear NMS kernel. This field is ignored if * gaussian kernel is selected. * * 7: An {@link %{OperandTypeLinkPfx}FLOAT32} scalar, specifying the sigma in * gaussian NMS kernel. This field is ignored if gaussian kernel is * not selected. * * 8: An {@link %{OperandTypeLinkPfx}FLOAT32} scalar, nms_score_threshold. * Boxes with scores lower than the threshold are dropped during the * score updating phase in soft NMS. * * Outputs: * * 0: A 1-D Tensor of the same {@link %{OperandType}} as input0, with shape * [num_output_rois], specifying the score of each output box. The boxes * are grouped by batches, but the sequential order in each batch is not * guaranteed. For type of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, %kind aidl canonical ndk hal_1.3+ * guaranteed. For type of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * or {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED}, %else * guaranteed. For type of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %/kind * the scale and zero point must be the same as input0. * * 1: A 2-D Tensor of the same {@link %{OperandType}} as input1, with shape * [num_output_rois, 4], specifying the coordinates of each * output bounding box with the same format as input1. The sequential * order of the boxes corresponds with output0. For type of * {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_ASYMM}, the scale must be * 0.125 and the zero point must be 0. * * 2: A 1-D {@link %{OperandTypeLinkPfx}TENSOR_INT32} tensor, of shape * [num_output_rois], specifying the class of each output box. The * sequential order of the boxes corresponds with output0. * * 3: A 1-D {@link %{OperandTypeLinkPfx}TENSOR_INT32} tensor, of shape * [num_output_rois], specifying the batch index of each box. Boxes * with the same batch index are grouped together. %insert AVAIL3 */ %{DeclareOperation_1.2 BOX_WITH_NMS_LIMIT 44}, /** * Casts a tensor to a type. * * This operation ignores the scale and zeroPoint of quanized tensors, * e.g. it treats a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} input * as a tensor of uint8 values. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * Since %{NNAPILevel4}, casting tensors of the following * {@link %{OperandType}} to the same {@link %{OperandType}} is supported: * * {@link %{OperandTypeLinkPfx}TENSOR_BOOL8} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_ASYMM} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_SYMM} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM} %/kind * * Supported tensor rank: from 1 * * Inputs: * * 0: A tensor. * * Outputs: * * 0: A tensor with the same shape as input0. %insert AVAIL3 */ %{DeclareOperation_1.2 CAST 45}, /** * Shuffle the channels of the input tensor. * * Given an input tensor and a integer value of num_groups, CHANNEL_SHUFFLE * divide the channel dimension into num_groups groups, and reorganize the * channels by grouping channels with the same index in each group. * * Along the channel dimension, the output is calculated using this formula: * * output_channel[k * num_groups + g] = input_channel[g * group_size + k] * * where group_size = num_channels / num_groups * * The number of channels must be divisible by num_groups. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: up to 4 * * Inputs: * * 0: An n-D tensor, specifying the tensor to be shuffled. * * 1: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the number of * groups. * * 2: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the dimension * channel shuffle would be performed on. Negative index is used to * specify axis from the end (e.g. -1 for the last axis). Must be in * the range [-n, n). * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} and same shape as input0. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint must be the same as input0. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint must be the same as input0. %/kind %insert AVAIL3 */ %{DeclareOperation_1.2 CHANNEL_SHUFFLE 46}, /** * Apply postprocessing steps to bounding box detections. * * Bounding box detections are generated by applying transformation on a set * of predefined anchors with the bounding box deltas from bounding box * regression. A final step of hard NMS is applied to limit the number of * returned boxes. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * Inputs: * * 0: A 3-D Tensor of shape [batches, num_anchors, num_classes], specifying * the score of each anchor with each class. Class 0 for each * [batches, num_anchors, 0] is background and will be ignored. * * 1: A 3-D Tensor of shape [batches, num_anchors, length_box_encoding], with * the first four values in length_box_encoding specifying the bounding * box deltas. The box deltas are encoded in the order of [dy, dx, dh, dw], * where dy and dx is the linear-scale relative correction factor for the * center position of the bounding box with respect to the width and height, * dh and dw is the log-scale relative correction factor for the width and * height. All the entries in length_box_encoding beyond the first four * values are ignored in this operation. * * 2: A 2-D Tensor of shape [num_anchors, 4], specifying the shape of each * predefined anchor, with format [ctr_y, ctr_x, h, w], where ctr_y and * ctr_x are the center position of the box, and h and w are the height * and the width. * * 3: An {@link %{OperandTypeLinkPfx}FLOAT32} scalar, specifying the scaling * factor for dy in bounding box deltas. * * 4: An {@link %{OperandTypeLinkPfx}FLOAT32} scalar, specifying the scaling * factor for dx in bounding box deltas. * * 5: An {@link %{OperandTypeLinkPfx}FLOAT32} scalar, specifying the scaling * factor for dh in bounding box deltas. * * 6: An {@link %{OperandTypeLinkPfx}FLOAT32} scalar, specifying the scaling * factor for dw in bounding box deltas. * * 7: An {@link %{OperandTypeLinkPfx}BOOL} scalar, set to true to use regular * multi-class NMS algorithm that do NMS separately for each class, * set to false for a faster algorithm that only do one single NMS * using the highest class score.. * * 8: An {@link %{OperandTypeLinkPfx}INT32} scalar, max_num_detections, specifying * the maximum number of boxes for the output. Boxes with the lowest * scores are discarded to meet the limit. * * 9: An {@link %{OperandTypeLinkPfx}INT32} scalar, only used when input7 is * set to false, specifying the maximum number of classes per detection. * * 10: An {@link %{OperandTypeLinkPfx}INT32} scalar, only used when input7 is * set to true, specifying the maximum number of detections when * applying NMS algorithm for each single class. * * 11: A scalar, score_threshold. Boxes with scores lower than the * threshold are filtered before sending to the NMS algorithm. The * scalar must be of {@link %{OperandTypeLinkPfx}FLOAT16} if input0 is of * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} and of * {@link %{OperandTypeLinkPfx}FLOAT32} if input0 is of * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32}. * * 12: A scalar, specifying the IoU threshold for hard NMS. The scalar * must be of {@link %{OperandTypeLinkPfx}FLOAT16} if input0 is of * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} and of * {@link %{OperandTypeLinkPfx}FLOAT32} if input0 is of * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32}. * * 13: An {@link %{OperandTypeLinkPfx}BOOL} scalar, set to true to include * background class in the list of label map for the output, set * to false to not include the background. When the background * class is included, it has label 0 and the output classes start * at 1 in the label map, otherwise, the output classes start at 0. * * Outputs: * * 0: A 2-D tensor of the same {@link %{OperandType}} as input0, with shape * [batches, max_num_detections], specifying the score of each output * detections. * * 1: A 3-D tensor of shape [batches, max_num_detections, 4], specifying the * coordinates of each output bounding box, with format * [y1, x1, y2, x2]. * * 2: A 2-D {@link %{OperandTypeLinkPfx}TENSOR_INT32} tensor, of shape * [batches, max_num_detections], specifying the class label for each * output detection. * * 3: An 1-D {@link %{OperandTypeLinkPfx}TENSOR_INT32} tensor, of shape [batches], * specifying the number of valid output detections for each batch. %insert AVAIL3 */ %{DeclareOperation_1.2 DETECTION_POSTPROCESSING 47}, /** * For input tensors x and y, computes x == y elementwise. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_BOOL8} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: from 1 * * This operation supports broadcasting. * * Inputs: * * 0: A tensor. * * 1: A tensor of the same {@link %{OperandType}} and dimensions compatible * with input0. * * Outputs: * * 0: A tensor of {@link %{OperandTypeLinkPfx}TENSOR_BOOL8}. %insert AVAIL3 */ %{DeclareOperation_1.2 EQUAL 48}, /** * Computes exponential of x element-wise. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * Supported tensor rank: from 1. * * Inputs: * * 0: A tensor. * * Outputs: * * 0: The output tensor of same shape as input0. %insert AVAIL3 */ %{DeclareOperation_1.2 EXP 49}, /** * Inserts a dimension of 1 into a tensor's shape. * * Given a tensor input, this operation inserts a dimension of 1 at the * given dimension index of input's shape. The dimension index starts at * zero; if you specify a negative dimension index, it is counted backward * from the end. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: from 1 * * Inputs: * * 0: An n-D tensor. * * 1: An {@link %{OperandTypeLinkPfx}INT32} scalar specifying the dimension * index to expand. Must be in the range [-(n + 1), (n + 1)). * * Outputs: * * 0: An (n + 1)-D tensor with the same {@link %{OperandType}} and data as * input0. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, %/kind * the scale and zeroPoint must be the same as input0. %insert AVAIL3 */ %{DeclareOperation_1.2 EXPAND_DIMS 50}, /** * Gathers values along an axis. * * Produces an output tensor with shape * input0.dimension[:axis] + indices.dimension + input0.dimension[axis + 1:] * where: * # Vector indices (output is rank(input0)). * output[a_0, ..., a_n, i, b_0, ..., b_n] = * input0[a_0, ..., a_n, indices[i], b_0, ..., b_n] * * # Higher rank indices (output is rank(input0) + rank(indices) - 1). * output[a_0, ..., a_n, i, ..., j, b_0, ... b_n] = * input0[a_0, ..., a_n, indices[i, ..., j], b_0, ..., b_n] * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: from 1 * * Inputs: * * 0: An n-D tensor from which to gather values. * * 1: An {@link %{OperandTypeLinkPfx}INT32} scalar specifying the axis. * Negative index is used to specify axis from the end * (e.g. -1 for the last axis). Must be in the range [-n, n). * * 2: A k-D tensor {@link %{OperandTypeLinkPfx}TENSOR_INT32} of indices. * The values must be in the bounds of the corresponding dimensions * of input0. * * Outputs: * * 0: An (n + k - 1)-D tensor with the same {@link %{OperandType}} as input0. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, %/kind * the scale and zeroPoint must be the same as input0. %insert AVAIL3 */ %{DeclareOperation_1.2 GATHER 51}, /** * Generate aixs-aligned bounding box proposals. * * Bounding box proposals are generated by applying transformation on a set * of predefined anchors with the bounding box deltas from bounding box * regression. A final step of hard NMS is applied to limit the number of * returned boxes. * * Axis-aligned bounding boxes are represented by its upper-left corner * coordinate (x1,y1) and lower-right corner coordinate (x2,y2). A valid * bounding box should satisfy x1 <= x2 and y1 <= y2. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Inputs: * * 0: A 4-D Tensor specifying the score of each anchor at each * location. With "NHWC" data layout, the tensor shape is * [batches, height, width, num_anchors]. With "NCHW" data layout, * the tensor shape is [batches, num_anchors, height, width]. * * 1: A 4-D Tensor specifying the bounding box deltas. With "NHWC" data * layout, the tensor shape is [batches, height, width, num_anchors * 4]. * With "NCHW" data layout, the tensor shape is * [batches, num_anchors * 4, height, width]. The box deltas are encoded * in the order of [dx, dy, dw, dh], where dx and dy is the linear-scale * relative correction factor for the center position of the bounding box * with respect to the width and height, dw and dh is the log-scale * relative correction factor for the width and height. The last * dimensions is the channel dimension. * * 2: A 2-D Tensor of shape [num_anchors, 4], specifying the shape of each * predefined anchor, with format [x1, y1, x2, y2]. For input0 of type %kind aidl canonical ndk hal_1.3+ * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} or * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED}, this tensor should be of %else * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, this tensor should be of %/kind * {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_SYMM}, with scale of 0.125. * * 3: A 2-D Tensor of shape [batches, 2], specifying the size of * each image in the batch, with format [image_height, image_width]. %kind aidl canonical ndk hal_1.3+ * For input0 of type {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} or * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED}, this %else * For input0 of type {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, this %/kind * tensor should be of {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_SYMM}, with * scale of 0.125. * * 4: An {@link %{OperandTypeLinkPfx}FLOAT32} scalar, specifying the ratio * from the height of original image to the height of feature map. * * 5: An {@link %{OperandTypeLinkPfx}FLOAT32} scalar, specifying the ratio * from the width of original image to the width of feature map. * * 6: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the maximum * number of boxes before going into the hard NMS algorithm. Boxes * with the lowest scores are discarded to meet the limit. Set to * a non-positive value for unlimited number. * * 7: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the maximum * number of boxes returning from the hard NMS algorithm. Boxes * with the lowest scores are discarded to meet the limit. Set to * a non-positive value for unlimited number. * * 8: An {@link %{OperandTypeLinkPfx}FLOAT32} scalar, specifying the IoU * threshold for hard NMS. * * 9: An {@link %{OperandTypeLinkPfx}FLOAT32} scalar, min_size. Boxes with * height or width lower than the absolute threshold are filtered out. * * 10: An {@link %{OperandTypeLinkPfx}BOOL} scalar, set to true to specify * NCHW data layout for input0 and input1. Set to false for NHWC. * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} as input0, of shape * [num_output_rois], specifying the score of each output box. * The boxes are grouped by batches, but the sequential order in * each batch is not guaranteed. For type of %kind aidl canonical ndk hal_1.3+ * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} or * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED}, the scale and zero %else * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, the scale and zero %/kind * point must be the same as input0. * * 1: A tensor of the same {@link %{OperandType}} as input3, of shape * [num_output_rois, 4], specifying the coordinates of each output * bounding box for each class, with format [x1, y1, x2, y2]. * The sequential order of the boxes corresponds with output0. * For type of {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_ASYMM}, the * scale must be 0.125 and the zero point must be 0. * * 2: A 1-D {@link %{OperandTypeLinkPfx}TENSOR_INT32} tensor, of shape * [num_output_rois], specifying the batch index of each box. Boxes * with the same batch index are grouped together. %insert AVAIL3 */ %{DeclareOperation_1.2 GENERATE_PROPOSALS 52}, /** * For input tensors x and y, computes x > y elementwise. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_BOOL8} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: from 1 * * This operation supports broadcasting. * * Inputs: * * 0: A tensor. * * 1: A tensor of the same {@link %{OperandType}} and dimensions compatible * with input0. * * Outputs: * * 0: A tensor of {@link %{OperandTypeLinkPfx}TENSOR_BOOL8}. %insert AVAIL3 */ %{DeclareOperation_1.2 GREATER 53}, /** * For input tensors x and y, computes x >= y elementwise. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_BOOL8} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: from 1 * * This operation supports broadcasting. * * Inputs: * * 0: A tensor. * * 1: A tensor of the same {@link %{OperandType}} and dimensions compatible * with input0. * * Outputs: * * 0: A tensor of {@link %{OperandTypeLinkPfx}TENSOR_BOOL8}. %insert AVAIL3 */ %{DeclareOperation_1.2 GREATER_EQUAL 54}, /** * Performs a grouped 2-D convolution operation. * * Given an input tensor of shape [batches, height, width, depth_in] and a * filter tensor of shape [depth_out, filter_height, filter_width, depth_group] * containing depth_out convolutional filters of depth depth_group, GROUPED_CONV * applies a group of different filters to each input channel group, then * concatenates the results together. * * Specifically, the input channels are divided into num_groups groups, each with * depth depth_group, i.e. depth_in = num_groups * depth_group. The convolutional * filters are also divided into num_groups groups, i.e. depth_out is divisible * by num_groups. GROUPED_CONV applies each group of filters to the corresponding * input channel group, and the result are concatenated together. * * The output dimensions are functions of the filter dimensions, stride, and * padding. * * The values in the output tensor are computed as: * * output[b, i, j, g * channel_multiplier + q] = * sum_{di, dj, dk} ( * input[b, strides[1] * i + di, strides[2] * j + dj, * g * depth_group + dk] * * filter[g * channel_multiplier + q, di, dj, dk] * ) + bias[channel] * * where channel_multiplier = depth_out / num_groups * * Supported tensor {@link %{OperandType}} configurations: * * 16 bit floating point: * * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} for input, filter, output, and bias. * * * 32 bit floating point: * * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} for input, filter, output, and bias. * * * Quantized: * * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} for input, filter, and output. * * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} for bias (with scale set to * * * input.scale * filter.scale). %kind aidl canonical ndk hal_1.3+ * * * Quantized signed (since %{NNAPILevel4}): * * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} for input, filter, and output. * * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} for bias (with scale set to * * * input.scale * filter.scale). %/kind * * * Quantized with symmetric per channel quantization for the filter: * * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} for input, and output. * * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM_PER_CHANNEL} for filter. * * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} for bias (scale set to 0.0, * * * each value scaling is separate and equal to input.scale * filter.scales[channel]). %kind aidl canonical ndk hal_1.3+ * * * Quantized signed with filter symmetric per channel quantization * (since %{NNAPILevel4}): * * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} for input, and output. * * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM_PER_CHANNEL} for filter. * * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} for bias (scale set to 0.0, * * * each value scaling is separate and equal to input.scale * filter.scales[channel]). %/kind * * Supported tensor rank: 4, with "NHWC" or "NCHW" data layout. * With the default data layout NHWC, the data is stored in the order of: * [batch, height, width, channels]. Alternatively, the data layout could * be NCHW, the data storage order of: [batch, channels, height, width]. * * Both explicit padding and implicit padding are supported. * * Inputs (explicit padding): * * 0: A 4-D tensor, of shape [batches, height, width, depth_in], * specifying the input, where depth_in = num_groups * depth_group. * * 1: A 4-D tensor, of shape * [depth_out, filter_height, filter_width, depth_group], specifying * the filter, where depth_out must be divisible by num_groups. For * tensor of type {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM_PER_CHANNEL} * the channel dimension (channelDim at * {@link %{Ann}SymmPerChannelQuantParams}) must be set to 0. * * 2: A 1-D tensor, of shape [depth_out], specifying the bias. For input * tensor of type {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} or * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16}, the bias must be of the same type. %kind aidl canonical ndk hal_1.3+ * For filter tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} %else * For filter tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, %/kind * the bias should be of {@link %{OperandTypeLinkPfx}TENSOR_INT32}, with zeroPoint * of 0 and bias_scale == input_scale * filter_scale. For filter tensor * of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM_PER_CHANNEL}, the bias * should be of {@link %{OperandTypeLinkPfx}TENSOR_INT32}, with zeroPoint of * 0 and bias_scale of 0. The actual scale of each value 'i' is equal to * bias_scale[i] = input_scale * filter_scale[i]. * * 3: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the left, in the ‘width’ dimension. * * 4: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the right, in the ‘width’ dimension. * * 5: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the top, in the ‘height’ dimension. * * 6: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the bottom, in the ‘height’ dimension. * * 7: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘width’ dimension. * * 8: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘height’ dimension. * * 9: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the number of * groups. * * 10: An {@link %{OperandTypeLinkPfx}INT32} scalar, and has to be one of the * {@link %{FusedActivationFunc}} values. Specifies the activation to * invoke on the result. * * 11: An {@link %{OperandTypeLinkPfx}BOOL} scalar, set to true to specify * NCHW data layout for input0 and output0. Set to false for NHWC. * * Inputs (implicit padding): * * 0: A 4-D tensor, of shape [batches, height, width, depth_in], * specifying the input, where depth_in = num_groups * depth_group. * * 1: A 4-D tensor, of shape * [depth_out, filter_height, filter_width, depth_group], specifying * the filter, where depth_out must be divisible by num_groups. For * tensor of type {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM_PER_CHANNEL} * the channel dimension (%{Ann}SymmPerChannelQuantParams::channelDim) * must be set to 0. * * 2: A 1-D tensor, of shape [depth_out], specifying the bias. For input * tensor of type {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} or * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16}, the bias must be of the same * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16}, the bias must be of the same type. %kind aidl canonical ndk hal_1.3+ * For filter tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} %else * For filter tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, %/kind * the bias should be of {@link %{OperandTypeLinkPfx}TENSOR_INT32}, with zeroPoint * of 0 and bias_scale == input_scale * filter_scale. For filter tensor * of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM_PER_CHANNEL}, the bias * should be of {@link %{OperandTypeLinkPfx}TENSOR_INT32}, with zeroPoint of * 0 and bias_scale of 0. The actual scale of each value 'i' is equal to * bias_scale[i] = input_scale * filter_scale[i]. * * 3: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the implicit * padding scheme, has to be one of the %insert PaddingCodeValues * * 4: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘width’ dimension. * * 5: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘height’ dimension. * * 6: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the number of * groups. * * 7: An {@link %{OperandTypeLinkPfx}INT32} scalar, and has to be one of the * {@link %{FusedActivationFunc}} values. Specifies the activation to * invoke on the result. * * 8: An {@link %{OperandTypeLinkPfx}BOOL} scalar, set to true to specify * NCHW data layout for input0 and output0. Set to false for NHWC. * * Outputs: * * 0: The output 4-D tensor, of shape * [batches, out_height, out_width, depth_out]. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint can be different from inputs' scale and zeroPoint. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint can be different from inputs' scale and zeroPoint. %/kind %insert AVAIL3 */ %{DeclareOperation_1.2 GROUPED_CONV_2D 55}, /** * Localize the maximum keypoints from heatmaps. * * This operation approximates the accurate maximum keypoint scores and * indices after bicubic upscaling by using Taylor expansion up to the * quadratic term. * * The bounding box is represented by its upper-left corner coordinate * (x1,y1) and lower-right corner coordinate (x2,y2) in the original image. * A valid bounding box should satisfy x1 <= x2 and y1 <= y2. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: 4, with "NHWC" or "NCHW" data layout. * With the default data layout NHWC, the data is stored in the order of: * [batch, height, width, channels]. Alternatively, the data layout could * be NCHW, the data storage order of: [batch, channels, height, width]. * * Inputs: * * 0: A 4-D Tensor of shape * [num_boxes, heatmap_size, heatmap_size, num_keypoints], * specifying the heatmaps, the height and width of heatmaps should * be the same, and must be greater than or equal to 2. * * 1: A 2-D Tensor of shape [num_boxes, 4], specifying the bounding boxes, * each with format [x1, y1, x2, y2]. For input0 of type * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, this tensor should * be of {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_ASYMM}, with zeroPoint * of 0 and scale of 0.125. %kind aidl canonical ndk hal_1.3+ * For input0 of type * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED}, this tensor * should be of {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_ASYMM}, with * zeroPoint of -128 and scale of 0.125. %/kind * * 2: An {@link %{OperandTypeLinkPfx}BOOL} scalar, set to true to specify * NCHW data layout for input0. Set to false for NHWC. * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} as input0, with shape * [num_boxes, num_keypoints], specifying score of the keypoints. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} or * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint can be different from input0 scale and zeroPoint. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint can be different from input0 scale and zeroPoint. %/kind * * 1: A tensor of the same {@link %{OperandType}} as input1, with shape * [num_boxes, num_keypoints, 2], specifying the location of * the keypoints, the second dimension is organized as * [keypoint_x, keypoint_y]. * For type of {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_ASYMM}, the * scale must be 0.125 and the zero point must be 0. %insert AVAIL3 */ %{DeclareOperation_1.2 HEATMAP_MAX_KEYPOINT 56}, /** * Applies instance normalization to the input tensor. * * The values in the output tensor are computed as: * * output[b, h, w, c] = * (input[b, h, w, c] - mean[b, c]) * gamma / * sqrt(var[b, c] + epsilon) + beta * * Where the mean and variance are computed across the spatial dimensions: * * mean[b, c] = * sum_{h, w}(input[b, h, w, c]) / sum(1) * * var[b, c] = * sum_{h, w}(pow(input[b, h, w, c] - mean[b, c], 2)) / sum(1) * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * Supported tensor rank: 4, with "NHWC" or "NCHW" data layout. * With the default data layout NHWC, the data is stored in the order of: * [batch, height, width, channels]. Alternatively, the data layout could * be NCHW, the data storage order of: [batch, channels, height, width]. * * Inputs: * * 0: An n-D tensor, specifying the tensor to be normalized. * * 1: A scalar, specifying gamma, the scale applied to the normalized * tensor. The scalar must be of {@link %{OperandTypeLinkPfx}FLOAT16} if * input0 is of {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} and of * {@link %{OperandTypeLinkPfx}FLOAT32} if input0 is of * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32}. * * 2: A scalar, specifying beta, the offset applied to the normalized * tensor. The scalar must be of {@link %{OperandTypeLinkPfx}FLOAT16} if * input0 is of {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} and of * {@link %{OperandTypeLinkPfx}FLOAT32} if input0 is of * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32}. * * 3: A scalar, specifying epsilon, the small value added to variance to * avoid dividing by zero. The scalar must be of {@link %{OperandTypeLinkPfx}FLOAT16} if * input0 is of {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} and of * {@link %{OperandTypeLinkPfx}FLOAT32} if input0 is of * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32}. * * 4: An {@link %{OperandTypeLinkPfx}BOOL} scalar, set to true to specify * NCHW data layout for input0 and output0. Set to false for NHWC. * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} and same shape as input0. %insert AVAIL3 */ %{DeclareOperation_1.2 INSTANCE_NORMALIZATION 57}, /** * For input tensors x and y, computes x < y elementwise. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_BOOL8} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: from 1 * * This operation supports broadcasting. * * Inputs: * * 0: A tensor. * * 1: A tensor of the same {@link %{OperandType}} and dimensions compatible * with input0. * * Outputs: * * 0: A tensor of {@link %{OperandTypeLinkPfx}TENSOR_BOOL8}. %insert AVAIL3 */ %{DeclareOperation_1.2 LESS 58}, /** * For input tensors x and y, computes x <= y elementwise. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_BOOL8} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: from 1 * * This operation supports broadcasting. * * Inputs: * * 0: A tensor. * * 1: A tensor of the same {@link %{OperandType}} and dimensions compatible * with input0. * * Outputs: * * 0: A tensor of {@link %{OperandTypeLinkPfx}TENSOR_BOOL8}. %insert AVAIL3 */ %{DeclareOperation_1.2 LESS_EQUAL 59}, /** * Computes natural logarithm of x element-wise. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * Supported tensor rank: from 1. * * Inputs: * * 0: A tensor. * * Outputs: * * 0: The output tensor of same shape as input0. %insert AVAIL3 */ %{DeclareOperation_1.2 LOG 60}, /** * Returns the truth value of x AND y element-wise. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_BOOL8} * * Supported tensor rank: from 1 * * This operation supports broadcasting. * * Inputs: * * 0: A tensor of {@link %{OperandTypeLinkPfx}TENSOR_BOOL8}. * * 1: A tensor of {@link %{OperandTypeLinkPfx}TENSOR_BOOL8} and dimensions * compatible with input0. * * Outputs: * * 0: A tensor of {@link %{OperandTypeLinkPfx}TENSOR_BOOL8}. %insert AVAIL3 */ %{DeclareOperation_1.2 LOGICAL_AND 61}, /** * Computes the truth value of NOT x element-wise. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_BOOL8} * * Supported tensor rank: from 1. * * Inputs: * * 0: A tensor. * * Outputs: * * 0: The output tensor of same shape as input0. %insert AVAIL3 */ %{DeclareOperation_1.2 LOGICAL_NOT 62}, /** * Returns the truth value of x OR y element-wise. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_BOOL8} * * Supported tensor rank: from 1 * * This operation supports broadcasting. * * Inputs: * * 0: A tensor of {@link %{OperandTypeLinkPfx}TENSOR_BOOL8}. * * 1: A tensor of {@link %{OperandTypeLinkPfx}TENSOR_BOOL8} and dimensions * compatible with input0. * * Outputs: * * 0: A tensor of {@link %{OperandTypeLinkPfx}TENSOR_BOOL8}. %insert AVAIL3 */ %{DeclareOperation_1.2 LOGICAL_OR 63}, /** * Computes the log softmax activations given logits. * * The output is calculated using this formula: * * output = logits * beta - log(reduce_sum(exp(logits * beta), axis)) * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * Supported tensor rank: from 1. * * Inputs: * * 0: A tensor specifying the input logits. * * 1: A scalar, specifying the positive scaling factor for the exponent, * beta. * For input tensor of {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16}, the beta * value must be of {@link %{OperandTypeLinkPfx}FLOAT16}. * For input tensor of {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32}, the beta * value must be of {@link %{OperandTypeLinkPfx}FLOAT32}. * * 2: An {@link %{OperandTypeLinkPfx}INT32} scalar specifying the axis to * reduce across. Negative index is used to specify axis from the * end (e.g. -1 for the last axis). Must be in the range [-n, n). * * Outputs: * * 0: The output tensor of the same {@link %{OperandType}} and shape as * input0. %insert AVAIL3 */ %{DeclareOperation_1.2 LOG_SOFTMAX 64}, /** * Returns the element-wise maximum of two tensors. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: from 1. * * Inputs: * * 0: A tensor. * * 1: A tensor of the same {@link %{OperandType}} and compatible dimensions * with input0. * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scales and zeroPoint can be different from input0 scale and zeroPoint. * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} as input0. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint can be different from inputs' scale and zeroPoint. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint can be different from inputs' scale and zeroPoint. %/kind %insert AVAIL3 */ %{DeclareOperation_1.2 MAXIMUM 65}, /** * Returns the element-wise minimum of two tensors. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: from 1. * * Inputs: * * 0: A tensor. * * 1: A tensor of the same {@link %{OperandType}} and compatible dimensions * with input0. * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scales and zeroPoint can be different from input0 scale and zeroPoint. * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} as input0. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint can be different from inputs' scale and zeroPoint. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint can be different from inputs' scale and zeroPoint. %/kind %insert AVAIL3 */ %{DeclareOperation_1.2 MINIMUM 66}, /** * Computes numerical negative value element-wise. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * Supported tensor rank: from 1. * * Inputs: * * 0: A tensor. * * Outputs: * * 0: The output tensor of same shape as input0. %insert AVAIL3 */ %{DeclareOperation_1.2 NEG 67}, /** * For input tensors x and y, computes x != y elementwise. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_BOOL8} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: from 1 * * This operation supports broadcasting. * * Inputs: * * 0: A tensor. * * 1: A tensor of the same {@link %{OperandType}} and dimensions compatible * with input0. * * Outputs: * * 0: A tensor of {@link %{OperandTypeLinkPfx}TENSOR_BOOL8}. %insert AVAIL3 */ %{DeclareOperation_1.2 NOT_EQUAL 68}, /** * Pads a tensor with the given constant value according to the specified * paddings. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: up to 4 * * Inputs: * * 0: An n-D tensor, specifying the tensor to be padded. * * 1: A 2-D Tensor of {@link %{OperandTypeLinkPfx}TENSOR_INT32}, the paddings * for each spatial dimension of the input tensor. The shape of the * tensor must be {rank(input0), 2}. * padding[i, 0] specifies the number of elements to be padded in the * front of dimension i. * padding[i, 1] specifies the number of elements to be padded after * the end of dimension i. * * 2: A scalar specifying the value to use for padding input0. * For input tensor of {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16}, the * pad value must be of {@link %{OperandTypeLinkPfx}FLOAT16}. * For input tensor of {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32}, the * pad value must be of {@link %{OperandTypeLinkPfx}FLOAT32}. %kind aidl canonical ndk hal_1.3+ * For input tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED}, %else * For input tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, %/kind * the pad value must be of {@link %{OperandTypeLinkPfx}INT32}. The * scale and zeroPoint are assumed to be the same as in input0. * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} as input0. The * output tensor has the same rank as input0, and each * dimension of the output tensor has the same size as the * corresponding dimension of the input tensor plus the size * of the padding: * output0.dimension[i] = * padding[i, 0] + input0.dimension[i] + padding[i, 1] %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint must be the same as input0. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint must be the same as input0. %/kind %insert AVAIL3 */ %{DeclareOperation_1.2 PAD_V2 69}, /** * Computes the power of one value to another. * * Given a tensor base and a tensor exponent, this operation computes * base^exponent elementwise. * * This operations supports broadcasting. The size of the output is the * maximum size along each dimension of the input operands. It starts with * the trailing dimensions, and works its way forward. * * For example: * base.dimension = {4, 1, 2} * exponent.dimension = {5, 4, 3, 1} * output.dimension = {5, 4, 3, 2} * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * Supported tensor rank: from 1 * * Inputs: * * 0: A tensor specifying the base. * * 1: A tensor specifying the exponent. * * Outputs: * * 0: An output tensor. %insert AVAIL3 */ %{DeclareOperation_1.2 POW 70}, /** * Parametric Rectified Linear Unit. * * It follows: f(x) = alpha * x for x < 0, f(x) = x for x >= 0, where alpha * is a learned array with the same {@link %{OperandType}} and compatible * dimensions as input x. * * Two dimensions are compatible when: * 1. they are equal, or * 2. one of them is 1 * * The size of the output is the maximum size along each dimension of the * input operands. It starts with the trailing dimensions, and works its way * forward. * * Example: * input.dimension = {4, 1, 2} * alpha.dimension = {5, 4, 3, 1} * output.dimension = {5, 4, 3, 2} * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: from 1 * * Inputs: * * 0: A tensor, specifying the input. * * 1: A tensor of the same {@link %{OperandType}}, and compatible dimensions * as input0, specifying the alpha. * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} as input0. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scales and zeroPoint can be different from input0 scale and zeroPoint. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scales and zeroPoint can be different from input0 scale and zeroPoint. %/kind %insert AVAIL3 */ %{DeclareOperation_1.2 PRELU 71}, /** * Quantizes the input tensor. * * The formula for {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} output tensor is: * * output = max(0, min(255, round(input / scale) + zeroPoint) * %kind aidl canonical ndk hal_1.3+ * The formula for {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} output * tensor is: * * output = max(-128, min(127, round(input / scale) + zeroPoint) * %/kind * Supported input tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * Supported output tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: from 1 * * Inputs: * * 0: A tensor, may be zero-sized. * * Outputs: * * 0: The output tensor of same shape as input0, but with %kind aidl canonical ndk hal_1.3+ * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} or. * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED}. %else * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}. %/kind %insert AVAIL3 */ %{DeclareOperation_1.2 QUANTIZE 72}, /** * A version of quantized LSTM, using 16 bit quantization for internal * state. * * There is no projection layer, so cell state size is equal to the output * size. * * Inputs: * * 0: A 2-D tensor of type {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * and shape [numBatches, inputSize] specifying the input to the LSTM * cell. Tensor is quantized with a fixed quantization range of * [-1, 127/128] (scale = 1/128, zeroPoint = 128). * * 1: The input-to-input weights. * A 2-D tensor of type {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * and shape [outputSize, inputSize] specifying input-to-input part of * weights for fully-connected layer inside the LSTM cell. * Quantization zero point and scale must be the same across all the * weights. * * 2: The input-to-forget weights. * A 2-D tensor of type {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * and shape [outputSize, inputSize] specifying input-to-forget part of * weights for fully-connected layer inside the LSTM cell. * Quantization zero point and scale must be the same across all the * weights. * * 3: The input-to-cell weights. * A 2-D tensor of type {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * and shape [outputSize, inputSize] specifying input-to-cell part of * weights for fully-connected layer inside the LSTM cell. * Quantization zero point and scale must be the same across all the * weights. * * 4: The input-to-output weights. * A 2-D tensor of type {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * and shape [outputSize, inputSize] specifying input-to-output part of * weights for fully-connected layer inside the LSTM cell. * Quantization zero point and scale must be the same across all the * weights. * * 5: The recurrent-to-input weights. * A 2-D tensor of type {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * and shape [outputSize, outputSize] specifying recurrent-to-input part * of weights for fully-connected layer inside the LSTM cell. * Quantization zero point and scale must be the same across all the * weights. * * 6: The recurrent-to-forget weights. * A 2-D tensor of type {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * and shape [outputSize, outputSize] specifying recurrent-to-forget * part of weights for fully-connected layer inside the LSTM cell. * Quantization zero point and scale must be the same across all the * weights. * * 7: The recurrent-to-cell weights. * A 2-D tensor of type {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * and shape [outputSize, outputSize] specifying recurrent-to-cell part * of weights for fully-connected layer inside the LSTM cell. * Quantization zero point and scale must be the same across all the * weights. * * 8: The recurrent-to-output weights. * A 2-D tensor of type {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * and shape [outputSize, outputSize] specifying recurrent-to-output * part of weights for fully-connected layer inside the LSTM cell. * Quantization zero point and scale must be the same across all the * weights. * * 9: The input gate bias. * A 1-D tensor of type {@link %{OperandTypeLinkPfx}TENSOR_INT32} and shape * [outputSize] specifying the bias for the fully-connected layer * inside the LSTM cell. Bias is quantized with scale being a product * of input and weights scales and zeroPoint equal to 0. * * 10:The forget gate bias. * A 1-D tensor of type {@link %{OperandTypeLinkPfx}TENSOR_INT32} and shape * [outputSize] specifying the bias for the fully-connected layer * inside the LSTM cell. Bias is quantized with scale being a product * of input and weights scales and zeroPoint equal to 0. * * 11:The cell bias. * A 1-D tensor of type {@link %{OperandTypeLinkPfx}TENSOR_INT32} and shape * [outputSize] specifying the bias for the fully-connected layer * inside the LSTM cell. Bias is quantized with scale being a product * of input and weights scales and zeroPoint equal to 0. * * 12:The output gate bias. * A 1-D tensor of type {@link %{OperandTypeLinkPfx}TENSOR_INT32} and shape * [outputSize] specifying the bias for the fully-connected layer * inside the LSTM cell. Bias is quantized with scale being a product * of input and weights scales and zeroPoint equal to 0. * * 13: A 2-D tensor of type {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_SYMM} * and shape [numBatches, outputSize] specifying the cell state from the * previous time step of the LSTM cell. It is quantized using a * quantization range of [-2^4, 2^4 * 32767/32768] (scale = 2^4 / * 32768, zeroPoint = 0). * * 14: A 2-D tensor of type {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * and shape [numBathes, outputSize] specifying the output of the LSTM * cell from previous time-step. Tensor is quantized with a fixed * quantization range of [-1, 127/128] (scale = 1/128, zeroPoint = * 128). * * * Outputs: * * 0: A 2-D tensor of type {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_SYMM} * and shape [numBatches, outputSize] which contains a cell state from * the current time step. Tensor is quantized using a quantization * range of [-2^4, 2^4 * 32767/32768] (scale = 2^4 / 32768, zeroPoint = * 0). * * 1: A 2-D tensor of type {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * and shape [numBathes, outputSize] which contains the output value. * Tensor is quantized with a fixed quantization range of [-1, 127/128] * (scale = 1/128, zeroPoint = 128). */ %{DeclareOperation_1.2 QUANTIZED_16BIT_LSTM 73}, /** * Draws samples from a multinomial distribution. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * Inputs: * * 0: A 2-D tensor with shape [batches, classes], specifying the * unnormalized log-probabilities for all classes. * * 1: A scalar {@link %{OperandTypeLinkPfx}INT32}, specifying the number of * independent samples to draw for each row slice. * * 2: A 1-D {@link %{OperandTypeLinkPfx}TENSOR_INT32} tensor with shape [2], * specifying seeds used to initialize the random distribution. If both * provided seeds are 0, both will be randomly generated. * Outputs: * * 0: A 2-D {@link %{OperandTypeLinkPfx}TENSOR_INT32} tensor with shape * [batches, samples], containing the drawn samples. %insert AVAIL3 */ %{DeclareOperation_1.2 RANDOM_MULTINOMIAL 74}, /** * Reduces a tensor by computing the "logical and" of elements along given * dimensions. * * If keep_dims is true, the reduced dimensions are * retained with length 1. Otherwise, the rank of the tensor is reduced by * 1 for each entry in dimensions. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_BOOL8} * * Supported tensor rank: up to 4 * * Inputs: * * 0: An n-D tensor. * * 1: A 1-D tensor of {@link %{OperandTypeLinkPfx}TENSOR_INT32}. The dimensions * to reduce. Dimension values must be in the range [-n, n). * * 2: An {@link %{OperandTypeLinkPfx}BOOL} scalar, keep_dims. If true, * retains reduced dimensions with length 1. * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} as input0. * If all dimensions are reduced and keep_dims is false, the output * shape is [1]. %insert AVAIL3 */ %{DeclareOperation_1.2 REDUCE_ALL 75}, /** * Reduces a tensor by computing the "logical or" of elements along given * dimensions. * * If keep_dims is true, the reduced dimensions are * retained with length 1. Otherwise, the rank of the tensor is reduced by * 1 for each entry in dimensions. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_BOOL8} * * Supported tensor rank: up to 4 * * Inputs: * * 0: An n-D tensor. * * 1: A 1-D tensor of {@link %{OperandTypeLinkPfx}TENSOR_INT32}. The dimensions * to reduce. Dimension values must be in the range [-n, n). * * 2: An {@link %{OperandTypeLinkPfx}BOOL} scalar, keep_dims. If true, * retains reduced dimensions with length 1. * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} as input0. * If all dimensions are reduced and keep_dims is false, the output * shape is [1]. %insert AVAIL3 */ %{DeclareOperation_1.2 REDUCE_ANY 76}, /** * Reduces a tensor by computing the maximum of elements along given * dimensions. * * If keep_dims is true, the reduced dimensions are * retained with length 1. Otherwise, the rank of the tensor is reduced by * 1 for each entry in dimensions. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: up to 4 * * Inputs: * * 0: An n-D tensor. * * 1: A 1-D tensor of {@link %{OperandTypeLinkPfx}TENSOR_INT32}. The dimensions * to reduce. Dimension values must be in the range [-n, n). * * 2: An {@link %{OperandTypeLinkPfx}BOOL} scalar, keep_dims. If true, * retains reduced dimensions with length 1. * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} as input0. * If all dimensions are reduced and keep_dims is false, the output * shape is [1]. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint must be the same as input0. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint must be the same as input0. %/kind %insert AVAIL3 */ %{DeclareOperation_1.2 REDUCE_MAX 77}, /** * Reduces a tensor by computing the minimum of elements along given * dimensions. * * If keep_dims is true, the reduced dimensions are * retained with length 1. Otherwise, the rank of the tensor is reduced by * 1 for each entry in dimensions. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: up to 4 * * Inputs: * * 0: An n-D tensor. * * 1: A 1-D tensor of {@link %{OperandTypeLinkPfx}TENSOR_INT32}. The dimensions * to reduce. Dimension values must be in the range [-n, n). * * 2: An {@link %{OperandTypeLinkPfx}BOOL} scalar, keep_dims. If true, * retains reduced dimensions with length 1. * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} as input0. * If all dimensions are reduced and keep_dims is false, the output * shape is [1]. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint must be the same as input0. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint must be the same as input0. %/kind %insert AVAIL3 */ %{DeclareOperation_1.2 REDUCE_MIN 78}, /** * Reduces a tensor by multiplying elements along given dimensions. * * If keep_dims is true, the reduced dimensions are * retained with length 1. Otherwise, the rank of the tensor is reduced by * 1 for each entry in dimensions. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * Supported tensor rank: up to 4 * * Inputs: * * 0: An n-D tensor. * * 1: A 1-D tensor of {@link %{OperandTypeLinkPfx}TENSOR_INT32}. The dimensions * to reduce. Dimension values must be in the range [-n, n). * * 2: An {@link %{OperandTypeLinkPfx}BOOL} scalar, keep_dims. If true, * retains reduced dimensions with length 1. * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} as input0. * If all dimensions are reduced and keep_dims is false, the output * shape is [1]. %insert AVAIL3 */ %{DeclareOperation_1.2 REDUCE_PROD 79}, /** * Reduces a tensor by summing elements along given dimensions. * * If keep_dims is true, the reduced dimensions are * retained with length 1. Otherwise, the rank of the tensor is reduced by * 1 for each entry in dimensions. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * Supported tensor rank: up to 4 * * Inputs: * * 0: An n-D tensor. * * 1: A 1-D tensor of {@link %{OperandTypeLinkPfx}TENSOR_INT32}. The dimensions * to reduce. Dimension values must be in the range [-n, n). * * 2: An {@link %{OperandTypeLinkPfx}BOOL} scalar, keep_dims. If true, * retains reduced dimensions with length 1. * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} as input0. * If all dimensions are reduced and keep_dims is false, the output * shape is [1]. %insert AVAIL3 */ %{DeclareOperation_1.2 REDUCE_SUM 80}, /** * Select and scale the feature map of each region of interest to a unified * output size by average pooling sampling points from bilinear interpolation. * * The region of interest is represented by its upper-left corner coordinate * (x1,y1) and lower-right corner coordinate (x2,y2) in the original image. * A spatial scaling factor is applied to map into feature map coordinate. * A valid region of interest should satisfy x1 <= x2 and y1 <= y2. * * No rounding is applied in this operation. The sampling points are unified * distributed in the pooling bin and their values are calculated by bilinear * interpolation. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: 4, with "NHWC" or "NCHW" data layout. * With the default data layout NHWC, the data is stored in the order of: * [batch, height, width, channels]. Alternatively, the data layout could * be NCHW, the data storage order of: [batch, channels, height, width]. * * Inputs: * * 0: A 4-D tensor, specifying the feature map. * * 1: A 2-D Tensor of shape [num_rois, 4], specifying the locations of * the regions of interest, each line with format [x1, y1, x2, y2]. * For input0 of type {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, * this tensor should be of {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_ASYMM}, * with zeroPoint of 0 and scale of 0.125. Zero num_rois is * supported for this tensor. * * 2: An 1-D {@link %{OperandTypeLinkPfx}TENSOR_INT32} tensor, of shape * [num_rois], specifying the batch index of each box. Boxes with * the same batch index are grouped together. Zero num_rois is * supported for this tensor. * * 3: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the output * height of the output tensor. * * 4: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the output * width of the output tensor. * * 5: An {@link %{OperandTypeLinkPfx}FLOAT32} scalar, specifying the ratio * from the height of original image to the height of feature map. * * 6: An {@link %{OperandTypeLinkPfx}FLOAT32} scalar, specifying the ratio * from the width of original image to the width of feature map. * * 7: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the number of * sampling points in height dimension used to compute the output. * Set to 0 for adaptive value of ceil(roi_height/out_height). * * 8: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the number of * sampling points in width dimension used to compute the output. * Set to 0 for adaptive value of ceil(roi_width/out_width). * * 9: An {@link %{OperandTypeLinkPfx}BOOL} scalar, set to true to specify * NCHW data layout for input0 and output0. Set to false for NHWC. * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} as input0. The output * shape is [num_rois, out_height, out_width, depth]. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, %/kind * the scale and zeroPoint can be different from the input0 scale and zeroPoint. %insert AVAIL3 */ %{DeclareOperation_1.2 ROI_ALIGN 81}, /** * Select and scale the feature map of each region of interest to a unified * output size by max-pooling. * * The region of interest is represented by its upper-left corner coordinate * (x1,y1) and lower-right corner coordinate (x2,y2) in the original image. * A spatial scaling factor is applied to map into feature map coordinate. * A valid region of interest should satisfy x1 <= x2 and y1 <= y2. * * Rounding is applied in this operation to ensure integer boundary for * regions of interest and pooling bins. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: 4, with "NHWC" or "NCHW" data layout. * With the default data layout NHWC, the data is stored in the order of: * [batch, height, width, channels]. Alternatively, the data layout could * be NCHW, the data storage order of: [batch, channels, height, width]. * * Inputs: * * 0: A 4-D tensor, specifying the feature map. * * 1: A 2-D Tensor of shape [num_rois, 4], specifying the locations of * the regions of interest, each line with format [x1, y1, x2, y2]. %kind aidl canonical ndk hal_1.3+ * For input0 of type {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, %else * For input0 of type {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, %/kind * this tensor should be of {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_ASYMM}, * with zeroPoint of 0 and scale of 0.125. * * 2: An 1-D {@link %{OperandTypeLinkPfx}TENSOR_INT32} tensor, of shape * [num_rois], specifying the batch index of each box. Boxes with * the same batch index are grouped together. * * 3: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the output * height of the output tensor. * * 4: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the output * width of the output tensor. * * 5: An {@link %{OperandTypeLinkPfx}FLOAT32} scalar, specifying the ratio * from the height of original image to the height of feature map. * * 6: An {@link %{OperandTypeLinkPfx}FLOAT32} scalar, specifying the ratio * from the width of original image to the width of feature map. * * 7: An {@link %{OperandTypeLinkPfx}BOOL} scalar, set to true to specify * NCHW data layout for input0 and output0. Set to false for NHWC. * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} as input0. The output * shape is [num_rois, out_height, out_width, depth]. %kind aidl canonical ndk hal_1.3+ * For input0 of type {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, %/kind * the scale and zeroPoint must be the same as input0. %insert AVAIL3 */ %{DeclareOperation_1.2 ROI_POOLING 82}, /** * Computes reciprocal of square root of x element-wise. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} %kind aidl canonical ndk * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} (since %{NNAPILevel7}) * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel7}) %/kind * * Supported tensor rank: from 1. * * Inputs: * * 0: A tensor. * * Outputs: * * 0: The output tensor of same shape as input0. %kind aidl canonical ndk * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint can be different from inputs' scale and zeroPoint. %/kind %insert AVAIL3 */ %{DeclareOperation_1.2 RSQRT 83}, /** * Using a tensor of booleans c and input tensors x and y select values * elementwise from both input tensors: * * O[i] = C[i] ? x[i] : y[i]. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3 * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: from 1 * * Inputs: * * 0: A tensor of type {@link %{OperandTypeLinkPfx}TENSOR_BOOL8} acting as a * mask that chooses, based on the value at each element, whether the * corresponding element in the output should be taken from input1 (if * true) or input2 (if false). * * 1: An input tensor of the same shape as input0. %kind hal_1.2 * * 2: An input tensor of the same shape and type as input1. * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scales and zeroPoint can be different from input1 scale and zeroPoint. %else * * 2: An input tensor of the same shape and type as input1. * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * and {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scales and zeroPoint can be different from input1 scale and zeroPoint. %/kind * * Outputs: * * 0: A tensor of the same type and shape as input1 and input2. * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint can be different from inputs' scale and zeroPoint. %insert AVAIL3 */ %{DeclareOperation_1.2 SELECT 84}, /** * Computes sin of x element-wise. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * Supported tensor rank: from 1. * * Inputs: * * 0: A tensor. * * Outputs: * * 0: The output tensor of same shape as input0. %insert AVAIL3 */ %{DeclareOperation_1.2 SIN 85}, /** * Extracts a slice of specified size from the input tensor starting at a * specified location. * * The starting location is specified as a 1-D tensor containing offsets * for each dimension. The size is specified as a 1-D tensor containing * either size of a slice along corresponding dimension or -1. In the latter * case, all the remaining elements in dimension are included in the slice. * * A sum of begin offset and a size of a slice must not exceed size of a * corresponding dimension. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: from 1 * * Inputs: * * 0: An n-D tensor to take slice from, may be zero-sized. * * 1: A 1-D tensor of type {@link %{OperandTypeLinkPfx}TENSOR_INT32} specifying * the beginning indices of the slice in each dimension. * * 2: A 1-D tensor of type {@link %{OperandTypeLinkPfx}TENSOR_INT32} specifying * the size of the slice in each dimension. * * Outputs: * * 0: An n-D tensor of the same type as the input containing the slice. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, %/kind * its scale and zeroPoint has to be same as the input0 scale and zeroPoint. %insert AVAIL3 */ %{DeclareOperation_1.2 SLICE 86}, /** * Splits a tensor along a given axis into num_splits subtensors. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: from 1 * * Inputs: * * 0: An n-D tensor to split. * * 1: An {@link %{OperandTypeLinkPfx}INT32} scalar specifying the axis along * which to split. * * 2: An {@link %{OperandTypeLinkPfx}INT32} scalar indicating the number of * splits along given axis. Must evenly divide axis size. * * Outputs: * * 0 ~ (num_splits - 1): Resulting subtensors. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint must be the same as input0. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint must be the same as input0. %/kind %insert AVAIL3 */ %{DeclareOperation_1.2 SPLIT 87}, /** * Computes square root of x element-wise. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * Supported tensor rank: from 1. * * Inputs: * * 0: A tensor. * * Outputs: * * 0: The output tensor of same shape as input0. %insert AVAIL3 */ %{DeclareOperation_1.2 SQRT 88}, /** * Constructs a tensor by tiling a given tensor. * * This operation creates a new tensor by replicating `input` `multiples` * times. The output tensor's i-th dimension has `input.dims(i) * multiples[i]` * elements, and the values of `input` are replicated `multiples[i]` times * along the i-th dimension. * For example, tiling `[a b c d]` by `[2]` produces `[a b c d a b c d]`. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: from 1 * * Inputs: * * 0: input, an n-D tensor specifying the input. * * 1: multiples, a 1-D tensor of {@link %{OperandTypeLinkPfx}TENSOR_INT32}. * The length of multiples must be n. * * Outputs: * * 0: A tiled tensor of the same {@link %{OperandType}} and rank as `input`. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint must be the same as input0. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint must be the same as input0. %/kind %insert AVAIL3 */ %{DeclareOperation_1.2 TILE 89}, /** * Finds values and indices of the k largest entries for the last dimension. * * Resulting values in each dimensions are sorted in descending order. If * two values are equal, the one with larger index appears first. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind aidl canonical ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: from 1 * * Inputs: * * 0: input, an n-D tensor specifying the input. * * 1: k, an {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the number of * top elements to look for along the last dimension. * * Outputs: * * 0: An n-D tensor of the same type as the input, containing the k * largest elements along each last dimensional slice. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint must be the same as input0. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint must be the same as input0. %/kind * * 1: An n-D tensor of type {@link %{OperandTypeLinkPfx}TENSOR_INT32} * containing the indices of values within the last dimension of input. %insert AVAIL3 */ %{DeclareOperation_1.2 TOPK_V2 90}, /** * Performs the transpose of 2-D convolution operation. * * This operation is sometimes called "deconvolution" after Deconvolutional * Networks, but is actually the transpose (gradient) of * {@link %{OperandTypeLinkPfx}CONV_2D} rather than an actual deconvolution. * * The output dimensions are functions of the filter dimensions, stride, and * padding. * * Supported tensor {@link %{OperandType}} configurations: * * 16 bit floating point: * * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} for input, filter, output, and bias. * * * 32 bit floating point: * * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} for input, filter, output, and bias. * * * Quantized: * * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} for input, filter, and output. * * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} for bias (with scale set to * * * input.scale * filter.scale). * * * Quantized with symmetric per channel quantization for the filter: * * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} for input, and output. * * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM_PER_CHANNEL} for filter. * * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} for bias (scale set to 0.0, * * * each value scaling is separate and equal to input.scale * filter.scales[channel]). %kind aidl canonical ndk hal_1.3+ * * Available since %{NNAPILevel4}: * * Quantized signed (since %{NNAPILevel4}): * * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} for input, filter, and output. * * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} for bias (with scale set to * * * input.scale * filter.scale). * * * Quantized signed with filter symmetric per channel quantization * (since %{NNAPILevel4}): * * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} for input, and output. * * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM_PER_CHANNEL} for filter. * * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} for bias (scale set to 0.0, * * * each value scaling is separate and equal to input.scale * filter.scales[channel]). %/kind * * Supported tensor rank: 4, with "NHWC" or "NCHW" data layout. * With the default data layout NHWC, the data is stored in the order of: * [batch, height, width, channels]. Alternatively, the data layout could * be NCHW, the data storage order of: [batch, channels, height, width]. * * Both explicit padding and implicit padding are supported. * * Inputs (explicit padding): * * 0: A 4-D tensor, of shape [batches, height, width, depth_in], * specifying the input. %kind ndk * Since API level 29, zero batches is supported for this tensor. %/kind * * 1: A 4-D tensor, of shape * [depth_out, filter_height, filter_width, depth_in], specifying the * filter. For tensor of type * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM_PER_CHANNEL} the channel * dimension (%{Ann}SymmPerChannelQuantParams::channelDim) must be set to 0. * * 2: A 1-D tensor, of shape [depth_out], specifying the bias. For input * tensor of type {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} or * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16}, the bias must be of the * same type. %kind aidl canonical ndk hal_1.3+ * For filter tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * and {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED}, %else * For filter tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, %/kind * the bias should be of {@link %{OperandTypeLinkPfx}TENSOR_INT32}, * with zeroPoint of 0 and bias_scale == input_scale * filter_scale. * For filter tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM_PER_CHANNEL}, * the bias must be of {@link %{OperandTypeLinkPfx}TENSOR_INT32}, with zeroPoint of 0 * and bias_scale of 0. The actual scale of each value 'i' is equal to * bias_scale[i] = input_scale * filter_scale[i]. * * 3: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the left, in the ‘width’ dimension. * * 4: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the right, in the ‘width’ dimension. * * 5: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the top, in the ‘height’ dimension. * * 6: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the padding on * the bottom, in the ‘height’ dimension. * * 7: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘width’ dimension. * * 8: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘height’ dimension. * * 9: An {@link %{OperandTypeLinkPfx}INT32} scalar, and has to be one of the * {@link %{FusedActivationFunc}} values. Specifies the activation to * invoke on the result. * * 10: An {@link %{OperandTypeLinkPfx}BOOL} scalar, set to true to specify * NCHW data layout for input0 and output0. Set to false for NHWC. * * Inputs (implicit padding): * * 0: A 4-D tensor, of shape [batches, height, width, depth_in], * specifying the input. %kind ndk * Since API level 29, zero batches is supported for this tensor. %/kind * * 1: A 4-D tensor, of shape * [depth_out, filter_height, filter_width, depth_in], specifying the * filter. For tensor of type * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM_PER_CHANNEL} the channel * dimension (%{Ann}SymmPerChannelQuantParams::channelDim) must be set to 0. * * 2: A 1-D tensor, of shape [depth_out], specifying the bias. For input * tensor of type {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} or * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16}, the bias should be of the * same type. %kind aidl canonical ndk hal_1.3+ * For filter tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * and {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED}, %else * For filter tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM}, %/kind * the bias should be of {@link %{OperandTypeLinkPfx}TENSOR_INT32}, * with zeroPoint of 0 and bias_scale == input_scale * filter_scale. * For filter tensor of {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM_PER_CHANNEL}, * the bias must be of {@link %{OperandTypeLinkPfx}TENSOR_INT32}, with zeroPoint of 0 * and bias_scale of 0. The actual scale of each value 'i' is equal to * bias_scale[i] = input_scale * filter_scale[i]. * * 3: An {@link %{OperandTypeLinkPfx}TENSOR_INT32} tensor, specifying the output * tensor shape. * * 4: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the implicit * padding scheme, has to be one of the %insert PaddingCodeValues * * 5: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘width’ dimension. * * 6: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the stride when * walking through input in the ‘height’ dimension. * * 7: An {@link %{OperandTypeLinkPfx}INT32} scalar, and has to be one of the * {@link %{FusedActivationFunc}} values. Specifies the activation to * invoke on the result. * * 8: An {@link %{OperandTypeLinkPfx}BOOL} scalar, set to true to specify * NCHW data layout for input0 and output0. Set to false for NHWC. * * Outputs: * * 0: The output 4-D tensor, of shape * [batches, out_height, out_width, depth_out]. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, %/kind * the scale and zeroPoint can be different from inputs' scale and zeroPoint. %insert AVAIL3 */ %{DeclareOperation_1.2 TRANSPOSE_CONV_2D 91}, /** * A recurrent neural network specified by an LSTM cell. * * Performs (fully) dynamic unrolling of input. * * This Op unrolls the input along the time dimension, and implements the * following operation for each element in the sequence * s = 1...sequence_length: * outputs[s] = projection(state = activation(LSTMOp(inputs[s]))) * * Where LSTMOp is the LSTM op as in {@link %{OperandTypeLinkPfx}LSTM}, * the "projection" is an optional projection layer from state and output * and the “activation” is the function passed as the * “fused_activation_function” argument (if not “NONE”). * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * Supported tensor rank: 3, either time-major or batch-major. * * All input and output tensors must be of the same type. * * Inputs: * * 0: The input (\f$x_t\f$). * A 3-D tensor of shape: * If time-major: [max_time, batch_size, input_size] * If batch-major: [batch_size, max_time, input_size] * where “max_time” is the number of timesteps (sequence length), * “batch_size” corresponds to the batching dimension, and * “input_size” is the size of the input. * * 1: The input-to-input weights (\f$W_{xi}\f$). Optional. * A 2-D tensor of shape [num_units, input_size], where “num_units” * corresponds to the number of cell units. * * 2: The input-to-forget weights (\f$W_{xf}\f$). * A 2-D tensor of shape [num_units, input_size]. * * 3: The input-to-cell weights (\f$W_{xc}\f$). * A 2-D tensor of shape [num_units, input_size]. * * 4: The input-to-output weights (\f$W_{xo}\f$). * A 2-D tensor of shape [num_units, input_size]. * * 5: The recurrent-to-input weights (\f$W_{hi}\f$). Optional. * A 2-D tensor of shape [num_units, output_size], where “output_size” * corresponds to either the number of cell units (i.e., “num_units”), * or the second dimension of the “projection_weights”, if defined. * * 6: The recurrent-to-forget weights (\f$W_{hf}\f$). * A 2-D tensor of shape [num_units, output_size]. * * 7: The recurrent-to-cell weights (\f$W_{hc}\f$). * A 2-D tensor of shape [num_units, output_size]. * * 8: The recurrent-to-output weights (\f$W_{ho}\f$). * A 2-D tensor of shape [num_units, output_size]. * * 9: The cell-to-input weights (\f$W_{ci}\f$). Optional. * A 1-D tensor of shape [num_units]. * * 10:The cell-to-forget weights (\f$W_{cf}\f$). Optional. * A 1-D tensor of shape [num_units]. * * 11:The cell-to-output weights (\f$W_{co}\f$). Optional. * A 1-D tensor of shape [num_units]. * * 12:The input gate bias (\f$b_i\f$). Optional. * A 1-D tensor of shape [num_units]. * * 13:The forget gate bias (\f$b_f\f$). * A 1-D tensor of shape [num_units]. * * 14:The cell bias (\f$b_c\f$). * A 1-D tensor of shape [num_units]. * * 15:The output gate bias (\f$b_o\f$). * A 1-D tensor of shape [num_units]. * * 16:The projection weights (\f$W_{proj}\f$). Optional. * A 2-D tensor of shape [output_size, num_units]. * * 17:The projection bias (\f$b_{proj}\f$). Optional. * A 1-D tensor of shape [output_size]. * * 18:The output state (in) (\f$h_{t-1}\f$). * A 2-D tensor of shape [batch_size, output_size]. * * 19:The cell state (in) (\f$C_{t-1}\f$). * A 2-D tensor of shape [batch_size, num_units]. * * 20:The activation function (\f$g\f$). * A value indicating the activation function: * * * 21:The clipping threshold (\f$t_{cell}\f$) for the cell state, such * that values are bound within [-cell_clip, cell_clip]. If set to 0.0 * then clipping is disabled. * * 22:The clipping threshold (\f$t_{proj}\f$) for the output from the * projection layer, such that values are bound within * [-proj_clip, proj_clip]. If set to 0.0 then clipping is disabled. * * 23:Time-major if true, batch-major if false. * * 24:The input layer normalization weights. Optional. * A 1-D tensor of shape [num_units]. Used to rescale normalized inputs * to activation at input gate. * * 25:The forget layer normalization weights. Optional. * A 1-D tensor of shape [num_units]. Used to rescale normalized inputs * to activation at forget gate. * * 26:The cell layer normalization weights. Optional. * A 1-D tensor of shape [num_units]. Used to rescale normalized inputs * to activation at cell gate. * * 27:The output layer normalization weights. Optional. * A 1-D tensor of shape [num_units]. Used to rescale normalized inputs * to activation at output gate. * * Outputs: * * 0: The output (\f$o_t\f$). * A 3-D tensor of shape: * If time-major: [max_time, batch_size, output_size] * If batch-major: [batch_size, max_time, output_size] %kind aidl canonical ndk hal_1.3+ * * 1: A tensor of shape [batch_size, output_size] containing a hidden * state from the last time step in the sequence. This output is * optional and can be omitted. If this output is present then * output #2 must be present as well. * Available since %{NNAPILevel4}. * * 2: A tensor of shape [batch_size, cell_size] containing a cell state * from the last time step in the sequence. This output is optional * and can be omitted. * Available since %{NNAPILevel4}. %/kind %insert AVAIL3 %insert OutputState */ %{DeclareOperation_1.2 UNIDIRECTIONAL_SEQUENCE_LSTM 92}, /** * A recurrent neural network layer that applies a basic RNN cell to a * sequence of inputs. * * This layer unrolls the input along the sequence dimension, and implements * the following operation * for each element in the sequence s = 1...sequence_length: * outputs[s] = state = activation(inputs[s] * input_weights’ + state * * recurrent_weights’ + bias) * * Where: * * “input_weights” is a weight matrix that multiplies the inputs; * * “recurrent_weights” is a weight matrix that multiplies the current * “state” which itself is the output from the previous time step * computation; * * “bias” is a bias vector (added to each output vector in the batch); * * “activation” is the function passed as the “fused_activation_function” * argument (if not “NONE”). * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * The input tensors must all be the same type. * * Inputs: * * 0: input. * A 3-D tensor. The shape is defined by the input 6 (timeMajor). If * it is set to 1, then the input has a shape [maxTime, batchSize, * inputSize], otherwise the input has a shape [batchSize, maxTime, * inputSize]. * * 1: weights. * A 2-D tensor of shape [numUnits, inputSize]. * * 2: recurrent_weights. * A 2-D tensor of shape [numUnits, numUnits]. * * 3: bias. * A 1-D tensor of shape [numUnits]. * * 4: hidden state * A 2-D tensor of shape [batchSize, numUnits]. Specifies a hidden * state input for the first time step of the computation. * * 5: fusedActivationFunction. * A {@link %{FusedActivationFunc}} value indicating the activation function. If * “NONE” is specified then it results in a linear activation. * * 6: timeMajor * An {@link %{OperandTypeLinkPfx}INT32} scalar specifying the shape format * of input and output tensors. Must be set to either 0 or 1. * Outputs: * * 0: output. * A 3-D tensor. The shape is defined by the input 6 (timeMajor). If * it is set to 1, then the output has a shape [maxTime, batchSize, * numUnits], otherwise the output has a shape [batchSize, maxTime, * numUnits]. %kind ndk hal_1.3+ * * 1: A tensor of shape [batchSize, numUnits] containing hidden state * from the last time step in the sequence. This output is optional * and can be omitted. * Available since %{NNAPILevel4}. %/kind %insert AVAIL3 %insert OutputState */ %{DeclareOperation_1.2 UNIDIRECTIONAL_SEQUENCE_RNN 93}, /** * Resizes images to given size using the nearest neighbor interpretation. * * Resized images must be distorted if their output aspect ratio is not the * same as input aspect ratio. The corner pixels of output may not be the * same as corner pixels of input. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} %kind ndk hal_1.3+ * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} (since %{NNAPILevel4}) %/kind * * Supported tensor rank: 4, with "NHWC" or "NCHW" data layout. * With the default data layout NHWC, the data is stored in the order of: * [batch, height, width, channels]. Alternatively, the data layout could * be NCHW, the data storage order of: [batch, channels, height, width]. * * Both resizing by shape and resizing by scale are supported. * * Inputs (resizing by shape): * * 0: A 4-D tensor, of shape [batches, height, width, depth], specifying * the input. Zero batches is supported for this tensor. * * 1: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the output * width of the output tensor. * * 2: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the output * height of the output tensor. * * 3: An {@link %{OperandTypeLinkPfx}BOOL} scalar, default to false. * Set to true to specify NCHW data layout for input0 and output0. %kind aidl canonical ndk hal_1.3+ * * 4: Align corners. An optional {@link %{OperandTypeLinkPfx}BOOL} * scalar, default to false. If True, the centers of the 4 corner * pixels of the input and output tensors are aligned, preserving the * values at the corner pixels. * Available since %{NNAPILevel4}. * * 5: Half pixel centers. An optional {@link %{OperandTypeLinkPfx}BOOL} * scalar, default to false. If True, the pixel centers are assumed to * be at (0.5, 0.5). This is the default behavior of image.resize in * TF 2.0. If this parameter is True, then align_corners parameter * must be False. * Available since %{NNAPILevel4}. %/kind * * Inputs (resizing by scale): * * 0: A 4-D tensor, of shape [batches, height, width, depth], specifying * the input. Zero batches is supported for this tensor. * * 1: A scalar, specifying width_scale, the scaling factor of the width * dimension from the input tensor to the output tensor. The output * width is calculated as new_width = floor(width * width_scale). * The scalar must be of {@link %{OperandTypeLinkPfx}FLOAT16} if input0 is * of {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} and of * {@link %{OperandTypeLinkPfx}FLOAT32} otherwise. * * 2: A scalar, specifying height_scale, the scaling factor of the height * dimension from the input tensor to the output tensor. The output * height is calculated as new_height = floor(height * height_scale). * The scalar must be of {@link %{OperandTypeLinkPfx}FLOAT16} if input0 is * of {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} and of * {@link %{OperandTypeLinkPfx}FLOAT32} otherwise. * * 3: An {@link %{OperandTypeLinkPfx}BOOL} scalar, default to false. * Set to true to specify NCHW data layout for input0 and output0. %kind aidl canonical ndk hal_1.3+ * * 4: Align corners. An optional {@link %{OperandTypeLinkPfx}BOOL} * scalar, default to false. If True, the centers of the 4 corner * pixels of the input and output tensors are aligned, preserving the * values at the corner pixels. * Available since %{NNAPILevel4}. * * 5: Half pixel centers. An optional {@link %{OperandTypeLinkPfx}BOOL} * scalar, default to false. If True, the pixel centers are assumed to * be at (0.5, 0.5). This is the default behavior of image.resize in * TF 2.0. If this parameter is True, then align_corners parameter * must be False. * Available since %{NNAPILevel4}. %/kind * * Outputs: * * 0: The output 4-D tensor, of shape * [batches, new_height, new_width, depth]. %kind aidl canonical ndk hal_1.3+ * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint must be the same as input0. %else * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} tensor, * the scale and zeroPoint must be the same as input0. %/kind %insert AVAIL3 */ %{DeclareOperation_1.2 RESIZE_NEAREST_NEIGHBOR 94}, %/section %section Operation_1.2_MAX FUNDAMENTAL_MAX = 94, %/section %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% HAL OperandType for 1.3 %% NDK OperandCode for API 30 %section Operand_1.3 /** * A tensor of 8 bit signed integers that represent real numbers. * * Attached to this tensor are two numbers that can be used to convert the * 8 bit integer to the real value and vice versa. These two numbers are: * - scale: a 32 bit floating point value greater than zero. * - zeroPoint: a 32 bit integer, in range [-128, 127]. * * The formula is: * real_value = (integer_value - zeroPoint) * scale. %insert AVAIL4 */ %{ANN}TENSOR_QUANT8_ASYMM_SIGNED = 14, %insert canonical_empty_line /** * A reference to a %{model_or_subgraph}. %kind ndk * * {@link ANeuralNetworksModel_setOperandValueFromModel} must be used to set * the value for an Operand of this type. %/kind %kind aidl canonical hal* * * Must have the lifetime {@link %{OperandLifeTime}::SUBGRAPH}. %/kind %insert AVAIL4 */ %{ANN}%{MODEL_or_SUBGRAPH} = 15, %/section %section Operand_1.3_MAX FUNDAMENTAL_MAX = 15, %/section %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% HAL OperationType for 1.3 %% NDK OperationCode for API 30 %section Operation_1.3 /** * Quantized version of {@link %{OperationTypeLinkPfx}LSTM}. * * The input and the output use asymmetric quantized types, while the rest * use symmetric ones. * * Inputs: * * 0: The input to the LSTM cell. * Type: {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} * Shape: [batchSize, inputSize] * * 1: The input-to-input weights. Optional. * Type: {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM} * Shape: [numUnits, inputSize] * * 2: The input-to-forget weights. * Type: {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM} * Shape: [numUnits, inputSize] * * 3: The input-to-cell weights. * Type: {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM} * Shape: [numUnits, inputSize] * * 4: The input-to-output weights. * Type: {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM} * Shape: [numUnits, inputSize] * * 5: The recurrent-to-input weights. Optional. * Type: {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM} * Shape: [numUnits, outputSize] * * 6: The recurrent-to-forget weights. * Type: {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM} * Shape: [numUnits, outputSize] * * 7: The recurrent-to-cell weights. * Type: {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM} * Shape: [numUnits, outputSize] * * 8: The recurrent-to-output weights. * Type: {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM} * Shape: [numUnits, outputSize] * * 9: The cell-to-input weights (for peephole). Optional. * Type: {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_SYMM} * Shape: [numUnits] * * 10: The cell-to-forget weights (for peephole). Optional. * Type: {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_SYMM} * Shape: [numUnits] * * 11: The cell-to-output weights (for peephole). Optional. * Type: {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_SYMM} * Shape: [numUnits] * * 12: The input gate bias. Quantized with scale being the * product of input and weights scales and zeroPoint equal to 0. * Optional. * Type: {@link %{OperandTypeLinkPfx}TENSOR_INT32} * Shape: [numUnits] * * 13: The forget gate bias. Quantized with scale being the * product of input and weights scales and zeroPoint equal to 0. * Type: {@link %{OperandTypeLinkPfx}TENSOR_INT32} * Shape: [numUnits] * * 14: The cell bias. Quantized with scale being the * product of input and weights scales and zeroPoint equal to 0. * Type: {@link %{OperandTypeLinkPfx}TENSOR_INT32} * Shape: [numUnits] * * 15: The output gate bias. Quantized with scale being the * product of input and weights scales and zeroPoint equal to 0. * Type: {@link %{OperandTypeLinkPfx}TENSOR_INT32} * Shape: [numUnits] * * 16: The projection weights. Optional. * Type: {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM} * Shape: [outputSize, numUnits] * * 17: The projection bias. Quantized with scale being the * product of input and weights scales and zeroPoint equal to 0. * Optional. * Type: {@link %{OperandTypeLinkPfx}TENSOR_INT32} * Shape: [outputSize] * * 18: The output from the previous time step. * Type: {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} * Shape: [batchSize, outputSize] * * 19: The cell state from the previous time step. * Type: {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_SYMM} * Shape: [batchSize, numUnits] * * 20: The input layer normalization weights. Used to rescale * normalized inputs to activation at input gate. Optional. * Type: {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_SYMM} * Shape: [numUnits] * * 21: The forget layer normalization weights. Used to * rescale normalized inputs to activation at forget gate. Optional. * Type: {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_SYMM} * Shape: [numUnits] * * 22: The cell layer normalization weights. Used to rescale * normalized inputs to activation at cell gate. Optional. * Type: {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_SYMM} * Shape: [numUnits] * * 23: The output layer normalization weights. Used to * rescale normalized inputs to activation at output gate. Optional. * Type: {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_SYMM} * Shape: [numUnits] * * 24: The cell clip. If provided the cell state is clipped * by this value prior to the cell output activation. Optional. * Type: {@link %{OperandTypeLinkPfx}FLOAT32}. * * 25: The projection clip. If provided and projection is enabled, * this is used for clipping the projected values. Optional. * Type: {@link %{OperandTypeLinkPfx}FLOAT32}. * * 26: The scale of the intermediate result of matmul, * i.e. input to layer normalization, at input gate. * Type: {@link %{OperandTypeLinkPfx}FLOAT32}. * * 27: The scale of the intermediate result of matmul, * i.e. input to layer normalization, at forget gate. * Type: {@link %{OperandTypeLinkPfx}FLOAT32}. * * 28: The scale of the intermediate result of matmul, * i.e. input to layer normalization, at cell gate. * Type: {@link %{OperandTypeLinkPfx}FLOAT32}. * * 29: The scale of the intermediate result of matmul, * i.e. input to layer normalization, at output gate. * Type: {@link %{OperandTypeLinkPfx}FLOAT32}. * * 30: The zero point of the hidden state, i.e. input to * projection. * Type: {@link %{OperandTypeLinkPfx}INT32}. * * 31: The scale of the hidden state, i.e. input to * projection. * Type: {@link %{OperandTypeLinkPfx}FLOAT32}. * * Outputs: * * 0: The output state (out). * Type: {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} * Shape: [batchSize, outputSize] * * 1: The cell state (out). * Type: {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_SYMM} * Shape: [batchSize, numUnits] * * 2: The output. This is effectively the same as the current * "output state (out)" value. * Type: {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} * Shape: [batchSize, outputSize] %insert AVAIL4 */ %{DeclareOperation_1.3 QUANTIZED_LSTM 95}, /** * Executes one of the two referenced %{model_or_subgraph}s as determined by a boolean * value. * * The inputs and outputs of the two referenced %{model_or_subgraph}s must agree with the * signature of this operation. That is, if the operation has (3 + n) inputs * and m outputs, both %{model_or_subgraph}s must have n inputs and m outputs with the same * types, ranks%{NDK_if_specified}, dimensions%{NDK_if_specified}, scales, * zeroPoints, and %{otherOperandParameters} as the corresponding operation * inputs and outputs. %kind aidl canonical hal* * All of the operands mentioned must have fully specified dimensions. %/kind * * Inputs: * * 0: A value of type {@link %{OperandTypeLinkPfx}TENSOR_BOOL8} and shape [1] * that determines which of the two referenced %{model_or_subgraph}s to execute. * The operand must have fully specified dimensions. * * 1: A {@link %{OperandTypeLinkPfx}%{MODEL_or_SUBGRAPH}} reference to the %{model_or_subgraph} to be * executed if the condition is true. * * 2: A {@link %{OperandTypeLinkPfx}%{MODEL_or_SUBGRAPH}} reference to the %{model_or_subgraph} to be * executed if the condition is false. * * 3 ~ (n + 2): Inputs to be passed to the %{model_or_subgraph} selected for execution. * * Outputs: * * 0 ~ (m - 1): Outputs produced by the selected %{model_or_subgraph}. %insert AVAIL4 */ %{DeclareOperation_1.3 IF 96}, /** * Executes the body %{model_or_subgraph} until the condition %{model_or_subgraph} outputs false. * * The inputs to this operation are the condition %{model_or_subgraph}, the body %{model_or_subgraph}, * and operand values for the first iteration of the loop. The values are * implicitly split into three groups of input-output, state-only, and * input-only values, as described below. * * The outputs of this operation are the final values of input-output * operands. * * Both the condition and body %{model_or_subgraph} receive (m + k + n) inputs. * * The first m (m >= 1) inputs are input-output operands. For the first * iteration, these are initialized from the corresponding inputs of the * WHILE operation. In subsequent iterations, their values come from the * corresponding outputs of the body %{model_or_subgraph} produced during the previous * iteration. * * The next k (k >= 0) inputs are state-only operands. They are similar to * the input-output operands, except that their values are no longer * available after the loop terminates. * * The last n (n >= 0) inputs are input-only operands. Their values come * from the corresponding inputs of the WHILE operation. * * The body %{model_or_subgraph} produces (m + k) outputs. * * The first m outputs are input-output operands. They become the outputs * of the WHILE operation when a termination condition is reached. * * The last k outputs are state-only operands. Their values are no longer * available after the loop terminates. * * The numbers m, k, and n are inferred by the %{runtime_or_driver} as follows: * m = (WHILE operation output count) * k = (body %{model_or_subgraph} output count) - m * n = (body %{model_or_subgraph} input count) - m - k * * The pseudo-code below illustrates the flow of a WHILE operation with * inputs condition, body, initial_input_output, initial_state, input_only * (m = 1, k = 1, n = 1): * * input_output = initial_input_output * state = initial_state * while condition(input_output, state, input_only): * input_output, state = body(input_output, state, input_only) * return input_output * %kind ndk * To prevent infinite loops, there is an implicit execution timeout * associated with each loop ("loop timeout duration"). See {@link * ANeuralNetworksExecution_setLoopTimeout}. * %/kind * Inputs: * * 0: A {@link %{OperandTypeLinkPfx}%{MODEL_or_SUBGRAPH}} reference to the condition * %{model_or_subgraph}. The %{model_or_subgraph} must have (m + k + n) inputs with * the same types, ranks%{NDK_if_specified}, dimensions%{NDK_if_specified}, * scales, zeroPoints, and %{otherOperandParameters} as the * corresponding inputs of the WHILE operation and exactly one output * of {@link %{OperandTypeLinkPfx}TENSOR_BOOL8} and shape [1]. %kind ndk * The output operand must have fully specified dimensions. %/kind %kind aidl canonical hal* * All of the operands mentioned must have fully specified dimensions. %/kind * * 1: A {@link %{OperandTypeLinkPfx}%{MODEL_or_SUBGRAPH}} reference to the body %{model_or_subgraph}. * The %{model_or_subgraph} must have (m + k + n) inputs and (m + k) outputs with * the same types, ranks%{NDK_if_specified}, dimensions%{NDK_if_specified}, * scales, zeroPoints, and %{otherOperandParameters} as the * corresponding inputs and outputs of the WHILE operation. %kind aidl canonical hal* * All of the operands mentioned must have fully specified dimensions. %/kind * * (m inputs): Initial values for input-output operands. * * (k inputs): Initial values for state-only operands. * * (n inputs): Values for input-only operands. * * Outputs: * * 0 ~ (m - 1): Outputs produced by the loop. %insert AVAIL4 */ %{DeclareOperation_1.3 WHILE 97}, /** * Computes exponential linear activation on the input tensor element-wise. * * The output is calculated using the following formula: * * ELU(x) = max(0, x) + min(0, alpha * (exp(x) - 1)) * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * Supported tensor rank: from 1. * * Inputs: * * 0: A tensor, specifying the input. May be zero-sized. * * 1: A scalar, specifying the alpha parameter. * For input tensor of {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16}, * the alpha value must be of {@link %{OperandTypeLinkPfx}FLOAT16}. * For input tensor of {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32}, * the alpha value must be of {@link %{OperandTypeLinkPfx}FLOAT32}. * * Outputs: * * 0: The output tensor of same shape and type as input0. %insert AVAIL4 */ %{DeclareOperation_1.3 ELU 98}, /** * Computes hard-swish activation on the input tensor element-wise. * * Hard swish activation is introduced in * https://arxiv.org/pdf/1905.02244.pdf * * The output is calculated using the following formula: * * h-swish(x) = x * max(0, min(6, (x + 3))) / 6 * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} * * Supported tensor rank: from 1. * * Inputs: * * 0: A tensor, specifying the input. May be zero-sized. * * Outputs: * * 0: The output tensor of same shape and type as input0. * Scale and zero point of this tensor may be different from the input * tensor's parameters. %insert AVAIL4 */ %{DeclareOperation_1.3 HARD_SWISH 99}, /** * Creates a tensor filled with a scalar value. * * Supported output tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * Supported tensor rank: from 1. * * Inputs: * * 0: A 1-D tensor, specifying the desired output tensor shape. * * 1: A scalar, specifying the value to fill the output tensors with. * For output tensor of {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16}, * the scalar must be of {@link %{OperandTypeLinkPfx}FLOAT16}. * For output tensor of {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32}, * the scalar must be of {@link %{OperandTypeLinkPfx}FLOAT32}. * For output tensor of {@link %{OperandTypeLinkPfx}TENSOR_INT32}, * the scalar must be of {@link %{OperandTypeLinkPfx}INT32}. * * Outputs: * * 0: The output tensor. %insert AVAIL4 */ %{DeclareOperation_1.3 FILL 100}, /** * Returns the rank of a tensor. * * The rank of a tensor is the number of dimensions in it. Also known as * "order", "degree", "ndims". * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_SYMM} * * {@link %{OperandTypeLinkPfx}TENSOR_BOOL8} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM_PER_CHANNEL} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT16_ASYMM} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_SYMM} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} * * Supported tensor rank: from 1. * * Inputs: * * 0: The input tensor. * * Outputs: * * 0: A scalar of {@link %{OperandTypeLinkPfx}INT32}, specifying the rank * of the input tensor. %insert AVAIL4 */ %{DeclareOperation_1.3 RANK 101}, %/section %section Operation_1.3_MAX FUNDAMENTAL_MAX = 101, %/section %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% NDK OperationCode and HAL OperationType for Feature Level 6 %section Operation_fl6 /** * Performs multiplication of two tensors in batches. * * Multiplies all slices of two input tensors and arranges the individual * results in a single output tensor of the same batch size. Each pair of * slices in the same batch have identical {@link %{OperandType}}. Each * slice can optionally be adjointed (transpose and conjugate) before * multiplication. * * The two input tensors and the output tensor must be 2-D or higher and * have the same batch size. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * Supported tensor rank: at least 2 and up to 4 * * Inputs: * * 0: A tensor with 2-D or higher shape [..., r_x, c_x]. * * 1: A tensor with 2-D or higher shape [..., r_y, c_y]. It has the same * {@link %{OperandType}} and batch size as input0. * * 2: An optional {@link %{OperandTypeLinkPfx}BOOL} scalar adj_x, default * to false. Set to true to adjoint the slices of input0. * * 3: An optional {@link %{OperandTypeLinkPfx}BOOL} scalar adj_y, default * to false. Set to true to adjoint the slices of input1. * * Outputs: * * 0: A tensor with 2-D or higher shape [..., r_o, c_o], where * r_o = c_x if adj_x else r_x * c_o = r_y if adj_y else c_y %insert AVAIL6 */ %{DeclareOperation_fl6 BATCH_MATMUL 102}, /** * Packs N input tensors (N >= 1) of rank R into one output tensor of rank R+1. * The tensors are packed along a given axis. * * The input tensors must have identical {@link %{OperandType}} and dimensions. * * For example, suppose there are N input tensors of shape (A, B, C). * If axis is 0, the output tensor will have shape (N, A, B, C). * If axis is 1, the output tensor will have shape (A, N, B, C). * * All dimensions through the axis dimension determine the output tile count; * the remaining dimensions determine the tile shape. * * Return to the example of N input tensors of shape (A, B, C). * If axis is 0, there are N tiles in the output, each of shape (A, B, C). * If axis is 1, there are A*N tiles in the output, each of shape (B, C). * * The coordinates of a tile within the output tensor are (t[0],...,t[axis]). * The coordinates of a tile within an input tensor are (t[0],...,t[axis-1]). * (If axis is 0, an input tensor consists of a single tile.) * If we index input tensors starting with 0 (rather than by operand number), * then output_tile[t[0],...,t[axis]] = input_tile[t[axis]][t[0],...,t[axis-1]]. * That is, all output tile coordinates except for the axis coordinate select * the corresponding location within some input tensor; and the axis coordinate * selects the input tensor. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * Supported input tensor rank: from 1 * * Inputs: * * 0: A scalar of type {@link %{OperandTypeLinkPfx}INT32}, specifying * the axis along which to pack. The valid range is [0, R+1). * * 1 ~ N: Input tensors to be packed together. * For {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensors, * the scales and zeroPoint must be the same for all input tensors, * and will be the same for the output tensor. * * Outputs: * * 0: The packed tensor. %insert AVAIL6 */ %{DeclareOperation_fl6 PACK 103}, %/section %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% NDK OperationCode and HAL OperationType for Feature Level 7 %section Operation_fl7 /** * Pads a tensor with mirrored values. * * This operator specifies one of two padding modes: REFLECT or SYMMETRIC. * In the case of REFLECT mode, the mirroring excludes the border element * on the padding side. * In the case of SYMMETRIC mode, the mirroring includes the border element * on the padding side. * * For example, if the input is the 1-D tensor `[1, 2, 3]` and the padding * is `[0, 2]` (i.e., pad no elements before the first (and only) dimension, * and two elements after the first (and only) dimension), then: * - REFLECT mode produces the output `[1, 2, 3, 2, 1]` * - SYMMETRIC mode produces the output `[1, 2, 3, 3, 2]` * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * Supported tensor rank: from 1. * * Inputs: * * 0: An n-D tensor, specifying the tensor to be padded. * * 1: A 2-D tensor of {@link %{OperandTypeLinkPfx}TENSOR_INT32}, the paddings * for each spatial dimension of the input tensor. The shape of the * tensor must be {rank(input0), 2}. * padding[i, 0] specifies the number of elements to be padded in the * front of dimension i. * padding[i, 1] specifies the number of elements to be padded after the * end of dimension i. * Each padding value must be nonnegative. * In the case of REFLECT mode, each padding value must be less than the * corresponding dimension. * In the case of SYMMETRIC mode, each padding value must be less than or * equal to the corresponding dimension. * * 2: An {@link %{OperandTypeLinkPfx}INT32} scalar, specifying the mode. * Options are 0:REFLECT and 1:SYMMETRIC. * * Outputs: * * 0: A tensor of the same {@link %{OperandType}} as input0. The * output tensor has the same rank as input0, and each * dimension of the output tensor has the same size as the * corresponding dimension of the input tensor plus the size * of the padding: * output0.dimension[i] = * padding[i, 0] + input0.dimension[i] + padding[i, 1] * For a {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensor, * the scale and zeroPoint must be the same as input0. %insert AVAIL7 */ %{DeclareOperation_fl7 MIRROR_PAD 104}, /** * Reverses a specified dimension of a tensor. * * Supported tensor {@link %{OperandType}}: * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT16} * * {@link %{OperandTypeLinkPfx}TENSOR_FLOAT32} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} * * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} * * {@link %{OperandTypeLinkPfx}TENSOR_INT32} * * Supported tensor rank: up to 8. * * Inputs: * * 0: Input tensor of rank n. * * 1: Axis tensor of type {@link %{OperandTypeLinkPfx}TENSOR_INT32} and shape [1], * specifying which dimension of the input tensor is to be reversed. The dimension * must be in the range [0, n). * * Outputs: * * 0: The reversed tensor of the same shape as the input tensor. * For {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM} and * {@link %{OperandTypeLinkPfx}TENSOR_QUANT8_ASYMM_SIGNED} tensors, * the scales and zeroPoint must be the same as input0. %insert AVAIL7 */ %{DeclareOperation_fl7 REVERSE 105}, %/section %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% Misc HAL types %section OperandLifeTime /** * How an operand is used. */ %kind canonical enum class LifeTime { %else %{enum OperandLifeTime int32_t} { %/kind /** * The operand is internal to the model. It's created by an operation and * consumed by other operations. It must be an output operand of * exactly one operation. */ %{DeclareEnumValue TEMPORARY_VARIABLE 0}, /** * The operand is an input of %{the_model_or_a_subgraph}. It must not be an output * operand of any operation. * * An operand can't be both input and output of a %{model_or_subgraph}. */ %kind hal_1.0 %{DeclareEnumValue MODEL_INPUT 1}, %else %{DeclareEnumValue SUBGRAPH_INPUT 1}, %/kind /** * The operand is an output of %{the_model_or_a_subgraph}. It must be an output * operand of exactly one operation. * * An operand can't be both input and output of a %{model_or_subgraph}. */ %kind hal_1.0 %{DeclareEnumValue MODEL_OUTPUT 2}, %else %{DeclareEnumValue SUBGRAPH_OUTPUT 2}, %/kind /** * The operand is a constant found in Model%{::}operandValues. It must * not be an output operand of any operation. */ %{DeclareEnumValue CONSTANT_COPY 3}, /** * The operand is a constant that was specified via a Memory * object. It must not be an output operand of any operation. */ %{DeclareEnumValue CONSTANT_REFERENCE 4}, /** * The operand does not have a value. This is valid only for optional * arguments of operations. */ %{DeclareEnumValue NO_VALUE 5}, %kind canonical hal_1.3+ /** * The operand is a reference to a subgraph. It must be an input to one * or more {@link OperationType::IF} or {@link OperationType::WHILE} * operations. */ %{DeclareEnumValue SUBGRAPH 6}, %/kind %kind canonical /** * This operand is a constant found in a user buffer. It must not be an * output operand of any operation. */ %{DeclareEnumValue POINTER 7}, %/kind }; %/section %section DeviceStatus /** * Status of a device. */ %{enum DeviceStatus int32_t} { %{DeclareEnumValue AVAILABLE 0}, %{DeclareEnumValue BUSY 1}, %{DeclareEnumValue OFFLINE 2}, %{DeclareEnumValue UNKNOWN 3}, }; %/section %kind canonical %define init_execTime = kDefaultExecTime %define init_powerUsage = kDefaultPowerUsage %else %define init_execTime %define init_powerUsage %/kind %section PerformanceInfo /** * Performance information for the reference workload. * * Used by a driver to report its performance characteristics. */ struct PerformanceInfo { /** * Ratio of the time taken by the driver to execute the * workload compared to the time the CPU would take for the * same workload. A lower number is better. */ float execTime%{init_execTime}; /** * Ratio of the energy used by the driver compared to what * the CPU would use for doing the same workload. A lower number * is better. */ float powerUsage%{init_powerUsage}; }; %/section %section OutputShape /** * Describes the shape information of an output operand after execution. */ struct OutputShape { /** * Dimensions of the operand. */ %{vec} dimensions; /** * Whether the provided buffer size is sufficient for the output. */ bool isSufficient%{init_bool}; }; %/section %section MeasureTiming /** * Specifies whether or not to measure timing information during execution. */ %{enum MeasureTiming int32_t} { NO = 0, YES = 1, }; %/section %section ExecutionPreference /** * Execution preferences. %insert AVAIL1Short */ %{enum ExecutionPreference int32_t} { /** * Prefer executing in a way that minimizes battery drain. * This is desirable for compilations that will be executed often. */ %{DeclareExecutionPreference LOW_POWER 0}, /** * Prefer returning a single answer as fast as possible, even if this causes * more power consumption. */ %{DeclareExecutionPreference FAST_SINGLE_ANSWER 1}, /** * Prefer maximizing the throughput of successive frames, for example when * processing successive frames coming from the camera. */ %{DeclareExecutionPreference SUSTAINED_SPEED 2}, %kind canonical DEFAULT = FAST_SINGLE_ANSWER, %/kind }%{ndk_enum_name PreferenceCode}; %/section %section DeviceType /** * Device types. * * The type of NNAPI device. */ %{enum DeviceType int32_t} { %kind hal* // Leaving 0 unused as it means unknown type in NDK NNAPI. There is no // HAL equivalent of unknown type and a 1.2 HAL implementation must belong // to one of the categories below. %else /** The device type cannot be provided. */ %{DeclareDeviceType UNKNOWN 0}, %/kind /** The device does not fall into any category below. */ %{DeclareDeviceType OTHER 1}, /** The device runs NNAPI models on single or multi-core CPU. */ %{DeclareDeviceType CPU 2}, /** The device can run NNAPI models and also accelerate graphics APIs such * as OpenGL ES and Vulkan. */ %{DeclareDeviceType GPU 3}, /** Dedicated accelerator for Machine Learning workloads. */ %{DeclareDeviceType ACCELERATOR 4}, }%{ndk_enum_name DeviceTypeCode}; %/section %% NOTE: This is different from the NDK PriorityCode. %section Priority /** %kind ndk * Relative execution priority. * * Available since NNAPI feature level 4. %else * Priority given to a prepared model for execution. %/kind */ %{enum Priority int32_t} { %{DeclarePriority LOW 0 90}, %{DeclarePriority MEDIUM 1 100}, %{DeclarePriority HIGH 2 110}, %kind canonical ndk %{DeclarePriority DEFAULT MEDIUM ANEURALNETWORKS_PRIORITY_MEDIUM}, %/kind }%{ndk_enum_name PriorityCode}; %/section %kind canonical %define OptionalDuration OptionalDuration %else %define OptionalDuration uint64_t %/kind %section Timing /** %kind hal_1.2 %/kind * Timing information measured during execution. Each time is a duration from * the beginning of some task to the end of that task, including time when that * task is not active (for example, preempted by some other task, or * waiting for some resource to become available). * %kind hal* * Times are measured in microseconds. * When a time is not available, it must be reported as UINT64_MAX. %else * Times are measured in nanoseconds. %/kind */ struct Timing { /** Execution time on device (not driver, which runs on host processor). */ %{OptionalDuration} timeOnDevice; /** Execution time in driver (including time on device). */ %{OptionalDuration} timeInDriver; }; %/section %section Capabilities_float_quant_performance /** * Driver performance when operating on float32 data. */ PerformanceInfo float32Performance; /** * Driver performance when operating on asymmetric 8-bit quantized data. */ PerformanceInfo quantized8Performance; %/section %kind canonical %define OperandPerformanceTable OperandPerformanceTable %else %define OperandPerformanceTable vec %/kind %section Capabilities_relaxedPerformance /** * Driver performance when operating on float32 data but performing * calculations with range and/or precision as low as that of the IEEE * 754 16-bit floating-point format. */ %kind hal_1.1 PerformanceInfo relaxedFloat32toFloat16Performance; %else PerformanceInfo relaxedFloat32toFloat16PerformanceScalar; PerformanceInfo relaxedFloat32toFloat16PerformanceTensor; %/kind %/section %section Capabilities_operandPerformance /** * Performance by operand type. Must be sorted by OperandType. %kind hal_1.2 * If a particular OperandType is not present in operandPerformance, * its performance is treated as { .execTime = FLT_MAX, .powerUsage = FLT_MAX }. %else * * If a particular {@link OperandType} is not present in operandPerformance, * its performance is treated as * { .execTime = FLT_MAX, .powerUsage = FLT_MAX }. * * Performance does not apply to {@link OperandType::SUBGRAPH}, and a driver * must not report operand performance for {@link OperandType::SUBGRAPH}. %/kind */ %{OperandPerformanceTable} operandPerformance; %/section %section Capabilities_if_while_performance /** * Performance of an {@link OperationType::IF} operation is the sum of * {@link Capabilities::ifPerformance} and the mean of performance for the * two branch subgraphs, where performance for a subgraph is the sum of the * performance of all operations within the subgraph. */ PerformanceInfo ifPerformance; /** * Performance of a {@link OperationType::WHILE} operation is the sum of * {@link Capabilities::whilePerformance}, performance for the condition * subgraph and performance for the body subgraph, where performance for a * subgraph is the sum of the performance of all operations within the * subgraph. */ PerformanceInfo whilePerformance; %/section %section OperandPerformance /** * Driver performance when operating on a particular data type. * In the case of float32 data, this is used when the calculations * are not relaxed. */ struct OperandPerformance { OperandType type%{init_pod}; PerformanceInfo info; }; %/section %section Capabilities /** * The capabilities of a driver. %kind hal_1.2 * * Performance of an operation comes from the type of its first operand. * This represents performance for non extension operand types. %/kind %kind canonical hal_1.3+ * * This represents performance of non-extension operations. * * Performance of an operation other than {@link OperationType::IF} and * {@link OperationType::WHILE} comes from the type of its first operand. %/kind */ struct Capabilities { %kind canonical %insert-indented 4 PerformanceInfo %insert-indented 4 OperandPerformance class OperandPerformanceTable { public: static Result create( std::vector operandPerformances); PerformanceInfo lookup(OperandType type) const; const std::vector& asVector() const; private: explicit OperandPerformanceTable(std::vector operandPerformances); std::vector mSorted; }; %insert Capabilities_relaxedPerformance %insert Capabilities_operandPerformance %insert Capabilities_if_while_performance %/kind %kind hal_1.0 %insert Capabilities_float_quant_performance %/kind %kind hal_1.1 %insert Capabilities_float_quant_performance %insert Capabilities_relaxedPerformance %/kind %kind hal_1.2 %insert Capabilities_relaxedPerformance %insert-indented 4 OperandPerformance %insert Capabilities_operandPerformance %/kind %kind hal_1.3 %insert Capabilities_relaxedPerformance %insert-indented 4 OperandPerformance %insert Capabilities_operandPerformance %insert Capabilities_if_while_performance %/kind }; %/section %section DataLocation /** * Describes the location of a data object. */ struct DataLocation { %kind canonical /** * The address of the memory where the data is found. * * This field is only active when lifetime is POINTER. */ std::variant pointer; %/kind /** * The index of the memory pool where this location is found. */ uint32_t poolIndex%{init_int}; /** * Offset in bytes from the start of the pool. */ uint32_t offset%{init_int}; /** * The length of the data in bytes. */ uint32_t length%{init_int}; %kind canonical /** * The end padding of the specified memory region in bytes. */ uint32_t padding%{init_int}; %/kind }; %/section %section Extension_name /** * The extension name. * * The name must consist of lowercase latin letters, numbers, periods, and * underscore signs. The name must contain at least one period. * * The name must start with the reverse domain name of the vendor. * * Example: com.google.test_extension */ %{string} name; %/section %section Extension /** * Information about an extension. */ struct Extension { %kind hal* %insert Extension_name %/kind /** * Information about an extension operand type. */ struct OperandTypeInformation { /** * The extension operand type. */ uint16_t type%{init_int}; /** * Indicates whether the extension operand type represents a tensor or * a scalar. */ bool isTensor%{init_bool}; /** * The byte size of the operand (if scalar) or of a single element (if * tensor). */ uint32_t byteSize%{init_int}; }; %kind canonical %insert Extension_name %/kind /** * Information about operand types defined by the extension. */ %{vec} operandTypes; }; %/section %section Operation /** * Describes one operation of the model's graph. */ struct Operation { /** * The operation type. %kind hal_1.2+ * * Besides the values listed in {@link OperationType}, any value above * {@link OperationTypeRange::BASE_MAX} is possible and should be interpreted * as an extension type according to {@link Model::extensionNameToPrefix}. %/kind */ OperationType type%{init_pod}; /** * Describes the table that contains the indexes of the inputs of the * operation. The offset is the index in the operandIndexes table. */ %{vec} inputs; /** * Describes the table that contains the indexes of the outputs of the * operation. The offset is the index in the operandIndexes table. */ %{vec} outputs; }; %/section %section FusedActivationFunc /** * Fused activation function types. %insert AVAIL1Short */ %kind canonical enum class FusedActivationFunc : int32_t { %else %{enum FusedActivationFunc int32_t} { %/kind /** NO fused activation function. */ %{DeclareFusedActivationFunc NONE 0}, /** Fused ReLU activation function. */ %{DeclareFusedActivationFunc RELU 1}, /** Fused ReLU1 activation function. */ %{DeclareFusedActivationFunc RELU1 2}, /** Fused ReLU6 activation function. */ %{DeclareFusedActivationFunc RELU6 3}, }%{ndk_enum_name FuseCode}; %/section %section ExtraParams_Comment /** * Additional parameters specific to a particular operand type. */ %/section %section ExtraParams_none_Comment /** * No additional parameters. */ %/section %section ExtraParams_channelQuant_Comment /** * Symmetric per-channel quantization parameters. * * Only applicable to operands of type %{ANN}TENSOR_QUANT8_SYMM_PER_CHANNEL. */ %/section %section ExtraParams_extension_Comment /** * Extension operand parameters. * * The framework treats this as an opaque data blob. * The format is up to individual extensions. */ %/section %section SymmPerChannelQuantParams_Comment /** * Parameters for %{ANN}TENSOR_QUANT8_SYMM_PER_CHANNEL operand. */ %/section %section SymmPerChannelQuantParams %insert SymmPerChannelQuantParams_Comment struct SymmPerChannelQuantParams { /** Array of scaling values for each channel. Each value must be greater than zero. */ %{vec} scales; /** Index of the channel dimension */ uint32_t channelDim%{init_int}; }; %/section %kind canonical %section location_pointer_is_null * - location.pointer is null. %/section %else %section location_pointer_is_null %/section %/kind %% List item symbol %kind hal* %define li . %else %define li - %/kind %section Operand /** * Describes one operand of the model's graph. */ struct Operand { %kind canonical %insert-indented 4 OperandLifeTime %insert-indented 4 ExtraParams_none_Comment using NoParams = std::monostate; %insert-indented 4 SymmPerChannelQuantParams %insert-indented 4 ExtraParams_extension_Comment using ExtensionParams = std::vector; %insert-indented 4 ExtraParams_Comment using ExtraParams = std::variant; %/kind /** %kind canonical * The data type. * * Besides the values listed in {@link OperationType}, any value equal or over * (1 << kExtensionTypeBits) is possible and should be interpreted * as an extension type according to {@link Model::extensionNameToPrefix}. %/kind %kind hal_1.2+ * The data type. * * Besides the values listed in {@link OperandType}, any value above * {@link OperandTypeRange::BASE_MAX} is possible and should be interpreted * as an extension type according to {@link Model::extensionNameToPrefix}. %/kind %kind hal_1.0 * Data type of the operand. %/kind */ OperandType type%{init_pod}; /** * Dimensions of the operand. * * For a scalar operand, dimensions.size() must be 0. %kind hal_1.0 * * For a tensor operand, dimensions.size() must be at least 1; * however, any of the dimensions may be unspecified. %/kind * * A tensor operand with all dimensions specified has "fully * specified" dimensions. Whenever possible (i.e., whenever the * dimensions are known at model construction time), a tensor * operand should have (but is not required to have) fully * specified dimensions, in order to enable the best possible * performance. * * If a tensor operand's dimensions are not fully specified, the * dimensions of the operand are deduced from the operand * dimensions and values of the operation for which that operand %kind hal_1.0 hal_1.1 hal_1.2 * is an output. %else * is an output or from the corresponding {@link OperationType::IF} or * {@link OperationType::WHILE} operation input operand dimensions in the * case of referenced subgraph input operands. %/kind * * In the following situations, a tensor operand's dimensions must * be fully specified: * %kind canonical * %{li} The operand has lifetime CONSTANT_COPY, CONSTANT_REFERENCE, or * POINTER. %else * %{li} The operand has lifetime CONSTANT_COPY or * CONSTANT_REFERENCE. %/kind * %kind hal_1.0 * %{li} The operand has lifetime MODEL_INPUT or MODEL_OUTPUT. Fully * specified dimensions must either be present in the %/kind %kind hal_1.2 * %{li} The operand has lifetime MODEL_INPUT. Fully * specified dimensions must either be present in the %/kind %kind canonical hal_1.3+ * %{li} The operand has lifetime SUBGRAPH_INPUT and belongs to the main * subgraph. Fully specified dimensions must either be present in the %/kind * Operand or they must be provided in the corresponding * RequestArgument. %kind hal_1.0 * EXCEPTION: If the input or output is optional and omitted %else * EXCEPTION: If the input is optional and omitted %/kind * (by setting the hasNoValue field of the corresponding * RequestArgument to true) then it need not have fully * specified dimensions. * * A tensor operand with some number of unspecified dimensions is * represented by setting each unspecified dimension to 0. %kind canonical hal_1.2+ * * A tensor operand with unspecified rank is represented by providing * an empty dimensions vector. %/kind */ %{Dimensions} dimensions; %kind hal* /** * The number of times this operand appears as an operation input. * * (For example, if this operand appears once in one operation's * input list, and three times in another operation's input list, * then numberOfConsumers = 4.) */ uint32_t numberOfConsumers; %/kind /** * Quantized scale of the operand. * %kind hal_1.0 * Only applicable if the operand is of type TENSOR_QUANT8_ASYMM or * TENSOR_INT32. %else * Must be 0 when not applicable to an operand type. * * See {@link OperandType}. %/kind */ float scale%{init_float}; /** * Quantized zero-point offset of the operand. * %kind hal_1.0 * Only applicable if the operand is of type TENSOR_QUANT8_ASYMM. %else * Must be 0 when not applicable to an operand type. * * See {@link OperandType}. %/kind */ int32_t zeroPoint%{init_int}; /** * How the operand is used. */ %{concat_or_skip_first Operand LifeTime} lifetime%{init_pod}; /** * Where to find the data for this operand. %kind hal_1.0 hal_1.1 hal_1.2 * If the lifetime is TEMPORARY_VARIABLE, MODEL_INPUT, MODEL_OUTPUT, or * NO_VALUE: %else * If the lifetime is TEMPORARY_VARIABLE, SUBGRAPH_INPUT, SUBGRAPH_OUTPUT, * or NO_VALUE: %/kind * - All the fields must be 0. * If the lifetime is CONSTANT_COPY: %insert location_pointer_is_null * - location.poolIndex is 0. * - location.offset is the offset in bytes into Model%{::}operandValues. * - location.length is set. %kind canonical * - location.padding is 0. %/kind * If the lifetime is CONSTANT_REFERENCE: %insert location_pointer_is_null * - location.poolIndex is set. * - location.offset is the offset in bytes into the specified pool. * - location.length is set. %kind canonical * - location.padding is set. %/kind %kind canonical hal_1.3+ * If the lifetime is SUBGRAPH: %insert location_pointer_is_null * - location.poolIndex is 0. * - location.offset is the index of the referenced subgraph in * {@link Model::referenced}. * - location.length is 0. %/kind %kind canonical * - location.padding is 0. %/kind %kind canonical * If the lifetime is POINTER: * - location.pointer is non-null. * - location.poolIndex is 0. * - location.offset is 0. * - location.length is set. * - location.padding is 0. %/kind */ DataLocation location; %kind hal_1.2 %insert-indented 4 ExtraParams_Comment safe_union ExtraParams { %insert-indented 8 ExtraParams_none_Comment Monostate none; %insert-indented 8 ExtraParams_channelQuant_Comment SymmPerChannelQuantParams channelQuant; %insert-indented 8 ExtraParams_extension_Comment vec extension; } extraParams; %/kind %kind canonical hal_1.3 %insert-indented 4 ExtraParams_Comment %{concat_or_skip_first @1.2::Operand. ExtraParams} extraParams; %/kind }; %/section %kind canonical %define OperandValues OperandValues %define SharedMemory SharedMemory %else %define OperandValues vec %define SharedMemory memory %/kind %section ExtensionNameAndPrefix %kind canonical /** * The mapping between extension names and prefixes of values like operand and operation type, and * token in {@link TokenValuePair}. * * An operand or operation whose numeric type value is above {@link IDevice::OPERAND_TYPE_BASE_MAX} * or {@link IDevice::OPERATION_TYPE_BASE_MAX} respectively should be interpreted as an extension * operand/operation. The low kExtensionTypeBits bits of the value correspond to the type ID within * the extension and the high kExtensionPrefixBits bits encode the "prefix", which maps uniquely to * the extension name. The sign bit is always 0. * * For example, if a model contains an operation whose value is 0x7AAABBBB and * Model::extensionNameToPrefix contains an entry with prefix=0x7AAA and * name="vendor.test.test_extension", then the operation should be interpreted as the operation * 0xBBBB of the extension named vendor.test.test_extension. * * This is a one-to-one correspondence. That is, there must be at most one prefix corresponding to * each extension name and at most one extension name corresponding to each prefix. */ %/kind %kind hal_1.2 hal_1.3 /** * A correspondence between an extension name and a prefix of operand and * operation type values. */ %/kind struct ExtensionNameAndPrefix { /** * The extension name. * * See {@link Extension::name} for the format specification. */ %{string} name; %kind canonical /** * The extension prefix. Only the lowest 15 bits are used, so the value must be less than 32768. */ %/kind %kind hal_1.2 hal_1.3 /** * The unique extension identifier within the model. * * See {@link Model::extensionNameToPrefix}. */ %/kind uint16_t prefix%{init_int}; }; %/section %section Model_1.0 /** * A byte buffer containing operand data that were copied into the model. * * An operand's value must be located here if and only if Operand::lifetime * equals %{OperandLifeTime}::CONSTANT_COPY. */ %{OperandValues} operandValues; /** * A collection of shared memory pools containing operand values. * * An operand's value must be located here if and only if Operand::lifetime * equals %{OperandLifeTime}::CONSTANT_REFERENCE. */ %{vec}<%{SharedMemory}> pools; %/section %section Model_1.1 /** * 'true' indicates TENSOR_FLOAT32 may be calculated with range and/or * precision as low as that of the IEEE 754 16-bit floating-point format. * 'false' indicates TENSOR_FLOAT32 must be calculated using at least the * range and precision of the IEEE 754 32-bit floating-point format. */ bool relaxComputationFloat32toFloat16%{init_bool}; %/section %section Model_1.2 /** * The mapping between extension names and prefixes of operand and * operation type values. * %kind canonical * An operand or operation whose numeric type value is equal to or greater * than (1 << kExtensionTypeBits) should be interpreted %/kind %kind hal* * An operand or operation whose numeric type value is above * {@link OperandTypeRange::BASE_MAX} or * {@link OperationTypeRange::BASE_MAX} respectively should be interpreted %/kind * as an extension operand. The low %kind hal_1.2 * {@link Model::ExtensionTypeEncoding::LOW_BITS_TYPE} bits of the value * correspond to the type ID within the extension and the high * {@link Model::ExtensionTypeEncoding::HIGH_BITS_PREFIX} bits encode %/kind %kind hal_1.3 * {@link @1.2::Model::ExtensionTypeEncoding::LOW_BITS_TYPE} bits of the * value correspond to the type ID within the extension and the high * {@link @1.2::Model::ExtensionTypeEncoding::HIGH_BITS_PREFIX} bits encode %/kind %kind canonical * {@link kExtensionTypeBits} bits of the value correspond to the type ID * within the extension and the high {@link kExtensionPrefixBits} bits encode %/kind * the "prefix", which maps uniquely to the extension name. * * For example, if a model contains an operation whose value is * 0xAAAABBBB and extensionNameToPrefix contains an entry with * prefix=0xAAAA and name="vendor.test.test_extension", then * the operation should be interpreted as the operation 0xBBBB * of the extension named vendor.test.test_extension. * * This is a one-to-one correspondence. That is, there must be at most one * prefix corresponding to each extension name and at most one extension * name corresponding to each prefix. */ %kind hal_1.3 %{vec}<@1.2::Model.ExtensionNameAndPrefix> extensionNameToPrefix; %else %{vec} extensionNameToPrefix; %/kind %/section %section Model_1.3_main_and_referenced_subgraphs /** * The top-level subgraph. */ Subgraph main; /** * Referenced subgraphs. * * Each subgraph is referenced by the main subgraph or at least one other * referenced subgraph. * * There must be no reference cycles. */ %{vec} referenced; %/section %section Subgraph_fields /** * All operands included in the %{model_or_subgraph}. */ %{vec} operands; /** * All operations included in the %{model_or_subgraph}. * * The operations are sorted into execution order. Every operand * with lifetime %{MODEL_or_SUBGRAPH}_OUTPUT or TEMPORARY_VARIABLE must be * written before it is read. */ %{vec} operations; /** * Input indexes of the %{model_or_subgraph}. There must be at least one. * * Each value corresponds to the index of the operand in "operands". */ %{vec} inputIndexes; /** * Output indexes of the %{model_or_subgraph}. There must be at least one. * * Each value corresponds to the index of the operand in "operands". */ %{vec} outputIndexes; %/section %section Subgraph /** * An excerpt of the execution graph. */ struct Subgraph { %insert Subgraph_fields }; %/section %section ExtensionTypeEncoding /** * Numeric values of extension operand and operation types have the * following structure: * - 16 high bits represent the "prefix", which corresponds uniquely to the * extension name. * - 16 low bits represent the type ID within the extension. */ %kind canonical constexpr uint8_t kExtensionTypeBits = 16; constexpr uint8_t kExtensionPrefixBits = 16; constexpr uint32_t kTypeWithinExtensionMask = 0xFFFF; %else enum ExtensionTypeEncoding : uint8_t { HIGH_BITS_PREFIX = 16, LOW_BITS_TYPE = 16, }; %/kind %/section %section Model /** * A Neural Network Model. * * This includes not only the execution graph, but also constant data such as * weights or scalars added at construction time. The only information that %kind hal_1.0 * might not be known is the shape of the input tensors. %else * may not be known is the shape of the input tensors. %/kind */ struct Model { %kind canonical %insert-indented 4 Subgraph class OperandValues { public: OperandValues(); OperandValues(const uint8_t* data, size_t length); // Append a segment of memory (starting at `data` with `length` number of bytes) to the back // of `OperandValues`, adding padding as necessary so that the appended data is aligned. // Refer to `getAlignmentForLength` for more information on alignment (such as what the // current alignments are for different data lengths). DataLocation append(const uint8_t* data, size_t length); const uint8_t* data() const; size_t size() const; private: std::vector mData; }; %insert Model_1.3_main_and_referenced_subgraphs %insert Model_1.0 %insert Model_1.1 %insert Model_1.2 %/kind %kind hal_1.0 %insert Subgraph_fields %insert Model_1.0 %/kind %kind hal_1.1 %insert Subgraph_fields %insert Model_1.0 %insert Model_1.1 %/kind %kind hal_1.2 %insert Subgraph_fields %insert Model_1.0 %insert Model_1.1 %insert Model_1.2 %insert-indented 4 ExtensionNameAndPrefix %insert-indented 4 ExtensionTypeEncoding %/kind %kind hal_1.3 %insert Model_1.3_main_and_referenced_subgraphs %insert Model_1.0 %insert Model_1.1 %insert Model_1.2 %/kind }; %/section %section BufferDesc /** * A buffer descriptor. Describes the properties of a buffer. */ struct BufferDesc { /** * Dimensions of the buffer. May have unknown dimensions or rank. A buffer with some number * of unspecified dimensions is represented by setting each unspecified dimension to 0. A * buffer with unspecified rank is represented by providing an empty dimensions vector. */ %{Dimensions} dimensions; }; %/section %section BufferRole /** * Describes a role of an input or output to a prepared model. */ struct BufferRole { /** * The index of the IPreparedModel within the "preparedModel" argument passed in * IDevice::allocate. */ uint32_t modelIndex%{init_int}; /** * The index of the input or output operand. */ uint32_t ioIndex%{init_int}; /** * A floating-point value within the range (0.0, 1.0]. Describes how likely the * buffer is to be used in the specified role. This is provided as a hint to * optimize the case when multiple roles prefer different buffer locations or data * layouts. */ %kind canonical float probability%{init_float}; %else float frequency%{init_float}; %/kind }; %/section %kind aidl %define inputIndexes @@@NOT_DEFINED@@@ %define outputIndexes @@@NOT_DEFINED@@@ %/kind %kind canonical %define inputIndexes Model::main::inputIndexes %define outputIndexes Model::main::outputIndexes %/kind %kind hal_1.3 %define inputIndexes Model.main.inputIndexes %define outputIndexes Model.main.outputIndexes %/kind %kind hal_1.0 hal_1.1 hal_1.2 %define inputIndexes Model.inputIndexes %define outputIndexes Model.outputIndexes %/kind %kind ndk %define inputIndexes @@@NOT_DEFINED@@@ %define outputIndexes @@@NOT_DEFINED@@@ %/kind %kind canonical %define inputs inputs %define outputs outputs %else %define inputs input %define outputs output %/kind %section Request_inputs_and_outputs /** * Input data and information to be used in the execution of a prepared * model. * * The index of the input corresponds to the index in %{inputIndexes}. * E.g., %{inputs}[i] corresponds to %{inputIndexes}[i]. */ %{vec}<%{concat_or_skip_first Request Argument}> inputs; /** * Output data and information to be used in the execution of a prepared * model. * * The index of the output corresponds to the index in %{outputIndexes}. * E.g., %{outputs}[i] corresponds to %{outputIndexes}[i]. */ %{vec}<%{concat_or_skip_first Request Argument}> outputs; %/section %section Request_pools /** %kind hal_1.0 * A collection of shared memory pools containing operand data for both the %else * A collection of memory pools containing operand data for both the %/kind * inputs and the outputs to a model. */ %kind hal_1.0 vec pools; %else %{vec} pools; %/kind %/section %section Request_MemoryPool_Comment /** * A memory pool. */ %/section %section RequestArgument /** * Metadata information specifying the location of the input or output data and * any updates to the input or output operand. */ struct %{concat_or_skip_first Request Argument} { %kind canonical enum class LifeTime { POOL = 0, NO_VALUE = 1, POINTER = 2, }; %/kind %kind hal_1.0 /** * If true, the argument does not have a value. This can be used for * operations that take optional arguments. If true, the fields of location * are set to 0 and the dimensions vector is left empty. */ bool hasNoValue; %/kind %kind canonical LifeTime lifetime%{init_pod}; %/kind /** * The location within one of the memory pools passed in the Request. */ DataLocation location; /** * Updated dimension information. * * If dimensions.size() > 0, dimension information was provided * along with the argument. This can be the case for models that * accept inputs of varying size. This can't change the rank, just * the value of the dimensions that were unspecified in the * model. If dimensions.size() > 0, then all dimensions must be * specified here; and any dimension that was specified in the * model must have the same value here. * * If the dimensions in the model are not fully specified, then * they must be fully specified here, unless hasNoValue is set to * true. If the dimensions in the model are fully specified, then * either dimensions.size() may be 0, or the dimensions in the * model must be identical to the dimensions here. */ %{Dimensions} dimensions; }; %/section %section Request /** * Inputs to be sent to and outputs to be retrieved from a prepared model. * * A Request serves two primary tasks: * 1) Provides the input and output data to be used when executing the model. * 2) Specifies any updates to the input operand metadata that were left * unspecified at model preparation time. * * An output must not overlap with any other output, with an input, or * with an operand of lifetime CONSTANT_REFERENCE. */ struct Request { %kind canonical %insert-indented 4 RequestArgument /** * Specifies a driver-managed buffer. It is the token corresponding to an * IBuffer returned from IDevice::allocate, and is specific to the IDevice * object. */ enum class MemoryDomainToken : uint32_t {}; %insert-indented 4 Request_MemoryPool_Comment using MemoryPool = std::variant; %/kind %insert Request_inputs_and_outputs %kind hal_1.3 %insert-indented 4 Request_MemoryPool_Comment safe_union MemoryPool { /** * Specifies a client-managed shared memory pool. */ memory hidlMemory; /** * Specifies a driver-managed buffer. It is the token returned from IDevice::allocate, * and is specific to the IDevice object. */ uint32_t token; }; %/kind %insert Request_pools }; %/section %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%