aboutsummaryrefslogtreecommitdiff
path: root/arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NHWCKernel.h
diff options
context:
space:
mode:
Diffstat (limited to 'arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NHWCKernel.h')
-rw-r--r--arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NHWCKernel.h56
1 files changed, 34 insertions, 22 deletions
diff --git a/arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NHWCKernel.h b/arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NHWCKernel.h
index 7d0ecec13e..e8cca954e2 100644
--- a/arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NHWCKernel.h
+++ b/arm_compute/core/CL/kernels/CLDepthwiseConvolutionLayer3x3NHWCKernel.h
@@ -40,34 +40,46 @@ public:
/** Default move assignment operator. */
/** Initialize the function's source, destination, conv and border_size.
*
- * @param[in] input Source tensor. DataType supported: QASYMM8.
- * @param[in] weights Weights tensor. A 3D tensor with dimensions [IFM, 3, 3]. Data type supported: Same as @p input.
- * @param[in] biases Biases tensor. A 1D tensor with dimensions [IFM]. Must be nullptr if not needed.
- * Data type supported: Same as @p input.
- * @param[out] output Destination tensor. Data type supported: Same as @p input.
- * @param[in] conv_info Padding and stride information to use for the convolution.
- * @param[in] depth_multiplier (Optional) Multiplier to apply to the input's depth in order to retrieve the output's depth. Defaults to 1.
- * @param[in] act_info (Optional) Activation layer information in case of a fused activation. Only RELU, BOUNDED_RELU and LU_BOUNDED_RELU are supported.
- * @param[in] dilation (Optional) Dilation, in elements, across x and y. Defaults to (1, 1).
+ * @param[in] input Source tensor. DataType supported: QASYMM8.
+ * @param[in] weights Weights tensor. A 3D tensor with dimensions [IFM, 3, 3].
+ * Data type supported: Same as @p input or QASYMM8/QSYMM8_PER_CHANNEL when @p input is QASYMM8.
+ * @param[in] biases Biases tensor. A 1D tensor with dimensions [IFM]. Must be nullptr if not needed.
+ * Data type supported: Same as @p input, S32 when input is QASYMM8.
+ * @param[out] output Destination tensor. Data type supported: Same as @p input.
+ * @param[in] conv_info Padding and stride information to use for the convolution.
+ * @param[in] depth_multiplier (Optional) Multiplier to apply to the input's depth in order to retrieve the output's depth. Defaults to 1.
+ * @param[in] act_info (Optional) Activation layer information in case of a fused activation. Only RELU, BOUNDED_RELU and LU_BOUNDED_RELU are supported.
+ * @param[in] dilation (Optional) Dilation, in elements, across x and y. Defaults to (1, 1).
+ * @param[in] output_multipliers (Optional) Output multipliers tensor for quantized computations. In case of per-channel quantization,
+ * the number of multipliers must be equal to the number of filters (IFM). Supported data types: S32
+ * @param[in] output_shifts (Optional) Output shifts tensor for quantized computations. In case of per-channel quantization,
+ * the number of multipliers must be equal to the number of filters (IFM). Supported data types: S32
*/
- void configure(const ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const PadStrideInfo &conv_info, unsigned int depth_multiplier,
- ActivationLayerInfo act_info, const Size2D &dilation) override;
+ void configure(const ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const PadStrideInfo &conv_info,
+ unsigned int depth_multiplier = 1, ActivationLayerInfo act_info = ActivationLayerInfo(), const Size2D &dilation = Size2D(1U, 1U),
+ const ICLTensor *output_multipliers = nullptr, const ICLTensor *output_shifts = nullptr) override;
/** Static function to check if given info will lead to a valid configuration of @ref CLDepthwiseConvolutionLayer3x3NHWCKernel
*
- * @param[in] input Source tensor info. DataType supported: QASYMM8.
- * @param[in] weights Weights tensor info. A 3D tensor with dimensions [IFM, 3, 3]. Data type supported: Same as @p input.
- * @param[in] biases Biases tensor info. A 1D tensor with dimensions [IFM]. Must be nullptr if not needed.
- * Data type supported: Same as @p input.
- * @param[in] output Destination tensor info. Data type supported: Same as @p input.
- * @param[in] conv_info Padding and stride information to use for the convolution.
- * @param[in] depth_multiplier (Optional) Multiplier to apply to the input's depth in order to retrieve the output's depth. Defaults to 1.
- * @param[in] act_info (Optional) Activation layer information in case of a fused activation. Only RELU, BOUNDED_RELU and LU_BOUNDED_RELU are supported.
- * @param[in] dilation (Optional) Dilation, in elements, across x and y. Defaults to (1, 1).
+ * @param[in] input Source tensor info. DataType supported: QASYMM8.
+ * @param[in] weights Weights tensor info. A 3D tensor with dimensions [IFM, 3, 3].
+ * Data type supported: Same as @p input or QASYMM8/QSYMM8_PER_CHANNEL when @p input is QASYMM8.
+ * @param[in] biases Biases tensor info. A 1D tensor with dimensions [IFM]. Must be nullptr if not needed.
+ * Data type supported: Same as @p input, S32 when input is QASYMM8.
+ * @param[in] output Destination tensor info. Data type supported: Same as @p input.
+ * @param[in] conv_info Padding and stride information to use for the convolution.
+ * @param[in] depth_multiplier (Optional) Multiplier to apply to the input's depth in order to retrieve the output's depth. Defaults to 1.
+ * @param[in] act_info (Optional) Activation layer information in case of a fused activation. Only RELU, BOUNDED_RELU and LU_BOUNDED_RELU are supported.
+ * @param[in] dilation (Optional) Dilation, in elements, across x and y. Defaults to (1, 1).
+ * @param[in] output_multipliers (Optional) Output multipliers tensor info for quantized computations. In case of per-channel quantization,
+ * the number of multipliers must be equal to the number of filters (IFM). Supported data types: S32
+ * @param[in] output_shifts (Optional) Output shifts tensor for quantized computations. In case of per-channel quantization,
+ * the number of multipliers must be equal to the number of filters (IFM). Supported data types: S32
*
* @return a status
*/
- static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *output, const PadStrideInfo &conv_info, unsigned int depth_multiplier,
- ActivationLayerInfo act_info = ActivationLayerInfo(), const Size2D &dilation = Size2D(1U, 1U));
+ static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *output, const PadStrideInfo &conv_info,
+ unsigned int depth_multiplier = 1, ActivationLayerInfo act_info = ActivationLayerInfo(), const Size2D &dilation = Size2D(1U, 1U),
+ const ITensorInfo *output_multipliers = nullptr, const ITensorInfo *output_shifts = nullptr);
// Inherited methods overridden:
void run(const Window &window, cl::CommandQueue &queue) override;