COMPMID-3101 Fuse activation with floating point elementwise operation layers in CL
Signed-off-by: Giorgio Arena <giorgio.arena@arm.com>
Change-Id: I1693f8664ba7c0dc8c076bbe7365cef1e667bd25
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/2718
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
diff --git a/arm_compute/runtime/NEON/functions/NEPixelWiseMultiplication.h b/arm_compute/runtime/NEON/functions/NEPixelWiseMultiplication.h
index ede4327..2b31032 100644
--- a/arm_compute/runtime/NEON/functions/NEPixelWiseMultiplication.h
+++ b/arm_compute/runtime/NEON/functions/NEPixelWiseMultiplication.h
@@ -57,8 +57,10 @@
* Scale must be positive and its value must be either 1/255 or 1/2^n where n is between 0 and 15.
* @param[in] overflow_policy Overflow policy. ConvertPolicy cannot be WRAP if datatype is QASYMM8, QASYMM8_SIGNED or QSYMM16.
* @param[in] rounding_policy Rounding policy.
+ * @param[in] act_info (Optional) Activation layer information in case of a fused activation. Currently not supported.
*/
- void configure(ITensor *input1, ITensor *input2, ITensor *output, float scale, ConvertPolicy overflow_policy, RoundingPolicy rounding_policy);
+ void configure(ITensor *input1, ITensor *input2, ITensor *output, float scale, ConvertPolicy overflow_policy, RoundingPolicy rounding_policy,
+ const ActivationLayerInfo &act_info = ActivationLayerInfo());
/** Static function to check if given info will lead to a valid configuration of @ref NEPixelWiseMultiplication
*
* @note For @p scale equal to 1/255 only round to nearest even (implemented as round half up) is supported.
@@ -79,10 +81,12 @@
* Scale must be positive and its value must be either 1/255 or 1/2^n where n is between 0 and 15.
* @param[in] overflow_policy Overflow policy. ConvertPolicy cannot be WRAP if datatype is QASYMM8, QASYMM8_SIGNED or QSYMM16.
* @param[in] rounding_policy Rounding policy.
+ * @param[in] act_info (Optional) Activation layer information in case of a fused activation. Currently not supported.
*
* @return a status
*/
- static Status validate(const ITensorInfo *input1, const ITensorInfo *input2, const ITensorInfo *output, float scale, ConvertPolicy overflow_policy, RoundingPolicy rounding_policy);
+ static Status validate(const ITensorInfo *input1, const ITensorInfo *input2, const ITensorInfo *output, float scale, ConvertPolicy overflow_policy, RoundingPolicy rounding_policy,
+ const ActivationLayerInfo &act_info = ActivationLayerInfo());
};
/** Basic function to run @ref NEComplexPixelWiseMultiplicationKernel. */
@@ -91,20 +95,22 @@
public:
/** Initialise the kernel's inputs, output.
*
- * @param[in, out] input1 An input tensor. Data types supported: F32. Number of channels supported: 2 (complex tensor).
- * The input tensor is [in, out] because its TensorInfo might be modified inside the kernel in case of broadcasting of dimension 0.
- * @param[in, out] input2 An input tensor. Data types supported: same as @p input1. Number of channels supported: same as @p input1.
- * The input tensor is [in, out] because its TensorInfo might be modified inside the kernel in case of broadcasting of dimension 0.
- * @param[out] output The output tensor. Data types supported: same as @p input1. Number of channels: same as @p input1.
+ * @param[in, out] input1 An input tensor. Data types supported: F32. Number of channels supported: 2 (complex tensor).
+ * The input tensor is [in, out] because its TensorInfo might be modified inside the kernel in case of broadcasting of dimension 0.
+ * @param[in, out] input2 An input tensor. Data types supported: same as @p input1. Number of channels supported: same as @p input1.
+ * The input tensor is [in, out] because its TensorInfo might be modified inside the kernel in case of broadcasting of dimension 0.
+ * @param[out] output The output tensor. Data types supported: same as @p input1. Number of channels: same as @p input1.
+ * @param[in] act_info (Optional) Activation layer information in case of a fused activation. Currently not supported.
*/
- void configure(ITensor *input1, ITensor *input2, ITensor *output);
+ void configure(ITensor *input1, ITensor *input2, ITensor *output, const ActivationLayerInfo &act_info = ActivationLayerInfo());
/** Static function to check if given info will lead to a valid configuration of @ref NEComplexPixelWiseMultiplication
*
- * @param[in] input1 An input tensor info. Data types supported: F32. Number of channels supported: 2 (complex tensor).
- * @param[in] input2 An input tensor info. Data types supported: same as @p input1. Number of channels supported: same as @p input1.
- * @param[in] output The output tensor info. Data types supported: same as @p input1. Number of channels supported: same as @p input1.
+ * @param[in] input1 An input tensor info. Data types supported: F32. Number of channels supported: 2 (complex tensor).
+ * @param[in] input2 An input tensor info. Data types supported: same as @p input1. Number of channels supported: same as @p input1.
+ * @param[in] output The output tensor info. Data types supported: same as @p input1. Number of channels supported: same as @p input1.
+ * @param[in] act_info (Optional) Activation layer information in case of a fused activation. Currently not supported.
*/
- static Status validate(const ITensorInfo *input1, const ITensorInfo *input2, const ITensorInfo *output);
+ static Status validate(const ITensorInfo *input1, const ITensorInfo *input2, const ITensorInfo *output, const ActivationLayerInfo &act_info = ActivationLayerInfo());
};
}
#endif /*ARM_COMPUTE_NEPIXELWISEMULTIPLICATION_H */