* Copyright (c) 2025 Huawei Technologies Co., Ltd.
* This program is free software, you can redistribute it and/or modify it under the terms and conditions of
* CANN Open Software License Agreement Version 2.0 (the "License").
* Please refer to the License for details. You may not use this file except in compliance with the License.
* THIS SOFTWARE IS PROVIDED ON AN "AS IS" BASIS, WITHOUT WARRANTIES OF ANY KIND, EITHER EXPRESS OR IMPLIED,
* INCLUDING BUT NOT LIMITED TO NON-INFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR A PARTICULAR PURPOSE.
* See LICENSE in the root of the software repository for the full text of the License.
*/
* \file add_rms_norm_cast_proto.h
* \brief
*/
#ifndef OPS_NORM_ADD_RMS_NORM_CAST_PROTO_H_
#define OPS_NORM_ADD_RMS_NORM_CAST_PROTO_H_
#include "graph/operator_reg.h"
namespace ge {
* @brief AddRmsNormCast operator interface implementation. \n
* calculating: x1, x2, gamma \n
* x = x1 + x2 \n
* rstd = np.rsqrt(np.mean(np.power(x,2), reduce_axis, keepdims=True) + epsilon)) \n
* y2 = gamma * (x * rstd) \n
* y1 = cast16232(y2) \n
* @par Inputs
* Three inputs, including:
* @li x1: A tensor. Support dtype: float16/bfloat16, support format: ND.
* @li x2: A tensor. Support dtype: float16/bfloat16, support format: ND.
* @li gamma: A tensor. Support dtype: float16/bfloat16, support format: ND.
* @par Attributes
* epsilon: Input eps in the formula, which is used to prevent division-by-zero errors.
* An optional attribute, the type is float. Defaults to 1e-6.
* @par Outputs
* Four outputs, including:
* @li y1: A tensor. Support dtype: float32, support format: ND.
* @li y2: A tensor. Support dtype: float16/bfloat16, support format: ND.
* @li rstd: A tensor. Describing the reciprocal of (x1 + x2)'s standard deviation.
* Support dtype: float32, support format: ND.
* @li x: A tensor. Support dtype: float16/bfloat16, support format: ND.
*/
REG_OP(AddRmsNormCast)
.INPUT(x1, TensorType({DT_FLOAT16, DT_BF16}))
.INPUT(x2, TensorType({DT_FLOAT16, DT_BF16}))
.INPUT(gamma, TensorType({DT_FLOAT16, DT_BF16}))
.OUTPUT(y1, TensorType({DT_FLOAT}))
.OUTPUT(y2, TensorType({DT_FLOAT16, DT_BF16}))
.OUTPUT(rstd, TensorType({DT_FLOAT}))
.OUTPUT(x, TensorType({DT_FLOAT16, DT_BF16}))
.ATTR(epsilon, Float, 1e-6f)
.OP_END_FACTORY_REG(AddRmsNormCast)
}
#endif