* 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 reduce_std_v2_proto.h
* \brief
*/
#ifndef OPS_OP_PROTO_INC_REDUCE_STD_V2_H_
#define OPS_OP_PROTO_INC_REDUCE_STD_V2_H_
#include "graph/operator_reg.h"
#include "graph/types.h"
namespace ge {
* @brief Calculates the standard deviation and average value of tensors.
* @par Inputs:
* x: A tensor. Format supports ND. Must be one of the following types: float32, float16, bfloat16. \n
* @par Attributes:
* Four Attributes, including:
* @li dim: The dimensions to reduce. An optional listint, Defaults to "None".
* If None (the default), reduces all dimensions.
* Must be in the range [-rank(x), rank(x)).
* @li correction: An optional int. Used for Bessel's correction. Defaults to 1.
* @li keepdim: An optional bool. Defaults to "False".
* If "True", Keep the original tensor dimension.
* If "False", Do not keep the original tensor dimension.
* @li is_mean_out: An optional bool. Defaults to "True".
* If "True", Output the mean.
* If "False", Do not output the mean. \n
* @par Outputs:
* Two Outputs, including:
* @li std: A tensor, the standard deviation of x. Has the same type and format as "x".
* @li mean: A tensor, the mean of x. Has the same type and format as "x". \n
* @par Third-party framework compatibility
* Compatible with the Pytorch operator std and std_mean.
*/
REG_OP(ReduceStdV2)
.INPUT(x, TensorType({DT_FLOAT, DT_FLOAT16, DT_BF16}))
.OUTPUT(std, TensorType({DT_FLOAT, DT_FLOAT16, DT_BF16}))
.OUTPUT(mean, TensorType({DT_FLOAT, DT_FLOAT16, DT_BF16}))
.ATTR(dim, ListInt, {})
.ATTR(correction, Int, 1)
.ATTR(keepdim, Bool, false)
.ATTR(is_mean_out, Bool, true)
.OP_END_FACTORY_REG(ReduceStdV2)
}
#endif
