xref: /aosp_15_r20/external/ComputeLibrary/arm_compute/runtime/NEON/functions/NEBatchNormalizationLayer.h (revision c217d954acce2dbc11938adb493fc0abd69584f3)
1 /*
2  * Copyright (c) 2017-2021 Arm Limited.
3  *
4  * SPDX-License-Identifier: MIT
5  *
6  * Permission is hereby granted, free of charge, to any person obtaining a copy
7  * of this software and associated documentation files (the "Software"), to
8  * deal in the Software without restriction, including without limitation the
9  * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
10  * sell copies of the Software, and to permit persons to whom the Software is
11  * furnished to do so, subject to the following conditions:
12  *
13  * The above copyright notice and this permission notice shall be included in all
14  * copies or substantial portions of the Software.
15  *
16  * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
17  * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
18  * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
19  * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
20  * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
21  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
22  * SOFTWARE.
23  */
24 #ifndef ARM_COMPUTE_NEBATCHNORMALIZATIONLAYER_H
25 #define ARM_COMPUTE_NEBATCHNORMALIZATIONLAYER_H
26 
27 #include "arm_compute/core/Types.h"
28 #include "arm_compute/runtime/IFunction.h"
29 #include "arm_compute/runtime/NEON/functions/NEActivationLayer.h"
30 
31 #include <memory>
32 
33 namespace arm_compute
34 {
35 class ITensor;
36 class NEBatchNormalizationLayerKernel;
37 
38 /** Basic function to run @ref NENormalizationLayerKernel and simulate a batch normalization layer.
39  *
40  * Batch normalization is calculated by:
41  * @f[ out_i = \gamma * (\frac{in_i - \mu_{B}}{\sqrt{\sigma^2_{B} + \epsilon}}) + \beta \equiv BN_{\gamma,\beta}(in_i) @f]
42  *
43  */
44 class NEBatchNormalizationLayer : public IFunction
45 {
46 public:
47     /** Constructor */
48     NEBatchNormalizationLayer();
49     /** Prevent instances of this class from being copied (As this class contains pointers) */
50     NEBatchNormalizationLayer(const NEBatchNormalizationLayer &) = delete;
51     /** Prevent instances of this class from being copied (As this class contains pointers) */
52     NEBatchNormalizationLayer &operator=(const NEBatchNormalizationLayer &) = delete;
53     /** Prevent instances of this class from being moved (As this class contains non movable objects) */
54     NEBatchNormalizationLayer(NEBatchNormalizationLayer &&) = delete;
55     /** Prevent instances of this class from being moved (As this class contains non movable objects) */
56     NEBatchNormalizationLayer &operator=(NEBatchNormalizationLayer &&) = delete;
57     /** Default destructor */
58     ~NEBatchNormalizationLayer();
59     /** Set the input and output tensors.
60      *
61      * Valid data layouts:
62      * - NHWC
63      * - NCHW
64      *
65      * Valid data type configurations:
66      * |src            |dst            |
67      * |:--------------|:--------------|
68      * |F32            |F32            |
69      * |F16            |F16            |
70      *
71      * @note If the output tensor is a nullptr or is equal to the input, the batch normalization function will be performed in-place
72      *
73      * @param[in, out] input    Source tensor. In case of @p output tensor = nullptr, this tensor will store the result.
74      *                          3 lower dimensions represent a single input with dimensions [width, height, FM].
75      *                          The rest are optional and used for representing batches. Data types supported: F16/F32.
76      * @param[out]     output   Destination tensor. Output will have the same number of dimensions as input. Data type supported: same as @p input
77      * @param[in]      mean     Mean values tensor. 1 dimension with size equal to the feature maps [FM]. Data types supported: Same as @p input
78      * @param[in]      var      Variance values tensor. 1 dimension with size equal to the feature maps [FM]. Data types supported: Same as @p input
79      * @param[in]      beta     (Optional) Beta values tensor info. 1 dimension with size equal to the feature maps [FM]. If not provided, default value for beta is 0. Data types supported: Same as @p input
80      * @param[in]      gamma    (Optional) Gamma values tensor info. 1 dimension with size equal to the feature maps [FM]. If not provided, default value for gamma is 1. Data types supported: Same as @p input
81      * @param[in]      epsilon  (Optional) Small value to avoid division with zero. Default value is 0.001f.
82      * @param[in]      act_info (Optional) Activation layer information in case of a fused activation. Only RELU, BOUNDED_RELU and LU_BOUNDED_RELU supported.
83      */
84     void configure(ITensor *input, ITensor *output, const ITensor *mean, const ITensor *var, const ITensor *beta = nullptr, const ITensor *gamma = nullptr, float epsilon = 0.001f,
85                    ActivationLayerInfo act_info = ActivationLayerInfo());
86     /** Static function to check if given info will lead to a valid configuration of @ref NEBatchNormalizationLayer
87      *
88      * @param[in] input    Source tensor info. In case of @p output tensor = nullptr, this tensor will store the result.
89      *                     3 lower dimensions represent a single input with dimensions [width, height, FM].
90      *                     The rest are optional and used for representing batches. Data types supported: F16/F32.
91      * @param[in] output   Destination tensor info. Output will have the same number of dimensions as input. Data type supported: same as @p input
92      * @param[in] mean     Mean values tensor info. 1 dimension with size equal to the feature maps [FM]. Data types supported: Same as @p input
93      * @param[in] var      Variance values tensor info. 1 dimension with size equal to the feature maps [FM]. Data types supported: Same as @p input
94      * @param[in] beta     (Optional) Beta values tensor info. 1 dimension with size equal to the feature maps [FM]. If not provided, default value for beta is 0. Data types supported: Same as @p input
95      * @param[in] gamma    (Optional) Gamma values tensor info. 1 dimension with size equal to the feature maps [FM]. If not provided, default value for gamma is 1. Data types supported: Same as @p input
96      * @param[in] epsilon  (Optional) Small value to avoid division with zero. Default value is 0.001f.
97      * @param[in] act_info (Optional) Activation layer information in case of a fused activation. Only RELU, BOUNDED_RELU and LU_BOUNDED_RELU supported.
98      *
99      * @return a status
100      */
101     static Status validate(const ITensorInfo *input, const ITensorInfo *output,
102                            const ITensorInfo *mean, const ITensorInfo *var,
103                            const ITensorInfo *beta = nullptr, const ITensorInfo *gamma = nullptr,
104                            float epsilon = 0.001f, ActivationLayerInfo act_info = ActivationLayerInfo());
105 
106     // Inherited methods overridden:
107     void run() override;
108 
109 private:
110     std::unique_ptr<NEBatchNormalizationLayerKernel> _norm_kernel; /**< Batch normalization layer kernel */
111 };
112 }
113 #endif /* ARM_COMPUTE_NEBATCHNORMALIZATIONLAYER_H */
114