Compute Library
 22.05
CLNormalizationLayer.h
Go to the documentation of this file.
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_CLNORMALIZATIONLAYER_H
25 #define ARM_COMPUTE_CLNORMALIZATIONLAYER_H
26 
27 #include "arm_compute/core/Types.h"
30 
31 #include <memory>
32 
33 namespace arm_compute
34 {
35 class CLCompileContext;
36 class CLFillBorderKernel;
37 class CLNormalizationLayerKernel;
38 class ICLTensor;
39 class ITensorInfo;
40 
41 /** Basic function to compute a normalization layer. This function calls the following CL kernels:
42  *
43  * -# @ref CLFillBorderKernel
44  * -# @ref CLNormalizationLayerKernel
45  *
46  */
48 {
49 public:
50  /** Default constructor */
52  /** Prevent instances of this class from being copied */
54  /** Prevent instances of this class from being copied */
56  /** Prevent instances of this class to be moved */
58  /** Prevent instances of this class to be moved */
60  /** Default destructor */
62  /** Set the input and output tensors.
63  *
64  * Valid data layouts:
65  * - NHWC
66  * - NCHW
67  *
68  * Valid data type configurations:
69  * |src |dst |
70  * |:--------|:---------|
71  * |F32 |F32 |
72  * |F16 |F16 |
73  *
74  * @param[in, out] input Source tensor. 3 lower dims represent a single input with dimensions [width, height, IFM],
75  * and an optional 4th dimension for batch of inputs. Data types supported: F16/F32 (Written to by the border handler).
76  * Data layouts supported: NCHW/NHWC.
77  * @param[out] output Destination tensor. Dimensions, data type and number of channels must match the input ones.
78  * Data types supported: same as @p input. Data layouts supported: same as @p input.
79  * @param[in] norm_info Normalization layer information like the normalization type, normalization size and other parameters.
80  */
81  void configure(ICLTensor *input, ICLTensor *output, const NormalizationLayerInfo &norm_info);
82  /** Set the input and output tensors.
83  *
84  * @param[in] compile_context The compile context to be used.
85  * @param[in, out] input Source tensor. 3 lower dims represent a single input with dimensions [width, height, IFM],
86  * and an optional 4th dimension for batch of inputs. Data types supported: F16/F32 (Written to by the border handler).
87  * Data layouts supported: NCHW/NHWC.
88  * @param[out] output Destination tensor. Dimensions, data type and number of channels must match the input ones.
89  * Data types supported: same as @p input. Data layouts supported: same as @p input.
90  * @param[in] norm_info Normalization layer information like the normalization type, normalization size and other parameters.
91  */
92  void configure(const CLCompileContext &compile_context, ICLTensor *input, ICLTensor *output, const NormalizationLayerInfo &norm_info);
93  /** Static function to check if given info will lead to a valid configuration of @ref CLNormalizationLayer
94  *
95  * @param[in] input Source tensor. 3 lower dims represent a single input with dimensions [width, height, IFM],
96  * and an optional 4th dimension for batch of inputs. Data types supported: F16/F32. Data layouts supported: NCHW/NHWC.
97  * @param[in] output Destination tensor. Dimensions, data type and number of channels must match the input ones.
98  * Data layouts supported: same as @p input.
99  * @param[in] norm_info Normalization layer information like the normalization type, normalization size and other parameters.
100  *
101  * @return a status
102  */
103  static Status validate(const ITensorInfo *input, const ITensorInfo *output, const NormalizationLayerInfo &norm_info);
104 
105  // Inherited methods overridden:
106  void run() override;
107 
108 private:
109  std::unique_ptr<CLNormalizationLayerKernel> _norm_kernel; /**< Normalization layer kernel to run */
110  std::unique_ptr<CLFillBorderKernel> _border_handler; /**< Kernel to handle borders */
111 };
112 } // namespace arm_compute
113 #endif /* ARM_COMPUTE_CLNORMALIZATIONLAYER_H */
Base class for all functions.
Definition: IFunction.h:30
Normalization Layer Information class.
Definition: Types.h:1726
CLNormalizationLayer()
Default constructor.
Store the tensor&#39;s metadata.
Definition: ITensorInfo.h:40
Status class.
Definition: Error.h:52
void configure(ICLTensor *input, ICLTensor *output, const NormalizationLayerInfo &norm_info)
Set the input and output tensors.
~CLNormalizationLayer()
Default destructor.
Copyright (c) 2017-2022 Arm Limited.
CLNormalizationLayer & operator=(const CLNormalizationLayer &)=delete
Prevent instances of this class from being copied.
void run() override
Run the kernels contained in the function.
Basic function to compute a normalization layer.
CLCompileContext class.
Interface for OpenCL tensor.
Definition: ICLTensor.h:42
static Status validate(const ITensorInfo *input, const ITensorInfo *output, const NormalizationLayerInfo &norm_info)
Static function to check if given info will lead to a valid configuration of CLNormalizationLayer.