blob: 42cfc06fc45f82cf94c3ea3a886316e1242c27b6 [file] [log] [blame]
Anthony Barbier6ff3b192017-09-04 18:44:23 +01001/*
2 * Copyright (c) 2017 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_CLSOFTMAXLAYER_H__
25#define __ARM_COMPUTE_CLSOFTMAXLAYER_H__
26
27#include "arm_compute/core/CL/kernels/CLSoftmaxLayerKernel.h"
28#include "arm_compute/runtime/CL/CLTensor.h"
29#include "arm_compute/runtime/IFunction.h"
30
31namespace arm_compute
32{
33class ICLTensor;
34
35/** Basic function to compute a SoftmaxLayer.
36 *
37 * Softmax is calculated by :
38 * @f[ out = exp(x - max(x)) / sum(exp(x - max(x))) @f]
39 *
40 * This function runs the following kernels:
41 * -# @ref CLLogits1DMaxKernel
42 * -# @ref CLLogits1DShiftExpSumKernel
43 * -# @ref CLLogits1DNormKernel
44 */
45class CLSoftmaxLayer : public IFunction
46{
47public:
48 /** Constructor */
49 CLSoftmaxLayer();
50 /** Set the input and output tensors.
51 *
52 * @param[in] input Source tensor. Data types supported: F16, F32. Number of channels must be 1.
53 * @param[out] output Destination tensor. Matching input type and channel number.
54 */
55 void configure(const ICLTensor *input, ICLTensor *output);
56
57 // Inherited methods overridden:
58 void run() override;
59
60private:
61 CLLogits1DMaxKernel _max_kernel;
62 CLLogits1DShiftExpSumKernel _shift_exp_sum_kernel;
63 CLLogits1DNormKernel _norm_kernel;
64 CLTensor _max;
65 CLTensor _sum;
66 CLTensor _tmp;
67};
68}
69#endif /* __ARM_COMPUTE_CLSOFTMAXLAYER_H__ */