xref: /aosp_15_r20/external/ComputeLibrary/src/cpu/kernels/fuse_batch_normalization/nhwc/neon/fp16.cpp (revision c217d954acce2dbc11938adb493fc0abd69584f3)
1 /*
2  * Copyright (c) 2021-2022 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 #if defined(__ARM_FEATURE_FP16_VECTOR_ARITHMETIC) && defined(ENABLE_FP16_KERNELS)
25 
26 #include "src/cpu/kernels/fuse_batch_normalization/generic/impl.h"
27 #include "src/cpu/kernels/fuse_batch_normalization/nhwc/neon/impl.h"
28 
29 namespace arm_compute
30 {
31 namespace cpu
32 {
fused_batch_normalization_dwc_nhwc_f16(const ITensor * dwc_weights,const ITensor * dwc_bias,ITensor * fused_weights,ITensor * fused_bias,const ITensor * bn_mean,const ITensor * bn_var,const ITensor * bn_beta,const ITensor * bn_gamma,float epsilon,const Window & window)33 void fused_batch_normalization_dwc_nhwc_f16(const ITensor *dwc_weights, const ITensor *dwc_bias, ITensor *fused_weights, ITensor *fused_bias,
34                                             const ITensor *bn_mean, const ITensor *bn_var, const ITensor *bn_beta, const ITensor *bn_gamma, float epsilon, const Window &window)
35 {
36     return fused_batch_normalization_dwc_nhwc<float16_t>(dwc_weights, dwc_bias, fused_weights, fused_bias,
37                                                          bn_mean, bn_var, bn_beta, bn_gamma, epsilon, window);
38 }
39 
40 } // namespace cpu
41 } // namespace arm_compute
42 #endif /* defined(__ARM_FEATURE_FP16_VECTOR_ARITHMETIC) && defined(ENABLE_FP16_KERNELS) */
43