1 /// Copyright (c) 2021 ARM Limited and Contributors. All rights reserved.
3 /// SPDX-License-Identifier: MIT
9 @page serializer Serializer
12 The `armnnSerializer` is a library for serializing an Arm NN network to a stream.
14 @section serializersupport Supported Layers
16 This reference guide provides a list of layers which can be serialized by the Arm NN SDK.
18 @subsection serializersupportflully Fully supported
20 The Arm NN SDK Serializer currently supports the following layers:
32 - DepthwiseConvolution2d
34 - DetectionPostProcess
42 - InstanceNormalization
75 - TransposeConvolution2d
77 More machine learning layers will be supported in future releases.
79 @subsection serializersupportdeprecated Deprecated layers
81 Some layers have been deprecated and replaced by others layers. In order to maintain backward compatibility, serializations of these deprecated layers will deserialize to the layers that have replaced them, as follows:
83 - Abs will deserialize as ElementwiseUnary
84 - Equal will deserialize as Comparison
85 - Greater will deserialize as Comparison
86 - Merger will deserialize as Concat
87 - ResizeBilinear will deserialize as Resize
88 - Rsqrt will deserialize as ElementwiseUnary
91 @page deserializer Deserializer
94 The `armnnDeserializer` is a library for loading neural networks defined by Arm NN FlatBuffers files
95 into the Arm NN runtime.
97 @section deserializersupport Supported Layers
99 This reference guide provides a list of layers which can be deserialized by the Arm NN SDK.
101 @subsection deserializersupportfully Fully supported
103 The Arm NN SDK Deserialize parser currently supports the following layers:
116 - DepthwiseConvolution2d
118 - DetectionPostProcess
126 - InstanceNormalization
160 - TransposeConvolution2d
162 More machine learning layers will be supported in future releases.
164 @subsection deserializersupportdeprecated Deprecated layers
166 Some layers have been deprecated and replaced by others layers. In order to maintain backward compatibility, serializations of these deprecated layers will deserialize to the layers that have replaced them, as follows:
168 - Equal will deserialize as Comparison
169 - Merger will deserialize as Concat
170 - Greater will deserialize as Comparison
171 - ResizeBilinear will deserialize as Resize