Gitiles
Code Review Sign In
review.mlplatform.org / ml / armnn / c6ab02a626e15b4a12fc09ecd844eb8b95380c3c / . / docs / 01_02_deserializer_serializer.dox
blob: 1df3999d0a4058105d343571b189a5dc55ecb91a [file] [log] [blame]
Jan Eilers31a7c892021-01-29 14:18:13 +0000[diff] [blame]1/// Copyright (c) 2021 ARM Limited and Contributors. All rights reserved.
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]2///
3/// SPDX-License-Identifier: MIT
4///
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]5
6namespace armnn
7{
8/**
Jan Eilers1ed388f2021-01-28 14:45:09 +0000[diff] [blame]9@page serializer Serializer
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]10@tableofcontents
11
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]12The `armnnSerializer` is a library for serializing an Arm NN network to a stream.
13
Jan Eilers1ed388f2021-01-28 14:45:09 +0000[diff] [blame]14@section serializersupport Supported Layers
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]15
Jan Eilers1ed388f2021-01-28 14:45:09 +0000[diff] [blame]16This reference guide provides a list of layers which can be serialized by the Arm NN SDK.
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]17
Jan Eilers1ed388f2021-01-28 14:45:09 +0000[diff] [blame]18@subsection serializersupportflully Fully supported
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]19
20The Arm NN SDK Serializer currently supports the following layers:
21
22- Activation
23- Addition
24- ArgMinMax
25- BatchToSpaceNd
26- BatchNormalization
mathad01b392e982021-04-07 12:07:30 +0100[diff] [blame]27- Cast
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]28- Comparison
29- Concat
30- Constant
31- Convolution2d
32- DepthToSpace
33- DepthwiseConvolution2d
34- Dequantize
35- DetectionPostProcess
36- Division
37- ElementwiseUnary
Jan Eilers1ed388f2021-01-28 14:45:09 +0000[diff] [blame]38- Fill
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]39- Floor
40- FullyConnected
41- Gather
42- Input
43- InstanceNormalization
44- L2Normalization
James Ward590f3f72020-12-02 10:20:48 +0000[diff] [blame]45- Logical
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]46- LogSoftmax
47- Lstm
48- Maximum
49- Mean
50- Merge
51- Minimum
52- Multiplication
53- Normalization
54- Output
55- Pad
56- Permute
57- Pooling2d
58- Prelu
James Ward590f3f72020-12-02 10:20:48 +0000[diff] [blame]59- QLstm
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]60- Quantize
61- QuantizedLstm
James Ward590f3f72020-12-02 10:20:48 +0000[diff] [blame]62- Rank
Jan Eilers1121c6b2021-02-19 09:38:03 +0000[diff] [blame]63- Reduce
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]64- Reshape
65- Resize
66- Slice
67- Softmax
68- SpaceToBatchNd
69- SpaceToDepth
70- Splitter
71- Stack
72- StandIn
73- StridedSlice
74- Subtraction
75- Switch
James Ward590f3f72020-12-02 10:20:48 +0000[diff] [blame]76- Transpose
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]77- TransposeConvolution2d
78
79More machine learning layers will be supported in future releases.
80
Jan Eilers1ed388f2021-01-28 14:45:09 +0000[diff] [blame]81@subsection serializersupportdeprecated Deprecated layers
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]82
83Some 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:
84
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]85- Abs will deserialize as ElementwiseUnary
Jan Eilers1ed388f2021-01-28 14:45:09 +0000[diff] [blame]86- Equal will deserialize as Comparison
87- Greater will deserialize as Comparison
88- Merger will deserialize as Concat
89- ResizeBilinear will deserialize as Resize
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]90- Rsqrt will deserialize as ElementwiseUnary
91<br/><br/><br/><br/>
92
Jan Eilers1ed388f2021-01-28 14:45:09 +0000[diff] [blame]93@page deserializer Deserializer
94@tableofcontents
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]95
96The `armnnDeserializer` is a library for loading neural networks defined by Arm NN FlatBuffers files
97into the Arm NN runtime.
98
Jan Eilers1ed388f2021-01-28 14:45:09 +0000[diff] [blame]99@section deserializersupport Supported Layers
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]100
Jan Eilers1ed388f2021-01-28 14:45:09 +0000[diff] [blame]101This reference guide provides a list of layers which can be deserialized by the Arm NN SDK.
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]102
Jan Eilers1ed388f2021-01-28 14:45:09 +0000[diff] [blame]103@subsection deserializersupportfully Fully supported
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]104
105The Arm NN SDK Deserialize parser currently supports the following layers:
106
107- Abs
108- Activation
109- Addition
110- ArgMinMax
111- BatchToSpaceNd
112- BatchNormalization
mathad01b392e982021-04-07 12:07:30 +0100[diff] [blame]113- Cast
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]114- Concat
115- Comparison
116- Constant
117- Convolution2d
118- DepthToSpace
119- DepthwiseConvolution2d
120- Dequantize
121- DetectionPostProcess
122- Division
James Ward590f3f72020-12-02 10:20:48 +0000[diff] [blame]123- ElementwiseUnary
124- Fill
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]125- Floor
126- FullyConnected
127- Gather
128- Input
129- InstanceNormalization
130- L2Normalization
James Ward590f3f72020-12-02 10:20:48 +0000[diff] [blame]131- Logical
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]132- LogSoftmax
133- Lstm
134- Maximum
135- Mean
136- Merge
137- Minimum
138- Multiplication
139- Normalization
140- Output
141- Pad
142- Permute
143- Pooling2d
144- Prelu
145- Quantize
Jan Eilers1ed388f2021-01-28 14:45:09 +0000[diff] [blame]146- QLstm
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]147- QuantizedLstm
James Ward590f3f72020-12-02 10:20:48 +0000[diff] [blame]148- Rank
Jan Eilers1121c6b2021-02-19 09:38:03 +0000[diff] [blame]149- Reduce
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]150- Reshape
James Ward590f3f72020-12-02 10:20:48 +0000[diff] [blame]151- Resize
Jan Eilers1ed388f2021-01-28 14:45:09 +0000[diff] [blame]152- ResizeBilinear
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]153- Slice
154- Softmax
155- SpaceToBatchNd
156- SpaceToDepth
157- Splitter
158- Stack
159- StandIn
160- StridedSlice
161- Subtraction
162- Switch
Jan Eilers1121c6b2021-02-19 09:38:03 +0000[diff] [blame]163- Transpose
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]164- TransposeConvolution2d
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]165
166More machine learning layers will be supported in future releases.
167
Jan Eilers1ed388f2021-01-28 14:45:09 +0000[diff] [blame]168@subsection deserializersupportdeprecated Deprecated layers
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]169
170Some 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:
171
172- Equal will deserialize as Comparison
173- Merger will deserialize as Concat
174- Greater will deserialize as Comparison
175- ResizeBilinear will deserialize as Resize
176
177**/
178}