Gitiles
Code Review Sign In
review.mlplatform.org / ml / armnn / 2e5d0b2e2a212ceb803681b717cbaf821f5e0929 / . / docs / 01_02_deserializer_serializer.dox
blob: 5d4dc43a74ee90d6773570538588defcb9b86e50 [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
Simon Obute51f67772021-09-03 15:50:13 +0100[diff] [blame]28- ChannelShuffle
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]29- Comparison
30- Concat
31- Constant
32- Convolution2d
Matthew Sloyanb63a3112021-09-08 13:05:51 +0100[diff] [blame]33- Convolution3d
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]34- DepthToSpace
35- DepthwiseConvolution2d
36- Dequantize
37- DetectionPostProcess
38- Division
39- ElementwiseUnary
Jan Eilers1ed388f2021-01-28 14:45:09 +0000[diff] [blame]40- Fill
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]41- Floor
42- FullyConnected
43- Gather
44- Input
45- InstanceNormalization
46- L2Normalization
James Ward590f3f72020-12-02 10:20:48 +0000[diff] [blame]47- Logical
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]48- LogSoftmax
49- Lstm
50- Maximum
51- Mean
52- Merge
53- Minimum
54- Multiplication
55- Normalization
56- Output
Matthew Sloyan2e5d0b22021-10-21 14:05:31 +0100[diff] [blame^]57- Pad (Constant, Symmetric, Reflect)
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]58- Permute
59- Pooling2d
60- Prelu
James Ward590f3f72020-12-02 10:20:48 +0000[diff] [blame]61- QLstm
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]62- Quantize
63- QuantizedLstm
James Ward590f3f72020-12-02 10:20:48 +0000[diff] [blame]64- Rank
Jan Eilers1121c6b2021-02-19 09:38:03 +0000[diff] [blame]65- Reduce
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]66- Reshape
67- Resize
Keith Davis3ae3f972021-05-21 16:33:48 +0100[diff] [blame]68- Shape
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]69- Slice
70- Softmax
71- SpaceToBatchNd
72- SpaceToDepth
73- Splitter
74- Stack
75- StandIn
76- StridedSlice
77- Subtraction
78- Switch
James Ward590f3f72020-12-02 10:20:48 +0000[diff] [blame]79- Transpose
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]80- TransposeConvolution2d
Narumol Prangnawarata0162e12021-07-23 14:47:49 +0100[diff] [blame]81- UnidirectionalSequenceLstm
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]82
83More machine learning layers will be supported in future releases.
84
Jan Eilers1ed388f2021-01-28 14:45:09 +0000[diff] [blame]85@subsection serializersupportdeprecated Deprecated layers
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]86
87Some 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:
88
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]89- Abs will deserialize as ElementwiseUnary
Jan Eilers1ed388f2021-01-28 14:45:09 +0000[diff] [blame]90- Equal will deserialize as Comparison
91- Greater will deserialize as Comparison
92- Merger will deserialize as Concat
93- ResizeBilinear will deserialize as Resize
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]94- Rsqrt will deserialize as ElementwiseUnary
95<br/><br/><br/><br/>
96
Jan Eilers1ed388f2021-01-28 14:45:09 +0000[diff] [blame]97@page deserializer Deserializer
98@tableofcontents
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]99
100The `armnnDeserializer` is a library for loading neural networks defined by Arm NN FlatBuffers files
101into the Arm NN runtime.
102
Jan Eilers1ed388f2021-01-28 14:45:09 +0000[diff] [blame]103@section deserializersupport Supported Layers
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]104
Jan Eilers1ed388f2021-01-28 14:45:09 +0000[diff] [blame]105This 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]106
Jan Eilers1ed388f2021-01-28 14:45:09 +0000[diff] [blame]107@subsection deserializersupportfully Fully supported
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]108
109The Arm NN SDK Deserialize parser currently supports the following layers:
110
111- Abs
112- Activation
113- Addition
114- ArgMinMax
115- BatchToSpaceNd
116- BatchNormalization
mathad01b392e982021-04-07 12:07:30 +0100[diff] [blame]117- Cast
Simon Obute51f67772021-09-03 15:50:13 +0100[diff] [blame]118- ChannelShuffle
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]119- Concat
120- Comparison
121- Constant
122- Convolution2d
123- DepthToSpace
124- DepthwiseConvolution2d
125- Dequantize
126- DetectionPostProcess
127- Division
James Ward590f3f72020-12-02 10:20:48 +0000[diff] [blame]128- ElementwiseUnary
129- Fill
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]130- Floor
131- FullyConnected
132- Gather
133- Input
134- InstanceNormalization
135- L2Normalization
James Ward590f3f72020-12-02 10:20:48 +0000[diff] [blame]136- Logical
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]137- LogSoftmax
138- Lstm
139- Maximum
140- Mean
141- Merge
142- Minimum
143- Multiplication
144- Normalization
145- Output
146- Pad
147- Permute
148- Pooling2d
149- Prelu
150- Quantize
Jan Eilers1ed388f2021-01-28 14:45:09 +0000[diff] [blame]151- QLstm
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]152- QuantizedLstm
James Ward590f3f72020-12-02 10:20:48 +0000[diff] [blame]153- Rank
Jan Eilers1121c6b2021-02-19 09:38:03 +0000[diff] [blame]154- Reduce
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]155- Reshape
James Ward590f3f72020-12-02 10:20:48 +0000[diff] [blame]156- Resize
Jan Eilers1ed388f2021-01-28 14:45:09 +0000[diff] [blame]157- ResizeBilinear
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]158- Slice
159- Softmax
160- SpaceToBatchNd
161- SpaceToDepth
162- Splitter
163- Stack
164- StandIn
165- StridedSlice
166- Subtraction
167- Switch
Jan Eilers1121c6b2021-02-19 09:38:03 +0000[diff] [blame]168- Transpose
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]169- TransposeConvolution2d
Narumol Prangnawarata0162e12021-07-23 14:47:49 +0100[diff] [blame]170- UnidirectionalSequenceLstm
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]171
172More machine learning layers will be supported in future releases.
173
Jan Eilers1ed388f2021-01-28 14:45:09 +0000[diff] [blame]174@subsection deserializersupportdeprecated Deprecated layers
Ryan OSheaf3a43232020-02-12 16:15:27 +0000[diff] [blame]175
176Some 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:
177
178- Equal will deserialize as Comparison
179- Merger will deserialize as Concat
180- Greater will deserialize as Comparison
181- ResizeBilinear will deserialize as Resize
182
183**/
184}