Gitiles
Code Review Sign In
review.mlplatform.org / ml / armnn / 36eed46563a2a254e674865eb8ede62a1558c997 / . / Documentation / _tensor_flow_lite_support_8md_source.html
blob: 63bc427fa6479d7dcde373d8c73b9f33b7af6751 [file] [log] [blame]
Ryan OSheade36e4a2020-03-13 16:26:19 +0000[diff] [blame]1<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
2<html xmlns="http://www.w3.org/1999/xhtml">
3<head>
4<meta http-equiv="Content-Type" content="text/xhtml;charset=UTF-8"/>
5<meta http-equiv="X-UA-Compatible" content="IE=9"/>
6<meta name="generator" content="Doxygen 1.8.13"/>
7<meta name="viewport" content="width=device-width, initial-scale=1"/>
8<title>ArmNN: src/armnnTfLiteParser/TensorFlowLiteSupport.md Source File</title>
9<link href="tabs.css" rel="stylesheet" type="text/css"/>
10<script type="text/javascript" src="jquery.js"></script>
11<script type="text/javascript" src="dynsections.js"></script>
12<link href="navtree.css" rel="stylesheet" type="text/css"/>
13<script type="text/javascript" src="resize.js"></script>
14<script type="text/javascript" src="navtreedata.js"></script>
15<script type="text/javascript" src="navtree.js"></script>
16<script type="text/javascript">
17 $(document).ready(initResizable);
18</script>
19<link href="search/search.css" rel="stylesheet" type="text/css"/>
20<script type="text/javascript" src="search/searchdata.js"></script>
21<script type="text/javascript" src="search/search.js"></script>
22<link href="doxygen.css" rel="stylesheet" type="text/css" />
23</head>
24<body>
25<div id="top"><!-- do not remove this div, it is closed by doxygen! -->
26<div id="titlearea">
27<table cellspacing="0" cellpadding="0">
28 <tbody>
29 <tr style="height: 56px;">
30 <td id="projectalign" style="padding-left: 0.5em;">
31 <div id="projectname">ArmNN
32 &#160;<span id="projectnumber">NotReleased</span>
33 </div>
34 </td>
35 </tr>
36 </tbody>
37</table>
38</div>
39<!-- end header part -->
40<!-- Generated by Doxygen 1.8.13 -->
41<script type="text/javascript">
42var searchBox = new SearchBox("searchBox", "search",false,'Search');
43</script>
44<script type="text/javascript" src="menudata.js"></script>
45<script type="text/javascript" src="menu.js"></script>
46<script type="text/javascript">
47$(function() {
48 initMenu('',true,false,'search.php','Search');
49 $(document).ready(function() { init_search(); });
50});
51</script>
52<div id="main-nav"></div>
53</div><!-- top -->
54<div id="side-nav" class="ui-resizable side-nav-resizable">
55 <div id="nav-tree">
56 <div id="nav-tree-contents">
57 <div id="nav-sync" class="sync"></div>
58 </div>
59 </div>
60 <div id="splitbar" style="-moz-user-select:none;"
61 class="ui-resizable-handle">
62 </div>
63</div>
64<script type="text/javascript">
65$(document).ready(function(){initNavTree('_tensor_flow_lite_support_8md.html','');});
66</script>
67<div id="doc-content">
68<!-- window showing the filter options -->
69<div id="MSearchSelectWindow"
70 onmouseover="return searchBox.OnSearchSelectShow()"
71 onmouseout="return searchBox.OnSearchSelectHide()"
72 onkeydown="return searchBox.OnSearchSelectKey(event)">
73</div>
74
75<!-- iframe showing the search results (closed by default) -->
76<div id="MSearchResultsWindow">
77<iframe src="javascript:void(0)" frameborder="0"
78 name="MSearchResults" id="MSearchResults">
79</iframe>
80</div>
81
82<div class="header">
83 <div class="headertitle">
84<div class="title">src/armnnTfLiteParser/TensorFlowLiteSupport.md</div> </div>
85</div><!--header-->
86<div class="contents">
87<a href="_tensor_flow_lite_support_8md.html">Go to the documentation of this file.</a><div class="fragment"><div class="line"><a name="l00001"></a><span class="lineno"> 1</span>&#160;# TensorFlow Lite operators that the Arm NN SDK supports</div><div class="line"><a name="l00002"></a><span class="lineno"> 2</span>&#160;</div><div class="line"><a name="l00003"></a><span class="lineno"> 3</span>&#160;This reference guide provides a list of TensorFlow Lite operators the Arm NN SDK currently supports.</div><div class="line"><a name="l00004"></a><span class="lineno"> 4</span>&#160;</div><div class="line"><a name="l00005"></a><span class="lineno"> 5</span>&#160;## Fully supported</div><div class="line"><a name="l00006"></a><span class="lineno"> 6</span>&#160;</div><div class="line"><a name="l00007"></a><span class="lineno"> 7</span>&#160;The Arm NN SDK TensorFlow Lite parser currently supports the following operators:</div><div class="line"><a name="l00008"></a><span class="lineno"> 8</span>&#160;</div><div class="line"><a name="l00009"></a><span class="lineno"> 9</span>&#160;* ADD</div><div class="line"><a name="l00010"></a><span class="lineno"> 10</span>&#160;</div><div class="line"><a name="l00011"></a><span class="lineno"> 11</span>&#160;* AVERAGE_POOL_2D, Supported Fused Activation: RELU , RELU6 , TANH, NONE</div><div class="line"><a name="l00012"></a><span class="lineno"> 12</span>&#160;</div><div class="line"><a name="l00013"></a><span class="lineno"> 13</span>&#160;* BATCH_TO_SPACE</div><div class="line"><a name="l00014"></a><span class="lineno"> 14</span>&#160;</div><div class="line"><a name="l00015"></a><span class="lineno"> 15</span>&#160;* CONCATENATION, Supported Fused Activation: RELU , RELU6 , TANH, NONE</div><div class="line"><a name="l00016"></a><span class="lineno"> 16</span>&#160;</div><div class="line"><a name="l00017"></a><span class="lineno"> 17</span>&#160;* CONV_2D, Supported Fused Activation: RELU , RELU6 , TANH, NONE</div><div class="line"><a name="l00018"></a><span class="lineno"> 18</span>&#160;</div><div class="line"><a name="l00019"></a><span class="lineno"> 19</span>&#160;* DEPTHWISE_CONV_2D, Supported Fused Activation: RELU , RELU6 , TANH, NONE</div><div class="line"><a name="l00020"></a><span class="lineno"> 20</span>&#160;</div><div class="line"><a name="l00021"></a><span class="lineno"> 21</span>&#160;* DEQUANTIZE</div><div class="line"><a name="l00022"></a><span class="lineno"> 22</span>&#160;</div><div class="line"><a name="l00023"></a><span class="lineno"> 23</span>&#160;* FULLY_CONNECTED, Supported Fused Activation: RELU , RELU6 , TANH, NONE</div><div class="line"><a name="l00024"></a><span class="lineno"> 24</span>&#160;</div><div class="line"><a name="l00025"></a><span class="lineno"> 25</span>&#160;* LOGISTIC</div><div class="line"><a name="l00026"></a><span class="lineno"> 26</span>&#160;</div><div class="line"><a name="l00027"></a><span class="lineno"> 27</span>&#160;* L2_NORMALIZATION</div><div class="line"><a name="l00028"></a><span class="lineno"> 28</span>&#160;</div><div class="line"><a name="l00029"></a><span class="lineno"> 29</span>&#160;* MAX_POOL_2D, Supported Fused Activation: RELU , RELU6 , TANH, NONE</div><div class="line"><a name="l00030"></a><span class="lineno"> 30</span>&#160;</div><div class="line"><a name="l00031"></a><span class="lineno"> 31</span>&#160;* MAXIMUM</div><div class="line"><a name="l00032"></a><span class="lineno"> 32</span>&#160;</div><div class="line"><a name="l00033"></a><span class="lineno"> 33</span>&#160;* MEAN</div><div class="line"><a name="l00034"></a><span class="lineno"> 34</span>&#160;</div><div class="line"><a name="l00035"></a><span class="lineno"> 35</span>&#160;* MINIMUM</div><div class="line"><a name="l00036"></a><span class="lineno"> 36</span>&#160;</div><div class="line"><a name="l00037"></a><span class="lineno"> 37</span>&#160;* MUL</div><div class="line"><a name="l00038"></a><span class="lineno"> 38</span>&#160;</div><div class="line"><a name="l00039"></a><span class="lineno"> 39</span>&#160;* PACK</div><div class="line"><a name="l00040"></a><span class="lineno"> 40</span>&#160;</div><div class="line"><a name="l00041"></a><span class="lineno"> 41</span>&#160;* PAD</div><div class="line"><a name="l00042"></a><span class="lineno"> 42</span>&#160;</div><div class="line"><a name="l00043"></a><span class="lineno"> 43</span>&#160;* QUANTIZE</div><div class="line"><a name="l00044"></a><span class="lineno"> 44</span>&#160;</div><div class="line"><a name="l00045"></a><span class="lineno"> 45</span>&#160;* RELU</div><div class="line"><a name="l00046"></a><span class="lineno"> 46</span>&#160;</div><div class="line"><a name="l00047"></a><span class="lineno"> 47</span>&#160;* RELU6</div><div class="line"><a name="l00048"></a><span class="lineno"> 48</span>&#160;</div><div class="line"><a name="l00049"></a><span class="lineno"> 49</span>&#160;* RESHAPE</div><div class="line"><a name="l00050"></a><span class="lineno"> 50</span>&#160;</div><div class="line"><a name="l00051"></a><span class="lineno"> 51</span>&#160;* RESIZE_BILINEAR</div><div class="line"><a name="l00052"></a><span class="lineno"> 52</span>&#160;</div><div class="line"><a name="l00053"></a><span class="lineno"> 53</span>&#160;* RESIZE_NEAREST_NEIGHBOR</div><div class="line"><a name="l00054"></a><span class="lineno"> 54</span>&#160;</div><div class="line"><a name="l00055"></a><span class="lineno"> 55</span>&#160;* SLICE</div><div class="line"><a name="l00056"></a><span class="lineno"> 56</span>&#160;</div><div class="line"><a name="l00057"></a><span class="lineno"> 57</span>&#160;* SOFTMAX</div><div class="line"><a name="l00058"></a><span class="lineno"> 58</span>&#160;</div><div class="line"><a name="l00059"></a><span class="lineno"> 59</span>&#160;* SPACE_TO_BATCH</div><div class="line"><a name="l00060"></a><span class="lineno"> 60</span>&#160;</div><div class="line"><a name="l00061"></a><span class="lineno"> 61</span>&#160;* SPLIT</div><div class="line"><a name="l00062"></a><span class="lineno"> 62</span>&#160;</div><div class="line"><a name="l00063"></a><span class="lineno"> 63</span>&#160;* SQUEEZE</div><div class="line"><a name="l00064"></a><span class="lineno"> 64</span>&#160;</div><div class="line"><a name="l00065"></a><span class="lineno"> 65</span>&#160;* STRIDED_SLICE</div><div class="line"><a name="l00066"></a><span class="lineno"> 66</span>&#160;</div><div class="line"><a name="l00067"></a><span class="lineno"> 67</span>&#160;* SUB</div><div class="line"><a name="l00068"></a><span class="lineno"> 68</span>&#160;</div><div class="line"><a name="l00069"></a><span class="lineno"> 69</span>&#160;* TANH</div><div class="line"><a name="l00070"></a><span class="lineno"> 70</span>&#160;</div><div class="line"><a name="l00071"></a><span class="lineno"> 71</span>&#160;* TRANSPOSE</div><div class="line"><a name="l00072"></a><span class="lineno"> 72</span>&#160;</div><div class="line"><a name="l00073"></a><span class="lineno"> 73</span>&#160;* TRANSPOSE_CONV</div><div class="line"><a name="l00074"></a><span class="lineno"> 74</span>&#160;</div><div class="line"><a name="l00075"></a><span class="lineno"> 75</span>&#160;* UNPACK</div><div class="line"><a name="l00076"></a><span class="lineno"> 76</span>&#160;</div><div class="line"><a name="l00077"></a><span class="lineno"> 77</span>&#160;## Custom Operator</div><div class="line"><a name="l00078"></a><span class="lineno"> 78</span>&#160;</div><div class="line"><a name="l00079"></a><span class="lineno"> 79</span>&#160;* TFLite_Detection_PostProcess</div><div class="line"><a name="l00080"></a><span class="lineno"> 80</span>&#160;</div><div class="line"><a name="l00081"></a><span class="lineno"> 81</span>&#160;## Tested networks</div><div class="line"><a name="l00082"></a><span class="lineno"> 82</span>&#160;</div><div class="line"><a name="l00083"></a><span class="lineno"> 83</span>&#160;Arm tested these operators with the following TensorFlow Lite neural network:</div><div class="line"><a name="l00084"></a><span class="lineno"> 84</span>&#160;</div><div class="line"><a name="l00085"></a><span class="lineno"> 85</span>&#160;* [Quantized MobileNet](http://download.tensorflow.org/models/mobilenet_v1_2018_02_22/mobilenet_v1_1.0_224_quant.tgz)</div><div class="line"><a name="l00086"></a><span class="lineno"> 86</span>&#160;</div><div class="line"><a name="l00087"></a><span class="lineno"> 87</span>&#160;* [Quantized SSD MobileNet](http://download.tensorflow.org/models/object_detection/ssd_mobilenet_v1_quantized_300x300_coco14_sync_2018_07_18.tar.gz)</div><div class="line"><a name="l00088"></a><span class="lineno"> 88</span>&#160;</div><div class="line"><a name="l00089"></a><span class="lineno"> 89</span>&#160;* DeepSpeech v1 converted from [TensorFlow model](https://github.com/mozilla/DeepSpeech/releases/tag/v0.4.1)</div><div class="line"><a name="l00090"></a><span class="lineno"> 90</span>&#160;</div><div class="line"><a name="l00091"></a><span class="lineno"> 91</span>&#160;* DeepSpeaker</div><div class="line"><a name="l00092"></a><span class="lineno"> 92</span>&#160;</div><div class="line"><a name="l00093"></a><span class="lineno"> 93</span>&#160;More machine learning operators will be supported in future releases.</div></div><!-- fragment --></div><!-- contents -->
88</div><!-- doc-content -->
89<!-- start footer part -->
90<div id="nav-path" class="navpath"><!-- id is needed for treeview function! -->
91 <ul>
92 <li class="navelem"><a class="el" href="_tensor_flow_lite_support_8md.html">TensorFlowLiteSupport.md</a></li>
93 <li class="footer">Generated on Fri Mar 13 2020 16:06:55 for ArmNN by
94 <a href="http://www.doxygen.org/index.html">
95 <img class="footer" src="doxygen.png" alt="doxygen"/></a> 1.8.13 </li>
96 </ul>
97</div>
98</body>
99</html>