-
Notifications
You must be signed in to change notification settings - Fork 1
/
export_ONNX.html
336 lines (223 loc) · 15.1 KB
/
export_ONNX.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
<!DOCTYPE html>
<html class="writer-html5" lang="en" >
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>Export: ONNX — N2D2 documentation</title>
<link rel="stylesheet" href="_static/css/theme.css" type="text/css" />
<link rel="stylesheet" href="_static/pygments.css" type="text/css" />
<!--[if lt IE 9]>
<script src="_static/js/html5shiv.min.js"></script>
<![endif]-->
<script type="text/javascript" id="documentation_options" data-url_root="./" src="_static/documentation_options.js"></script>
<script src="_static/jquery.js"></script>
<script src="_static/underscore.js"></script>
<script src="_static/doctools.js"></script>
<script src="_static/language_data.js"></script>
<script async="async" src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.7/latest.js?config=TeX-AMS-MML_HTMLorMML"></script>
<script type="text/javascript" src="_static/js/theme.js"></script>
<link rel="index" title="Index" href="genindex.html" />
<link rel="search" title="Search" href="search.html" />
</head>
<body class="wy-body-for-nav">
<div class="wy-grid-for-nav">
<nav data-toggle="wy-nav-shift" class="wy-nav-side">
<div class="wy-side-scroll">
<div class="wy-side-nav-search" >
<a href="index.html" class="icon icon-home" alt="Documentation Home"> N2D2
</a>
<div role="search">
<form id="rtd-search-form" class="wy-form" action="search.html" method="get">
<input type="text" name="q" placeholder="Search docs" />
<input type="hidden" name="check_keywords" value="yes" />
<input type="hidden" name="area" value="default" />
</form>
</div>
</div>
<div class="wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
<p class="caption"><span class="caption-text">Introduction:</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="intro/intro.html">Presentation</a></li>
<li class="toctree-l1"><a class="reference internal" href="intro/about.html">About N2D2-IP</a></li>
<li class="toctree-l1"><a class="reference internal" href="intro/simus.html">Performing simulations</a></li>
<li class="toctree-l1"><a class="reference internal" href="intro/perfs_tools.html">Performance evaluation tools</a></li>
<li class="toctree-l1"><a class="reference internal" href="intro/tuto.html">Tutorials</a></li>
</ul>
<p class="caption"><span class="caption-text">ONNX Import:</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="onnx/convert.html">Obtain ONNX models</a></li>
<li class="toctree-l1"><a class="reference internal" href="onnx/import.html">Import ONNX models</a></li>
<li class="toctree-l1"><a class="reference internal" href="onnx/transfer.html">Train from ONNX models</a></li>
</ul>
<p class="caption"><span class="caption-text">Quantization and Export:</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="quant/post.html">Post-training quantization</a></li>
<li class="toctree-l1"><a class="reference internal" href="quant/qat.html">[NEW] Quantization-Aware Training</a></li>
<li class="toctree-l1"><a class="reference internal" href="export/CPP.html">Export: C++</a></li>
<li class="toctree-l1"><a class="reference internal" href="export/CPP_STM32.html">Export: C++/STM32</a></li>
<li class="toctree-l1"><a class="reference internal" href="export/TensorRT.html">Export: TensorRT</a></li>
<li class="toctree-l1"><a class="reference internal" href="export/DNeuro.html">Export: DNeuro</a></li>
<li class="toctree-l1"><a class="reference internal" href="export/ONNX.html">Export: ONNX</a></li>
<li class="toctree-l1"><a class="reference internal" href="export/legacy.html">Export: other / legacy</a></li>
</ul>
<p class="caption"><span class="caption-text">INI File Interface:</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="ini/intro.html">Introduction</a></li>
<li class="toctree-l1"><a class="reference internal" href="ini/databases.html">Databases</a></li>
<li class="toctree-l1"><a class="reference internal" href="ini/data_analysis.html">Stimuli data analysis</a></li>
<li class="toctree-l1"><a class="reference internal" href="ini/environment.html">Stimuli provider (Environment)</a></li>
<li class="toctree-l1"><a class="reference internal" href="ini/layers.html">Network Layers</a></li>
<li class="toctree-l1"><a class="reference internal" href="ini/target.html">Targets (outputs & losses)</a></li>
<li class="toctree-l1"><a class="reference internal" href="adversarial.html">Adversarial module</a></li>
</ul>
<p class="caption"><span class="caption-text">Python API:</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="python_api/intro.html">Introduction</a></li>
<li class="toctree-l1"><a class="reference internal" href="python_api/databases.html">Databases</a></li>
<li class="toctree-l1"><a class="reference internal" href="python_api/cells.html">Cells</a></li>
<li class="toctree-l1"><a class="reference internal" href="python_api/tensor.html">Tensor</a></li>
<li class="toctree-l1"><a class="reference internal" href="python_api/interoperability.html">Interoperability</a></li>
<li class="toctree-l1"><a class="reference internal" href="python_api/example.html">Example</a></li>
</ul>
<p class="caption"><span class="caption-text">C++/Python core:</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="core/core.html">Core N2D2</a></li>
<li class="toctree-l1"><a class="reference internal" href="core/example.html">Example</a></li>
</ul>
<p class="caption"><span class="caption-text">C++ API / Developer:</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="dev_intro.html">Introduction</a></li>
</ul>
</div>
</div>
</nav>
<section data-toggle="wy-nav-shift" class="wy-nav-content-wrap">
<nav class="wy-nav-top" aria-label="top navigation">
<i data-toggle="wy-nav-top" class="fa fa-bars"></i>
<a href="index.html">N2D2</a>
</nav>
<div class="wy-nav-content">
<div class="rst-content">
<div role="navigation" aria-label="breadcrumbs navigation">
<ul class="wy-breadcrumbs">
<li><a href="index.html" class="icon icon-home"></a> »</li>
<li>Export: ONNX</li>
<li class="wy-breadcrumbs-aside">
<a href="_sources/export_ONNX.rst.txt" rel="nofollow"> View page source</a>
</li>
</ul>
<hr/>
</div>
<div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
<div itemprop="articleBody">
<div class="section" id="export-onnx">
<span id="export-onnx-label"></span><h1>Export: ONNX<a class="headerlink" href="#export-onnx" title="Permalink to this headline">¶</a></h1>
<dl class="simple">
<dt>Export type: <code class="docutils literal notranslate"><span class="pre">ONNX</span></code></dt><dd><p>ONNX export.</p>
</dd>
</dl>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">n2d2</span> <span class="n">MobileNet_ONNX</span><span class="o">.</span><span class="n">ini</span> <span class="o">-</span><span class="n">seed</span> <span class="mi">1</span> <span class="o">-</span><span class="n">w</span> <span class="o">/</span><span class="n">dev</span><span class="o">/</span><span class="n">null</span> <span class="o">-</span><span class="n">export</span> <span class="n">ONNX</span>
</pre></div>
</div>
<div class="section" id="principle">
<h2>Principle<a class="headerlink" href="#principle" title="Permalink to this headline">¶</a></h2>
<p>The ONNX export allows you to generate an ONNX model from a N2D2 model. The
generated ONNX model is optimized for inference and can be quantized beforehand
with either post-training quantization or Quantization Aware Training (QAT).</p>
<div class="section" id="graph-optimizations">
<h3>Graph optimizations<a class="headerlink" href="#graph-optimizations" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li><p>Weights are equalized between layers when possible;</p></li>
<li><p><code class="docutils literal notranslate"><span class="pre">BatchNorm</span></code> is automatically fused with the preceding <code class="docutils literal notranslate"><span class="pre">Conv</span></code> or <code class="docutils literal notranslate"><span class="pre">Fc</span></code> when possible;</p></li>
<li><p><code class="docutils literal notranslate"><span class="pre">Padding</span></code> layers are fused with <code class="docutils literal notranslate"><span class="pre">Conv</span></code> when possible;</p></li>
<li><p><code class="docutils literal notranslate"><span class="pre">Dropout</span></code> layers are removed.</p></li>
</ul>
</div>
<div class="section" id="export-parameters">
<h3>Export parameters<a class="headerlink" href="#export-parameters" title="Permalink to this headline">¶</a></h3>
<p>Extra parameters can be passed during export using the
<code class="docutils literal notranslate"><span class="pre">-export-parameters</span> <span class="pre">params.ini</span></code> command line argument. The parameters must be
saved in an INI-like file.</p>
<p>List of available parameters:</p>
<table class="docutils align-default">
<colgroup>
<col style="width: 35%" />
<col style="width: 65%" />
</colgroup>
<thead>
<tr class="row-odd"><th class="head"><p>Argument [default value]</p></th>
<th class="head"><p>Description</p></th>
</tr>
</thead>
<tbody>
<tr class="row-even"><td><p><code class="docutils literal notranslate"><span class="pre">ImplicitCasting</span></code> [0]</p></td>
<td><p>If true (1), casting in the graph is implicit and <code class="docutils literal notranslate"><span class="pre">Cast</span></code> ONNX operators are not inserted</p></td>
</tr>
<tr class="row-odd"><td><p><code class="docutils literal notranslate"><span class="pre">FakeQuantization</span></code> [0]</p></td>
<td><p>If true (1), the graph is fake quantized, meaning floating-point ONNX operators are used for the computation</p></td>
</tr>
</tbody>
</table>
<p>The parameters <code class="docutils literal notranslate"><span class="pre">ImplicitCasting</span></code> and <code class="docutils literal notranslate"><span class="pre">FakeQuantization</span></code> are useful only for
quantized networks. In this case, a full integer ONNX graph is generated when
possible, notably using the ONNX <em>ConvInteger</em> and <em>MatMulInteger</em> when
<code class="docutils literal notranslate"><span class="pre">-nbbits</span></code> is ≤ 8 bits. An example of generated graph is shown below, with a
<code class="docutils literal notranslate"><span class="pre">Single-shift</span></code> activation rescaling mode (<code class="docutils literal notranslate"><span class="pre">-act-rescaling-mode</span></code>, see
<a class="reference internal" href="quant/post.html#post-quant-label"><span class="std std-ref">Post-training quantization</span></a>):</p>
<div class="figure align-center">
<img alt="Example of fully integer, quantized, exported ONNX graph." src="_images/export_ONNX_quant.svg" /></div>
<p>By default, strict adherence to the ONNX standard is enforced, by adding
explicit <code class="docutils literal notranslate"><span class="pre">Cast</span></code> operators when required. The automatic insertion of <code class="docutils literal notranslate"><span class="pre">Cast</span></code>
operators can be disabled by setting the <code class="docutils literal notranslate"><span class="pre">ImplicitCasting</span></code> export parameter
to true. This results in the simplified graph below:</p>
<div class="figure align-center">
<img alt="Example of fully integer, quantized, exported ONNX graph without ``Cast`` operators (with ``ImplicitCasting`` set to 1)." src="_images/export_ONNX_quant_implicit_cast.svg" /></div>
<p>The <code class="docutils literal notranslate"><span class="pre">FakeQuantization</span></code> parameter allows to export a quantized network using
fake quantization, meaning the parameters of the network are quantized (integer)
but their representation remains in floating-point and the computation is done
with floating-point operators. However, the output values of the network
should be almost identical to when the computation is done in integer. The
differences are due to numerical errors as all integers cannot be represented
exactly with floating-point.</p>
<div class="figure align-center">
<img alt="Example of fully integer, quantized, exported ONNX graph with fake quantization (``FakeQuantization`` set to 1)." src="_images/export_ONNX_quant_fake.svg" /></div>
<div class="admonition note">
<p class="admonition-title">Note</p>
<p>The <code class="docutils literal notranslate"><span class="pre">FakeQuantization</span></code>, when set, implies <code class="docutils literal notranslate"><span class="pre">ImplicitCasting</span></code>, as no
casting operator is required in a fully floating-point graph.</p>
</div>
</div>
</div>
<div class="section" id="example">
<h2>Example<a class="headerlink" href="#example" title="Permalink to this headline">¶</a></h2>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">n2d2</span> <span class="n">MobileNet_ONNX</span><span class="o">.</span><span class="n">ini</span> <span class="o">-</span><span class="n">seed</span> <span class="mi">1</span> <span class="o">-</span><span class="n">w</span> <span class="o">/</span><span class="n">dev</span><span class="o">/</span><span class="n">null</span> <span class="o">-</span><span class="n">export</span> <span class="n">ONNX</span> <span class="o">-</span><span class="n">nbbits</span> <span class="mi">8</span> <span class="o">-</span><span class="n">calib</span> <span class="o">-</span><span class="mi">1</span> <span class="o">-</span><span class="n">db</span><span class="o">-</span><span class="n">export</span> <span class="mi">100</span> <span class="o">-</span><span class="n">test</span>
</pre></div>
</div>
<p>This command generates a 8-bits integer quantized ONNX model in the sub-directory
<code class="docutils literal notranslate"><span class="pre">export_ONNX_int8</span></code>.</p>
</div>
</div>
</div>
</div>
<footer>
<hr/>
<div role="contentinfo">
<p>
© Copyright 2019, CEA LIST
</p>
</div>
Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a
<a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a>
provided by <a href="https://readthedocs.org">Read the Docs</a>.
</footer>
</div>
</div>
</section>
</div>
<script type="text/javascript">
jQuery(function () {
SphinxRtdTheme.Navigation.enable(true);
});
</script>
</body>
</html>