Deci-AI
/
super-gradients
connected to https://github.com/Deci-AI/super-gradients.git


  
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

	
337

	
338

	
339

	
340

	
341

	
            <!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>super_gradients.training.models.shufflenetv2 &mdash; SuperGradients 1.0 documentation</title>
      <link rel="stylesheet" href="../../../../_static/pygments.css" type="text/css" />
      <link rel="stylesheet" href="../../../../_static/css/theme.css" type="text/css" />
  <!--[if lt IE 9]>
    <script src="../../../../_static/js/html5shiv.min.js"></script>
  <![endif]-->
  
        <script data-url_root="../../../../" id="documentation_options" 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/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"> SuperGradients
          </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="Navigation menu">
              <p class="caption"><span class="caption-text">Contents:</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../../../intro.html">Introduction</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../../../intro.html#installation">Installation</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../../../super_gradients.training.html">super_gradients.training package</a></li>
</ul>

        </div>
      </div>
    </nav>

    <section data-toggle="wy-nav-shift" class="wy-nav-content-wrap"><nav class="wy-nav-top" aria-label="Mobile navigation menu" >
          <i data-toggle="wy-nav-top" class="fa fa-bars"></i>
          <a href="../../../../index.html">SuperGradients</a>
      </nav>

      <div class="wy-nav-content">
        <div class="rst-content">
          <div role="navigation" aria-label="Page navigation">
  <ul class="wy-breadcrumbs">
      <li><a href="../../../../index.html" class="icon icon-home"></a> &raquo;</li>
          <li><a href="../../../index.html">Module code</a> &raquo;</li>
      <li>super_gradients.training.models.shufflenetv2</li>
      <li class="wy-breadcrumbs-aside">
      </li>
  </ul>
  <hr/>
</div>
          <div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
           <div itemprop="articleBody">
             
  <h1>Source code for super_gradients.training.models.shufflenetv2</h1><div class="highlight"><pre>
<span></span><span class="sd">&quot;&quot;&quot;</span>
<span class="sd">ShuffleNetV2 in PyTorch.</span>

<span class="sd">See the paper &quot;ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design&quot; for more details.</span>
<span class="sd">(https://arxiv.org/abs/1807.11164)</span>

<span class="sd">Code taken from torchvision/models/shufflenetv2.py</span>
<span class="sd">&quot;&quot;&quot;</span>
<span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">List</span>

<span class="kn">import</span> <span class="nn">torch</span>
<span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">Tensor</span>
<span class="kn">import</span> <span class="nn">torch.nn</span> <span class="k">as</span> <span class="nn">nn</span>

<span class="kn">from</span> <span class="nn">super_gradients.training.utils</span> <span class="kn">import</span> <span class="n">HpmStruct</span>
<span class="kn">from</span> <span class="nn">super_gradients.training.models.sg_module</span> <span class="kn">import</span> <span class="n">SgModule</span>


<span class="n">__all__</span> <span class="o">=</span> <span class="p">[</span>
    <span class="s1">&#39;ShuffleNetV2Base&#39;</span><span class="p">,</span> <span class="s1">&#39;ShufflenetV2_x0_5&#39;</span><span class="p">,</span> <span class="s1">&#39;ShufflenetV2_x1_0&#39;</span><span class="p">,</span>
    <span class="s1">&#39;ShufflenetV2_x1_5&#39;</span><span class="p">,</span> <span class="s1">&#39;ShufflenetV2_x2_0&#39;</span><span class="p">,</span> <span class="s1">&#39;CustomizedShuffleNetV2&#39;</span>
<span class="p">]</span>


<span class="k">class</span> <span class="nc">ChannelShuffleInvertedResidual</span><span class="p">(</span><span class="n">nn</span><span class="o">.</span><span class="n">Module</span><span class="p">):</span>
    <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">    Implement Inverted Residual block as in [https://arxiv.org/abs/1807.11164] in Fig.3 (c) &amp; (d):</span>

<span class="sd">    * When stride &gt; 1</span>
<span class="sd">      - the whole input goes through branch1,</span>
<span class="sd">      - the whole input goes through branch2 ,</span>
<span class="sd">      and the arbitrary number of output channels are produced.</span>
<span class="sd">    * When stride == 1</span>
<span class="sd">      - half of input channels in are passed as identity,</span>
<span class="sd">      - another half of input channels goes through branch2,</span>
<span class="sd">      and the number of output channels after the block remains the same as in input.</span>

<span class="sd">    Channel shuffle is performed on a concatenation in both cases.</span>
<span class="sd">    &quot;&quot;&quot;</span>
    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">inp</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">out</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">stride</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
        <span class="nb">super</span><span class="p">(</span><span class="n">ChannelShuffleInvertedResidual</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>

        <span class="k">assert</span> <span class="mi">1</span> <span class="o">&lt;=</span> <span class="n">stride</span> <span class="o">&lt;=</span> <span class="mi">3</span><span class="p">,</span> <span class="s2">&quot;Illegal stride value&quot;</span>
        <span class="k">assert</span> <span class="p">(</span><span class="n">stride</span> <span class="o">!=</span> <span class="mi">1</span><span class="p">)</span> <span class="ow">or</span> <span class="p">(</span><span class="n">inp</span> <span class="o">==</span> <span class="n">out</span><span class="p">),</span> \
            <span class="s2">&quot;When stride == 1 num of input channels should be equal to the requested num of out output channels&quot;</span>

        <span class="bp">self</span><span class="o">.</span><span class="n">stride</span> <span class="o">=</span> <span class="n">stride</span>
        <span class="c1"># half of requested out channels will be produced by each branch</span>
        <span class="n">branch_features</span> <span class="o">=</span> <span class="n">out</span> <span class="o">//</span> <span class="mi">2</span>

        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">stride</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">:</span>
            <span class="bp">self</span><span class="o">.</span><span class="n">branch1</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
                <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">inp</span><span class="p">,</span> <span class="n">inp</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">stride</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">groups</span><span class="o">=</span><span class="n">inp</span><span class="p">),</span>
                <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">inp</span><span class="p">),</span>

                <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">inp</span><span class="p">,</span> <span class="n">branch_features</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
                <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">branch_features</span><span class="p">),</span>
                <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
            <span class="p">)</span>
        <span class="k">else</span><span class="p">:</span>
            <span class="c1"># won&#39;t be called if self.stride == 1</span>
            <span class="bp">self</span><span class="o">.</span><span class="n">branch1</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Identity</span><span class="p">()</span>

        <span class="bp">self</span><span class="o">.</span><span class="n">branch2</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
            <span class="c1"># branch 2 operates on the whole input when stride &gt; 1 and on half of it otherwise</span>
            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">inp</span> <span class="k">if</span> <span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">stride</span> <span class="o">&gt;</span> <span class="mi">1</span><span class="p">)</span> <span class="k">else</span> <span class="n">inp</span> <span class="o">//</span> <span class="mi">2</span><span class="p">,</span> <span class="n">branch_features</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
                      <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">branch_features</span><span class="p">),</span>
            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>

            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">branch_features</span><span class="p">,</span> <span class="n">branch_features</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">stride</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
                      <span class="n">groups</span><span class="o">=</span><span class="n">branch_features</span><span class="p">),</span>
            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">branch_features</span><span class="p">),</span>

            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">branch_features</span><span class="p">,</span> <span class="n">branch_features</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">branch_features</span><span class="p">),</span>
            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
        <span class="p">)</span>

    <span class="nd">@staticmethod</span>
    <span class="k">def</span> <span class="nf">channel_shuffle</span><span class="p">(</span><span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">,</span> <span class="n">groups</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        From &quot;ShuffleNet V2: Practical Guidelines for EfficientCNN Architecture Design&quot; (https://arxiv.org/abs/1807.11164):</span>
<span class="sd">            A “channel shuffle” operation is then introduced to enable</span>
<span class="sd">            information communication between different groups of channels and improve accuracy.</span>

<span class="sd">        The operation preserves x.size(), but shuffles its channels in the manner explained further in the example.</span>

<span class="sd">        Example:</span>
<span class="sd">            If group = 2 (2 branches with the same # of activation maps were concatenated before channel_shuffle),</span>
<span class="sd">            then activation maps in x are:</span>
<span class="sd">            from_B1, from_B1, ... from_B2, from_B2</span>
<span class="sd">            After channel_shuffle activation maps in x will be:</span>
<span class="sd">            from_B1, from_B2, ... from_B1, from_B2</span>
<span class="sd">        &quot;&quot;&quot;</span>

        <span class="n">batch_size</span><span class="p">,</span> <span class="n">num_channels</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">()</span>
        <span class="n">channels_per_group</span> <span class="o">=</span> <span class="n">num_channels</span> <span class="o">//</span> <span class="n">groups</span>

        <span class="c1"># reshape</span>
        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">batch_size</span><span class="p">,</span> <span class="n">groups</span><span class="p">,</span> <span class="n">channels_per_group</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">)</span>
        <span class="n">x</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">transpose</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span><span class="o">.</span><span class="n">contiguous</span><span class="p">()</span>

        <span class="c1"># flatten</span>
        <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">batch_size</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">)</span>
        <span class="k">return</span> <span class="n">x</span>

    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">stride</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
            <span class="c1"># num channels remains the same due to assert that inp == out in __init__</span>
            <span class="n">x1</span><span class="p">,</span> <span class="n">x2</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">chunk</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
            <span class="n">out</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="n">x1</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch2</span><span class="p">(</span><span class="n">x2</span><span class="p">)),</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
        <span class="k">else</span><span class="p">:</span>
            <span class="c1"># inp num channels can change to a requested arbitrary out num channels</span>
            <span class="n">out</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="bp">self</span><span class="o">.</span><span class="n">branch1</span><span class="p">(</span><span class="n">x</span><span class="p">),</span> <span class="bp">self</span><span class="o">.</span><span class="n">branch2</span><span class="p">(</span><span class="n">x</span><span class="p">)),</span> <span class="n">dim</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>

        <span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">channel_shuffle</span><span class="p">(</span><span class="n">out</span><span class="p">,</span> <span class="mi">2</span><span class="p">)</span>
        <span class="k">return</span> <span class="n">out</span>


<div class="viewcode-block" id="ShuffleNetV2Base"><a class="viewcode-back" href="../../../../super_gradients.training.models.html#super_gradients.training.models.shufflenetv2.ShuffleNetV2Base">[docs]</a><span class="k">class</span> <span class="nc">ShuffleNetV2Base</span><span class="p">(</span><span class="n">SgModule</span><span class="p">):</span>

    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">structure</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span> <span class="n">stages_out_channels</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">],</span>
                 <span class="n">backbone_mode</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1000</span><span class="p">,</span>
                 <span class="n">block</span><span class="p">:</span> <span class="n">nn</span><span class="o">.</span><span class="n">Module</span> <span class="o">=</span> <span class="n">ChannelShuffleInvertedResidual</span><span class="p">):</span>
        <span class="nb">super</span><span class="p">(</span><span class="n">ShuffleNetV2Base</span><span class="p">,</span> <span class="bp">self</span><span class="p">)</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>

        <span class="bp">self</span><span class="o">.</span><span class="n">backbone_mode</span> <span class="o">=</span> <span class="n">backbone_mode</span>

        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">structure</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">3</span><span class="p">:</span>
            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s1">&#39;expected structure as list of 3 positive ints&#39;</span><span class="p">)</span>
        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">stages_out_channels</span><span class="p">)</span> <span class="o">!=</span> <span class="mi">5</span><span class="p">:</span>
            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s1">&#39;expected stages_out_channels as list of 5 positive ints&#39;</span><span class="p">)</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">structure</span> <span class="o">=</span> <span class="n">structure</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span> <span class="o">=</span> <span class="n">stages_out_channels</span>

        <span class="n">input_channels</span> <span class="o">=</span> <span class="mi">3</span>
        <span class="n">output_channels</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">conv1</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">input_channels</span><span class="p">,</span> <span class="n">output_channels</span><span class="p">,</span> <span class="mi">3</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">output_channels</span><span class="p">),</span>
            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
        <span class="p">)</span>
        <span class="n">input_channels</span> <span class="o">=</span> <span class="n">output_channels</span>

        <span class="bp">self</span><span class="o">.</span><span class="n">maxpool</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">MaxPool2d</span><span class="p">(</span><span class="n">kernel_size</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">padding</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>

        <span class="c1"># Static annotations for mypy</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">layer2</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_make_layer</span><span class="p">(</span><span class="n">block</span><span class="p">,</span> <span class="n">input_channels</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">structure</span><span class="p">[</span><span class="mi">0</span><span class="p">])</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">layer3</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_make_layer</span><span class="p">(</span><span class="n">block</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span><span class="p">[</span><span class="mi">1</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">structure</span><span class="p">[</span><span class="mi">1</span><span class="p">])</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">layer4</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_make_layer</span><span class="p">(</span><span class="n">block</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span><span class="p">[</span><span class="mi">2</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span><span class="p">[</span><span class="mi">3</span><span class="p">],</span> <span class="bp">self</span><span class="o">.</span><span class="n">structure</span><span class="p">[</span><span class="mi">2</span><span class="p">])</span>

        <span class="n">input_channels</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span><span class="p">[</span><span class="mi">3</span><span class="p">]</span>
        <span class="n">output_channels</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">out_channels</span><span class="p">[</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">conv5</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span>
            <span class="n">nn</span><span class="o">.</span><span class="n">Conv2d</span><span class="p">(</span><span class="n">input_channels</span><span class="p">,</span> <span class="n">output_channels</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">bias</span><span class="o">=</span><span class="kc">False</span><span class="p">),</span>
            <span class="n">nn</span><span class="o">.</span><span class="n">BatchNorm2d</span><span class="p">(</span><span class="n">output_channels</span><span class="p">),</span>
            <span class="n">nn</span><span class="o">.</span><span class="n">ReLU</span><span class="p">(</span><span class="n">inplace</span><span class="o">=</span><span class="kc">True</span><span class="p">),</span>
        <span class="p">)</span>

        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">backbone_mode</span><span class="p">:</span>
            <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">AdaptiveAvgPool2d</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
            <span class="bp">self</span><span class="o">.</span><span class="n">fc</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Linear</span><span class="p">(</span><span class="n">output_channels</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">)</span>

    <span class="nd">@staticmethod</span>
    <span class="k">def</span> <span class="nf">_make_layer</span><span class="p">(</span><span class="n">block</span><span class="p">,</span> <span class="n">input_channels</span><span class="p">,</span> <span class="n">output_channels</span><span class="p">,</span> <span class="n">repeats</span><span class="p">):</span>
        <span class="c1"># add first block with stride 2 to downsize the input</span>
        <span class="n">seq</span> <span class="o">=</span> <span class="p">[</span><span class="n">block</span><span class="p">(</span><span class="n">input_channels</span><span class="p">,</span> <span class="n">output_channels</span><span class="p">,</span> <span class="mi">2</span><span class="p">)]</span>

        <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">repeats</span> <span class="o">-</span> <span class="mi">1</span><span class="p">):</span>
            <span class="n">seq</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">block</span><span class="p">(</span><span class="n">output_channels</span><span class="p">,</span> <span class="n">output_channels</span><span class="p">,</span> <span class="mi">1</span><span class="p">))</span>
        <span class="k">return</span> <span class="n">nn</span><span class="o">.</span><span class="n">Sequential</span><span class="p">(</span><span class="o">*</span><span class="n">seq</span><span class="p">)</span>

<div class="viewcode-block" id="ShuffleNetV2Base.load_state_dict"><a class="viewcode-back" href="../../../../super_gradients.training.models.html#super_gradients.training.models.shufflenetv2.ShuffleNetV2Base.load_state_dict">[docs]</a>    <span class="k">def</span> <span class="nf">load_state_dict</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">state_dict</span><span class="p">,</span> <span class="n">strict</span><span class="o">=</span><span class="kc">True</span><span class="p">):</span>
        <span class="sd">&quot;&quot;&quot;</span>
<span class="sd">        load_state_dict - Overloads the base method and calls it to load a modified dict for usage as a backbone</span>
<span class="sd">        :param state_dict:  The state_dict to load</span>
<span class="sd">        :param strict:      strict loading (see super() docs)</span>
<span class="sd">        &quot;&quot;&quot;</span>
        <span class="n">pretrained_model_weights_dict</span> <span class="o">=</span> <span class="n">state_dict</span><span class="o">.</span><span class="n">copy</span><span class="p">()</span>

        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">backbone_mode</span><span class="p">:</span>
            <span class="c1"># removing fc weights first not to break strict loading</span>
            <span class="n">fc_weights_keys</span> <span class="o">=</span> <span class="p">[</span><span class="n">k</span> <span class="k">for</span> <span class="n">k</span> <span class="ow">in</span> <span class="n">pretrained_model_weights_dict</span> <span class="k">if</span> <span class="s1">&#39;fc&#39;</span> <span class="ow">in</span> <span class="n">k</span><span class="p">]</span>

            <span class="k">for</span> <span class="n">key</span> <span class="ow">in</span> <span class="n">fc_weights_keys</span><span class="p">:</span>
                <span class="n">pretrained_model_weights_dict</span><span class="o">.</span><span class="n">pop</span><span class="p">(</span><span class="n">key</span><span class="p">)</span>

        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">load_state_dict</span><span class="p">(</span><span class="n">pretrained_model_weights_dict</span><span class="p">,</span> <span class="n">strict</span><span class="p">)</span></div>

<div class="viewcode-block" id="ShuffleNetV2Base.forward"><a class="viewcode-back" href="../../../../super_gradients.training.models.html#super_gradients.training.models.shufflenetv2.ShuffleNetV2Base.forward">[docs]</a>    <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">Tensor</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Tensor</span><span class="p">:</span>
        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv1</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">maxpool</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">layer2</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">layer3</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">layer4</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
        <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv5</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>

        <span class="k">if</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">backbone_mode</span><span class="p">:</span>
            <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">avgpool</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
            <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">),</span> <span class="o">-</span><span class="mi">1</span><span class="p">)</span>
            <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">fc</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
        <span class="k">return</span> <span class="n">x</span></div></div>


<div class="viewcode-block" id="ShufflenetV2_x0_5"><a class="viewcode-back" href="../../../../super_gradients.training.models.html#super_gradients.training.models.shufflenetv2.ShufflenetV2_x0_5">[docs]</a><span class="k">class</span> <span class="nc">ShufflenetV2_x0_5</span><span class="p">(</span><span class="n">ShuffleNetV2Base</span><span class="p">):</span>

    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">arch_params</span><span class="p">:</span> <span class="n">HpmStruct</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1000</span><span class="p">,</span> <span class="n">backbone_mode</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">):</span>
        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">([</span><span class="mi">4</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span> <span class="p">[</span><span class="mi">24</span><span class="p">,</span> <span class="mi">48</span><span class="p">,</span> <span class="mi">96</span><span class="p">,</span> <span class="mi">192</span><span class="p">,</span> <span class="mi">1024</span><span class="p">],</span>
                         <span class="n">backbone_mode</span><span class="o">=</span><span class="n">backbone_mode</span><span class="p">,</span>
                         <span class="n">num_classes</span><span class="o">=</span><span class="n">num_classes</span> <span class="ow">or</span> <span class="n">arch_params</span><span class="o">.</span><span class="n">num_classes</span><span class="p">)</span></div>


<div class="viewcode-block" id="ShufflenetV2_x1_0"><a class="viewcode-back" href="../../../../super_gradients.training.models.html#super_gradients.training.models.shufflenetv2.ShufflenetV2_x1_0">[docs]</a><span class="k">class</span> <span class="nc">ShufflenetV2_x1_0</span><span class="p">(</span><span class="n">ShuffleNetV2Base</span><span class="p">):</span>

    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">arch_params</span><span class="p">:</span> <span class="n">HpmStruct</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1000</span><span class="p">,</span> <span class="n">backbone_mode</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">):</span>
        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">([</span><span class="mi">4</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span> <span class="p">[</span><span class="mi">24</span><span class="p">,</span> <span class="mi">116</span><span class="p">,</span> <span class="mi">232</span><span class="p">,</span> <span class="mi">464</span><span class="p">,</span> <span class="mi">1024</span><span class="p">],</span>
                         <span class="n">backbone_mode</span><span class="o">=</span><span class="n">backbone_mode</span><span class="p">,</span>
                         <span class="n">num_classes</span><span class="o">=</span><span class="n">num_classes</span> <span class="ow">or</span> <span class="n">arch_params</span><span class="o">.</span><span class="n">num_classes</span><span class="p">)</span></div>


<div class="viewcode-block" id="ShufflenetV2_x1_5"><a class="viewcode-back" href="../../../../super_gradients.training.models.html#super_gradients.training.models.shufflenetv2.ShufflenetV2_x1_5">[docs]</a><span class="k">class</span> <span class="nc">ShufflenetV2_x1_5</span><span class="p">(</span><span class="n">ShuffleNetV2Base</span><span class="p">):</span>

    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">arch_params</span><span class="p">:</span> <span class="n">HpmStruct</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1000</span><span class="p">,</span> <span class="n">backbone_mode</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">):</span>
        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">([</span><span class="mi">4</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span> <span class="p">[</span><span class="mi">24</span><span class="p">,</span> <span class="mi">176</span><span class="p">,</span> <span class="mi">352</span><span class="p">,</span> <span class="mi">704</span><span class="p">,</span> <span class="mi">1024</span><span class="p">],</span>
                         <span class="n">backbone_mode</span><span class="o">=</span><span class="n">backbone_mode</span><span class="p">,</span>
                         <span class="n">num_classes</span><span class="o">=</span><span class="n">num_classes</span> <span class="ow">or</span> <span class="n">arch_params</span><span class="o">.</span><span class="n">num_classes</span><span class="p">)</span></div>


<div class="viewcode-block" id="ShufflenetV2_x2_0"><a class="viewcode-back" href="../../../../super_gradients.training.models.html#super_gradients.training.models.shufflenetv2.ShufflenetV2_x2_0">[docs]</a><span class="k">class</span> <span class="nc">ShufflenetV2_x2_0</span><span class="p">(</span><span class="n">ShuffleNetV2Base</span><span class="p">):</span>

    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">arch_params</span><span class="p">:</span> <span class="n">HpmStruct</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1000</span><span class="p">,</span> <span class="n">backbone_mode</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">):</span>
        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">([</span><span class="mi">4</span><span class="p">,</span> <span class="mi">8</span><span class="p">,</span> <span class="mi">4</span><span class="p">],</span> <span class="p">[</span><span class="mi">24</span><span class="p">,</span> <span class="mi">244</span><span class="p">,</span> <span class="mi">488</span><span class="p">,</span> <span class="mi">976</span><span class="p">,</span> <span class="mi">2048</span><span class="p">],</span>
                         <span class="n">backbone_mode</span><span class="o">=</span><span class="n">backbone_mode</span><span class="p">,</span>
                         <span class="n">num_classes</span><span class="o">=</span><span class="n">num_classes</span> <span class="ow">or</span> <span class="n">arch_params</span><span class="o">.</span><span class="n">num_classes</span><span class="p">)</span></div>


<div class="viewcode-block" id="CustomizedShuffleNetV2"><a class="viewcode-back" href="../../../../super_gradients.training.models.html#super_gradients.training.models.shufflenetv2.CustomizedShuffleNetV2">[docs]</a><span class="k">class</span> <span class="nc">CustomizedShuffleNetV2</span><span class="p">(</span><span class="n">ShuffleNetV2Base</span><span class="p">):</span>

    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">arch_params</span><span class="p">:</span> <span class="n">HpmStruct</span><span class="p">,</span> <span class="n">num_classes</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1000</span><span class="p">,</span> <span class="n">backbone_mode</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span><span class="p">):</span>
        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">arch_params</span><span class="o">.</span><span class="n">structure</span><span class="p">,</span> <span class="n">arch_params</span><span class="o">.</span><span class="n">stages_out_channels</span><span class="p">,</span>
                         <span class="n">backbone_mode</span><span class="o">=</span><span class="n">backbone_mode</span><span class="p">,</span>
                         <span class="n">num_classes</span><span class="o">=</span><span class="n">num_classes</span> <span class="ow">or</span> <span class="n">arch_params</span><span class="o">.</span><span class="n">num_classes</span><span class="p">)</span></div>
</pre></div>

           </div>
          </div>
          <footer>

  <hr/>

  <div role="contentinfo">
    <p>&#169; Copyright 2021, SuperGradients team.</p>
  </div>

  Built with <a href="https://www.sphinx-doc.org/">Sphinx</a> using a
    <a href="https://github.com/readthedocs/sphinx_rtd_theme">theme</a>
    provided by <a href="https://readthedocs.org">Read the Docs</a>.
   

</footer>
        </div>
      </div>
    </section>
  </div>
  <script>
      jQuery(function () {
          SphinxRtdTheme.Navigation.enable(true);
      });
  </script> 

</body>
</html>