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

	
            <!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.params &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" />
      <link rel="stylesheet" href="../../../_static/graphviz.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">SuperGradients</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../../welcome.html">SuperGradients</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../../super_gradients.common.html">Common</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../../super_gradients.training.html">Training</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.params</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.params</h1><div class="highlight"><pre>
<span></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="n">DEFAULT_TRAINING_PARAMS</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;lr_warmup_epochs&quot;</span><span class="p">:</span> <span class="mi">0</span><span class="p">,</span>
                           <span class="s2">&quot;cosine_final_lr_ratio&quot;</span><span class="p">:</span> <span class="mf">0.01</span><span class="p">,</span>
                           <span class="s2">&quot;optimizer&quot;</span><span class="p">:</span> <span class="s2">&quot;SGD&quot;</span><span class="p">,</span>
                           <span class="s2">&quot;criterion_params&quot;</span><span class="p">:</span> <span class="p">{},</span>
                           <span class="s2">&quot;ema&quot;</span><span class="p">:</span> <span class="kc">False</span><span class="p">,</span>
                           <span class="s2">&quot;batch_accumulate&quot;</span><span class="p">:</span> <span class="mi">1</span><span class="p">,</span>  <span class="c1"># number of batches to accumulate before every backward pass</span>
                           <span class="s2">&quot;ema_params&quot;</span><span class="p">:</span> <span class="p">{},</span>
                           <span class="s2">&quot;zero_weight_decay_on_bias_and_bn&quot;</span><span class="p">:</span> <span class="kc">False</span><span class="p">,</span>
                           <span class="s2">&quot;load_opt_params&quot;</span><span class="p">:</span> <span class="kc">True</span><span class="p">,</span>
                           <span class="s2">&quot;run_validation_freq&quot;</span><span class="p">:</span> <span class="mi">1</span><span class="p">,</span>
                           <span class="s2">&quot;save_model&quot;</span><span class="p">:</span> <span class="kc">True</span><span class="p">,</span>
                           <span class="s2">&quot;metric_to_watch&quot;</span><span class="p">:</span> <span class="s2">&quot;Accuracy&quot;</span><span class="p">,</span>
                           <span class="s2">&quot;launch_tensorboard&quot;</span><span class="p">:</span> <span class="kc">False</span><span class="p">,</span>
                           <span class="s2">&quot;tb_files_user_prompt&quot;</span><span class="p">:</span> <span class="kc">False</span><span class="p">,</span>  <span class="c1"># Asks User for Tensorboard Deletion Prompt</span>
                           <span class="s2">&quot;silent_mode&quot;</span><span class="p">:</span> <span class="kc">False</span><span class="p">,</span>  <span class="c1"># Silents the Print outs</span>
                           <span class="s2">&quot;mixed_precision&quot;</span><span class="p">:</span> <span class="kc">False</span><span class="p">,</span>
                           <span class="s2">&quot;tensorboard_port&quot;</span><span class="p">:</span> <span class="kc">None</span><span class="p">,</span>
                           <span class="s2">&quot;save_ckpt_epoch_list&quot;</span><span class="p">:</span> <span class="p">[],</span>  <span class="c1"># indices where the ckpt will save automatically</span>
                           <span class="s2">&quot;average_best_models&quot;</span><span class="p">:</span> <span class="kc">True</span><span class="p">,</span>
                           <span class="s2">&quot;dataset_statistics&quot;</span><span class="p">:</span> <span class="kc">False</span><span class="p">,</span>  <span class="c1"># add a dataset statistical analysis and sample images to tensorboard</span>
                           <span class="s2">&quot;save_tensorboard_to_s3&quot;</span><span class="p">:</span> <span class="kc">False</span><span class="p">,</span>
                           <span class="s2">&quot;lr_schedule_function&quot;</span><span class="p">:</span> <span class="kc">None</span><span class="p">,</span>
                           <span class="s2">&quot;train_metrics_list&quot;</span><span class="p">:</span> <span class="p">[],</span>
                           <span class="s2">&quot;valid_metrics_list&quot;</span><span class="p">:</span> <span class="p">[],</span>
                           <span class="s2">&quot;loss_logging_items_names&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;Loss&quot;</span><span class="p">],</span>
                           <span class="s2">&quot;greater_metric_to_watch_is_better&quot;</span><span class="p">:</span> <span class="kc">True</span><span class="p">,</span>
                           <span class="s2">&quot;precise_bn&quot;</span><span class="p">:</span> <span class="kc">False</span><span class="p">,</span>
                           <span class="s2">&quot;precise_bn_batch_size&quot;</span><span class="p">:</span> <span class="kc">None</span><span class="p">,</span>
                           <span class="s2">&quot;seed&quot;</span><span class="p">:</span> <span class="mi">42</span><span class="p">,</span>
                           <span class="s2">&quot;lr_mode&quot;</span><span class="p">:</span> <span class="kc">None</span><span class="p">,</span>
                           <span class="s2">&quot;phase_callbacks&quot;</span><span class="p">:</span> <span class="p">[],</span>
                           <span class="s2">&quot;log_installed_packages&quot;</span><span class="p">:</span> <span class="kc">True</span><span class="p">,</span>
                           <span class="s2">&quot;save_full_train_log&quot;</span><span class="p">:</span> <span class="kc">False</span><span class="p">,</span>
                           <span class="s2">&quot;sg_logger&quot;</span><span class="p">:</span> <span class="s2">&quot;base_sg_logger&quot;</span><span class="p">,</span>
                           <span class="s2">&quot;sg_logger_params&quot;</span><span class="p">:</span>
                               <span class="p">{</span><span class="s2">&quot;tb_files_user_prompt&quot;</span><span class="p">:</span> <span class="kc">False</span><span class="p">,</span>  <span class="c1"># Asks User for Tensorboard Deletion Prompt</span>
                                <span class="s2">&quot;project_name&quot;</span><span class="p">:</span> <span class="s2">&quot;&quot;</span><span class="p">,</span>
                                <span class="s2">&quot;launch_tensorboard&quot;</span><span class="p">:</span> <span class="kc">False</span><span class="p">,</span>
                                <span class="s2">&quot;tensorboard_port&quot;</span><span class="p">:</span> <span class="kc">None</span><span class="p">,</span>
                                <span class="s2">&quot;save_checkpoints_remote&quot;</span><span class="p">:</span> <span class="kc">False</span><span class="p">,</span>  <span class="c1"># upload checkpoint files to s3</span>
                                <span class="s2">&quot;save_tensorboard_remote&quot;</span><span class="p">:</span> <span class="kc">False</span><span class="p">,</span>  <span class="c1"># upload tensorboard files to s3</span>
                                <span class="s2">&quot;save_logs_remote&quot;</span><span class="p">:</span> <span class="kc">False</span><span class="p">},</span>  <span class="c1"># upload log files to s3</span>
                           <span class="s2">&quot;warmup_mode&quot;</span><span class="p">:</span> <span class="s2">&quot;linear_step&quot;</span>
                           <span class="p">}</span>

<span class="n">DEFAULT_OPTIMIZER_PARAMS_SGD</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;weight_decay&quot;</span><span class="p">:</span> <span class="mf">1e-4</span><span class="p">,</span> <span class="s2">&quot;momentum&quot;</span><span class="p">:</span> <span class="mf">0.9</span><span class="p">}</span>

<span class="n">DEFAULT_OPTIMIZER_PARAMS_ADAM</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;weight_decay&quot;</span><span class="p">:</span> <span class="mf">1e-4</span><span class="p">}</span>

<span class="n">DEFAULT_OPTIMIZER_PARAMS_RMSPROP</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;weight_decay&quot;</span><span class="p">:</span> <span class="mf">1e-4</span><span class="p">,</span> <span class="s2">&quot;momentum&quot;</span><span class="p">:</span> <span class="mf">0.9</span><span class="p">}</span>

<span class="n">DEFAULT_OPTIMIZER_PARAMS_RMSPROPTF</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;weight_decay&quot;</span><span class="p">:</span> <span class="mf">1e-4</span><span class="p">,</span> <span class="s2">&quot;momentum&quot;</span><span class="p">:</span> <span class="mf">0.9</span><span class="p">}</span>

<span class="n">TRAINING_PARAM_SCHEMA</span> <span class="o">=</span> <span class="p">{</span><span class="s2">&quot;type&quot;</span><span class="p">:</span> <span class="s2">&quot;object&quot;</span><span class="p">,</span>
                         <span class="s2">&quot;properties&quot;</span><span class="p">:</span> <span class="p">{</span>
                             <span class="s2">&quot;max_epochs&quot;</span><span class="p">:</span> <span class="p">{</span><span class="s2">&quot;type&quot;</span><span class="p">:</span> <span class="s2">&quot;number&quot;</span><span class="p">,</span> <span class="s2">&quot;minimum&quot;</span><span class="p">:</span> <span class="mi">1</span><span class="p">,</span> <span class="s2">&quot;maximum&quot;</span><span class="p">:</span> <span class="mi">800</span><span class="p">},</span>

                             <span class="c1"># FIXME: CHECK THE IMPORTANCE OF THE COMMENTED SCHEMA- AS IT CAUSES HYDRA USE TO CRASH</span>

                             <span class="c1"># &quot;lr_updates&quot;: {&quot;type&quot;: &quot;array&quot;, &quot;minItems&quot;: 1},</span>
                             <span class="s2">&quot;lr_decay_factor&quot;</span><span class="p">:</span> <span class="p">{</span><span class="s2">&quot;type&quot;</span><span class="p">:</span> <span class="s2">&quot;number&quot;</span><span class="p">,</span> <span class="s2">&quot;minimum&quot;</span><span class="p">:</span> <span class="mi">0</span><span class="p">,</span> <span class="s2">&quot;maximum&quot;</span><span class="p">:</span> <span class="mi">1</span><span class="p">},</span>
                             <span class="s2">&quot;lr_warmup_epochs&quot;</span><span class="p">:</span> <span class="p">{</span><span class="s2">&quot;type&quot;</span><span class="p">:</span> <span class="s2">&quot;number&quot;</span><span class="p">,</span> <span class="s2">&quot;minimum&quot;</span><span class="p">:</span> <span class="mi">0</span><span class="p">,</span> <span class="s2">&quot;maximum&quot;</span><span class="p">:</span> <span class="mi">10</span><span class="p">},</span>
                             <span class="s2">&quot;initial_lr&quot;</span><span class="p">:</span> <span class="p">{</span><span class="s2">&quot;type&quot;</span><span class="p">:</span> <span class="s2">&quot;number&quot;</span><span class="p">,</span> <span class="s2">&quot;exclusiveMinimum&quot;</span><span class="p">:</span> <span class="mi">0</span><span class="p">,</span> <span class="s2">&quot;maximum&quot;</span><span class="p">:</span> <span class="mi">10</span><span class="p">}</span>
                         <span class="p">},</span>
                         <span class="s2">&quot;if&quot;</span><span class="p">:</span> <span class="p">{</span>
                             <span class="s2">&quot;properties&quot;</span><span class="p">:</span> <span class="p">{</span><span class="s2">&quot;lr_mode&quot;</span><span class="p">:</span> <span class="p">{</span><span class="s2">&quot;const&quot;</span><span class="p">:</span> <span class="s2">&quot;step&quot;</span><span class="p">}}</span>
                         <span class="p">},</span>
                         <span class="s2">&quot;then&quot;</span><span class="p">:</span> <span class="p">{</span>
                             <span class="s2">&quot;required&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;lr_updates&quot;</span><span class="p">,</span> <span class="s2">&quot;lr_decay_factor&quot;</span><span class="p">]</span>
                         <span class="p">},</span>
                         <span class="s2">&quot;required&quot;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;max_epochs&quot;</span><span class="p">,</span> <span class="s2">&quot;lr_mode&quot;</span><span class="p">,</span> <span class="s2">&quot;initial_lr&quot;</span><span class="p">,</span> <span class="s2">&quot;loss&quot;</span><span class="p">]</span>
                         <span class="p">}</span>


<div class="viewcode-block" id="TrainingParams"><a class="viewcode-back" href="../../../super_gradients.training.html#super_gradients.training.params.TrainingParams">[docs]</a><span class="k">class</span> <span class="nc">TrainingParams</span><span class="p">(</span><span class="n">HpmStruct</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="o">**</span><span class="n">entries</span><span class="p">):</span>
        <span class="c1"># WE initialize by the default training params, overridden by the provided params</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="o">**</span><span class="n">DEFAULT_TRAINING_PARAMS</span><span class="p">)</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">set_schema</span><span class="p">(</span><span class="n">TRAINING_PARAM_SCHEMA</span><span class="p">)</span>
        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">entries</span><span class="p">)</span> <span class="o">&gt;</span> <span class="mi">0</span><span class="p">:</span>
            <span class="bp">self</span><span class="o">.</span><span class="n">override</span><span class="p">(</span><span class="o">**</span><span class="n">entries</span><span class="p">)</span>

<div class="viewcode-block" id="TrainingParams.override"><a class="viewcode-back" href="../../../super_gradients.training.html#super_gradients.training.params.TrainingParams.override">[docs]</a>    <span class="k">def</span> <span class="nf">override</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">**</span><span class="n">entries</span><span class="p">):</span>
        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">override</span><span class="p">(</span><span class="o">**</span><span class="n">entries</span><span class="p">)</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">validate</span><span class="p">()</span></div></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>