Add WARNING messages explanation.

Author: ueno-hiroshi002

docs/_sources/guidelines/XQuant_Extension_Tool.rst.txt

@@ -81,7 +81,7 @@ To be more specific, execute the following steps:
 
 .. note::
 
-  If log of *mct.set_log_folder* does not exist, the *Unbalanced Concatnation* described below will not be executed.
+  If log of *mct.set_log_folder* does not exist, the *Unbalanced Concatenation* described below will not be executed.
 
 XQuantConfig Format and Examples
 ======================================
@@ -158,14 +158,34 @@ Understanding the Judgeable Troubleshoots
 =======================================================
 
 The following items are automatically identified by the XQuant Extension Tool.
-
-Please refer to the respective Troubleshooting Manuals for details.
+When this tool detects these issues, corresponding WARNING messages are displayed in your console.
+Please refer to the respective Troubleshooting Manuals and change the configuration as needed.
 
 * `Outlier Removal <https://sonysemiconductorsolutions.github.io/mct-model-optimization/docs_troubleshoot/troubleshoots/outlier_removal.html#ug-outlier-removal>`_
+
+::
+
+    WARNING:Model Compression Toolkit:There are output values that deviate significantly from the average. Refer to the following images and the TroubleShooting Documentation (MCT XQuant Extension Tool) of 'Outlier Removal'.
+
+
 * `Shift Negative Activation <https://sonysemiconductorsolutions.github.io/mct-model-optimization/docs_troubleshoot/troubleshoots/shift_negative_activation.html#ug-shift-negative-activation>`_
+
+::
+
+    WARNING:Model Compression Toolkit:There are activations that contain negative values. Refer to the troubleshooting manual of "Shift Negative Activation".
+
 * `Unbalanced "concatenation" <https://sonysemiconductorsolutions.github.io/mct-model-optimization/docs_troubleshoot/troubleshoots/unbalanced_concatenation.html#ug-unbalanced-concatenation>`_
+
+::
+
+    WARNING:Model Compression Toolkit:There are unbalanced range layers concatnated. Refer to the troubleshooting manual of 'Unbalanced "concatenation"'.
+
 * `Mixed Precision with model output loss objective <https://sonysemiconductorsolutions.github.io/mct-model-optimization/docs_troubleshoot/troubleshoots/mixed_precision_with_model_output_loss_objective.html#ug-mixed-precision-with-model-output-loss-objective>`_
 
+::
+
+    WARNING:Model Compression Toolkit:the quantization bitwidth of the last layer is an extremely small number. Refer to the troubleshooting manual of 'Mixed Precision with model output loss objective'.
+
 Understanding the General Troubleshoots
 ============================================
 

docs/docs_troubleshoot/_sources/troubleshoots/unbalanced_concatenation.rst.txt

@@ -46,3 +46,9 @@ When you discover the concatenation layers with inputs that have notably differe
 
 
 2. Add a scaling operation between problematic "concatenation" layers in your float model.
+
+When the Unbalanced "concatenation" layers detected, the following formula message will be displayed.
+You can add the displayed scaling operation between the layers to your network.
+
+::
+    WARNING:Model Compression Toolkit:first layer:features.15.conv.2, second layer:features.15.conv.3, if you add a scaling operation, recommended scaling:first layer * 5.758747418625537

docs/docs_troubleshoot/searchindex.js

@@ -79,6 +79,12 @@ <h2>Solution<a class="headerlink" href="#solution" title="Link to this heading">
 <ol class="arabic simple" start="2">
 <li><p>Add a scaling operation between problematic “concatenation” layers in your float model.</p></li>
 </ol>
+<p>When the Unbalanced “concatenation” layers detected, the following formula message will be displayed.
+You can add the displayed scaling operation between the layers to your network.</p>
+<dl class="simple">
+<dt>::</dt><dd><p>WARNING:Model Compression Toolkit:first layer:features.15.conv.2, second layer:features.15.conv.3, if you add a scaling operation, recommended scaling:first layer * 5.758747418625537</p>
+</dd>
+</dl>
 </section>
 </section>
 

docs/docs_troubleshoot/troubleshoots/unbalanced_concatenation.html

@@ -123,7 +123,7 @@ <h2>How to Run<a class="headerlink" href="#how-to-run" title="Link to this headi
 </div>
 <div class="admonition note">
 <p class="admonition-title">Note</p>
-<p>If log of <em>mct.set_log_folder</em> does not exist, the <em>Unbalanced Concatnation</em> described below will not be executed.</p>
+<p>If log of <em>mct.set_log_folder</em> does not exist, the <em>Unbalanced Concatenation</em> described below will not be executed.</p>
 </div>
 </section>
 <section id="xquantconfig-format-and-examples">
@@ -202,14 +202,33 @@ <h2>Understanding the Quantization Error Graph<a class="headerlink" href="#under
 </section>
 <section id="understanding-the-judgeable-troubleshoots">
 <h2>Understanding the Judgeable Troubleshoots<a class="headerlink" href="#understanding-the-judgeable-troubleshoots" title="Link to this heading">¶</a></h2>
-<p>The following items are automatically identified by the XQuant Extension Tool.</p>
-<p>Please refer to the respective Troubleshooting Manuals for details.</p>
+<p>The following items are automatically identified by the XQuant Extension Tool.
+When this tool detects these issues, corresponding WARNING messages are displayed in your console.
+Please refer to the respective Troubleshooting Manuals and change the configuration as needed.</p>
 <ul class="simple">
 <li><p><a class="reference external" href="https://sonysemiconductorsolutions.github.io/mct-model-optimization/docs_troubleshoot/troubleshoots/outlier_removal.html#ug-outlier-removal">Outlier Removal</a></p></li>
+</ul>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">WARNING</span><span class="p">:</span><span class="n">Model</span> <span class="n">Compression</span> <span class="n">Toolkit</span><span class="p">:</span><span class="n">There</span> <span class="n">are</span> <span class="n">output</span> <span class="n">values</span> <span class="n">that</span> <span class="n">deviate</span> <span class="n">significantly</span> <span class="kn">from</span><span class="w"> </span><span class="nn">the</span> <span class="n">average</span><span class="o">.</span> <span class="n">Refer</span> <span class="n">to</span> <span class="n">the</span> <span class="n">following</span> <span class="n">images</span> <span class="ow">and</span> <span class="n">the</span> <span class="n">TroubleShooting</span> <span class="n">Documentation</span> <span class="p">(</span><span class="n">MCT</span> <span class="n">XQuant</span> <span class="n">Extension</span> <span class="n">Tool</span><span class="p">)</span> <span class="n">of</span> <span class="s1">&#39;Outlier Removal&#39;</span><span class="o">.</span>
+</pre></div>
+</div>
+<ul class="simple">
 <li><p><a class="reference external" href="https://sonysemiconductorsolutions.github.io/mct-model-optimization/docs_troubleshoot/troubleshoots/shift_negative_activation.html#ug-shift-negative-activation">Shift Negative Activation</a></p></li>
+</ul>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">WARNING</span><span class="p">:</span><span class="n">Model</span> <span class="n">Compression</span> <span class="n">Toolkit</span><span class="p">:</span><span class="n">There</span> <span class="n">are</span> <span class="n">activations</span> <span class="n">that</span> <span class="n">contain</span> <span class="n">negative</span> <span class="n">values</span><span class="o">.</span> <span class="n">Refer</span> <span class="n">to</span> <span class="n">the</span> <span class="n">troubleshooting</span> <span class="n">manual</span> <span class="n">of</span> <span class="s2">&quot;Shift Negative Activation&quot;</span><span class="o">.</span>
+</pre></div>
+</div>
+<ul class="simple">
 <li><p><a class="reference external" href="https://sonysemiconductorsolutions.github.io/mct-model-optimization/docs_troubleshoot/troubleshoots/unbalanced_concatenation.html#ug-unbalanced-concatenation">Unbalanced “concatenation”</a></p></li>
+</ul>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">WARNING</span><span class="p">:</span><span class="n">Model</span> <span class="n">Compression</span> <span class="n">Toolkit</span><span class="p">:</span><span class="n">There</span> <span class="n">are</span> <span class="n">unbalanced</span> <span class="nb">range</span> <span class="n">layers</span> <span class="n">concatnated</span><span class="o">.</span> <span class="n">Refer</span> <span class="n">to</span> <span class="n">the</span> <span class="n">troubleshooting</span> <span class="n">manual</span> <span class="n">of</span> <span class="s1">&#39;Unbalanced &quot;concatenation&quot;&#39;</span><span class="o">.</span>
+</pre></div>
+</div>
+<ul class="simple">
 <li><p><a class="reference external" href="https://sonysemiconductorsolutions.github.io/mct-model-optimization/docs_troubleshoot/troubleshoots/mixed_precision_with_model_output_loss_objective.html#ug-mixed-precision-with-model-output-loss-objective">Mixed Precision with model output loss objective</a></p></li>
 </ul>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">WARNING</span><span class="p">:</span><span class="n">Model</span> <span class="n">Compression</span> <span class="n">Toolkit</span><span class="p">:</span><span class="n">the</span> <span class="n">quantization</span> <span class="n">bitwidth</span> <span class="n">of</span> <span class="n">the</span> <span class="n">last</span> <span class="n">layer</span> <span class="ow">is</span> <span class="n">an</span> <span class="n">extremely</span> <span class="n">small</span> <span class="n">number</span><span class="o">.</span> <span class="n">Refer</span> <span class="n">to</span> <span class="n">the</span> <span class="n">troubleshooting</span> <span class="n">manual</span> <span class="n">of</span> <span class="s1">&#39;Mixed Precision with model output loss objective&#39;</span><span class="o">.</span>
+</pre></div>
+</div>
 </section>
 <section id="understanding-the-general-troubleshoots">
 <h2>Understanding the General Troubleshoots<a class="headerlink" href="#understanding-the-general-troubleshoots" title="Link to this heading">¶</a></h2>

docs/guidelines/XQuant_Extension_Tool.html

@@ -81,7 +81,7 @@ To be more specific, execute the following steps:
 
 .. note::
 
-  If log of *mct.set_log_folder* does not exist, the *Unbalanced Concatnation* described below will not be executed.
+  If log of *mct.set_log_folder* does not exist, the *Unbalanced Concatenation* described below will not be executed.
 
 XQuantConfig Format and Examples
 ======================================
@@ -158,14 +158,34 @@ Understanding the Judgeable Troubleshoots
 =======================================================
 
 The following items are automatically identified by the XQuant Extension Tool.
-
-Please refer to the respective Troubleshooting Manuals for details.
+When this tool detects these issues, corresponding WARNING messages are displayed in your console.
+Please refer to the respective Troubleshooting Manuals and change the configuration as needed.
 
 * `Outlier Removal <https://sonysemiconductorsolutions.github.io/mct-model-optimization/docs_troubleshoot/troubleshoots/outlier_removal.html#ug-outlier-removal>`_
+
+::
+
+    WARNING:Model Compression Toolkit:There are output values that deviate significantly from the average. Refer to the following images and the TroubleShooting Documentation (MCT XQuant Extension Tool) of 'Outlier Removal'.
+
+
 * `Shift Negative Activation <https://sonysemiconductorsolutions.github.io/mct-model-optimization/docs_troubleshoot/troubleshoots/shift_negative_activation.html#ug-shift-negative-activation>`_
+
+::
+
+    WARNING:Model Compression Toolkit:There are activations that contain negative values. Refer to the troubleshooting manual of "Shift Negative Activation".
+
 * `Unbalanced "concatenation" <https://sonysemiconductorsolutions.github.io/mct-model-optimization/docs_troubleshoot/troubleshoots/unbalanced_concatenation.html#ug-unbalanced-concatenation>`_
+
+::
+
+    WARNING:Model Compression Toolkit:There are unbalanced range layers concatnated. Refer to the troubleshooting manual of 'Unbalanced "concatenation"'.
+
 * `Mixed Precision with model output loss objective <https://sonysemiconductorsolutions.github.io/mct-model-optimization/docs_troubleshoot/troubleshoots/mixed_precision_with_model_output_loss_objective.html#ug-mixed-precision-with-model-output-loss-objective>`_
 
+::
+
+    WARNING:Model Compression Toolkit:the quantization bitwidth of the last layer is an extremely small number. Refer to the troubleshooting manual of 'Mixed Precision with model output loss objective'.
+
 Understanding the General Troubleshoots
 ============================================
 

docs/searchindex.js

@@ -46,3 +46,9 @@ When you discover the concatenation layers with inputs that have notably differe
 
 
 2. Add a scaling operation between problematic "concatenation" layers in your float model.
+
+When the Unbalanced "concatenation" layers detected, the following formula message will be displayed.
+You can add the displayed scaling operation between the layers to your network.
+
+::
+    WARNING:Model Compression Toolkit:first layer:features.15.conv.2, second layer:features.15.conv.3, if you add a scaling operation, recommended scaling:first layer * 5.758747418625537