Deployed 1514810 with MkDocs version: 1.0.4

LightningModule/properties/index.html

@@ -344,6 +344,26 @@
   
 </li>
       
+        <li class="md-nav__item">
+  <a href="#debugging" title="Debugging" class="md-nav__link">
+    Debugging
+  </a>
+  
+    <nav class="md-nav">
+      <ul class="md-nav__list">
+        
+          <li class="md-nav__item">
+  <a href="#example_input_array" title="example_input_array" class="md-nav__link">
+    example_input_array
+  </a>
+  
+</li>
+        
+      </ul>
+    </nav>
+  
+</li>
+      
       
       
       
@@ -595,6 +615,26 @@
   
 </li>
       
+        <li class="md-nav__item">
+  <a href="#debugging" title="Debugging" class="md-nav__link">
+    Debugging
+  </a>
+  
+    <nav class="md-nav">
+      <ul class="md-nav__list">
+        
+          <li class="md-nav__item">
+  <a href="#example_input_array" title="example_input_array" class="md-nav__link">
+    example_input_array
+  </a>
+  
+</li>
+        
+      </ul>
+    </nav>
+  
+</li>
+      
       
       
       
@@ -647,6 +687,17 @@ self.experiment.add_scalars(...)
 self.trainer.current_epoch
 ...   
 </code></pre>
+
+<h2 id="debugging">Debugging</h2>
+<p>The LightningModule also offers these tricks to help debug.   </p>
+<hr />
+<h4 id="example_input_array">example_input_array</h4>
+<p>In the LightningModule init, you can set a dummy tensor for this property
+to get a print out of sizes coming into and out of every layer.   </p>
+<pre><code class="python">def __init__(self):
+    # put the dimensions of the first input to your system
+    self.example_input_array = torch.rand(5, 28 * 28)
+</code></pre>
                 
                   
                 

Trainer/Logging/index.html

@@ -675,6 +675,14 @@ Trainer(experiment=exp)
 
 <hr />
 <h3 id="tensorboard-support">Tensorboard support</h3>
+<p>In the LightningModule you can access the experiment logger by doing:</p>
+<pre><code class="python">self.experiment
+
+# add image
+# Look at PyTorch SummaryWriter docs for what you can do.   
+self.experiment.add_image(...)
+</code></pre>
+
 <p>The experiment object is a strict subclass of PyTorch SummaryWriter. However, this class
 also snapshots every detail about the experiment (data folder paths, code, hyperparams),
 and allows you to visualize it using tensorboard.</p>