wassname/pytorch-lightning
1
0
Fork 0
mirror of https://github.com/wassname/pytorch-lightning.git synced 2026-09-09 11:32:07 +08:00
Code Issues Packages Projects Releases Wiki Activity

Deployed 5a9320d with MkDocs version: 1.0.4

Browse Source
This commit is contained in:
William Falcon
2019-09-26 10:28:41 -05:00
parent 8f53e506cf
commit a5cc9ec2dc
23 changed files with 417 additions and 349 deletions
Download Patch File Download Diff File Expand all files Collapse all files
404.html
+4 -4
View File
@@ -34,7 +34,7 @@
<meta name="lang:search.tokenizer" content="[\s\-]+">
<link rel="shortcut icon" href="/assets/images/favicon.png">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.0">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.2">
@@ -42,7 +42,7 @@
<link rel="stylesheet" href="/assets/stylesheets/application.0284f74d.css">
<link rel="stylesheet" href="/assets/stylesheets/application.30686662.css">
@@ -167,7 +167,7 @@
<main class="md-main">
<main class="md-main" role="main">
<div class="md-main__inner md-grid" data-md-component="container">
@@ -501,7 +501,7 @@
</div>
<script src="/assets/javascripts/application.245445c6.js"></script>
<script src="/assets/javascripts/application.c648116f.js"></script>
<script>app.initialize({version:"1.0.4",url:{base:"/"}})</script>
LightningModule/RequiredTrainerInterface/index.html
+97 -97
View File
@@ -34,7 +34,7 @@
<meta name="lang:search.tokenizer" content="[\s\-]+">
<link rel="shortcut icon" href="../../assets/images/favicon.png">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.0">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.2">
@@ -42,7 +42,7 @@
<link rel="stylesheet" href="../../assets/stylesheets/application.0284f74d.css">
<link rel="stylesheet" href="../../assets/stylesheets/application.30686662.css">
@@ -171,7 +171,7 @@
<main class="md-main">
<main class="md-main" role="main">
<div class="md-main__inner md-grid" data-md-component="container">
@@ -276,21 +276,21 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#minimal-example" title="Minimal example" class="md-nav__link">
<a href="#minimal-example" class="md-nav__link">
Minimal example
</a>
</li>
<li class="md-nav__item">
<a href="#how-do-these-methods-fit-into-the-broader-training" title="How do these methods fit into the broader training?" class="md-nav__link">
<a href="#how-do-these-methods-fit-into-the-broader-training" class="md-nav__link">
How do these methods fit into the broader training?
</a>
</li>
<li class="md-nav__item">
<a href="#required-methods" title="Required Methods" class="md-nav__link">
<a href="#required-methods" class="md-nav__link">
Required Methods
</a>
@@ -298,22 +298,22 @@
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#training_step" title="training_step" class="md-nav__link">
<a href="#training_step" class="md-nav__link">
training_step
</a>
</li>
<li class="md-nav__item">
<a href="#tng_dataloader" title="tng_dataloader" class="md-nav__link">
tng_dataloader
<a href="#train_dataloader" class="md-nav__link">
train_dataloader
</a>
<nav class="md-nav">
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#return" title="Return" class="md-nav__link">
<a href="#return" class="md-nav__link">
Return
</a>
@@ -325,7 +325,7 @@
</li>
<li class="md-nav__item">
<a href="#configure_optimizers" title="configure_optimizers" class="md-nav__link">
<a href="#configure_optimizers" class="md-nav__link">
configure_optimizers
</a>
@@ -333,7 +333,7 @@
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#return_1" title="Return" class="md-nav__link">
<a href="#return_1" class="md-nav__link">
Return
</a>
@@ -350,7 +350,7 @@
</li>
<li class="md-nav__item">
<a href="#optional-methods" title="Optional Methods" class="md-nav__link">
<a href="#optional-methods" class="md-nav__link">
Optional Methods
</a>
@@ -358,35 +358,35 @@
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#validation_step" title="validation_step" class="md-nav__link">
<a href="#validation_step" class="md-nav__link">
validation_step
</a>
</li>
<li class="md-nav__item">
<a href="#validation_end" title="validation_end" class="md-nav__link">
<a href="#validation_end" class="md-nav__link">
validation_end
</a>
</li>
<li class="md-nav__item">
<a href="#test_step" title="test_step" class="md-nav__link">
<a href="#test_step" class="md-nav__link">
test_step
</a>
</li>
<li class="md-nav__item">
<a href="#test_end" title="test_end" class="md-nav__link">
<a href="#test_end" class="md-nav__link">
test_end
</a>
</li>
<li class="md-nav__item">
<a href="#on_save_checkpoint" title="on_save_checkpoint" class="md-nav__link">
<a href="#on_save_checkpoint" class="md-nav__link">
on_save_checkpoint
</a>
@@ -394,7 +394,7 @@
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#return_2" title="Return" class="md-nav__link">
<a href="#return_2" class="md-nav__link">
Return
</a>
@@ -406,7 +406,7 @@
</li>
<li class="md-nav__item">
<a href="#on_load_checkpoint" title="on_load_checkpoint" class="md-nav__link">
<a href="#on_load_checkpoint" class="md-nav__link">
on_load_checkpoint
</a>
@@ -414,7 +414,7 @@
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#return_3" title="Return" class="md-nav__link">
<a href="#return_3" class="md-nav__link">
Return
</a>
@@ -426,7 +426,7 @@
</li>
<li class="md-nav__item">
<a href="#val_dataloader" title="val_dataloader" class="md-nav__link">
<a href="#val_dataloader" class="md-nav__link">
val_dataloader
</a>
@@ -434,7 +434,7 @@
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#return_4" title="Return" class="md-nav__link">
<a href="#return_4" class="md-nav__link">
Return
</a>
@@ -446,7 +446,7 @@
</li>
<li class="md-nav__item">
<a href="#test_dataloader" title="test_dataloader" class="md-nav__link">
<a href="#test_dataloader" class="md-nav__link">
test_dataloader
</a>
@@ -454,7 +454,7 @@
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#return_5" title="Return" class="md-nav__link">
<a href="#return_5" class="md-nav__link">
Return
</a>
@@ -466,15 +466,15 @@
</li>
<li class="md-nav__item">
<a href="#update_tng_log_metrics" title="update_tng_log_metrics" class="md-nav__link">
update_tng_log_metrics
<a href="#update_training_log_metrics" class="md-nav__link">
update_training_log_metrics
</a>
<nav class="md-nav">
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#return_6" title="Return" class="md-nav__link">
<a href="#return_6" class="md-nav__link">
Return
</a>
@@ -486,7 +486,7 @@
</li>
<li class="md-nav__item">
<a href="#add_model_specific_args" title="add_model_specific_args" class="md-nav__link">
<a href="#add_model_specific_args" class="md-nav__link">
add_model_specific_args
</a>
@@ -494,7 +494,7 @@
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#return_7" title="Return" class="md-nav__link">
<a href="#return_7" class="md-nav__link">
Return
</a>
@@ -751,21 +751,21 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#minimal-example" title="Minimal example" class="md-nav__link">
<a href="#minimal-example" class="md-nav__link">
Minimal example
</a>
</li>
<li class="md-nav__item">
<a href="#how-do-these-methods-fit-into-the-broader-training" title="How do these methods fit into the broader training?" class="md-nav__link">
<a href="#how-do-these-methods-fit-into-the-broader-training" class="md-nav__link">
How do these methods fit into the broader training?
</a>
</li>
<li class="md-nav__item">
<a href="#required-methods" title="Required Methods" class="md-nav__link">
<a href="#required-methods" class="md-nav__link">
Required Methods
</a>
@@ -773,22 +773,22 @@
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#training_step" title="training_step" class="md-nav__link">
<a href="#training_step" class="md-nav__link">
training_step
</a>
</li>
<li class="md-nav__item">
<a href="#tng_dataloader" title="tng_dataloader" class="md-nav__link">
tng_dataloader
<a href="#train_dataloader" class="md-nav__link">
train_dataloader
</a>
<nav class="md-nav">
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#return" title="Return" class="md-nav__link">
<a href="#return" class="md-nav__link">
Return
</a>
@@ -800,7 +800,7 @@
</li>
<li class="md-nav__item">
<a href="#configure_optimizers" title="configure_optimizers" class="md-nav__link">
<a href="#configure_optimizers" class="md-nav__link">
configure_optimizers
</a>
@@ -808,7 +808,7 @@
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#return_1" title="Return" class="md-nav__link">
<a href="#return_1" class="md-nav__link">
Return
</a>
@@ -825,7 +825,7 @@
</li>
<li class="md-nav__item">
<a href="#optional-methods" title="Optional Methods" class="md-nav__link">
<a href="#optional-methods" class="md-nav__link">
Optional Methods
</a>
@@ -833,35 +833,35 @@
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#validation_step" title="validation_step" class="md-nav__link">
<a href="#validation_step" class="md-nav__link">
validation_step
</a>
</li>
<li class="md-nav__item">
<a href="#validation_end" title="validation_end" class="md-nav__link">
<a href="#validation_end" class="md-nav__link">
validation_end
</a>
</li>
<li class="md-nav__item">
<a href="#test_step" title="test_step" class="md-nav__link">
<a href="#test_step" class="md-nav__link">
test_step
</a>
</li>
<li class="md-nav__item">
<a href="#test_end" title="test_end" class="md-nav__link">
<a href="#test_end" class="md-nav__link">
test_end
</a>
</li>
<li class="md-nav__item">
<a href="#on_save_checkpoint" title="on_save_checkpoint" class="md-nav__link">
<a href="#on_save_checkpoint" class="md-nav__link">
on_save_checkpoint
</a>
@@ -869,7 +869,7 @@
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#return_2" title="Return" class="md-nav__link">
<a href="#return_2" class="md-nav__link">
Return
</a>
@@ -881,7 +881,7 @@
</li>
<li class="md-nav__item">
<a href="#on_load_checkpoint" title="on_load_checkpoint" class="md-nav__link">
<a href="#on_load_checkpoint" class="md-nav__link">
on_load_checkpoint
</a>
@@ -889,7 +889,7 @@
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#return_3" title="Return" class="md-nav__link">
<a href="#return_3" class="md-nav__link">
Return
</a>
@@ -901,7 +901,7 @@
</li>
<li class="md-nav__item">
<a href="#val_dataloader" title="val_dataloader" class="md-nav__link">
<a href="#val_dataloader" class="md-nav__link">
val_dataloader
</a>
@@ -909,7 +909,7 @@
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#return_4" title="Return" class="md-nav__link">
<a href="#return_4" class="md-nav__link">
Return
</a>
@@ -921,7 +921,7 @@
</li>
<li class="md-nav__item">
<a href="#test_dataloader" title="test_dataloader" class="md-nav__link">
<a href="#test_dataloader" class="md-nav__link">
test_dataloader
</a>
@@ -929,7 +929,7 @@
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#return_5" title="Return" class="md-nav__link">
<a href="#return_5" class="md-nav__link">
Return
</a>
@@ -941,15 +941,15 @@
</li>
<li class="md-nav__item">
<a href="#update_tng_log_metrics" title="update_tng_log_metrics" class="md-nav__link">
update_tng_log_metrics
<a href="#update_training_log_metrics" class="md-nav__link">
update_training_log_metrics
</a>
<nav class="md-nav">
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#return_6" title="Return" class="md-nav__link">
<a href="#return_6" class="md-nav__link">
Return
</a>
@@ -961,7 +961,7 @@
</li>
<li class="md-nav__item">
<a href="#add_model_specific_args" title="add_model_specific_args" class="md-nav__link">
<a href="#add_model_specific_args" class="md-nav__link">
add_model_specific_args
</a>
@@ -969,7 +969,7 @@
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#return_7" title="Return" class="md-nav__link">
<a href="#return_7" class="md-nav__link">
Return
</a>
@@ -1012,7 +1012,7 @@
<p><strong>Required</strong>: </p>
<ul>
<li><a href="./#training_step">training_step</a> </li>
<li><a href="./#tng_dataloader">tng_dataloader</a> </li>
<li><a href="./#train_dataloader">train_dataloader</a> </li>
<li><a href="./#configure_optimizers">configure_optimizers</a> </li>
</ul>
<p><strong>Optional</strong>: </p>
@@ -1025,7 +1025,7 @@
<li><a href="./#test_dataloader">test_dataloader</a> </li>
<li><a href="./#on_save_checkpoint">on_save_checkpoint</a> </li>
<li><a href="./#on_load_checkpoint">on_load_checkpoint</a> </li>
<li><a href="./#update_tng_log_metrics">update_tng_log_metrics</a> </li>
<li><a href="./#update_training_log_metrics">update_training_log_metrics</a> </li>
<li><a href="./#add_model_specific_args">add_model_specific_args</a> </li>
</ul>
<hr />
@@ -1147,7 +1147,7 @@
<span class="k">return</span> <span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">optim</span><span class="o">.</span><span class="n">Adam</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">parameters</span><span class="p">(),</span> <span class="n">lr</span><span class="o">=</span><span class="mf">0.02</span><span class="p">)]</span>
<span class="nd">@pl.data_loader</span>
<span class="k">def</span> <span class="nf">tng_dataloader</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="k">def</span> <span class="nf">train_dataloader</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="k">return</span> <span class="n">DataLoader</span><span class="p">(</span><span class="n">MNIST</span><span class="p">(</span><span class="n">os</span><span class="o">.</span><span class="n">getcwd</span><span class="p">(),</span> <span class="n">train</span><span class="o">=</span><span class="bp">True</span><span class="p">,</span> <span class="n">download</span><span class="o">=</span><span class="bp">True</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transforms</span><span class="o">.</span><span class="n">ToTensor</span><span class="p">()),</span> <span class="n">batch_size</span><span class="o">=</span><span class="mi">32</span><span class="p">)</span>
<span class="nd">@pl.data_loader</span>
@@ -1175,7 +1175,7 @@
<h2 id="required-methods">Required Methods</h2>
<h3 id="training_step">training_step</h3>
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre>1</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="k">def</span> <span class="nf">training_step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">data_batch</span><span class="p">,</span> <span class="n">batch_nb</span><span class="p">)</span>
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre>1</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="k">def</span> <span class="nf">training_step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">batch</span><span class="p">,</span> <span class="n">batch_nb</span><span class="p">)</span>
</pre></div>
</td></tr></table>
@@ -1190,7 +1190,7 @@
</thead>
<tbody>
<tr>
<td>data_batch</td>
<td>batch</td>
<td>The output of your dataloader. A tensor, tuple or list</td>
</tr>
<tr>
@@ -1216,7 +1216,7 @@
<td>Y</td>
</tr>
<tr>
<td>prog</td>
<td>progress</td>
<td>Dict for progress bar display. Must have only tensors</td>
<td>N</td>
</tr>
@@ -1236,8 +1236,8 @@
11
12
13
14</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="k">def</span> <span class="nf">training_step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">data_batch</span><span class="p">,</span> <span class="n">batch_nb</span><span class="p">):</span>
<span class="n">x</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">z</span> <span class="o">=</span> <span class="n">data_batch</span>
14</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="k">def</span> <span class="nf">training_step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">batch</span><span class="p">,</span> <span class="n">batch_nb</span><span class="p">):</span>
<span class="n">x</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">z</span> <span class="o">=</span> <span class="n">batch</span>
<span class="c1"># implement your own</span>
<span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">forward</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
@@ -1245,7 +1245,7 @@
<span class="n">output</span> <span class="o">=</span> <span class="p">{</span>
<span class="s1">&#39;loss&#39;</span><span class="p">:</span> <span class="n">loss</span><span class="p">,</span> <span class="c1"># required</span>
<span class="s1">&#39;prog&#39;</span><span class="p">:</span> <span class="p">{</span><span class="s1">&#39;tng_loss&#39;</span><span class="p">:</span> <span class="n">loss</span><span class="p">,</span> <span class="s1">&#39;batch_nb&#39;</span><span class="p">:</span> <span class="n">batch_nb</span><span class="p">}</span> <span class="c1"># optional</span>
<span class="s1">&#39;progress&#39;</span><span class="p">:</span> <span class="p">{</span><span class="s1">&#39;training_loss&#39;</span><span class="p">:</span> <span class="n">loss</span><span class="p">}</span> <span class="c1"># optional (MUST ALL BE TENSORS)</span>
<span class="p">}</span>
<span class="c1"># return a dict</span>
@@ -1260,7 +1260,7 @@
4
5
6</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="c1"># Multiple optimizers (ie: GANs) </span>
<span class="k">def</span> <span class="nf">training_step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">data_batch</span><span class="p">,</span> <span class="n">batch_nb</span><span class="p">,</span> <span class="n">optimizer_idx</span><span class="p">):</span>
<span class="k">def</span> <span class="nf">training_step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">batch</span><span class="p">,</span> <span class="n">batch_nb</span><span class="p">,</span> <span class="n">optimizer_idx</span><span class="p">):</span>
<span class="k">if</span> <span class="n">optimizer_idx</span> <span class="o">==</span> <span class="mi">0</span><span class="p">:</span>
<span class="c1"># do training_step with encoder</span>
<span class="k">if</span> <span class="n">optimizer_idx</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
@@ -1269,10 +1269,10 @@
</td></tr></table>
<hr />
<h3 id="tng_dataloader">tng_dataloader</h3>
<h3 id="train_dataloader">train_dataloader</h3>
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre>1
2</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="nd">@pl.data_loader</span>
<span class="k">def</span> <span class="nf">tng_dataloader</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
<span class="k">def</span> <span class="nf">train_dataloader</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
</pre></div>
</td></tr></table>
@@ -1290,7 +1290,7 @@
8
9
10</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="nd">@pl.data_loader</span>
<span class="k">def</span> <span class="nf">tng_dataloader</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="k">def</span> <span class="nf">train_dataloader</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
<span class="n">transform</span> <span class="o">=</span> <span class="n">transforms</span><span class="o">.</span><span class="n">Compose</span><span class="p">([</span><span class="n">transforms</span><span class="o">.</span><span class="n">ToTensor</span><span class="p">(),</span> <span class="n">transforms</span><span class="o">.</span><span class="n">Normalize</span><span class="p">((</span><span class="mf">0.5</span><span class="p">,),</span> <span class="p">(</span><span class="mf">1.0</span><span class="p">,))])</span>
<span class="n">dataset</span> <span class="o">=</span> <span class="n">MNIST</span><span class="p">(</span><span class="n">root</span><span class="o">=</span><span class="s1">&#39;/path/to/mnist/&#39;</span><span class="p">,</span> <span class="n">train</span><span class="o">=</span><span class="bp">True</span><span class="p">,</span> <span class="n">transform</span><span class="o">=</span><span class="n">transform</span><span class="p">,</span> <span class="n">download</span><span class="o">=</span><span class="bp">True</span><span class="p">)</span>
<span class="n">loader</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">utils</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">DataLoader</span><span class="p">(</span>
@@ -1362,10 +1362,10 @@ the <a href="https://williamfalcon.github.io/pytorch-lightning/Trainer/hooks/#op
3
4
5</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="c1"># if you have one val dataloader:</span>
<span class="k">def</span> <span class="nf">validation_step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">data_batch</span><span class="p">,</span> <span class="n">batch_nb</span><span class="p">)</span>
<span class="k">def</span> <span class="nf">validation_step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">batch</span><span class="p">,</span> <span class="n">batch_nb</span><span class="p">)</span>
<span class="c1"># if you have multiple val dataloaders: </span>
<span class="k">def</span> <span class="nf">validation_step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">data_batch</span><span class="p">,</span> <span class="n">batch_nb</span><span class="p">,</span> <span class="n">dataloader_idx</span><span class="p">)</span>
<span class="k">def</span> <span class="nf">validation_step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">batch</span><span class="p">,</span> <span class="n">batch_nb</span><span class="p">,</span> <span class="n">dataloader_idxdx</span><span class="p">)</span>
</pre></div>
</td></tr></table>
@@ -1383,7 +1383,7 @@ If you don't need to validate you don't need to implement this method. In this s
</thead>
<tbody>
<tr>
<td>data_batch</td>
<td>batch</td>
<td>The output of your dataloader. A tensor, tuple or list</td>
</tr>
<tr>
@@ -1391,7 +1391,7 @@ If you don't need to validate you don't need to implement this method. In this s
<td>Integer displaying which batch this is</td>
</tr>
<tr>
<td>dataloader_i</td>
<td>dataloader_idx</td>
<td>Integer displaying which dataloader this is (only if multiple val datasets used)</td>
</tr>
</tbody>
@@ -1441,8 +1441,8 @@ If you don't need to validate you don't need to implement this method. In this s
25
26
27</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="c1"># CASE 1: A single validation dataset</span>
<span class="k">def</span> <span class="nf">validation_step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">data_batch</span><span class="p">,</span> <span class="n">batch_nb</span><span class="p">):</span>
<span class="n">x</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">data_batch</span>
<span class="k">def</span> <span class="nf">validation_step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">batch</span><span class="p">,</span> <span class="n">batch_nb</span><span class="p">):</span>
<span class="n">x</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">batch</span>
<span class="c1"># implement your own</span>
<span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">forward</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
@@ -1474,7 +1474,7 @@ If you don't need to validate you don't need to implement this method. In this s
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre>1
2
3</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="c1"># CASE 2: multiple validation datasets</span>
<span class="k">def</span> <span class="nf">validation_step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">data_batch</span><span class="p">,</span> <span class="n">batch_nb</span><span class="p">,</span> <span class="n">dataset_idx</span><span class="p">):</span>
<span class="k">def</span> <span class="nf">validation_step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">batch</span><span class="p">,</span> <span class="n">batch_nb</span><span class="p">,</span> <span class="n">dataset_idx</span><span class="p">):</span>
<span class="c1"># dataset_idx tells you which dataset this is. </span>
</pre></div>
</td></tr></table>
@@ -1552,8 +1552,8 @@ If you don't need to validate you don't need to implement this method. In this s
<span class="n">val_loss_mean</span> <span class="o">/=</span> <span class="nb">len</span><span class="p">(</span><span class="n">outputs</span><span class="p">)</span>
<span class="n">val_acc_mean</span> <span class="o">/=</span> <span class="nb">len</span><span class="p">(</span><span class="n">outputs</span><span class="p">)</span>
<span class="n">tqdm_dic</span> <span class="o">=</span> <span class="p">{</span><span class="s1">&#39;val_loss&#39;</span><span class="p">:</span> <span class="n">val_loss_mean</span><span class="o">.</span><span class="n">item</span><span class="p">(),</span> <span class="s1">&#39;val_acc&#39;</span><span class="p">:</span> <span class="n">val_acc_mean</span><span class="o">.</span><span class="n">item</span><span class="p">()}</span>
<span class="k">return</span> <span class="n">tqdm_dic</span>
<span class="n">tqdm_dict</span> <span class="o">=</span> <span class="p">{</span><span class="s1">&#39;val_loss&#39;</span><span class="p">:</span> <span class="n">val_loss_mean</span><span class="o">.</span><span class="n">item</span><span class="p">(),</span> <span class="s1">&#39;val_acc&#39;</span><span class="p">:</span> <span class="n">val_acc_mean</span><span class="o">.</span><span class="n">item</span><span class="p">()}</span>
<span class="k">return</span> <span class="n">tqdm_dict</span>
</pre></div>
</td></tr></table>
@@ -1595,8 +1595,8 @@ each validation step for that dataloader.</p>
<span class="n">val_loss_mean</span> <span class="o">/=</span> <span class="n">i</span>
<span class="n">val_acc_mean</span> <span class="o">/=</span> <span class="n">i</span>
<span class="n">tqdm_dic</span> <span class="o">=</span> <span class="p">{</span><span class="s1">&#39;val_loss&#39;</span><span class="p">:</span> <span class="n">val_loss_mean</span><span class="o">.</span><span class="n">item</span><span class="p">(),</span> <span class="s1">&#39;val_acc&#39;</span><span class="p">:</span> <span class="n">val_acc_mean</span><span class="o">.</span><span class="n">item</span><span class="p">()}</span>
<span class="k">return</span> <span class="n">tqdm_dic</span>
<span class="n">tqdm_dict</span> <span class="o">=</span> <span class="p">{</span><span class="s1">&#39;val_loss&#39;</span><span class="p">:</span> <span class="n">val_loss_mean</span><span class="o">.</span><span class="n">item</span><span class="p">(),</span> <span class="s1">&#39;val_acc&#39;</span><span class="p">:</span> <span class="n">val_acc_mean</span><span class="o">.</span><span class="n">item</span><span class="p">()}</span>
<span class="k">return</span> <span class="n">tqdm_dict</span>
</pre></div>
</td></tr></table>
@@ -1606,10 +1606,10 @@ each validation step for that dataloader.</p>
3
4
5</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="c1"># if you have one test dataloader:</span>
<span class="k">def</span> <span class="nf">test_step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">data_batch</span><span class="p">,</span> <span class="n">batch_nb</span><span class="p">)</span>
<span class="k">def</span> <span class="nf">test_step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">batch</span><span class="p">,</span> <span class="n">batch_nb</span><span class="p">)</span>
<span class="c1"># if you have multiple test dataloaders: </span>
<span class="k">def</span> <span class="nf">test_step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">data_batch</span><span class="p">,</span> <span class="n">batch_nb</span><span class="p">,</span> <span class="n">dataloader_idx</span><span class="p">)</span>
<span class="k">def</span> <span class="nf">test_step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">batch</span><span class="p">,</span> <span class="n">batch_nb</span><span class="p">,</span> <span class="n">dataloader_idxdx</span><span class="p">)</span>
</pre></div>
</td></tr></table>
@@ -1628,7 +1628,7 @@ If you don't need to test you don't need to implement this method. In this step
</thead>
<tbody>
<tr>
<td>data_batch</td>
<td>batch</td>
<td>The output of your dataloader. A tensor, tuple or list</td>
</tr>
<tr>
@@ -1636,7 +1636,7 @@ If you don't need to test you don't need to implement this method. In this step
<td>Integer displaying which batch this is</td>
</tr>
<tr>
<td>dataloader_i</td>
<td>dataloader_idx</td>
<td>Integer displaying which dataloader this is (only if multiple test datasets used)</td>
</tr>
</tbody>
@@ -1680,8 +1680,8 @@ If you don't need to test you don't need to implement this method. In this step
19
20
21</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="c1"># CASE 1: A single test dataset</span>
<span class="k">def</span> <span class="nf">test_step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">data_batch</span><span class="p">,</span> <span class="n">batch_nb</span><span class="p">):</span>
<span class="n">x</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">data_batch</span>
<span class="k">def</span> <span class="nf">test_step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">batch</span><span class="p">,</span> <span class="n">batch_nb</span><span class="p">):</span>
<span class="n">x</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">batch</span>
<span class="c1"># implement your own</span>
<span class="n">out</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">forward</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
@@ -1707,7 +1707,7 @@ If you don't need to test you don't need to implement this method. In this step
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre>1
2
3</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="c1"># CASE 2: multiple test datasets</span>
<span class="k">def</span> <span class="nf">test_step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">data_batch</span><span class="p">,</span> <span class="n">batch_nb</span><span class="p">,</span> <span class="n">dataset_idx</span><span class="p">):</span>
<span class="k">def</span> <span class="nf">test_step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">batch</span><span class="p">,</span> <span class="n">batch_nb</span><span class="p">,</span> <span class="n">dataset_idx</span><span class="p">):</span>
<span class="c1"># dataset_idx tells you which dataset this is. </span>
</pre></div>
</td></tr></table>
@@ -1784,8 +1784,8 @@ If you don't need to test you don't need to implement this method. In this step
<span class="n">test_loss_mean</span> <span class="o">/=</span> <span class="nb">len</span><span class="p">(</span><span class="n">outputs</span><span class="p">)</span>
<span class="n">test_acc_mean</span> <span class="o">/=</span> <span class="nb">len</span><span class="p">(</span><span class="n">outputs</span><span class="p">)</span>
<span class="n">tqdm_dic</span> <span class="o">=</span> <span class="p">{</span><span class="s1">&#39;test_loss&#39;</span><span class="p">:</span> <span class="n">test_loss_mean</span><span class="o">.</span><span class="n">item</span><span class="p">(),</span> <span class="s1">&#39;test_acc&#39;</span><span class="p">:</span> <span class="n">test_acc_mean</span><span class="o">.</span><span class="n">item</span><span class="p">()}</span>
<span class="k">return</span> <span class="n">tqdm_dic</span>
<span class="n">tqdm_dict</span> <span class="o">=</span> <span class="p">{</span><span class="s1">&#39;test_loss&#39;</span><span class="p">:</span> <span class="n">test_loss_mean</span><span class="o">.</span><span class="n">item</span><span class="p">(),</span> <span class="s1">&#39;test_acc&#39;</span><span class="p">:</span> <span class="n">test_acc_mean</span><span class="o">.</span><span class="n">item</span><span class="p">()}</span>
<span class="k">return</span> <span class="n">tqdm_dict</span>
</pre></div>
</td></tr></table>
@@ -1827,8 +1827,8 @@ each validation step for that dataloader.</p>
<span class="n">test_loss_mean</span> <span class="o">/=</span> <span class="n">i</span>
<span class="n">test_acc_mean</span> <span class="o">/=</span> <span class="n">i</span>
<span class="n">tqdm_dic</span> <span class="o">=</span> <span class="p">{</span><span class="s1">&#39;test_loss&#39;</span><span class="p">:</span> <span class="n">test_loss_mean</span><span class="o">.</span><span class="n">item</span><span class="p">(),</span> <span class="s1">&#39;test_acc&#39;</span><span class="p">:</span> <span class="n">test_acc_mean</span><span class="o">.</span><span class="n">item</span><span class="p">()}</span>
<span class="k">return</span> <span class="n">tqdm_dic</span>
<span class="n">tqdm_dict</span> <span class="o">=</span> <span class="p">{</span><span class="s1">&#39;test_loss&#39;</span><span class="p">:</span> <span class="n">test_loss_mean</span><span class="o">.</span><span class="n">item</span><span class="p">(),</span> <span class="s1">&#39;test_acc&#39;</span><span class="p">:</span> <span class="n">test_acc_mean</span><span class="o">.</span><span class="n">item</span><span class="p">()}</span>
<span class="k">return</span> <span class="n">tqdm_dict</span>
</pre></div>
</td></tr></table>
@@ -1961,8 +1961,8 @@ If you don't need a test dataset and a test_step, you don't need to implement th
</td></tr></table>
<hr />
<h3 id="update_tng_log_metrics">update_tng_log_metrics</h3>
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre>1</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="k">def</span> <span class="nf">update_tng_log_metrics</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">logs</span><span class="p">)</span>
<h3 id="update_training_log_metrics">update_training_log_metrics</h3>
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre>1</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="k">def</span> <span class="nf">update_training_log_metrics</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">logs</span><span class="p">)</span>
</pre></div>
</td></tr></table>
@@ -1973,7 +1973,7 @@ This is a chance to amend or add to the metrics about to be logged.</p>
<p><strong>Example</strong></p>
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre>1
2
3</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="k">def</span> <span class="nf">update_tng_log_metrics</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">logs</span><span class="p">):</span>
3</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="k">def</span> <span class="nf">update_training_log_metrics</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">logs</span><span class="p">):</span>
<span class="c1"># modify or add to logs</span>
<span class="k">return</span> <span class="n">logs</span>
</pre></div>
@@ -2019,7 +2019,7 @@ The <a href="https://williamfalcon.github.io/test-tube/hyperparameter_optimizati
<span class="n">parser</span> <span class="o">=</span> <span class="n">HyperOptArgumentParser</span><span class="p">(</span><span class="n">strategy</span><span class="o">=</span><span class="n">parent_parser</span><span class="o">.</span><span class="n">strategy</span><span class="p">,</span> <span class="n">parents</span><span class="o">=</span><span class="p">[</span><span class="n">parent_parser</span><span class="p">])</span>
<span class="c1"># param overwrites</span>
<span class="c1"># parser.set_defaults(gradient_clip=5.0)</span>
<span class="c1"># parser.set_defaults(gradient_clip_val=5.0)</span>
<span class="c1"># network params</span>
<span class="n">parser</span><span class="o">.</span><span class="n">opt_list</span><span class="p">(</span><span class="s1">&#39;--drop_prob&#39;</span><span class="p">,</span> <span class="n">default</span><span class="o">=</span><span class="mf">0.2</span><span class="p">,</span> <span class="n">options</span><span class="o">=</span><span class="p">[</span><span class="mf">0.2</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">],</span> <span class="nb">type</span><span class="o">=</span><span class="nb">float</span><span class="p">,</span> <span class="n">tunable</span><span class="o">=</span><span class="bp">False</span><span class="p">)</span>
@@ -2107,7 +2107,7 @@ The <a href="https://williamfalcon.github.io/test-tube/hyperparameter_optimizati
</div>
<script src="../../assets/javascripts/application.245445c6.js"></script>
<script src="../../assets/javascripts/application.c648116f.js"></script>
<script>app.initialize({version:"1.0.4",url:{base:"../.."}})</script>
LightningModule/methods/index.html
+10 -10
View File
@@ -34,7 +34,7 @@
<meta name="lang:search.tokenizer" content="[\s\-]+">
<link rel="shortcut icon" href="../../assets/images/favicon.png">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.0">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.2">
@@ -42,7 +42,7 @@
<link rel="stylesheet" href="../../assets/stylesheets/application.0284f74d.css">
<link rel="stylesheet" href="../../assets/stylesheets/application.30686662.css">
@@ -171,7 +171,7 @@
<main class="md-main">
<main class="md-main" role="main">
<div class="md-main__inner md-grid" data-md-component="container">
@@ -284,21 +284,21 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#freeze" title="freeze" class="md-nav__link">
<a href="#freeze" class="md-nav__link">
freeze
</a>
</li>
<li class="md-nav__item">
<a href="#load_from_metrics" title="load_from_metrics" class="md-nav__link">
<a href="#load_from_metrics" class="md-nav__link">
load_from_metrics
</a>
</li>
<li class="md-nav__item">
<a href="#unfreeze" title="unfreeze" class="md-nav__link">
<a href="#unfreeze" class="md-nav__link">
unfreeze
</a>
@@ -531,21 +531,21 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#freeze" title="freeze" class="md-nav__link">
<a href="#freeze" class="md-nav__link">
freeze
</a>
</li>
<li class="md-nav__item">
<a href="#load_from_metrics" title="load_from_metrics" class="md-nav__link">
<a href="#load_from_metrics" class="md-nav__link">
load_from_metrics
</a>
</li>
<li class="md-nav__item">
<a href="#unfreeze" title="unfreeze" class="md-nav__link">
<a href="#unfreeze" class="md-nav__link">
unfreeze
</a>
@@ -716,7 +716,7 @@ The meta_tags.csv file can be found in the test-tube experiment save_dir.
</div>
<script src="../../assets/javascripts/application.245445c6.js"></script>
<script src="../../assets/javascripts/application.c648116f.js"></script>
<script>app.initialize({version:"1.0.4",url:{base:"../.."}})</script>
LightningModule/properties/index.html
+25 -25
View File
@@ -34,7 +34,7 @@
<meta name="lang:search.tokenizer" content="[\s\-]+">
<link rel="shortcut icon" href="../../assets/images/favicon.png">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.0">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.2">
@@ -42,7 +42,7 @@
<link rel="stylesheet" href="../../assets/stylesheets/application.0284f74d.css">
<link rel="stylesheet" href="../../assets/stylesheets/application.30686662.css">
@@ -171,7 +171,7 @@
<main class="md-main">
<main class="md-main" role="main">
<div class="md-main__inner md-grid" data-md-component="container">
@@ -296,56 +296,56 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#current_epoch" title="current_epoch" class="md-nav__link">
<a href="#current_epoch" class="md-nav__link">
current_epoch
</a>
</li>
<li class="md-nav__item">
<a href="#dtype" title="dtype" class="md-nav__link">
<a href="#dtype" class="md-nav__link">
dtype
</a>
</li>
<li class="md-nav__item">
<a href="#experiment" title="experiment" class="md-nav__link">
<a href="#experiment" class="md-nav__link">
experiment
</a>
</li>
<li class="md-nav__item">
<a href="#global_step" title="global_step" class="md-nav__link">
<a href="#global_step" class="md-nav__link">
global_step
</a>
</li>
<li class="md-nav__item">
<a href="#gradient_clip" title="gradient_clip" class="md-nav__link">
gradient_clip
<a href="#gradient_clip_val" class="md-nav__link">
gradient_clip_val
</a>
</li>
<li class="md-nav__item">
<a href="#on_gpu" title="on_gpu" class="md-nav__link">
<a href="#on_gpu" class="md-nav__link">
on_gpu
</a>
</li>
<li class="md-nav__item">
<a href="#trainer" title="trainer" class="md-nav__link">
<a href="#trainer" class="md-nav__link">
trainer
</a>
</li>
<li class="md-nav__item">
<a href="#debugging" title="Debugging" class="md-nav__link">
<a href="#debugging" class="md-nav__link">
Debugging
</a>
@@ -353,7 +353,7 @@
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#example_input_array" title="example_input_array" class="md-nav__link">
<a href="#example_input_array" class="md-nav__link">
example_input_array
</a>
@@ -579,56 +579,56 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#current_epoch" title="current_epoch" class="md-nav__link">
<a href="#current_epoch" class="md-nav__link">
current_epoch
</a>
</li>
<li class="md-nav__item">
<a href="#dtype" title="dtype" class="md-nav__link">
<a href="#dtype" class="md-nav__link">
dtype
</a>
</li>
<li class="md-nav__item">
<a href="#experiment" title="experiment" class="md-nav__link">
<a href="#experiment" class="md-nav__link">
experiment
</a>
</li>
<li class="md-nav__item">
<a href="#global_step" title="global_step" class="md-nav__link">
<a href="#global_step" class="md-nav__link">
global_step
</a>
</li>
<li class="md-nav__item">
<a href="#gradient_clip" title="gradient_clip" class="md-nav__link">
gradient_clip
<a href="#gradient_clip_val" class="md-nav__link">
gradient_clip_val
</a>
</li>
<li class="md-nav__item">
<a href="#on_gpu" title="on_gpu" class="md-nav__link">
<a href="#on_gpu" class="md-nav__link">
on_gpu
</a>
</li>
<li class="md-nav__item">
<a href="#trainer" title="trainer" class="md-nav__link">
<a href="#trainer" class="md-nav__link">
trainer
</a>
</li>
<li class="md-nav__item">
<a href="#debugging" title="Debugging" class="md-nav__link">
<a href="#debugging" class="md-nav__link">
Debugging
</a>
@@ -636,7 +636,7 @@
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#example_input_array" title="example_input_array" class="md-nav__link">
<a href="#example_input_array" class="md-nav__link">
example_input_array
</a>
@@ -690,7 +690,7 @@
<h4 id="global_step">global_step</h4>
<p>Total training batches seen across all epochs </p>
<hr />
<h4 id="gradient_clip">gradient_clip</h4>
<h4 id="gradient_clip_val">gradient_clip_val</h4>
<p>The current gradient clip value </p>
<hr />
<h4 id="on_gpu">on_gpu</h4>
@@ -788,7 +788,7 @@ to get a print out of sizes coming into and out of every layer. </p>
</div>
<script src="../../assets/javascripts/application.245445c6.js"></script>
<script src="../../assets/javascripts/application.c648116f.js"></script>
<script>app.initialize({version:"1.0.4",url:{base:"../.."}})</script>
Trainer/Checkpointing/index.html
+8 -8
View File
@@ -34,7 +34,7 @@
<meta name="lang:search.tokenizer" content="[\s\-]+">
<link rel="shortcut icon" href="../../assets/images/favicon.png">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.0">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.2">
@@ -42,7 +42,7 @@
<link rel="stylesheet" href="../../assets/stylesheets/application.0284f74d.css">
<link rel="stylesheet" href="../../assets/stylesheets/application.30686662.css">
@@ -171,7 +171,7 @@
<main class="md-main">
<main class="md-main" role="main">
<div class="md-main__inner md-grid" data-md-component="container">
@@ -344,14 +344,14 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#model-saving" title="Model saving" class="md-nav__link">
<a href="#model-saving" class="md-nav__link">
Model saving
</a>
</li>
<li class="md-nav__item">
<a href="#restoring-training-session" title="Restoring training session" class="md-nav__link">
<a href="#restoring-training-session" class="md-nav__link">
Restoring training session
</a>
@@ -524,14 +524,14 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#model-saving" title="Model saving" class="md-nav__link">
<a href="#model-saving" class="md-nav__link">
Model saving
</a>
</li>
<li class="md-nav__item">
<a href="#restoring-training-session" title="Restoring training session" class="md-nav__link">
<a href="#restoring-training-session" class="md-nav__link">
Restoring training session
</a>
@@ -725,7 +725,7 @@ the system isn't different. If you add a layer, for instance, it might not work.
</div>
<script src="../../assets/javascripts/application.245445c6.js"></script>
<script src="../../assets/javascripts/application.c648116f.js"></script>
<script>app.initialize({version:"1.0.4",url:{base:"../.."}})</script>
Trainer/Distributed training/index.html
+47 -23
View File
@@ -34,7 +34,7 @@
<meta name="lang:search.tokenizer" content="[\s\-]+">
<link rel="shortcut icon" href="../../assets/images/favicon.png">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.0">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.2">
@@ -42,7 +42,7 @@
<link rel="stylesheet" href="../../assets/stylesheets/application.0284f74d.css">
<link rel="stylesheet" href="../../assets/stylesheets/application.30686662.css">
@@ -171,7 +171,7 @@
<main class="md-main">
<main class="md-main" role="main">
<div class="md-main__inner md-grid" data-md-component="container">
@@ -356,56 +356,56 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#choosing-a-backend" title="Choosing a backend" class="md-nav__link">
<a href="#choosing-a-backend" class="md-nav__link">
Choosing a backend
</a>
</li>
<li class="md-nav__item">
<a href="#distributed-and-16-bit-precision" title="Distributed and 16-bit precision." class="md-nav__link">
<a href="#distributed-and-16-bit-precision" class="md-nav__link">
Distributed and 16-bit precision.
</a>
</li>
<li class="md-nav__item">
<a href="#cuda-flags" title="CUDA flags" class="md-nav__link">
<a href="#cuda-flags" class="md-nav__link">
CUDA flags
</a>
</li>
<li class="md-nav__item">
<a href="#16-bit-mixed-precision" title="16-bit mixed precision" class="md-nav__link">
<a href="#16-bit-mixed-precision" class="md-nav__link">
16-bit mixed precision
</a>
</li>
<li class="md-nav__item">
<a href="#single-gpu" title="Single-gpu" class="md-nav__link">
<a href="#single-gpu" class="md-nav__link">
Single-gpu
</a>
</li>
<li class="md-nav__item">
<a href="#multi-gpu" title="multi-gpu" class="md-nav__link">
<a href="#multi-gpu" class="md-nav__link">
multi-gpu
</a>
</li>
<li class="md-nav__item">
<a href="#multi-node" title="Multi-node" class="md-nav__link">
<a href="#multi-node" class="md-nav__link">
Multi-node
</a>
</li>
<li class="md-nav__item">
<a href="#self-balancing-architecture" title="Self-balancing architecture" class="md-nav__link">
<a href="#self-balancing-architecture" class="md-nav__link">
Self-balancing architecture
</a>
@@ -566,56 +566,56 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#choosing-a-backend" title="Choosing a backend" class="md-nav__link">
<a href="#choosing-a-backend" class="md-nav__link">
Choosing a backend
</a>
</li>
<li class="md-nav__item">
<a href="#distributed-and-16-bit-precision" title="Distributed and 16-bit precision." class="md-nav__link">
<a href="#distributed-and-16-bit-precision" class="md-nav__link">
Distributed and 16-bit precision.
</a>
</li>
<li class="md-nav__item">
<a href="#cuda-flags" title="CUDA flags" class="md-nav__link">
<a href="#cuda-flags" class="md-nav__link">
CUDA flags
</a>
</li>
<li class="md-nav__item">
<a href="#16-bit-mixed-precision" title="16-bit mixed precision" class="md-nav__link">
<a href="#16-bit-mixed-precision" class="md-nav__link">
16-bit mixed precision
</a>
</li>
<li class="md-nav__item">
<a href="#single-gpu" title="Single-gpu" class="md-nav__link">
<a href="#single-gpu" class="md-nav__link">
Single-gpu
</a>
</li>
<li class="md-nav__item">
<a href="#multi-gpu" title="multi-gpu" class="md-nav__link">
<a href="#multi-gpu" class="md-nav__link">
multi-gpu
</a>
</li>
<li class="md-nav__item">
<a href="#multi-node" title="Multi-node" class="md-nav__link">
<a href="#multi-node" class="md-nav__link">
Multi-node
</a>
</li>
<li class="md-nav__item">
<a href="#self-balancing-architecture" title="Self-balancing architecture" class="md-nav__link">
<a href="#self-balancing-architecture" class="md-nav__link">
Self-balancing architecture
</a>
@@ -768,10 +768,34 @@ SLURM will set these for you. </p>
<h4 id="16-bit-mixed-precision">16-bit mixed precision</h4>
<p>16 bit precision can cut your memory footprint by half. If using volta architecture GPUs it can give a dramatic training speed-up as well. <br />
First, install apex (if install fails, look <a href="https://github.com/NVIDIA/apex">here</a>):</p>
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre>1
2
3</pre></div></td><td class="code"><div class="codehilite"><pre><span></span>$ git clone https://github.com/NVIDIA/apex
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre> 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15</pre></div></td><td class="code"><div class="codehilite"><pre><span></span>$ git clone https://github.com/NVIDIA/apex
$ <span class="nb">cd</span> apex
<span class="c1"># ------------------------</span>
<span class="c1"># OPTIONAL: on your cluster you might need to load cuda 10 or 9</span>
<span class="c1"># depending on how you installed PyTorch</span>
<span class="c1"># see available modules</span>
module avail
<span class="c1"># load correct cuda before install</span>
module load cuda-10.0
<span class="c1"># ------------------------</span>
$ pip install -v --no-cache-dir --global-option<span class="o">=</span><span class="s2">&quot;--cpp_ext&quot;</span> --global-option<span class="o">=</span><span class="s2">&quot;--cuda_ext&quot;</span> ./
</pre></div>
</td></tr></table>
@@ -963,7 +987,7 @@ portion of your dataset onto each GPU. (World_size = gpus_per_node * nb_nodes).
</div>
<script src="../../assets/javascripts/application.245445c6.js"></script>
<script src="../../assets/javascripts/application.c648116f.js"></script>
<script>app.initialize({version:"1.0.4",url:{base:"../.."}})</script>
Trainer/Logging/index.html
+21 -21
View File
@@ -34,7 +34,7 @@
<meta name="lang:search.tokenizer" content="[\s\-]+">
<link rel="shortcut icon" href="../../assets/images/favicon.png">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.0">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.2">
@@ -42,7 +42,7 @@
<link rel="stylesheet" href="../../assets/stylesheets/application.0284f74d.css">
<link rel="stylesheet" href="../../assets/stylesheets/application.30686662.css">
@@ -171,7 +171,7 @@
<main class="md-main">
<main class="md-main" role="main">
<div class="md-main__inner md-grid" data-md-component="container">
@@ -368,49 +368,49 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#display-metrics-in-progress-bar" title="Display metrics in progress bar" class="md-nav__link">
<a href="#display-metrics-in-progress-bar" class="md-nav__link">
Display metrics in progress bar
</a>
</li>
<li class="md-nav__item">
<a href="#log-metric-row-every-k-batches" title="Log metric row every k batches" class="md-nav__link">
<a href="#log-metric-row-every-k-batches" class="md-nav__link">
Log metric row every k batches
</a>
</li>
<li class="md-nav__item">
<a href="#log-metric-row-every-k-batches_1" title="Log metric row every k batches" class="md-nav__link">
<a href="#log-metric-row-every-k-batches_1" class="md-nav__link">
Log metric row every k batches
</a>
</li>
<li class="md-nav__item">
<a href="#process-position" title="Process position" class="md-nav__link">
<a href="#process-position" class="md-nav__link">
Process position
</a>
</li>
<li class="md-nav__item">
<a href="#save-a-snapshot-of-all-hyperparameters" title="Save a snapshot of all hyperparameters" class="md-nav__link">
<a href="#save-a-snapshot-of-all-hyperparameters" class="md-nav__link">
Save a snapshot of all hyperparameters
</a>
</li>
<li class="md-nav__item">
<a href="#snapshot-code-for-a-training-run" title="Snapshot code for a training run" class="md-nav__link">
<a href="#snapshot-code-for-a-training-run" class="md-nav__link">
Snapshot code for a training run
</a>
</li>
<li class="md-nav__item">
<a href="#tensorboard-support" title="Tensorboard support" class="md-nav__link">
<a href="#tensorboard-support" class="md-nav__link">
Tensorboard support
</a>
@@ -418,7 +418,7 @@
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#write-logs-file-to-csv-every-k-batches" title="Write logs file to csv every k batches" class="md-nav__link">
<a href="#write-logs-file-to-csv-every-k-batches" class="md-nav__link">
Write logs file to csv every k batches
</a>
@@ -572,49 +572,49 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#display-metrics-in-progress-bar" title="Display metrics in progress bar" class="md-nav__link">
<a href="#display-metrics-in-progress-bar" class="md-nav__link">
Display metrics in progress bar
</a>
</li>
<li class="md-nav__item">
<a href="#log-metric-row-every-k-batches" title="Log metric row every k batches" class="md-nav__link">
<a href="#log-metric-row-every-k-batches" class="md-nav__link">
Log metric row every k batches
</a>
</li>
<li class="md-nav__item">
<a href="#log-metric-row-every-k-batches_1" title="Log metric row every k batches" class="md-nav__link">
<a href="#log-metric-row-every-k-batches_1" class="md-nav__link">
Log metric row every k batches
</a>
</li>
<li class="md-nav__item">
<a href="#process-position" title="Process position" class="md-nav__link">
<a href="#process-position" class="md-nav__link">
Process position
</a>
</li>
<li class="md-nav__item">
<a href="#save-a-snapshot-of-all-hyperparameters" title="Save a snapshot of all hyperparameters" class="md-nav__link">
<a href="#save-a-snapshot-of-all-hyperparameters" class="md-nav__link">
Save a snapshot of all hyperparameters
</a>
</li>
<li class="md-nav__item">
<a href="#snapshot-code-for-a-training-run" title="Snapshot code for a training run" class="md-nav__link">
<a href="#snapshot-code-for-a-training-run" class="md-nav__link">
Snapshot code for a training run
</a>
</li>
<li class="md-nav__item">
<a href="#tensorboard-support" title="Tensorboard support" class="md-nav__link">
<a href="#tensorboard-support" class="md-nav__link">
Tensorboard support
</a>
@@ -622,7 +622,7 @@
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#write-logs-file-to-csv-every-k-batches" title="Write logs file to csv every k batches" class="md-nav__link">
<a href="#write-logs-file-to-csv-every-k-batches" class="md-nav__link">
Write logs file to csv every k batches
</a>
@@ -668,7 +668,7 @@
<p>Every k batches lightning will make an entry in the metrics log</p>
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre>1
2</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="c1"># DEFAULT (ie: save a .csv log file every 10 batches)</span>
<span class="n">trainer</span> <span class="o">=</span> <span class="n">Trainer</span><span class="p">(</span><span class="n">add_log_row_interval</span><span class="o">=</span><span class="mi">10</span><span class="p">)</span>
<span class="n">trainer</span> <span class="o">=</span> <span class="n">Trainer</span><span class="p">(</span><span class="n">row_log_interval</span><span class="o">=</span><span class="mi">10</span><span class="p">)</span>
</pre></div>
</td></tr></table>
@@ -862,7 +862,7 @@ and allows you to visualize it using tensorboard.</p>
</div>
<script src="../../assets/javascripts/application.245445c6.js"></script>
<script src="../../assets/javascripts/application.c648116f.js"></script>
<script>app.initialize({version:"1.0.4",url:{base:"../.."}})</script>
Trainer/SLURM Managed Cluster/index.html
+8 -8
View File
@@ -34,7 +34,7 @@
<meta name="lang:search.tokenizer" content="[\s\-]+">
<link rel="shortcut icon" href="../../assets/images/favicon.png">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.0">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.2">
@@ -42,7 +42,7 @@
<link rel="stylesheet" href="../../assets/stylesheets/application.0284f74d.css">
<link rel="stylesheet" href="../../assets/stylesheets/application.30686662.css">
@@ -171,7 +171,7 @@
<main class="md-main">
<main class="md-main" role="main">
<div class="md-main__inner md-grid" data-md-component="container">
@@ -380,14 +380,14 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#running-grid-search-on-a-cluster" title="Running grid search on a cluster" class="md-nav__link">
<a href="#running-grid-search-on-a-cluster" class="md-nav__link">
Running grid search on a cluster
</a>
</li>
<li class="md-nav__item">
<a href="#walltime-auto-resubmit" title="Walltime auto-resubmit" class="md-nav__link">
<a href="#walltime-auto-resubmit" class="md-nav__link">
Walltime auto-resubmit
</a>
@@ -524,14 +524,14 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#running-grid-search-on-a-cluster" title="Running grid search on a cluster" class="md-nav__link">
<a href="#running-grid-search-on-a-cluster" class="md-nav__link">
Running grid search on a cluster
</a>
</li>
<li class="md-nav__item">
<a href="#walltime-auto-resubmit" title="Walltime auto-resubmit" class="md-nav__link">
<a href="#walltime-auto-resubmit" class="md-nav__link">
Walltime auto-resubmit
</a>
@@ -777,7 +777,7 @@ your SLURM script. </p>
</div>
<script src="../../assets/javascripts/application.245445c6.js"></script>
<script src="../../assets/javascripts/application.c648116f.js"></script>
<script>app.initialize({version:"1.0.4",url:{base:"../.."}})</script>
Trainer/Testing loop/index.html
+6 -6
View File
@@ -34,7 +34,7 @@
<meta name="lang:search.tokenizer" content="[\s\-]+">
<link rel="shortcut icon" href="../../assets/images/favicon.png">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.0">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.2">
@@ -42,7 +42,7 @@
<link rel="stylesheet" href="../../assets/stylesheets/application.0284f74d.css">
<link rel="stylesheet" href="../../assets/stylesheets/application.30686662.css">
@@ -171,7 +171,7 @@
<main class="md-main">
<main class="md-main" role="main">
<div class="md-main__inner md-grid" data-md-component="container">
@@ -392,7 +392,7 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#test" title="test" class="md-nav__link">
<a href="#test" class="md-nav__link">
test
</a>
@@ -517,7 +517,7 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#test" title="test" class="md-nav__link">
<a href="#test" class="md-nav__link">
test
</a>
@@ -658,7 +658,7 @@ First case is where you test right after a full training routine.</p>
</div>
<script src="../../assets/javascripts/application.245445c6.js"></script>
<script src="../../assets/javascripts/application.c648116f.js"></script>
<script>app.initialize({version:"1.0.4",url:{base:"../.."}})</script>
Trainer/Training Loop/index.html
+61 -18
View File
@@ -34,7 +34,7 @@
<meta name="lang:search.tokenizer" content="[\s\-]+">
<link rel="shortcut icon" href="../../assets/images/favicon.png">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.0">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.2">
@@ -42,7 +42,7 @@
<link rel="stylesheet" href="../../assets/stylesheets/application.0284f74d.css">
<link rel="stylesheet" href="../../assets/stylesheets/application.30686662.css">
@@ -171,7 +171,7 @@
<main class="md-main">
<main class="md-main" role="main">
<div class="md-main__inner md-grid" data-md-component="container">
@@ -404,42 +404,49 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#accumulated-gradients" title="Accumulated gradients" class="md-nav__link">
<a href="#accumulated-gradients" class="md-nav__link">
Accumulated gradients
</a>
</li>
<li class="md-nav__item">
<a href="#force-training-for-min-or-max-epochs" title="Force training for min or max epochs" class="md-nav__link">
<a href="#force-training-for-min-or-max-epochs" class="md-nav__link">
Force training for min or max epochs
</a>
</li>
<li class="md-nav__item">
<a href="#force-disable-early-stop" title="Force disable early stop" class="md-nav__link">
<a href="#early-stopping" class="md-nav__link">
Early stopping
</a>
</li>
<li class="md-nav__item">
<a href="#force-disable-early-stop" class="md-nav__link">
Force disable early stop
</a>
</li>
<li class="md-nav__item">
<a href="#gradient-clipping" title="Gradient Clipping" class="md-nav__link">
<a href="#gradient-clipping" class="md-nav__link">
Gradient Clipping
</a>
</li>
<li class="md-nav__item">
<a href="#inspect-gradient-norms" title="Inspect gradient norms" class="md-nav__link">
<a href="#inspect-gradient-norms" class="md-nav__link">
Inspect gradient norms
</a>
</li>
<li class="md-nav__item">
<a href="#set-how-much-of-the-training-set-to-check" title="Set how much of the training set to check" class="md-nav__link">
<a href="#set-how-much-of-the-training-set-to-check" class="md-nav__link">
Set how much of the training set to check
</a>
@@ -552,42 +559,49 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#accumulated-gradients" title="Accumulated gradients" class="md-nav__link">
<a href="#accumulated-gradients" class="md-nav__link">
Accumulated gradients
</a>
</li>
<li class="md-nav__item">
<a href="#force-training-for-min-or-max-epochs" title="Force training for min or max epochs" class="md-nav__link">
<a href="#force-training-for-min-or-max-epochs" class="md-nav__link">
Force training for min or max epochs
</a>
</li>
<li class="md-nav__item">
<a href="#force-disable-early-stop" title="Force disable early stop" class="md-nav__link">
<a href="#early-stopping" class="md-nav__link">
Early stopping
</a>
</li>
<li class="md-nav__item">
<a href="#force-disable-early-stop" class="md-nav__link">
Force disable early stop
</a>
</li>
<li class="md-nav__item">
<a href="#gradient-clipping" title="Gradient Clipping" class="md-nav__link">
<a href="#gradient-clipping" class="md-nav__link">
Gradient Clipping
</a>
</li>
<li class="md-nav__item">
<a href="#inspect-gradient-norms" title="Inspect gradient norms" class="md-nav__link">
<a href="#inspect-gradient-norms" class="md-nav__link">
Inspect gradient norms
</a>
</li>
<li class="md-nav__item">
<a href="#set-how-much-of-the-training-set-to-check" title="Set how much of the training set to check" class="md-nav__link">
<a href="#set-how-much-of-the-training-set-to-check" class="md-nav__link">
Set how much of the training set to check
</a>
@@ -634,6 +648,35 @@
</pre></div>
</td></tr></table>
<hr />
<h4 id="early-stopping">Early stopping</h4>
<p>To enable ealry-stopping, define the callback and give it to the trainer.</p>
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre> 1
2
3
4
5
6
7
8
9
10
11
12</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="kn">from</span> <span class="nn">pytorch_lightning.callbacks</span> <span class="kn">import</span> <span class="n">EarlyStopping</span>
<span class="c1"># DEFAULTS</span>
<span class="n">early_stop_callback</span> <span class="o">=</span> <span class="n">EarlyStopping</span><span class="p">(</span>
<span class="n">monitor</span><span class="o">=</span><span class="s1">&#39;val_loss&#39;</span><span class="p">,</span>
<span class="n">min_delta</span><span class="o">=</span><span class="mf">0.00</span><span class="p">,</span>
<span class="n">patience</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
<span class="n">verbose</span><span class="o">=</span><span class="bp">False</span><span class="p">,</span>
<span class="n">mode</span><span class="o">=</span><span class="s1">&#39;auto&#39;</span>
<span class="p">)</span>
<span class="n">trainer</span> <span class="o">=</span> <span class="n">Trainer</span><span class="p">(</span><span class="n">early_stop_callback</span><span class="o">=</span><span class="n">early_stop_callback</span><span class="p">)</span>
</pre></div>
</td></tr></table>
<hr />
<h4 id="force-disable-early-stop">Force disable early stop</h4>
<p>Use this to turn off early stopping and run training to the <a href="#force-training-for-min-or-max-epochs">max_epoch</a></p>
@@ -652,10 +695,10 @@ Specifically, this will <a href="https://pytorch.org/docs/stable/nn.html#torch.n
3
4
5</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="c1"># DEFAULT (ie: don&#39;t clip)</span>
<span class="n">trainer</span> <span class="o">=</span> <span class="n">Trainer</span><span class="p">(</span><span class="n">gradient_clip</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
<span class="n">trainer</span> <span class="o">=</span> <span class="n">Trainer</span><span class="p">(</span><span class="n">gradient_clip_val</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
<span class="c1"># clip gradients with norm above 0.5</span>
<span class="n">trainer</span> <span class="o">=</span> <span class="n">Trainer</span><span class="p">(</span><span class="n">gradient_clip</span><span class="o">=</span><span class="mf">0.5</span><span class="p">)</span>
<span class="n">trainer</span> <span class="o">=</span> <span class="n">Trainer</span><span class="p">(</span><span class="n">gradient_clip_val</span><span class="o">=</span><span class="mf">0.5</span><span class="p">)</span>
</pre></div>
</td></tr></table>
@@ -758,7 +801,7 @@ Specifically, this will <a href="https://pytorch.org/docs/stable/nn.html#torch.n
</div>
<script src="../../assets/javascripts/application.245445c6.js"></script>
<script src="../../assets/javascripts/application.c648116f.js"></script>
<script>app.initialize({version:"1.0.4",url:{base:"../.."}})</script>
Trainer/Validation loop/index.html
+14 -14
View File
@@ -34,7 +34,7 @@
<meta name="lang:search.tokenizer" content="[\s\-]+">
<link rel="shortcut icon" href="../../assets/images/favicon.png">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.0">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.2">
@@ -42,7 +42,7 @@
<link rel="stylesheet" href="../../assets/stylesheets/application.0284f74d.css">
<link rel="stylesheet" href="../../assets/stylesheets/application.30686662.css">
@@ -171,7 +171,7 @@
<main class="md-main">
<main class="md-main" role="main">
<div class="md-main__inner md-grid" data-md-component="container">
@@ -416,35 +416,35 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#check-validation-every-n-epochs" title="Check validation every n epochs" class="md-nav__link">
<a href="#check-validation-every-n-epochs" class="md-nav__link">
Check validation every n epochs
</a>
</li>
<li class="md-nav__item">
<a href="#set-how-much-of-the-validation-set-to-check" title="Set how much of the validation set to check" class="md-nav__link">
<a href="#set-how-much-of-the-validation-set-to-check" class="md-nav__link">
Set how much of the validation set to check
</a>
</li>
<li class="md-nav__item">
<a href="#set-how-much-of-the-test-set-to-check" title="Set how much of the test set to check" class="md-nav__link">
<a href="#set-how-much-of-the-test-set-to-check" class="md-nav__link">
Set how much of the test set to check
</a>
</li>
<li class="md-nav__item">
<a href="#set-validation-check-frequency-within-1-training-epoch" title="Set validation check frequency within 1 training epoch" class="md-nav__link">
<a href="#set-validation-check-frequency-within-1-training-epoch" class="md-nav__link">
Set validation check frequency within 1 training epoch
</a>
</li>
<li class="md-nav__item">
<a href="#set-the-number-of-validation-sanity-steps" title="Set the number of validation sanity steps" class="md-nav__link">
<a href="#set-the-number-of-validation-sanity-steps" class="md-nav__link">
Set the number of validation sanity steps
</a>
@@ -545,35 +545,35 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#check-validation-every-n-epochs" title="Check validation every n epochs" class="md-nav__link">
<a href="#check-validation-every-n-epochs" class="md-nav__link">
Check validation every n epochs
</a>
</li>
<li class="md-nav__item">
<a href="#set-how-much-of-the-validation-set-to-check" title="Set how much of the validation set to check" class="md-nav__link">
<a href="#set-how-much-of-the-validation-set-to-check" class="md-nav__link">
Set how much of the validation set to check
</a>
</li>
<li class="md-nav__item">
<a href="#set-how-much-of-the-test-set-to-check" title="Set how much of the test set to check" class="md-nav__link">
<a href="#set-how-much-of-the-test-set-to-check" class="md-nav__link">
Set how much of the test set to check
</a>
</li>
<li class="md-nav__item">
<a href="#set-validation-check-frequency-within-1-training-epoch" title="Set validation check frequency within 1 training epoch" class="md-nav__link">
<a href="#set-validation-check-frequency-within-1-training-epoch" class="md-nav__link">
Set validation check frequency within 1 training epoch
</a>
</li>
<li class="md-nav__item">
<a href="#set-the-number-of-validation-sanity-steps" title="Set the number of validation sanity steps" class="md-nav__link">
<a href="#set-the-number-of-validation-sanity-steps" class="md-nav__link">
Set the number of validation sanity steps
</a>
@@ -739,7 +739,7 @@ Lightning will run 5 steps of validation in the beginning of training as a sanit
</div>
<script src="../../assets/javascripts/application.245445c6.js"></script>
<script src="../../assets/javascripts/application.c648116f.js"></script>
<script>app.initialize({version:"1.0.4",url:{base:"../.."}})</script>
Trainer/debugging/index.html
+16 -16
View File
@@ -34,7 +34,7 @@
<meta name="lang:search.tokenizer" content="[\s\-]+">
<link rel="shortcut icon" href="../../assets/images/favicon.png">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.0">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.2">
@@ -42,7 +42,7 @@
<link rel="stylesheet" href="../../assets/stylesheets/application.0284f74d.css">
<link rel="stylesheet" href="../../assets/stylesheets/application.30686662.css">
@@ -171,7 +171,7 @@
<main class="md-main">
<main class="md-main" role="main">
<div class="md-main__inner md-grid" data-md-component="container">
@@ -428,42 +428,42 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#fast-dev-run" title="Fast dev run" class="md-nav__link">
<a href="#fast-dev-run" class="md-nav__link">
Fast dev run
</a>
</li>
<li class="md-nav__item">
<a href="#inspect-gradient-norms" title="Inspect gradient norms" class="md-nav__link">
<a href="#inspect-gradient-norms" class="md-nav__link">
Inspect gradient norms
</a>
</li>
<li class="md-nav__item">
<a href="#make-model-overfit-on-subset-of-data" title="Make model overfit on subset of data" class="md-nav__link">
<a href="#make-model-overfit-on-subset-of-data" class="md-nav__link">
Make model overfit on subset of data
</a>
</li>
<li class="md-nav__item">
<a href="#print-the-parameter-count-by-layer" title="Print the parameter count by layer" class="md-nav__link">
<a href="#print-the-parameter-count-by-layer" class="md-nav__link">
Print the parameter count by layer
</a>
</li>
<li class="md-nav__item">
<a href="#print-which-gradients-are-nan" title="Print which gradients are nan" class="md-nav__link">
<a href="#print-which-gradients-are-nan" class="md-nav__link">
Print which gradients are nan
</a>
</li>
<li class="md-nav__item">
<a href="#log-gpu-usage" title="Log GPU usage" class="md-nav__link">
<a href="#log-gpu-usage" class="md-nav__link">
Log GPU usage
</a>
@@ -552,42 +552,42 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#fast-dev-run" title="Fast dev run" class="md-nav__link">
<a href="#fast-dev-run" class="md-nav__link">
Fast dev run
</a>
</li>
<li class="md-nav__item">
<a href="#inspect-gradient-norms" title="Inspect gradient norms" class="md-nav__link">
<a href="#inspect-gradient-norms" class="md-nav__link">
Inspect gradient norms
</a>
</li>
<li class="md-nav__item">
<a href="#make-model-overfit-on-subset-of-data" title="Make model overfit on subset of data" class="md-nav__link">
<a href="#make-model-overfit-on-subset-of-data" class="md-nav__link">
Make model overfit on subset of data
</a>
</li>
<li class="md-nav__item">
<a href="#print-the-parameter-count-by-layer" title="Print the parameter count by layer" class="md-nav__link">
<a href="#print-the-parameter-count-by-layer" class="md-nav__link">
Print the parameter count by layer
</a>
</li>
<li class="md-nav__item">
<a href="#print-which-gradients-are-nan" title="Print which gradients are nan" class="md-nav__link">
<a href="#print-which-gradients-are-nan" class="md-nav__link">
Print which gradients are nan
</a>
</li>
<li class="md-nav__item">
<a href="#log-gpu-usage" title="Log GPU usage" class="md-nav__link">
<a href="#log-gpu-usage" class="md-nav__link">
Log GPU usage
</a>
@@ -740,7 +740,7 @@ Use this to debug a full run of your program quickly</p>
</div>
<script src="../../assets/javascripts/application.245445c6.js"></script>
<script src="../../assets/javascripts/application.c648116f.js"></script>
<script>app.initialize({version:"1.0.4",url:{base:"../.."}})</script>
Trainer/hooks/index.html
+28 -28
View File
@@ -34,7 +34,7 @@
<meta name="lang:search.tokenizer" content="[\s\-]+">
<link rel="shortcut icon" href="../../assets/images/favicon.png">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.0">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.2">
@@ -42,7 +42,7 @@
<link rel="stylesheet" href="../../assets/stylesheets/application.0284f74d.css">
<link rel="stylesheet" href="../../assets/stylesheets/application.30686662.css">
@@ -171,7 +171,7 @@
<main class="md-main">
<main class="md-main" role="main">
<div class="md-main__inner md-grid" data-md-component="container">
@@ -444,70 +444,70 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#on_epoch_start" title="on_epoch_start" class="md-nav__link">
<a href="#on_epoch_start" class="md-nav__link">
on_epoch_start
</a>
</li>
<li class="md-nav__item">
<a href="#on_epoch_end" title="on_epoch_end" class="md-nav__link">
<a href="#on_epoch_end" class="md-nav__link">
on_epoch_end
</a>
</li>
<li class="md-nav__item">
<a href="#on_batch_start" title="on_batch_start" class="md-nav__link">
<a href="#on_batch_start" class="md-nav__link">
on_batch_start
</a>
</li>
<li class="md-nav__item">
<a href="#on_batch_end" title="on_batch_end" class="md-nav__link">
<a href="#on_batch_end" class="md-nav__link">
on_batch_end
</a>
</li>
<li class="md-nav__item">
<a href="#on_pre_performance_check" title="on_pre_performance_check" class="md-nav__link">
<a href="#on_pre_performance_check" class="md-nav__link">
on_pre_performance_check
</a>
</li>
<li class="md-nav__item">
<a href="#on_post_performance_check" title="on_post_performance_check" class="md-nav__link">
<a href="#on_post_performance_check" class="md-nav__link">
on_post_performance_check
</a>
</li>
<li class="md-nav__item">
<a href="#on_tng_metrics" title="on_tng_metrics" class="md-nav__link">
on_tng_metrics
<a href="#on_training_metrics" class="md-nav__link">
on_training_metrics
</a>
</li>
<li class="md-nav__item">
<a href="#optimizer_step" title="optimizer_step" class="md-nav__link">
<a href="#optimizer_step" class="md-nav__link">
optimizer_step
</a>
</li>
<li class="md-nav__item">
<a href="#on_before_zero_grad" title="on_before_zero_grad" class="md-nav__link">
<a href="#on_before_zero_grad" class="md-nav__link">
on_before_zero_grad
</a>
</li>
<li class="md-nav__item">
<a href="#on_after_backward" title="on_after_backward" class="md-nav__link">
<a href="#on_after_backward" class="md-nav__link">
on_after_backward
</a>
@@ -586,70 +586,70 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#on_epoch_start" title="on_epoch_start" class="md-nav__link">
<a href="#on_epoch_start" class="md-nav__link">
on_epoch_start
</a>
</li>
<li class="md-nav__item">
<a href="#on_epoch_end" title="on_epoch_end" class="md-nav__link">
<a href="#on_epoch_end" class="md-nav__link">
on_epoch_end
</a>
</li>
<li class="md-nav__item">
<a href="#on_batch_start" title="on_batch_start" class="md-nav__link">
<a href="#on_batch_start" class="md-nav__link">
on_batch_start
</a>
</li>
<li class="md-nav__item">
<a href="#on_batch_end" title="on_batch_end" class="md-nav__link">
<a href="#on_batch_end" class="md-nav__link">
on_batch_end
</a>
</li>
<li class="md-nav__item">
<a href="#on_pre_performance_check" title="on_pre_performance_check" class="md-nav__link">
<a href="#on_pre_performance_check" class="md-nav__link">
on_pre_performance_check
</a>
</li>
<li class="md-nav__item">
<a href="#on_post_performance_check" title="on_post_performance_check" class="md-nav__link">
<a href="#on_post_performance_check" class="md-nav__link">
on_post_performance_check
</a>
</li>
<li class="md-nav__item">
<a href="#on_tng_metrics" title="on_tng_metrics" class="md-nav__link">
on_tng_metrics
<a href="#on_training_metrics" class="md-nav__link">
on_training_metrics
</a>
</li>
<li class="md-nav__item">
<a href="#optimizer_step" title="optimizer_step" class="md-nav__link">
<a href="#optimizer_step" class="md-nav__link">
optimizer_step
</a>
</li>
<li class="md-nav__item">
<a href="#on_before_zero_grad" title="on_before_zero_grad" class="md-nav__link">
<a href="#on_before_zero_grad" class="md-nav__link">
on_before_zero_grad
</a>
</li>
<li class="md-nav__item">
<a href="#on_after_backward" title="on_after_backward" class="md-nav__link">
<a href="#on_after_backward" class="md-nav__link">
on_after_backward
</a>
@@ -737,12 +737,12 @@ To enable a hook, simply override the method in your LightningModule and the tra
</td></tr></table>
<hr />
<h4 id="on_tng_metrics">on_tng_metrics</h4>
<h4 id="on_training_metrics">on_training_metrics</h4>
<p>Called in the training loop, right before metrics are logged.
Although you can log at any time by using self.experiment, you can use
this callback to modify what will be logged.</p>
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre>1
2</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="k">def</span> <span class="nf">on_tng_metrics</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">metrics</span><span class="p">):</span>
2</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="k">def</span> <span class="nf">on_training_metrics</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">metrics</span><span class="p">):</span>
<span class="c1"># do something before validation end</span>
</pre></div>
</td></tr></table>
@@ -922,7 +922,7 @@ This is the ideal place to inspect or log gradient information </p>
</div>
<script src="../../assets/javascripts/application.245445c6.js"></script>
<script src="../../assets/javascripts/application.c648116f.js"></script>
<script>app.initialize({version:"1.0.4",url:{base:"../.."}})</script>
Trainer/index.html
+4 -4
View File
@@ -34,7 +34,7 @@
<meta name="lang:search.tokenizer" content="[\s\-]+">
<link rel="shortcut icon" href="../assets/images/favicon.png">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.0">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.2">
@@ -42,7 +42,7 @@
<link rel="stylesheet" href="../assets/stylesheets/application.0284f74d.css">
<link rel="stylesheet" href="../assets/stylesheets/application.30686662.css">
@@ -171,7 +171,7 @@
<main class="md-main">
<main class="md-main" role="main">
<div class="md-main__inner md-grid" data-md-component="container">
@@ -655,7 +655,7 @@
</div>
<script src="../assets/javascripts/application.245445c6.js"></script>
<script src="../assets/javascripts/application.c648116f.js"></script>
<script>app.initialize({version:"1.0.4",url:{base:".."}})</script>
assets/javascripts/application.245445c6.js → assets/javascripts/application.c648116f.js
+1 -1
View File
File diff suppressed because one or more lines are too long
assets/stylesheets/application-palette.01803549.css → assets/stylesheets/application-palette.a8b3c06d.css
+1 -1
View File
File diff suppressed because one or more lines are too long
assets/stylesheets/application.0284f74d.css
-1
View File
File diff suppressed because one or more lines are too long
assets/stylesheets/application.30686662.css
+1
View File
File diff suppressed because one or more lines are too long
examples/Examples/index.html
+14 -14
View File
@@ -34,7 +34,7 @@
<meta name="lang:search.tokenizer" content="[\s\-]+">
<link rel="shortcut icon" href="../../assets/images/favicon.png">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.0">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.2">
@@ -42,7 +42,7 @@
<link rel="stylesheet" href="../../assets/stylesheets/application.0284f74d.css">
<link rel="stylesheet" href="../../assets/stylesheets/application.30686662.css">
@@ -171,7 +171,7 @@
<main class="md-main">
<main class="md-main" role="main">
<div class="md-main__inner md-grid" data-md-component="container">
@@ -476,14 +476,14 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#template-model-definition" title="Template model definition" class="md-nav__link">
<a href="#template-model-definition" class="md-nav__link">
Template model definition
</a>
</li>
<li class="md-nav__item">
<a href="#trainer-example" title="Trainer Example" class="md-nav__link">
<a href="#trainer-example" class="md-nav__link">
Trainer Example
</a>
@@ -491,21 +491,21 @@
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#cpu-hyperparameter-search" title="CPU hyperparameter search" class="md-nav__link">
<a href="#cpu-hyperparameter-search" class="md-nav__link">
CPU hyperparameter search
</a>
</li>
<li class="md-nav__item">
<a href="#hyperparameter-search-on-a-single-or-multiple-gpus" title="Hyperparameter search on a single or multiple GPUs" class="md-nav__link">
<a href="#hyperparameter-search-on-a-single-or-multiple-gpus" class="md-nav__link">
Hyperparameter search on a single or multiple GPUs
</a>
</li>
<li class="md-nav__item">
<a href="#hyperparameter-search-on-a-slurm-hpc-cluster" title="Hyperparameter search on a SLURM HPC cluster" class="md-nav__link">
<a href="#hyperparameter-search-on-a-slurm-hpc-cluster" class="md-nav__link">
Hyperparameter search on a SLURM HPC cluster
</a>
@@ -551,14 +551,14 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#template-model-definition" title="Template model definition" class="md-nav__link">
<a href="#template-model-definition" class="md-nav__link">
Template model definition
</a>
</li>
<li class="md-nav__item">
<a href="#trainer-example" title="Trainer Example" class="md-nav__link">
<a href="#trainer-example" class="md-nav__link">
Trainer Example
</a>
@@ -566,21 +566,21 @@
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#cpu-hyperparameter-search" title="CPU hyperparameter search" class="md-nav__link">
<a href="#cpu-hyperparameter-search" class="md-nav__link">
CPU hyperparameter search
</a>
</li>
<li class="md-nav__item">
<a href="#hyperparameter-search-on-a-single-or-multiple-gpus" title="Hyperparameter search on a single or multiple GPUs" class="md-nav__link">
<a href="#hyperparameter-search-on-a-single-or-multiple-gpus" class="md-nav__link">
Hyperparameter search on a single or multiple GPUs
</a>
</li>
<li class="md-nav__item">
<a href="#hyperparameter-search-on-a-slurm-hpc-cluster" title="Hyperparameter search on a SLURM HPC cluster" class="md-nav__link">
<a href="#hyperparameter-search-on-a-slurm-hpc-cluster" class="md-nav__link">
Hyperparameter search on a SLURM HPC cluster
</a>
@@ -938,7 +938,7 @@ argument parser you get the default arguments in the argument parser.</p>
</div>
<script src="../../assets/javascripts/application.245445c6.js"></script>
<script src="../../assets/javascripts/application.c648116f.js"></script>
<script>app.initialize({version:"1.0.4",url:{base:"../.."}})</script>
index.html
+35 -34
View File
@@ -34,7 +34,7 @@
<meta name="lang:search.tokenizer" content="[\s\-]+">
<link rel="shortcut icon" href="assets/images/favicon.png">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.0">
<meta name="generator" content="mkdocs-1.0.4, mkdocs-material-4.4.2">
@@ -42,7 +42,7 @@
<link rel="stylesheet" href="assets/stylesheets/application.0284f74d.css">
<link rel="stylesheet" href="assets/stylesheets/application.30686662.css">
@@ -171,7 +171,7 @@
<main class="md-main">
<main class="md-main" role="main">
<div class="md-main__inner md-grid" data-md-component="container">
@@ -239,105 +239,105 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#new-project-quick-start" title="New project Quick Start" class="md-nav__link">
<a href="#new-project-quick-start" class="md-nav__link">
New project Quick Start
</a>
</li>
<li class="md-nav__item">
<a href="#case-1-bert" title="Case 1: BERT" class="md-nav__link">
<a href="#case-1-bert" class="md-nav__link">
Case 1: BERT
</a>
</li>
<li class="md-nav__item">
<a href="#case-2-cooler-not-bert" title="Case 2: COOLER NOT BERT" class="md-nav__link">
<a href="#case-2-cooler-not-bert" class="md-nav__link">
Case 2: COOLER NOT BERT
</a>
</li>
<li class="md-nav__item">
<a href="#rapid-research-flow" title="Rapid research flow" class="md-nav__link">
<a href="#rapid-research-flow" class="md-nav__link">
Rapid research flow
</a>
</li>
<li class="md-nav__item">
<a href="#templates" title="Templates" class="md-nav__link">
<a href="#templates" class="md-nav__link">
Templates
</a>
</li>
<li class="md-nav__item">
<a href="#docs-shortcuts" title="Docs shortcuts" class="md-nav__link">
<a href="#docs-shortcuts" class="md-nav__link">
Docs shortcuts
</a>
</li>
<li class="md-nav__item">
<a href="#quick-start-examples" title="Quick start examples" class="md-nav__link">
<a href="#quick-start-examples" class="md-nav__link">
Quick start examples
</a>
</li>
<li class="md-nav__item">
<a href="#checkpointing" title="Checkpointing" class="md-nav__link">
<a href="#checkpointing" class="md-nav__link">
Checkpointing
</a>
</li>
<li class="md-nav__item">
<a href="#computing-cluster-slurm" title="Computing cluster (SLURM)" class="md-nav__link">
<a href="#computing-cluster-slurm" class="md-nav__link">
Computing cluster (SLURM)
</a>
</li>
<li class="md-nav__item">
<a href="#debugging" title="Debugging" class="md-nav__link">
<a href="#debugging" class="md-nav__link">
Debugging
</a>
</li>
<li class="md-nav__item">
<a href="#distributed-training" title="Distributed training" class="md-nav__link">
<a href="#distributed-training" class="md-nav__link">
Distributed training
</a>
</li>
<li class="md-nav__item">
<a href="#experiment-logging" title="Experiment Logging" class="md-nav__link">
<a href="#experiment-logging" class="md-nav__link">
Experiment Logging
</a>
</li>
<li class="md-nav__item">
<a href="#training-loop" title="Training loop" class="md-nav__link">
<a href="#training-loop" class="md-nav__link">
Training loop
</a>
</li>
<li class="md-nav__item">
<a href="#validation-loop" title="Validation loop" class="md-nav__link">
<a href="#validation-loop" class="md-nav__link">
Validation loop
</a>
</li>
<li class="md-nav__item">
<a href="#testing-loop" title="Testing loop" class="md-nav__link">
<a href="#testing-loop" class="md-nav__link">
Testing loop
</a>
@@ -613,105 +613,105 @@
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="#new-project-quick-start" title="New project Quick Start" class="md-nav__link">
<a href="#new-project-quick-start" class="md-nav__link">
New project Quick Start
</a>
</li>
<li class="md-nav__item">
<a href="#case-1-bert" title="Case 1: BERT" class="md-nav__link">
<a href="#case-1-bert" class="md-nav__link">
Case 1: BERT
</a>
</li>
<li class="md-nav__item">
<a href="#case-2-cooler-not-bert" title="Case 2: COOLER NOT BERT" class="md-nav__link">
<a href="#case-2-cooler-not-bert" class="md-nav__link">
Case 2: COOLER NOT BERT
</a>
</li>
<li class="md-nav__item">
<a href="#rapid-research-flow" title="Rapid research flow" class="md-nav__link">
<a href="#rapid-research-flow" class="md-nav__link">
Rapid research flow
</a>
</li>
<li class="md-nav__item">
<a href="#templates" title="Templates" class="md-nav__link">
<a href="#templates" class="md-nav__link">
Templates
</a>
</li>
<li class="md-nav__item">
<a href="#docs-shortcuts" title="Docs shortcuts" class="md-nav__link">
<a href="#docs-shortcuts" class="md-nav__link">
Docs shortcuts
</a>
</li>
<li class="md-nav__item">
<a href="#quick-start-examples" title="Quick start examples" class="md-nav__link">
<a href="#quick-start-examples" class="md-nav__link">
Quick start examples
</a>
</li>
<li class="md-nav__item">
<a href="#checkpointing" title="Checkpointing" class="md-nav__link">
<a href="#checkpointing" class="md-nav__link">
Checkpointing
</a>
</li>
<li class="md-nav__item">
<a href="#computing-cluster-slurm" title="Computing cluster (SLURM)" class="md-nav__link">
<a href="#computing-cluster-slurm" class="md-nav__link">
Computing cluster (SLURM)
</a>
</li>
<li class="md-nav__item">
<a href="#debugging" title="Debugging" class="md-nav__link">
<a href="#debugging" class="md-nav__link">
Debugging
</a>
</li>
<li class="md-nav__item">
<a href="#distributed-training" title="Distributed training" class="md-nav__link">
<a href="#distributed-training" class="md-nav__link">
Distributed training
</a>
</li>
<li class="md-nav__item">
<a href="#experiment-logging" title="Experiment Logging" class="md-nav__link">
<a href="#experiment-logging" class="md-nav__link">
Experiment Logging
</a>
</li>
<li class="md-nav__item">
<a href="#training-loop" title="Training loop" class="md-nav__link">
<a href="#training-loop" class="md-nav__link">
Training loop
</a>
</li>
<li class="md-nav__item">
<a href="#validation-loop" title="Validation loop" class="md-nav__link">
<a href="#validation-loop" class="md-nav__link">
Validation loop
</a>
</li>
<li class="md-nav__item">
<a href="#testing-loop" title="Testing loop" class="md-nav__link">
<a href="#testing-loop" class="md-nav__link">
Testing loop
</a>
@@ -886,6 +886,7 @@ You would define a single LightningModule and use flags to switch between your d
<ul>
<li><a href="https://williamfalcon.github.io/pytorch-lightning/Trainer/Training%20Loop/#accumulated-gradients">Accumulate gradients</a></li>
<li><a href="https://williamfalcon.github.io/pytorch-lightning/Trainer/Training%20Loop/#force-training-for-min-or-max-epochs">Force training for min or max epochs</a></li>
<li><a href="https://williamfalcon.github.io/pytorch-lightning/Trainer/Training%20Loop/#early-stopping">Early stopping</a></li>
<li><a href="https://williamfalcon.github.io/pytorch-lightning/Trainer/Training%20Loop/#force-disable-early-stop">Force disable early stop</a></li>
<li><a href="https://williamfalcon.github.io/pytorch-lightning/Trainer/Training%20Loop/#gradient-clipping">Gradient Clipping</a></li>
<li><a href="https://williamfalcon.github.io/pytorch-lightning/Trainer/hooks/">Hooks</a></li>
@@ -962,7 +963,7 @@ You would define a single LightningModule and use flags to switch between your d
</div>
<script src="assets/javascripts/application.245445c6.js"></script>
<script src="assets/javascripts/application.c648116f.js"></script>
<script>app.initialize({version:"1.0.4",url:{base:"."}})</script>
search/search_index.json
+1 -1
View File
File diff suppressed because one or more lines are too long
sitemap.xml
+15 -15
View File
@@ -2,77 +2,77 @@
<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
<url>
<loc>None</loc>
<lastmod>2019-09-17</lastmod>
<lastmod>2019-09-26</lastmod>
<changefreq>daily</changefreq>
</url>
<url>
<loc>None</loc>
<lastmod>2019-09-17</lastmod>
<lastmod>2019-09-26</lastmod>
<changefreq>daily</changefreq>
</url>
<url>
<loc>None</loc>
<lastmod>2019-09-17</lastmod>
<lastmod>2019-09-26</lastmod>
<changefreq>daily</changefreq>
</url>
<url>
<loc>None</loc>
<lastmod>2019-09-17</lastmod>
<lastmod>2019-09-26</lastmod>
<changefreq>daily</changefreq>
</url>
<url>
<loc>None</loc>
<lastmod>2019-09-17</lastmod>
<lastmod>2019-09-26</lastmod>
<changefreq>daily</changefreq>
</url>
<url>
<loc>None</loc>
<lastmod>2019-09-17</lastmod>
<lastmod>2019-09-26</lastmod>
<changefreq>daily</changefreq>
</url>
<url>
<loc>None</loc>
<lastmod>2019-09-17</lastmod>
<lastmod>2019-09-26</lastmod>
<changefreq>daily</changefreq>
</url>
<url>
<loc>None</loc>
<lastmod>2019-09-17</lastmod>
<lastmod>2019-09-26</lastmod>
<changefreq>daily</changefreq>
</url>
<url>
<loc>None</loc>
<lastmod>2019-09-17</lastmod>
<lastmod>2019-09-26</lastmod>
<changefreq>daily</changefreq>
</url>
<url>
<loc>None</loc>
<lastmod>2019-09-17</lastmod>
<lastmod>2019-09-26</lastmod>
<changefreq>daily</changefreq>
</url>
<url>
<loc>None</loc>
<lastmod>2019-09-17</lastmod>
<lastmod>2019-09-26</lastmod>
<changefreq>daily</changefreq>
</url>
<url>
<loc>None</loc>
<lastmod>2019-09-17</lastmod>
<lastmod>2019-09-26</lastmod>
<changefreq>daily</changefreq>
</url>
<url>
<loc>None</loc>
<lastmod>2019-09-17</lastmod>
<lastmod>2019-09-26</lastmod>
<changefreq>daily</changefreq>
</url>
<url>
<loc>None</loc>
<lastmod>2019-09-17</lastmod>
<lastmod>2019-09-26</lastmod>
<changefreq>daily</changefreq>
</url>
<url>
<loc>None</loc>
<lastmod>2019-09-17</lastmod>
<lastmod>2019-09-26</lastmod>
<changefreq>daily</changefreq>
</url>
</urlset>
sitemap.xml.gz
BIN
View File
Binary file not shown.