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 b4c14a4 with MkDocs version: 1.0.4

Browse Source
This commit is contained in:
William Falcon
2019-08-07 08:03:37 -05:00
parent 343e3df1c0
commit 4e9d3fc53b
9 changed files with 47 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files
Trainer/Checkpointing/index.html
+29
View File
@@ -350,6 +350,13 @@
</li>
<li class="md-nav__item">
<a href="#restoring-training-session" title="Restoring training session" class="md-nav__link">
Restoring training session
</a>
</li>
@@ -511,6 +518,13 @@
</li>
<li class="md-nav__item">
<a href="#restoring-training-session" title="Restoring training session" class="md-nav__link">
Restoring training session
</a>
</li>
@@ -548,6 +562,21 @@ checkpoint_callback = ModelCheckpoint(
trainer = Trainer(checkpoint_callback=checkpoint_callback)
</code></pre>
<hr />
<h3 id="restoring-training-session">Restoring training session</h3>
<p>You might want to not only load a model but also continue training it. Use this method to
restore the trainer state as well. This will continue from the epoch and global step you last left off.<br />
However, the dataloaders will start from the first batch again (if you shuffled it shouldn't matter). </p>
<p>Lightning will restore the session if you pass an experiment with the same version and there's a saved checkpoint. </p>
<pre><code class="python">from test_tube import Experiment
exp = Experiment(version=a_previous_version_with_a_saved_checkpoint)
Trainer(experiment=exp)
trainer = Trainer(checkpoint_callback=checkpoint_callback)
# the trainer is now restored
</code></pre>
Trainer/Distributed training/index.html
+1 -1
View File
@@ -698,7 +698,7 @@ cluster.add_command('export NCCL_SOCKET_IFNAME=^docker0,lo')
cluster.add_command('export NCCL_DEBUG=INFO')
# setting a master port here is a good idea.
cluster.add_command(f'export MASTER_PORT={PORT}')
cluster.add_command('export MASTER_PORT=%r' % PORT)
# good to load the latest NCCL version
cluster.load_modules(['NCCL/2.4.7-1-cuda.10.0'])
Trainer/index.html
+1
View File
@@ -508,6 +508,7 @@ trainer.fit(model)
<ul>
<li><a href="https://williamfalcon.github.io/pytorch-lightning/Trainer/Checkpointing/#model-saving">Model saving</a></li>
<li><a href="https://williamfalcon.github.io/pytorch-lightning/LightningModule/methods/#load-from-metrics">Model loading</a> </li>
<li><a href="https://williamfalcon.github.io/pytorch-lightning/Trainer/Checkpointing/#restoring-training-session">Restoring training session</a></li>
</ul>
<p><strong>Computing cluster (SLURM)</strong> </p>
<ul>
index.html
+1
View File
@@ -687,6 +687,7 @@ one could be a seq-2-seq model, both (optionally) ran by the same trainer file.<
<ul>
<li><a href="https://williamfalcon.github.io/pytorch-lightning/Trainer/Checkpointing/#model-saving">Model saving</a></li>
<li><a href="https://williamfalcon.github.io/pytorch-lightning/LightningModule/methods/#load-from-metrics">Model loading</a> </li>
<li><a href="https://williamfalcon.github.io/pytorch-lightning/Trainer/Checkpointing/#restoring-training-session">Restoring training session</a></li>
</ul>
<h6 id="computing-cluster-slurm">Computing cluster (SLURM)</h6>
<ul>
search/search_index.json
+1 -1
View File
File diff suppressed because one or more lines are too long
sitemap.xml
+14 -14
View File
@@ -2,72 +2,72 @@
<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
<url>
<loc>None</loc>
<lastmod>2019-08-05</lastmod>
<lastmod>2019-08-07</lastmod>
<changefreq>daily</changefreq>
</url>
<url>
<loc>None</loc>
<lastmod>2019-08-05</lastmod>
<lastmod>2019-08-07</lastmod>
<changefreq>daily</changefreq>
</url>
<url>
<loc>None</loc>
<lastmod>2019-08-05</lastmod>
<lastmod>2019-08-07</lastmod>
<changefreq>daily</changefreq>
</url>
<url>
<loc>None</loc>
<lastmod>2019-08-05</lastmod>
<lastmod>2019-08-07</lastmod>
<changefreq>daily</changefreq>
</url>
<url>
<loc>None</loc>
<lastmod>2019-08-05</lastmod>
<lastmod>2019-08-07</lastmod>
<changefreq>daily</changefreq>
</url>
<url>
<loc>None</loc>
<lastmod>2019-08-05</lastmod>
<lastmod>2019-08-07</lastmod>
<changefreq>daily</changefreq>
</url>
<url>
<loc>None</loc>
<lastmod>2019-08-05</lastmod>
<lastmod>2019-08-07</lastmod>
<changefreq>daily</changefreq>
</url>
<url>
<loc>None</loc>
<lastmod>2019-08-05</lastmod>
<lastmod>2019-08-07</lastmod>
<changefreq>daily</changefreq>
</url>
<url>
<loc>None</loc>
<lastmod>2019-08-05</lastmod>
<lastmod>2019-08-07</lastmod>
<changefreq>daily</changefreq>
</url>
<url>
<loc>None</loc>
<lastmod>2019-08-05</lastmod>
<lastmod>2019-08-07</lastmod>
<changefreq>daily</changefreq>
</url>
<url>
<loc>None</loc>
<lastmod>2019-08-05</lastmod>
<lastmod>2019-08-07</lastmod>
<changefreq>daily</changefreq>
</url>
<url>
<loc>None</loc>
<lastmod>2019-08-05</lastmod>
<lastmod>2019-08-07</lastmod>
<changefreq>daily</changefreq>
</url>
<url>
<loc>None</loc>
<lastmod>2019-08-05</lastmod>
<lastmod>2019-08-07</lastmod>
<changefreq>daily</changefreq>
</url>
<url>
<loc>None</loc>
<lastmod>2019-08-05</lastmod>
<lastmod>2019-08-07</lastmod>
<changefreq>daily</changefreq>
</url>
</urlset>
sitemap.xml.gz
BIN
View File
Binary file not shown.
source/_static/lightning_logo_medium.png
BIN
View File
Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 8.3 KiB

source/_static/lightning_logo_small.png
BIN
View File
Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 2.6 KiB