<?xml version="1.0"?>
<feed xmlns="http://www.w3.org/2005/Atom" xml:lang="fr">
	<id>http://wiki.backprop.fr/index.php?action=history&amp;feed=atom&amp;title=Dreambooth</id>
	<title>Dreambooth - Historique des versions</title>
	<link rel="self" type="application/atom+xml" href="http://wiki.backprop.fr/index.php?action=history&amp;feed=atom&amp;title=Dreambooth"/>
	<link rel="alternate" type="text/html" href="http://wiki.backprop.fr/index.php?title=Dreambooth&amp;action=history"/>
	<updated>2026-05-09T14:21:19Z</updated>
	<subtitle>Historique des versions pour cette page sur le wiki</subtitle>
	<generator>MediaWiki 1.38.4</generator>
	<entry>
		<id>http://wiki.backprop.fr/index.php?title=Dreambooth&amp;diff=23&amp;oldid=prev</id>
		<title>Jboscher : Page créée avec « Dreambooth is a kind of fine-tuning that attempts to introduce new subjects by providing just a few images of the new subject. The goal is similar to that of Textual Inversion, but the process is different. Instead of creating a new token as Textual Inversion does, we select an existing token in the vocabulary (usually a rarely used one), and fine-tune the model for a few hundred steps to bring that token close to the images we provide. This is a regular fine-tun... »</title>
		<link rel="alternate" type="text/html" href="http://wiki.backprop.fr/index.php?title=Dreambooth&amp;diff=23&amp;oldid=prev"/>
		<updated>2023-01-05T13:54:38Z</updated>

		<summary type="html">&lt;p&gt;Page créée avec « Dreambooth is a kind of fine-tuning that attempts to introduce new subjects by providing just a few images of the new subject. The goal is similar to that of Textual Inversion, but the process is different. Instead of creating a new token as Textual Inversion does, we select an existing token in the vocabulary (usually a rarely used one), and fine-tune the model for a few hundred steps to bring that token close to the images we provide. This is a regular fine-tun... »&lt;/p&gt;
&lt;p&gt;&lt;b&gt;Nouvelle page&lt;/b&gt;&lt;/p&gt;&lt;div&gt;Dreambooth is a kind of fine-tuning that attempts to introduce new subjects by providing just a few images of the new subject. The goal is similar to that of Textual Inversion, but the process is different. Instead of creating a new token as Textual Inversion does, we select an existing token in the vocabulary (usually a rarely used one), and fine-tune the model for a few hundred steps to bring that token close to the images we provide. This is a regular fine-tuning process in which all modules are unfrozen.&lt;/div&gt;</summary>
		<author><name>Jboscher</name></author>
	</entry>
</feed>