If you develop plugins, please join the Plugin developer announcement chat channel.
When developing Hydra plugins, keep the following things in mind:
- Hydra plugins can be either a standalone Python package, or a part of your existing Python package.
In both cases - They should be in the namespace module
hydra_plugins(This is a top level module, Your plugin will NOT be discovered if you place it in
- Do NOT place an
hydra_plugins(doing so may break other installed Hydra plugins).
The plugin discovery process runs whenever Hydra starts. During plugin discovery, Hydra scans for plugins in all the submodules of
hydra_plugins. Hydra will import each module and look for plugins defined in that module.
Any module under
hydra_plugins that is slow to import will slow down the startup of ALL Hydra applicaitons.
Plugins with expensive imports can exclude individual files from Hydra's plugin discovery process by prefixing them with
_ (but not
For example, the file
_my_plugin_lib.py would not be imported and scanned, while
my_plugin_lib.py would be.
The best way to get started developing a Hydra plugin is to base your new plugin on one of the example plugins:
- Copy the subtree of the relevant example plugin into a standalone project.
setup.py, rename the plugin module, for example from
- Install the new plugin (Run this in the plugin directory:
pip install -e .)
- Run the included example app and make sure that the plugin is discovered:
- Run the example application to see that that your plugin is doing something.
- [Optional] If you want the plugin be embedded in your existing application/library, move the
hydra_pluginsdirectory and make sure that it's included as a namespace module in your final Python package. See the
setup.pyfile included with the example plugin for hints (typically this involves using
- Hack on your plugin, Ensure that the recommended tests and any tests you want to add are passing.