Skip to main content
Version: Next

Getting started

Introduction​

Hydra is an open-source Python framework that simplifies the development of research and other complex applications. The key feature is the ability to dynamically create a hierarchical configuration by composition and override it through config files and the command line. The name Hydra comes from its ability to run multiple similar jobs - much like a Hydra with multiple heads.

Key features:​

  • Hierarchical configuration composable from multiple sources
  • Configuration can be specified or overridden from the command line
  • Dynamic command line tab completion
  • Run your application locally or launch it to run remotely
  • Run multiple jobs with different arguments with a single command

Versions​

Hydra supports Linux, Mac and Windows.
Use the version switcher in the top bar to switch between documentation versions.

VersionRelease notesPython Versions
β–Ί1.3 (Stable)Release notes3.6 - 3.11
1.2Release notes3.6 - 3.10
1.1Release notes3.6 - 3.9
1.0Release notes3.6 - 3.8
0.11Release notes2.7, 3.5 - 3.8

Quick start guide​

This guide will show you some of the most important features you get by writing your application as a Hydra app. If you only want to use Hydra for config composition, check out Hydra's compose API for an alternative. Please also read the full tutorial to gain a deeper understanding.

Installation​

pip install hydra-core --upgrade

Basic example​

Config:

conf/config.yaml
db:
driver: mysql
user: omry
pass: secret

Application:

my_app.py
import hydra
from omegaconf import DictConfig, OmegaConf

@hydra.main(version_base=None, config_path="conf", config_name="config")
def my_app(cfg : DictConfig) -> None:
print(OmegaConf.to_yaml(cfg))

if __name__ == "__main__":
my_app()

You can learn more about OmegaConf here later.

config.yaml is loaded automatically when you run your application

$ python my_app.py
db:
driver: mysql
pass: secret
user: omry

You can override values in the loaded config from the command line:

$ python my_app.py db.user=root db.pass=1234
db:
driver: mysql
user: root
pass: 1234

Composition example​

You may want to alternate between two different databases. To support this create a config group named db, and place one config file for each alternative inside: The directory structure of our application now looks like:

β”œβ”€β”€ conf
β”‚Β Β  β”œβ”€β”€ config.yaml
β”‚Β Β  β”œβ”€β”€ db
β”‚Β Β  β”‚Β Β  β”œβ”€β”€ mysql.yaml
β”‚Β Β  β”‚Β Β  └── postgresql.yaml
β”‚Β Β  └── __init__.py
└── my_app.py

Here is the new config:

conf/config.yaml
defaults:
- db: mysql

defaults is a special directive telling Hydra to use db/mysql.yaml when composing the configuration object. The resulting cfg object is a composition of configs from defaults with configs specified in your config.yaml.

You can now choose which database configuration to use and override values from the command line:

$ python my_app.py db=postgresql db.timeout=20
db:
driver: postgresql
pass: drowssap
timeout: 20
user: postgres_user

You can have as many config groups as you need.

Multirun​

You can run your function multiple times with different configurations easily with the --multirun|-m flag.

$ python my_app.py --multirun db=mysql,postgresql
[HYDRA] Sweep output dir : multirun/2020-01-09/01-16-29
[HYDRA] Launching 2 jobs locally
[HYDRA] #0 : db=mysql
db:
driver: mysql
pass: secret
user: omry

[HYDRA] #1 : db=postgresql
db:
driver: postgresql
pass: drowssap
timeout: 10
user: postgres_user

There is a whole lot more to Hydra. Read the tutorial to learn more.

Other stuff​

Community​

Ask questions on github or StackOverflow (Use the tag #fb-hydra):

Follow Hydra on Twitter and Facebook:

Citing Hydra​

If you use Hydra in your research please use the following BibTeX entry:

@Misc{Yadan2019Hydra,
author = {Omry Yadan},
title = {Hydra - A framework for elegantly configuring complex applications},
howpublished = {Github},
year = {2019},
url = {https://github.com/facebookresearch/hydra}
}