This repo hosts an example Hydra app.
Create a model in a Notebook or another script by calling create_model
.
>>> from hydra_app import create_model
>>>
>>> net = create_model()
>>> print(net)
NeuralNetwork(num_hidden=[200, 200], num_latent=20, dropout=0.2, lr=0.0001)
Override base configuration by supplying an additional config.
>>> from omegaconf import OmegaConf
>>>
>>> experiment_conf = OmegaConf.load("experiment.yaml")
>>>
>>> net = create_model(experiment_conf)
>>> print(net)
NeuralNetwork(num_hidden=[200, 200], num_latent=16, dropout=0.0, lr=0.1)
Create an my_conf/experiment
folder and place inside a config file (in the
example below it's called test.yaml
). Then, from the command line, call the
app, specify the name of the config file without extension as experiment=test
and the path to the folder using the -cd
argument.
>>> python hydra_app experiment=test -cd myconf
NeuralNetwork(num_hidden=[200, 200], num_latent=16, dropout=0.0, lr=0.1)
You can check the default options using --help
:
python hydra_app --help
or load the "other" model which is specified in conf/model/other.yaml
python hydra_app model=other --help