-
Notifications
You must be signed in to change notification settings - Fork 13
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
Showing
1 changed file
with
31 additions
and
23 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -1,41 +1,49 @@ | ||
train | ||
====== | ||
####### | ||
train | ||
####### | ||
|
||
Use this command to create a train a model: | ||
|
||
.. code-block:: bash | ||
.. code:: bash | ||
% anemoi-training train config.yaml | ||
% anemoi-training train config.yaml | ||
The command will read the default configuration and override it with the values in the provided configuration file. | ||
The configuration file should be a YAML file with the structure defined in the `Configuration` section. The file `config.yaml` will typically destribes | ||
the model to be trained, the dataset to be used, and the training hyperparameters: | ||
The command will read the default configuration and override it with the | ||
values in the provided configuration file. The configuration file should | ||
be a YAML file with the structure defined in the `Configuration` | ||
section. The file `config.yaml` will typically destribes the model to be | ||
trained, the dataset to be used, and the training hyperparameters: | ||
|
||
.. literalinclude:: train.yaml | ||
:language: yaml | ||
:language: yaml | ||
|
||
You can provide more that one configuration file, in which case the values will be merged in the order they are provided. A typical usage would be | ||
to split the training configurations into model description, training hyperparameters and runtime options | ||
You can provide more that one configuration file, in which case the | ||
values will be merged in the order they are provided. A typical usage | ||
would be to split the training configurations into model description, | ||
training hyperparameters and runtime options | ||
|
||
.. code-block:: bash | ||
.. code:: bash | ||
% anemoi-training train model.yaml hyperparameters.yaml slurm.yaml | ||
% anemoi-training train model.yaml hyperparameters.yaml slurm.yaml | ||
Furthermore, you can also provide values directly on the command line, which will override any values in the configuration files: | ||
Furthermore, you can also provide values directly on the command line, | ||
which will override any values in the configuration files: | ||
|
||
.. code-block:: bash | ||
.. code:: bash | ||
% anemoi-training train config.yaml tracker.mlflow.tracking_uri=http://localhost:5000 | ||
% anemoi-training train config.yaml tracker.mlflow.tracking_uri=http://localhost:5000 | ||
If the file `~/.config/anemoi/train.yaml` exists, it will be loaded after the defaults and before any other configuration file. | ||
This allows you to provide values such as passwords or other sensitive information that you do not want to store a git repository. | ||
If the file `~/.config/anemoi/train.yaml` exists, it will be loaded | ||
after the defaults and before any other configuration file. This allows | ||
you to provide values such as passwords or other sensitive information | ||
that you do not want to store a git repository. | ||
|
||
********************* | ||
******************** | ||
Command line usage | ||
********************* | ||
******************** | ||
|
||
.. argparse:: | ||
:module: anemoi.training.__main__ | ||
:func: create_parser | ||
:prog: anemoi-training | ||
:path: train | ||
:module: anemoi.training.__main__ | ||
:func: create_parser | ||
:prog: anemoi-training | ||
:path: train |