Documentation Addition: Training YAML Reference #835
Conversation
This file contains hidden or 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
There was a problem hiding this comment.
I think this should replace custom_dataset_conf and advanced_base_config, right?
| ``metatrain`` uses a YAML file to specify the parameters for model training, | ||
| accessed via ``mtt train options.yaml``. In this section, we provide a complete reference | ||
| for the parameters provided by the training YAML input. | ||
|
|
There was a problem hiding this comment.
We could add something like
A `sample yaml file <XXX>` is available. This should be appropriate to start a first training. Edit it to suit your specific needs and desires.
There was a problem hiding this comment.
My overall idea of this file would be similar to the main option file of GROMACS:
https://manual.gromacs.org/current/user-guide/mdp-options.html
You can maybe scroll through for some inspiration. They have for example the parameters highlighted in red. We could do the same. As they are also using sphinx it should be possible :-)
|
|
||
| Computational Parameters | ||
| ====================================== | ||
| The computational parameters define the computational device, precision and seed. These parameters are optional. |
There was a problem hiding this comment.
Suggested change
| The computational parameters define the computational device, precision and seed. These parameters are optional. | |
| The computational parameters define the computational ``device``, ``precision`` and ``seed``. These parameters are optional. |
| precision: 32 | ||
| seed: 0 | ||
|
|
||
| :param device [optional]: The computational device used for model training. The script automatically |
There was a problem hiding this comment.
Suggested change
| :param device [optional]: The computational device used for model training. The script automatically | |
| :param device [optional]: The computational device used for model training. The metatrain automatically |
| ``float16`` respectively. The datatypes that can be supported also depends on the model architecture used. | ||
| :param seed [optional]: The seed used for non-deterministic operations and is used to set the seed for ``numpy.random``, | ||
| ``random``, ``torch`` and ``torch.cuda``. The input must be a non-negative integer. This parameter is important for ensuring | ||
| reproducibility. If not specified, the seed is generated randomly. |
There was a problem hiding this comment.
Suggested change
| reproducibility. If not specified, the seed is generated randomly. | |
| reproducibility. If not specified, the seed is generated randomly and reported in the log. |
Comment on lines
+42
to
+44
| The next set of parameters are also optional and deals with integration with Weights and Biases (wandb) logging. Leaving this | ||
| section blank will simply disable wandb integration. The parameters for this section is the same as that in | ||
| `wandb.init <https://docs.wandb.ai/ref/python/init/>`_. Here we provide a minimal example for the YAML input |
There was a problem hiding this comment.
I thunk this can be shortened a bit maybe along the lines of:
Suggested change
| The next set of parameters are also optional and deals with integration with Weights and Biases (wandb) logging. Leaving this | |
| section blank will simply disable wandb integration. The parameters for this section is the same as that in | |
| `wandb.init <https://docs.wandb.ai/ref/python/init/>`_. Here we provide a minimal example for the YAML input | |
| Optional section dealing with integration with `Weights and Biases (wandb) <link>`_ logging. Leaving this | |
| section blank will simply disable wandb integration. The parameters for this section is the same as that in | |
| `wandb.init <https://docs.wandb.ai/ref/python/init/>`_. Here we provide a minimal example for the YAML input |
| - tag2 | ||
| notes: This is a test run | ||
|
|
||
| All parameters of your options file will be automatically added to the wandb run so |
There was a problem hiding this comment.
Suggested change
| All parameters of your options file will be automatically added to the wandb run so | |
| All parameters of your ``options.yaml`` file will be automatically added to the wandb run so |
Comment on lines
+73
to
+74
| Loss | ||
| =================== |
There was a problem hiding this comment.
Just to please my eyes
Suggested change
| Loss | |
| =================== | |
| Loss | |
| ==== |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Added a reference page for the training YAML, I have yet to remove the redundant pages
📚 Documentation preview 📚: https://metatrain--835.org.readthedocs.build/en/835/