User guide for PyTorch Training

[andreyvelich]

We should create user guide for PyTorch training.

Initially, it could be a simple guide explaining how to get available runtimes, how to configure train() API, and use the Kubeflow Trainer client APIs to fetch TrainJob results.

Feel free to propose more guides that we should create for Kubeflow Trainer V2 @varodrig @hbelmiro @franciscojavierarceo!

/good-first-issue
/area docs

[google-oss-prow]

@andreyvelich:
This request has been marked as suitable for new contributors.

Please ensure the request meets the requirements listed here.

If this request no longer meets these requirements, the label can be removed
by commenting with the /remove-good-first-issue command.

In response to this:

We should create user guide for PyTorch training.

Initially, it could be a simple guide explaining how to get available runtimes, how to configure train() API, and use the Kubeflow Trainer client APIs to fetch TrainJob results.

Feel free to propose more guides that we should create for Kubeflow Trainer V2 @varodrig @hbelmiro @franciscojavierarceo!

/good-first-issue
/area docs

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[izuku-sds]

Hello @andreyvelich, I would like work on this issue. However, I'm not fully familiar with these concepts, could you provide resources which I can refer for this.

[andreyvelich]

Thank you for your interest, sure @izuku-sds!
Please start to explore our TrainerClient APIs, to understand what APIs users can use while developing PyTorch models: https://github.com/kubeflow/trainer/blob/master/sdk/kubeflow/trainer/api/trainer_client.py
/assign @izuku-sds