mirror of
https://github.com/hanzo-ml/trainer.git
synced 2026-09-24 00:31:34 +00:00
Alex LembiyeuskiandGitHub
3daf40fe6a
* Add Python SDK for Kubeflow Training Operator * Update example notebooks * Update SDK generation tooling and docs * Re-generate SDK * Allow to specify container name in 'get_logs' methods * Generalize job labels * Check if attribute exists * Address code review comments
715 B
715 B
V1PyTorchJobSpec
PyTorchJobSpec is a desired state description of the PyTorchJob.
Properties
| Name | Type | Description | Notes |
|---|---|---|---|
| pytorch_replica_specs | dict(str, V1ReplicaSpec) | A map of PyTorchReplicaType (type) to ReplicaSpec (value). Specifies the PyTorch cluster configuration. For example, { "Master": PyTorchReplicaSpec, "Worker": PyTorchReplicaSpec, } | |
| run_policy | V1RunPolicy |