TorchX is a universal job launcher for PyTorch applications. TorchX is designed to have fast iteration time for training/research and support for E2E production ML pipelines when you're ready.
Please note that this issue tracker is not a help form and this issue will be closed.
Before submitting, please ensure you have gone through our
documentation.
Question
The examples on the documentation refer to using torchx via the cli implementation. I was wondering if there was any way that torchx can be used in a sdk format. For instance:
class MyCustomComponent
class MyScheduler
runner = torchx.runner()
runner.with_scheduler(MyScheduler())
runner.run_component(MyCustomComponent())
If its possible, is there any documentation or a sample project that provides an example of how this can be used, in particular using a custom scheduler and component?
❓ Questions and Help
Please note that this issue tracker is not a help form and this issue will be closed.
Before submitting, please ensure you have gone through our documentation.
Question
The examples on the documentation refer to using torchx via the cli implementation. I was wondering if there was any way that torchx can be used in a sdk format. For instance:
If its possible, is there any documentation or a sample project that provides an example of how this can be used, in particular using a custom scheduler and component?
Thank You!