Platform#
Generic platform#
- class pfl.internal.platform.generic_platform.Platform#
Collection of properties and methods related to compute platforms
- abstract consume_metrics(metrics, iteration, context=None)#
Process task status and metrics.
- abstract create_checkpoint_directories(subdir_names)#
Given a list of directories, creates the directories either in the current working directory if executed on a local computer or on a remote compute platform.
- abstract get_distributed_addresses(verbose=False)#
Get all IP addresses of compute nodes when running in a distributed context.
- abstract get_platform_context()#
Get a string-keyed dict of platform context information.
- Return type:
Dict[str,str]
- abstract get_default_callbacks()#
Get a list of callbacks to add to algorithms using this platform.
- Return type:
List[TrainingProcessCallback]
- class pfl.internal.platform.generic_platform.GenericPlatform#
- consume_metrics(metrics, iteration, context=None)#
Send metrics to platform, which may e.g. output them on the terminal or display on a dashboard.
- Parameters:
metrics (
Metrics) – Metrics to output.iteration (
int) – Current iteration.context (
Optional[Dict]) – An optional dict defining the context, e.g. hyperparameters used.
- create_checkpoint_directories(subdir_names)#
Given a list of directories, creates the directories in appropriate location on the current platform.
- Parameters:
subdir_names – A list of directory names to create.
- Return type:
List- Returns:
A list of the paths for every new directory.
- get_distributed_addresses(verbose=False)#
Get all IP addresses of compute nodes when running in a distributed context.
- Return type:
Tuple
- get_platform_context()#
Get a string-keyed dict of platform context information.
- Return type:
Dict[str,str]
- get_default_callbacks()#
Get a list of callbacks to add to algorithms using this platform.
- Return type:
List[TrainingProcessCallback]
Selector#
- pfl.internal.platform.selector.get_platform()#
Get the current platform. If the platform is not set, return the default platform
pfl.internal.platform.GenericPlatform.
- pfl.internal.platform.selector.set_platform(module)#
Manually set the platform module. Can be useful when you are running on an unsupported platform and made a custom integration for it.