flip.nvflare.executors
FLIP Executors module containing NVFLARE executor wrappers.
These executors wrap user-provided training, validation, and evaluation logic.
- Exports:
RUN_TRAINER: Wrapper executor for user’s FLIP_TRAINER class
RUN_VALIDATOR: Wrapper executor for user’s FLIP_VALIDATOR class
RUN_EVALUATOR: Wrapper executor for user’s FLIP_EVALUATOR class
Submodules
Classes
Executes the uploaded evaluator and handles any errors. |
|
Wrapper executor that runs user-provided FLIP_TRAINER implementations. |
|
Wrapper executor that runs user-provided FLIP_VALIDATOR implementations. |
Package Contents
- class flip.nvflare.executors.RUN_EVALUATOR(evaluate_task_name=PTConstants.EvalTaskName, project_id='', query='')
Bases:
nvflare.apis.executor.ExecutorExecutes the uploaded evaluator and handles any errors.
- _evaluate_task_name = 'evaluation'
- _project_id = ''
- _query = ''
- _evaluator = None
- execute(task_name: str, shareable: nvflare.apis.shareable.Shareable, fl_ctx: nvflare.apis.fl_context.FLContext, abort_signal: nvflare.apis.signal.Signal) nvflare.apis.shareable.Shareable
- class flip.nvflare.executors.RUN_TRAINER(train_task_name=AppConstants.TASK_TRAIN, submit_model_task_name=AppConstants.TASK_SUBMIT_MODEL, exclude_vars=None, project_id='', query='')
Bases:
nvflare.apis.executor.ExecutorWrapper executor that runs user-provided FLIP_TRAINER implementations.
This executor handles: - Dynamic importing of the user’s FLIP_TRAINER class - Error handling and exception logging - Lazy initialization of the trainer instance
- _train_task_name
- _submit_model_task_name
- _exclude_vars = None
- _project_id = ''
- _query = ''
- _flip_trainer = None
- _epochs = None
- execute(task_name: str, shareable: nvflare.apis.shareable.Shareable, fl_ctx: nvflare.apis.fl_context.FLContext, abort_signal: nvflare.apis.signal.Signal) nvflare.apis.shareable.Shareable
Execute the training task.
This method: 1. Lazily imports and initializes the user’s FLIP_TRAINER 2. Delegates execution to the user’s trainer 3. Catches and reports any exceptions
- Parameters:
task_name – The name of the task to execute
shareable – The input shareable data
fl_ctx – The FL context
abort_signal – Signal for aborting the task
- Returns:
The result of the training task
- Return type:
Shareable
- class flip.nvflare.executors.RUN_VALIDATOR(validate_task_name=AppConstants.TASK_VALIDATION, project_id='', query='')
Bases:
nvflare.apis.executor.ExecutorWrapper executor that runs user-provided FLIP_VALIDATOR implementations.
This executor handles: - Dynamic importing of the user’s FLIP_VALIDATOR class - Error handling and exception logging - Lazy initialization of the validator instance
- _validate_task_name
- _project_id = ''
- _query = ''
- _validator = None
- execute(task_name: str, shareable: nvflare.apis.shareable.Shareable, fl_ctx: nvflare.apis.fl_context.FLContext, abort_signal: nvflare.apis.signal.Signal) nvflare.apis.shareable.Shareable
Execute the validation task.
This method: 1. Lazily imports and initializes the user’s FLIP_VALIDATOR 2. Delegates execution to the user’s validator 3. Catches and reports any exceptions
- Parameters:
task_name – The name of the task to execute
shareable – The input shareable data
fl_ctx – The FL context
abort_signal – Signal for aborting the task
- Returns:
The result of the validation task
- Return type:
Shareable