flytekit.core.array_node_map_task

Directory

Classes

Methods

Variables

Methods

array_node_map_task()

def array_node_map_task(
    task_function: flytekit.core.python_function_task.PythonFunctionTask,
    concurrency: typing.Optional[int],
    min_success_ratio: float,
    kwargs,
)

Map task that uses the ArrayNode construct..

This is an experimental drop-in replacement for ~flytekit.map_task.

map_task()

def map_task(
    target: typing.Union[flytekit.core.launch_plan.LaunchPlan, flytekit.core.python_function_task.PythonFunctionTask, ForwardRef('FlyteLaunchPlan')],
    concurrency: typing.Optional[int],
    min_successes: typing.Optional[int],
    min_success_ratio: float,
    kwargs,
)

Wrapper that creates a map task utilizing either the existing ArrayNodeMapTask or the drop in replacement ArrayNode implementation

flytekit.core.array_node_map_task.ArrayNodeMapTask

Base Class for all Tasks with a Python native Interface. This should be directly used for task types, that do not have a python function to be executed. Otherwise refer to :py:class:flytekit.PythonFunctionTask.

class ArrayNodeMapTask(
    python_function_task: typing.Union[flytekit.core.python_function_task.PythonFunctionTask, flytekit.core.python_function_task.PythonInstanceTask, functools.partial],
    concurrency: typing.Optional[int],
    min_successes: typing.Optional[int],
    min_success_ratio: typing.Optional[float],
    bound_inputs: typing.Optional[typing.Set[str]],
    bound_inputs_values: typing.Optional[typing.Dict[str, typing.Any]],
    kwargs,
)

Methods

compile()

def compile(
    ctx: flytekit.core.context_manager.FlyteContext,
    args,
    kwargs,
) -> typing.Union[typing.Tuple[flytekit.core.promise.Promise], flytekit.core.promise.Promise, flytekit.core.promise.VoidPromise, NoneType]

Generates a node that encapsulates this task in a workflow definition.

construct_node_metadata()

def construct_node_metadata()

This returns metadata for the parent ArrayNode, not the sub-node getting mapped over

dispatch_execute()

def dispatch_execute(
    ctx: flytekit.core.context_manager.FlyteContext,
    input_literal_map: flytekit.models.literals.LiteralMap,
) -> typing.Union[flytekit.models.literals.LiteralMap, flytekit.models.dynamic_job.DynamicJobSpec, typing.Coroutine]

This method translates Flyte’s Type system based input values and invokes the actual call to the executor This method is also invoked during runtime.

• VoidPromise is returned in the case when the task itself declares no outputs.
• Literal Map is returned when the task returns either one more outputs in the declaration. Individual outputs may be none
• DynamicJobSpec is returned when a dynamic workflow is executed

execute()

def execute(
    kwargs,
) -> typing.Any

This method will be invoked to execute the task.

find_lhs()

def find_lhs()

get_command()

def get_command(
    settings: flytekit.configuration.SerializationSettings,
) -> typing.List[str]

TODO ADD bound variables to the resolver. Maybe we need a different resolver?

get_config()

def get_config(
    settings: flytekit.configuration.SerializationSettings,
) -> typing.Optional[typing.Dict[str, str]]

Returns the task config as a serializable dictionary. This task config consists of metadata about the custom defined for this task.

get_container()

def get_container(
    settings: flytekit.configuration.SerializationSettings,
) -> flytekit.models.task.Container

Returns the container definition (if any) that is used to run the task on hosted Flyte.

get_custom()

def get_custom(
    settings: flytekit.configuration.SerializationSettings,
) -> typing.Dict[str, typing.Any]

Return additional plugin-specific custom data (if any) as a serializable dictionary.

get_extended_resources()

def get_extended_resources(
    settings: flytekit.configuration.SerializationSettings,
) -> typing.Optional[flyteidl.core.tasks_pb2.ExtendedResources]

Returns the extended resources to allocate to the task on hosted Flyte.

get_input_types()

def get_input_types()

Returns the names and python types as a dictionary for the inputs of this task.

get_k8s_pod()

def get_k8s_pod(
    settings: flytekit.configuration.SerializationSettings,
) -> flytekit.models.task.K8sPod

Returns the kubernetes pod definition (if any) that is used to run the task on hosted Flyte.

get_sql()

def get_sql(
    settings: flytekit.configuration.SerializationSettings,
) -> flytekit.models.task.Sql

Returns the Sql definition (if any) that is used to run the task on hosted Flyte.

get_type_for_input_var()

def get_type_for_input_var(
    k: str,
    v: typing.Any,
) -> typing.Type[typing.Any]

Returns the python type for an input variable by name.

get_type_for_output_var()

def get_type_for_output_var(
    k: str,
    v: typing.Any,
) -> type

We override this method from flytekit.core.base_task Task because the dispatch_execute method uses this interface to construct outputs. Each instance of an container_array task will however produce outputs according to the underlying run_task interface and the array plugin handler will actually create a collection from these individual outputs as the final output value.

local_execute()

def local_execute(
    ctx: flytekit.core.context_manager.FlyteContext,
    kwargs,
) -> typing.Union[typing.Tuple[flytekit.core.promise.Promise], flytekit.core.promise.Promise, flytekit.core.promise.VoidPromise, typing.Coroutine, NoneType]

This function is used only in the local execution path and is responsible for calling dispatch execute. Use this function when calling a task with native values (or Promises containing Flyte literals derived from Python native values).

local_execution_mode()

def local_execution_mode()

post_execute()

def post_execute(
    user_params: typing.Optional[flytekit.core.context_manager.ExecutionParameters],
    rval: typing.Any,
) -> typing.Any

Post execute is called after the execution has completed, with the user_params and can be used to clean-up, or alter the outputs to match the intended tasks outputs. If not overridden, then this function is a No-op

pre_execute()

def pre_execute(
    user_params: typing.Optional[flytekit.core.context_manager.ExecutionParameters],
) -> typing.Optional[flytekit.core.context_manager.ExecutionParameters]

This is the method that will be invoked directly before executing the task method and before all the inputs are converted. One particular case where this is useful is if the context is to be modified for the user process to get some user space parameters. This also ensures that things like SparkSession are already correctly setup before the type transformers are called

This should return either the same context of the mutated context

prepare_target()

def prepare_target()

Alters the underlying run_task command to modify it for map task execution and then resets it after.

sandbox_execute()

def sandbox_execute(
    ctx: flytekit.core.context_manager.FlyteContext,
    input_literal_map: flytekit.models.literals.LiteralMap,
) -> flytekit.models.literals.LiteralMap

Call dispatch_execute, in the context of a local sandbox execution. Not invoked during runtime.

set_command_prefix()

def set_command_prefix(
    cmd: typing.Optional[typing.List[str]],
)

Properties

flytekit.core.array_node_map_task.ArrayNodeMapTaskResolver

Special resolver that is used for ArrayNodeMapTasks. This exists because it is possible that ArrayNodeMapTasks are created using nested “partial” subtasks. When a maptask is created its interface is interpolated from the interface of the subtask - the interpolation, simply converts every input into a list/collection input.

For example: interface -> (i: int, j: str) -> str => map_task interface -> (i: List[int], j: List[str]) -> List[str]

But in cases in which j is bound to a fixed value by using functools.partial we need a way to ensure that the interface is not simply interpolated, but only the unbound inputs are interpolated.

def foo((i: int, j: str) -> str:
    ...

mt = map_task(functools.partial(foo, j=10))

print(mt.interface)
    ```

output:

        (i: List[int], j: str) -> List[str]

But, at runtime this information is lost. To reconstruct this, we use ArrayNodeMapTaskResolver that records the "bound vars"
and then at runtime reconstructs the interface with this knowledge


```python
class ArrayNodeMapTaskResolver(
    args,
    kwargs,
)

Methods

find_lhs()

def find_lhs()

get_all_tasks()

def get_all_tasks()

Future proof method. Just making it easy to access all tasks (Not required today as we auto register them)

load_task()

def load_task(
    loader_args: typing.List[str],
    max_concurrency: int,
) -> flytekit.core.array_node_map_task.ArrayNodeMapTask

Loader args should be of the form vars “var1,var2,..” resolver “resolver” [resolver_args]

loader_args()

def loader_args(
    settings: flytekit.configuration.SerializationSettings,
    t: flytekit.core.array_node_map_task.ArrayNodeMapTask,
) -> typing.List[str]

Return a list of strings that can help identify the parameter Task

name()

def name()

task_name()

def task_name(
    t: flytekit.core.base_task.Task,
) -> typing.Optional[str]

Overridable function that can optionally return a custom name for a given task

Properties