flytekit.extras.pytorch.PyTorchCheckpointTransformer#
- class flytekit.extras.pytorch.PyTorchCheckpointTransformer[source]#
TypeTransformer that supports serializing and deserializing checkpoint.
Methods
- from_binary_idl(binary_idl_object, expected_python_type)[source]#
This function primarily handles deserialization for untyped dicts, dataclasses, Pydantic BaseModels, and attribute access.`
For untyped dict, dataclass, and pydantic basemodel: Life Cycle (Untyped Dict as example):
- python val -> msgpack bytes -> binary literal scalar -> msgpack bytes -> python val
(to_literal) (from_binary_idl)
For attribute access: Life Cycle:
- python val -> msgpack bytes -> binary literal scalar -> resolved golang value -> binary literal scalar -> msgpack bytes -> python val
(to_literal) (propeller attribute access) (from_binary_idl)
- Parameters:
binary_idl_object (Binary)
expected_python_type (Type[T])
- Return type:
T | None
- from_generic_idl(generic, expected_python_type)[source]#
TODO: Support all Flyte Types. This is for dataclass attribute access from input created from the Flyte Console.
Note: - This can be removed in the future when the Flyte Console support generate Binary IDL Scalar as input.
- Parameters:
generic (Struct)
expected_python_type (Type[T])
- Return type:
T | None
- get_literal_type(t)[source]#
Converts the python type to a Flyte LiteralType
- Parameters:
t (Type[PyTorchCheckpoint])
- Return type:
- guess_python_type(literal_type)[source]#
Converts the Flyte LiteralType to a python object type.
- Parameters:
literal_type (LiteralType)
- Return type:
- to_html(ctx, python_val, expected_python_type)[source]#
Converts any python val (dataframe, int, float) to a html string, and it will be wrapped in the HTML div
- Parameters:
ctx (FlyteContext)
python_val (T)
expected_python_type (Type[T])
- Return type:
- to_literal(ctx, python_val, python_type, expected)[source]#
Converts a given python_val to a Flyte Literal, assuming the given python_val matches the declared python_type. Implementers should refrain from using type(python_val) instead rely on the passed in python_type. If these do not match (or are not allowed) the Transformer implementer should raise an AssertionError, clearly stating what was the mismatch :param ctx: A FlyteContext, useful in accessing the filesystem and other attributes :param python_val: The actual value to be transformed :param python_type: The assumed type of the value (this matches the declared type on the function) :param expected: Expected Literal Type
- Parameters:
ctx (FlyteContext)
python_val (PyTorchCheckpoint)
python_type (Type[PyTorchCheckpoint])
expected (LiteralType)
- Return type:
- to_python_value(ctx, lv, expected_python_type)[source]#
Converts the given Literal to a Python Type. If the conversion cannot be done an AssertionError should be raised :param ctx: FlyteContext :param lv: The received literal Value :param expected_python_type: Expected native python type that should be returned
- Parameters:
ctx (FlyteContext)
lv (Literal)
expected_python_type (Type[PyTorchCheckpoint])
- Return type:
Attributes
- PYTORCH_CHECKPOINT_FORMAT = 'PyTorchCheckpoint'
- is_async
- name
- python_type
This returns the python type
- type_assertions_enabled
Indicates if the transformer wants type assertions to be enabled at the core type engine layer