tfx.types.Channel

Tfx Channel.

View aliases

Main aliases

tfx.types.channel.Channel

tfx.types.Channel(
    type: Type[tfx.types.Artifact],
    artifacts: Optional[Iterable[Artifact]] = None,
    matching_channel_name: Optional[Text] = None,
    producer_component_id: Optional[Text] = None,
    output_key: Optional[Text] = None
)

Used in the notebooks

Used in the tutorials
TFX Estimator Component Tutorial TFX Keras Component Tutorial Graph-based Neural Structured Learning in TFX

TFX Channel is an abstract concept that connects data producers and data consumers. It contains restriction of the artifact type that should be fed into or read from it.

Args
type	Subclass of Artifact that represents the type of this Channel.
artifacts	(Optional) A collection of artifacts as the values that can be read from the Channel. This is used to construct a static Channel.
matching_channel_name	This targets to the key of an input Channel dict in a Component. The artifacts count of this channel will be decided at runtime in Driver, based on the artifacts count of the target channel. Only one of artifacts and matching_channel_name should be set.
producer_component_id	(Optional) Producer component id of the Channel.
output_key	(Optional) The output key when producer component produces the artifacts in this Channel.

Attributes
type	The artifact type class that the Channel takes.
type_name

Methods

from_json_dict

View source

@classmethod
from_json_dict(
    dict_data: Dict[Text, Any]
) -> Any

Convert from dictionary data to an object.

get

View source

get() -> Iterable[tfx.types.Artifact]

Returns all artifacts that can be get from this Channel.

Returns
An artifact collection.

to_json_dict

View source

to_json_dict() -> Dict[Text, Any]

Convert from an object to a JSON serializable dictionary.