Pipeline

Pipeline defines a set of steps that can be executed together.

Attributes

Name	Type	Description
pipeline_id	uuid.UUID	technical id of the pipeline.
experiment_id	uuid.UUID	technical id of the experiment linked to the pipeline.
key	str	logical string unique id of the pipeline.
steps	list	list of steps connected between by their inputs and outputs.
template_id	uuid.UUID	template id associated to the pipeline.
variables	dict	dictionary of variable name with they wizata_dsapi.VarType.
createdById	uuid.UUID	unique identifier of creating user.
createdDate	int	timestamp of created date.
updatedById	uuid.UUID	unique identifier of updating user.
updatedDate	int	timestamp of updated date.

add a model step

Name	Type	Default	Description
config	MLModelConfig		model configuration to define how pipeline should train and use your model.
input_df			str, dict or Pipeline I/O defining input dataframe properties
output_df		None	str, dict or Pipeline I/O defining input dataframe properties

add a writer step

Name	Type	Default	Description
script			script configuration to define how pipeline should execute the plot script.
df_name	str	None	str deprecated usage
input_df	PipelineIO	None	str, dict or Pipeline I/O defining input dataframe properties

add a query step

Name	Type	Default	Description
request	Request		request definition to add.
df_name	str	query_df	output name ot use for the dataframe - use df_output for more features.
use_template	bool	True	by default, if pipeline is link to a template, the query will be too. set to false to disable forcing it.
output_df	PipelineIO	None	output df - can set a mapping.

add a transformation script

Name	Type	Default	Description
script			name, Script or ScriptConfig.
inputs	list	None	list of Pipeline I/O or dict or str for dataframe input names.
outputs	list	None	list of Pipeline I/O or dict or str for dataframe output names.
input_df_names	list	None	deprecated support.
output_df_names	list	None	deprecated support.

add a writer step

Name	Type	Default	Description
config	WriteConfig		writer configuration to define how pipeline should write data into platform.
input_df			str, dict or Pipeline I/O defining input dataframe properties

Id of the pipeline

return: string formatted UUID of the Pipeline.

validate that steps create a valid path.

return true if path is valid, otherwise raise errors

verify that variables dict is a valid { "name" : "VarType" } dictionary.

Name of the endpoints used to manipulate pipeline.

return: Endpoint name.

load from JSON dictionary representation

Name	Type	Default	Description
obj

specify the id_value neutrally

Name	Type	Default	Description
id_value

return:

Convert to a json version of Execution definition.

By default, use DS API format.

Name	Type	Default	Description
target	str	None