loadPipeline: Loads the 'AnalysisPipeline' or 'StreamingAnalysisPipeline'...

In analysisPipelines: Compose Interoperable Analysis Pipelines & Put Them in Production

Description

Loads the AnalysisPipeline or StreamingAnalysisPipeline object from the file system

Usage

loadPipeline(path, input = data.frame(), filePath = "")

Arguments

path	the path at which the .Rds file containing the pipeline is located
input	(optional) data frame with which the pipeline object should be initialized
filePath	(optional) path where a dataset in .CSV format is present which is to be loaded

Details

The AnalysisPipeline or StreamingAnalysisPipeline object is loaded into the file system from the file system based on the path specified.

Optionally, the input parameter can be provided to initialize the AnalysisPipeline or StreamingAnalysisPipeline object with an R data frame or Streaming Spark DataFrame (in case of StreamingAnalysisPipeline object) present in the R session.

Another provided option, is to specify a filePath where the input dataset is present (in a .CSV format) and the object will be initialized with this data frame. The filePath parameter takes precedence over input parameter. This is applicable only from AnalysisPipeline objects

Note - When a pipeline is loaded, the existing registry is overwritten with the registry saved with the pipeline

Value

An AnalysisPipeline or StreamingAnalysisPipeline object, optinally initialized with the data frame provided

See Also

Other Package core functions: BaseAnalysisPipeline-class, MetaAnalysisPipeline-class, assessEngineSetUp, checkSchemaMatch, createPipelineInstance, exportAsMetaPipeline, generateOutput, genericPipelineException, getInput, getLoggerDetails, getOutputById, getPipelinePrototype, getPipeline, getRegistry, initDfBasedOnType, initialize,BaseAnalysisPipeline-method, loadMetaPipeline, loadPredefinedFunctionRegistry, loadRegistry, prepExecution, registerFunction, savePipeline, saveRegistry, setInput, setLoggerDetails, updateObject, visualizePipeline

Examples

## Not run: 
  library(analysisPipelines)
  loadPipeline(path = "./pipeline.RDS")

## End(Not run)

analysisPipelines documentation built on July 1, 2020, 7:09 p.m.