External Pipeline Execution

External Pipeline Execution


1.1 Overview

  1. This document outlines the steps required to execute the Pipeline using the Rubiscape API.  

  2. The process involves obtaining a login token, preparing the necessary parameters, defining the API endpoint and request body, making the API call and handling the response.

 

​

1.2 Steps Involved in API Connection


API Endpoint

https://hostname/users/loginUser/

HTTP Request Method:

POST

Parameters

"username": "U2FsdGVkX187ruk+Uu3O89+ava/mT9ZAciikLGEKZyAyRV88fK0tR8eDgTYPTNRFwC",

"password": "U2FsdGVkX19gw/Hhy3zUM+RqYjXfagLtufrGbg2XCmQqIA="

(Note: These are dummy values. Reach out to Rubiscape Support Team to get these for your credentials)

 

Notes

·         Username and Password must be in encrypted format.

·         For Encrypted Username and Password, contact Rubiscape Team.

·         Make POST API call to above login API.

·         Use the obtained Sessionkey in the subsequent API calls.


Parameters

Example

Remark

backurl_R:

“http://localhost:8000/”

Adjust according to host.

PipelineName:

‘Pipeline_Name’

Adjust according to required Pipeline.

withData:

False

Set this flag to True if data needs to be sent.

inputData:

{“feature1”:[values],”feature2”:[values]}

If withData flag is True, send data in dictionary format

projectName:

‘project_Name’

Adjust according to project name.

workspaceName

‘workspace_Name

Adjust according to workspace name.

SessionKey

‘y05cqb9wibt7xcaii2vbh38xvqhb62wf’

Adjust the session key according to the authenticated session key obtained from the login API.

setWFVariables

{"abc":40,"pqr":41}

These are variable names and values which are set when the Pipeline is executed.

Define the API with the Rubiscape backend URL and concatenate it with the following endpoint:
Pipeline/externalPipelineRun/


data = {'sessionKey':sessionKey,'PipelineName': PipelineName,"withData": withData,'inputData':inputData,'projectName':projectName,"workspaceName":workspaceName
"setWFVariables":{"abc":40,"pqr":41}}


Step 5: Call API Using Request Body and Other Parameters

Write code to make a POST HTTP request to the external Pipeline execution API and pass the above data as the request body.


{“resultData”:{…}, “message” : “Pipeline executed successfully.”, “status” : 200}

In above response body “resultData” is the results of executed Pipeline.


1.3 Sample Python Code

Below code is end to end flow for external execution - Login and execution.




1.4 Set Variable Values from External Execution of Pipeline

  1. This functionality lets you update the variable values runtime while executing pipeline externally.
  2. These values get considered in the pipeline execution and in the tasks wherever referenced.

How to setVariables

  1. To set variable values, as can be seen in sample python code added above, you need to add the 'setWFVariables' key in the payload json of pipeline run.
  2. This key has value of dictionary with variable name and variable values.
  3. Note that, the name of variables defined at Pipeline level should match with the names defined in the dictionary at setWFVariables. However if not matched then this will not fail the Pipeline execution, only the proper update of variable values won't take place and values defined at pipeline level will get considered.
  4. The external pipeline execution will not directly update the respective variable values on the modal, these get only considered during that pipeline run in the respective references in the task.

    • Related Articles

    • Setting Variable Values from Scheduled Pipeline Execution

      This functionality allows users to assign specific values to pipeline variables at the schedule level. When a pipeline runs through a schedule, the variable values defined in that schedule override the global variable values. This enables dynamic, ...

    • Multiple Execution

      Rubiscape provides a feature that enables the multiple execution of a complete pipeline. This allows the entire pipeline to be executed multiple times, creating a new instance for each run. Each instance can be configured independently. For multiple ...

    • MLOPS - Machine Learning Operations

      Introduction: Why Rubiscape MLOps? Rubiscape MLOps provides an end-to-end environment for building, tracking, publishing, and serving machine learning models. It ensures experiment reproducibility, streamlined deployment, and centralized model ...

    • Profile

      Overview: The Profile node provides statistical summaries and visual insights for selected columns in a dataset. It helps users understand distributions, missing values, and basic metrics before further transformations. Location: Pipeline → Data ...

    • RubiR

      RubiR is a feature within Code Fusion on the Rubiscape platform to write your code in the R programming language for building models. You can use RubiR as an independent node or connect it to your Reader node (dataset) or other algorithm nodes. This ...