Skip to content

Creating Inference pipelines

Introduction

Here we discuss running an inference pipeline for on-cloud implementation. On-premise implementation is almost identical, and the difference would be highlighted for each of the components individually. In order to create a training pipeline, click on “create training pipeline, “and you will proceed to the first step – “Pipeline details.”

Pipeline details
Inference pipeline details

In this section, the following fields are filled by the user. The name of the pipeline and the description would be displayed on the pipeline run page. The git username and git token give access to your code to Robotika. You can skip the git credentials and directly upload the files when configuring the components. The organization name is fixed and cannot be modified.

Pipeline component config
Pipeline component config

For the pipeline to run the following components should be configured:

  1. ExampleGen
  2. BatchInference
  3. CSVWriter

The components that are not configured on this step but contains valuable information during the debugging and after the deployment are:

  1. StatisticGen
  2. SchemaGen

Please refer to the page corresponding to the specific component for more details.

Scheduler
Scheduler

The inference pipeline scheduler is responsible for scheduling batch inferencing for the ML pipeline. Currently, Robotika supports only retraining based on the time schedule. Users can choose the periodicity of the Inference. Please note that the first execution starts as soon as the user clicks the button “Save.”