Code Process Orchestration

Code process orchestration refers to creating streams by using a composition method similar to a deep learning framework to avoid writing complex pipeline files. In this way, the process orchestration can be quickly and efficiently implemented.

Basic Concepts

Table 1 Parameter Description

Name

Class Name

Description

Description

PluginNode

PluginNode

Plugin node

Defines a plugin, including its name, type, and properties.

Stream

Stream

Service stream

A complete inference service stream, which is concatenated by plugins.

Sequential Stream

SequentialStream

Sequential service stream

Processes sequential services, that is, the plugins are in a sequential relationship.

Functional Stream

FunctionalStream

Functional service stream

Processes complex relationships such as multi-input multi-out (MIMO) process.

Sample Overview

Figure 1 Example of orchestrating code

The code orchestration process is divided into four steps.

  1. (Optional) Configure properties based on the plugin requirements.
  2. Create service streams and configure the processor. Service flows are used for subsequent plugin combination and data processing.
  3. Create plugin nodes and construct inference process.
  4. Process data. After an input object is created, SendData and GetResult of the stream are used to transfer data and obtain results.