Start Building

Pipelogic is made for product teams and therefore comes with all the tools teams need to build solutions that work in real-world conditions. If you are a researcher or developer, you may be interested to use our Pipelogic CLI or browser-based Code Editor to build an efficient component in one of our supported languages, e.g. C++ or Python. But if you are a product manager or business analyst, you may be interested to browse through pre-made template pipelines or use our visual programming tool called Pipelogic Studio to quickly prototype your business cases.

Pipelines

The most high-level abstraction that hides the complexity of AI solutions via simplified visual representation of data flows and components. When looking at the pipeline it becomes immediately clear what are its data sources, outputs, main AI components and transformations. The pipeline could for instance connect to a camera to detect traffic jams, or use a microphone to recognize speech commands and turn off your shutters or heating at home.

A pipeline is made of components, where the output of one component serves as the input for the next. It is a systematic workflow that involves the processing, transformation, and movement of data or tasks through various stages optimized for stream processing.

Create Pipelines

To create your own pipeline, navigate to Pipelines page and click on Create new pipeline button. Enter the pipeline name, and then proceed to Pipelogic Studio by clicking on the Create button. Your pipeline can be shared with the rest of the users on your workspace, or it can be featured on your public profile.

Use Template Pipelines

Pipelogic provides a range of pre-existing pipelines that can be used out of the box or customized for a specific purpose. To browse through pre-existing public pipelines, navigate to Pipelines → Explore Templates page.

In simple terms component is a container for compiled code to execute a particular task. It can be shared and reused in multiple pipelines. Components may have typed inputs and outputs, making sure they connect with other components only if types match. A component could for instance load the model and run inference with pre- and post- processing, or take a sequence of bounding boxes and apply Kalman filter for tracking of objects.

Select A Template

Go to https://app.pipelogic.ai/components and click on Create new Component. There you can choose different component templates for C++ and Python.

For this example select Double Value, enter your component name.

Congratulations, you created your first component.

Build

First, the component needs to be built and compiled before it can be released.

Click on the build:

Release

Click on release button:

Confirm release:

Congratulations, you released your first component.

Component Structure

├── component.yml -> Component configuration file, where component metadata is defined
├── icon.svg -> icon of component
├── README.md
└── src -> contains source code with all the input/output sources
    ├── requirements.txt
    └── main.py