With the release of LightlyStudio 0.4.6, we’re excited to introduce plugins, a new way to extend and customize LightlyStudio. Plugins allow users to integrate their own functionality or leverage third-party tools directly within the LightlyStudio workflow.
.png)
Plugins make it easier than ever to tailor the platform to specific use cases, automate repetitive tasks, and experiment with new ideas without leaving the environment.Â
Adding your own plugin is very straightforward: all you need to do is implement a so-called operator in Python. Based on the operator’s definitions, the required GUI inputs are automatically assembled for you, removing the need to manually build or maintain a user interface.Â
This lets you focus entirely on the core logic of your plugin.
For more information, check out LightlyStudio Plugins.
As part of this release, we at Lightly are introducing a plugin that brings LightlyTrain directly into LightlyStudio.Â
This integration enables you to use LightlyTrain for automatic labeling of your datasets straight from the LightlyStudio interface. By combining dataset exploration, curation, and training-based auto-labeling in one place, you can significantly speed up your data preparation workflow.
To get started, take a look at the LightlyStudio Plugins Documentation.
Self-Supervised Pretraining
Leverage self-supervised learning to pretrain models
Smart Data Capturing on Device
Find only the most valuable data directly on device
.png)
.png)
.png)
.svg)