3LC Version 2.0 Release Notes#

  1. November 2023

We proudly present 3LC Version 2.0!

This release is given to early adopters of 3LC, to be deployed in their own environment. 3LC no longer requires any form of data conversion and will work natively with training data in common dataset formats.

This version provides many enhancements to computer vision workflows and adds sophisticated tools for viewing and modifying bounding-boxes.

Major Features#

  • Vastly improved Dashboard with hundreds of enhancements.

  • Python Notebook API that can be integrated into Jupyter Notebooks.

    • Works natively with PyTorch datasets, no conversion required.

  • Numerous example notebooks illustrating how to use 3LC with common computer vision models.

  • Greatly improved documentation.


This release is provided to enterprise clients, who have been given access to install it from our private CloudRepo Python repository. Clients will be able to locally install Python wheels which contain the notebook API, the 3LC Dashboard, and documentation from Python packages.

In addition to the credentials to access the private CloudRepo, clients will also need a license key in order to use the software.

Supported Platforms#

  • Python 3.8 - Python 3.11

    • Both Conda and “vanilla” Python environments should work

  • Microsoft Windows 10 and 11 (x86)

  • Ubuntu 20.04 (x86-64) is our supported Linux platform

    • Most other GLibc based Linux distributions are expected to work, but these are untested and unsupported.

  • Chrome and Edge web-browsers, with GPU acceleration enabled


  • Notebooks written against the version 1.0 API are no longer functional.

  • The 3LC-Dataformat has been deprecated and no longer supported.

  • The Ideas panel has been removed from the Dashboard, since it was not a good idea.

Known Issues#

  • Tables for dataset revisions are currently always stored at a location next to the input table, and it is not possible to override that behavior. This means, for example, that writing a dataset revision with an input table stored in a read-only location (on disk, in cloud storage, etc.) is not supported.

  • The Table object in the tlc Python API is designed to represent immutable columnar data, but it currently returns objects by reference when iterating or indexing. Consequently, it is possible to modify the in-memory representation of the Table, which could then get cached to disk. In general, users of the API should not make such modifications.

  • It is not currently possible to deactivate samples so that metrics are collected on only active samples.

  • The order of columns in the Dashboard filter panel does not follow the order in the tables panel, where the columns are more logically ordered. This will be addressed in an upcoming release.

  • When rendering more than ~500k samples (depending on hardware) in a chart in the Dashboard, interactivity may suffer as the number of samples increases.

  • The 3LC public S3 buckets, accessed when invoking the 3lc command with the --with-public-examples flag, are intrinsically public. However, due to AWS IAM policies, they cannot be accessed by IAM entities from other accounts unless those entities are explicitly permitted to do so by policy. Users should launch the service without any S3 credentials, or they should attach a specific IAM policy to their IAM entities to enable access to the 3LC public S3 buckets. A policy for doing so is provided in the documentation.