mlex_lightlytrain_segmentation_prototype

lightly_train DINOv3 EoMT models for image segmentation in MLExchange.

Supported models

For a full list of supported lightly_train models, see the lightly_train documentation.

lightly_train model
dinov3/vits16-eomt
dinov3/vits16-eomt-coco
dinov3/vits16-eomt-cityscapes

Usage

# Train
python mlex_lightly/main.py example_yamls/example_lightly_train.yaml --train

# Inference
python mlex_lightly/main.py example_yamls/example_lightly_train.yaml

Docker

docker build -t ghcr.io/mlexchange/mlex_lightlytrain_segmentation_prototype:main .

Copyright

MLExchange Copyright (c) 2026, The Regents of the University of California, through Lawrence Berkeley National Laboratory (subject to receipt of any required approvals from the U.S. Dept. of Energy). All rights reserved.

If you have questions about your rights to use or distribute this software, please contact Berkeley Lab's Intellectual Property Office at IPO@lbl.gov.

NOTICE. This Software was developed under funding from the U.S. Department of Energy and the U.S. Government consequently retains certain rights. As such, the U.S. Government has been granted for itself and others acting on its behalf a paid-up, nonexclusive, irrevocable, worldwide license in the Software to reproduce, distribute copies to the public, prepare derivative works, and perform publicly and display publicly, and to permit others to do so.