PoLAr-MAE β€” semantic segmentation

PoLAr-MAE fine-tuned for 4-class LArTPC semantic segmentation (shower, track, Michel, delta); reproduces the paper mF1 β‰ˆ 0.82.

  • pimm type: PoLArMAE-SemSeg Β· 4 classes

Loading

import pimm
model = pimm.from_pretrained("hf://deeplearnphysics/polar-mae-semantic")

Provenance

Repackaged from the original PoLAr-MAE release checkpoints into the pimm export format. Inherits the source repo license.

Downloads last month
26
Safetensors
Model size
22.9M params
Tensor type
F32
Β·
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support

Dataset used to train DeepLearnPhysics/PoLAr-MAE-Semantic

Collection including DeepLearnPhysics/PoLAr-MAE-Semantic

Paper for DeepLearnPhysics/PoLAr-MAE-Semantic