Skip to page content
SAVRN
Search Plan a project

Open-weight model

AdaOcc

by Wjl wjldragon/AdaOcc

AdaOcc is a model from Wjl, released under other (access requested at publisher). Its published files total 7.7 GB.

Accepted to NeurIPS 2026. - Project page (videos and real-robot demos): - Code, data preparation, and reproduction docs: This Hugging Face repository only hosts the public AdaOcc checkpoint assets.

Parameters—
Context—
Weights7.6 GB
Licenseother
AccessAccess requested at publisher
Monthly Downloads—

Model Card

Accepted to NeurIPS 2026. - Project page (videos and real-robot demos): - Code, data preparation, and reproduction docs: This Hugging Face repository only hosts the public AdaOcc checkpoint assets. It does not include OccScanNet data, generated labels/depth maps, RADIO weights, the Depth-Anything-V2 fine-tuned checkpoint, or upstream MapAnything assets referenced by the config snapshots. The full-split checkpoint, config snapshot, and training/evaluation log come from the same run. The epoch-100 validation line reports mIoU=59.67 and IoU=65.29; this is also the best validation result recorded in that log. Here, full refers to the full OccScanNet data split, not to an ablation label. The run…

Excerpt from the card by Wjl, licensed other.

Identity and Version

Repository
wjldragon/AdaOcc
Publisher
Wjl
Task
Not stated by the source
Modality
Other
Library
pytorch
Parameters
Not stated by the source
Languages
Not stated by the source
Revision
bbb3f373ce6858f98371426d79372ec99773ec63
First published
2026-06-03
Last updated
2026-09-29

Files and Weights

11 files, 7.7 GB in total. The weights are 3 files totalling 7.6 GB in pth.

Weights3 files · 7.6 GB
Configuration2 files · 104.0 KB
Documentation1 file · 4.8 KB
Other4 files · 50.1 MB
Repository1 file · 1.6 KB
Every file
FileTypeSizeSHA-256
checkpoints/adaocc_online_depth_occscannet_mini_epoch200.pthWeights4.0 GB —
checkpoints/adaocc_radio_occscannet_full_epoch100.pthWeights3.6 GB —
pretrain/fusion_pretrain_model.pthWeights10.8 MB —
configs/radio_occscannet_full_training_snapshot.pyConfiguration57.9 KB —
configs/radio_occscannet_mini_training_snapshot.pyConfiguration46.1 KB —
README.mdDocumentation4.8 KB —
SHA256SUMSOther801 B —
logs/online_depth_occscannet_mini_epoch200.logOther8.2 MB —
logs/radio_occscannet_full_epoch100.logOther41.8 MB —
upload_to_hf.shOther214 B —
.gitattributesRepository1.6 KB —

License and Download

License
other
Access
Access requested at publisher
Download size
7.6 GB
Request access from Wjl

Wjl grants access through its official repository on Hugging Face.

Memory Requirements

PrecisionWeights in memory
As published7.6 GB

Weights only, from the published parameter count; the key-value cache and runtime add to this.

Questions About AdaOcc

What license is AdaOcc released under?

other, as its publisher declares it. Read the license text before commercial use.