Skip to content

Latest commit

 

History

History
21 lines (17 loc) · 429 Bytes

File metadata and controls

21 lines (17 loc) · 429 Bytes

World Model

Post-train dataset processing

# here we provide the post-train dataset processing script
cd tools/process/post_train

# take oxe toto for example, remember to change the oxe path and output path
python oxe_toto.py

World Model Training

# train the world model
bash scripts/pretrain/train_video_1node.sh

Inference example

python models/inference/inference_vision.py