Instructions to use LeeLiaoLiao/MoLA-ckpt with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use LeeLiaoLiao/MoLA-ckpt with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("LeeLiaoLiao/MoLA-ckpt", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
MoLA Checkpoints
This repository contains pretrained checkpoints for From Imagined Futures to Executable Actions: Mixture of Latent Actions for Robot Manipulation.
Contents
calvin/
action_model/
saved_models/
last.pt
idms/
flow.bin
depth.bin
semantic.bin
libero/
action_model/
libero_spatial/
last.pt
libero_object/
last.pt
libero_goal/
last.pt
libero_10/
last.pt
idms/
flow.bin
depth.bin
semantic.bin
video_model/
model_index.json
feature_extractor/
image_encoder/
scheduler/
unet/
vae/
Usage
Use these checkpoints with the MoLA codebase:
git clone https://github.com/LeeLiaoLiao/MoLA-OpenSource.git
For LIBERO evaluation, pass the corresponding action model directory, video model directory, and suite name:
bash scripts/rollout_libero.sh \
/path/to/MoLA-ckpt/libero/action_model/libero_10 \
/path/to/MoLA-ckpt/libero/video_model \
/path/to/clip-vit-base-patch32 \
/path/to/LIBERO \
1 \
libero_10
For CALVIN evaluation:
bash scripts/rollout_calvin.sh \
/path/to/MoLA-ckpt/libero/video_model \
/path/to/MoLA-ckpt/calvin/action_model/saved_models \
/path/to/clip-vit-base-patch32 \
/path/to/calvin/task_ABC_D \
1
- Downloads last month
- -