VideoActionModel

(★ 164)

VaViM and VaVAM: Autonomous Driving through Video Generative Modeling (official repository).

File Explorer

Download Latest Version (.zip)

- - frontal_0103_run_45.gif
- - callbacks_action_training.yaml
  - callbacks_debug.yaml
  - callbacks_finetuning.yaml
  - callbacks_grid_search.yaml
  - callbacks_nuplan_training.yaml
  - callbacks_opendv_training.yaml
  - callbacks_opendv_validate.yaml
  - callbacks_training.yaml
  - device_stats_monitor.yaml
  - early_stopping.yaml
  - learning_rate_monitor.yaml
  - model_checkpoint.yaml
  - model_summary.yaml
  - none.yaml
  - rich_progress_bar.yaml
  - tqdm_progress_bar.yaml
  - trajectory_logging.yaml
- - ego_trajectory_dataset.yaml
  - finetuning_mix.yaml
  - tokenized_sequence_opendv.yaml
- - action_learning.yaml
  - finetune_mix_complet.yaml
  - video_pretraining_GPT2_llamagen_ds16_16384_opendv.yaml
- - default.yaml
- - csv.yaml
  - many_loggers.yaml
  - none.yaml
  - tensorboard.yaml
  - wandb.yaml
- - muAdamW.yaml
- - default.yaml
  - eval_paths_jeanzay.yaml
  - eval_paths_local.yaml
  - jeanzay_debug.yaml
  - jeanzay_nuplan.yaml
- - none.yaml
  - step_lr.yaml
  - warmup_stable_drop.yaml
- - ddp.yaml
  - deepspeed2.yaml
  - default.yaml
  - gpu.yaml
- __init__.py
- train.yaml
- Dockerfile
- - _dispatch_scenario.slurm
  - _singularity_job.sh
  - aggregate_results.py
  - build_singularity_images.slurm
  - create_mp4.py
  - create_mp4.slurm
  - grid_ncap_evaluation.sh
  - run_neuro_ncap_eval.sh
  - run_single_job.sh
- README.md
- runner.py
- server.py
- actionexpert_24layers_baseattentiondim256_baseembeddingdim64.bsh
- gpt2_24layers_basewidth256.bsh
- qualitative_study_diffusion.ipynb
- qualitative_study_video_generation.ipynb
- scaling_laws.ipynb
- create_vam_env_jeanzay.sh
- depth_anything_a_dataset.py
- download_pretrained_models.py
- evaluate_ego_trajectory.py
- evaluate_ego_trajectory.slurm
- fuse_multiple_checkpoints.py
- fused_checkpoint.py
- fused_checkpoint.sh
- handle_checkpoints.py
- hbird_evaluation.py
- hbird_evaluation.slurm
- hbird_evaluation_depth.slurm
- nxt_evaluation.py
- nxt_evaluation.slurm
- prepare_checkpoint_release.sh
- quality_evaluation.py
- quality_evaluation.slurm
- README.md
- regain_index_from_train.py
- video_qualitative_results.py
- - __init__.py
  - action_learning.py
  - joint_model.py
  - mup_action_expert.py
  - video_action_model.py
- - __init__.py
  - create_opendv_tokens.py
  - data_mixing.py
  - ego_trajectory_datamodule.py
  - ego_trajectory_dataset.py
  - finetuning_datamodule.py
  - opendv_tokens_datamodule.py
  - opendv_tokens_dataset.py
  - README.md
  - stateful_dataloader.py
  - token_creator.py
  - transforms.py
- - __init__.py
  - depth.py
  - ego_trajectory_forecasting.py
  - hbird.py
  - miou.py
  - quality.py
- - __init__.py
  - boolean_flag.py
  - cmd_line_logging.py
  - create_mp4_from_folder.py
  - expand_path.py
  - hyperparam_logging.py
  - info_printing.py
  - instantiators.py
  - plot_utils.py
  - read_eval_config.py
  - task_utils.py
  - tensorboard_logging.py
  - torch_dtype.py
  - trajectory_logging.py
  - warmup_stable_drop.py
- - __init__.py
  - mup_gpt2.py
  - next_token_predictor.py
  - prepare_token_sequence.py
- __init__.py
- train.py
.flake8
.gitignore
CITATION.cff
jeanzay_slurm_job_submit.py
LICENSE
LICENSE_DATA
LICENSE_MODEL
lint.sh
MODELS.md
pyproject.toml
README.md

# Installation Guide

1. Get the code

git clone https://github.com/valeoai/VideoActionModel

Downloads the entire project code from GitHub to your computer.

cd VideoActionModel

Moves into the project folder you just downloaded.

2. Official Install Script

Easy Recommended

Prerequisites

Python 3 Python is required to use pip.

pip install scann

Installs the package published on PyPI directly — no need to clone the source.

✅ After installing, open a new terminal and run the program's version command (e.g. --version) to confirm it worked.

Pulled directly from this repo's README.

3. Docker

Easy

Prerequisites

Git Needed to download the project code from GitHub.
Docker Desktop Needed to build and run containers. Install it and keep it running in the background.

docker build -f docker/Dockerfile -t videoactionmodel .

Builds a runnable image based on the Dockerfile.

docker run -p 8080:80 videoactionmodel

Runs the built image as an actual container.

✅ Run docker compose ps to check the containers are Up. If the README mentions a port, open http://localhost:PORT in your browser.

4. Python

Easy

Prerequisites

Git Needed to download the project code from GitHub.
Python 3 On Windows, be sure to check 'Add Python to PATH' during installation.

pip install -e .

Installs the Python libraries listed in requirements.txt (or similar).

pip install scann

Installs the package published on PyPI directly — no need to clone the source.

✅ If it runs without errors and prints output in the terminal, it worked.

Pulled directly from this repo's README.