Skip to content

install.md

Latest commit

 

History

History
84 lines (67 loc) · 2.44 KB

install.md

File metadata and controls

84 lines (67 loc) · 2.44 KB
Raw

Download and installation

1. Clone the navsim-devkit

Clone the repository

git clone https://github.com/AIR-THU/StyleDrive.git
cd StyleDrive

2. Download StyleDrive dataset

Download styletrain.json and styletest.json from huggingface.

mkdir dataset/extra_data
cd dataset/extra_data
wget https://huggingface.co/datasets/Ryhn98/StyleDrive-Dataset/blob/main/styletest.json
wget https://huggingface.co/datasets/Ryhn98/StyleDrive-Dataset/blob/main/styletrain.json
cd ../..

Then, you need to download the OpenScene logs and sensor blobs, as well as the nuPlan maps. We provide scripts to download the nuplan maps, the mini split and the test split. Navigate to the download directory and download the maps

cd download && ./download_maps

Next download the following data splits:

./download_navtrain
./download_test

This will download the splits into the download directory. From there, move it to create the following structure.

~/StyleDrive
├── navsim (containing the devkit)
├── exp
└── dataset
    ├── extra_data
    |    ├── styletrain.json
    |    ├── styletest.json
    ├── maps
    ├── navsim_logs
    |    ├── test
    |    ├── trainval
    └── sensor_blobs
    |    ├── test
    |    ├── trainval

Set the required environment variables, by adding the following to your ~/.bashrc file Based on the structure above, the environment variables need to be defined as:

export NUPLAN_MAP_VERSION="nuplan-maps-v1.0"
export NUPLAN_MAPS_ROOT="$HOME/StyleDrive/dataset/maps"
export NAVSIM_EXP_ROOT="$HOME/StyleDrive/exp"
export NAVSIM_DEVKIT_ROOT="$HOME/StyleDrive/navsim"
export OPENSCENE_DATA_ROOT="$HOME/StyleDrive/dataset"

3. Install the navsim-devkit and specific packages for baseline models

Then, install navsim. To this end, create a new environment and install the required dependencies:

conda env create --name navsim -f environment.yml
conda activate navsim
pip install -e .
pip install diffusers einops

4. Modify Path in Agents' Configs

Finally, modify the styletrain_path and styletest_path in Agents' configs:

  1. $NAVSIM_DEVKIT_ROOT/planning/script/config/common/agent/transfuser_style_agent.yaml
  2. $NAVSIM_DEVKIT_ROOT/planning/script/config/common/agent/diffusiondrive_style_agent.yaml
  3. $NAVSIM_DEVKIT_ROOT/planning/script/config/common/agent/ego_status_mlp_style_agent.yaml