trajdata: A Unified Interface to Multiple Human Trajectory Datasets

Announcements

Sept 2023: Our paper about trajdata has been accepted to the NeurIPS 2023 Datasets and Benchmarks Track!

Installation

The easiest way to install trajdata is through PyPI with

pip install trajdata

In case you would also like to use datasets such as nuScenes, Lyft Level 5, or Waymo Open Motion Dataset (which require their own devkits to access raw data or additional package dependencies), the following will also install the respective devkits and/or package dependencies.

# For nuScenes
pip install "trajdata[nusc]"

# For Lyft
pip install "trajdata[lyft]"

# For Waymo
pip install "trajdata[waymo]"

# For INTERACTION
pip install "trajdata[interaction]"

# All
pip install "trajdata[nusc,lyft,waymo,interaction]"

Then, download the raw datasets (nuScenes, Lyft Level 5, ETH/UCY, etc.) in case you do not already have them. For more information about how to structure dataset folders/files, please see DATASETS.md.

Package Developer Installation

First, in whichever environment you would like to use (conda, venv, ...), make sure to install all required dependencies with

pip install -r requirements.txt

Then, install trajdata itself in editable mode with

pip install -e .

Data Preprocessing [Optional]

The dataloader operates via a two-stage process, visualized below. While optional, we recommend first preprocessing data into a canonical format. Take a look at the examples/preprocess_data.py script for an example script that does this. Data preprocessing will execute the first part of the diagram above and create data caches for each specified dataset.

Note: Explicitly preprocessing datasets like this is not necessary; the dataloader will always internally check if there exists a cache for any requested data and will create one if not.

Data Loading

At a minimum, batches of data for training/evaluation/etc can be loaded the following way:

import os
from torch.utils.data import DataLoader
from trajdata import AgentBatch, UnifiedDataset

# See below for a list of already-supported datasets and splits.
dataset = UnifiedDataset(
    desired_data=["nusc_mini"],
    data_dirs={  # Remember to change this to match your filesystem!
        "nusc_mini": "~/datasets/nuScenes"
    },
)

dataloader = DataLoader(
    dataset,
    batch_size=64,
    shuffle=True,
    collate_fn=dataset.get_collate_fn(),
    num_workers=os.cpu_count(), # This can be set to 0 for single-threaded loading, if desired.
)

batch: AgentBatch
for batch in dataloader:
    # Train/evaluate/etc.
    pass

For a more comprehensive example, please see examples/batch_example.py.

For more information on all of the possible UnifiedDataset constructor arguments, please see src/trajdata/dataset.py.

Supported Datasets

Currently, the dataloader supports interfacing with the following datasets:

Dataset	ID	Splits	Locations	Description	dt	Maps
nuScenes Train/TrainVal/Val	nusc_trainval	train, train_val, val	boston, singapore	nuScenes prediction challenge training/validation/test splits (500/200/150 scenes)	0.5s (2Hz)	✅
nuScenes Test	nusc_test	test	boston, singapore	nuScenes test split, no annotations (150 scenes)	0.5s (2Hz)	✅
nuScenes Mini	nusc_mini	mini_train, mini_val	boston, singapore	nuScenes mini training/validation splits (8/2 scenes)	0.5s (2Hz)	✅
nuPlan Train	nuplan_train	N/A	boston, singapore, pittsburgh, las_vegas	nuPlan training split (947.42 GB)	0.05s (20Hz)	✅
nuPlan Validation	nuplan_val	N/A	boston, singapore, pittsburgh, las_vegas	nuPlan validation split (90.30 GB)	0.05s (20Hz)	✅
nuPlan Test	nuplan_test	N/A	boston, singapore, pittsburgh, las_vegas	nuPlan testing split (89.33 GB)	0.05s (20Hz)	✅
nuPlan Mini	nuplan_mini	mini_train, mini_val, mini_test	boston, singapore, pittsburgh, las_vegas	nuPlan mini training/validation/test splits (942/197/224 scenes, 7.96 GB)	0.05s (20Hz)	✅
Waymo Open Motion Training	waymo_train	train	N/A	Waymo Open Motion Dataset training split	0.1s (10Hz)	✅
Waymo Open Motion Validation	waymo_val	val	N/A	Waymo Open Motion Dataset validation split	0.1s (10Hz)	✅
Waymo Open Motion Testing	waymo_test	test	N/A	Waymo Open Motion Dataset testing split	0.1s (10Hz)	✅
Lyft Level 5 Train	lyft_train	train	palo_alto	Lyft Level 5 training data - part 1/2 (8.4 GB)	0.1s (10Hz)	✅
Lyft Level 5 Train Full	lyft_train_full	train	palo_alto	Lyft Level 5 training data - part 2/2 (70 GB)	0.1s (10Hz)	✅
Lyft Level 5 Validation	lyft_val	val	palo_alto	Lyft Level 5 validation data (8.2 GB)	0.1s (10Hz)	✅
Lyft Level 5 Sample	lyft_sample	mini_train, mini_val	palo_alto	Lyft Level 5 sample data (100 scenes, randomly split 80/20 for training/validation)	0.1s (10Hz)	✅
Argoverse 2 Motion Forecasting	av2_motion_forecasting	train, val, test	N/A	250,000 motion forecasting scenarios of 11s each	0.1s (10Hz)	✅
INTERACTION Dataset Single-Agent	interaction_single	train, val, test, test_conditional	usa, china, germany, bulgaria	Single-agent split of the INTERACTION Dataset (where the goal is to predict one target agents' future motion)	0.1s (10Hz)	✅
INTERACTION Dataset Multi-Agent	interaction_multi	train, val, test, test_conditional	usa, china, germany, bulgaria	Multi-agent split of the INTERACTION Dataset (where the goal is to jointly predict multiple agents' future motion)	0.1s (10Hz)	✅
ETH - Univ	eupeds_eth	train, val, train_loo, val_loo, test_loo	zurich	The ETH (University) scene from the ETH BIWI Walking Pedestrians dataset	0.4s (2.5Hz)
ETH - Hotel	eupeds_hotel	train, val, train_loo, val_loo, test_loo	zurich	The Hotel scene from the ETH BIWI Walking Pedestrians dataset	0.4s (2.5Hz)
UCY - Univ	eupeds_univ	train, val, train_loo, val_loo, test_loo	cyprus	The University scene from the UCY Pedestrians dataset	0.4s (2.5Hz)
UCY - Zara1	eupeds_zara1	train, val, train_loo, val_loo, test_loo	cyprus	The Zara1 scene from the UCY Pedestrians dataset	0.4s (2.5Hz)
UCY - Zara2	eupeds_zara2	train, val, train_loo, val_loo, test_loo	cyprus	The Zara2 scene from the UCY Pedestrians dataset	0.4s (2.5Hz)
Stanford Drone Dataset	sdd	train, val, test	stanford	Stanford Drone Dataset (60 scenes, randomly split 42/9/9 (70%/15%/15%) for training/validation/test)	0.0333...s (30Hz)

Adding New Datasets

The code that interfaces the original datasets (dealing with their unique formats) can be found in src/trajdata/dataset_specific.

To add a new dataset, one needs to:

• Create a new folder under src/trajdata/dataset_specific which will contain all the code specific to a particular dataset (e.g., for extracting data into our canonical format). In particular, there must be:
  • An __init__.py file.
  • A file that defines a subclass of RawDataset and implements some of its functions. Reference implementations can be found in the nusc/nusc_dataset.py, lyft/lyft_dataset.py, and eth_ucy_peds/eupeds_dataset.py files.
• Add a subclass of NamedTuple to src/trajdata/dataset_specific/scene_records.py which contains the minimal set of information sufficient to describe a scene. This "scene record" will be used in conjunction with the raw dataset class above and relates to how scenes are stored and efficiently accessed later.
• Add a section to the DATASETS.md file which outlines how users should store the raw dataset locally.
• Add a section to src/trajdata/utils/env_utils.py which allows users to get the raw dataset via its name, and specify if the dataset is a good candidate for parallel processing (e.g., does its native dataset object have a large memory footprint which might not allow it to be loaded in multiple processes, such as nuScenes?) and if it has maps.

Examples

Please see the examples/ folder for more examples, below are just a few demonstrations of core capabilities.

Multiple Datasets

The following will load data from both the nuScenes mini dataset as well as the ETH - University scene from the ETH BIWI Walking Pedestrians dataset.

dataset = UnifiedDataset(
    desired_data=["nusc_mini", "eupeds_eth"],
    data_dirs={  # Remember to change this to match your filesystem!
        "nusc_mini": "~/datasets/nuScenes",
        "eupeds_eth": "~/datasets/eth_ucy_peds"
    },
    desired_dt=0.1, # Please see the note below about common dt!
)

Note: Be careful about loading multiple datasets without an associated desired_dt argument; many datasets do not share the same underlying data annotation frequency. To address this, we've implemented timestep interpolation to a common frequency which will ensure that all batched data shares the same dt. Interpolation can only be performed to integer multiples of the original data annotation frequency. For example, nuScenes' dt=0.5 and the ETH BIWI dataset's dt=0.4 can be interpolated to a common desired_dt=0.1.

Map API

trajdata also provides an API to access the raw vector map information from datasets that provide it.

from pathlib import Path
from trajdata import MapAPI, VectorMap

cache_path = Path("~/.unified_data_cache").expanduser()
map_api = MapAPI(cache_path)

vector_map: VectorMap = map_api.get_map("nusc_mini:boston-seaport")

Simulation Interface

One additional feature of trajdata is that it can be used to initialize simulations from real data and track resulting agent motion, metrics, etc.

At a minimum, a simulation can be initialized and stepped through as follows (also present in examples/simple_sim_example.py):

from typing import Dict # Just for type annotations

import numpy as np

from trajdata import AgentBatch, UnifiedDataset
from trajdata.data_structures.scene_metadata import Scene # Just for type annotations
from trajdata.simulation import SimulationScene

# See below for a list of already-supported datasets and splits.
dataset = UnifiedDataset(
    desired_data=["nusc_mini"],
    data_dirs={  # Remember to change this to match your filesystem!
        "nusc_mini": "~/datasets/nuScenes",
    },
)

desired_scene: Scene = dataset.get_scene(scene_idx=0)
sim_scene = SimulationScene(
    env_name="nusc_mini_sim",
    scene_name="sim_scene",
    scene=desired_scene,
    dataset=dataset,
    init_timestep=0,
    freeze_agents=True,
)

obs: AgentBatch = sim_scene.reset()
for t in range(1, sim_scene.scene.length_timesteps):
    new_xyh_dict: Dict[str, np.ndarray] = dict()

    # Everything inside the forloop just sets
    # agents' next states to their current ones.
    for idx, agent_name in enumerate(obs.agent_name):
        curr_yaw = obs.curr_agent_state[idx, -1]
        curr_pos = obs.curr_agent_state[idx, :2]

        next_state = np.zeros((3,))
        next_state[:2] = curr_pos
        next_state[2] = curr_yaw
        new_xyh_dict[agent_name] = next_state

    obs = sim_scene.step(new_xyh_dict)

examples/sim_example.py contains a more comprehensive example which initializes a simulation from a scene in the nuScenes mini dataset, steps through it by replaying agents' GT motions, and computes metrics based on scene statistics (e.g., displacement error from the original GT data, velocity/acceleration/jerk histograms).

Citation

If you use this software, please cite it as follows:

@Inproceedings{ivanovic2023trajdata,
  author = {Ivanovic, Boris and Song, Guanyu and Gilitschenski, Igor and Pavone, Marco},
  title = {{trajdata}: A Unified Interface to Multiple Human Trajectory Datasets},
  booktitle = {{Proceedings of the Neural Information Processing Systems (NeurIPS) Track on Datasets and Benchmarks}},
  month = dec,
  year = {2023},
  address = {New Orleans, USA},
  url = {https://arxiv.org/abs/2307.13924}
}

TODO

• Create a method like finalize() which writes all the batch information to a TFRecord/WebDataset/some other format which is (very) fast to read from for higher epoch training.