### Environment Setup and Dependency Installation

Source: https://github.com/tsinghua-fib-lab/worldmove/blob/main/README.md

Instructions for setting up a Python virtual environment using conda or virtualenv and installing project dependencies from requirements.txt.

```bash
conda create -n move python=3.10
conda activate move
```

```bash
virtualenv -p python3.10 move
source move/bin/activate
```

```bash
pip install -r requirements.txt
```

--------------------------------

### Project Dependencies

Source: https://github.com/tsinghua-fib-lab/worldmove/blob/main/requirements.txt

This section lists the Python packages and their exact versions used in the project. These dependencies are crucial for setting up a consistent development and execution environment.

```python
matplotlib==3.8.4
numpy==2.0.0
protobuf==5.27.1
pytorch_lightning==2.2.1
PyYAML==6.0.1
ray==2.9.3
scikit_learn==1.3.2
setproctitle==1.3.3
sortedcontainers==2.4.0
tensorflow==2.16.1
torch==2.1.2
torch_cluster==1.6.3+pt21cu121
torch_geometric==2.5.3
tqdm==4.66.2
```

--------------------------------

### Generate Population Data

Source: https://github.com/tsinghua-fib-lab/worldmove/blob/main/README.md

Script to acquire population data from WorldPop, using a configuration file and saving the output to a specified directory.

```python
python3 scripts/generate_pop.py --config configs/{city}.yml --save /path/to/pop_dir
```

--------------------------------

### Generate Location Profiling Data

Source: https://github.com/tsinghua-fib-lab/worldmove/blob/main/README.md

Script to create location profiling data by integrating population distribution and POI attributes. It uses a configuration file and saves the output to a specified directory.

```python
python3 scripts/generate_profile.py --config configs/{city}.yml --save /path/to/pop_dir
```

--------------------------------

### Generate Mobility Data

Source: https://github.com/tsinghua-fib-lab/worldmove/blob/main/README.md

Script to generate mobility data using prepared location profiling and population data. It requires a configuration file and saves the output to a specified directory.

```python
python3 scripts/generate_mobility --config configs/{city}.yml --save /path/to/pop_dir
```

--------------------------------

### Train Diffusion Model

Source: https://github.com/tsinghua-fib-lab/worldmove/blob/main/README.md

Command to train the diffusion model for generating mobility trajectories. Requires a configuration file specifying dataset path and hyperparameters. The trained model is saved to a log directory.

```python
python3 train.py --config configs/{city}.yml --save /path/to/log_dir
```

--------------------------------

### Citation Information

Source: https://github.com/tsinghua-fib-lab/worldmove/blob/main/README.md

Provides BibTeX entries for citing the research papers associated with the WorldMove project. These entries include author, title, journal, year, and DOI information.

```bibtex
@article{10.1093/pnasnexus/pgaf081,
    author = {Yuan, Yuan and Ding, Jingtao and Jin, Depeng and Li, Yong},
    title = {Learning the complexity of urban mobility with deep generative network},
    journal = {PNAS Nexus},
    volume = {4},
    number = {5},
    pages = {pgaf081},
    year = {2025},
    month = {05},
    issn = {2752-6542},
    doi = {10.1093/pnasnexus/pgaf081},
    url = {https://doi.org/10.1093/pnasnexus/pgaf081},
    eprint = {https://academic.oup.com/pnasnexus/article-pdf/4/5/pgaf081/63066168/pgaf081.pdf},
}

@inproceedings{10.1145/3696410.3714516,
    author = {Zhang, Yuheng and Yuan, Yuan and Ding, Jingtao and Yuan, Jian and Li, Yong},
    title = {Noise Matters: Diffusion Model-based Urban Mobility Generation with Collaborative Noise Priors},
    year = {2025},
    isbn = {9798400712746},
    publisher = {Association for Computing Machinery},
    address = {New York, NY, USA},
    url = {https://doi.org/10.1145/3696410.3714516},
    doi = {10.1145/3696410.3714516},
    booktitle = {Proceedings of the ACM on Web Conference 2025},
    pages = {5352–5363},
    numpages = {12},
    keywords = {diffusion models, sustainable cities, urban mobility},
    location = {Sydney NSW, Australia},
    series = {WWW '25}
}

@misc{yuan2025worldmoveglobalopendata,
    title={WorldMove, a global open data for human mobility},
    author={Yuan Yuan and Yuheng Zhang and Jingtao Ding and Yong Li},
    year={2025},
    eprint={2504.10506},
    archivePrefix={arXiv},
    primaryClass={cs.SI},
    url={https://arxiv.org/abs/2504.10506},
}
```

--------------------------------

### Train Location Feature Encoder

Source: https://github.com/tsinghua-fib-lab/worldmove/blob/main/README.md

Command to train the location feature encoder, which compresses regional characteristics into a unified embedding space. The output is saved to a specified directory.

```python
python3 loc_encoder.py --save /path/to/save_dir
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.