density_field_properties

Tools for computing and analyzing properties derived from the dark matter density field in cosmological simulations.

This repository provides methods and scripts for:

• Building the density field from dark matter particles using grid assignment schemes (CIC, TSC, etc.).
• Computing the tidal tensor and its eigenvalues/eigenvectors.
• Deriving related quantities such as tidal anisotropy and linear bias.
• Performing auxiliary analyses on density fields, including smoothing, Fourier transforms, and memory-efficient batching.

---

Installation

Clone the repository and create the environment:

git clone https://github.com/<your-username>/density_field_properties.git
cd density_field_properties
conda env create -f environment.yml
conda activate density_field

Alternatively, install it as a local editable package:

pip install -e .

---

Pre-commit installation

To ensure consistent code style and quality, this repository uses pre-commit hooks configured with:

• Black — automatic code formatter
• isort — import sorting
• flake8 — code linting

Install pre-commit

pip install pre-commit

Set up hooks in your local repository

pre-commit install

Run checks manually (optional)

pre-commit run --all-files

The hooks will automatically run on every git commit, ensuring that code is formatted, imports are sorted, and style issues are caught early.

---

Main dependencies

• numpy
• scipy
• h5py
• matplotlib
• tqdm
• numba (optional, for JIT acceleration)

---

Scientific context

This codebase is designed for studies involving:

• Assembly bias and extended HOD models (e.g. hod_madrid_py).
• Tidal tensor and anisotropy in large-scale structure (Haloscope project).
• Applications to UNIT (UNIT simulations), FastPM (FastPM GitHub), and CAMELS (CAMELS project).
• Integration within cosmological survey frameworks such as Euclid (ESA Euclid mission) and DESI (DESI collaboration).

---

FastPM halo catalogs (Rockstar .list)

Rockstar text catalogs produced after FastPM runs (for example out_0.list under rockstar_out_pm or rockstar_out_nbody) are read with the same RockstarCatalogReader as standalone Rockstar outputs. Do not use FastPMCatalogReader for those .list files; that reader is only for native halos stored in FastPM BigFile snapshots.

Design: reuse RockstarCatalogReader only—no duplicate parser in fastpm.py and no separate fastpm_rockstar catalog alias unless a future requirement justifies it.

CLI example:

python scripts/your_pipeline.py \
  --halo_catalog_name rockstar \
  --halo_file /path/to/fastpm_tfm/rockstar_out_pm/out_0.list

Typical folder layout on shared storage is documented in config/fastpm_folders.md (rockstar_out_pm vs rockstar_out_nbody).