dagzoo 0.14.4

Synthetic tabular data generator for causal modeling

dagzoo

dagzoo generates reproducible synthetic tabular corpora from sampled causal structure. The stable adoption layer is a small set of named recipe packs plus stable artifact contracts; repo-internal authoring under configs/ remains available for advanced work, but it is not the primary public entrypoint.

Start

Use the packaged CLI when you want the public workflow without a repo checkout:

uv tool install dagzoo
dagzoo recipe list
dagzoo generate --config recipe:default-baseline --num-datasets 25 --out data/default_baseline
dagzoo generate --config recipe:tabpfn-v1-prior-approx --num-datasets 25 --out data/tabpfn_prior
dagzoo filter --in data/default_baseline --out data/default_baseline_filter

Use a repo checkout when you want to edit configs, run docs tooling, or work on the codebase:

./scripts/dev bootstrap
source .venv/bin/activate
./scripts/dev verify quick

For in-process training loops, use the same recipe references through the PyTorch bridge:

from dagzoo import build_dataloader

loader = build_dataloader(
    "recipe:default-baseline",
    num_datasets=10,
    seed=7,
    device="cpu",
)
sample = next(iter(loader))
print(sample["X_train"].shape)

Public Surface

• dagzoo recipe list shows the curated public catalog.
• dagzoo generate --config recipe:<name> is the primary reproducible CLI path.
• build_dataloader("recipe:<name>", ...) is the programmatic equivalent.
• recipes/*.yaml are the published recipe sources behind those stable names.
• configs/*.yaml remain useful for advanced/internal authoring, but they move faster than the named recipe surface.

Docs

• Start
• Reference Packs
• Advanced Controls
• Artifacts & API
• How It Works
• Feature Guides
• Roadmap

Community

• CITATION.cff
• Contributing
• Security