⭐ What's new in 0.17.1
0.17.1 makes Andrey easier to enter and states its standards. The homepage shows one method per
family, and a Methods page lists every method by the kind of data it reads. A standards page says
what we require of every method and how we check it, with new tests that rescale each column and
require the same graph. The docs gain a concepts guide and a learning path, and
score_func="normal_scores", experimental, keeps a search's graph under increasing transforms of a
column. 0.17.0 added SCORE, NoGAM, and AGES.
Methods ·
Standards ·
0.17.0 release card
Andrey is a causal discovery package for Python.
- Fast: The fastest causal discovery package.
- Accurate: more than 80% fewer errors than other popular packages.
- One API: 49 supported methods, each one function call.
- Any data: continuous, categorical, mixed, time series, nonlinear, hidden confounders.
- Agent-ready: a command line that answers in JSON, and a skill for coding agents.
uv add andrey-core
pip install andrey-core
You install andrey-core and then import andrey.
- Optional extras:
andrey-core[torch] for GPUs and CALM, andrey-core[viz] for charts and
graph layouts
- From GitHub, the latest release:
uv add "andrey-core @ git+https://github.com/Abel-ai-lab/andrey@main"
- Installation guide: the
extras, a PyTorch build for your GPU, and building Andrey from source
What's inside
| Family | Methods | Returns |
|---|
| Constraint-based | pc, fci, gfci, cdnod, fci_jci, icp, iamb, inter_iamb, mmpc | CPDAG, PAG, DAG, Undirected graph |
| Score-based | ges, gies, gnies, xges, ages, hc, tabu, exact_search, calm, notears, dagma*, mmhc, |
* Experimental:
not supported yet (see its Limitations), and the API may change.
Quick start
Learn the protein-signaling network of Sachs et al. (2005) from 853 cells, and score it against the
known network:
import andrey
from andrey.data import load_dataset
from andrey.metrics import score
sachs = load_dataset("sachs") # 853 cells, 11 proteins, and the known network
out = andrey.pc(sachs.data) # learn a graph
print(out)
scores = score(out.structure, sachs.graph)
print({k: round(scores[k], 2) for k in ("shd", "skeleton_precision", "skeleton_recall")})
This is the beginning of the README. Read the rest at the original. Read the full README ↗
