arboreto

1# Arboreto
2 
3## When to use
4 
5Use Arboreto to rank candidate regulator-target associations from expression
6measurements. GRNBoost2 fits stochastic gradient boosting regressions; GENIE3
7fits random forests. Each target is predicted from candidate regulators, excluding
8itself. These are observational predictive associations, not proof of direct
9binding, activation/repression, or causal regulation.
10 
11The latest PyPI release checked is **0.1.6** (2021-02-09). Read the Docs still
12labels its documentation 0.1.5; the current GitHub source contains fixes that are
13**not in the PyPI wheel**. Do not assume a successful unpinned installation can
14run inference. See [compatibility and distribution details](references/distributed_computing.md).
15 
16## Installation and compatibility
17 
18The following exact stack passed dense and CSC-sparse GRNBoost2, dense GENIE3,
19custom GBM/RF, and wrapper smoke tests on macOS arm64 with Python 3.11.11:
20 
21```bash
22uv venv --python 3.11 .venv-arboreto
23uv pip install --python .venv-arboreto/bin/python \
24  'arboreto==0.1.6' 'dask[complete]==2024.7.1' 'distributed==2024.7.1' \
25  'numpy==1.26.4' 'pandas==2.2.3' 'scikit-learn==1.5.2' 'scipy==1.13.1'
26```
27 
28This is a bounded compatibility recipe, not a claim that current releases of all
29dependencies work. PyPI Arboreto builds an empty metadata graph that the newer
30Dask dataframe implementation rejects. For this pinned Dask version, select its
31legacy dataframe backend **before importing Arboreto or `dask.dataframe`**:
32…