R Integration¶
The causaliq_core.r module provides subprocess-based R session management,
data conversion utilities, and bnlearn graph utilities. It was introduced in
v0.8.0 as the shared R foundation for CausalIQ packages that interact with R.
Overview¶
R is invoked via Rscript --vanilla - with scripts piped through stdin. No
rpy2, no Rtools, and no C compiler are required — a plain CRAN R installation
is sufficient.
from causaliq_core.r import (
run_r_script,
is_r_available,
is_r_package_available,
)
if is_r_available() and is_r_package_available("bnlearn"):
output = run_r_script('cat("hello from R\\n")')
print(output) # hello from R
Submodules¶
| Submodule | Purpose |
|---|---|
| session | Rscript discovery and script execution |
| availability | R and R package availability detection |
| convert | Python ↔ R data conversion utilities |
| bnlearn | bnlearn graph utilities (CPDAG, compare, import) |
| exceptions | R-specific exception types |
Prerequisites¶
Install R from CRAN — no additional Python packages are needed:
Install bnlearn from within R for graph utilities:
See Also¶
- R Language Integration Architecture — design rationale, call flow, and test strategy