Skip to content
View rootcoder007's full-sized avatar
🌀
🌀

Block or report rootcoder007

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
rootcoder007/README.md
Header

Building tools that make science reproducible and accessible.

University of Toronto · causal inference · spatial statistics · reproducibility · post-quantum crypto

The MORIE ecosystem

Multi-domain Open Research and Inferential Estimation — one methodology, shipped as a family of packages. Home of the MRM (Multilevel Reconciliation Methodology) framework.

Package What it is
morie The full dual-language toolkit — Python and R, at parity — with bundled open datasets in a portable SQLite layer, a shared C/C++ core, and the complete MRM framework. The batteries-included home of the ecosystem.
rmorie The lean R-only carve for CRAN / rOpenSci — the same statistical core, 2,000+ exported functions across causal inference (DML, AIPW, IPW, matching), spatial statistics (LISA, kriging, Kulldorff scans), Hawkes processes, psychometrics/IRT, and survey-weighted inference. Dependency-light; rOpenSci statistical-software standards addressed across all nine categories.
rmoriedata Data-only companion — 40+ integrated open datasets (OTIS, CPADS, …) plus privacy primitives (differential privacy, k-anonymity).
rmorie-bricklayer Brick-proof reproducibility capsules — compiled C provenance core, CKAN resolution, SHA-256 + Wayback provenance, synthetic fallback.

Native, not thin wrappers. Both toolkits implement their statistical methods natively — no heavy third-party runtime dependencies. Every estimator is cross-validated to machine precision against the reference package it replaces, benchmarked, and CI-gated on Linux; several run markedly faster (e.g. double-ML 40–60× over DoubleML, optimal matching 7–14× over optmatch).

PyPI rmorie on r-universe License: AGPL v3

pip install morie                                        # Python
Rscript -e 'install.packages("rmorie",
  repos = "https://rootcoder007.r-universe.dev")'        # R
brew tap rootcoder007/morie && brew install morie        # macOS/Linux CLI

Every result-emitting function returns a RichResult — a self-describing report (summary, tables, warnings, interpretation) that round-trips to JSON. One design, (data, treatment, outcome, covariates), flows through any estimator; the MRM framework runs ~10 of them on one design and reconciles the answers.

Documentation r-universe

Python R C C++ Linux

Coding setup animation

@rootcoder007's activity is private