Repository for different network models related to flow/disparity (ECCV 18)
-
Updated
Jul 17, 2019 - Python
Repository for different network models related to flow/disparity (ECCV 18)
Hybrid (non)-Gaussian solver for Distributed Factor Graphs, with clique recycling; and is officially supported as part of the NavAbility(TM) toolchain and Caesar.jl ecosystem.
Claude Code plugin & MCP server for Phase 2/3 clinical trial design. 9 tools: single-primary (binary/continuous/PH+NPH survival), multi-hypothesis (co-primary, multi-population, Maurer-Bretz). Backed by gsDesign, gsDesign2, graphicalMCP. Word/PDF reporting, Monte-Carlo verification, 176-case benchmark corpus.
A deterministic, zero-dependency, local-first text structure and plagiarism screening engine. Built to evaluate text without relying on probabilistic AI models.
Model-agnostic Agent Skill: keep multiple hypotheses alive, test them against evidence, revive alternatives, collapse to the best-supported answer. Ships the skill contract, a reference branch controller and a benchmark harness. No model evaluation published yet, so no performance claim.
To associate your repository with the multi-hypothesis topic, visit your repo's landing page and select "manage topics."