Skip to contents

Benchmark Runner

Functions for executing the full benchmark suite against fixture projects and ground-truth datasets.

run_full_benchmark()
Run the full rrlmgraph benchmark
compute_benchmark_statistics()
Compute benchmark statistics from a full results data frame
count_hallucinations()
Count hallucinations produced by an LLM in a generated code snippet
rrlmgraphbench rrlmgraphbench-package
rrlmgraphbench: Benchmark Suite for rrlmgraph