plot

A module to plot the results of the benchmark

---

source

create_radar_plot

def create_radar_plot(
    benchmark_results:Sequence[Mapping[str, Any]], # sequence of benchmark result dicts
    model_names:Sequence[str] | None=None, # optional list of model names for the legend
    reference_max:Mapping[str, float] | None=None, # optional dict mapping emoji to max value
    specs:dict[str, tuple] | None=None, # custom specs override (defaults to SPECS)
)->go.Figure:

Create a radar plot comparing benchmark results across models.