make_row(experiment_group="correctness", dataset="validation_k4", variant="B0", declared_classes=4, color_limit=4), make_row(experiment_group="correctness", dataset ...
Standard improvement ratio vs. each comparator. relative_oracle (dast - comp) / (oracle - comp) Fraction of oracle gap captured by DAST over each comparator. Requires oracle_profits_impl in pkl.
Some results have been hidden because they may be inaccessible to you
Show inaccessible results