Benchmark

Pipeline

bm.benchmark(adata, net[, metrics, groupby, ...])

Benchmark enrichment methods or networks on a given set of perturbation experiments.

Metrics

bm.metric.auc
bm.metric.fscore
bm.metric.qrank
bm.metric.hmean(df[, metrics, beta])	Computes the harmonic mean between two metric statistics.

Plotting

bm.pl.auc(df[, hue, palette, thr_auroc, ...])	Plot auroc and auprc.
bm.pl.fscore(df[, hue, palette])	Plot precision and recall as scatterplot.
bm.pl.qrank(df[, hue, palette, thr_rank, ...])	Plot 1-qrank and p-value.
bm.pl.bar(df, x, y[, hue, palette])	Plot the harmonic mean between two metric statistics as a barplot.
bm.pl.summary(df, y[, metrics, cmap_y, ...])	Summarizes metrics into a final score by computing the quantile-normalized rank across them.