Skip to content

Should np.median be replaced with np.mean for plotting 'Mean code coverage growth over time'? #2073

Description

@leonasdev

I noticed that when plotting the 'Mean code coverage growth over time' graph, np.median is used as the estimator (L166):

axes = sns.lineplot(
y=column_of_interest,
x='time',
hue='fuzzer',
hue_order=fuzzer_order,
data=benchmark_df[benchmark_df.time <= snapshot_time],
ci=None if bugs or self._quick else 95,
estimator=np.median,
palette=self._fuzzer_colors,
style='fuzzer',
dashes=False,
markers=self._fuzzer_markers,
ax=axes)

Shouldn't np.mean be used instead? For example, the order of this graph uses benchmark_rank_by_mean:

fuzzer_order = data_utils.benchmark_rank_by_mean(

Is there a specific reason for using np.median here, or am I missing something?

Looking forward to your insights. Thanks.

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions