def compare(results):
"""Plots run time comparison of all results, one subplot per stencil.
Args:
results: List of `result.Result` objects.
"""
stencils, stenciltimes = result.times_by_stencil(results)
cols = math.ceil(math.sqrt(len(stencils)) / (0.5 * len(results)))
rows = math.ceil(len(stencils) / cols)
fig, axarr = plt.subplots(rows,
cols,
squeeze=False,
figsize=figsize(cols * len(results) / 2, rows))
axes = itertools.chain(*axarr)
colors = discrete_colors(len(results))
suptitle, titles = get_titles(results)
fig.suptitle(suptitle,
wrap=True,
y=1 - 0.1 / rows,
verticalalignment='center')
xticks = list(range(1, len(results) + 1))
for ax, stencil, times in itertools.zip_longest(axes, stencils,
stenciltimes):
if stencil:
ax.set_title(stencil.title())
medians = [statistics.median(t) for t in times]
ax.bar(xticks, medians, width=0.9, color=colors)
ax.boxplot(times, widths=0.9, medianprops={'color': 'black'})
ax.set_xticklabels(titles, wrap=True)
else:
ax.set_visible(False)
fig.tight_layout(rect=[0, 0, 1, 1 - 0.2 / rows])
return fig
def history(results, key='job', limit=None):
"""Plots run time history of all results. Depending on the argument `job`,
The results are either ordered by commit/build time or job time
(i.e. when the job was run).
Args:
results: List of `result.Result` objects.
key: Either 'job' or 'build'.
limit: Optionally limits the number of plotted results to the given
number, i.e. only displays the most recent results. If `None`,
all given results are plotted.
"""
# get date/time either from the commit/build or job (when job was run)
def get_datetime(result):
if key == 'build':
datetime = result.runinfo.datetime
elif key == 'job':
datetime = result.datetime
else:
raise ValueError('"key" argument must be "build" or "job"')
return time.local_time(datetime)
# sort results by desired reference time
results = sorted(results, key=get_datetime)
#
if limit is not None:
if not isinstance(limit, int) or limit <= 0:
raise ValueError('"limit" must be a positive integer')
results = results[-limit:]
with warnings.catch_warnings():
warnings.filterwarnings('ignore', category=RuntimeWarning)
percentiles = [result.times_by_stencil(results,
missing=[np.nan],
func=functools.partial(
np.percentile, q=q))
for q in (0, 25, 50, 75, 100)]
stencils = percentiles[0].keys()
percentiles = {stencil: [p[stencil] for p in percentiles]
for stencil in stencils}
dates = [matplotlib.dates.date2num(get_datetime(r)) for r in results]
if len(dates) > len(set(dates)):
log.warning('Non-unique datetimes in history plot')
fig, ax = plt.subplots(figsize=figsize(2, 1))
locator = matplotlib.dates.AutoDateLocator()
formatter = matplotlib.dates.AutoDateFormatter(locator)
formatter.scaled[1 / 24] = '%y-%m-%d %H:%M'
formatter.scaled[1 / (24 * 60)] = '%y-%m-%d %H:%M'
formatter.scaled[1 / (24 * 60 * 60)] = '%y-%m-%d %H:%M:%S'
ax.xaxis.set_major_locator(locator)
ax.xaxis.set_major_formatter(formatter)
colors = discrete_colors(len(stencils))
for color, (stencil, qs) in zip(colors, sorted(percentiles.items())):
mint, q1, q2, q3, maxt = qs
ax.fill_between(dates, mint, maxt, alpha=0.2, color=color)
ax.fill_between(dates, q1, q3, alpha=0.5, color=color)
ax.plot(dates, q2, '|-', label=stencil.title(), color=color)
ax.legend(loc='upper left')
fig.autofmt_xdate()
fig.tight_layout()
return fig