# Filename: plot.py
# pylint: disable=C0103
# pragma: no cover
"""
Common Plotting utils.
"""
from __future__ import absolute_import, print_function, division
try:
import _tkinter
except ImportError:
import matplotlib
matplotlib.use('agg')
import matplotlib as mpl # noqa
import matplotlib.pyplot as plt
import numpy as np
try:
import seaborn as sns
except ImportError:
HAS_SEABORN = False
from km3pipe.stats import bincenters
__author__ = "Moritz Lotze"
__copyright__ = "Copyright 2018, Tamas Gal and the KM3NeT collaboration."
__credits__ = []
__license__ = "BSD-3"
__maintainer__ = "Moritz Lotze"
__email__ = "mlotze@km3net.de"
__status__ = "Development"
[docs]def hexbin(x, y, color="purple", **kwargs):
"""Seaborn-compatible hexbin plot.
See also: http://seaborn.pydata.org/tutorial/axis_grids.html#mapping-custom-functions-onto-the-grid
"""
if HAS_SEABORN:
cmap = sns.light_palette(color, as_cmap=True)
else:
cmap = "Purples"
plt.hexbin(x, y, cmap=cmap, **kwargs)
[docs]def get_ax(ax=None):
"""Grab last ax if none specified, or pass through."""
if ax is None:
ax = plt.gca()
return ax
[docs]def diag(ax=None, linecolor='0.0', linestyle='--', **kwargs):
"""Plot the diagonal."""
ax = get_ax(ax)
xy_min = np.min((ax.get_xlim(), ax.get_ylim()))
xy_max = np.max((ax.get_ylim(), ax.get_xlim()))
return ax.plot([xy_min, xy_max], [xy_min, xy_max],
ls=linestyle,
c=linecolor,
**kwargs)
[docs]def automeshgrid(
x, y, step=0.02, xstep=None, ystep=None, pad=0.5, xpad=None, ypad=None
):
"""Make a meshgrid, inferred from data."""
if xpad is None:
xpad = pad
if xstep is None:
xstep = step
if ypad is None:
ypad = pad
if ystep is None:
ystep = step
xmin = x.min() - xpad
xmax = x.max() + xpad
ymin = y.min() - ypad
ymax = y.max() + ypad
return meshgrid(xmin, xmax, step, ymin, ymax, ystep)
[docs]def meshgrid(x_min, x_max, x_step, y_min=None, y_max=None, y_step=None):
"""Make a meshgrid, e.g. when plotting decision surfaces."""
if y_min is None:
y_min = x_min
if y_max is None:
y_max = x_max
if y_step is None:
y_step = x_step
xx, yy = np.meshgrid(
np.arange(x_min, x_max, x_step), np.arange(y_min, y_max, y_step)
)
return xx, yy
[docs]def prebinned_hist(counts, binlims, ax=None, *args, **kwargs):
"""Plot a histogram with counts, binlims already given.
Example
=======
>>> gaus = np.random.normal(size=100)
>>> counts, binlims = np.histogram(gaus, bins='auto')
>>> prebinned_hist(countsl binlims)
"""
ax = get_ax(ax)
x = bincenters(binlims)
weights = counts
return ax.hist(x, bins=binlims, weights=weights, *args, **kwargs)
[docs]def joint_hex(x, y, **kwargs):
"""Seaborn Joint Hexplot with marginal KDE + hists."""
return sns.jointplot(
x, y, kind='hex', stat_func=None, marginal_kws={'kde': True}, **kwargs
)
[docs]def plot_convexhull(xy, ax=None, plot_points=True):
from scipy.spatial import ConvexHull
ch = ConvexHull(xy)
ax = get_ax(ax)
if plot_points:
ax.plot(xy[:, 0], xy[:, 1], 'o')
for simplex in ch.simplices:
ax.plot(xy[simplex, 0], xy[simplex, 1], 'k-')
return ax