def __plot_variance(self):
pointsMin = self.__calc_min()
pointsMax = self.__calc_max()
polys = []
variance = []
varMin = 1000
varMax = 0
lastX = None
lastYMin = None
lastYMax = None
for x in pointsMin.iterkeys():
if lastX is None:
lastX = x
if lastYMin is None:
lastYMin = pointsMin[x]
if lastYMax is None:
lastYMax = pointsMax[x]
polys.append([[x, pointsMin[x]],
[x, pointsMax[x]],
[lastX, lastYMax],
[lastX, lastYMin],
[x, pointsMin[x]]])
lastX = x
lastYMin = pointsMin[x]
lastYMax = pointsMax[x]
var = pointsMax[x] - pointsMin[x]
variance.append(var)
varMin = min(varMin, var)
varMax = max(varMax, var)
norm = Normalize(vmin=varMin, vmax=varMax)
sm = ScalarMappable(norm, self.colourMap)
colours = sm.to_rgba(variance)
pc = PolyCollection(polys)
pc.set_gid('plot')
pc.set_norm(norm)
pc.set_color(colours)
self.axes.add_collection(pc)
return None, None
def __plot_variance(self):
pointsMin = self.__calc_min()
pointsMax = self.__calc_max()
polys = []
variance = []
varMin = 1000
varMax = 0
lastX = None
lastYMin = None
lastYMax = None
for x in pointsMin.iterkeys():
if lastX is None:
lastX = x
if lastYMin is None:
lastYMin = pointsMin[x]
if lastYMax is None:
lastYMax = pointsMax[x]
polys.append([[x, pointsMin[x]],
[x, pointsMax[x]],
[lastX, lastYMax],
[lastX, lastYMin],
[x, pointsMin[x]]])
lastX = x
lastYMin = pointsMin[x]
lastYMax = pointsMax[x]
var = pointsMax[x] - pointsMin[x]
variance.append(var)
varMin = min(varMin, var)
varMax = max(varMax, var)
norm = Normalize(vmin=varMin, vmax=varMax)
sm = ScalarMappable(norm, self.colourMap)
colours = sm.to_rgba(variance)
pc = PolyCollection(polys)
pc.set_gid('plot')
pc.set_norm(norm)
pc.set_color(colours)
self.axes.add_collection(pc)
return None, None
def shapes(shapes,
data=None,
colorbar=False,
colorbar_ticklabels=None,
norm=None,
with_labels=False,
edgecolors=None,
facecolors=None,
fontsize=None,
ax=None,
**kwds):
"""
Plot `data` on the basis of a dictionary of shapes. `data`
must be given as a pandas Series with the corresponding keys
of shapes as index.
Parameters
----------
shapes : dict | pd.Series
Dictionary of shapes
data : pd.Series
Float valued data to be plotted. If data is omitted,
np.arange(N) will be used.
with_labels : bool
Whether to plot the name of each shape at its centroid
Returns
-------
collection : PolyCollection
"""
if 'outline' in kwds:
# deprecated
if kwds.pop('outline'): facecolors = 'none'
if 'colour' in kwds:
# deprecated
edgecolors = kwds.pop('colour')
if ax is None:
ax = plt.gca()
if not isinstance(shapes, pd.Series):
shapes = pd.Series(shapes)
def flatten_multipolygons(shapes):
flat_shapes = []
flat_index = []
for n, sh in shapes.iteritems():
if isinstance(sh, MultiPolygon):
flat_shapes += list(sh)
flat_index += [n] * len(sh)
else:
flat_shapes.append(sh)
flat_index.append(n)
return pd.Series(flat_shapes, index=flat_index)
if isinstance(edgecolors, pd.Series):
shapes = shapes.reindex(edgecolors.index)
flat_shapes = flatten_multipolygons(shapes)
edgecolors = edgecolors.reindex(flat_shapes.index)
elif isinstance(facecolors, pd.Series):
shapes = shapes.reindex(facecolors.index)
flat_shapes = flatten_multipolygons(shapes)
facecolors = facecolors.reindex(flat_shapes.index)
elif isinstance(data, pd.Series):
shapes = shapes.reindex(data.index)
flat_shapes = flatten_multipolygons(shapes)
data = data.reindex(flat_shapes.index)
else:
flat_shapes = flatten_multipolygons(shapes)
if facecolors is None:
data = pd.Series(np.arange(len(shapes)),
index=shapes.index).reindex(flat_shapes.index)
coll = PolyCollection((np.asarray(x.exterior) for x in flat_shapes),
transOffset=ax.transData,
facecolors=facecolors,
edgecolors=edgecolors,
**kwds)
if data is not None:
coll.set_array(data)
if norm is not None:
coll.set_norm(norm)
ax.add_collection(coll, autolim=True)
if colorbar:
kwargs = dict()
if isinstance(colorbar, dict):
kwargs.update(colorbar)
## FIXME : sounds like a bug to me, but hey
if norm is not None:
norm.autoscale(data)
cbar = plt.colorbar(mappable=coll, **kwargs)
if colorbar_ticklabels is not None:
cbar.ax.set_yticklabels(colorbar_ticklabels)
if with_labels:
for k, v in iteritems(shapes.reindex(data.index)):
x, y = np.asarray(v.centroid)
plt.text(x,
y,
k,
fontsize=fontsize,
horizontalalignment='center',
verticalalignment='center')
ax.autoscale_view()
return coll
def tripcolor(ax, *args, **kwargs):
"""
Create a pseudocolor plot of an unstructured triangular grid to
the :class:`~matplotlib.axes.Axes`.
The triangulation can be specified in one of two ways; either::
tripcolor(triangulation, ...)
where triangulation is a :class:`~matplotlib.tri.Triangulation`
object, or
::
tripcolor(x, y, ...)
tripcolor(x, y, triangles, ...)
tripcolor(x, y, triangles=triangles, ...)
tripcolor(x, y, mask=mask, ...)
tripcolor(x, y, triangles, mask=mask, ...)
in which case a Triangulation object will be created. See
:class:`~matplotlib.tri.Triangulation` for a explanation of these
possibilities.
The next argument must be *C*, the array of color values, one per
point in the triangulation. The colors used for each triangle
are from the mean C of the triangle's three points.
The remaining kwargs are the same as for
:meth:`~matplotlib.axes.Axes.pcolor`.
**Example:**
.. plot:: mpl_examples/pylab_examples/tripcolor_demo.py
"""
if not ax._hold: ax.cla()
alpha = kwargs.pop('alpha', 1.0)
norm = kwargs.pop('norm', None)
cmap = kwargs.pop('cmap', None)
vmin = kwargs.pop('vmin', None)
vmax = kwargs.pop('vmax', None)
shading = kwargs.pop('shading', 'flat')
tri, args, kwargs = Triangulation.get_from_args_and_kwargs(*args, **kwargs)
x = tri.x
y = tri.y
triangles = tri.get_masked_triangles()
# Vertices of triangles.
verts = np.concatenate((x[triangles][...,np.newaxis],
y[triangles][...,np.newaxis]), axis=2)
C = np.asarray(args[0])
if C.shape != x.shape:
raise ValueError('C array must have same length as triangulation x and'
' y arrays')
# Color values, one per triangle, mean of the 3 vertex color values.
C = C[triangles].mean(axis=1)
if shading == 'faceted':
edgecolors = (0,0,0,1),
linewidths = (0.25,)
else:
edgecolors = 'face'
linewidths = (1.0,)
kwargs.setdefault('edgecolors', edgecolors)
kwargs.setdefault('antialiaseds', (0,))
kwargs.setdefault('linewidths', linewidths)
collection = PolyCollection(verts, **kwargs)
collection.set_alpha(alpha)
collection.set_array(C)
if norm is not None: assert(isinstance(norm, Normalize))
collection.set_cmap(cmap)
collection.set_norm(norm)
if vmin is not None or vmax is not None:
collection.set_clim(vmin, vmax)
else:
collection.autoscale_None()
ax.grid(False)
minx = tri.x.min()
maxx = tri.x.max()
miny = tri.y.min()
maxy = tri.y.max()
corners = (minx, miny), (maxx, maxy)
ax.update_datalim( corners)
ax.autoscale_view()
ax.add_collection(collection)
return collection
def tripcolor(ax, *args, **kwargs):
"""
Create a pseudocolor plot of an unstructured triangular grid to
the :class:`~matplotlib.axes.Axes`.
The triangulation can be specified in one of two ways; either::
tripcolor(triangulation, ...)
where triangulation is a :class:`~matplotlib.tri.Triangulation`
object, or
::
tripcolor(x, y, ...)
tripcolor(x, y, triangles, ...)
tripcolor(x, y, triangles=triangles, ...)
tripcolor(x, y, mask, ...)
tripcolor(x, y, mask=mask, ...)
tripcolor(x, y, triangles, mask, ...)
tripcolor(x, y, triangles, mask=mask, ...)
in which case a Triangulation object will be created. See
:class:`~matplotlib.tri.Triangulation` for a explanation of these
possibilities.
The next argument must be *C*, the array of color values, one per
point in the triangulation. The colors used for each triangle
are from the mean C of the triangle's three points.
The remaining kwargs are the same as for
:meth:`~matplotlib.axes.Axes.pcolor`.
**Example:**
.. plot:: mpl_examples/pylab_examples/tripcolor_demo.py
"""
if not ax._hold: ax.cla()
alpha = kwargs.pop('alpha', 1.0)
norm = kwargs.pop('norm', None)
cmap = kwargs.pop('cmap', None)
vmin = kwargs.pop('vmin', None)
vmax = kwargs.pop('vmax', None)
shading = kwargs.pop('shading', 'flat')
tri, args, kwargs = Triangulation.get_from_args_and_kwargs(*args, **kwargs)
x = tri.x
y = tri.y
triangles = tri.get_masked_triangles()
# Vertices of triangles.
verts = np.concatenate(
(x[triangles][..., np.newaxis], y[triangles][..., np.newaxis]), axis=2)
C = np.asarray(args[0])
if C.shape != x.shape:
raise ValueError('C array must have same length as triangulation x and'
' y arrays')
# Color values, one per triangle, mean of the 3 vertex color values.
C = C[triangles].mean(axis=1)
if shading == 'faceted':
edgecolors = (0, 0, 0, 1),
linewidths = (0.25, )
else:
edgecolors = 'face'
linewidths = (1.0, )
kwargs.setdefault('edgecolors', edgecolors)
kwargs.setdefault('antialiaseds', (0, ))
kwargs.setdefault('linewidths', linewidths)
collection = PolyCollection(verts, **kwargs)
collection.set_alpha(alpha)
collection.set_array(C)
if norm is not None: assert (isinstance(norm, Normalize))
collection.set_cmap(cmap)
collection.set_norm(norm)
if vmin is not None or vmax is not None:
collection.set_clim(vmin, vmax)
else:
collection.autoscale_None()
ax.grid(False)
minx = tri.x.min()
maxx = tri.x.max()
miny = tri.y.min()
maxy = tri.y.max()
corners = (minx, miny), (maxx, maxy)
ax.update_datalim(corners)
ax.autoscale_view()
ax.add_collection(collection)
return collection
