def map_to_colors(buff, cmap_name):
try:
lut = cmd.color_map_luts[cmap_name]
except KeyError as e:
try:
# if cmap is tuple, then we're using palettable or brewer2mpl cmaps
if isinstance(cmap_name, tuple):
cmap = get_brewer_cmap(cmap_name)
else:
cmap = mcm.get_cmap(cmap_name)
cmap(0.0)
lut = cmap._lut.T
except ValueError:
raise KeyError(
"Your color map (%s) was not found in either the extracted"
" colormap file or matplotlib colormaps" % cmap_name) from e
if isinstance(cmap_name, tuple):
# If we are using the colorbrewer maps, don't interpolate
shape = buff.shape
# We add float_eps so that digitize doesn't go out of bounds
x = np.mgrid[0.0:1.0 + np.finfo(np.float32).eps:lut[0].shape[0] * 1j]
inds = np.digitize(buff.ravel(), x)
inds.shape = (shape[0], shape[1])
mapped = np.dstack([(v[inds] * 255).astype("uint8") for v in lut])
del inds
else:
x = np.mgrid[0.0:1.0:lut[0].shape[0] * 1j]
mapped = np.dstack([(np.interp(buff, x, v) * 255).astype("uint8")
for v in lut])
return mapped.copy("C")
def get_colormap_lut(cmap_id: Union[Tuple[str, str], str]):
# "lut" stands for "lookup table". This function provides a consistent and
# reusable accessor to a hidden (and by defaut, uninitialized) attribute
# (`_lut`) in registered colormaps, from matplotlib or palettable.
# colormap "lookup tables" are RGBA arrays in matplotlib,
# and contain sufficient data to reconstruct the colormaps entierly.
# This exists mostly for historical reasons, hence the custom output format.
# It isn't meant as part of yt's public api.
if isinstance(cmap_id, tuple) and len(cmap_id) == 2:
cmap = get_brewer_cmap(cmap_id)
elif isinstance(cmap_id, str):
cmap = get_cmap(cmap_id)
else:
raise TypeError(
"Expected a string or a 2-tuple of strings as a colormap id. "
f"Received: {cmap_id}"
)
if not cmap._isinit:
cmap._init()
r = cmap._lut[:-3, 0]
g = cmap._lut[:-3, 1]
b = cmap._lut[:-3, 2]
a = np.ones(b.shape)
return [r, g, b, a]
def _init_image(self, data, cbnorm, cblinthresh, cmap, extent, aspect):
"""Store output of imshow in image variable"""
if (cbnorm == 'log10'):
norm = matplotlib.colors.LogNorm()
elif (cbnorm == 'linear'):
norm = matplotlib.colors.Normalize()
elif (cbnorm == 'symlog'):
if cblinthresh is None:
cblinthresh = float((np.nanmax(data) - np.nanmin(data)) / 10.)
norm = matplotlib.colors.SymLogNorm(cblinthresh,
vmin=float(np.nanmin(data)),
vmax=float(np.nanmax(data)))
extent = [float(e) for e in extent]
# tuple colormaps are from palettable (or brewer2mpl)
if isinstance(cmap, tuple):
cmap = get_brewer_cmap(cmap)
vmin = float(self.zmin) if self.zmax is not None else None
vmax = float(self.zmax) if self.zmax is not None else None
self.image = self.axes.imshow(data.to_ndarray(),
origin='lower',
extent=extent,
norm=norm,
vmin=vmin,
vmax=vmax,
aspect=aspect,
cmap=cmap,
interpolation='nearest')
if (cbnorm == 'symlog'):
if LooseVersion(matplotlib.__version__) < LooseVersion("2.0.0"):
formatter_kwargs = {}
else:
formatter_kwargs = dict(linthresh=cblinthresh)
formatter = matplotlib.ticker.LogFormatterMathtext(
**formatter_kwargs)
self.cb = self.figure.colorbar(self.image,
self.cax,
format=formatter)
if np.nanmin(data) >= 0.0:
yticks = [np.nanmin(data).v] + list(10**np.arange(
np.rint(np.log10(cblinthresh)),
np.ceil(np.log10(np.nanmax(data))) + 1))
elif np.nanmax(data) <= 0.0:
yticks = list(
-10**np.arange(
np.floor(np.log10(-np.nanmin(data))),
np.rint(np.log10(cblinthresh)) - 1, -1)) + \
[np.nanmax(data).v]
else:
yticks = list(
-10**np.arange(np.floor(np.log10(-np.nanmin(data))),
np.rint(np.log10(cblinthresh))-1, -1)) + \
[0] + \
list(10**np.arange(np.rint(np.log10(cblinthresh)),
np.ceil(np.log10(np.nanmax(data)))+1))
self.cb.set_ticks(yticks)
else:
self.cb = self.figure.colorbar(self.image, self.cax)
for which in ['major', 'minor']:
self.cax.tick_params(which=which, axis='y', direction='in')
def _init_image(self, data, cbnorm, cblinthresh, cmap, extent, aspect):
"""Store output of imshow in image variable"""
cbnorm_kwargs = dict(
vmin=float(self.zmin) if self.zmin is not None else None,
vmax=float(self.zmax) if self.zmax is not None else None,
)
if cbnorm == "log10":
cbnorm_cls = matplotlib.colors.LogNorm
elif cbnorm == "linear":
cbnorm_cls = matplotlib.colors.Normalize
elif cbnorm == "symlog":
if cblinthresh is None:
cblinthresh = float((np.nanmax(data) - np.nanmin(data)) / 10.0)
cbnorm_kwargs.update(
dict(
linthresh=cblinthresh,
vmin=float(np.nanmin(data)),
vmax=float(np.nanmax(data)),
))
MPL_VERSION = LooseVersion(matplotlib.__version__)
if MPL_VERSION >= "3.2.0":
# note that this creates an inconsistency between mpl versions
# since the default value previous to mpl 3.4.0 is np.e
# but it is only exposed since 3.2.0
cbnorm_kwargs["base"] = 10
cbnorm_cls = matplotlib.colors.SymLogNorm
else:
raise ValueError(f"Unknown value `cbnorm` == {cbnorm}")
norm = cbnorm_cls(**cbnorm_kwargs)
extent = [float(e) for e in extent]
# tuple colormaps are from palettable (or brewer2mpl)
if isinstance(cmap, tuple):
cmap = get_brewer_cmap(cmap)
if self._transform is None:
# sets the transform to be an ax.TransData object, where the
# coordiante system of the data is controlled by the xlim and ylim
# of the data.
transform = self.axes.transData
else:
transform = self._transform
if hasattr(self.axes, "set_extent"):
# CartoPy hangs if we do not set_extent before imshow if we are
# displaying a small subset of the globe. What I believe happens is
# that the transform for the points on the outside results in
# infinities, and then the scipy.spatial cKDTree hangs trying to
# identify nearest points.
#
# Also, set_extent is defined by cartopy, so not all axes will have
# it as a method.
#
# A potential downside is that other images may change, but I believe
# the result of imshow is to set_extent *regardless*. This just
# changes the order in which it happens.
#
# NOTE: This is currently commented out because it breaks in some
# instances. It is left as a historical note because we will
# eventually need some form of it.
# self.axes.set_extent(extent)
pass
self.image = self.axes.imshow(
data.to_ndarray(),
origin="lower",
extent=extent,
norm=norm,
aspect=aspect,
cmap=cmap,
interpolation="nearest",
transform=transform,
)
if cbnorm == "symlog":
if LooseVersion(matplotlib.__version__) < LooseVersion("2.0.0"):
formatter_kwargs = {}
else:
formatter_kwargs = dict(linthresh=cblinthresh)
formatter = matplotlib.ticker.LogFormatterMathtext(
**formatter_kwargs)
self.cb = self.figure.colorbar(self.image,
self.cax,
format=formatter)
if np.nanmin(data) >= 0.0:
yticks = [np.nanmin(data).v] + list(10**np.arange(
np.rint(np.log10(cblinthresh)),
np.ceil(np.log10(np.nanmax(data))) + 1,
))
elif np.nanmax(data) <= 0.0:
yticks = list(-(10**np.arange(
np.floor(np.log10(-np.nanmin(data))),
np.rint(np.log10(cblinthresh)) - 1,
-1,
))) + [np.nanmax(data).v]
else:
yticks = (list(-(10**np.arange(
np.floor(np.log10(-np.nanmin(data))),
np.rint(np.log10(cblinthresh)) - 1,
-1,
))) + [0] + list(10**np.arange(
np.rint(np.log10(cblinthresh)),
np.ceil(np.log10(np.nanmax(data))) + 1,
)))
self.cb.set_ticks(yticks)
else:
self.cb = self.figure.colorbar(self.image, self.cax)
for which in ["major", "minor"]:
self.cax.tick_params(which=which, axis="y", direction="in")
def _init_image(self, data, cbnorm, cblinthresh, cmap, extent, aspect):
"""Store output of imshow in image variable"""
cbnorm_kwargs = dict(
vmin=float(self.zmin) if self.zmin is not None else None,
vmax=float(self.zmax) if self.zmax is not None else None,
)
zmin = float(self.zmin) if self.zmin is not None else np.nanmin(data)
zmax = float(self.zmax) if self.zmax is not None else np.nanmax(data)
if cbnorm == "symlog":
# if cblinthresh is not specified, try to come up with a reasonable default
min_abs_val = np.min(np.abs((zmin, zmax)))
if cblinthresh is None:
cblinthresh = np.nanmin(np.absolute(data)[data != 0])
elif zmin * zmax > 0 and cblinthresh < min_abs_val:
warnings.warn(
f"Cannot set a symlog norm with linear threshold {cblinthresh} "
f"lower than the minimal absolute data value {min_abs_val} . "
"Switching to log norm.")
cbnorm = "log10"
if cbnorm == "log10":
cbnorm_cls = matplotlib.colors.LogNorm
elif cbnorm == "linear":
cbnorm_cls = matplotlib.colors.Normalize
elif cbnorm == "symlog":
cbnorm_kwargs.update(dict(linthresh=cblinthresh))
if MPL_VERSION >= Version("3.2.0"):
# note that this creates an inconsistency between mpl versions
# since the default value previous to mpl 3.4.0 is np.e
# but it is only exposed since 3.2.0
cbnorm_kwargs["base"] = 10
cbnorm_cls = matplotlib.colors.SymLogNorm
else:
raise ValueError(f"Unknown value `cbnorm` == {cbnorm}")
norm = cbnorm_cls(**cbnorm_kwargs)
extent = [float(e) for e in extent]
# tuple colormaps are from palettable (or brewer2mpl)
if isinstance(cmap, tuple):
cmap = get_brewer_cmap(cmap)
if self._transform is None:
# sets the transform to be an ax.TransData object, where the
# coordinate system of the data is controlled by the xlim and ylim
# of the data.
transform = self.axes.transData
else:
transform = self._transform
if hasattr(self.axes, "set_extent"):
# CartoPy hangs if we do not set_extent before imshow if we are
# displaying a small subset of the globe. What I believe happens is
# that the transform for the points on the outside results in
# infinities, and then the scipy.spatial cKDTree hangs trying to
# identify nearest points.
#
# Also, set_extent is defined by cartopy, so not all axes will have
# it as a method.
#
# A potential downside is that other images may change, but I believe
# the result of imshow is to set_extent *regardless*. This just
# changes the order in which it happens.
#
# NOTE: This is currently commented out because it breaks in some
# instances. It is left as a historical note because we will
# eventually need some form of it.
# self.axes.set_extent(extent)
# possibly related issue (operation order dependency)
# https://github.com/SciTools/cartopy/issues/1468
# in cartopy 0.19 (or 0.20), some intented behaviour changes produced an
# incompatibility here where default values for extent would lead to a crash.
# A solution is to let the transform object set the image extents internally
# see https://github.com/SciTools/cartopy/issues/1955
extent = None
self.image = self.axes.imshow(
data.to_ndarray(),
origin="lower",
extent=extent,
norm=norm,
aspect=aspect,
cmap=cmap,
interpolation="nearest",
transform=transform,
)
if cbnorm == "symlog":
formatter = matplotlib.ticker.LogFormatterMathtext(
linthresh=cblinthresh)
self.cb = self.figure.colorbar(self.image,
self.cax,
format=formatter)
if zmin >= 0.0:
yticks = [zmin] + list(10**np.arange(
np.rint(np.log10(cblinthresh)),
np.ceil(np.log10(zmax)),
))
elif zmax <= 0.0:
if MPL_VERSION >= Version("3.5.0b"):
offset = 0
else:
offset = 1
yticks = (list(-(10**np.arange(
np.floor(np.log10(-zmin)),
np.rint(np.log10(cblinthresh)) - offset,
-1,
))) + [zmax])
else:
yticks = (list(-(10**np.arange(
np.floor(np.log10(-zmin)),
np.rint(np.log10(cblinthresh)) - 1,
-1,
))) + [0] + list(10**np.arange(
np.rint(np.log10(cblinthresh)),
np.ceil(np.log10(zmax)),
)))
self.cb.set_ticks(yticks)
else:
self.cb = self.figure.colorbar(self.image, self.cax)
self.cax.tick_params(which="both", axis="y", direction="in")
def _init_image(self, data, cbnorm, cblinthresh, cmap, extent, aspect):
"""Store output of imshow in image variable"""
if (cbnorm == 'log10'):
norm = matplotlib.colors.LogNorm()
elif (cbnorm == 'linear'):
norm = matplotlib.colors.Normalize()
elif (cbnorm == 'symlog'):
if cblinthresh is None:
cblinthresh = float((np.nanmax(data)-np.nanmin(data))/10.)
norm = matplotlib.colors.SymLogNorm(
cblinthresh, vmin=float(np.nanmin(data)),
vmax=float(np.nanmax(data)))
extent = [float(e) for e in extent]
# tuple colormaps are from palettable (or brewer2mpl)
if isinstance(cmap, tuple):
cmap = get_brewer_cmap(cmap)
vmin = float(self.zmin) if self.zmax is not None else None
vmax = float(self.zmax) if self.zmax is not None else None
if self._transform is None:
# sets the transform to be an ax.TransData object, where the
# coordiante system of the data is controlled by the xlim and ylim
# of the data.
transform = self.axes.transData
else:
transform = self._transform
if hasattr(self.axes, "set_extent"):
# CartoPy hangs if we do not set_extent before imshow if we are
# displaying a small subset of the globe. What I believe happens is
# that the transform for the points on the outside results in
# infinities, and then the scipy.spatial cKDTree hangs trying to
# identify nearest points.
#
# Also, set_extent is defined by cartopy, so not all axes will have
# it as a method.
#
# A potential downside is that other images may change, but I believe
# the result of imshow is to set_extent *regardless*. This just
# changes the order in which it happens.
#
# NOTE: This is currently commented out because it breaks in some
# instances. It is left as a historical note because we will
# eventually need some form of it.
# self.axes.set_extent(extent)
pass
self.image = self.axes.imshow(
data.to_ndarray(), origin='lower', extent=extent, norm=norm,
vmin=vmin, vmax=vmax, aspect=aspect, cmap=cmap,
interpolation='nearest', transform=transform)
if (cbnorm == 'symlog'):
if LooseVersion(matplotlib.__version__) < LooseVersion("2.0.0"):
formatter_kwargs = {}
else:
formatter_kwargs = dict(linthresh=cblinthresh)
formatter = matplotlib.ticker.LogFormatterMathtext(
**formatter_kwargs)
self.cb = self.figure.colorbar(
self.image, self.cax, format=formatter)
if np.nanmin(data) >= 0.0:
yticks = [np.nanmin(data).v] + list(
10**np.arange(np.rint(np.log10(cblinthresh)),
np.ceil(np.log10(np.nanmax(data))) + 1))
elif np.nanmax(data) <= 0.0:
yticks = list(
-10**np.arange(
np.floor(np.log10(-np.nanmin(data))),
np.rint(np.log10(cblinthresh)) - 1, -1)) + \
[np.nanmax(data).v]
else:
yticks = list(
-10**np.arange(np.floor(np.log10(-np.nanmin(data))),
np.rint(np.log10(cblinthresh))-1, -1)) + \
[0] + \
list(10**np.arange(np.rint(np.log10(cblinthresh)),
np.ceil(np.log10(np.nanmax(data)))+1))
self.cb.set_ticks(yticks)
else:
self.cb = self.figure.colorbar(self.image, self.cax)
for which in ['major', 'minor']:
self.cax.tick_params(which=which, axis='y', direction='in')
