from warnings import warn
from mizani.bounds import rescale_mid
from mizani.palettes import (hue_pal, brewer_pal, grey_pal,
gradient_n_pal, cmap_pal, cmap_d_pal,
desaturate_pal)
from ..utils import alias
from ..doctools import document
from ..exceptions import PlotnineWarning
from .scale import scale_discrete, scale_continuous, scale_datetime
# Discrete color scales #
# Note: plotnine operates in the hcl space
[docs]@document
class scale_color_hue(scale_discrete):
"""
Qualitative color scale with evenly spaced hues
Parameters
----------
h : float
first hue. Must be in the range [0, 1]
Default is ``0.01``
l : float
lightness. Must be in the range [0, 1]
Default is ``0.6``
s : float
saturation. Must be in the range [0, 1]
Default is ``0.65``
colorspace : str in ``['hls', 'husl']``
Color space to use.
`hls <https://en.wikipedia.org/wiki/HSL_and_HSV>`_
`husl <http://www.husl-colors.org/>`_
{superclass_parameters}
na_value : str
Color of missing values. Default is ``'#7F7F7F'``
"""
_aesthetics = ['color']
na_value = '#7F7F7F'
def __init__(self, h=.01, l=.6, s=.65, color_space='hls', **kwargs):
self.palette = hue_pal(h, l, s, color_space=color_space)
scale_discrete.__init__(self, **kwargs)
[docs]@document
class scale_fill_hue(scale_color_hue):
"""
Qualitative color scale with evenly spaced hues
Parameters
----------
{superclass_parameters}
"""
_aesthetics = ['fill']
[docs]@document
class scale_color_brewer(scale_discrete):
"""
Sequential, diverging and qualitative discrete color scales
See `colorbrewer.org <http://colorbrewer2.org/>`_
Parameters
----------
type : str in ``['seq', 'div', 'qual']``
Type of data. Sequential, diverging or qualitative
palette : int | str
If a string, will use that named palette.
If a number, will index into the list of palettes
of appropriate type. Default is 1
direction: int in ``[-1, 1]``
Sets the order of colors in the scale. If 1,
the default, colors are as output by
mizani.palettes.brewer_pal(). If -1,
the order of colors is reversed.
{superclass_parameters}
na_value : str
Color of missing values. Default is ``'None'``
"""
_aesthetics = ['color']
na_value = '#7F7F7F'
def __init__(self, type='seq', palette=1, direction=1, **kwargs):
self.palette = brewer_pal(type, palette, direction=direction)
scale_discrete.__init__(self, **kwargs)
[docs]@document
class scale_fill_brewer(scale_color_brewer):
"""
Sequential, diverging and qualitative color scales
Parameters
----------
{superclass_parameters}
"""
_aesthetics = ['fill']
[docs]@document
class scale_color_grey(scale_discrete):
"""
Sequential grey color scale.
Parameters
----------
start : float
grey value at low end of palette.
Default is 0.2
end : float
grey value at high end of palette
Default is 0.8
{superclass_parameters}
"""
_aesthetics = ['color']
def __init__(self, start=0.2, end=0.8, **kwargs):
self.palette = grey_pal(start, end)
scale_discrete.__init__(self, **kwargs)
[docs]@document
class scale_fill_grey(scale_color_grey):
"""
Sequential grey color scale.
Parameters
----------
{superclass_parameters}
"""
_aesthetics = ['fill']
# Continuous color scales #
[docs]@document
class scale_color_gradient(scale_continuous):
"""
Create a 2 point color gradient
Parameters
----------
low : str
low color
high : str
high color
{superclass_parameters}
na_value : str
Color of missing values. Default is ``'None'``
See Also
--------
:class:`.scale_color_gradient2`
:class:`.scale_color_gradientn`
"""
_aesthetics = ['color']
guide = 'colorbar'
na_value = '#7F7F7F'
def __init__(self, low='#132B43', high='#56B1F7', **kwargs):
"""
Create colormap that will be used by the palette
"""
self.palette = gradient_n_pal([low, high],
name='gradient')
scale_continuous.__init__(self, **kwargs)
[docs]@document
class scale_fill_gradient(scale_color_gradient):
"""
Create a 2 point color gradient
Parameters
----------
{superclass_parameters}
"""
_aesthetics = ['fill']
[docs]@document
class scale_color_desaturate(scale_continuous):
"""
Create a desaturated color gradient
Parameters
----------
color : str, optional (Default: 'red')
Color to desaturate
prop : float, optional (Default: 0)
Saturation channel of color will be multiplied by
this value.
reverse : bool, optional (Default: False)
Whether to go from color to desaturated color
or desaturated color to color.
{superclass_parameters}
na_value : str
Color of missing values. Default is ``'None'``
"""
_aesthetics = ['color']
guide = 'colorbar'
na_value = '#7F7F7F'
def __init__(self, color='red', prop=0, reverse=False,
**kwargs):
self.palette = desaturate_pal(color, prop, reverse)
scale_continuous.__init__(self, **kwargs)
[docs]@document
class scale_fill_desaturate(scale_color_desaturate):
"""
Create a desaturated color gradient
Parameters
----------
{superclass_parameters}
"""
_aesthetics = ['fill']
[docs]@document
class scale_color_gradient2(scale_continuous):
"""
Create a 3 point diverging color gradient
Parameters
----------
low : str, optional
low color
mid : str, optional
mid point color
high : str, optional
high color
midpoint : float, optional (Default: 0)
Mid point of the input data range.
{superclass_parameters}
na_value : str
Color of missing values. Default is ``'None'``
See Also
--------
:class:`.scale_color_gradient`
:class:`.scale_color_gradientn`
"""
_aesthetics = ['color']
guide = 'colorbar'
na_value = '#7F7F7F'
def __init__(self, low='#832424', mid='#FFFFFF',
high='#3A3A98', midpoint=0,
**kwargs):
# All rescale functions should have the same signature
def _rescale_mid(*args, **kwargs):
return rescale_mid(*args, mid=midpoint, **kwargs)
kwargs['rescaler'] = _rescale_mid
self.palette = gradient_n_pal([low, mid, high],
name='gradient2')
scale_continuous.__init__(self, **kwargs)
[docs]@document
class scale_fill_gradient2(scale_color_gradient2):
"""
Create a 3 point diverging color gradient
Parameters
----------
{superclass_parameters}
"""
_aesthetics = ['fill']
[docs]@document
class scale_color_gradientn(scale_continuous):
"""
Create a n color gradient
Parameters
----------
colors : list
list of colors
values : list, optional
list of points in the range [0, 1] at which to
place each color. Must be the same size as
`colors`. Default to evenly space the colors
{superclass_parameters}
na_value : str
Color of missing values. Default is ``'None'``
See Also
--------
:class:`.scale_color_gradient`
:class:`.scale_color_gradientn`
"""
_aesthetics = ['color']
guide = 'colorbar'
na_value = '#7F7F7F'
def __init__(self, colors, values=None, **kwargs):
self.palette = gradient_n_pal(colors, values, 'gradientn')
scale_continuous.__init__(self, **kwargs)
[docs]@document
class scale_fill_gradientn(scale_color_gradientn):
"""
Create a n color gradient
Parameters
----------
{superclass_parameters}
"""
_aesthetics = ['fill']
[docs]@document
class scale_color_distiller(scale_color_gradientn):
"""
Sequential and diverging continuous color scales
This is a convinience scale around :class:`.scale_color_gradientn`
with colors from `colorbrewer.org <http://colorbrewer2.org/>`_.
It smoothly interpolates 7 colors from a brewer palette to create
a continuous palette.
Parameters
----------
type : str in ``['seq', 'div']``
Type of data. Sequential, diverging or qualitative
palette : int | str
If a string, will use that named palette.
If a number, will index into the list of palettes
of appropriate type. Default is 1
values : list, optional
list of points in the range [0, 1] at which to
place each color. Must be the same size as
`colors`. Default to evenly space the colors
direction: int in ``[-1, 1]``
Sets the order of colors in the scale. If 1
colors are as output by mizani.palettes.brewer_pal().
If -1, the default, the order of colors is reversed.
{superclass_parameters}
na_value : str
Color of missing values. Default is ``'None'``
"""
_aesthetics = ['color']
guide = 'colorbar'
na_value = '#7F7F7F'
def __init__(self, type='seq', palette=1,
values=None, direction=-1, **kwargs):
"""
Create colormap that will be used by the palette
"""
if type.lower() in ('qual', 'qualitative'):
warn("Using a discrete color palette in a continuous scale."
"Consider using type = 'seq' or type = 'div' instead",
PlotnineWarning)
# Grab 7 colors from brewer and create a gradient palette
# An odd number matches the midpoint of the palette to that
# of the data
colors = brewer_pal(type, palette, direction=direction)(7)
scale_color_gradientn.__init__(self, colors, values, **kwargs)
[docs]@document
class scale_fill_distiller(scale_color_distiller):
"""
Sequential, diverging continuous color scales
Parameters
----------
{superclass_parameters}
"""
_aesthetics = ['fill']
# matplotlib colormaps
[docs]@document
class scale_color_cmap(scale_continuous):
"""
Create color scales using Matplotlib colormaps
Parameters
----------
cmap_name : str
A standard Matplotlib colormap name. The default is
`viridis`. For the list of names checkout the output
of ``matplotlib.cm.cmap_d.keys()`` or see the
`documentation <http://matplotlib.org/users/colormaps.html>`_.
lut : None | int
This is the number of entries desired in the
lookup table. Default is `None`, leave it up
Matplotlib.
{superclass_parameters}
na_value : str
Color of missing values. Default is ``'None'``
See Also
--------
:mod:`matplotlib.cm`
:mod:`matplotlib.colors`
"""
_aesthetics = ['color']
guide = 'colorbar'
na_value = '#7F7F7F'
def __init__(self, cmap_name='viridis', lut=None, **kwargs):
self.palette = cmap_pal(cmap_name, lut)
super(scale_color_cmap, self).__init__(**kwargs)
[docs]@document
class scale_fill_cmap(scale_color_cmap):
"""
Create color scales using Matplotlib colormaps
Parameters
----------
{superclass_parameters}
"""
_aesthetics = ['fill']
@document
class scale_color_cmap_d(scale_discrete):
"""
A discrete color scales using Matplotlib colormaps
Parameters
----------
name : str
A standard Matplotlib colormap name. It must be of type
:class:`matplotlib.colors.ListedColormap`.
. The default is `viridis`. For the list of names checkout
the output of ``matplotlib.cm.cmap_d.keys()`` or see the
`documentation <http://matplotlib.org/users/colormaps.html>`_.
lut : None | int
This is the number of entries desired in the
lookup table. Default is `None`, leave it up
Matplotlib.
{superclass_parameters}
na_value : str
Color of missing values. Default is ``'None'``
See Also
--------
:mod:`matplotlib.cm`
:mod:`matplotlib.colors`
"""
_aesthetics = ['color']
na_value = '#7F7F7F'
def __init__(self, name='viridis', lut=None, **kwargs):
self.palette = cmap_d_pal(name, lut)
super(scale_color_cmap_d, self).__init__(**kwargs)
@document
class scale_fill_cmap_d(scale_color_cmap_d):
"""
Create color scales using Matplotlib colormaps
Parameters
----------
{superclass_parameters}
"""
_aesthetics = ['fill']
[docs]@document
class scale_color_datetime(scale_datetime, scale_color_cmap):
"""
Datetime color scale
Parameters
----------
{superclass_parameters}
"""
[docs]@document
class scale_fill_datetime(scale_datetime, scale_fill_cmap):
"""
Datetime fill scale
Parameters
----------
{superclass_parameters}
"""
# Default scales
alias('scale_color_discrete', scale_color_hue)
alias('scale_color_continuous', scale_color_cmap)
alias('scale_color_ordinal', scale_color_cmap_d)
alias('scale_fill_discrete', scale_fill_hue)
alias('scale_fill_continuous', scale_fill_cmap)
alias('scale_fill_ordinal', scale_fill_cmap_d)
# American to British spelling
alias('scale_colour_hue', scale_color_hue)
alias('scale_color_gray', scale_color_grey)
alias('scale_colour_grey', scale_color_grey)
alias('scale_colour_gray', scale_color_grey)
alias('scale_fill_gray', scale_fill_grey)
alias('scale_colour_brewer', scale_color_brewer)
alias('scale_colour_desaturate', scale_color_desaturate)
alias('scale_colour_gradient', scale_color_gradient)
alias('scale_colour_gradient2', scale_color_gradient2)
alias('scale_colour_gradientn', scale_color_gradientn)
alias('scale_colour_discrete', scale_color_hue)
alias('scale_colour_continuous', scale_color_cmap)
alias('scale_colour_distiller', scale_color_distiller)
alias('scale_colour_cmap', scale_color_cmap)
alias('scale_colour_cmap_d', scale_color_cmap_d)
alias('scale_colour_datetime', scale_color_datetime)
alias('scale_colour_ordinal', scale_color_cmap_d)
