调色板创建与蟒蛇散景(Colour palette creation with python bokeh)

调色板创建与蟒蛇散景(Colour palette creation with python bokeh)

我很努力地想知道如何为数据集中第3列的bokeh.scatter图生成调色板。 这是一个例子：

from bokeh.plotting import figure, output_file, show import pandas as pd data_headers = ['eastings', 'northings', 'obs'] data_points = pd.read_csv('~/Documents/ds/data.csv', header=None, names=data_headers) my_color_map = [(20, green_val, blue_val) for blue_val, green_val in zip(data_points.obs, 255-data_points.obs)] output_file("scatter.html") p=figure(x_range=(0, 5), y_range=(0, 10)) p.scatter(x=data_points.eastings, y=data_points.northings, size=5, fill_color=my_color_map) show(p)

数据集是：

1,3,123 1,4,97 1,5,83 1,6,192 2,3,126 2,3.5,97 2,4.6,102 2,5.8,45

而色彩图是一个合理的价差：

>>> my_color_map [(20, 132, 123), (20, 158, 97), (20, 172, 83), (20, 63, 192), (20, 129, 126), (20, 158, 97), (20, 153, 102), (20, 210, 45)]

代码工作得很好，除了它没有填充圆圈。 我可以用fill_color（100,100,100）硬连线，以满足自己我可以批量更改颜色，但我想让颜色成为第3列的功能。

我正在尝试学习如何手动执行此操作（即，只填充RGB元组中的绿色和蓝色通道），但我认为我的最终目标是使用调制器调色板（例如，绘制x，y点的颜色） ）从booked.palettes相应地着色点。

我也可以在matplotlib中工作，但这里的目的是看看散景是如何工作的！ 谢谢你的帮助 ：-）

I'm struggling to get my head around how to generate a colour palette for a bokeh.scatter plot from the 3rd column in a dataset. Here's an example:

from bokeh.plotting import figure, output_file, show import pandas as pd data_headers = ['eastings', 'northings', 'obs'] data_points = pd.read_csv('~/Documents/ds/data.csv', header=None, names=data_headers) my_color_map = [(20, green_val, blue_val) for blue_val, green_val in zip(data_points.obs, 255-data_points.obs)] output_file("scatter.html") p=figure(x_range=(0, 5), y_range=(0, 10)) p.scatter(x=data_points.eastings, y=data_points.northings, size=5, fill_color=my_color_map) show(p)

The data set is:

1,3,123 1,4,97 1,5,83 1,6,192 2,3,126 2,3.5,97 2,4.6,102 2,5.8,45

And the colour map is a reasonable spread thus:

>>> my_color_map [(20, 132, 123), (20, 158, 97), (20, 172, 83), (20, 63, 192), (20, 129, 126), (20, 158, 97), (20, 153, 102), (20, 210, 45)]

The code works fine apart from it doesn't fill the circles. I can hardwire it with fill_color(100,100,100) to satisfy myself that I can change the colours in bulk but I want to make the colour a function of the 3rd column.

I'm trying to learn how to do this manually (i.e. just populate the green and blue channels in the RGB tuple) but I think my ultimate aim would be to plot the colour of a point at x,y using a brewer palette (etc) from booked.palettes to colour the point accordingly.

I can also get it working in matplotlib but the aim here is to see how bokeh works! Thanks for any help :-)

最满意答案

传递颜色元组列表目前不起作用 。 相反，您可以将颜色定义转换为字符串，并将字符串列表传递给fill_color

from bokeh.plotting import figure, show, output_notebook
import pandas as pd
output_notebook()

TESTDATA=StringIO("""1,3,123
1,4,97
1,5,83
1,6,192
2,3,126
2,3.5,97
2,4.6,102
2,5.8,45
""")

data_headers = ['eastings', 'northings', 'obs']
data_points = pd.read_csv(TESTDATA, header=None, names=data_headers)

my_color_map = ["rgb({!s},{!s},{!s})".format(20, green_val, blue_val) for blue_val, green_val in zip(data_points.obs, 255-data_points.obs)]

p=figure(x_range=(0, 5), y_range=(0, 10))
p.scatter(x=data_points.eastings, y=data_points.northings, size=25, fill_color=my_color_map)
show(p)
Passing a list of color tuples doesn't work at the moment. Instead, you can turn your color definitions into strings and pass the list of strings to fill_color
 
from bokeh.plotting import figure, show, output_notebook
import pandas as pd
output_notebook()

TESTDATA=StringIO("""1,3,123
1,4,97
1,5,83
1,6,192
2,3,126
2,3.5,97
2,4.6,102
2,5.8,45
""")

data_headers = ['eastings', 'northings', 'obs']
data_points = pd.read_csv(TESTDATA, header=None, names=data_headers)

my_color_map = ["rgb({!s},{!s},{!s})".format(20, green_val, blue_val) for blue_val, green_val in zip(data_points.obs, 255-data_points.obs)]

p=figure(x_range=(0, 5), y_range=(0, 10))
p.scatter(x=data_points.eastings, y=data_points.northings, size=25, fill_color=my_color_map)
show(p)调色板创建与蟒蛇散景(Colour palette creation with python bokeh)

                
 我很努力地想知道如何为数据集中第3列的bokeh.scatter图生成调色板。 这是一个例子： 
 
from bokeh.plotting import figure, output_file, show
import pandas as pd

data_headers = ['eastings', 'northings', 'obs']
data_points = pd.read_csv('~/Documents/ds/data.csv', header=None, names=data_headers)

my_color_map = [(20, green_val, blue_val) for blue_val, green_val in zip(data_points.obs, 255-data_points.obs)]

output_file("scatter.html")

p=figure(x_range=(0, 5), y_range=(0, 10))
p.scatter(x=data_points.eastings, y=data_points.northings, size=5, fill_color=my_color_map)
show(p)
 
 数据集是： 
 
1,3,123
1,4,97
1,5,83
1,6,192
2,3,126
2,3.5,97
2,4.6,102
2,5.8,45
 
 而色彩图是一个合理的价差： 
 
>>> my_color_map
[(20, 132, 123), (20, 158, 97), (20, 172, 83), (20, 63, 192), (20, 129, 126), (20, 158, 97), (20, 153, 102), (20, 210, 45)]
 
 代码工作得很好，除了它没有填充圆圈。 我可以用fill_color（100,100,100）硬连线，以满足自己我可以批量更改颜色，但我想让颜色成为第3列的功能。 
 
 我正在尝试学习如何手动执行此操作（即，只填充RGB元组中的绿色和蓝色通道），但我认为我的最终目标是使用调制器调色板（例如，绘制x，y点的颜色） ）从booked.palettes相应地着色点。 
 
 我也可以在matplotlib中工作，但这里的目的是看看散景是如何工作的！ 谢谢你的帮助 ：-） 
I'm struggling to get my head around how to generate a colour palette for a bokeh.scatter plot from the 3rd column in a dataset. Here's an example:
 
from bokeh.plotting import figure, output_file, show
import pandas as pd

data_headers = ['eastings', 'northings', 'obs']
data_points = pd.read_csv('~/Documents/ds/data.csv', header=None, names=data_headers)

my_color_map = [(20, green_val, blue_val) for blue_val, green_val in zip(data_points.obs, 255-data_points.obs)]

output_file("scatter.html")

p=figure(x_range=(0, 5), y_range=(0, 10))
p.scatter(x=data_points.eastings, y=data_points.northings, size=5, fill_color=my_color_map)
show(p)
 
The data set is:
 
1,3,123
1,4,97
1,5,83
1,6,192
2,3,126
2,3.5,97
2,4.6,102
2,5.8,45
 
And the colour map is a reasonable spread thus:
 
>>> my_color_map
[(20, 132, 123), (20, 158, 97), (20, 172, 83), (20, 63, 192), (20, 129, 126), (20, 158, 97), (20, 153, 102), (20, 210, 45)]
 
The code works fine apart from it doesn't fill the circles. I can hardwire it with fill_color(100,100,100) to satisfy myself that I can change the colours in bulk but I want to make the colour a function of the 3rd column.
 
I'm trying to learn how to do this manually (i.e. just populate the green and blue channels in the RGB tuple) but I think my ultimate aim would be to plot the colour of a point at x,y using a brewer palette (etc) from booked.palettes to colour the point accordingly.
 
I can also get it working in matplotlib but the aim here is to see how bokeh works! Thanks for any help :-)
                
最满意答案
                
                    
                        
 传递颜色元组列表目前不起作用 。 相反，您可以将颜色定义转换为字符串，并将字符串列表传递给fill_color 
 
from bokeh.plotting import figure, show, output_notebook
import pandas as pd
output_notebook()

TESTDATA=StringIO("""1,3,123
1,4,97
1,5,83
1,6,192
2,3,126
2,3.5,97
2,4.6,102
2,5.8,45
""")

data_headers = ['eastings', 'northings', 'obs']
data_points = pd.read_csv(TESTDATA, header=None, names=data_headers)

my_color_map = ["rgb({!s},{!s},{!s})".format(20, green_val, blue_val) for blue_val, green_val in zip(data_points.obs, 255-data_points.obs)]

p=figure(x_range=(0, 5), y_range=(0, 10))
p.scatter(x=data_points.eastings, y=data_points.northings, size=25, fill_color=my_color_map)
show(p)
Passing a list of color tuples doesn't work at the moment. Instead, you can turn your color definitions into strings and pass the list of strings to fill_color
 
from bokeh.plotting import figure, show, output_notebook
import pandas as pd
output_notebook()

TESTDATA=StringIO("""1,3,123
1,4,97
1,5,83
1,6,192
2,3,126
2,3.5,97
2,4.6,102
2,5.8,45
""")

data_headers = ['eastings', 'northings', 'obs']
data_points = pd.read_csv(TESTDATA, header=None, names=data_headers)

my_color_map = ["rgb({!s},{!s},{!s})".format(20, green_val, blue_val) for blue_val, green_val in zip(data_points.obs, 255-data_points.obs)]

p=figure(x_range=(0, 5), y_range=(0, 10))
p.scatter(x=data_points.eastings, y=data_points.northings, size=25, fill_color=my_color_map)
show(p)