 # categorical ranges x_range

Good morning,

Is there any way to adjust the x_range “extent” when dealing with categorical ranges?

As you can see in the picture attached I would like to crate bar charts but instead of stacking I am interested in putting each bar side to side. However, if I tried to separate the x values (centers of rectangles) to create more space between the categories or if I try to make them wider, I always end with part of the plot missing (the part at the right).

I will appreciate any suggestion.

Thanks you very much ans best regards,

Oj. Categorical axes are an area that can use improvement. Can you submit a GitHub issue with this image and problem description? In the mean time, you can use "categorical coordinates" and thinner bars to get a plot that fits. Note that the "categorical coordinate" syntax will probably change at some point, because it is extremely clunky (as you can see). Here is a short example:

from bokeh.plotting import *
import numpy as np

cats = ["A", "B", "C"]

x0 = ["A:0.65", "B:0.65", "C:0.65"]
y0 = np.array([3, 5, 2])

x1 = ["A:0.35", "B:0.35", "C:0.35"]
y1 = np.array([2, 4, 6])

output_notebook()

p = figure(x_range=cats, tools="")
p.rect(x=x0, y=y0/2, color="red", width=0.3, height=y0)
p.rect(x=x1, y=y1/2, color="green", width=0.3, height=y1)

show(p)

Bryan

···

On Dec 2, 2014, at 7:55 AM, Oscar Delgado <[email protected]> wrote:

Good morning,

Is there any way to adjust the x_range "extent" when dealing with categorical ranges?

As you can see in the picture attached I would like to crate bar charts but instead of stacking I am interested in putting each bar side to side. However, if I tried to separate the x values (centers of rectangles) to create more space between the categories or if I try to make them wider, I always end with part of the plot missing (the part at the right).

I will appreciate any suggestion.

Thanks you very much ans best regards,

Oj.

--
You received this message because you are subscribed to the Google Groups "Bokeh Discussion - Public" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [email protected].
To post to this group, send email to [email protected].
<test.png>

hello Bryan,

I submitted the GitHub issue with the graph as you suggested.

Thanks a lot for the example with the categorical coordinates.

regards,

Oj.

···

Le mardi 2 décembre 2014 11:10:58 UTC-5, Bryan Van de ven a écrit :

Categorical axes are an area that can use improvement. Can you submit a GitHub issue with this image and problem description? In the mean time, you can use “categorical coordinates” and thinner bars to get a plot that fits. Note that the “categorical coordinate” syntax will probably change at some point, because it is extremely clunky (as you can see). Here is a short example:

from bokeh.plotting import *

import numpy as np

cats = [“A”, “B”, “C”]

x0 = [“A:0.65”, “B:0.65”, “C:0.65”]

y0 = np.array([3, 5, 2])

x1 = [“A:0.35”, “B:0.35”, “C:0.35”]

y1 = np.array([2, 4, 6])

output_notebook()

p = figure(x_range=cats, tools="")

p.rect(x=x0, y=y0/2, color=“red”, width=0.3, height=y0)

p.rect(x=x1, y=y1/2, color=“green”, width=0.3, height=y1)

show§

Bryan

On Dec 2, 2014, at 7:55 AM, Oscar Delgado [email protected] wrote:

Good morning,

Is there any way to adjust the x_range “extent” when dealing with categorical ranges?

As you can see in the picture attached I would like to crate bar charts but instead of stacking I am interested in putting each bar side to side. However, if I tried to separate the x values (centers of rectangles) to create more space between the categories or if I try to make them wider, I always end with part of the plot missing (the part at the right).

I will appreciate any suggestion.

Thanks you very much ans best regards,

Oj.

You received this message because you are subscribed to the Google Groups “Bokeh Discussion - Public” group.

To unsubscribe from this group and stop receiving emails from it, send an email to [email protected].

To post to this group, send email to [email protected].

<test.png>

Just for reference, this is the opened issue at github: https://github.com/bokeh/bokeh/issues/1492

···

On Tuesday, December 2, 2014 2:30:26 PM UTC-3, Oscar Delgado wrote:

hello Bryan,

I submitted the GitHub issue with the graph as you suggested.

Thanks a lot for the example with the categorical coordinates.

regards,

Oj.

Le mardi 2 décembre 2014 11:10:58 UTC-5, Bryan Van de ven a écrit :

Categorical axes are an area that can use improvement. Can you submit a GitHub issue with this image and problem description? In the mean time, you can use “categorical coordinates” and thinner bars to get a plot that fits. Note that the “categorical coordinate” syntax will probably change at some point, because it is extremely clunky (as you can see). Here is a short example:

from bokeh.plotting import *

import numpy as np

cats = [“A”, “B”, “C”]

x0 = [“A:0.65”, “B:0.65”, “C:0.65”]

y0 = np.array([3, 5, 2])

x1 = [“A:0.35”, “B:0.35”, “C:0.35”]

y1 = np.array([2, 4, 6])

output_notebook()

p = figure(x_range=cats, tools="")

p.rect(x=x0, y=y0/2, color=“red”, width=0.3, height=y0)

p.rect(x=x1, y=y1/2, color=“green”, width=0.3, height=y1)

show§

Bryan

On Dec 2, 2014, at 7:55 AM, Oscar Delgado [email protected] wrote:

Good morning,

Is there any way to adjust the x_range “extent” when dealing with categorical ranges?

As you can see in the picture attached I would like to crate bar charts but instead of stacking I am interested in putting each bar side to side. However, if I tried to separate the x values (centers of rectangles) to create more space between the categories or if I try to make them wider, I always end with part of the plot missing (the part at the right).

I will appreciate any suggestion.

Thanks you very much ans best regards,

Oj.

You received this message because you are subscribed to the Google Groups “Bokeh Discussion - Public” group.

To unsubscribe from this group and stop receiving emails from it, send an email to [email protected].

To post to this group, send email to [email protected].