Archived discussions · Captured October 28, 2026

seaborn jointplot with bokeh

FabianJun 30, 2017 #1

Hello,

I would like to get a seaborn jointplot into bokeh.

The lines below work with the kdeplot, but not with the jointplot.

import seaborn as sns
from bokeh import mpl
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np
from bokeh.io import show

df = pd.DataFrame(np.random.randint(0,100,size=(100, 4)), columns=list('ABCD'))
ax = sns.jointplot(x="A", y="B", data=df, kind="kde")
# ax = sns.kdeplot(df["A"], df["B"])
show(mpl.to_bokeh())

Do you have a suggestion how to proceed with this? I assume that I need somehow create several bokeh plots out of the jointplot...

Thank you!

Best Regards

Fabian

⌁ Link to post

BryanJun 30, 2017 #2

Hi,

I would actually recommend that you check out the new HoloViews release:

[http://holoviews.org/](http://holoviews.org/)

HoloViews is a very high-level API for data exploration that integrates tightly with Bokeh to produce statistical and other visualizations with less code. We will be advocating for users to use Holoviews (instead of the old bkcharts package) going forward.

Here is an example of a Bokeh joint plot created using Holoviews:

···

On Jun 30, 2017, at 11:33, ‘Fabian Braennstroem’ via Bokeh Discussion - Public bokeh@continuum.io wrote:

Hello,

I would like to get a seaborn jointplot into bokeh.

The lines below work with the kdeplot, but not with the jointplot.

import seaborn as sns
from bokeh import mpl
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np
from bokeh.io import show

df = pd.DataFrame(np.random.randint(0,100,size=(100, 4)), columns=list(‘ABCD’))
ax = sns.jointplot(x=“A”, y=“B”, data=df, kind=“kde”)

ax = sns.kdeplot(df[“A”], df[“B”])

show(mpl.to_bokeh())

Do you have a suggestion how to proceed with this? I assume that I need somehow create several bokeh plots out of the jointplot…

Thank you!

Best Regards

Fabian

–
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 bokeh+unsubscribe@continuum.io.
To post to this group, send email to bokeh@continuum.io.
To view this discussion on the web visit https://groups.google.com/a/continuum.io/d/msgid/bokeh/cf39bca1-658c-9cf1-d486-d2d61c4155b1%40googlemail.com.
For more options, visit https://groups.google.com/a/continuum.io/d/optout.

⌁ Link to post

BryanJun 30, 2017 #3

I was just informed this can be done even more simply

  import numpy as np
  import holoviews as hv
  from holoviews.operation import histogram
  hv.extension('bokeh')

data = np.random.multivariate_normal((0, 0), [[1, 0.1], [0.1, 1]], (1000,))
points = hv.Points(data)

points.hist(dimension=['x','y'])

Thanks,

Bryan

···

On Jun 30, 2017, at 12:11, Bryan Van de ven <bryan@bokeh.org> wrote:

Hi,

I would actually recommend that you check out the new HoloViews release:

http://holoviews.org/

HoloViews is a very high-level API for data exploration that integrates tightly with Bokeh to produce statistical and other visualizations with less code. We will be advocating for users to use Holoviews (instead of the old bkcharts package) going forward.

Here is an example of a Bokeh joint plot created using Holoviews:

<image.png>

On Jun 30, 2017, at 11:33, 'Fabian Braennstroem' via Bokeh Discussion - Public <bokeh@continuum.io> wrote:

Hello,

I would like to get a seaborn jointplot into bokeh.

The lines below work with the kdeplot, but not with the jointplot.

import seaborn as sns
from bokeh import mpl
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np
from bokeh.io import show

df = pd.DataFrame(np.random.randint(0,100,size=(100, 4)), columns=list('ABCD'))
ax = sns.jointplot(x="A", y="B", data=df, kind="kde")
# ax = sns.kdeplot(df["A"], df["B"])
show(mpl.to_bokeh())

Do you have a suggestion how to proceed with this? I assume that I need somehow create several bokeh plots out of the jointplot...

Thank you!

Best Regards

Fabian

--
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 bokeh+unsubscribe@continuum.io.
To post to this group, send email to bokeh@continuum.io.
To view this discussion on the web visit https://groups.google.com/a/continuum.io/d/msgid/bokeh/cf39bca1-658c-9cf1-d486-d2d61c4155b1%40googlemail.com\.
For more options, visit https://groups.google.com/a/continuum.io/d/optout\.

⌁ Link to post

BryanJun 30, 2017 #4

Another cool updated version, in case you want multiple distributions on one plot:

···

On Jun 30, 2017, at 12:16, Bryan Van de ven bryan@bokeh.org wrote:

I was just informed this can be done even more simply

import numpy as np
import holoviews as hv
from holoviews.operation import histogram
hv.extension(‘bokeh’)

data = np.random.multivariate_normal((0, 0), [[1, 0.1], [0.1, 1]], (1000,))
points = hv.Points(data)

points.hist(dimension=[‘x’,‘y’])

Thanks,

Bryan
On Jun 30, 2017, at 12:11, Bryan Van de ven bryanv@continuum.io wrote:

Hi,

I would actually recommend that you check out the new HoloViews release:
[http://holoviews.org/](http://holoviews.org/)
HoloViews is a very high-level API for data exploration that integrates tightly with Bokeh to produce statistical and other visualizations with less code. We will be advocating for users to use Holoviews (instead of the old bkcharts package) going forward.

Here is an example of a Bokeh joint plot created using Holoviews:

<image.png>

On Jun 30, 2017, at 11:33, ‘Fabian Braennstroem’ via Bokeh Discussion - Public bokeh@continuum.io wrote:

Hello,

I would like to get a seaborn jointplot into bokeh.

The lines below work with the kdeplot, but not with the jointplot.

import seaborn as sns
from bokeh import mpl
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np
from bokeh.io import show

df = pd.DataFrame(np.random.randint(0,100,size=(100, 4)), columns=list(‘ABCD’))
ax = sns.jointplot(x=“A”, y=“B”, data=df, kind=“kde”)

ax = sns.kdeplot(df[“A”], df[“B”])

show(mpl.to_bokeh())

Do you have a suggestion how to proceed with this? I assume that I need somehow create several bokeh plots out of the jointplot…

Thank you!

Best Regards

Fabian

–
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 bokeh+unsubscribe@continuum.io.
To post to this group, send email to bokeh@continuum.io.
To view this discussion on the web visit https://groups.google.com/a/continuum.io/d/msgid/bokeh/cf39bca1-658c-9cf1-d486-d2d61c4155b1%40googlemail.com.
For more options, visit https://groups.google.com/a/continuum.io/d/optout.

⌁ Link to post

FabianJun 30, 2017 #5

Hi. Thank you very much for the suggestions!
Fabian

···

On Jun 30, 2017 7:21 PM, “Bryan Van de ven” bryan@bokeh.org wrote:

Another cool updated version, in case you want multiple distributions on one plot:

On Jun 30, 2017, at 12:11, Bryan Van de ven bryan@bokeh.org wrote:

Hi,

I would actually recommend that you check out the new HoloViews release:
[http://holoviews.org/](http://holoviews.org/)
HoloViews is a very high-level API for data exploration that integrates tightly with Bokeh to produce statistical and other visualizations with less code. We will be advocating for users to use Holoviews (instead of the old bkcharts package) going forward.

Here is an example of a Bokeh joint plot created using Holoviews:

<image.png>

On Jun 30, 2017, at 11:33, ‘Fabian Braennstroem’ via Bokeh Discussion - Public bokeh@continuum.io wrote:

Hello,

I would like to get a seaborn jointplot into bokeh.

The lines below work with the kdeplot, but not with the jointplot.

import seaborn as sns
from bokeh import mpl
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np
from bokeh.io import show

df = pd.DataFrame(np.random.randint(0,100,size=(100, 4)), columns=list(‘ABCD’))
ax = sns.jointplot(x=“A”, y=“B”, data=df, kind=“kde”)

ax = sns.kdeplot(df[“A”], df[“B”])

show(mpl.to_bokeh())

Do you have a suggestion how to proceed with this? I assume that I need somehow create several bokeh plots out of the jointplot…

Thank you!

Best Regards

Fabian

–
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 bokeh+unsubscribe@continuum.io.
To post to this group, send email to bokeh@continuum.io.
To view this discussion on the web visit https://groups.google.com/a/continuum.io/d/msgid/bokeh/cf39bca1-658c-9cf1-d486-d2d61c4155b1%40googlemail.com.
For more options, visit https://groups.google.com/a/continuum.io/d/optout.
On Jun 30, 2017, at 12:16, Bryan Van de ven bryan@bokeh.org wrote:

I was just informed this can be done even more simply
import numpy as np
import holoviews as hv
from holoviews.operation import histogram
hv.extension('bokeh')


data = np.random.multivariate_normal((0, 0), [[1, 0.1], [0.1, 1]], (1000,))
points = hv.Points(data)

points.hist(dimension=['x','y'])
Thanks,

Bryan

–

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 bokeh+unsubscribe@continuum.io.

To post to this group, send email to bokeh@continuum.io.

To view this discussion on the web visit https://groups.google.com/a/continuum.io/d/msgid/bokeh/B2397DB4-A496-44B3-8FA5-C636FA6C7E30%40continuum.io.

For more options, visit https://groups.google.com/a/continuum.io/d/optout.

⌁ Link to post

FabianJul 06, 2017 #6

Hello Bryan,

do you have an idea, if there is in option for a kde (see the 'kind' of jointplot in seaborn) plot as well? I could not figure this out yet; actually it does not look like that it exists.

Thank you in advance!
Fabian

···

On 06/30/2017 07:21 PM, Bryan Van de ven wrote:

Another cool updated version, in case you want multiple distributions on one plot:

On Jun 30, 2017, at 12:16, Bryan Van de ven <bryan@bokeh.org >> <mailto:bryan@bokeh.org>> wrote:

I was just informed this can be done even more simply

import numpy as np
import holoviews as hv
from holoviews.operation import histogram
hv.extension('bokeh')

data = np.random.multivariate_normal((0, 0), [[1, 0.1], [0.1, 1]], (1000,))
points = hv.Points(data)

points.hist(dimension=['x','y'])

Thanks,

Bryan

On Jun 30, 2017, at 12:11, Bryan Van de ven <bryan@bokeh.org >>> <mailto:bryan@bokeh.org>> wrote:

Hi,

I would actually recommend that you check out the new HoloViews release:

http://holoviews.org/

HoloViews is a very high-level API for data exploration that integrates tightly with Bokeh to produce statistical and other visualizations with less code. We will be advocating for users to use Holoviews (instead of the old bkcharts package) going forward.

Here is an example of a Bokeh joint plot created using Holoviews:

<image.png>

On Jun 30, 2017, at 11:33, 'Fabian Braennstroem' via Bokeh >>>> Discussion - Public <bokeh@continuum.io> wrote:

Hello,

I would like to get a seaborn jointplot into bokeh.

The lines below work with the kdeplot, but not with the jointplot.

import seaborn as sns
from bokeh import mpl
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np
from bokeh.io import show

df = pd.DataFrame(np.random.randint(0,100,size=(100, 4)), columns=list('ABCD'))
ax = sns.jointplot(x="A", y="B", data=df, kind="kde")
# ax = sns.kdeplot(df["A"], df["B"])
show(mpl.to_bokeh())

Do you have a suggestion how to proceed with this? I assume that I need somehow create several bokeh plots out of the jointplot...

Thank you!

Best Regards

Fabian

--
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 bokeh+unsubscribe@continuum.io.
To post to this group, send email to bokeh@continuum.io.
To view this discussion on the web visit https://groups.google.com/a/continuum.io/d/msgid/bokeh/cf39bca1-658c-9cf1-d486-d2d61c4155b1%40googlemail.com\.
For more options, visit https://groups.google.com/a/continuum.io/d/optout\.

--
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 bokeh+unsubscribe@continuum.io <mailto:bokeh+unsubscribe@continuum.io>.
To post to this group, send email to bokeh@continuum.io <mailto:bokeh@continuum.io>.
To view this discussion on the web visit https://groups.google.com/a/continuum.io/d/msgid/bokeh/B2397DB4-A496-44B3-8FA5-C636FA6C7E30%40continuum.io <https://groups.google.com/a/continuum.io/d/msgid/bokeh/B2397DB4-A496-44B3-8FA5-C636FA6C7E30%40continuum.io?utm_medium=email&utm_source=footer>\.
For more options, visit https://groups.google.com/a/continuum.io/d/optout\.

⌁ Link to post

philipp.jfrJul 06, 2017 #7

Hi Fabian,

We used to have support for KDE plots in HoloViews via the matplotlib/seaborn compatibility that used to be in bokeh but that has since been dropped. We are hoping to add Elements representing 1D and 2D KDEs to HoloViews in one of the upcoming releases but can’t promise how soon that will happen. For now you could compute the KDE yourself and plot it as a Curve or Area.

All the Best,

Philipp

···

On Thursday, 6 July 2017 15:49:06 UTC+1, Fabian Braennstroem wrote:

Hello Bryan,
  do you have an idea, if there is in option for a kde (see the
‘kind’ of jointplot in seaborn) plot as well? I could not figure
this out yet; actually it does not look like that it exists.
  Thank you in advance!

  Fabian



  On 06/30/2017 07:21 PM, Bryan Van de ven wrote:

    Another cool updated version, in case you want
multiple distributions on one plot:

1596×1264 181 KB

      On Jun 30, 2017, at 12:11, > > > > Bryan Van de ven <bry...@continuum.io          > > > > > wrote:



      Hi,



      I would actually recommend that you check out the new

HoloViews release:

      	[http://holoviews.org/](http://holoviews.org/)



      HoloViews is a very high-level API for data exploration that

integrates tightly with Bokeh to produce statistical and other
visualizations with less code. We will be advocating for users
to use Holoviews (instead of the old bkcharts package) going
forward.

      Here is an example of a Bokeh joint plot created using

Holoviews:

      <image.png>

        On Jun 30, 2017, at 11:33, > > > > > 'Fabian Braennstroem' via Bokeh Discussion - Public > > > > > <bo...@continuum.io> wrote:



        Hello,



        I would like to get a seaborn jointplot into bokeh.



        The lines below work with the kdeplot, but not with the

jointplot.

        import seaborn as sns

        from bokeh import mpl

        import matplotlib.pyplot as plt

        import pandas as pd

        import numpy as np

        from [bokeh.io](http://bokeh.io) import show



        df = pd.DataFrame(np.random.            randint(0,100,size=(100, 4)),

columns=list(‘ABCD’))

        ax = sns.jointplot(x="A", y="B", data=df, kind="kde")

        # ax = sns.kdeplot(df["A"], df["B"])

        show(mpl.to_bokeh())



        Do you have a suggestion how to proceed with this? I assume

that I need somehow create several bokeh plots out of the
jointplot…

        Thank you!



        Best Regards



        Fabian



        -- 

        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 bokeh+un...@continuum.io.

        To post to this group, send email to bo...@continuum.io.

        To view this discussion on the web visit

https://groups.google.com/a/continuum.io/d/msgid/bokeh/cf39bca1-658c-9cf1-d486-d2d61c4155b1%40googlemail.com.

        For more options, visit

https://groups.google.com/a/continuum.io/d/optout.

    On Jun 30, 2017, at 12:16, Bryan > > > Van de ven <bry...@continuum.io> wrote:



    I was just informed this can be done even more simply



    	        import

numpy as np

    	        import

holoviews as hv

    	        from

holoviews.operation import histogram

    	hv.extension('bokeh')





    	        data

= np.random.multivariate_normal( (0, 0), [[1, 0.1], [0.1, 1]],
(1000,))

    	        points

= hv.Points(data)

    	points.hist(dimension=['x','y'])





    Thanks,



    Bryan

  --

  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 bokeh+un...@continuum.io.

  To post to this group, send email to bo...@continuum.io.

  To view this discussion on the web visit [https://groups.google.com/a/continuum.io/d/msgid/bokeh/B2397DB4-A496-44B3-8FA5-C636FA6C7E30%40continuum.io](https://groups.google.com/a/continuum.io/d/msgid/bokeh/B2397DB4-A496-44B3-8FA5-C636FA6C7E30%40continuum.io?utm_medium=email&utm_source=footer).

  For more options, visit [https://groups.google.com/a/continuum.io/d/optout](https://groups.google.com/a/continuum.io/d/optout).

⌁ Link to post

FabianJul 07, 2017 #8

Hello Philipp,

thank you for the quick information! No problem, I will check how I can do this.

Best Regards
Fabian

···

On 07/06/2017 04:56 PM, philipp.jfr@gmail.com wrote:

Hi Fabian,

We used to have support for KDE plots in HoloViews via the matplotlib/seaborn compatibility that used to be in bokeh but that has since been dropped. We are hoping to add Elements representing 1D and 2D KDEs to HoloViews in one of the upcoming releases but can't promise how soon that will happen. For now you could compute the KDE yourself and plot it as a Curve or Area.

All the Best,

Philipp

On Thursday, 6 July 2017 15:49:06 UTC+1, Fabian Braennstroem wrote:

    Hello Bryan,

    do you have an idea, if there is in option for a kde (see the
    'kind' of jointplot in seaborn) plot as well? I could not figure
    this out yet; actually it does not look like that it exists.

    Thank you in advance!
    Fabian

    On 06/30/2017 07:21 PM, Bryan Van de ven wrote:

    Another cool updated version, in case you want multiple
    distributions on one plot:

    On Jun 30, 2017, at 12:16, Bryan Van de ven <bry...@continuum.io >>> <javascript:>> wrote:

    I was just informed this can be done even more simply

    import numpy as np
    import holoviews as hv
    from holoviews.operation import histogram
    hv.extension('bokeh')

    data = np.random.multivariate_normal((0, 0), [[1, 0.1], [0.1,
    1]], (1000,))
    points = hv.Points(data)

    points.hist(dimension=['x','y'])

    Thanks,

    Bryan

    On Jun 30, 2017, at 12:11, Bryan Van de ven >>>> <bry...@continuum.io <javascript:>> wrote:

    Hi,

    I would actually recommend that you check out the new HoloViews
    release:

    http://holoviews.org/

    HoloViews is a very high-level API for data exploration that
    integrates tightly with Bokeh to produce statistical and other
    visualizations with less code. We will be advocating for users
    to use Holoviews (instead of the old bkcharts package) going
    forward.

    Here is an example of a Bokeh joint plot created using Holoviews:

    <image.png>

    On Jun 30, 2017, at 11:33, 'Fabian Braennstroem' via Bokeh >>>>> Discussion - Public <bo...@continuum.io> <javascript:> wrote:

    Hello,

    I would like to get a seaborn jointplot into bokeh.

    The lines below work with the kdeplot, but not with the jointplot.

    import seaborn as sns
    from bokeh import mpl
    import matplotlib.pyplot as plt
    import pandas as pd
    import numpy as np
    from bokeh.io <http://bokeh.io> import show

    df = pd.DataFrame(np.random.randint(0,100,size=(100, 4)),
    columns=list('ABCD'))
    ax = sns.jointplot(x="A", y="B", data=df, kind="kde")
    # ax = sns.kdeplot(df["A"], df["B"])
    show(mpl.to_bokeh())

    Do you have a suggestion how to proceed with this? I assume
    that I need somehow create several bokeh plots out of the
    jointplot...

    Thank you!

    Best Regards

    Fabian

    -- 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 bokeh+un...@continuum.io <javascript:>.
    To post to this group, send email to bo...@continuum.io
    <javascript:>.
    To view this discussion on the web visit
    https://groups.google.com/a/continuum.io/d/msgid/bokeh/cf39bca1-658c-9cf1-d486-d2d61c4155b1%40googlemail.com
    <https://groups.google.com/a/continuum.io/d/msgid/bokeh/cf39bca1-658c-9cf1-d486-d2d61c4155b1%40googlemail.com>\.
    For more options, visit
    https://groups.google.com/a/continuum.io/d/optout
    <https://groups.google.com/a/continuum.io/d/optout>\.

    -- 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 bokeh+un...@continuum.io <javascript:>.
    To post to this group, send email to bo...@continuum.io
    <javascript:>.
    To view this discussion on the web visit
    https://groups.google.com/a/continuum.io/d/msgid/bokeh/B2397DB4-A496-44B3-8FA5-C636FA6C7E30%40continuum.io
    <https://groups.google.com/a/continuum.io/d/msgid/bokeh/B2397DB4-A496-44B3-8FA5-C636FA6C7E30%40continuum.io?utm_medium=email&utm_source=footer>\.
    For more options, visit
    https://groups.google.com/a/continuum.io/d/optout
    <https://groups.google.com/a/continuum.io/d/optout>\.

--
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 bokeh+unsubscribe@continuum.io <mailto:bokeh+unsubscribe@continuum.io>.
To post to this group, send email to bokeh@continuum.io <mailto:bokeh@continuum.io>.
To view this discussion on the web visit https://groups.google.com/a/continuum.io/d/msgid/bokeh/c119d7bc-e000-4c53-8ee7-64521a705cdf%40continuum.io <https://groups.google.com/a/continuum.io/d/msgid/bokeh/c119d7bc-e000-4c53-8ee7-64521a705cdf%40continuum.io?utm_medium=email&utm_source=footer>\.
For more options, visit https://groups.google.com/a/continuum.io/d/optout\.

⌁ Link to post

systemJul 21, 2021 #9

⌁ Link to post

End of archived discussion · Replies are disabled