Hi Bryan,
thanks for the quick reply. It seems to work (Raspberry Pi is the server, MacBook Pro is the client). I have the bokeh-server running in one ssh terminal, and I call the python graph script example in a second terminal. I just need to be patient while the page loads in Safari on my MacBook. Here's a screen capture of it working:
http://youtu.be/da_loaEaytc
I've modified the color_scatter.py example as follows:
import numpy as np
from six.moves import zip
from bokeh.plotting import *
# new line added: -js
from bokeh.session import Session
N = 4000
# new line added: -js
session = Session(load_from_config=False)
x = np.random.random(size=N) * 100
y = np.random.random(size=N) * 100
radii = np.random.random(size=N) * 1.5
colors = ["#%02x%02x%02x" % (r, g, 150) for r, g in zip(np.floor(50+2*x), np.floor(30+2*y))]
# change line: -js
output_server("color_scatter",url='http://192.168.1.136:5006')
scatter(x,y, radius=radii, radius_units="data",
fill_color=colors, fill_alpha=0.6,
line_color=None, tools="select,pan,wheel_zoom,box_zoom,reset,previewsave", name="color_scatter_example")
show() # open a browser
all the best,
James
On Wednesday, September 24, 2014 10:01:56 AM UTC-4, Bryan Van de ven wrote:
Hi James,
I think you have exposed a need for better documentation about a core component. I mentioned before that "Running the server does not create any plots, just makes a place for plots to be stored" but that's only half the story, I left out the other half, which is that once you have a server running, you can run all the scripts in the example directory, separately, in a new shell. *Those* scripts store plots on the server when you run them, and they should open a browser for you to the right link by default. Let's run through a complete example:
* In a first terminal, do this:
bryan@laptop (git:feature/selection) ~/work/bokeh $ bokeh-server
Bokeh Server Configuration
==========================
python version : 2.7.8
bokeh version : 0.6.1.dev.20140923
listening : 127.0.0.1:5006
backend : shelve
python options : debug:OFF, verbose:OFF, filter-logs:OFF, multi-user:OFF
js options : splitjs:OFF, debugjs:OFF
data-directory : None
zmqpub starting
Starting Bokeh plot server on port 5006...
View http://127.0.0.1:5006/bokeh to see plots
* Leave the server running, in a NEW terminal, do this:
bryan@laptop (git:feature/selection) ~/work/bokeh/examples/plotting/server $ python color_scatter.py
Using saved session configuration for http://localhost:5006/
To override, pass 'load_from_config=False' to Session
This should automatically open a browser to the single document page for this plot, it will look something like this:
http://localhost:5006/bokeh/doc/<big long UUID>/<another big UUID>
However, maybe it does not work if the server is running on a remote computer. However in any case, now if you navigate to the Bokeh server main page:
http://localhost:5006/bokeh/
You should see the color scatterplot in the list of plots. Note, above it has "localhost" but if you are passing the --ip parameter to the server, then then you would use whatever ip address you specified for the server to listen on instead of "localhost". However, Bokeh outputs to the localhost server by default, so you will also have to modify the "color_scatterplot.py" example to point it at your specific server address by passing in a "url" parameter to the output_server(...) function (there's no way for it to magically know where you are running your server). I would suggest trying everything on a single machine with the default server settings, first.
Let me know if this does not work for you.
Byran
On Sep 24, 2014, at 9:46 AM, onni...@gmail.com wrote:
> Hi,
>
> I'm running Python2.7 on an ArchLinux distro (no X) on a Raspberry Pi and am trying to get bokeh-server working. Unfortunately, I only seem to get the same message "Bokeh Documents for default user" on my browser (running on a different computer). When I fire up bokeh-server I get the following:
>
> [jasmith@alarmpi server]$ bokeh-server --ip 192.168.1.136
>
> Bokeh Server Configuration
> ==========================
> python version : 2.7.8
> bokeh version : 0.6.0
> listening : 192.168.1.136:5006
> backend : shelve
> python options : debug:OFF, verbose:OFF, filter-logs:OFF, multi-user:OFF
> js options : splitjs:OFF, debugjs:OFF
> data-directory : None
>
> zmqpub starting
>
> Starting Bokeh plot server on port 5006...
> View http://192.168.1.136:5006/bokeh to see plots
>
> Bryan, you suggested downloading some of the example plots from Github. I've done that... now where should I put these example plots on the RPi? How do I tell bokeh-server where to look? I tried
>
> bokeh-server --ip 192.168.1.136 --data-directory /home/jasmith/temp/bokeh-master/examples/plotting/server
>
> but I got the same "Bokeh Documents for default user" message on my browser.
>
> Any suggestions for how to get the bokeh-server to find those example files and display them on the browser would be much appreciated.
>
> all the best,
>
> James
>
> On Saturday, August 16, 2014 12:05:08 PM UTC-4, Bryan Van de ven wrote:
> Hi HS,
>
> Thanks for your interest in bokeh! Running the server does not create any plots, just makes a place for plots to be stored. You need to actually create a plot and store it on the server. You can see lots of examples in the examples/plotting/server directory of the GitHub source tree:
>
> https://github.com/ContinuumIO/bokeh/tree/master/examples/plotting/server
>
> Let us know if we can help more!
>
> Bryan
>
>
> On Aug 16, 2014, at 11:01 AM, chan...@gmail.com wrote:
>
> > Hi,
> > I am new to Anaconda and Bokeh and I just installed a fresh Anaconda 3.4 x64.
> > '3.4.1 |Anaconda 2.0.1 (64-bit)| (default, Jun 11 2014, 17:27:11) [MSC v.1600 64 bit (AMD64)]'
> > I installed bokeh using "conda bokeh"
> >
> > bokeh.server.run()
> > and got:
> > Bokeh Server Configuration
> > ==========================
> > listening : 127.0.0.1:5006
> > backend : shelve
> > python options : debug:OFF, verbose:OFF, filter-logs:OFF, multi-user:OFF
> > js options : splitjs:OFF, debugjs:OFF
> > data-directory : None
> >
> >
> >
> >
> > Starting Bokeh plot server on port 5006...127.0.0.1 - - [16/Aug/2014 18:53:28] "GET / HTTP/1.1" 302 221
> > 127.0.0.1 - - [16/Aug/2014 18:53:28] "GET /bokeh/ HTTP/1.1" 200 479
> > 127.0.0.1 - - [16/Aug/2014 18:53:30] "GET /bokeh/userinfo/ HTTP/1.1" 200 39
> > 127.0.0.1 - - [16/Aug/2014 18:53:46] "GET /bokeh/sub HTTP/1.1" 200 4
> >
> >
> > View http://127.0.0.1:5006/bokeh to see plots
> >
> > I then visited http://localhost:5006/bokeh/ but I see only "Bokeh Documents for defaultuser" and "logout" with no examples appearing.
> >
> > I have also tried to run
> > bokeh-server.exe
> >
> > but it didn't help.
> >
> > Additionally, how do I upload new plot to the server, assuming it is working correctly?
> > Moreover, why do I need redis.io for?
> >
> > Thanks,
> > HS.
> >
> > --
> > 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/a0d1995e-9767-4700-a823-12ca3d799a50%40continuum.io\.
> > 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+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/89b3d159-e97d-4bc9-9265-5ced5e37fbc6%40continuum.io\.
> 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 [email protected].
To post to this group, send email to [email protected].
To view this discussion on the web visit https://groups.google.com/a/continuum.io/d/msgid/bokeh/209e53e5-e009-4c68-bc28-7bf0bee57606%40continuum.io\.
For more options, visit https://groups.google.com/a/continuum.io/d/optout\.
