Inspired by IPython, built with love

IPython-Dashboard

A stand alone, light-weight web server for building, sharing graphs created in IPython. Build for data science, data analysis guys. Aiming at building an interactive visualization, collaborated dashboard, and real-time streaming graph.

Usage

• Install prerequisite

  • install the latest stable IPython-Dashboard: pip install ipython-dashboard --upgrade
  • install redis 2.6+ : install guide
  • install IPython-Dashboard requirements [unneeded sometimes]:
    • cd ~/your python package path/IPython-Dashboard
    • pip install -r requirements.txt

• Config mysql

  • create a user and grant privileges;
  • create tables;

• Create logging path

  • create a folder to store log files. I put it under mnt currently: /mnt/ipython-dashboard/logs
  • make sure the log folder is write-able, using chmod and ls -l to confirm.

  chenshan@mac007:/mnt/ipython-dashboard$ls -l
  total 0
  drwxrwxrwx  9 root  wheel  306 Dec 15 22:09 logs
  

• Config IPython-Dashboard server : IPython-Dashboard/dashboard/config.py

  • app_host='ip_address:port'

• Start redis and IPython-Dashboard server

  chenshan@mac007:~/Desktop/github/IPython-Dashboard$redis-server &
  
  chenshan@mac007:~/Desktop/github/IPython-Dashboard$dash-server --help
  usage: dash-server [-h] [-H HOST] [-p PORT] [-d DEBUG]
  
  Start your IPython-Dashboard server ...
  
  optional arguments:
    -h, --help            show this help message and exit
    -H HOST, --host HOST  server host, default localhost
    -p PORT, --port PORT  server port, default 9090
    -d DEBUG, --debug DEBUG
                          server port, default true
  
  chenshan@mac007:~/Desktop/github/IPython-Dashboard$dash-server
  Namespace(debug=True, host='0.0.0.0', port=9090)
   * Running on http://0.0.0.0:9090/
   * Restarting with reloader
  Namespace(debug=True, host='0.0.0.0', port=9090)
  

• Do your exploring

  • IPython-Dashboard-Tutorial.ipynb : On nbviewer or On github

Goal

• support raw html visualization
• support python object visualization
• Editable
• Real-time fresh when rendering a variable python object
• Can be shared, both public and private [ need password ]
• In the notebook, can share an object to a dashboard [ that's visualize that object in that dashboard ]

Use Case

• exploring in notebook, share/send the result/summary to people, without the details.
• share some data in a private notebook.
• disappointed with the complicated code when drawing a graceful/staic graph using matplotlib/seaborn/mpld3 etc.
• want an interactive graph, allow people to zoom in/out, resize, get hover tips, change graph type easily.
• want a real-time graph.
• want an collaborated graph/dashboard.

Screenshot and Demo

Demo on Youtube
Demo on Youku

Run tests

just run sudo nosetests --with-coverage --cover-package=dashboard under this repo

aaron@aarons-MacBook-Pro:~/Desktop/github/IPython-Dashboard$sudo nosetests --with-coverage --cover-package=dahboard
....
Name                                    Stmts   Miss  Cover   Missing
---------------------------------------------------------------------
dashboard.py                               11      0   100%
dashboard/client.py                         1      0   100%
dashboard/client/sender.py                 11      3    73%   26-27, 33
dashboard/config.py                        11      0   100%
dashboard/server.py                         1      0   100%
dashboard/server/resources.py               0      0   100%
dashboard/server/resources/dash.py         35     10    71%   55-56, 67-69, 82-89
dashboard/server/resources/home.py         40     12    70%   25, 28-30, 83-91
dashboard/server/resources/status.py        8      1    88%   19
dashboard/server/resources/storage.py      15      5    67%   28, 43-47
dashboard/server/utils.py                  31      6    81%   18-22, 29, 43
dashboard/server/views.py                  12      0   100%
---------------------------------------------------------------------
TOTAL                                     176     37    79%
----------------------------------------------------------------------
Ran 4 tests in 3.120s

OK

Change Log

• future

  • front side, databricks style
  • pep 8, code clean up & restructure
  • hover tips
  • edit modal can be resized
  • Share one graph
  • Share one dashboard
  • Presentation mode
  • footer
  • unified message display center
  • SQL Editor
  • login management
  • unified logger and exception report
  • server side log
  • client side log
  • support python3
  • create examples

• V 0.1.6 : [ ]

  • Dashboard

    • re-structure code, follow pep8 style

  • SQL Editor

    • optimize page

• V 0.1.5 : sql-server-log [ current develop version ]

  • Dashboard

    • create 1 example
    • server side log
    • support x-axis as date format
    • research on real-time updated dataframe

  • SQL Editor

    • sql server develop : render sql result as table view

• V 0.1.4 : sql-ui-optimize : [ current stable release ]

  • Dashboard

    • create 1 example
    • hover tips
    • unified message display
    • make the redis-server/dash-server configurable

  • SQL Editor

    • sql editor web UI.

Related Projects & Products

• mpld3
• lighting
• bokeh
• matplotlib
• zeppelin
• yhat
• hue
• plotly
• datadog
• databricks
• nvd3
• c3js
• periscope
• folium
• metabase
• gridstack
• gridster
• dashboards
• js, css, html code style