RootInteractive 0.0.25

No project description provided

RootInteractive

Code for interactive visualization of multidimensional data in ROOT or native Python formats (Panda, numpy) Support for ROOT data structures:

• THn - for multi-dimensional histograms
• TTree and TTreeFormula, Aliases ..
• TFormula, resp. any static Root/AliRoot functions

No python packages dependencies

• ROOT
• aliBuild, AliRoot
  • only one library libSTAT needed - to be written as standalone package ...

RootInteractive content:

• Visualization part wrappers

  • client/server application (Jupyter, Bokeh, bqplot (in future))
  • client application - (Bokeh standalone)

• Machine learning part (plans)

  • We would like to provide set of wrappers which will simplify/and user analysis using tabular data (trees,csv, pandas ...) error estimates, robust statistic, handling of extrapolation errors set of predefined (parametrizable layouts - e.g autonecoders). Similar approach was choosen recently by GOOGLE (announcement -09.04.2019) and Microsoft (announcement 02.05.2019)

• Integration of the ML part with visualization tools

  • Similar approach as was chosen by Ilastic project

RootInteractive Information

• RootInteractive github (source code)
  • https://github.com/miranov25/RootInteractive
  • JIRA: https://alice.its.cern.ch/jira/browse/PWGPP-485
• RootIteractive tutorial github
  • https://github.com/miranov25/RootInteractiveTest
  • Mostly example Jupyter notebooks using Alice data
  • JIRA https://alice.its.cern.ch/jira/browse/PWGPP-532
• Data server at CERN
  • https://rootinteractive.web.cern.ch/RootInteractive/data/
  • data on eos:
    • /eos/user/r/rootinteractive/www/testData/
    • /eos/user/r/rootinteractive/www/data/
• Documentation server at CERN
  • https://rootinteractive.web.cern.ch/RootInteractive/html/
  • /eos/user/r/rootinteractive/www/html/

Detailed description for ALICE user in JIRA

For ALICE CERN see detailed description in issue tracker in JIRA:

• Jupyter notebooks for interactive n-dimensional analysis. Ipywidgets + bokeh. RootInteractive https://alice.its.cern.ch/jira/browse/PWGPP-485
• ROOT tabular metadata description
• ....

Galery:

• Unit test dashboards:

  • source:
    • dashboard: https://rootinteractive.web.cern.ch/RootInteractive/testData/JIRA/PWGPP-532/test_bokehDrawSAArray.html
    • https://github.com/miranov25/RootInteractive/blob/master/RootInteractive/InteractiveDrawing/bokeh/test_bokehDrawSA.py
  • source: https://github.com/miranov25/RootInteractive/blob/master/RootInteractive/InteractiveDrawing/bokeh/test_bokehDrawArray.py
    • dashborad: https://rootinteractive.web.cern.ch/RootInteractive/testData/JIRA/PWGPP-532/test_BokehDrawArray_DraFromArray.html

• Real use case - using C++ interface - multidimensional dEdx calibration - video example:

  • video: https://drive.google.com/file/d/1Uo2IhIWo94egaKIBAdjqTRsoTyvjKdJv/view , https://drive.google.com/file/d/1HcZ9jFhaofdiAv62aTiHllhbGjdaCbD-/view
  • source code: https://github.com/miranov25/RootInteractiveTest/blob/master/JIRA/PWGPP-485/dEdxPerformance.C

• dashboards created by ROOT maco (dEdxPerformance.C):

  • https://rootinteractive.web.cern.ch/RootInteractive/testData/JIRA/PWGPP-538/alice/data/2015/LHC15o/pass1/40MeV_width/dedxPtElPi_0.html
  • https://rootinteractive.web.cern.ch/RootInteractive/testData/JIRA/PWGPP-538/alice/data/2018/LHC18q/pass1/40MeV_width/dedxPtElPi_0.html
  • https://rootinteractive.web.cern.ch/RootInteractive/testData/JIRA/PWGPP-538/alice/data/2018/LHC18q/pass1/40MeV_width/dedxPtElPi_0.html