Fusion Data Platform (FDP) is a data framework in Python for magnetic fusion experiments. FDP streamlines data discovery, analysis methods, and visualization.

• Github repository: https://github.com/Fusion-Data-Platform/fdp
• Documentation: http://Fusion-Data-Platform.github.io/

• Integrate data sources, data, and analysis methods in a single, extensible data object
  • Streamline data access for multiple facilities, diagnostics, and databases
  • Organize data and analysis methods in an intuitive object-oriented framework
• Promote collaborative code development and reduce inefficient code duplication
• Reduce barriers to entry for new students and scientists
• Free components and flexible usage
  • Python, Numpy, Matplotlib, Github, etc.
  • Platform-independent: desktop Mac/PC, Linux cluster
  • Command-line tool or import into code

Initiate an FDP session:

>>> import fdp
>>> nstxu = fdp.nstxu()

List diagnostics:

>>> dir(nstxu.s141000)
['bes', 'chers', 'equilibria', 'filterscopes', 'magnetics', 'mpts', 'mse', 'usxr']

View logbook entries:

>>> nstxu.s141000.logbook()

List signals:

>>> nstxu.s141000.equilibria.efit02.listSignals()
['psirz', 'qpsi', 'shot', 'userid', 'wmhd']
>>> nstxu.s204620.chers.listSignals()
['ft', 'nc', 'ti', 'valid', 'vt']

Plot a signal:

>>> nstxu.s141000.mpts.te.plot()

List methods for a signal:

>>> nstxu.s141000.bes.d1ch01.listMethods()
['animate', 'fft', 'loadConfig', 'plotfft']

Load shots for an XP:

>>> xp1013 = nstxu.filter_shots(xp=1013)
>>> dir(xp1013)
['s141382', 's141383', 's141384', 's141385', 's141386', 's141387', 
's141388', 's141389', 's141390', 's141391', 's141392', 's141393', 
's141394', 's141395', 's141396', 's141397', 's141398', 's141399', 
's141400', 's141401', 's141402', 's141403', 's141404', 's141405', 
's141406', 's141407', 's141408', 's141409', 's141410', 's141411', 
's141412', 's141413', 's141414']

As the examples above illustrate, the FDP data object is organized like this:

<machine>.<shot>.<diagnostic>.<signal>.<method>

or, for diagnostic sub-containers like spline profiles and x-ray arrays:

<machine>.<shot>.<diagnostic>.<sub-container>.<signal>.<method>

• David R. Smith, U. Wisconsin-Madison
• Kevin Tritz, The Johns Hopkins U.
• Howard Yuh, Nova Photonics

PPPL cluster support from John Schmitt