Clone wiki

yt-3.0-artio / functionality

Current 4/7

See for related scripts (+ working; - broken)

  • +Slices - yt-scripts/
  • +Basic Halo finding - yt-scripts/ :
required changes:
hash table filled up too quickly and apply to all particles.
  • +Star formation histories - yt-scripts/ :
required changes:
(1) volume() fails -- data_size-type issue (setting data_size=1 in identify_base_chunk -> fwidth issue) 
(1 fix) *HACK/NEW BUG* volume returns 1 always 
(2) script needs to reference ("stars", "creation_time") in order for ("stars", "particle_age") derived field to be accessible. 
(3) minor modifications -- tmin bin was .99*max lookback instead of 1.01; particle_age is returned in seconds; sign of dt fixed.


Making Simple Plots

  • Slices - Working
  • Probability Distribution Functions - Working
  • Phase Plot - Working, but pc = PlotCollection(pf, 'c') fails because ARTIOGeometryHandler needs attributes find_max-->find_max_cell_location-->grid levels). But, you can set the halo center manually.
  • 1D Histogram - Working for n vs. T
  • Projections - Working
  • Radial profiles - Not working, need to RadiusMpc needs to be included in the field
  • Multiple Fields Simultaneously - Working: Density, GasEnergy, Pressure, Temperature work; but Entropy & Vorticity need to be added
  • Accessing & Modifying Plots Directly - Have not yet tested, but should be fine
  • Off-Axis Slicing - Not working: pf.h.find_max() is a problem; pf.h.sphere() takes in pre-defined center (but do not need to define the argument with center='..') ; AngularMomentumVector needs to be fed as a derived quantity (?)
  • Volume Rendering - Not working: in in AttributeError: 'ARTIOGeometryHandler' object has no attribute 'grids'

Calculating Dataset Information

  • Average Field Value - Working
  • Mass Enclosed in a Sphere - Not working. artio doesn't have find_max. [ traceback]
  • Gobal Phase Plot - Not working, same error as above. [ traceback]
  • Radial Velocity Profile - Not working. "Particle support not implemented." [ traceback]
  • Simulation Analysis - Not working. There is a "simulation" method that doesn't appear to be set up for ART code. It appears to belong to SimulationTimeSeries, but neither is in the API listing.
  • Time Series Analysis - Working

A Few Complex Plots

  • Multi-Plot - Not working: problem in defining multiple keyword argument 'center' in add_slice() func
  • Multi-Width Image- Working
  • Multi-Plot Slice and Projections - Not working: pf.h.slice() function does not recognize keyword argument in fields: VelocityMagnitude also needs to be added on the field
  • Multi-Plot Multi-Panal - Not working: same issue as Multi-Plot
  • Advanced Multi-Plot Multi-Panel - Working
  • Projecting Off-Axis - Not working: projcam.snapshot()
  • Projecting Off-Axis with a Colorbar - Not working: same as above
  • Thin-Slice Projections -Not working: pc.add_projection, complaining about data_source=region set by pf.h.region() - try with another data container
  • Plotting Particles over Fluids - Not working: needs particle data
  • Overplotting Velocity Vectors - Working
  • Overplotting Contours - Working
  • Styling Radial Profile Plots - Not working
  • Moving a Volume Rendering Camera - Not working: in in AttributeError: 'ARTIOGeometryHandler' object has no attribute 'grids'
  • Zooming into an Image - Working
  • Opaque Volume Rendering - Not working: in in AttributeError: 'ARTIOGeometryHandler' object has no attribute 'grids'
  • Downsampling Data for Volume Rendering - Not working: in in AttributeError: 'ARTIOGeometryHandler' object has no attribute 'grids'

Cosmological Analysis

Constructing Data Objects

Advanced Recipes

Parallelism in yt

  • add a line " from yt.config import ytcfg; ytcfg["yt","serialize"]="False" " at the top of the line.
  • mpirun -np 2 python did not run: parallel (tried on BDM); error message: issue pc.add_projection() -- unexpected keyword argument 'source'
  • testing parallel projection routine
    • python2.7 -- ran and took about 37 sec
    • mpirun -np 1 python - ran and took about 37 sec
    • mpirun -np 2 python - ran and took about 38 sec

Position info for grid data

Manipulating particle data

  • creating particle fields