notebook.community

Edit and run 





In [1]:



    


import holoviews as hv
hv.notebook_extension('bokeh')



    


















    




















    




















  
  























In [2]:



    


from lsst.daf.persistence import Butler

repo_path = '/datasets/hsc/repo/rerun/RC/w_2018_04/DM-13256/'
butler = Butler(repo_path)



    











In [3]:



    


from explorer.catalog import CoaddCatalog, VisitCatalog, MultiMatchedCatalog
from explorer.utils import get_visits

tract = 9615
filt = 'HSC-I'
dataId = {'tract':tract, 'filter':filt}
catalog = CoaddCatalog(butler, dataId)
# visit_catalogs = [VisitCatalog(butler, {'tract': tract, 'filter':filt, 'visit':v}, name=v) for v in get_visits(butler, tract, filt)]
# matched_catalog = MultiMatchedCatalog(catalog, visit_catalogs, match_registry='t9615_demo.h5')



    











In [4]:



    


from explorer.functors import MagDiff, Seeing, DeconvolvedMoments, Column

magdiff = MagDiff('modelfit_CModel', 'base_PsfFlux')
magdiff_gauss = MagDiff('base_GaussianFlux', 'base_PsfFlux')

funcs = {'cmodel':magdiff, 'gauss':magdiff_gauss, 'count': Column('base_InputCount_value'), 'seeing':Seeing()}
matched_funcs = {'guass':magdiff_gauss, 'seeing':Seeing()}



    











In [5]:



    


from explorer.dataset import QADataset

flags = ['calib_psfUsed', 'qaBad_flag',
 'merge_measurement_i',
 'merge_measurement_r',
 'merge_measurement_z',
 'merge_measurement_y',
 'merge_measurement_g',
 'base_Footprint_nPix_flag',
 'base_PixelFlags_flag_inexact_psfCenter']

data = QADataset(catalog, funcs, flags=flags)
# matched_data = QADataset(matched_catalog, matched_funcs, flags=flags) #memory issues for demo



    











In [6]:



    


# Calculate dataframe; see how long it takes.
%time data.df.head()



    


















    






/ssd/lsstsw/stack3_20171021/python/miniconda3-4.3.21/lib/python3.6/site-packages/dask/local.py:271: RuntimeWarning: divide by zero encountered in log10
  return func(*args2)
/ssd/lsstsw/stack3_20171021/python/miniconda3-4.3.21/lib/python3.6/site-packages/dask/local.py:271: RuntimeWarning: invalid value encountered in log10
  return func(*args2)










    






CPU times: user 9.49 s, sys: 1.66 s, total: 11.1 s
Wall time: 19.3 s










    
Out[6]:











  
    

      
      base_Footprint_nPix_flag
      base_PixelFlags_flag_inexact_psfCenter
      calib_psfUsed
      cmodel
      count
      dec
      gauss
      label
      merge_measurement_g
      merge_measurement_i
      merge_measurement_r
      merge_measurement_y
      merge_measurement_z
      patchId
      qaBad_flag
      ra
      seeing
      x
    

    

      id
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
    

  
  
    

      42287217204264961
      False
      False
      False
      -0.160032
      7
      -0.095521
      -0.169774
      null
      False
      True
      False
      False
      False
      0,0
      True
      217.132928
      0.807439
      25.519317
    

    

      42287217204264968
      False
      False
      False
      -0.712531
      7
      -0.095429
      -0.631927
      null
      False
      True
      False
      False
      False
      0,0
      True
      217.092807
      0.808911
      22.879350
    

    

      42287217204264972
      False
      False
      False
      -0.000603
      8
      -0.094888
      0.006244
      star
      False
      True
      False
      False
      False
      0,0
      True
      217.058166
      0.773655
      22.771899
    

    

      42287217204264990
      False
      False
      False
      -0.316413
      8
      -0.095571
      -0.345569
      null
      False
      True
      False
      False
      False
      0,0
      True
      216.998380
      0.766290
      24.457679
    

    

      42287217204264993
      False
      False
      False
      -0.429495
      7
      -0.095523
      -0.314505
      null
      False
      True
      False
      False
      False
      0,0
      True
      217.068641
      0.755195
      25.700984
    

  




















In [7]:



    


from explorer.plots import FilterStream, multi_scattersky
filter_stream = FilterStream()
multi_scattersky(data.ds, filter_stream=filter_stream, width=900, height=300)



    


















    
Out[7]:

You can use the flags you defined the dataset with to select points in the above plots, as long as you pass the same FilterStream object.





In [8]:



    


from explorer.plots import FlagSetter
import parambokeh

flag_setter = FlagSetter(filter_stream=filter_stream, flags=data.flags, bad_flags=data.flags)
parambokeh.Widgets(flag_setter, callback=flag_setter.event, push=False, on_init=True)

Note that the above plots are "datashaded," meaning that they do not contain information about individual points. If you want to see a plot containing individual point information, you may use the .coadd_explore method of QADataset, as follows (note that the same FilterStream is being used, so the above FlagSetter can still be used).





In [9]:



    


%%output max_frames=10000
%%opts Points [width=500, height=500, tools=['hover'], colorbar=True] (cmap='coolwarm', size=4)

coadd_dmap = data.coadd_explore('seeing', filter_stream=filter_stream)
tap = hv.streams.Tap(source=coadd_dmap, rename={'x':'ra', 'y':'dec'})

coadd_dmap



    


















    
Out[9]:

Note how we defined a tap stream. Below, we connect this to a ginga display window such that when you click on the above sky plot, it shows the image below, with a crosshair centered on the closest object currently displayed in the plot.





In [10]:



    


import lsst.afw.display
lsst.afw.display.setDefaultBackend("ginga")

from explorer.display import CoaddDisplay
coadd_display = CoaddDisplay(butler, filt, dims=(500,500))
coadd_display.connect_tap(tap)
coadd_display.embed()



    


















    






 
 






















In [11]:



    


%%opts RGB [width=400, height=400]

from explorer.plots import skyplot, skyplot_layout
skyplots = [skyplot(data.ds.groupby('label'), vdim=d, filter_stream=filter_stream).relabel(d) for d in funcs]
skyplot_layout(skyplots).cols(3)



    


















    
Out[11]:








  

    

      
      




    






      
    

  

  

    

      
      
        

          
          
        

        
        
        
        

    






















In [ ]:

Content source: timothydmorton/qa_explorer

Similar notebooks:

notebook.community | gallery | about