Error in convertReferenceCatalog: "Failure preparing data"

Support
#1

We are creating a merged reference catalog for WFST by combining Pan-STARRS (g/r/i/z/y) and SDSS (u/g/r/i/z) data. Our goal is to use Pan-STARRS as the primary source and supplement u-band photometry from SDSS.

To do this, we have generated thousands of CSV files, with a schema like:

healpix_id,coord_ra,coord_dec,source_type,sdss_sep_arcsec,ps_id,ps_coord_ra,ps_coord_dec,ps_parent,ps_g_flux,ps_r_flux,ps_i_flux,ps_z_flux,ps_y_flux,ps_i_fluxSigma,ps_y_fluxSigma,ps_r_fluxSigma,ps_z_fluxSigma,ps_g_fluxSigma,ps_coord_ra_err,ps_coord_dec_err,ps_epoch,ps_pm_ra,ps_pm_dec,ps_pm_ra_err,ps_pm_dec_err,ps_footprint,sdss_id,sdss_coord_ra,sdss_coord_dec,sdss_u_flux,sdss_g_flux,sdss_r_flux,sdss_i_flux,sdss_z_flux,sdss_u_fluxErr,sdss_g_fluxErr,sdss_r_fluxErr,sdss_i_fluxErr,sdss_z_fluxErr,ps_g_mag,ps_g_mag_err,ps_r_mag,ps_r_mag_err,ps_i_mag,ps_i_mag_err,ps_z_mag,ps_z_mag_err,ps_y_mag,ps_y_mag_err,sdss_u_mag,sdss_u_mag_err,sdss_g_mag,sdss_g_mag_err,sdss_r_mag,sdss_r_mag_err,sdss_i_mag,sdss_i_mag_err,sdss_z_mag,sdss_z_mag_err
143800,1.2401491934085658,-0.5785334509620012,ps_only,,68220710552823264,1.2401491934085658,-0.5785334509620012,0,0.0130986403673887,0.0162190850824117,0.0172992087900638,0.0176045950502157,0.0177184622734785,0.00955989677459,0.0097915846854448,0.0089629981666803,0.0097286598756909,0.0072385766543447,0.0,0.0,1419873408.0,-0.000759758579079,-0.0074059693142771,0.0033094324171543,0.0032997245434671,0,0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,13.60700007619344,0.6000000400122356,13.374999742179286,0.6000000347305979,13.305000021391912,0.6000000443884193,13.28600052313188,0.6000000693390526,13.279000551756983,0.6000000356044211,,,,,,,,,,
143800,1.2403506985720716,-0.5782874534477238,ps_only,,68240710668420169,1.2403506985720716,-0.5782874534477238,0,0.0003948445082642,0.0007537476485595,0.0009279857040382,0.0010814981069415,0.0011706476798281,1.9196256744180573e-06,0.0006469238433055,0.000416536378907,0.0005976579268462,0.0002181991440011,0.0,0.0,1419834880.0,0.0024005575105547,-0.0075076259672641,0.0053496062755584,0.0116071421653032,0,0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,17.409000367379097,0.6000000140073345,16.707000696540142,0.6000000504711587,16.48121240777254,0.0022459474162469,16.315001213319594,0.5999999861976275,16.229000100505232,0.6000000260550253,,,,,,,,,,
143800,1.240528342891102,-0.579019387292847,ps_only,,68180710770269891,1.240528342891102,-0.579019387292847,0,0.0031508377287536,0.005567061714828,0.0072031673043966,0.0075444234535098,0.0080394493415951,2.4798862796160392e-05,0.004442763980478,0.0030764720868319,0.0041692028753459,0.0017412174493074,0.0,0.0,1419774976.0,-0.0045610321685671,-0.0070206765085458,0.003365577198565,0.006067348178476,0,0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,15.1540005293913,0.6000000596087445,14.53600053279219,0.600000017730073,14.256256867190974,0.0037379422185829,14.2060004810312,0.6000000576106188,14.136999864869004,0.6000000121703924,,,,,,,,,,
143800,1.2405629083846008,-0.5785754401720666,ps_only,,68220710790100381,1.2405629083846008,-0.5785754401720666,0,0.0118256760761141,0.0192852802574634,0.0224504955112934,0.0236606076359748,0.0247073490172624,0.0124066034331917,0.0136537859216332,0.0106574399396777,0.0130753358826041,0.0065351105295121,0.0,0.0,1419872512.0,0.0091399122029542,-0.0064788833260536,0.0032761241309344,0.0032660374417901,0,0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,13.718000675241218,0.6000000387562556,13.187000736031615,0.6000000123444055,13.022000795075792,0.600000054284627,12.96500088803995,0.6000000412301947,12.918000246933373,0.6000000120116724,,,,,,,,,,
143800,1.2408377203169474,-0.5792474266165331,ps_only,,68170710947804234,1.2408377203169474,-0.5792474266165331,0,6.020408909535036e-05,0.000141392374644,0.0002418772637611,0.0003749956085812,0.000443022523541,1.988488520510145e-06,0.000244823313551,7.813631964381784e-05,0.0002072302304441,3.327001104480587e-05,0.0,0.0,1419962624.0,0.0293840561062097,-0.0284562855958938,0.0104350410401821,0.0279433988034725,0,0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,19.451000647823875,0.6000000343305314,18.524000651305187,0.6000000449629148,17.941078004779264,0.00892590707325,17.465000167496978,0.5999999967061043,17.284001105816493,0.6000000477776969,,,,,,,,,,

We then created the config file based on the documentation at ConvertReferenceCatalogConfig:

# The name of the output reference catalog dataset.
config.dataset_config.ref_dataset_name = "ps_and_sdss"

# Gen3 butler wants all of our refcats have the same indexing depth.
config.dataset_config.indexer['HTM'].depth = 7

# Ingest the data in parallel with this many processes.
config.n_processes = 16

# These define the names of the fields from the gaia_source data model:
# https://gea.esac.esa.int/archive/documentation/GDR2/Gaia_archive/chap_datamodel/sec_dm_main_tables/ssec_dm_gaia_source.html
config.ra_name = "coord_ra"
config.dec_name = "coord_dec"

# NOTE: these names have `_flux` appended to them when the output Schema is created,
# while the Gaia-specific class handles the errors.
config.mag_column_list = ["ps_g_mag", "ps_r_mag", "ps_i_mag", "ps_z_mag", "ps_y_mag", "sdss_u_mag", "sdss_g_mag", "sdss_r_mag", "sdss_i_mag", "sdss_z_mag"]
config.mag_err_column_map = {"ps_g_mag":"ps_g_mag_err", "ps_r_mag":"ps_r_mag_err", "ps_i_mag":"ps_i_mag_err", "ps_z_mag":"ps_z_mag_err", "ps_y_mag":"ps_y_mag_err", "sdss_u_mag":"sdss_u_mag_err", "sdss_g_mag":"sdss_g_mag_err", "sdss_r_mag":"sdss_r_mag_err", "sdss_i_mag":"sdss_i_mag_err", "sdss_z_mag":"sdss_z_mag_err"}

We ran the command:

convertReferenceCatalog ps_and_sdss catalog_config.cfg "/data/xxx/pipeline/catalog/final_catalog/*.csv"

However, we encountered the following errors:

lsst.ConvertReferenceCatalogTask INFO: Creating 131072 file locks.
lsst.ConvertReferenceCatalogTask INFO: File locks created.
lsst.ConvertReferenceCatalogTask ERROR: Failure preparing data for: /data/xxx/pipeline/catalog/final_catalog/1438xx.csv
lsst.ConvertReferenceCatalogTask ERROR: Failure preparing data for: /data/xxx/pipeline/catalog/final_catalog/1948xx.csv
lsst.ConvertReferenceCatalogTask ERROR: Failure preparing data for: /data/xxx/pipeline/catalog/final_catalog/2273xx.csv

What could be causing the “Failure preparing data” error? We suspect it might be related to:

  • Missing values (we use nan for non-matched sources)?
  • Missing magnitude errors for SDSS (but we do provide sdss_u_mag_err, etc. in mag_err_column_map)?
  • Unit mismatch (we use radian degrees for RA/Dec)?

Thank you for any guidance!

#2

Then, the following error occurred::

multiprocessing.pool.RemoteTraceback: 
"""
Traceback (most recent call last):
  File "/data/public/lsst_stack_v29_2_0/conda/envs/lsst-scipipe-10.1.0/lib/python3.12/multiprocessing/pool.py", line 125, in worker
    result = (True, func(*args, **kwds))
                    ^^^^^^^^^^^^^^^^^^^
  File "/data/public/lsst_stack_v29_2_0/conda/envs/lsst-scipipe-10.1.0/lib/python3.12/multiprocessing/pool.py", line 51, in starmapstar
    return list(itertools.starmap(args[0], args[1]))
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/data/public/lsst_stack_v29_2_0/conda/envs/lsst-scipipe-10.1.0/share/eups/Linux64/meas_algorithms/g646c943bdb+dbb9921566/python/lsst/meas/algorithms/convertRefcatManager.py", line 156, in _convertOneFile
    fluxes = self._getFluxes(inputData)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/data/public/lsst_stack_v29_2_0/conda/envs/lsst-scipipe-10.1.0/share/eups/Linux64/meas_algorithms/g646c943bdb+dbb9921566/python/lsst/meas/algorithms/convertRefcatManager.py", line 360, in _getFluxes
    fluxErr = fluxErrFromABMagErr(inputData[error_col_name].copy(),
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
TypeError: fluxErrFromABMagErr(): incompatible function arguments. The following argument types are supported:
    1. (magErr: typing.SupportsFloat, mag: typing.SupportsFloat) -> float
    2. (magErr: numpy.ndarray, mag: numpy.ndarray) -> numpy.ndarray
    3. (magErr: numpy.ndarray, mag: numpy.ndarray) -> numpy.ndarray

Invoked with: array([0, 0, 0, ..., 0, 0, 0], shape=(110319,)), array([0, 0, 0, ..., 0, 0, 0], shape=(110319,))
"""

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "/data/public/lsst_stack_v29_2_0/conda/envs/lsst-scipipe-10.1.0/share/eups/Linux64/meas_algorithms/g646c943bdb+dbb9921566/bin/convertReferenceCatalog", line 7, in <module>
    main()
  File "/data/public/lsst_stack_v29_2_0/conda/envs/lsst-scipipe-10.1.0/share/eups/Linux64/meas_algorithms/g646c943bdb+dbb9921566/python/lsst/meas/algorithms/convertReferenceCatalog.py", line 704, in main
    run_convert(args.outputDir, args.configFile, args.fileglob[0])
  File "/data/public/lsst_stack_v29_2_0/conda/envs/lsst-scipipe-10.1.0/share/eups/Linux64/meas_algorithms/g646c943bdb+dbb9921566/python/lsst/meas/algorithms/convertReferenceCatalog.py", line 684, in run_convert
    converter.run(files)
  File "/data/public/lsst_stack_v29_2_0/conda/envs/lsst-scipipe-10.1.0/share/eups/Linux64/meas_algorithms/g646c943bdb+dbb9921566/python/lsst/meas/algorithms/convertReferenceCatalog.py", line 538, in run
    result = worker.run(inputFiles)
             ^^^^^^^^^^^^^^^^^^^^^^
  File "/data/public/lsst_stack_v29_2_0/conda/envs/lsst-scipipe-10.1.0/share/eups/Linux64/meas_algorithms/g646c943bdb+dbb9921566/python/lsst/meas/algorithms/convertRefcatManager.py", line 128, in run
    result = pool.starmap(self._convertOneFile, zip(inputFiles, itertools.repeat(fileLocks)))
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/data/public/lsst_stack_v29_2_0/conda/envs/lsst-scipipe-10.1.0/lib/python3.12/multiprocessing/pool.py", line 375, in starmap
    return self._map_async(func, iterable, starmapstar, chunksize).get()
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/data/public/lsst_stack_v29_2_0/conda/envs/lsst-scipipe-10.1.0/lib/python3.12/multiprocessing/pool.py", line 774, in get
    raise self._value
TypeError: fluxErrFromABMagErr(): incompatible function arguments. The following argument types are supported:
    1. (magErr: typing.SupportsFloat, mag: typing.SupportsFloat) -> float
    2. (magErr: numpy.ndarray, mag: numpy.ndarray) -> numpy.ndarray
    3. (magErr: numpy.ndarray, mag: numpy.ndarray) -> numpy.ndarray

Invoked with: array([0, 0, 0, ..., 0, 0, 0], shape=(110319,)), array([0, 0, 0, ..., 0, 0, 0], shape=(110319,))
#3

It looks like there are columns that need to be floats that are instead integers because the csv reader thinks 0 is an integer. If the error is unknown it should be set to 0.0 rather than 0 to make the csv reader happy.

1 Like
#4

Thank you for the reply — I only just saw this, and I really appreciate you taking the time to respond.

We’ve since resolved the issue internally, but your support means a lot. Thanks again for being part of the community!