A priori flagging miss 0319+415 if told to look for 3C84
Closed this issue · 4 comments
I ran the a-priori flagging step on a dataset which had 0319+415 in it. But I kept the name 3C84 in the input-file from the templace since it's the same source. I got the following output:
2017-11-04 10:49:07 | INFO | Starting pipeline
2017-11-04 10:49:07 | INFO | Running pipeline from: pipeline/
2017-11-04 10:49:07 | INFO | Pipeline version: v0.6.1
2017-11-04 10:49:07 | INFO | Using github branch: master
2017-11-04 10:49:07 | INFO | github last commit: ad51f0d
2017-11-04 10:49:07 | INFO | This log uses UTC times
2017-11-04 10:49:07 | INFO | Inputs from file: inputs.txt
2017-11-04 10:49:07 | INFO | Parameters in inputs file:
2017-11-04 10:49:07 | INFO | fits_path : ./fitsidi/
2017-11-04 10:49:07 | INFO | inbase : TSARP220_C_20171013
2017-11-04 10:49:07 | INFO | targets : 1534+2330
2017-11-04 10:49:07 | INFO | phscals : 1532+2344,1516+1932
2017-11-04 10:49:07 | INFO | fluxcal : 3C286
2017-11-04 10:49:07 | INFO | bpcal : OQ208
2017-11-04 10:49:07 | INFO | ptcal : 3C84
2017-11-04 10:49:07 | INFO | refant : Mk2
2017-11-04 10:49:07 | INFO | Lo_dropout_scans:
2017-11-04 10:49:07 | INFO | manual_flags_a : ./inputfg_a.flags
2017-11-04 10:49:07 | INFO | manual_flags_b : ./inputfg_b_avg.flags
2017-11-04 10:49:07 | INFO | run_importfits : 0
2017-11-04 10:49:07 | INFO | summary_weblog : 0
2017-11-04 10:49:07 | INFO | hanning : 0
2017-11-04 10:49:07 | INFO | ms2mms : 0
2017-11-04 10:49:07 | INFO | flag_0_aoflagger: 0
2017-11-04 10:49:07 | INFO | flag_1_apriori : 1
2017-11-04 10:49:07 | INFO | flag_2a_manual : 0
2017-11-04 10:49:07 | INFO | average_1 : 0
2017-11-04 10:49:07 | INFO | plot_data : 0
2017-11-04 10:49:07 | INFO | flag_2b_manual : 0
2017-11-04 10:49:07 | INFO | init_models : 0
2017-11-04 10:49:07 | INFO | bandpass_0 : 0
2017-11-04 10:49:07 | INFO | flag_3_tfcropBP : 0
2017-11-04 10:49:07 | INFO | delay : 0
2017-11-04 10:49:07 | INFO | gain_0_p_ap : 0
2017-11-04 10:49:07 | INFO | fluxscale : 0
2017-11-04 10:49:07 | INFO | bandpass_1_sp : 0
2017-11-04 10:49:07 | INFO | gain_1_amp_sp : 0
2017-11-04 10:49:07 | INFO | applycal_all : 0
2017-11-04 10:49:07 | INFO | flag_4_rflag : 0
2017-11-04 10:49:07 | INFO | plot_corrected : 0
2017-11-04 10:49:07 | INFO | weblog : 0
2017-11-04 10:49:07 | INFO | Generating empty flags list
2017-11-04 10:49:07 | INFO | Targets: 1534+2330
2017-11-04 10:49:07 | INFO | Phasecals: 1532+2344,1516+1932
2017-11-04 10:49:07 | INFO | Fluxcal: 3C286
2017-11-04 10:49:07 | INFO | Bandpass: OQ208
2017-11-04 10:49:07 | INFO | Pointcal: 3C84
2017-11-04 10:49:07 | INFO | Start flagdata1_apriori
2017-11-04 10:49:09 | INFO | Antennas in MS TSARP220_C_20171013.ms: ['Cm' 'Da' 'De' 'Kn' 'Mk2' 'Pi']
2017-11-04 10:49:10 | INFO | MS has 512 channels
2017-11-04 10:49:10 | INFO | Flagging edge channels *:03;508511
2017-11-04 10:49:24 SEVERE agentflagger:: (file ../../tools/flagging/agentflagger_cmpt.cc, line 39) Exception Reported: Field Expression: No match found for name "3C84"
2017-11-04 10:49:24 SEVERE agentflagger:: (file ../../tools/flagging/agentflagger_cmpt.cc, line 39)+ (near char. 14 in string "1516+1932,3C84,1532+2344,3C286,OQ208,1534+2330")
2017-11-04 10:49:24 SEVERE flagdata:::: Field Expression: No match found for name "3C84"
2017-11-04 10:49:24 SEVERE flagdata::::+ (near char. 14 in string "1516+1932,3C84,1532+2344,3C286,OQ208,1534+2330")
2017-11-04 10:49:24 | INFO | Flagging 5 min from bright calibrators
I think the pipeline should recognize the different source name for 3C84.
I think it is not a good idea to allow for aliases, I can imagine many situations in which that can create confusion. The idea is that allowing 3C84 would only work in that particular context, but would not work in the plots, tables, manual flags, etc.
I included those names in the inputs.txt template to force people change them manually. My idea is to have empty strings in all sources in the inputs file. In fact, I thought I have already changed it, several times. But I always forget.
Right, so my bad misunderstanding the intention of the default values. Probably a good idea to change them.
Perhaps the default sources could instead be reasonable names? Most observations use the
same ones, i.e.
fluxcal = '1331+305'
bpcal = '1407+284'
ptcal = '0319+415'
?