pipeline.qa.scorecalculator

Created on 9 Jan 2014

@author: sjw

Functions

calc_flags_per_agent(summaries[, scanids])

Calculate flagging statistics per agents.

calc_frac_newly_flagged(summaries[, agents, …])

Calculate fraction of data that is newly flagged, i.e. exclude pre-existing flags (assumed to be represented in first summary).

calc_frac_total_flagged(summaries[, agents, …])

Calculate total fraction of data that is flagged.

calc_vla_science_frac_total_flagged(summaries)

Calculate total fraction of vla science data that is flagged.

countbaddelays(m, delaytable, delaymax)

Parameters
  • m – measurement set object

direction_recover(ra, dec, org_direction)

generate_metric_mask(context, result, cs, mask)

Generate boolean mask array for metric calculation in score_sdimage_masked_pixels.

linear_score(x, x1, x2[, y1, y2])

Calculate the score for the given data value, assuming the score follows a linear gradient between the low and high values.

linear_score_fraction_newly_flagged(self, …)

linear_score_fraction_unflagged_newly_flagged_for_intent(…)

log_qa(method)

Decorator that logs QA evaluations as they return with a log level of INFO for scores between perfect and ‘slightly suboptimal’ scores and WARNING for any other level.

score_applycal_agents(self, *args, **kw)

score_applycmds_exist(self, *args, **kw)

score_bands(self, *args, **kw)

score_bwswitching(self, *args, **kw)

score_caltables_exist(self, *args, **kw)

score_checksources(self, *args, **kw)

score_contiguous_session(self, *args, **kw)

score_data_flagged_by_agents(ms, summaries, …)

Calculate a score for the agentflagger summaries based on the fraction of data flagged by certain flagging agents.

score_derived_fluxes_snr(self, *args, **kw)

score_ephemeris_coordinates(self, *args, **kw)

score_file_exists(self, *args, **kw)

score_flagged_vla_baddef(self, *args, **kw)

score_flagging_view_exists(self, *args, **kw)

score_flags_exist(self, *args, **kw)

score_fluxservice(result)

If the primary FS query fails and the backup is invoked, the severity level should be BLUE (below standard; numerically, on its own, 0.9).

score_fraction_newly_flagged(self, *args, **kw)

score_gfluxscale_k_spw(self, *args, **kw)

score_images_exist(self, *args, **kw)

score_missing_bandpass_snrs(self, *args, **kw)

score_missing_derived_fluxes(self, *args, **kw)

score_missing_intents(self, *args, **kw)

score_missing_phase_snrs(self, *args, **kw)

score_missing_phaseup_snrs(self, *args, **kw)

score_mom8_fc_image(mom8_fc_name, peak_snr, …)

Check the MOM8 FC image for outliers above a given SNR threshold.

score_ms_history_entries_present(self, …)

score_ms_model_data_column_present(all_mses, …)

Give a score for a group of mses based on the number with modeldata columns present.

score_mses_exist(self, *args, **kw)

score_multiply(self, *args, **kw)

score_number_antenna_offsets(self, *args, **kw)

score_online_shadow_template_agents(self, …)

score_parallactic_range(self, *args, **kw)

score_path_exists(self, *args, **kw)

score_phaseup_mapping_fraction(self, *args, **kw)

score_polintents(self, *args, **kw)

score_poor_bandpass_solutions(self, *args, **kw)

score_poor_phase_snrs(self, *args, **kw)

score_poor_phaseup_solutions(self, *args, **kw)

score_refspw_mapping_fraction(self, *args, **kw)

score_science_spw_names(mses, …)

Check that all MSs have the same set of spw names.

score_sd_line_detection(self, *args, **kw)

score_sd_line_detection_for_ms(self, *args, **kw)

score_sd_skycal_elevation_difference(self, …)

score_sdimage_masked_pixels(self, *args, **kw)

score_sdtotal_data_flagged(self, *args, **kw)

score_sdtotal_data_flagged_old(self, *args, **kw)

score_setjy_measurements(self, *args, **kw)

score_total_data_flagged(self, *args, **kw)

score_total_data_flagged_vla(self, *args, **kw)

score_total_data_flagged_vla_bandpass(self, …)

score_total_data_vla_delay(self, *args, **kw)

score_tsysspwmap(self, *args, **kw)

score_vla_agents(self, *args, **kw)

score_vla_flux_residual_rms(self, *args, **kw)

score_vla_science_data_flagged_by_agents(ms, …)

Calculate a score for the agentflagger summaries based on the fraction of VLA science data flagged by certain flagging agents.

score_wvrgcal(self, *args, **kw)

Classes

AgentStats(name, flagged, total)