reorganize the checking condition & message

take out z.sum() condition as it is not used

Ryan Magee authored 3 years ago and Patrick Godwin committed 3 years ago

bankchisq: these changes implement bankchisq at the filtering and background collection level. There is no impact on the LR at this time.

- if an SNR, \chi^2 PDF is all NaN, skip it instead of crashing

- teach plot_likelihood_ratio_pdf() how to skip extinction model if it cannot be constructed

- sort of:  it now only shows quantities related to noise rates.  a new plot will be constructed for signal related quantities.

- switch numerator and denominator PDFs to a rate.BinnedLnPDF + snglcoinc.LnLRDensity() -based implementation
- add trigger rate tracking feature to denominator, taking advantage of new high-speed coincidence rate machinery in snglcoinc
- add signal rate tracking feature to numerator, taking advantage of pre-computed P(instruments) factors in SNR PDF look-up table
- switch ranking statistic distribution to a rate.BinnedLnPDF -based implementation
- remove search_summary tables, attach all livetime tracking to the trigger rate data in the denominator object

this is a work in progress.  some notable features that are still missing:

- --min-log-L cut based candidate rate reduction feature in gstlal_inspiral is disabled
- online mode for gstlal_inspiral is non-functional

- temporary hack, remove after O2

- introduce internal cacheentry class to hold entry information
- remove limit on cache size
- when initializing, move loop over instruments into .add_to_cache()
- add min_instruments parameter to the cache key
- in P_instruments_given_signal(), handle off instruments better
- record P(instruments | signal) in the SNRPDF cache
- enables generalization of ranking statistic away from fixed P(instruments | signal) marginalized over horizon distance history

- remove ifo_snr keyword argument, code already had sngls
- add special case for a sngls list with just one entry to reproduce behaviour of ifo_snr keyword parameter
- generalize to 1D case
- indicate horizon distance ratio in figure title
- also, correct the first argument in the call to this function from lvalert_background_plotter

- this code uses double quotes for strings
- adjust whitespace

- ylo and yhi are in axes co-ordinates, not data co-ordinates
- and the default is to draw from top to bottom so no need to set them at all

- set font sizes and DPI to make plots consistent across programs
- also fixes bonkers aspect ratio of rates pie charts produced by some programs

- do not use linspace for x co-cordinates, use actual times of measurements to ensure that all features are reproduced on the plot
- get colours from plotutils
- change x axis to time from reference rather than absolute time to clean up the presentation
- explain what masses the horizon distance is for in the title
- change x axis major ticks to 1/2 hour intervals (human units)
- increase density of y axis ticks to help facilitate visual estimates of the values

- faster, makes smaller vector format output

- related to ranking statistic rework

- property is no longer being used by anything except plotfar.py
- re-implement as function in plotfar.py

- fap_after_trials()
- FAPFAR.ccdf_from_rank()
- FAPFAR.fap_from_rank()
- FAPFAR.rank_from_fap()
- FAPFAR.far_from_rank()
- FAPFAR.rank_from_far()

- there is a long-term plan to replace the use of segment lists for tracking instrument duty cycle in ranking statistic PDFs with horizon distances that go to 0 during times when instruments are off.  some code needs to be tweaked to handle this.
- refs #2668

- disregard min value of zero-lag curve if provided.