Recpies for Powerspectrum

[kashish2210]

this includes various methods of plotting: [powerspectrum and avg power spectrum]

basis setup:

using Plots
events = readevents("ni1200120104_0mpu7_cl.evt", sort=true)
lc = create_lightcurve(events, 0.1)

plotting :

plot(events, segment_size=256.0, dt=0.01, norm="frac", show_noise=true)

plot(events, segment_size=256.0, show_noise=true, axis_limits=[1,1/10])

# Create the power spectra objects
ps2 = AveragedPowerspectrum(events, 256.0; dt=0.01, norm="frac") 
ps3 = AveragedPowerspectrum(events, 256.0; dt=0.01, norm="leahy")
ps4 = AveragedPowerspectrum(events, 256.0; dt=0.01, norm="abs")
plot([ps2, ps3, ps4], 
     labels=[ "Fractional", "Leahy", "Absolute"], 
     colors=[ :red, :green, :orange], 
     linewidth=2.0,
     show_noise=true)

plot(events, segment_size=256.0, dt=0.01, norm="frac", show_noise=true, show_errors=true, freq_mult=true)

plot(events, segment_size=256.0, dt=0.01, norm="frac", show_noise=true, log_scale=false)

# you can use drawstyle if u want like
plot(events, segment_size=256.0, dt=0.01, norm="frac", drawstyle=:scatter)

plot(lc, segment_size=256, norm="frac", show_noise=true)

plot(lc, segment_size=128, norm="frac", show_noise=true, show_errors=true, freq_mult=true, log_scale=true, drawstyle=:steppost)

ps = AveragedPowerspectrum(events, 256.0, norm="frac")
plot(ps, show_noise=true, meanrate=ps.mean_rate)

plot(ps, show_noise=true, show_errors=true, subtract_noise=true, meanrate=ps.mean_rate)
plot(ps, freq_mult=true, show_noise=true, meanrate=ps.mean_rate)

ps1 = AveragedPowerspectrum(events, 128.0, norm="frac")
ps2 = AveragedPowerspectrum(events, 256.0, norm="frac") 
ps3 = AveragedPowerspectrum(events, 512.0, norm="frac")

plot([ps1, ps2, ps3], labels=["128s", "256s", "512s"])

ps_orig = Powerspectrum(lc, norm="frac")
lc_rebinned = rebin(lc, 100)
ps_rebin = Powerspectrum(lc_rebinned, norm="frac")

# Use vector syntax
plot([ps_orig, ps_rebin], 
     labels=["Original", "Rebinned"], 
     colors=[:blue, :red],
     show_noise=true)

plot(events,segment_size=256.0, norm="frac", freq_mult=true)

events_soft = filter_energy(e -> 30 <= e <= 200, events)    # ~0.3-2 keV
events_medium = filter_energy(e -> 200 <= e <= 800, events) # ~2-8 keV  
events_hard = filter_energy(e -> 800 <= e <= 1200, events)  # ~8-12 keV

# Now this should work:
ps_soft = AveragedPowerspectrum(events_soft, 256.0, norm="frac", dt=0.1)
ps_medium = AveragedPowerspectrum(events_medium, 256.0, norm="frac", dt=0.1)
ps_hard = AveragedPowerspectrum(events_hard, 256.0, norm="frac", dt=0.1)

plot([ps_soft, ps_medium, ps_hard], 
     labels=["Soft", "Medium", "Hard"], 
     colors=[:blue, :green, :red])
plot([ps_soft, ps_medium, ps_hard], segment_size=256.0, norm="frac",
     colors=[:blue, :green, :red], freq_mult=true)

powerlaw_model(f) = 100.0 * f^(-2.0) + 2.0
plot(ps, powerlaw_model, freq_mult=true, log_scale=true)

plot(ps, Val(:qpo), qpo_range=(0.1, 50), significance_level=5.0)

tagin @matteobachetti @stefanocovino @fjebaker for review

#33

[codecov-commenter]

Codecov Report

❌ Patch coverage is 14.56311% with 440 lines in your changes missing coverage. Please review.
✅ Project coverage is 63.55%. Comparing base (2a7829d) to head (c86ff8d).
⚠️ Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
src/plotting/plots_recipes_powerspectrum.jl	0.00%	201 Missing ⚠️
src/powerspectrum.jl	0.00%	190 Missing ⚠️
src/fourier.jl	33.33%	40 Missing ⚠️
src/gti.jl	87.50%	5 Missing ⚠️
src/lightcurve.jl	83.33%	4 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##             main      #59       +/-   ##
===========================================
- Coverage   88.06%   63.55%   -24.51%     
===========================================
  Files           5        7        +2     
  Lines        1081     1534      +453     
===========================================
+ Hits          952      975       +23     
- Misses        129      559      +430     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[matteobachetti]

When freq_multi is true, the Poisson noise level should always be subtracted

[kashish2210]

Member

yup I have removed the if condition @matteobachetti