Spectra

[kashish2210]

In this PR, I have updated the specra function to match our event list and light curve ones

so now we can do :

events = readevents("ni1200120104_0mpu7_cl.evt", load_gti=true, sort=true)
lc = create_lightcurve(events, 0.01)
# Calculate PDS from events with 100-second segments
segment_size = 100.0  # seconds
dt = 0.01  # time resolution
pds_result = avg_pds_from_eventlist(events, segment_size, dt; 
                                   norm="frac", 
                                   use_common_mean=true, 
                                   silent=false)

Output:

pds_result = avg_pds_from_lightcurve(lc, 100.0;  # 100-second segments
                                    norm="frac", 
                                    use_common_mean=true, 
                                    silent=false)

output

events2 = readevents("xmm_test.fits", load_gti=true, sort=true)
cs_result = avg_cs_from_eventlists(events, events2, segment_size, dt;
                                  norm="frac",
                                  use_common_mean=true,
                                  power_type="all")
lc2 = create_lightcurve(events2, 0.01)
cs_result = avg_cs_from_lightcurves(lc, lc2, 100.0;
                                   norm="frac",
                                   use_common_mean=true,
                                   power_type="all")

Output:

# some plotting examples, later I will put this in recipes
plot(pds_result.freq[subsample], pds_result.power[subsample],
      xscale=:log10, yscale=:log10,
      xlabel="Frequency (Hz)", ylabel="Power",
      title="Full PDS (Log-Log)", linewidth=1,
      label="PDS")
plot(histogram(log10.(pds_result.power), bins=100,
                      xlabel="log10(Power)", ylabel="Count",
                      title="Power Distribution", alpha=0.7,
                      label="Power Hist"))

Outputs:

For a complete analysis, you may visit this notebook

tagging @matteobachetti @fjebaker @stefanocovino for reviews

[kashish2210]

#33

[codecov-commenter]

Codecov Report

❌ Patch coverage is 82.97872% with 8 lines in your changes missing coverage. Please review.
✅ Project coverage is 88.92%. Comparing base (98935ce) to head (2902dee).
⚠️ Report is 8 commits behind head on main.

Files with missing lines	Patch %	Lines
src/fourier.jl	82.97%	8 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main      #58      +/-   ##
==========================================
- Coverage   89.26%   88.92%   -0.35%     
==========================================
  Files           5        5              
  Lines        1025     1065      +40     
==========================================
+ Hits          915      947      +32     
- Misses        110      118       +8     

☔ 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]

At the very least, there should be a test for every new function.

I suggest a test that verifies that the new spectra function from event lists give the exact same result as the corresponding functions that get the times instead.

In practice we should get something like
spectrum_fun(events) .== spectrum_fun(events.time)

[kashish2210]

this functions

[kashish2210]

I have added test cases @matteobachetti :)
I upgraded the format to make it simpler and compatible with Julia, as I have seen this type of format in @fjebaker repo
One more thing, I want to ask, I have kept these previous functions, shall I remove or just keep them? I thought it may help in vector time