feat: Implementing frequency rebinning for PowerSpectrum and LightCurve by Omiii-215 · Pull Request #71 · StingraySoftware/Stingray.jl

Omiii-215 · 2026-02-15T20:21:04Z

Summary

Adds frequency rebinning support for power spectra.

What’s included

Linear rebinning via rebin(ps, factor)
Logarithmic rebinning via logrebin(ps, f)
Integer-factor rebin overload for LightCurve
Tests for linear/log rebinning and error scaling

Notes

Keeps implementation minimal and focused on core functionality.
Designed to be extendable for future improvements.

Implement Workflow

Add fourier methods and tests

Implement GTI Reading and Handling

…nmath Bump NanMath requirements

Merge GTI modifications into main

…uple

It seems entirely overkill to run on 3 different operating systems and for two different Julia versions for a package that is still in heavy development. Currently we just need the CI to be rapid and automate code checking. There is nothing in this package that is specific to a given OS, and we can worry about compat with Julia versions once we approach a first stable version.

Simplify CI

Fix StingraySoftware#16 & StingraySoftware#17 Removed Metadata.jl dependency

Events

Unstable one

…ation and adding test includes

- Restore GTI filtering in - Harden - Rename to - Update docstrings and plotting recipes - Verify rebinning logic with tests

kashish2210

Hey @Omiii-215, thanks for ur sub PR you have really done a great work {u can once review my comments. specially this one try testing via some datas} :)

kashish2210 · 2026-02-17T04:36:46Z

 [deps]
 Documenter = "e30172f5-a6a5-5a46-863b-614d45cd2de4"
+Stingray = "2045e982-6adb-4e09-8364-0d2cc8a3fe4f"



While running the project, we can't include Stingray in the project.toml {this is the document config}
So I know what happened here, you may tried to make docs you can just use

using Stingray Stingray.someFuncunction()

avoid this scope

kashish2210 · 2026-02-17T04:39:55Z

        for (band_name, events) in events_dict
-            ps = AveragedPowerspectrum(events, segment_size; norm = norm)
+            ps = AveragedPowerSpectrum(events, segment_size; norm = norm)
            power_vec = freq_mult ? ps.power .* ps.freq : ps.power


Thanks for noticing this :)

kashish2210 · 2026-02-17T04:40:16Z

    counts::Vector{Int},
    method::Symbol;
-    gaussian_errors::Union{Nothing,Vector{<:Real}} = nothing,
+    gaussian_errors::Union{Nothing,Vector{<:Real}} = nothing


kashish2210 · 2026-02-17T04:42:59Z

 # Throws
 - `ArgumentError`: If custom errors length doesn't match bin count
 """
-function set_errors!(lc::LightCurve{T}) where {T}


you revert to

function set_errors!(lc::LightCurve{T}) where {T}

since"Curly braces become necessary when you have multiple type parameters." Otherwise, it is the same

kashish2210 · 2026-02-17T04:43:35Z

 - [`set_errors!`](@ref): Set error method and calculate errors
 """
-function calculate_errors!(lc::LightCurve{T}) where {T}
+function calculate_errors!(lc::LightCurve{T}) where T


same here for {T}

kashish2210 · 2026-02-17T04:47:59Z

-        merge!(extra_metadata, eventlist.meta.extra)
+    if !isnothing(eventlist.meta.gti)
+        extra_metadata["gti"] = eventlist.meta.gti
    end


This may hinder review it once since merging the gtis is crucial

kashish2210 · 2026-02-17T04:49:26Z

+            throw(ArgumentError("No events remain after GTI filtering"))
+        end
+    end
+


i didnt get what the use of this?

kashish2210 · 2026-02-17T04:51:08Z

+            new_powers,
+            new_errs,
+            ps.norm,
+            0.0, # df is variable


beware of df

kashish2210 · 2026-02-17T04:52:36Z

+using Test
+using Stingray
+using Statistics
+


u can include these files in the runtest.jl {it is already there}

kashish2210 · 2026-02-17T05:04:18Z

+            ps.metadata
+        )
+    end
+end


You need to test these new functions via data present here
Use clean data "ni1200120104_0mpu7_cl.evt" from "ni1200120104_0mpu7_cl.evt.gz" so u can run the test easily via setting up a Jupyter notebook and include powespectrum and plotting.jl and lightcurve and event.jl files in ur dirs and start testing, use your function via first using readevents function u may refer to:

and start making Plots : )

- LightCurve: Restore apply_filters, safe metadata merge, correct GTI logic - PowerSpectrum: Fix logrebin df calculation to avoid 0.0 - Docs: Fix Pkg.develop pattern and deps in docs/make.jl - Tests: Add robust invariant checks to test_rebinning.jl

Omiii-215 · 2026-02-18T16:50:37Z

@kashish2210 Verification with Real Data:

I have verified these changes locally using a clean event file (ni1200120104_0mpu7_cl.evt). The apply_filters, rebin (linear), and logrebin (logarithmic) functions were tested.

LightCurve Rebinning: Validated that total counts are preserved exactly (sum(lc_binned.counts) == sum(lc.counts)).
PowerSpectrum Log Rebinning: Confirmed that logrebin now produces a positive effective df (fixing the df=0.0 issue) and strictly monotonic frequencies.
Plots: Visually inspected the rebinned light curves and power spectra to ensure they match the expected behavior of the original data.

Loading events from ni1200120104_0mpu7_cl.evt...
Event file loaded successfully.
  Telescope: NICER
  Instrument: XTI
  Object: MAXI_J1820+070
  Events: 21244574

--- Testing LightCurve Rebinning ---
Original LC: 731950 bins, dt=0.1 s
Rebinned LC: 73195 bins, dt=1.0 s
Total Counts Check: Original=21244574, Rebinned=21244574
SUCCESS: Total counts preserved.

--- Testing PowerSpectrum Linear Rebinning ---
Original PDS: 365974 bins, df=1.3662135391761732e-5 Hz
Linearly Rebinned PDS: 91493 bins, df=5.464854156704693e-5 Hz

--- Testing PowerSpectrum Logarithmic Rebinning ---
Log Rebinned PDS: 127 bins
  Effective df: 0.0012056834483229716 Hz (should not be 0.0)
SUCCESS: Effective df is positive.
Monotonicity Check: Passed

Verification complete. Plots saved as PNG files.

kashish2210 · 2026-02-18T17:37:04Z

@kashish2210 Verification with Real Data:

I have verified these changes locally using a clean event file (ni1200120104_0mpu7_cl.evt). The apply_filters, rebin (linear), and logrebin (logarithmic) functions were tested.

LightCurve Rebinning: Validated that total counts are preserved exactly (sum(lc_binned.counts) == sum(lc.counts)).
PowerSpectrum Log Rebinning: Confirmed that logrebin now produces a positive effective df (fixing the df=0.0 issue) and strictly monotonic frequencies.
Plots: Visually inspected the rebinned light curves and power spectra to ensure they match the expected behavior of the original data.
Loading events from ni1200120104_0mpu7_cl.evt...
Event file loaded successfully.
  Telescope: NICER
  Instrument: XTI
  Object: MAXI_J1820+070
  Events: 21244574

--- Testing LightCurve Rebinning ---
Original LC: 731950 bins, dt=0.1 s
Rebinned LC: 73195 bins, dt=1.0 s
Total Counts Check: Original=21244574, Rebinned=21244574
SUCCESS: Total counts preserved.

--- Testing PowerSpectrum Linear Rebinning ---
Original PDS: 365974 bins, df=1.3662135391761732e-5 Hz
Linearly Rebinned PDS: 91493 bins, df=5.464854156704693e-5 Hz

--- Testing PowerSpectrum Logarithmic Rebinning ---
Log Rebinned PDS: 127 bins
  Effective df: 0.0012056834483229716 Hz (should not be 0.0)
SUCCESS: Effective df is positive.
Monotonicity Check: Passed

Verification complete. Plots saved as PNG files.

This looks good to me nice work @Omiii-215

kashish2210 · 2026-02-18T17:39:07Z

Just one more thing we should not focus on docs since, for this scope it is not needed so u can revert changes in that

Omiii-215 · 2026-02-18T20:12:21Z

Just one more thing we should not focus on docs since, for this scope it is not needed so u can revert changes in that

Done

kashish2210

Thank you @Omiii-215

kashish2210 · 2026-02-19T02:39:04Z

@matteobachetti and @fjebaker, we can start testing this branch, waiting for your review

matteobachetti · 2026-02-27T19:20:30Z

@Omiii-215 in the current state, the PR is full of formatting changes that really make review difficult. Also, it seems that you removed at least one function. Please clean up the PR, only maintain the actual important changes.

codecov-commenter · 2026-02-27T19:27:08Z

Codecov Report

❌ Patch coverage is 75.70093% with 26 lines in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (unstable@5d2763d). Learn more about missing BASE report.

Files with missing lines	Patch %	Lines
src/powerspectrum.jl	73.07%	21 Missing ⚠️
src/plotting/plots_recipes_powerspectrum.jl	0.00%	3 Missing ⚠️
src/lightcurve.jl	92.30%	2 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##             unstable      #71   +/-   ##
===========================================
  Coverage            ?   66.97%           
===========================================
  Files               ?       12           
  Lines               ?     3770           
  Branches            ?        0           
===========================================
  Hits                ?     2525           
  Misses              ?     1245           
  Partials            ?        0

☔ 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.

… naming changes

Omiii-215 · 2026-02-27T22:07:01Z

@matteobachetti I've completely cleaned up the PR, reverted all unrelated formatting/naming changes (like PowerSpectrum) and restored the accidentally removed functions. The branch now only cleanly introduces the new rebin and logrebin implementations, and all tests are passing. :)

matteobachetti and others added 30 commits March 14, 2022 10:16

Initial commit

7dabbfe

Implement Workflow

f8020ce

Fix minor issues

e00ec18

Merge pull request StingraySoftware#1 from matteobachetti/aman-ci

9de9cb7

Implement Workflow

Add fourier methods and tests

459ad64

Update Project.toml

2fc095d

Fix Issues

1d23a13

Fix Compat Error

8cfc449

Fix Performance Issues

2a4f3e0

Add more tests and Use NaNMath

6ee1d1c

Minor Fixes

0854d43

Implement GTI Reading and Handling

ac4923c

Merge pull request StingraySoftware#2 from matteobachetti/aman-fourier

7bac1ec

Add fourier methods and tests

Fix issues

a020c7e

Merge pull request StingraySoftware#3 from StingraySoftware/aman-gti

f2dacb5

Implement GTI Reading and Handling

Bump NanMath requirements

fafcc8d

Merge pull request StingraySoftware#5 from StingraySoftware/update_na…

61f774f

…nmath Bump NanMath requirements

Merge branch 'main' into aman-fourier

23e43ac

Merge pull request StingraySoftware#6 from StingraySoftware/aman-fourier

8dd50d4

Merge GTI modifications into main

Removed Metadata.jl dependency and replaced it with a struct

27328aa

Removed Metadata.jl dependency

fec2773

Updated PR with feedback, replaced FourierResults(Struct) with NamedT…

69503a7

…uple

Missing Return(Results)

1b45a9a

For Latest Julia Version(1.11)

d27cc8f

ci: bump action versions to latest

93ca2c2

ci: add cache action to docs

bf841c8

Merge pull request StingraySoftware#21 from StingraySoftware/fergus/ci

300f82b

Simplify CI

Merge branch 'main' into remove-metadata

a96164c

Merge pull request StingraySoftware#18 from JawadAhmadCS/remove-metadata

54bb952

Fix StingraySoftware#16 & StingraySoftware#17 Removed Metadata.jl dependency

kashish2210 and others added 8 commits August 26, 2025 14:06

adding sec_to_mjd function

a6cb814

Merge pull request StingraySoftware#61 from kashish2210/events

ddb5af3

Events

adding all updates

10a972b

formating

7595368

Merge pull request StingraySoftware#66 from kashish2210/unstable

f46a350

Unstable one

feat: implement frequency rebinning for PowerSpectrum and LightCurve

cdf9ff0

Merge upstream/unstable: resolve conflicts by keeping rebin implement…

78f26a3

…ation and adding test includes

Resolve merge conflicts, fix GTI metadata bug, and relax test tolerance

6c4f710

Omiii-215 marked this pull request as draft February 15, 2026 21:40

refactor: address code review feedback

fabffbe

- Restore GTI filtering in - Harden - Rename to - Update docstrings and plotting recipes - Verify rebinning logic with tests

Omiii-215 marked this pull request as ready for review February 16, 2026 09:56

Omiii-215 mentioned this pull request Feb 16, 2026

Fourier transforms and spectra #33

Open

kashish2210 reviewed Feb 17, 2026

View reviewed changes

Omiii-215 added 2 commits February 18, 2026 21:50

Remove unwanted comment lines

d811a92

revert: remove docs changes from scope

8c02221

kashish2210 approved these changes Feb 19, 2026

View reviewed changes

kashish2210 mentioned this pull request Feb 23, 2026

Refactor plotting recipes and add comprehensive test suite for Issue #29 #72

Open

Omiii-215 added 3 commits February 28, 2026 03:05

Cleanup PR: keep only rebin/logrebin additions, revert formatting and…

28628a8

… naming changes

Merge upstream/unstable and resolve conflicts

f5d6479

Include and fix test_rebinning.jl in main test suite

437f9af

Omiii-215 mentioned this pull request Mar 20, 2026

Stingray | JuliaAstro: Spectral Timing in Julia (Om S Habib) OpenAstronomy/gsoc-proposals#14

Open

5 tasks

Conversation

Omiii-215 commented Feb 15, 2026

Summary

What’s included

Notes

Uh oh!

kashish2210 left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Thanks for noticing this :)

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

kashish2210 Feb 17, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Omiii-215 commented Feb 18, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

kashish2210 commented Feb 18, 2026

Uh oh!

kashish2210 commented Feb 18, 2026

Uh oh!

Omiii-215 commented Feb 18, 2026

Uh oh!

kashish2210 left a comment

Choose a reason for hiding this comment

Uh oh!

kashish2210 commented Feb 19, 2026

Uh oh!

matteobachetti commented Feb 27, 2026

Uh oh!

codecov-commenter commented Feb 27, 2026

Codecov Report

Uh oh!

Omiii-215 commented Feb 27, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

7 participants

kashish2210 left a comment •

edited

Loading

kashish2210 Feb 17, 2026 •

edited

Loading

Omiii-215 commented Feb 18, 2026 •

edited

Loading