-
Notifications
You must be signed in to change notification settings - Fork 4
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Averaging issue in misaligned coarse channels #162
Comments
was not able to reproduce this issue in Marlu. MWATelescope/Marlu@b74adde#diff-c8aa32c8b3f0bfb2afad951e7427d2fa0e03a5218025e6afadaee1c07c9de949R434 |
might be an aoflagger thing, it only shows up with rfi enabled. export time_res_s=4;
birli \
--sel-time 50 53 \
--sel-ants 0 \
--avg-time-res $time_res_s \
--avg-freq-res 80 \
--flag-edge-width 0 \
-m demo/data/1128352792/raw/1128352792.metafits \
demo/data//1128352792/raw/1128352792_2*.fits \
-u demo/data//1128352792/prep/birli_${time_res_s}s_80kHz_1128352792.uvfits \
&& demo/04_ssins.py --no-diff --threshold 9999 --export-tsv $_ gps_time 187.5550 187.6350 187.7150 187.7950 187.8750 187.9550 188.0350 188.1150 188.1950 188.2750 188.3550 188.4350 188.5150 188.5950 188.6750 188.7550 188.8350 188.9150 188.9950 189.0750 189.1550 189.2350 189.3150 189.3950 189.4750 189.5550 189.6350 189.7150 189.7950 189.8750 189.9550 190.0350
1128352894.000 47849.730 47790.383 47580.281 47675.242 47718.297 47445.801 47126.418 46590.168 45567.430 45270.258 45027.156 44471.488 44114.973 44169.926 44952.309 46509.164 45658.797 45439.703 45166.535 45489.891 45577.422 45322.711 44957.766 44526.797 43751.695 43736.848 42851.535 42445.188 42365.641 42207.148 42911.750 44171.449
1128352898.000 48022.730 47645.055 47647.215 47793.059 47387.375 47573.438 46867.598 46307.125 45664.832 45274.172 44997.586 44495.887 44237.852 44579.625 44821.891 46410.742 45773.086 45434.098 45239.570 45228.828 45035.562 45534.949 22562.662 44734.473 44251.035 43521.223 21458.520 21254.477 21105.754 21093.191 42846.855 44288.273
gps_time 187.5550 187.6350 187.7150 187.7950 187.8750 187.9550 188.0350 188.1150 188.1950 188.2750 188.3550 188.4350 188.5150 188.5950 188.6750 188.7550 188.8350 188.9150 188.9950 189.0750 189.1550 189.2350 189.3150 189.3950 189.4750 189.5550 189.6350 189.7150 189.7950 189.8750 189.9550 190.0350
1128352894.000 47849.730 47790.383 47580.281 47675.242 47718.297 47445.801 47087.434 46590.168 45567.430 45270.258 44793.953 44471.445 44114.973 44169.926 44952.309 46469.344 45544.012 45439.703 45166.535 45491.965 45470.359 45322.711 44957.770 44526.797 43751.695 43538.270 42851.605 42445.207 42365.422 42205.293 42708.570 44171.449
1128352898.000 48022.730 47645.055 47647.215 47793.059 47387.375 47573.438 47126.207 46307.125 45664.832 45274.172 44831.289 44465.848 44237.852 44579.625 44821.891 46476.375 45650.793 45434.098 45239.570 45372.379 45081.391 45534.949 45125.328 44734.473 44251.035 43490.277 42917.289 42508.957 42211.570 42181.242 42733.941 44288.273 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
When preprocessing an observation where a coarse channel ends early, birli is not correctly averaging the final timestamp on that coarse channel.
we'll use mwa-demo to test this
birli preprocess only first antenna and last few timesteps, no averaging (2s)
mkdir -p demo/data//1128352792/prep/ birli \ -m demo/data/1128352792/raw/1128352792.metafits \ --sel-time 50 53 --sel-ants 0 --flag-edge-width 80 \ -u demo/data//1128352792/prep/birli_2s_1128352792.uvfits \ demo/data//1128352792/raw/1128352792_2*.fits
plot and export auto magnitudes to tsv
the last timestep on channel 148 is all negative, but that's fine because it's all flagged.
birli preprocess only first antenna and last few timesteps, averaging to 4s
birli \ -m demo/data/1128352792/raw/1128352792.metafits \ --sel-time 50 53 --sel-ants 0 --flag-edge-width 80 \ --avg-time-res 4 \ -u demo/data//1128352792/prep/birli_4s_1128352792.uvfits \ demo/data//1128352792/raw/1128352792_2*.fits demo/04_ssins.py demo/data/1128352792/prep/birli_4s_1128352792.uvfits --no-diff --threshold 9999 --export-tsv
Now some negative values (not all) have made it into the last timestamp, and because not all of the visibilities that went into these were all flagged, these values are not flagged, they're just down-weighted.
Average should not be including these flagged values.
The text was updated successfully, but these errors were encountered: