search

tikk3r / lofar-h5plot

A small GUI for plotting H5Parms produced during LOFAR calibration.
GNU General Public License v3.0
2 stars 0 forks source link

Fix plotting weights for all stations #31

Closed tikk3r closed 6 months ago

tikk3r commented 6 months ago

Plotting the weights for all stations did not transpose the data correctly, as reported in #30. This fixes that issue.

rvweeren commented 6 months ago

I think there are more issues, also plotting the data. Just an example from the amplitudes where top-bottom are reversed. Phases also seem to suffer from the same issue (and maybe the weights also, diffs, etc. did not check these). I'm not sure which is the correct one (top or bottom).

image image
tikk3r commented 6 months ago

Hmm, can I grab this h5parm from somewhere? I can't reproduce the inverted amplitudes with one I have lying around.

rvweeren commented 6 months ago

/net/rijn1/data1/rvweeren/Perseus_LBA/merged_selfcalcyle010_TC00.MS.avg.subtracted.copy.avg.h5

rvweeren commented 6 months ago

(this is another example, but it shows the same problem)

rvweeren commented 6 months ago

Ahh sorry for the confusion, the amplitude problem was fixed (together with the weights fix) as reported in https://github.com/tikk3r/lofar-h5plot/issues/30. I spotted the amplitude before trying the issue-30 fix. Turns out that both weights and amplitude plottings issue are adressed in branch issue-30

rvweeren commented 6 months ago

Conclusion: I can confirm your fix works and I think you can merge with master.