Closed bilby-bot closed 1 month ago
In GitLab by @git.ligo:gregory.ashton on Jul 7, 2019, 07:56
It sounds like you where missing a prior-line, but when added the new error is something related to a memory problem. That isn't something we typically expect.
@git.ligo:virginia.demilio , do I recall that you wrote this example? Would you mind checking if it still works with changes to the spins?
@git.ligo:christos.karathanasis could you say which versions of python and bilby you are using (the bilby version gets printed out in the logging) and give some more context to the error you see? E.g, what is happening just before it occurs, is there any other output?
In GitLab by @git.ligo:christos.karathanasis on Jul 7, 2019, 12:29
I am using bilby 0.5.1 and python 3.7.
Before this error there are outputs regarding the data(time,window,....), then there is this output multiple times : bilby INFO : Performing redundancy check using BBHPriorDict(self).test_redundancy , after this there are information about the priors like this one : bilby INFO : mass_1 = Uniform(minimum=1, maximum=2, name='mass_1', latex_label='$m1$', unit='$M{\odot}$', boundary=None) ,
then this appears : XLAL Error - XLALSimInspiralChooseFDWaveform: Non-zero transverse spins were given, but this is a non-precessing approximant. XLAL Error - XLALSimInspiralChooseFDWaveform (LALSimInspiral.c:1106): Invalid argument
and after this there are many lines of errors including the one above about the invalid pointer.
line 101, in
In GitLab by @git.ligo:virginia.demilio on Jul 8, 2019, 17:25
Hi @git.ligo:christos.karathanasis, thank you for pointing out these errors. As @git.ligo:gregory.ashton said I had put together this script at the pe f2f in February (at the time bilby == 0.3.5
), but bilby has been developed upon significantly since then so you'll have to take into account the changes to the spin (quickly looking at your error message you need to change to aligned spins prior as TaylorF2 is a non-precessing waveform).
I'm happy to re-run the script and make the necessary changes after the Amaldi conference is over next week. Also, as a side note, I remember you were wondering about the settings used for duration, psd_offset and psd_duration: these values were tweaked in order to fetch the open data using gwpy, as gwosc was not able to find the event data unless the gps interval included the trigger time (I believe this issue was subsequently solved as you can see here). I hope this is helpful and feel free to contact me if you need further help :)
In GitLab by @git.ligo:christos.karathanasis on Jul 10, 2019, 20:09
@git.ligo:gregory.ashton @git.ligo:virginia.demilio Yeah, indeed it was an error about spins and deformations in the waveform approximant. For the example to work you have to change the prior file to precescing_binary_neutron_star.prior and the waveform approximant to IMRPhenomPv2_NRTidal.
In GitLab by @git.ligo:gregory.ashton on Jul 30, 2019, 02:11
@git.ligo:virginia.demilio would you be able to update the example when you get a moment?
In GitLab by @git.ligo:virginia.demilio on Jul 30, 2019, 17:11
@git.ligo:gregory.ashton I've updated the example and made a GW170817.prior file for it. I'm currently running it to check the results are as expected. I'll send a WIP merge request in the mean time. Cheers!
In GitLab by @git.ligo:gregory.ashton on Aug 20, 2019, 11:14
Closing as this was resolved by !570
In GitLab by @git.ligo:gregory.ashton on Aug 20, 2019, 11:14
closed
In GitLab by @git.ligo:michael.williams on Oct 3, 2024, 17:55
unassigned @git.ligo:virginia.demilio
In GitLab by @git.ligo:christos.karathanasis on Jul 6, 2019, 12:25
An error appears when running the example for GW170817 (https://git.ligo.org/lscsoft/bilby/blob/master/examples/open_data_examples/GW170817.py). When running this I get the following error: File "/nfs/pic.es/user/c/ckaratha/miniconda3/lib/python3.7/site-packages/bilby/gw/waveform_generator.py", line 218, in parameters new_parameters.pop(key) KeyError: 'tilt_1'
I included the following priors to fix this and other similar errors: priors['a_1'] = bilby.core.prior.Uniform(name='a_1', minimum=0, maximum=0.05, boundary='reflective',latex_label='$a_1$') priors['a_2'] = bilby.core.prior.Uniform(name='a_2', minimum=0, maximum=0.05, boundary='reflective',latex_label='$a_2$') priors['tilt_1']=bilby.core.prior.Sine(name='tilt_1', boundary='reflective') priors['tilt_2']=bilby.core.prior.Sine(name='tilt_2', boundary='reflective') priors['phi_jl']=bilby.core.prior.Uniform(name='phi_jl', minimum=0, maximum=2 np.pi, boundary='periodic') priors['phi_12']=bilby.core.prior.Uniform(name='phi_12', minimum=0, maximum=2 np.pi, boundary='periodic')
But then I get an other error: Error in `python': free(): invalid pointer: 0x00007f0c958f8be8
The same error I also get when I change the prior file from binary_neutron_stars.prior to precessing_binary_neutron_stars.prior.