Issue in parsing component spins for aligned spin BNS run

Issue in parsing component spins for BNS run to the WF call in LalSimulation

I tried to run aligned-spin BNS system with the default example, but it seemed to fail with the following error:

In [1]: %run binary_neutron_star_example.py
14:35 bilby INFO    : Running bilby version: 0.4.1: (CLEAN) 8af0cce 2019-03-04 23:22:41 -0600
/Users/arunava/anaconda3/lib/python3.7/site-packages/bilby-0.4.1-py3.7.egg/bilby/gw/detector.py:1997: RuntimeWarning: invalid value encountered in multiply
  frequency_domain_strain = self.__power_spectral_density_interpolated(frequencies) ** 0.5 * white_noise
XLAL Error - XLALSimInspiralChooseFDWaveform: Non-zero transverse spins were given, but this is a non-precessing approximant.
XLAL Error - XLALSimInspiralChooseFDWaveform (LALSimInspiral.c:1057): Invalid argument
---------------------------------------------------------------------------
RuntimeError                              Traceback (most recent call last)

/anaconda3/lib/python3.7/site-packages/bilby-0.4.1-py3.7.egg/bilby/gw/utils.py in lalsim_SimInspiralChooseFDWaveform(mass_1, mass_2, spin_1x, spin_1y, spin_1z, spin_2x, spin_2y, spin_2z, luminosity_distance, iota, phase, longitude_ascending_nodes, eccentricity, mean_per_ano, delta_frequency, minimum_frequency, maximum_frequency, reference_frequency, waveform_dictionary, approximant)
    797         longitude_ascending_nodes, eccentricity, mean_per_ano, delta_frequency,
    798         minimum_frequency, maximum_frequency, reference_frequency,
--> 799         waveform_dictionary, approximant)
    800
    801

RuntimeError: Invalid argument

In [2]:

• It seemed to be an error in parsing the spin components return values to the lalsim.SimInspiralChooseFDWaveform call in line no 794 of bilby/gw/utils.py for this version (bilby version: 0.4.1: (CLEAN) 8af0cce9 2019-03-04 23:22:41 -0600): here

Issue in computing the numbers in spin components

• While I am running it for the values fixed prior in chi_1 = chi_2 = 0.02

In [2]: list(priors.items())
Out[2]:
[('chi_1', DeltaFunction(peak=0.02, name=None, latex_label=None, unit=None)),
 ('chi_2', DeltaFunction(peak=0.02, name=None, latex_label=None, unit=None)),
 ('luminosity_distance',
  DeltaFunction(peak=50.0, name=None, latex_label=None, unit=None)),
 ('dec', DeltaFunction(peak=-1.2108, name=None, latex_label=None, unit=None)),
 ('ra', DeltaFunction(peak=1.375, name=None, latex_label=None, unit=None)),
 ('theta_jn', DeltaFunction(peak=0.4, name=None, latex_label=None, unit=None)),
 ('psi', DeltaFunction(peak=2.659, name=None, latex_label=None, unit=None)),
 ('phase', DeltaFunction(peak=1.3, name=None, latex_label=None, unit=None)),
 ('geocent_time',
  DeltaFunction(peak=1126259642.413, name=None, latex_label=None, unit=None)),
 ('chirp_mass',
  Gaussian(mu=1.215, sigma=0.1, name='chirp_mass', latex_label='$\\mathcal{M}$', unit='$M_{\\odot}$')),
 ('symmetric_mass_ratio',
  Uniform(minimum=0.1, maximum=0.25, name='symmetric_mass_ratio', latex_label='$\\eta$', unit=None)),
 ('lambda_tilde',
  Uniform(minimum=0, maximum=5000, name='lambda_tilde', latex_label='$\\tilde{\\Lambda}$', unit=None)),
 ('delta_lambda',
  Uniform(minimum=-5000, maximum=5000, name='delta_lambda', latex_label='$\\delta\\Lambda$', unit=None))]

I still see that spin components are passed to the lalsim.SimInspiralChooseFDWaveform call as:

spin_1x=-0.00493474, spin_1y=0.0193775, spin_1z=0.0004, spin_2x=-0.00493474, spin_2y=0.0193775, spin_2z=0.0004

• It seems to me to be incorrect.
• It also explains why the aligned spin waveform like TaylorF2 is failing. (see #334 (closed))