EOS-DATA-2019-01: Posterior samples for Bbar -> D(*)lnu form factors in the heavy-quark expansion to order 1/m_c^2

Ancillary files for arXiv:1908.09398

For the fit model 3/2/1 we provide 50k posterior samples obtained from fits to a
theory-only likelihood (th-only) and a likelihood with theoretical and all experimental
constraints (th+all-exp):

  anc/pmc_321_th-only_50k.hdf5
  anc/pmc_321_th+all-exp_50k.hdf5

The posterior samples can be accessed through the following few lines of Python code:

  import h5py
  f = h5py.File('anc/pmc_321_th-only_50k.hdf5', 'r')
  samples = f['/data/final/samples'][:]

The first 23 (i=0..22) columnns of 'samples' correspond to the HQE parameters in the following order:

  xi'(1), xi''(1), xi'''(1),
  chi_2(1), chi_2'(1), chi_2''(1)
  chi_3'(1), chi_3''(1),
  eta(1), eta'(1), eta''(1)
  l_1(1), l_1'(1),
  l_2(1), l_2'(1),
  l_3(1), l_3'(1),
  l_4(1), l_4'(1),
  l_5(1), l_5'(1),
  l_6(1), l_6'(1)

The 24th (i=23) column corresponds to the index of the generating component in the Population Monte
Carlo algorithm. The 25th (i=24) column corresponds to the log(posterior) value, and the 26th (i=25)
column corresponds to the log(weight) of the sample.