mirror of https://github.com/QMCPACK/qmcpack.git
![]() * Fix type + Format * Add type check to determinant_tools script and generalize alpha/beta groups. * Use unsigned integers in multidet h5 files * determinant_tools transforms CI vecs to uint64 p utils/determinants_tools.py tests/molecules/LiH_ae_MSD/QuantumPackage/LiH.orbs.h5 --transform Example on LiH_ae_MSD 2166c2166 < DATATYPE H5T_STD_I64LE --- > DATATYPE H5T_STD_U64LE 2701c2701 < (532,0): -9223372036854775807, 0, --- > (532,0): 9223372036854775809, 0, 3111c3111 < (942,0): -9223372036854775807, 0, --- > (942,0): 9223372036854775809, 0, 3243c3243 < (1074,0): -9223372036854775807, 0, --- > (1074,0): 9223372036854775809, 0, 3659c3659 < DATATYPE H5T_STD_I64LE --- > DATATYPE H5T_STD_U64LE 4194c4194 < (532,0): -9223372036854775807, 0, --- > (532,0): 9223372036854775809, 0, 4605c4605 < (943,0): -9223372036854775807, 0, --- > (943,0): 9223372036854775809, 0, 4737c4737 < (1075,0): -9223372036854775807, 0, --- > (1075,0): 9223372036854775809, 0, * Transform test inputs * Always transform in determinant_tools if the file contains signed integers * Ammended docs * revert to transform flag in determinant_tools.py * Add functions to check the type a dataset in hdf file * Check that the CI vectors are stored as uint64 in the HDF5 file * Update hdf5 to have unsigned determinants * Converters should use uint64_t |
||
---|---|---|
.. | ||
Bi.mol | ||
df.inp | ||
df_Bi.out | ||
gold.orbs.h5 | ||
gold.qmc.in-wfnoj.xml | ||
gold.structure.xml | ||
gold.wfnoj.xml |