Weekly Progress of DeePMD-kit 20221128

[HuangJiameng]

What's new

• skip get_gpus subprocess when TF is cpu only #2135
• fix the missing header in api_c tests #2131
• c: fix bug in hpp float interface and add tests #2128
• add precompile headers #2123
• fix invalid escape sequence in strings #2122
• speed up tests with CLI #2121
• speed up test_neighbor_stat #2120
• c: fix ComputeNlist -> ComputeNList in the C API #2119
• print out the time of each Python unit test #2116
• docs: add a FAQ about precision #2115
• docs: fix spellings and grammars #2114
• c: support DeepPot model deviation in the C API #2112
• add shuffle_exclude_empty to DeepPot.compute(nlist, atomic) #2111

Issue

• I want to ask some questions about DeepMD and dp-gen #2129
• I want to ask how the network map the Descriptor into the output energy, #2126
• 'sel' adjustment support error #2118
• [BUG] TestInferDeepPotA.cpu_lmp_nlist_type_sel failed when requesting atomic energies and atomic virial #2105
• [Feature Request] Support neighbor list input in the C API #2086
• [Feature Request] some unit tests are quite slow #2117

Discussion

• Can the DPLR model simulate external electric field interactions? #2136
• [LAMMPS] Potential Comparison #2133
• I want to ask some questions about DeepMD and dp-gen #2134
• [For Third-party Developers] Migrate from C++ library to C library #2124
• Help with some questions about the ZBL form #2130
• can gromacs be uese with deepmd-kit which is build on tensorflow 1.15 #2125
• I want to ask how the network map the Descriptor into the output energy, #2127