MPI running on gap_fitting

[GUANGZChen]

Hi GAP developers,
I am currently trying MPI gap fitting on an HPC system. I successfully finished the installation part.

For using part, I followed the instruction as follows,

1. Calculate without MPI to get the sparseX output files
   Use sparsify_only_no_fit=T to just create the sparseX files
2. Convert output files to input files (QUIP script gap_prepare_sparsex_input.py gp.xml)
   Optionally rename them to something shorter (e.g. 1.input, 2.input etc.)
3. Change input method (sparse_method=FILE sparse_file=1.input)
   For more than one species use add_species=F and explicit input
4. Check the number of braces (gap={{$gap1}:{$gap2}:{$gap3}})
   Run with mpirun -np … (or srun for Slurm)
   MPI can trip up on the ….xyz.idx file. Usually works if it's there from the start, so retry can help.

I simply add sparsify_only_no_fit=T at the end of my original code which is attached below.

gap_fit at_file=CN_train_dataset.xyz e0={C:-148.6822264562:N:-271.4749770587} gap={distance_2b n_sparse=15 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=4.5 delta=2.0:angle_3b n_sparse=200 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=2.5 delta=0.05:soap n_max=12 l_max=4 atom_sigma=0.5 zeta=4.0 cutoff=4.5 cutoff_transition_width=1.0 central_weight=1.0 n_sparse=10000 delta=0.2 covariance_type=dot_product sparse_method=cur_points radial_decay=-0.5} default_sigma={0.001 0.01 0.05 0.0} config_type_sigma={Liquid:0.050:0.5:0.5:0.0:Liquid_Interface:0.050:0.5:0.5:0.0:Amorphous_Bulk:0.005:0.2:0.2:0.0:Amorphous_Surfaces:0.005:0.2:0.2:0.0:Surfaces:0.002:0.1:0.2:0.0:Dimer:0.002:0.1:0. :0.0:Fullerenes:0.002:0.1:0.2:0.0:Defects:0.001:0.01:0.05:0.0:Crystalline_Bulk:0.001:0.01:0.05:0.0:Nanotubes:0.001:0.01:0.05:0.0:Graphite:0.001:0.01:0.05:0.0:Diamond:0.001:0.01:0.05:0.0:Graphene:0.001:0.01:0.05:0.0:Graphite_Layer_Sep:0.001:0.01:0.05:0.0:Single_Atom:0.0001:0.001:0.05:0.0} energy_parameter_name=energy force_parameter_name=force sparse_jitter=1.0e-8 do_copy_at_file=F openmp_chunk_size=10000 gp_file=gap.xml sparsify_only_no_fit=T| tee out

Also, the output of this code is as follows:
libAtoms::Hello World: 2023-06-29 00:41:14
libAtoms::Hello World: git version https://github.com/libAtoms/QUIP,v0.9.12-dirty
libAtoms::Hello World: QUIP_ARCH linux_x86_64_gfortran_openmp
libAtoms::Hello World: compiled on Jan 15 2023 at 16:26:45
libAtoms::Hello World: OpenMP parallelisation with 1 threads
WARNING: libAtoms::Hello World: environment variable OMP_STACKSIZE not set explicitly. The default value - system and compiler dependent - may be too small for some applications.
libAtoms::Hello World: Random Seed = 2474568
libAtoms::Hello World: global verbosity = 0

Calls to system_timer will do nothing by default

================================ Input parameters ==============================

config_file =
atoms_filename = //MANDATORY//
at_file = CN_train_dataset.xyz
gap = "distance_2b n_sparse=15 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=4.5 delta=2.0:angle_3b n_sparse=200 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=2.5 delta=0.05:soap n_max=12 l_max=4 atom_sigma=0.5 zeta=4.0 cutoff=4.5 cutoff_transition_width=1.0 central_weight=1.0 n_sparse=10000 delta=0.2 covariance_type=dot_product sparse_method=cur_points radial_decay=-0.5"
e0 = C:-148.6822264562:N:-271.4749770587
local_property0 = 0.0
e0_offset = 0.0
e0_method = isolated
default_kernel_regularisation = //MANDATORY//
default_sigma = "0.001 0.01 0.05 0.0"
default_kernel_regularisation_local_property = 0.001
default_local_property_sigma = 0.001
sparse_jitter = 1.0e-8
hessian_displacement = 1.0e-2
hessian_delta = 1.0e-2
baseline_param_filename = quip_params.xml
core_param_file = quip_params.xml
baseline_ip_args =
core_ip_args =
energy_parameter_name = energy
local_property_parameter_name = local_property
force_parameter_name = force
virial_parameter_name = virial
stress_parameter_name = stress
hessian_parameter_name = hessian
config_type_parameter_name = config_type
kernel_regularisation_parameter_name = sigma
sigma_parameter_name = sigma
force_mask_parameter_name = force_mask
parameter_name_prefix =
config_type_kernel_regularisation =
config_type_sigma = Liquid:0.050:0.5:0.5:0.0:Liquid_Interface:0.050:0.5:0.5:0.0:Amorphous_Bulk:0.005:0.2:0.2:0.0:Amorphous_Surfaces:0.005:0.2:0.2:0.0:Surfaces:0.002:0.1:0.2:0.0:Dimer:0.002:0.1:0.2:0.0:Fullerenes:0.002:0.1:0.2:0.0:Defects:0.001:0.01:0.05:0.0:Crystalline_Bulk:0.001:0.01:0.05:0.0:Nanotubes:0.001:0.01:0.05:0.0:Graphite:0.001:0.01:0.05:0.0:Diamond:0.001:0.01:0.05:0.0:Graphene:0.001:0.01:0.05:0.0:Graphite_Layer_Sep:0.001:0.01:0.05:0.0:Single_Atom:0.0001:0.001:0.05:0.0
kernel_regularisation_is_per_atom = T
sigma_per_atom = T
do_copy_atoms_file = T
do_copy_at_file = F
sparse_separate_file = T
sparse_use_actual_gpcov = F
gap_file = gap_new.xml
gp_file = gap.xml
verbosity = NORMAL
rnd_seed = -1
openmp_chunk_size = 10000
do_ip_timing = F
template_file = template.xyz
sparsify_only_no_fit = T
dryrun = F
condition_number_norm =
linear_system_dump_file =
mpi_blocksize_rows = 0
mpi_blocksize_cols = 100
mpi_print_all = F

======================================== ======================================

WARNING: sparsify_only_no_fit == T: force, virial, hessian, stress parameters are ignored.

============== Gaussian Approximation Potentials - Database fitting ============

Initial parsing of command line arguments finished.
Found 3 GAPs.
Descriptors have been parsed
XYZ file read
Old GAP: {distance_2b n_sparse=15 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=4.5 delta=2.0}
New GAP: {distance_2b n_sparse=15 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=4.5 delta=2.0 Z1=6 Z2=6}
New GAP: {distance_2b n_sparse=15 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=4.5 delta=2.0 Z1=6 Z2=7}
New GAP: {distance_2b n_sparse=15 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=4.5 delta=2.0 Z1=7 Z2=7}
Old GAP: {angle_3b n_sparse=200 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=2.5 delta=0.05}
New GAP: {angle_3b n_sparse=200 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=2.5 delta=0.05 Z=6 Z1=6 Z2=6}
New GAP: {angle_3b n_sparse=200 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=2.5 delta=0.05 Z=6 Z1=6 Z2=7}
New GAP: {angle_3b n_sparse=200 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=2.5 delta=0.05 Z=6 Z1=7 Z2=7}
New GAP: {angle_3b n_sparse=200 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=2.5 delta=0.05 Z=7 Z1=6 Z2=6}
New GAP: {angle_3b n_sparse=200 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=2.5 delta=0.05 Z=7 Z1=6 Z2=7}
New GAP: {angle_3b n_sparse=200 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=2.5 delta=0.05 Z=7 Z1=7 Z2=7}
Old GAP: {soap n_max=12 l_max=4 atom_sigma=0.5 zeta=4.0 cutoff=4.5 cutoff_transition_width=1.0 central_weight=1.0 n_sparse=10000 delta=0.2 covariance_type=dot_product sparse_method=cur_points radial_decay=-0.5}
New GAP: {soap n_max=12 l_max=4 atom_sigma=0.5 zeta=4.0 cutoff=4.5 cutoff_transition_width=1.0 central_weight=1.0 n_sparse=10000 delta=0.2 covariance_type=dot_product sparse_method=cur_points radial_decay=-0.5 n_species=2 Z=6 species_Z={6 7 }}
New GAP: {soap n_max=12 l_max=4 atom_sigma=0.5 zeta=4.0 cutoff=4.5 cutoff_transition_width=1.0 central_weight=1.0 n_sparse=10000 delta=0.2 covariance_type=dot_product sparse_method=cur_points radial_decay=-0.5 n_species=2 Z=7 species_Z={6 7 }}
Sparse points and target errors per pre-defined types of configurations
Liquid 0.50000000000000003E-001 0.50000000000000000E+000 0.50000000000000000E+000 0.00000000000000000E+000
Liquid_Interface 0.50000000000000003E-001 0.50000000000000000E+000 0.50000000000000000E+000 0.00000000000000000E+000
Amorphous_Bulk 0.50000000000000001E-002 0.20000000000000001E+000 0.20000000000000001E+000 0.00000000000000000E+000
Amorphous_Surfaces 0.50000000000000001E-002 0.20000000000000001E+000 0.20000000000000001E+000 0.00000000000000000E+000
Surfaces 0.20000000000000000E-002 0.10000000000000001E+000 0.20000000000000001E+000 0.00000000000000000E+000
Dimer 0.20000000000000000E-002 0.10000000000000001E+000 0.20000000000000001E+000 0.00000000000000000E+000
Fullerenes 0.20000000000000000E-002 0.10000000000000001E+000 0.20000000000000001E+000 0.00000000000000000E+000
Defects 0.10000000000000000E-002 0.10000000000000000E-001 0.50000000000000003E-001 0.00000000000000000E+000
Crystalline_Bulk 0.10000000000000000E-002 0.10000000000000000E-001 0.50000000000000003E-001 0.00000000000000000E+000
Nanotubes 0.10000000000000000E-002 0.10000000000000000E-001 0.50000000000000003E-001 0.00000000000000000E+000
Graphite 0.10000000000000000E-002 0.10000000000000000E-001 0.50000000000000003E-001 0.00000000000000000E+000
Diamond 0.10000000000000000E-002 0.10000000000000000E-001 0.50000000000000003E-001 0.00000000000000000E+000
Graphene 0.10000000000000000E-002 0.10000000000000000E-001 0.50000000000000003E-001 0.00000000000000000E+000
Graphite_Layer_Sep 0.10000000000000000E-002 0.10000000000000000E-001 0.50000000000000003E-001 0.00000000000000000E+000
Single_Atom 0.10000000000000000E-003 0.10000000000000000E-002 0.50000000000000003E-001 0.00000000000000000E+000
default 0.10000000000000000E-002 0.10000000000000000E-001 0.50000000000000003E-001 0.00000000000000000E+000
Multispecies support added where requested

===================== Report on number of descriptors found ====================

---

Descriptor 1: distance_2b n_sparse=15 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=4.5 delta=2.0 Z1=6 Z2=6
Number of descriptors: 7382902
Number of partial derivatives of descriptors: 0

Descriptor 2: distance_2b n_sparse=15 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=4.5 delta=2.0 Z1=6 Z2=7
Number of descriptors: 44
Number of partial derivatives of descriptors: 0

Descriptor 3: distance_2b n_sparse=15 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=4.5 delta=2.0 Z1=7 Z2=7
Number of descriptors: 44
Number of partial derivatives of descriptors: 0

Descriptor 4: angle_3b n_sparse=200 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=2.5 delta=0.05 Z=6 Z1=6 Z2=6
Number of descriptors: 9039580
Number of partial derivatives of descriptors: 0

Descriptor 5: angle_3b n_sparse=200 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=2.5 delta=0.05 Z=6 Z1=6 Z2=7
Number of descriptors: 0
Number of partial derivatives of descriptors: 0

Descriptor 6: angle_3b n_sparse=200 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=2.5 delta=0.05 Z=6 Z1=7 Z2=7
Number of descriptors: 0
Number of partial derivatives of descriptors: 0

Descriptor 7: angle_3b n_sparse=200 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=2.5 delta=0.05 Z=7 Z1=6 Z2=6
Number of descriptors: 0
Number of partial derivatives of descriptors: 0

Descriptor 8: angle_3b n_sparse=200 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=2.5 delta=0.05 Z=7 Z1=6 Z2=7
Number of descriptors: 0
Number of partial derivatives of descriptors: 0

Descriptor 9: angle_3b n_sparse=200 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=2.5 delta=0.05 Z=7 Z1=7 Z2=7
Number of descriptors: 0
Number of partial derivatives of descriptors: 0

Descriptor 10: soap n_max=12 l_max=4 atom_sigma=0.5 zeta=4.0 cutoff=4.5 cutoff_transition_width=1.0 central_weight=1.0 n_sparse=10000 delta=0.2 covariance_type=dot_product sparse_method=cur_points radial_decay=-0.5 n_species=2 Z=6 species_Z={6 7 }
Number of descriptors: 178167
Number of partial derivatives of descriptors: 0

Descriptor 11: soap n_max=12 l_max=4 atom_sigma=0.5 zeta=4.0 cutoff=4.5 cutoff_transition_width=1.0 central_weight=1.0 n_sparse=10000 delta=0.2 covariance_type=dot_product sparse_method=cur_points radial_decay=-0.5 n_species=2 Z=7 species_Z={6 7 }
Number of descriptors: 79
Number of partial derivatives of descriptors: 0

======================================== ======================================

========================= Memory Estimate (per process) ========================

Descriptors
Descriptor 1 :: x 1 7382902 memory 59 MB
Descriptor 1 :: xPrime 1 0 memory 0 B
Descriptor 2 :: x 1 44 memory 352 B
Descriptor 2 :: xPrime 1 0 memory 0 B
Descriptor 3 :: x 1 44 memory 352 B
Descriptor 3 :: xPrime 1 0 memory 0 B
Descriptor 4 :: x 3 9039580 memory 216 MB
Descriptor 4 :: xPrime 3 0 memory 0 B
Descriptor 5 :: x 3 0 memory 0 B
Descriptor 5 :: xPrime 3 0 memory 0 B
Descriptor 6 :: x 3 0 memory 0 B
Descriptor 6 :: xPrime 3 0 memory 0 B
Descriptor 7 :: x 3 0 memory 0 B
Descriptor 7 :: xPrime 3 0 memory 0 B
Descriptor 8 :: x 3 0 memory 0 B
Descriptor 8 :: xPrime 3 0 memory 0 B
Descriptor 9 :: x 3 0 memory 0 B
Descriptor 9 :: xPrime 3 0 memory 0 B
Descriptor 10 :: x 1501 178167 memory 2139 MB
Descriptor 10 :: xPrime 1501 0 memory 0 B
Descriptor 11 :: x 1501 79 memory 948 KB
Descriptor 11 :: xPrime 1501 0 memory 0 B
Subtotal 2416 MB

Covariances
yY 21245 2965 memory 503 MB * 2
yy 21245 21245 memory 3610 MB
A 21245 24210 memory 4114 MB * 2
Subtotal 12 GB

Peak1 2920 MB
Peak2 12 GB
PEAK 12 GB

Free system memory 373 GB
Total system memory 405 GB

======================================== ======================================

========== Report on number of target properties found in training XYZ: ========

Number of target energies (property name: energy) found: 2965
Number of target local_properties (property name: local_property) found: 0
Number of target forces (property name: //IGNORE//) found: 0
Number of target virials (property name: //IGNORE//) found: 0
Number of target Hessian eigenvalues (property name: //IGNORE//) found: 0

================================= End of report ================================

===== Report on per-configuration/per-atom sigma (error parameter) settings ====

Number of per-configuration setting of energy_sigma found: 0
Number of per-configuration setting of force_sigma found: 0
Number of per-configuration setting of virial_sigma found: 0
Number of per-configuration setting of hessian_sigma found: 0
Number of per-configuration setting of local_propery_sigma found:0
Number of per-atom setting of force_atom_sigma found: 0
Number of per-component setting of force_component_sigma found: 0
Number of per-component setting of virial_component_sigma found: 0

================================= End of report ================================

WARNING: gpCoordinates_sparsify: number of data points (0) less than the number of sparse points (200), number of sparse points changed to 0
WARNING: gpCoordinates_sparsify: affected descriptor : angle_3b n_sparse=200 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=2.5 delta=0.05 Z=6 Z1=6 Z2=7
WARNING: gpCoordinates_sparsify: number of data points (0) less than the number of sparse points (200), number of sparse points changed to 0
WARNING: gpCoordinates_sparsify: affected descriptor : angle_3b n_sparse=200 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=2.5 delta=0.05 Z=6 Z1=7 Z2=7
WARNING: gpCoordinates_sparsify: number of data points (0) less than the number of sparse points (200), number of sparse points changed to 0
WARNING: gpCoordinates_sparsify: affected descriptor : angle_3b n_sparse=200 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=2.5 delta=0.05 Z=7 Z1=6 Z2=6
WARNING: gpCoordinates_sparsify: number of data points (0) less than the number of sparse points (200), number of sparse points changed to 0
WARNING: gpCoordinates_sparsify: affected descriptor : angle_3b n_sparse=200 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=2.5 delta=0.05 Z=7 Z1=6 Z2=7
WARNING: gpCoordinates_sparsify: number of data points (0) less than the number of sparse points (200), number of sparse points changed to 0
WARNING: gpCoordinates_sparsify: affected descriptor : angle_3b n_sparse=200 theta_uniform=1.0 sparse_method=uniform covariance_type=ard_se cutoff=2.5 delta=0.05 Z=7 Z1=7 Z2=7
Started CUR decomposition

However, there was no *.xml file generated as expected and only a *.idx file was created.
it would be appreciated if someone could help out with this and provide more details on the MPI running instruction.

Thank you so much in advance.

[Sideboard]

It looks like your calculation did not finish. Was it killed by the queuing system?

(When you want to quote raw text on Github issues, you should use three backticks ` above and below the quoted text, so it's not parsed as markdown.)

[GUANGZChen]

Hi,
After rewriting the inputs, I set up the MPI gap_fit run. However, it was stopped and here is the output messages. Could some help out with this? Thank you so much in advance.

libAtoms::Hello World: 2023-07-05 15:24:01
libAtoms::Hello World: git version https://github.com/libAtoms/QUIP.git,v0.9.14-dirty
libAtoms::Hello World: QUIP_ARCH linux_x86_64_gfortran_openmpi+openmp
libAtoms::Hello World: compiled on Jul 5 2023 at 06:44:18
libAtoms::Hello World: MPI parallelisation with 48 processes
libAtoms::Hello World: OpenMP parallelisation with 1 threads
WARNING: libAtoms::Hello World: environment variable OMP_STACKSIZE not set explicitly. The default value - system and compiler dependent - may be too small for some applications.
libAtoms::Hello World: MPI run with the same seed on each process
libAtoms::Hello World: Random Seed = -1633787871
libAtoms::Hello World: global verbosity = 0

Calls to system_timer will do nothing by default

MPI hostnames :: arc-c009
MPI host refs :: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
MPI my_host : 0
MPI hostname : arc-c009

================================ Input parameters ==============================

config_file =
atoms_filename = //MANDATORY//
at_file = CN_train_dataset.xyz
gap = "distance_2b n_sparse=15 theta_uniform=1.0 sparse_method=FILE sparse_file=1.input covariance_type=ard_se cutoff=4.5 delta=2.0 add_species=F:distance_2b n_sparse=15 theta_uniform=1.0 sparse_method=FILE sparse_file=2.input covariance_type=ard_se cutoff=4.5 delta=2.0 add_species=F:distance_2b n_sparse=15 theta_uniform=1.0 sparse_method=FILE sparse_file=3.input covariance_type=ard_se cutoff=4.5 delta=2.0 add_species=F:angle_3b n_sparse=200 theta_uniform=1.0 sparse_method=FILE sparse_file=4.input covariance_type=ard_se cutoff=2.5 delta=0.05 add_species=F:angle_3b n_sparse=0 theta_uniform=1.0 sparse_method=FILE sparse_file=5.input covariance_type=ard_se cutoff=2.5 delta=0.05 add_species=F:angle_3b n_sparse=0 theta_uniform=1.0 sparse_method=FILE sparse_file=6.input covariance_type=ard_se cutoff=2.5 delta=0.05 add_species=F:angle_3b n_sparse=0 theta_uniform=1.0 sparse_method=FILE sparse_file=7.input covariance_type=ard_se cutoff=2.5 delta=0.05 add_species=F:angle_3b n_sparse=0 theta_uniform=1.0 sparse_method=FILE sparse_file=8.input covariance_type=ard_se cutoff=2.5 delta=0.05 add_species=F:angle_3b n_sparse=0 theta_uniform=1.0 sparse_method=FILE sparse_file=9.input covariance_type=ard_se cutoff=2.5 delta=0.05 add_species=F:soap n_max=12 l_max=4 atom_sigma=0.5 zeta=4.0 cutoff=4.5 cutoff_transition_width=1.0 central_weight=1.0 n_sparse=45 delta=0.2 covariance_type=dot_product sparse_method=FILE sparse_file=10.input radial_decay=-0.5 add_species=F soap_exponent=1:soap n_max=12 l_max=4 atom_sigma=0.5 zeta=4.0 cutoff=4.5 cutoff_transition_width=1.0 central_weight=1.0 n_sparse=45 delta=0.2 covariance_type=dot_product sparse_method=FILE sparse_file=11.input radial_decay=-0.5 add_species=F soap_exponent=1"
e0 = C:-148.6822264562:N:-271.4749770587
local_property0 = 0.0
e0_offset = 0.0
e0_method = isolated
default_kernel_regularisation = //MANDATORY//
default_sigma = "0.001 0.01 0.05 0.0"
default_kernel_regularisation_local_property = 0.001
default_local_property_sigma = 0.001
sparse_jitter = 1.0e-8
hessian_displacement = 1.0e-2
hessian_delta = 1.0e-2
baseline_param_filename = quip_params.xml
core_param_file = quip_params.xml
baseline_ip_args =
core_ip_args =
energy_parameter_name = energy
local_property_parameter_name = local_property
force_parameter_name = force
virial_parameter_name = virial
stress_parameter_name = stress
hessian_parameter_name = hessian
config_type_parameter_name = config_type
kernel_regularisation_parameter_name = sigma
sigma_parameter_name = sigma
force_mask_parameter_name = force_mask
local_property_mask_parameter_name = local_property_mask
parameter_name_prefix =
config_type_kernel_regularisation =
config_type_sigma = Liquid:0.050:0.5:0.5:0.0:Liquid_Interface:0.050:0.5:0.5:0.0:Amorphous_Bulk:0.005:0.2:0.2:0.0:Amorphous_Surfaces:0.005:0.2:0.2:0.0:Surfaces:0.002:0.1:0.2:0.0:Dimer:0.002:0.1:0.2:0.0:Fullerenes:0.002:0.1:0.2:0.0:Defects:0.001:0.01:0.05:0.0:Crystalline_Bulk:0.001:0.01:0.05:0.0:Nanotubes:0.001:0.01:0.05:0.0:Graphite:0.001:0.01:0.05:0.0:Diamond:0.001:0.01:0.05:0.0:Graphene:0.001:0.01:0.05:0.0:Graphite_Layer_Sep:0.001:0.01:0.05:0.0:Single_Atom:0.0001:0.001:0.05:0.0
kernel_regularisation_is_per_atom = T
sigma_per_atom = T
do_copy_atoms_file = T
do_copy_at_file = F
sparse_separate_file = T
sparse_use_actual_gpcov = F
gap_file = gap_new.xml
gp_file = gap.xml
verbosity = NORMAL
rnd_seed = -1
openmp_chunk_size = 10000
do_ip_timing = F
template_file = template.xyz
sparsify_only_no_fit = F
dryrun = F
condition_number_norm =
linear_system_dump_file =
mpi_blocksize_rows = 0
mpi_blocksize_cols = 100
mpi_print_all = F
export_covariance = F

======================================== ======================================

============== Gaussian Approximation Potentials - Database fitting ============

Initial parsing of command line arguments finished.
Found 11 GAPs.
Descriptors have been parsed
XYZ file read
Unchanged GAP: {distance_2b n_sparse=15 theta_uniform=1.0 sparse_method=FILE sparse_file=1.input covariance_type=ard_se cutoff=4.5 delta=2.0 add_species=F}
Unchanged GAP: {distance_2b n_sparse=15 theta_uniform=1.0 sparse_method=FILE sparse_file=2.input covariance_type=ard_se cutoff=4.5 delta=2.0 add_species=F}
Unchanged GAP: {distance_2b n_sparse=15 theta_uniform=1.0 sparse_method=FILE sparse_file=3.input covariance_type=ard_se cutoff=4.5 delta=2.0 add_species=F}
Unchanged GAP: {angle_3b n_sparse=200 theta_uniform=1.0 sparse_method=FILE sparse_file=4.input covariance_type=ard_se cutoff=2.5 delta=0.05 add_species=F}
Unchanged GAP: {angle_3b n_sparse=0 theta_uniform=1.0 sparse_method=FILE sparse_file=5.input covariance_type=ard_se cutoff=2.5 delta=0.05 add_species=F}
Unchanged GAP: {angle_3b n_sparse=0 theta_uniform=1.0 sparse_method=FILE sparse_file=6.input covariance_type=ard_se cutoff=2.5 delta=0.05 add_species=F}
Unchanged GAP: {angle_3b n_sparse=0 theta_uniform=1.0 sparse_method=FILE sparse_file=7.input covariance_type=ard_se cutoff=2.5 delta=0.05 add_species=F}
Unchanged GAP: {angle_3b n_sparse=0 theta_uniform=1.0 sparse_method=FILE sparse_file=8.input covariance_type=ard_se cutoff=2.5 delta=0.05 add_species=F}
Unchanged GAP: {angle_3b n_sparse=0 theta_uniform=1.0 sparse_method=FILE sparse_file=9.input covariance_type=ard_se cutoff=2.5 delta=0.05 add_species=F}
Unchanged GAP: {soap n_max=12 l_max=4 atom_sigma=0.5 zeta=4.0 cutoff=4.5 cutoff_transition_width=1.0 central_weight=1.0 n_sparse=45 delta=0.2 covariance_type=dot_product sparse_method=FILE sparse_file=10.input radial_decay=-0.5 add_species=F soap_exponent=1}
Unchanged GAP: {soap n_max=12 l_max=4 atom_sigma=0.5 zeta=4.0 cutoff=4.5 cutoff_transition_width=1.0 central_weight=1.0 n_sparse=45 delta=0.2 covariance_type=dot_product sparse_method=FILE sparse_file=11.input radial_decay=-0.5 add_species=F soap_exponent=1}
SYSTEM ABORT: proc=47 Traceback (most recent call last)
File "/data/png-nanocrystals/wolf5816/quip/QUIP/src/GAP/gp_fit.f95", line 470 kind unspecified
count_entries_in_sparse_file: file 10.input contains 13518000 lines, not conforming with descriptor
size 391

MPI_ABORT was invoked on rank 47 in communicator MPI_COMM_WORLD
with errorcode 1.

NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes.
You may or may not see output from other processes, depending on
exactly when Open MPI kills them.

[gabor1]

In the latest version we don't require this two-step fitting. can I close this?