Closed tiandijunhao closed 4 years ago
Paste here the first few lines of your XYZ file.
-- Gábor
On 19 Apr 2020, at 07:02, tiandijunhao notifications@github.com wrote:
Hello,
When I put several datasets into a XYZ file for training, the following error is always reported:
fields[0] = Nb, length 2 fields[1] = 0.00000000, length 10 fields[2] = 0.00000000, length 10 fields[3] = 0.00000000, length 10 fields[4] = 0.00000000, length 10 fields[5] = -0.00000000, length 11 fields[6] = -0.00000000, length 11 fields[7] = , length 1 SYSTEM ABORT: Traceback (most recent call last) File "/BIGDATA1/caep_swcc_hygeng_1/software/QUIP-public/src/libAtoms/xyz.c", line 1047 kind IO incomplete row, frame 0 atom 0 - got 8/7 entries line 0
libAtoms::Hello World: 19/04/2020 13:51:41 libAtoms::Hello World: git version NOT_A_GIT_REPOSITORY libAtoms::Hello World: QUIP_ARCH linux_x86_64_gfortran libAtoms::Hello World: compiled on Apr 13 2020 at 14:14:16 libAtoms::Hello World: Random Seed = 49901384 libAtoms::Hello World: global verbosity = 0
Calls to system_timer will do nothing by default
================================ Input parameters ==============================
atoms_filename = //MANDATORY// at_file = Nb_54.xyz gap = "soap l_max=12 n_max=12 cutoff=5.0 cutoff_transition_width=1.0 delta=1.0 atom_sigma=0.5 zeta=4 config_type_n_sparse={elastics:100:md_300k:500:md_1200k:500:default:1000} sparse_method=cur_points covariance_type=dot_product" e0 = Nb:-0.70796320 local_property0 = 0.0 e0_offset = 0.0 e0_method = isolated default_kernel_regularisation = //MANDATORY// default_sigma = "0.005 0.2 1.0 0.0" sparse_jitter = 1e-12 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 = free_energy local_property_parameter_name = local_property force_parameter_name = forces virial_parameter_name = stress hessian_parameter_name = hessian config_type_parameter_name = config_type kernel_regularisation_parameter_name = sigma sigma_parameter_name = sigma config_type_kernel_regularisation = config_type_sigma = elastics:0.0001:0.01:0.01:0.0:md_300k:0.001:0.05:1.0:0.0:md_1200k:0.001:0.05:1.0:0.0 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_Nb.xml verbosity = NORMAL rnd_seed = -1 do_ip_timing = F template_file = template.xyz sparsify_only_no_fit = F
======================================== ======================================
============== Gaussian Approximation Potentials - Database fitting ============
Initial parsing of command line arguments finished. Found 1 GAPs. Descriptors have been parsed When I train different data sets separately, it can be carried out normally.
This is my command:
gap_fit energy_parameter_name=free_energy force_parameter_name=forces virial_parameter_name = stress do_copy_at_file=F sparse_separate_file=T gp_file=GAP_Nb.xml at_file=Nb_54.xyz \ gap={soap l_max=12 n_max=12 \ cutoff=5.0 cutoff_transition_width=1.0 delta=1.0 \ atom_sigma=0.5 zeta=4 config_type_n_sparse={elastics:100:md_300k:500:md_1200k:500:default:2000} \ sparse_method=cur_points \ covariance_type=dot_product} \ sparse_jitter=1e-12 \ default_sigma={0.005 0.2 1.0 0.0} \ config_type_sigma={elastics:0.0001:0.01:0.01:0.0:md_300k:0.001:0.05:1.0:0.0:md_1200k:0.001:0.05:1.0:0.0} \ e0={Nb:-0.70796320} How can I solve it?
Best regards, Wang
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or unsubscribe.
BTW your l_max=12 is likely overkill and l_max=6 is just as good and faster
-- Gábor
On 19 Apr 2020, at 07:02, tiandijunhao notifications@github.com wrote:
Hello,
When I put several datasets into a XYZ file for training, the following error is always reported:
fields[0] = Nb, length 2 fields[1] = 0.00000000, length 10 fields[2] = 0.00000000, length 10 fields[3] = 0.00000000, length 10 fields[4] = 0.00000000, length 10 fields[5] = -0.00000000, length 11 fields[6] = -0.00000000, length 11 fields[7] = , length 1 SYSTEM ABORT: Traceback (most recent call last) File "/BIGDATA1/caep_swcc_hygeng_1/software/QUIP-public/src/libAtoms/xyz.c", line 1047 kind IO incomplete row, frame 0 atom 0 - got 8/7 entries line 0
libAtoms::Hello World: 19/04/2020 13:51:41 libAtoms::Hello World: git version NOT_A_GIT_REPOSITORY libAtoms::Hello World: QUIP_ARCH linux_x86_64_gfortran libAtoms::Hello World: compiled on Apr 13 2020 at 14:14:16 libAtoms::Hello World: Random Seed = 49901384 libAtoms::Hello World: global verbosity = 0
Calls to system_timer will do nothing by default
================================ Input parameters ==============================
atoms_filename = //MANDATORY// at_file = Nb_54.xyz gap = "soap l_max=12 n_max=12 cutoff=5.0 cutoff_transition_width=1.0 delta=1.0 atom_sigma=0.5 zeta=4 config_type_n_sparse={elastics:100:md_300k:500:md_1200k:500:default:1000} sparse_method=cur_points covariance_type=dot_product" e0 = Nb:-0.70796320 local_property0 = 0.0 e0_offset = 0.0 e0_method = isolated default_kernel_regularisation = //MANDATORY// default_sigma = "0.005 0.2 1.0 0.0" sparse_jitter = 1e-12 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 = free_energy local_property_parameter_name = local_property force_parameter_name = forces virial_parameter_name = stress hessian_parameter_name = hessian config_type_parameter_name = config_type kernel_regularisation_parameter_name = sigma sigma_parameter_name = sigma config_type_kernel_regularisation = config_type_sigma = elastics:0.0001:0.01:0.01:0.0:md_300k:0.001:0.05:1.0:0.0:md_1200k:0.001:0.05:1.0:0.0 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_Nb.xml verbosity = NORMAL rnd_seed = -1 do_ip_timing = F template_file = template.xyz sparsify_only_no_fit = F
======================================== ======================================
============== Gaussian Approximation Potentials - Database fitting ============
Initial parsing of command line arguments finished. Found 1 GAPs. Descriptors have been parsed When I train different data sets separately, it can be carried out normally.
This is my command:
gap_fit energy_parameter_name=free_energy force_parameter_name=forces virial_parameter_name = stress do_copy_at_file=F sparse_separate_file=T gp_file=GAP_Nb.xml at_file=Nb_54.xyz \ gap={soap l_max=12 n_max=12 \ cutoff=5.0 cutoff_transition_width=1.0 delta=1.0 \ atom_sigma=0.5 zeta=4 config_type_n_sparse={elastics:100:md_300k:500:md_1200k:500:default:2000} \ sparse_method=cur_points \ covariance_type=dot_product} \ sparse_jitter=1e-12 \ default_sigma={0.005 0.2 1.0 0.0} \ config_type_sigma={elastics:0.0001:0.01:0.01:0.0:md_300k:0.001:0.05:1.0:0.0:md_1200k:0.001:0.05:1.0:0.0} \ e0={Nb:-0.70796320} How can I solve it?
Best regards, Wang
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or unsubscribe.
Paste here the first few lines of your XYZ file.
Thanks, I will try to set l_max. This is my file. Nb_54.zip
your file has Windows style newlines. that usually screws things up. blame Bill Gates. remove the extra carriage return characters using “dos2unix” utility or something similar
the internet says this command should also do it : perl -pi -e 's/\r\n/\n/g' *
-- Gábor
Gábor Csányi Professor of Molecular Modelling Engineering Laboratory, University of Cambridge Pembroke College Cambridge
Pembroke College supports CARA. A Lifeline to Academics at Risk. http://www.cara.ngo/
On 19 Apr 2020, at 10:32, tiandijunhao notifications@github.com wrote:
Paste here the first few lines of your XYZ file.
Thanks, I will try to set l_max. This is my file. Nb_54.zip
— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.
here is your example file converted. Nb_54.xyz.zip
here is your example file converted. Nb_54.xyz.zip
Thank you very much for your help.
Best regards, Wang
did it work?
-- Gábor
Gábor Csányi Professor of Molecular Modelling Engineering Laboratory, University of Cambridge Pembroke College Cambridge
Pembroke College supports CARA. A Lifeline to Academics at Risk. http://www.cara.ngo/
On 19 Apr 2020, at 13:10, tiandijunhao notifications@github.com wrote:
here is your example file converted. Nb_54.xyz.zip
Thank you very much for your help.
Best regards, Wang
— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.
did it work? … -- Gábor Gábor Csányi Professor of Molecular Modelling Engineering Laboratory, University of Cambridge Pembroke College Cambridge Pembroke College supports CARA. A Lifeline to Academics at Risk. http://www.cara.ngo/ On 19 Apr 2020, at 13:10, tiandijunhao @.***> wrote: here is your example file converted. Nb_54.xyz.zip Thank you very much for your help. Best regards, Wang — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.
Yes, very thanks.
Hello,
When I put several datasets into a XYZ file for training, the following error is always reported:
When I train different data sets separately, it can be carried out normally.
This is my command:
How can I solve it?
Best regards, Wang