add some additional NNP and helium related data/inputs

This commit is contained in:
Harald Forbert 2025-06-27 15:26:31 +02:00 committed by hforbert
parent c9ed8e0e9c
commit 93bc24a11a
15 changed files with 7081 additions and 3 deletions

View file

@ -0,0 +1,13 @@
#### Neural Network IP for Helium and Protonated Water (H<sub>3</sub>O<sup>+</sup>-He)
This is an updated version of the originally developed NNP to use CCSD(T\*)-F12a/AVTZ-cp
calculations.
Original publication of the methodology: "High-dimensional neural network potentials for solvation:
The case of protonated water clusters in helium" Reference:
[J. Chem. Phys. 148 (10), 102310 (2018)](https://doi.org/10.1063/1.4996819). Preprint available on
[arxiv](https://arxiv.org/abs/2103.13123).
Publication of the updated version: "Properties of hydrogen bonding at ultra-low temperatures in
superfluid quantum solvents.", C. Schran, PhD thesis, Ruhr-Universitaet Bochum, 2019.
(https://doi.org/10.13154/294-9852)

View file

@ -0,0 +1,304 @@
## #############################################################
### This is the input file for RuNNer (version 0_44 and upwards)
### #############################################################
### General remarks:
### - commands can be switched off by using the # character at the BEGINNING of the line
### - the input file can be structured by blank lines and comment lines
### - the order of the keywords is arbitrary
### - if keywords are missing, default values will be used and written to runner.out
### - if mandatory keywords or keyword options are missing, RuNNer will stop with an error message
element_decoupled_kalman
########################################################################################################################
### general keywords
########################################################################################################################
nn_type_short 1 # 1=Behler-Parrinello, 2=Pair NN
runner_mode 2 # 1=calculate symmetry functions, 2=fitting mode, 3=predicition mode (mode)
number_of_elements 2 # number of elements (nelem) MODE1+2+3+4
elements O H # specification of elements (element) MODE1+2+3+4
random_seed 22 # seed for initial random weight parameters and train/test splitting (iseed) MODE1+2
random_number_type 1 # 1=ran0, 2=ran1, 3=ran2, 4=ran3
remove_atom_energies # remove atomic energies before fitting (lremoveatomenergies) MODE1+2+3+4
atom_energy O -74.94518524 # free atom reference energy (atomic zora)
atom_energy H -0.45890771 # free atom reference energy (atomic zora)
#energy_threshold 100.0d0 # energythreshold for fitting data in Ha per atom (fitethres) MODE1
#bond_threshold 0.4d0 # threshold for the shortest bond in structure (rmin) MODE1+2+3
########################################################################################################################
### NN structure of the short-range NN
########################################################################################################################
use_short_nn # use NN for short range interactions (lshort)
global_hidden_layers_short 2 # number of hidden layers (num_layersshort-1)
global_nodes_short 30 30 # number of nodes in hidden layers (nodes_short)
global_activation_short t t l # activation functions (actfunc_short)
########################################################################################################################
### symmetry function generation ( mode 1):
########################################################################################################################
#use_atom_energies # use atomic energies for fitting (not implemented) (luseatomenergies) MODE1+2+3+4
use_atom_charges # use atomic charges for fitting(set always true!) (luseatomcharges) MODE1+2+3+4
test_fraction 0.1 # threshold for splitting between fitting and test set (splitthres) MODE1
#CAUTION: don't forget use_short_forces below (if you want to generate the training files for the forces)
########################################################################################################################
### symmetry function definitions (all modes):
########################################################################################################################
# INFO: not yet implemented in this file
# INFO: symfunction format: reference atom, type, neighbor element 1 (and neighbor element 2), symfunction parameters
#
# GLOBAL SYMMETRY FUNCTIONS FOR SHORT RANGE NN
# SAMPLE TYPE 1: global_symfunction_short 1 2.0 ! type funccutoff
# SAMPLE TYPE 2: global_symfunction_short 2 7.14214 0.0 11.338 ! type eta rshift funccutoff
# SAMPLE TYPE 3: global_symfunction_short 3 0.03571 -1.0 16.0 7.55891 ! type eta lambda zeta funccutoff
# SAMPLE TYPE 4: global_symfunction_short 4 7.14214 11.338 ! type eta funccutoff
# SAMPLE TYPE 5: global_symfunction_short O 5 1.000 ! central_atom type eta ! CARTESIAN COORDINATES
# SAMPLE TYPE 6: global_symfunction_short O 6 11.338 ! central_atom type funccutoff ! BOND LENGTH
#
# ELEMENT-SPECIFIC SYMMETRY FUNCTIONS FOR SHORT RANGE NN
# SAMPLE TYPE 1: element_symfunction_short O 1 2.0 ! central_atom type funccutoff
# SAMPLE TYPE 2: element_symfunction_short O 2 7.14214 0.0 11.338 ! central_atom type eta rshift funccutoff
# SAMPLE TYPE 3: element_symfunction_short O 3 0.03571 -1.0 16.0 7.55891 ! central_atom type eta lambda zeta funccutoff
# SAMPLE TYPE 4: element_symfunction_short O 4 7.14214 11.338 ! central_atom type eta funccutoff
# SAMPLE TYPE 5: element_symfunction_short O 5 1.000 ! central_atom type eta ! CARTESIAN COORDINATES
# SAMPLE TYPE 6: element_symfunction_short O 6 11.338 ! central_atom type funccutoff ! BOND LENGTH
#
# CUSTOMIZED SYMMETRY FUNCTIONS FOR SHORT RANGE NN
# SAMPLE TYPE 1: symfunction_short O 1 O 2.0 ! central_atom type neighbor_atom funccutoff
# SAMPLE TYPE 2: symfunction_short O 2 O 7.14214 0.0 11.338 ! central_atom type neighbor_atom eta rshift funccutoff
# SAMPLE TYPE 3: symfunction_short O 3 Zn Zn 0.03571 -1.0 16.0 7.55891 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
# SAMPLE TYPE 4: symfunction_short O 4 O 7.14214 11.338 ! central_atom type neighbor_atom eta funccutoff
# SAMPLE TYPE 5: symfunction_short O 5 1.000 ! central_atom type eta ! CARTESIAN COORDIATES
# SAMPLE TYPE 6: symfunction_short O 6 O 11.338 ! central_atom type neighbor_atom funccutoff ! BOND LENGTH
cutoff_type 1
symfunction_short H 2 O 0.00100 0.0 4.000 ! eta rshift funccutoff
symfunction_short H 2 O 0.01000 0.0 4.000 ! eta rshift funccutoff
symfunction_short H 2 O 0.03000 0.0 4.000 ! eta rshift funccutoff
symfunction_short H 2 O 0.06000 0.0 4.000 ! eta rshift funccutoff
symfunction_short H 2 O 0.15000 0.0 4.000 ! eta rshift funccutoff
symfunction_short H 2 O 0.15000 1.5 4.000 ! eta rshift funccutoff
symfunction_short H 2 O 0.30000 1.5 4.000 ! eta rshift funccutoff
symfunction_short H 2 O 0.60000 1.5 4.000 ! eta rshift funccutoff
symfunction_short H 2 H 0.00100 0.0 6.000 ! eta rshift funccutoff
symfunction_short H 2 H 0.01000 0.0 6.000 ! eta rshift funccutoff
symfunction_short H 2 H 0.03000 0.0 6.000 ! eta rshift funccutoff
symfunction_short H 2 H 0.06000 0.0 6.000 ! eta rshift funccutoff
symfunction_short H 2 H 0.15000 0.0 6.000 ! eta rshift funccutoff
symfunction_short H 2 H 0.15000 1.5 6.000 ! eta rshift funccutoff
symfunction_short H 2 H 0.30000 1.5 6.000 ! eta rshift funccutoff
symfunction_short H 2 H 0.00100 0.0 12.000 ! eta rshift funccutoff
symfunction_short H 2 O 0.00100 0.0 12.000 ! eta rshift funccutoff
symfunction_short H 2 H 0.01000 0.0 12.000 ! eta rshift funccutoff
symfunction_short H 2 O 0.01000 0.0 12.000 ! eta rshift funccutoff
symfunction_short H 2 H 0.03000 0.0 12.000 ! eta rshift funccutoff
symfunction_short H 2 O 0.03000 0.0 12.000 ! eta rshift funccutoff
symfunction_short H 2 H 0.06000 0.0 12.000 ! eta rshift funccutoff
symfunction_short H 2 O 0.06000 0.0 12.000 ! eta rshift funccutoff
symfunction_short H 2 H 0.15000 0.0 12.000 ! eta rshift funccutoff
symfunction_short H 2 O 0.15000 0.0 12.000 ! eta rshift funccutoff
symfunction_short H 2 H 0.15000 1.5 12.000 ! eta rshift funccutoff
symfunction_short H 2 O 0.15000 1.5 12.000 ! eta rshift funccutoff
symfunction_short H 2 H 0.30000 1.5 12.000 ! eta rshift funccutoff
symfunction_short H 2 O 0.30000 1.5 12.000 ! eta rshift funccutoff
symfunction_short H 2 O 0.60000 1.5 12.000 ! eta rshift funccutoff
symfunction_short H 3 H O 0.010 -1.0 1.0 6.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short H 3 H O 0.010 1.0 1.0 6.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short H 3 H O 0.010 -1.0 2.0 6.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short H 3 H O 0.010 1.0 2.0 6.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short H 3 H O 0.010 1.0 3.0 6.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short H 3 H O 0.010 -1.0 1.0 8.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short H 3 H O 0.010 1.0 1.0 8.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short H 3 H O 0.010 -1.0 2.0 8.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short H 3 H O 0.010 1.0 2.0 8.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short H 3 H O 0.010 1.0 3.0 8.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short H 3 O O 0.001 -1.0 4.0 12.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short H 3 O O 0.001 1.0 4.0 12.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short H 3 H O 0.010 -1.0 4.0 12.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short H 3 H O 0.010 1.0 4.0 12.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short H 3 H O 0.030 -1.0 1.0 12.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short H 3 O O 0.030 -1.0 1.0 12.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short H 3 H O 0.030 1.0 1.0 12.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short H 3 O O 0.030 1.0 1.0 12.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short H 3 H O 0.070 -1.0 1.0 12.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short H 3 H O 0.070 1.0 1.0 12.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short H 3 H O 0.200 1.0 1.0 12.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
# O:
symfunction_short O 2 H 0.00100 0.0 4.000 ! eta rshift funccutoff
symfunction_short O 2 H 0.01000 0.0 4.000 ! eta rshift funccutoff
symfunction_short O 2 H 0.03000 0.0 4.000 ! eta rshift funccutoff
symfunction_short O 2 H 0.06000 0.0 4.000 ! eta rshift funccutoff
symfunction_short O 2 H 0.15000 0.0 4.000 ! eta rshift funccutoff
symfunction_short O 2 H 0.15000 1.5 4.000 ! eta rshift funccutoff
symfunction_short O 2 H 0.30000 1.5 4.000 ! eta rshift funccutoff
symfunction_short O 2 H 0.60000 1.5 4.000 ! eta rshift funccutoff
symfunction_short O 2 H 0.00100 0.0 12.000 ! eta rshift funccutoff
symfunction_short O 2 O 0.00100 0.0 12.000 ! eta rshift funccutoff
symfunction_short O 2 H 0.01000 0.0 12.000 ! eta rshift funccutoff
symfunction_short O 2 O 0.01000 0.0 12.000 ! eta rshift funccutoff
symfunction_short O 2 H 0.03000 1.5 12.000 ! eta rshift funccutoff
symfunction_short O 2 O 0.03000 0.0 12.000 ! eta rshift funccutoff
symfunction_short O 2 O 0.06000 1.5 12.000 ! eta rshift funccutoff
symfunction_short O 2 H 0.06000 0.0 12.000 ! eta rshift funccutoff
symfunction_short O 2 O 0.06000 0.0 12.000 ! eta rshift funccutoff
symfunction_short O 2 O 0.09000 0.0 12.000 ! eta rshift funccutoff
symfunction_short O 2 O 0.15000 0.0 12.000 ! eta rshift funccutoff
symfunction_short O 2 H 0.15000 0.0 12.000 ! eta rshift funccutoff
symfunction_short O 2 H 0.30000 0.0 12.000 ! eta rshift funccutoff
symfunction_short O 2 H 0.60000 0.0 12.000 ! eta rshift funccutoff
symfunction_short O 3 H H 0.010 -1.0 1.0 6.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short O 3 H H 0.010 1.0 1.0 6.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short O 3 H H 0.010 -1.0 2.0 6.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short O 3 H H 0.010 1.0 2.0 6.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short O 3 H H 0.010 -1.0 3.0 6.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short O 3 H H 0.010 -1.0 1.0 8.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short O 3 H H 0.010 1.0 1.0 8.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short O 3 H H 0.010 -1.0 2.0 8.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short O 3 H H 0.010 1.0 2.0 8.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short O 3 H H 0.010 -1.0 3.0 8.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short O 3 H O 0.001 -1.0 4.0 12.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short O 3 O O 0.001 -1.0 4.0 12.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short O 3 H O 0.001 1.0 4.0 12.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short O 3 O O 0.001 1.0 4.0 12.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short O 3 H H 0.010 -1.0 4.0 12.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short O 3 H H 0.010 1.0 4.0 12.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short O 3 H H 0.030 -1.0 1.0 12.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short O 3 H O 0.030 -1.0 1.0 12.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short O 3 O O 0.030 -1.0 1.0 12.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short O 3 H H 0.030 1.0 1.0 12.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short O 3 H O 0.030 1.0 1.0 12.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short O 3 O O 0.030 1.0 1.0 12.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short O 3 H H 0.070 -1.0 1.0 12.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
symfunction_short O 3 H H 0.070 1.0 1.0 12.00000 ! central_atom type neighbor_atom1 neighbor_atom2 eta lambda zeta funccutoff
#
# SYMMETRY FUNCTIONS FOR ELECTROSTATIC NN HAVE THE SAME FORMAT, BUT REPLACE "short" by "electrostatic"
########################################################################################################################
### fitting (mode 2):general inputs for short range AND electrostatic part:
########################################################################################################################
epochs 80 # number of epochs (nepochs) MODE2
points_in_memory 2000 # max number of structures in memory (nblock) MODE2
mix_all_points # training with random order of points (lrandomtrain) MODE2
scale_symmetry_functions # scale symmetry functions (lscalesym) MODE2+3+4
center_symmetry_functions # remove center of mass of structure function values (lcentersym) MODE2+3+4
#fix_weights # fix some weights (lfixweights) MODE2
#growth_mode 11 6 # growth mode (lgrowth,ngrowth,growthstep) MODE2
#use_damping 0.00001d0 # use weight decay (ldampw,dampw) MODE2
#pdate_single_element 8 # do weight update just for one element (lupdatebyelement,elemupdate) MODE2
fitting_unit eV # unit for error output in mode 2 (eV or Ha)
#joint_energy_force_update # for each atom do one update for energy and averaged forces together (not yet working well)
########################################################################################################################
### fitting options ( mode 2): short range part only:
########################################################################################################################
print_mad
optmode_short_energy 1 # optimization mode short range energies(optmodee, 1=Kalman filter, 2=conjugate gradient, 3=steepest descent)
optmode_short_force 1 # optimization mode short range forces (optmodef, 1=Kalman filter, 2=conjugate gradient, 3=steepest descent)
short_energy_error_threshold 0.8 # threshold of adaptive Kalman filter short E (kalmanthreshold) MODE2
short_force_error_threshold 1.0 # threshold of adaptive Kalman filter short F (kalmanthresholdf) MODE2
kalman_lambda_short 0.98000 # Kalman parameter short E/F (kalmanlambda) MODE2
kalman_nue_short 0.99870 # Kalman parameter short E/F (kalmannue) MODE2
#steepest_descent_step_energy_short 0.01d0 # step size for steepest descent energy (steepeststepe) MODE2
#steepest_descent_step_force_short 0.01d0 # step size for steepest descent force (steepeststepf) MODE2
#use_old_weights_short # restart fitting with old weight parameters for short (luseoldweightsshort) MODE2
#update_worst_short_energies 0.1d0 # percentage of the worst energies used for update (worste) MODE2
#update_worst_short_forces 0.1d0 # percentage of the worst forces used for update (worstf) MODE2
#force_update_scaling -1.0d0 # scaling factor for the force update (negative value means automatic scaling) (scalefactorf) MODE2
#short_energy_group 1 # group energies for update (nenergygroup) MODE2
#short_energy_fraction 1.00 # percentage of energies used for fitting 100%=1.0 (energyrnd) MODE2
#short_force_group 1 # group forces for update (nforcegroup) MODE2
#short_force_fraction 0.05 # percentage of forces used for fitting 100%=1.0 (forcernd) MODE2
#use_short_forces # use forces for fitting (luseforces) MODE2
#weight_constraint H all fixed # "all" switch
#weight_constraint O interlayer 1 2 free # "interlayer" layer1 layer2 switch
#weight_constraint Zn bias 1 2 free # "bias layer" node switch
#weight_constraint Zn weight 1 3 2 3 free # "weight" layer1 node1 layer2 node2 switch
#weight_constraint Zn node 1 1 free # "node" layer node switch
weights_min -1.0 # minimum value for initial random short range weights
weights_max 1.0 # maximum value for initial random short range weights
precondition_weights # precondition initial weights (lprecond)
#normalize_nodes # normalize input of nodes
repeated_energy_update # calculate error of
nguyen_widrow_weights_short # initialize short
########################################################################################################################
### fitting ( mode 2): electrostatic part only:
########################################################################################################################
#optmode_charge 1 # optimization mode atomic charges (optmodeq, 1=Kalman filter, 2=conjugate gradient, 3=steepest descent)
#charge_error_threshold 1.0 # threshold of adaptive Kalman filter charge (kalmanthresholde) MODE2
#kalman_lambda_charge 0.98000 # Kalman parameter charge (kalmanlambdae) MODE2
#kalman_nue_charge 0.99870 # Kalman parameter charge (kalmannuee) MODE2
#steepest_descent_step_charge 0.01d0 # step size for steepest descent charge (steepeststepq) MODE2
#use_old_weights_charge # restart fitting with old weight parameters for charge(luseoldweightscharge) MODE2
#update_worst_charges 0.1d0 # percentage of the worst charges used for update (worstq) MODE2
#charge_group 20 # group charges for update (nchargegroup) MODE2
#charge_fraction 1.00 # percentage of charges used for fitting 100%=1.0 (chargernd) MODE2
#weighte_constraint O all fixed # "all" switch
#weighte_constraint O interlayer 1 2 free # "interlayer" layer1 layer2 switch
#weighte_constraint Zn bias 1 2 free # "bias layer" node switch
#weighte_constraint Zn weight 1 3 2 3 free # "weight" layer1 node1 layer2 node2 switch
#weighte_constraint Zn node 2 1 free # "node" layer node switch
#precondition_weights # precondition initial weights (lprecond)
#nguyen_widrow_weights_ewald # initialize short
#weightse_min -1.0 # minimum value for initial random charge weights
#weightse_max 1.0 # maximum value for initial random charge weights
########################################################################################################################
### options for charge constraint in mode 2 (not tested! not parallel!)
########################################################################################################################
#use_charge_constraint # use total charge constraint (lchargeconstraint) MODE2
#total_charge_error_threshold 0.0000001 # threshold of adaptive Kalman filter charge constraint(kalmanthresholdc) MODE2
#kalman_lambda_charge_constraint 0.98000 # Kalman parameter charge constraint (kalmanlambdac) MODE2
#kalman_nue_charge_constraint 0.99870 # Kalman parameter charge constraint (kalmannuec) MODE2
########################################################################################################################
### output options for mode 2 (fitting):
########################################################################################################################
#write_weights_epoch 1 # write set of weight parameters every ith epoch (iwriteweight) MODE2
#write_temporary_weights # write temporary weights each data block (lwritetmpweights) MODE2
#write_trainpoints # write trainpoints.out and testpoints.out files (lwritetrainpoints) MODE2
#write_traincharges # write traincharges.out and testcharges.out files (lwritetraincharges) MODE2
#write_trainforces # write trainforces.out and testforces.out files (lwritetrainforces) MODE2
########################################################################################################################
### output options for mode 3 (prediction):
########################################################################################################################
calculate_forces # calculate forces (ldoforces) MODE3
#calculate_stress # calculate stress (ldostress)(not fully implemented) MODE3
#write_pdb # write predicted structure in pdb format (lwritepdb) MODE3
#write_xyz # write predicted structure in xyz format (lwritexyz) MODE3
#write_pov # write predicted structure in pov format (lwritepov) MODE3
#write_pwscf # write predicted structure in pwscf format (lwritepw) MODE3
########################################################################################################################
### output options for debug.out file
########################################################################################################################
#print_all_short_weights
#print_all_electrostatic_weights
########################################################################################################################
### options for mode 4 (not yet working)
########################################################################################################################
#symfunction_check_threshold 0.001d0 # threshold for symmetry function check (symthres) MODE4
#charge_check_threshold 0.0002d0 # threshold for atomic charge check (chargethres) MODE4
#force_check_threshold 0.0003d0 # threshold for atomic force check (forcethres) MODE4
#energy_check_threshold 0.0004d0 # threshold for atomic energy check (energythres)(not yet implemented) MODE4

View file

@ -0,0 +1,5 @@
#### A Neural Network PES for Protonated Water Cluster (H<sub>2</sub>O, H<sub>3</sub>O<sup>+</sup>, H<sub>5</sub>O<sub>2</sub><sup>+</sup>, H<sub>7</sub>O<sub>3</sub><sup>+</sup>, H<sub>9</sub>O<sub>4</sub><sup>+</sup>)
Reference:
[J. Chem. Theo. Comput. 16 88-99 (2020)](https://pubs.acs.org/doi/abs/10.1021/acs.jctc.9b00805) -
Preprint available on [arxiv](https://arxiv.org/abs/1908.08734).

View file

@ -0,0 +1,98 @@
1 1 0.165866306 1.150251278 0.596258580
1 2 0.153557832 1.117121138 0.576151625
1 3 0.129377333 1.046884305 0.534099497
1 4 0.100054708 0.949743512 0.477212046
1 5 0.046278101 0.709253284 0.342606118
1 6 0.123261629 1.137980874 0.576333141
1 7 0.090819048 1.122213144 0.556886439
1 8 0.049303191 1.091394050 0.523892578
1 9 0.237444578 2.231342694 0.927071145
1 10 0.205030225 2.056354519 0.839559591
1 11 0.147966614 1.724095491 0.676015706
1 12 0.090715711 1.339980859 0.492654192
1 13 0.020904454 0.674654564 0.200282009
1 14 0.091811226 1.615739187 0.581957504
1 15 0.034925820 1.254779643 0.384809986
1 16 0.751433278 5.981202053 2.941799272
1 17 0.866855577 3.527665557 1.839341041
1 18 0.655964897 5.212588917 2.391783325
1 19 0.808725291 3.318433461 1.595858867
1 20 0.485006608 3.844183495 1.636759612
1 21 0.693123352 2.897164460 1.256359093
1 22 0.308353043 2.448769562 1.047554995
1 23 0.541873300 2.364074731 0.984905817
1 24 0.066848987 1.064338339 0.372634018
1 25 0.238344311 1.287189548 0.628012203
1 26 0.313718706 2.773703678 1.152109163
1 27 0.639670343 2.954118854 1.132760729
1 28 0.108662427 1.969479056 0.705507231
1 29 0.467086800 2.460981749 1.031731943
1 30 0.253524242 1.787085849 0.936409559
1 31 0.001604317 0.488283013 0.085071113
1 32 0.152752128 1.892027291 0.789749004
1 33 0.000007644 0.228082000 0.011503735
1 34 0.137125573 1.708147539 0.716181627
1 35 0.123097616 1.618411278 0.650668981
1 36 0.004782681 2.045360990 0.265908405
1 37 0.482050526 5.378313230 1.738291957
1 38 0.000029793 1.165464167 0.104040644
1 39 0.398656831 4.498416406 1.576424196
1 40 0.312949092 4.024125448 1.439974415
1 41 0.000000000 1.638977784 0.277336855
1 42 0.000000000 2.345366024 0.418105282
1 43 0.000000002 2.411407835 0.355802422
1 44 0.291716290 7.076306974 3.044028766
1 45 0.005110613 2.911861916 0.364993787
1 46 0.000000000 1.737800060 0.124517952
1 47 0.559743272 6.287422066 2.055158900
1 48 0.000000000 1.015661692 0.109207393
1 49 0.002417389 0.540743798 0.102010577
1 50 0.174404163 1.944323477 0.811707172
1 51 0.005005377 0.323217524 0.081737304
2 1 0.657274824 2.035556327 1.387251737
2 2 0.624101027 1.990896131 1.340471012
2 3 0.556439880 1.899792923 1.242632777
2 4 0.468851968 1.770966345 1.110278766
2 5 0.282152111 1.434986601 0.797105398
2 6 0.583874553 2.036121388 1.340893327
2 7 0.517651379 2.034158158 1.295648745
2 8 0.411264390 2.030246850 1.218885418
2 9 1.816106690 6.423665448 4.279400147
2 10 0.000000000 2.189541276 1.033133655
2 11 1.732440032 5.578661499 3.712915940
2 12 0.000000000 1.886951428 0.795224774
2 13 0.000000000 1.356753513 0.466887254
2 14 1.791632237 5.344368008 3.536483686
2 15 1.333358805 3.319570974 2.291476135
2 16 0.000000000 0.828594178 0.224874979
2 17 0.000000000 1.500470197 0.470830607
2 18 0.000000000 0.507075315 0.112303320
2 19 0.747962094 2.131986148 1.461129531
2 20 0.000000000 0.191093591 0.029261491
2 21 0.312388927 1.426100462 0.817025816
2 22 0.058038559 0.709369295 0.280292723
2 23 0.122039701 1.730637203 0.670932369
2 24 0.009773230 1.214363043 0.346743324
2 25 0.044741762 1.262354388 0.453445211
2 26 0.000246995 0.691187474 0.129256166
2 27 0.010857802 0.943006651 0.311434796
2 28 0.270028501 3.357922699 1.405435122
2 29 0.019403340 2.831690039 0.926045318
2 30 0.065567757 2.241016565 0.960026987
2 31 0.000490372 1.841475028 0.480637183
2 32 0.015911795 1.533406115 0.671402062
2 33 0.000000000 2.357307473 0.519029987
2 34 0.000000000 0.389055598 0.018429503
2 35 0.000000000 8.530817821 3.314077670
2 36 0.000000000 0.828396152 0.126674366
2 37 0.005022603 2.426018991 1.048668182
2 38 0.000000487 3.848184040 1.363272649
2 39 0.277493995 3.784158320 1.611777182
2 40 0.000000000 1.397576616 0.217060308
2 41 0.000000000 0.138455586 0.004478793
2 42 0.020733761 3.520744462 1.203579386
2 43 0.000000000 3.596744231 0.870507742
2 44 0.000000000 0.379638470 0.012041681
2 45 0.121234911 1.712147007 0.676217801
2 46 0.009807347 1.313200307 0.386707395
-0.1693317533 -0.0480493525

File diff suppressed because it is too large Load diff

File diff suppressed because it is too large Load diff

View file

@ -31,10 +31,10 @@
SAMPLING_METHOD WORM
SOLUTE_INTERACTION NNP
&NNP
NNP_INPUT_FILE_NAME ./input.nn
SCALE_FILE_NAME ./scaling.data
NNP_INPUT_FILE_NAME NNP/protH2OHelium-interaction-nnp/input.nn
SCALE_FILE_NAME NNP/protH2OHelium-interaction-nnp/scaling.data
&MODEL
WEIGHTS ./weights
WEIGHTS NNP/protH2OHelium-interaction-nnp/weights
&END MODEL
&SR_CUTOFF
ELEMENT H

View file

@ -0,0 +1,95 @@
&GLOBAL
PRINT_LEVEL low
PROJECT_NAME he32_only_worm_p50
RUN_TYPE PINT
&END GLOBAL
&MOTION
&PINT
# NUM_STEPS 400000
NUM_STEPS 1000
&HELIUM
CELL_SHAPE OCTAHEDRON
DENSITY 0.0218457
HELIUM_ONLY
NATOMS 32
NBEADS 50
N_OUTER 10000
PERIODIC T
POTENTIAL_FILE_NAME helium_aziz95_80k.potx
PRESAMPLE T
SAMPLING_METHOD WORM
&PRINT
&MOMENT_OF_INERTIA OFF
&END MOMENT_OF_INERTIA
&PROJECTED_AREA_2_AVG OFF
&END PROJECTED_AREA_2_AVG
&RDF ON
&EACH
PINT 1000
&END EACH
&END RDF
&WINDING_NUMBER ON
&EACH
PINT 1
&END EACH
&END WINDING_NUMBER
&END PRINT
&RDF T
HE_HE T
MAXR 15.0
NBIN 300
SOLUTE_HE F
&END RDF
&END HELIUM
&END PINT
&PRINT
&RESTART
&EACH
PINT 1000
&END EACH
&END RESTART
&RESTART_HISTORY OFF
&END RESTART_HISTORY
&END PRINT
&END MOTION
#dummy force_eval
#will be ignore due to MOTION&PINT&HELIUM&HELIUM_ONLY
#but some valid force_eval is needed.
&FORCE_EVAL
METHOD FIST
&MM
&FORCEFIELD
&CHARGE
ATOM Ar
CHARGE 0.0
&END CHARGE
&NONBONDED
&LENNARD-JONES
ATOMS Ar Ar
EPSILON 119.8
RCUT 8.4
SIGMA 3.405
&END LENNARD-JONES
&END NONBONDED
&END FORCEFIELD
&POISSON
&EWALD
EWALD_TYPE none
&END EWALD
&END POISSON
&END MM
&SUBSYS
&CELL
ABC 8.0 8.0 8.0
PERIODIC NONE
&END CELL
&COORD
Ar 0.0 0.0 0.0
Ar 4.0 4.0 4.0
&END COORD
&TOPOLOGY
&END TOPOLOGY
&END SUBSYS
&END FORCE_EVAL

View file

@ -0,0 +1,193 @@
&GLOBAL
PRINT_LEVEL SILENT
PROJECT_NAME "droplet_h3o+_8He_1K"
RUN_TYPE PINT
&END GLOBAL
&MOTION
&PINT
DT 0.1
HARM_INT EXACT
NRESPA 1
# NUM_STEPS 1000000
NUM_STEPS 1000
P 160
PROPAGATOR RPMD
TEMP 1.0000000000000000E+000
TRANSFORMATION NORMAL
&HELIUM T
CELL_SHAPE OCTAHEDRON
CELL_SIZE 1.9115700000000000E+001
DROPLET_RADIUS 15.0
GET_FORCES LAST
NATOMS 8 # 8 helium atoms
NBEADS 80
NUM_ENV 1
N_OUTER 1000
PERIODIC F
POTENTIAL_FILE_NAME ./helium_aziz95_80k.potx
PRESAMPLE T
RNG_SEED 92404581
SAMPLING_METHOD WORM
SOLUTE_INTERACTION NNP
&NNP
NNP_INPUT_FILE_NAME NNP/protH2OHelium-interaction-nnp/input.nn
SCALE_FILE_NAME NNP/protH2OHelium-interaction-nnp/scaling.data
&MODEL
WEIGHTS NNP/protH2OHelium-interaction-nnp/weights
&END MODEL
&SR_CUTOFF
ELEMENT "H"
RADIUS 1.25
&END SR_CUTOFF
&SR_CUTOFF
ELEMENT "He"
RADIUS 0.0000000000000000E+000
&END SR_CUTOFF
&SR_CUTOFF
ELEMENT "O"
RADIUS 2.05
&END SR_CUTOFF
&END NNP
&PRINT
&COORDINATES ON
&EACH
PINT 100
&END EACH
&END COORDINATES
&ENERGY SILENT
&EACH
PINT 100
&END EACH
&END ENERGY
&MOMENT_OF_INERTIA ON
&EACH
PINT 1
&END EACH
&END MOMENT_OF_INERTIA
&MOMENT_OF_INERTIA_AVG ON
&EACH
PINT 100
&END EACH
&END MOMENT_OF_INERTIA_AVG
&PERM ON
&EACH
PINT 100
&END EACH
&END PERM
&PLENGTH OFF
&END PLENGTH
&PROJECTED_AREA ON
&EACH
PINT 1
&END EACH
&END PROJECTED_AREA
&PROJECTED_AREA_2_AVG ON
&EACH
PINT 100
&END EACH
&END PROJECTED_AREA_2_AVG
&RDF ON
&EACH
PINT 100
&END EACH
&END RDF
&WINDING_NUMBER OFF
&END WINDING_NUMBER
&WINDING_NUMBER_2_AVG OFF
&END WINDING_NUMBER_2_AVG
&END PRINT
&RDF T
HE_HE T
MAXR 30.0
NBIN 1000
SOLUTE_HE T
&END RDF
&WORM
ALLOW_OPEN T
CENTROID_DRMAX 0.25
MAX_OPEN_CYCLES 100
OPEN_CLOSE_SCALE 5.0000000000000000E+001
SHOW_STATISTICS T
STAGING_L 6
&END WORM
&END HELIUM
&INIT
CENTROID_SPEED F
RANDOMIZE_POS F
THERMOSTAT_SEED 40
&END INIT
&PILE
LAMBDA 0.5
TAU 200
&END PILE
&PRINT
&ACTION OFF
&END ACTION
&CENTROID_GYR OFF
&END CENTROID_GYR
&CENTROID_POS SILENT
&EACH
PINT 100
&END EACH
&END CENTROID_POS
&CENTROID_VEL SILENT
&EACH
PINT 100
&END EACH
&END CENTROID_VEL
&ENERGY SILENT
&EACH
PINT 100
&END EACH
&END ENERGY
&END PRINT
&END PINT
&PRINT
&RESTART SILENT
BACKUP_COPIES 1
&EACH
PINT 1000
&END EACH
&END RESTART
&RESTART_HISTORY OFF
&END RESTART_HISTORY
&TRAJECTORY OFF
&END TRAJECTORY
&VELOCITIES OFF
&END VELOCITIES
&END PRINT
&END MOTION
&FORCE_EVAL
METHOD NNP
&NNP
# Reference: [J. Chem. Theo. Comput. 16 88-99 (2020)](https://pubs.acs.org/doi/abs/10.1021/acs.jctc.9b00805)
NNP_INPUT_FILE_NAME NNP/protH2Ocluster-jctc2020-nnp/input.nn
SCALE_FILE_NAME NNP/protH2Ocluster-jctc2020-nnp/scaling.data
&MODEL
WEIGHTS NNP/protH2Ocluster-jctc2020-nnp/weights
&END MODEL
&END NNP
&SUBSYS
&CELL
ABC 8.5000000000000053E+000 8.5000000000000053E+000 8.5000000000000053E+000
PERIODIC NONE
&END CELL
&COORD
O 0.0000000000 0.0000000000 0.0000000000
H 1.0503793728 0.0000000000 0.0000000000
H -0.4976819981 0.7788636207 0.0000000000
H -0.5579356431 -0.5112027017 -0.4870267348
&END COORD
&TOPOLOGY
NUMBER_OF_ATOMS 4
&END TOPOLOGY
&VELOCITY
0.0000000000000000E+000 0.0000000000000000E+000 0.0000000000000000E+000
0.0000000000000000E+000 0.0000000000000000E+000 0.0000000000000000E+000
0.0000000000000000E+000 0.0000000000000000E+000 0.0000000000000000E+000
0.0000000000000000E+000 0.0000000000000000E+000 0.0000000000000000E+000
&END VELOCITY
&END SUBSYS
&END FORCE_EVAL

File diff suppressed because it is too large Load diff