compatibility with new lammps-user-pace library and cleanup

examples/PACKAGES/pace/README.md

@@ -0,0 +1,9 @@
+# This folder contains examples for pace in LAMMPS
+
+
+## Compute pace usage
+compute/latte_cell_0.data            # lammps data file with C-H-O structure
+compute/latte_cell_0.xyz             # xyz file with C-H-O structure 
+compute/coupling_coefficients.yace   # .yace file containing coupling coefficients (or ACE potential parameters)
+compute/in.lammps                    # input file for calling `compute pace`
+

examples/PACKAGES/pace/compute/in.lammps

@@ -0,0 +1,22 @@
+#info all out log
+units  metal
+atom_style  atomic
+boundary    p p p
+atom_modify    map hash
+boundary  p p p
+read_data  latte_cell_0.data
+mass  1 1.00
+mass  2 14.00
+mass  3 15.999
+
+        # potential settings
+
+pair_style     zero 5.7
+pair_coeff     * *
+
+compute     pace all pace coupling_coefficients.yace 1 0
+
+thermo 1 
+thermo_style  custom step temp c_pace[1][183]
+
+run 0

examples/pace/multi-elem/README.md

@@ -1,5 +0,0 @@
-Check that seg fault doesn't occur by doing:
-
-    ./loop.sh
-
-which will loop through many runs of `python test_en.py`.

examples/pace/multi-elem/loop.sh

@@ -1,8 +0,0 @@
-for (( ; ; ))
-do 
-    python test_en.py
-    # terminate loop if seg fault
-    if [[ $? -eq 139 ]]; then 
-        break 
-    fi
-done

examples/pace/multi-elem/test_en.py

@@ -1,86 +0,0 @@
-from __future__ import print_function
-import sys, os
-import ctypes
-import numpy as np
-from ase.io import read,write
-from lammps import lammps, LMP_TYPE_ARRAY, LMP_STYLE_GLOBAL
-
-# get MPI settings from LAMMPS
-def run_struct(f):
-    file_prefix = f.split('.')[0]
-    atoms = read(f)
-    lmp = lammps()
-
-    me = lmp.extract_setting("world_rank")
-    nprocs = lmp.extract_setting("world_size")
-
-    write('%s.data' % file_prefix,atoms,format='lammps-data')
-
-    cmds = ["-screen", "none", "-log", "none"]
-    lmp = lammps(cmdargs = cmds)
-
-    print("Made LAMMPS instance")
-
-    def run_lammps(dgradflag):
-
-        # simulation settings
-        fname = file_prefix
-        lmp.command("clear")
-        lmp.command("info all out log")
-        lmp.command('units  metal')
-        lmp.command('atom_style  atomic')
-        lmp.command("boundary    p p p")
-        lmp.command("atom_modify    map hash")
-        lmp.command('neighbor  2.3 bin')
-        # boundary
-        lmp.command('boundary  p p p')
-        # read atoms
-        lmp.command('read_data  %s.data' % fname )
-        lmp.command('mass  1 1.00')
-        lmp.command('mass  2 14.00')
-        lmp.command('mass  3 15.999')
-
-        # potential settings
-
-        lmp.command(f"pair_style     zero 5.7")
-        lmp.command(f"pair_coeff     * *")
-
-
-        if dgradflag:
-            lmp.command(f"compute     pace all pace coupling_coefficients.yace 1 1")
-        else:
-            lmp.command(f"compute     pace all pace coupling_coefficients.yace 1 0 ")
-
-        # run
-
-        lmp.command(f"thermo         100")
-        lmp.command(f"run {nsteps}")
-
-    # declare simulation/structure variables
-
-    nsteps = 0
-    ntypes = 3
-
-    # declare compute pace variables
-
-
-    bikflag = 1
-
-    # NUMBER of descriptors
-    nd = 91
-
-    dgradflag = 0
-    
-    run_lammps(dgradflag)
-    
-    lmp_pace = lmp.numpy.extract_compute("pace", LMP_STYLE_GLOBAL, LMP_TYPE_ARRAY)
-    print ('global shape',np.shape(lmp_pace))
-    np.save('%s_chi_i.npy' % file_prefix,lmp_pace)
-    lmp.close()
-    del lmp
-    return None
-
-import glob
-for f in sorted(glob.glob('*.xyz')):
-    print ('running %s' % f)
-    run_struct(f)

src/ML-PACE/compute_pace.cpp

@@ -77,6 +77,7 @@ ComputePACE::ComputePACE(LAMMPS *lmp, int narg, char **arg) :
 
   //read in file with CG coefficients or c_tilde coefficients
   auto potential_file_name = utils::get_potential_file_path(arg[3]);
+  delete acecimpl -> basis_set;
   acecimpl -> basis_set = new ACECTildeBasisSet(potential_file_name);
   double cut = acecimpl -> basis_set->cutoffmax;
   cutmax = acecimpl -> basis_set->cutoffmax;
@@ -138,6 +139,7 @@ ComputePACE::ComputePACE(LAMMPS *lmp, int narg, char **arg) :
 
 ComputePACE::~ComputePACE()
 {
+  delete acecimpl;
   memory->destroy(pace);
   memory->destroy(paceall);
   memory->destroy(cutsq);
@@ -244,10 +246,6 @@ void ComputePACE::compute_array()
   NS_TYPE *ns;
   LS_TYPE *ls;
 
-  int n_r1, n_rp = 0;
-  n_r1 = acecimpl -> basis_set->total_basis_size_rank1[0];
-  n_rp = acecimpl -> basis_set->total_basis_size[0];
-
   const int inum = list->inum;
   const int* const ilist = list->ilist;
   const int* const numneigh = list->numneigh;
@@ -285,15 +283,18 @@ void ComputePACE::compute_array()
       const int typeoffset_local = ndims_peratom*nvalues*(itype-1);
       const int typeoffset_global = nvalues*(itype-1);
 
+      delete acecimpl -> ace;
       acecimpl -> ace = new ACECTildeEvaluator(*acecimpl -> basis_set);
+      acecimpl -> ace->compute_projections = 1;
+      acecimpl -> ace->compute_b_grad = 1;
       int n_r1, n_rp = 0;
-      n_r1 = basis_set->total_basis_size_rank1[0];
+      n_r1 = acecimpl -> basis_set->total_basis_size_rank1[0];
-      n_rp = basis_set->total_basis_size[0];
+      n_rp = acecimpl -> basis_set->total_basis_size[0];
 
       int ncoeff = n_r1 + n_rp;
       acecimpl -> ace->element_type_mapping.init(ntypes+1);
       for (int ik = 1; ik <= ntypes; ik++) {
-        for(int mu = 0; mu < basis_set->nelements; mu++){
+        for(int mu = 0; mu < acecimpl -> basis_set ->nelements; mu++){
           if (mu != -1) {
             if (mu == ik - 1) {
               map[ik] = mu;
@@ -336,7 +337,7 @@ void ComputePACE::compute_array()
       // resize the neighbor cache after setting the basis
       acecimpl -> ace->resize_neighbours_cache(max_jnum);
       acecimpl -> ace->compute_atom(i, atom->x, atom->type, list->numneigh[i], list->firstneigh[i]);
-      Array1D<DOUBLE_TYPE> Bs =ace->B_all;
+      Array1D<DOUBLE_TYPE> Bs = acecimpl -> ace -> projections;
 
       for (int jj = 0; jj < jnum; jj++) {
         const int j = jlist[jj];
@@ -345,14 +346,13 @@ void ComputePACE::compute_array()
           double *pacedi = pace_peratom[i]+typeoffset_local;
           double *pacedj = pace_peratom[j]+typeoffset_local;
 
-          Array3D<DOUBLE_TYPE> fs = acecimpl -> ace->neighbours_dB;
           //force array in (func_ind,neighbour_ind,xyz_ind) format
           // dimension: (n_descriptors,max_jnum,3)
           //example to access entries for neighbour jj after running compute_atom for atom i:
           for (int func_ind =0; func_ind < n_r1 + n_rp; func_ind++){
-            DOUBLE_TYPE fx_dB = fs(func_ind,jj,0);
+            DOUBLE_TYPE fx_dB = acecimpl -> ace -> neighbours_dB(func_ind,jj,0);
-            DOUBLE_TYPE fy_dB = fs(func_ind,jj,1);
+            DOUBLE_TYPE fy_dB = acecimpl -> ace -> neighbours_dB(func_ind,jj,1);
-            DOUBLE_TYPE fz_dB = fs(func_ind,jj,2);
+            DOUBLE_TYPE fz_dB = acecimpl -> ace -> neighbours_dB(func_ind,jj,2);
             pacedi[func_ind] += fx_dB;
             pacedi[func_ind+yoffset] += fy_dB;
             pacedi[func_ind+zoffset] += fz_dB;
@@ -362,15 +362,14 @@ void ComputePACE::compute_array()
             }
          } else {
             //printf("inside dBi/dRj logical : ncoeff = %d \n", ncoeff);
-            Array3D<DOUBLE_TYPE> fs = acecimpl -> ace->neighbours_dB;
             for (int iicoeff = 0; iicoeff < ncoeff; iicoeff++) {
 
               // add to pace array for this proc
               //printf("inside dBi/dRj loop\n");
               // dBi/dRj
-              DOUBLE_TYPE fx_dB = fs(iicoeff,jj,0);
+              DOUBLE_TYPE fx_dB = acecimpl -> ace -> neighbours_dB(iicoeff,jj,0);
-              DOUBLE_TYPE fy_dB = fs(iicoeff,jj,1);
+              DOUBLE_TYPE fy_dB = acecimpl -> ace -> neighbours_dB(iicoeff,jj,1);
-              DOUBLE_TYPE fz_dB = fs(iicoeff,jj,2);
+              DOUBLE_TYPE fz_dB = acecimpl -> ace -> neighbours_dB(iicoeff,jj,2);
               pace[bik_rows + ((atom->tag[j]-1)*3*natoms) + 3*(atom->tag[i]-1) + 0][iicoeff+3] -= fx_dB;
               pace[bik_rows + ((atom->tag[j]-1)*3*natoms) + 3*(atom->tag[i]-1) + 1][iicoeff+3] -= fy_dB;
               pace[bik_rows + ((atom->tag[j]-1)*3*natoms) + 3*(atom->tag[i]-1) + 2][iicoeff+3] -= fz_dB;
@@ -395,10 +394,8 @@ void ComputePACE::compute_array()
           pace[irow][k++] += Bs(icoeff);
         }
       }
-    delete acecimpl -> ace;
     } //group bit
   } // for ii loop
-  delete acecimpl -> basis_set;
   // accumulate force contributions to global array
   if (!dgradflag){
     for (int itype = 0; itype < atom->ntypes; itype++) {