def __init__(self, build=True, gpuid=0, reslist=[], **kwargs):
Calculator.__init__(self, **kwargs)
if build:
anipath = os.path.dirname(__file__)
cnstfile = anipath + '/../ANI-c08f-ntwk/rHCNO-4.6A_16-3.1A_a4-8.params'
saefile = anipath + '/../ANI-c08f-ntwk/sae_6-31gd.dat'
nnfdir = anipath + '/../ANI-c08f-ntwk/networks/'
self.nc = pync.molecule(cnstfile, saefile, nnfdir, gpuid)
self.Setup = True
self.reslist = reslist
def __init__(self, build=True, gpuid=0, reslist=[], **kwargs):
Calculator.__init__(self, **kwargs)
if build:
anipath = os.path.dirname(__file__)
cnstfile = anipath + '/ani-1x_dft_x8ens/rHCNO-5.2R_16-3.5A_a4-8.params'
saefile = anipath + '/ani-1x_dft_x8ens/sae_linfit.dat'
nnfdir = anipath + '/ani-1x_dft_x8ens//train0/networks/'
self.nc = pync.molecule(cnstfile, saefile, nnfdir, gpuid)
self.Setup = True
self.reslist = reslist
def __init__(self,
cnstfile,
saefile,
nnfprefix,
Nnet,
gpuid=0,
sinet=False):
# Number of networks
self.Nn = Nnet
# Construct pyNeuroChem molecule classes
self.ncl = [
pync.molecule(cnstfile, saefile, nnfprefix + str(i) + '/networks/',
gpuid, sinet) for i in range(self.Nn)
]
r = re.compile('#Smiles:(.+?)\n')
s = r.search(f)
return s.group(1).strip()
# Set required files for pyNeuroChem
anipath = '/home/jujuman/Dropbox/ChemSciencePaper.AER/networks/ANI-c08f-ntwk/'
cnstfile = anipath + '/rHCNO-4.6A_16-3.1A_a4-8.params'
saefile = anipath + '/sae_6-31gd.dat'
nnfdir = anipath + '/networks/'
idir = '/home/jujuman/Scratch/Research/GDB-11-wB97X-6-31gd/dnntsgdb11_07/inputs/'
sdir = '/home/jujuman/Scratch/Research/GDB-11-wB97X-6-31gd/dnntsgdb11_07/inputs_new/'
# Construct pyNeuroChem class
nc = pync.molecule(cnstfile, saefile, nnfdir, 0)
files = [f for f in os.listdir(idir) if f.split(".")[1] == "ipt"]
for i, f in enumerate(files):
data = hdn.read_rcdb_coordsandnm(idir + f)
X = data['coordinates']
S = data['species']
mol = Atoms(positions=X, symbols=S)
mol.set_calculator(ANI(False))
mol.calc.setnc(nc)
dyn = LBFGS(mol, logfile='optimization.log')
dyn.run(fmax=0.00001, steps=1000)
def molecule_worker(task_queue, gpuid, net_list, energy, forces, net_dict):
print('Building...')
ncl = [
pync.molecule(net_dict['cns'], net_dict['sae'],
net_dict['nnf'] + str(i) + '/networks/', gpuid, False)
for i in net_list
]
Nn = net_dict['Nn']
if net_dict['epw']:
for nc in ncl:
nc.setPairWise()
set_pbc = False
Sp = []
while True:
next_task = task_queue.get()
if next_task is None:
# Poison pill means shutdown
time.sleep(1)
print('Exiting')
task_queue.task_done()
break
if not set_pbc:
pbc = next_task['pbc']
for i, netid in enumerate(net_list):
#print('PBC:',pbc)
ncl[i].setPBC(pbc[0], pbc[1], pbc[2])
set_pbc = True
# Atomic elements
S = next_task['S']
# Cell
cell = next_task['cell']
# pbc_inv
pinv = next_task['pinv']
# make it two-dimensional
for i, netid in enumerate(net_list):
# Set the molecule/coordinates
if Sp == S:
if 'bynet' in next_task:
ncl[i].setCoordinates(
coords=next_task['X'][netid].astype(np.float32))
else:
ncl[i].setCoordinates(
coords=next_task['X'].astype(np.float32))
else:
if 'bynet' in next_task:
ncl[i].setMolecule(
next_task['X'][netid].astype(np.float32), S)
else:
ncl[i].setMolecule(next_task['X'].astype(np.float32), S)
# Set the cell
ncl[i].setCell(cell, pinv)
#print('CELL:',cell)
energy[netid] = ncl[i].energy().copy()
forces[netid] = ncl[i].force().copy()
#if netid == 0 and net_dict['epw']:
#energy[netid] += ncl[i].pwenergy()
#forces[netid] += ncl[i].pwforce()
#print(ncl[i].pwenergy().copy())
#print(-ncl[i].pwforce().copy()[0])
if net_dict['epw']:
energy[netid] += ncl[i].pwenergy()
forces[netid] += ncl[i].pwforce()
Sp = S
task_queue.task_done()
return
import os
import ase
import pyNeuroChem
import ase_interface
import numpy
path = os.path.dirname(os.path.realpath(__file__))
builtin_path = os.path.join(path, '../../torchani/resources/ani-1x_dft_x8ens/')
const_file = os.path.join(builtin_path, 'rHCNO-5.2R_16-3.5A_a4-8.params')
sae_file = os.path.join(builtin_path, 'sae_linfit.dat')
network_dir = os.path.join(builtin_path, 'train0/networks/')
radial_length = 64
conv_au_ev = 27.21138505
all_species = ['H', 'C', 'N', 'O']
species_indices = {all_species[i]: i for i in range(len(all_species))}
nc = pyNeuroChem.molecule(const_file, sae_file, network_dir, 0)
def calc():
calc = ase_interface.ANI(False)
calc.setnc(nc)
return calc
class NeuroChem:
def _get_radial_part(self, fullaev):
return fullaev[:, :, :radial_length]
def _get_angular_part(self, fullaev):
return fullaev[:, :, radial_length:]
def __init__(self,cnstfile,saefile,nnfdir,gpuid=0, sinet=False):
# Construct pyNeuroChem class
self.nc = pync.molecule(cnstfile, saefile, nnfdir, gpuid, sinet)
smfile = '/home/jujuman/Research/RawGDB11Database/gdb11_size09.smi' # Smiles file
wkdir = '/home/jujuman/Research/CrossValidation/'
cnstfile = wkdir + 'rHCNO-4.6A_16-3.1A_a4-8.params'
saefile = wkdir + 'sae_6-31gd.dat'
At = ['C', 'O', 'N'] # Hydrogens added after check
#-------------------------------------------
#nnfdir = wkdir + 'cv_c08e_ntw_' + str(0) + '/networks/'
# Construct pyNeuroChem classes
#nc = pync.molecule(cnstfile, saefile, nnfdir, 0)
nc = [
pync.molecule(cnstfile, saefile,
wkdir + 'cv_c08e_ntw_' + str(l) + '/networks/', 0)
for l in range(5)
]
molecules = Chem.SmilesMolSupplier(smfile, nameColumn=0)
#mols = [molecules[i] for i in range(0,1000)]
#print(mols)
f = open(
'/home/jujuman/Research/CrossValidation/GDB-09-High-sdev/gdb-09-0.5sdev.dat',
'w')
for k, m in enumerate(molecules):
if m is None: continue
typecount = 0
wkdir3 = '/home/jujuman/Dropbox/ChemSciencePaper.AER/networks/ANI-c08f09dd-ntwk-cv/'
wkdir4 = '/home/jujuman/Dropbox/ChemSciencePaper.AER/networks/ANI-c08f09div-ntwk-cv/'
cnstfile = 'rHCNO-4.6A_16-3.1A_a4-8.params'
saefile = 'sae_6-31gd.dat'
At = ['C', 'O', 'N'] # Hydrogens added after check
Nnc = 5
#-------------------------------------------
#nnfdir = wkdir + 'cv_c08e_ntw_' + str(0) + '/networks/'
# Construct pyNeuroChem classes
nc1 = [
pync.molecule(wkdir1 + cnstfile, wkdir1 + saefile,
wkdir1 + 'cv_c08e_ntw_' + str(l) + '/networks/', 0)
for l in range(Nnc)
]
nc2 = [
pync.molecule(wkdir2 + cnstfile, wkdir2 + saefile,
wkdir2 + 'cv_c08e_ntw_' + str(l) + '/networks/', 0)
for l in range(Nnc)
]
nc3 = [
pync.molecule(wkdir3 + cnstfile, wkdir3 + saefile,
wkdir3 + 'cv_c08e_ntw_' + str(l) + '/networks/', 0)
for l in range(Nnc)
]
nc4 = [
pync.molecule(wkdir4 + cnstfile, wkdir4 + saefile,
wkdir4 + 'cv_c08e_ntw_' + str(l) + '/networks/', 0)
cnstfile = wkdir + 'rHCNO-4.6A_16-3.1A_a4-8.params'
saefile = wkdir + 'sae_6-31gd.dat'
At = ['C', 'O', 'N'] # Hydrogens added after check
T = 800.0
dt = 0.25
stdir = '/home/jujuman/Research/CrossValidation/MD_CV/'
#-------------------------------------------
# Construct pyNeuroChem classes
print('Constructing CV network list...')
ncl = [
pync.molecule(cnstfile, saefile, wkdir + 'train' + str(l) + '/networks/',
0, False) for l in range(5)
]
print('Complete.')
# Set required files for pyNeuroChem
#anipath = '/home/jujuman/Dropbox/ChemSciencePaper.AER/ANI-c08e-ccdissotest1-ntwk'
#cnstfile = anipath + '/rHCNO-4.6A_16-3.1A_a4-8.params'
#saefile = anipath + '/sae_6-31gd.dat'
#nnfdir = anipath + '/networks/'
# Construct pyNeuroChem class
print('Constructing MD network...')
nc = ncl[1]
#nc = pync.molecule(cnstfile, saefile, nnfdir, 0)
print('FINISHED')
def plot_irc_data(axes, file, rcf, title):
xyz, typ, Eact = hdt.readncdat(file, np.float32)
Rc = np.load(rcf)
# Set required files for pyNeuroChem
wkdir = '/home/jujuman/Dropbox/ChemSciencePaper.AER/networks/ANI-c08f-ntwk-cv/'
cnstfile = 'rHCNO-4.6A_16-3.1A_a4-8.params'
saefile = 'sae_6-31gd.dat'
nc = [
pync.conformers(wkdir + cnstfile, wkdir + saefile,
wkdir + 'cv_c08e_ntw_' + str(l) + '/networks/', 0)
for l in range(5)
]
rcdir = '/home/jujuman/Research/ANI-DATASET/RXN1_TNET/training/rxn1to6/ani_benz_rxn_ntwk/'
ncr1 = pync.conformers(rcdir + '../../' + cnstfile,
rcdir + '../../' + saefile, rcdir + '/networks/', 0)
ncr2 = pync.molecule(rcdir + '../../' + cnstfile,
rcdir + '../../' + saefile, rcdir + '/networks/', 0)
ncr3 = pync.molecule(rcdir + '../../' + cnstfile,
rcdir + '../../' + saefile, rcdir + '/networks/', 0)
# Compute reactant E
ncr2.setMolecule(coords=xyz[0], types=list(typ))
Er = ncr2.energy()
# Compute product E
ncr3.setMolecule(coords=xyz[-1], types=list(typ))
Ep = ncr3.energy()
#Eact = Eact[::-1]
dE_ani = hdt.hatokcal * (Er - Ep)
dE_dft = hdt.hatokcal * (Eact[0] - Eact[-1])
print('Delta E R/P ANI:', dE_ani, 'Delta E R/P ANI:', dE_dft, 'Diff:',
abs(dE_ani - dE_dft))
# Set the conformers in NeuroChem
ncr1.setConformers(confs=xyz, types=list(typ))
# Compute Energies of Conformations
E1 = ncr1.energy()
# Shift
E1 = E1 - E1[0]
Eact = Eact - Eact[0]
# Plot
errn = hdt.calculaterootmeansqrerror(hdt.hatokcal * E1,
hdt.hatokcal * Eact)
axes.plot(Rc['x'][:, 1],
hdt.hatokcal * (E1),
color='red',
label="{:.2f}".format(errn),
linewidth=2)
axes.plot(Rc['x'][:, 1],
hdt.hatokcal * (Eact),
'r--',
color='black',
linewidth=3)
err = []
for n, net in enumerate(nc):
# Set the conformers in NeuroChem
net.setConformers(confs=xyz, types=list(typ))
# Compute Energies of Conformations
E1 = net.energy()
E1 = E1 - E1[0]
err.append(
hdt.calculaterootmeansqrerror(hdt.hatokcal * E1,
hdt.hatokcal * Eact))
# Plot
if n == len(nc) - 1:
mean = np.mean(np.asarray(err))
axes.plot(Rc['x'][:, 1],
hdt.hatokcal * (E1),
color='blue',
label="{:.2f}".format(mean),
linewidth=1)
else:
axes.plot(Rc['x'][:, 1],
hdt.hatokcal * (E1),
color='blue',
linewidth=1)
axes.plot(Rc['x'][:, 1],
hdt.hatokcal * (E1),
color='blue',
linewidth=1)
axes.set_xlim([Rc['x'][:, 1].min(), Rc['x'][:, 1].max()])
axes.legend(loc="upper right", fontsize=8)
axes.set_title(title)
return np.array([errn, np.mean(err)])
