Hey @ImranNasrullah,
It’s difficult to say what’s going on in your code (there’s a lot going on 
). I made this code that is heavily based on this demo. The water Hamiltonian has 14 qubits with the default basis set (sto-3g), so it’s taking me a bit to run on my humble laptop. But, maybe it will give you some ideas for why your code is stalling out.
import pennylane as qml
import pennylane.numpy as np
def finite_diff(f, x, delta=0.01):
"""Compute the central-difference finite difference of a function"""
gradient = []
for i in range(len(x)):
shift = np.zeros_like(x)
shift[i] += 0.5 * delta
res = (f(x + shift) - f(x - shift)) * delta**-1
gradient.append(res)
return gradient
def optimize_coordinates():
electrons = 10
orbitals = 7
active_orbitals = 4
active_electrons = 4
qubits = 14 # from previous calculation
def water_hamiltonian(coordinates):
H = qml.qchem.molecular_hamiltonian(
["H", "O", "H"],
coordinates,
charge=0,
method="dhf",
)[0]
return H
init_coords = np.array([-1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 1.0, 0.0, 0.0], requires_grad=True)
hf_state = qml.qchem.hf_state(active_electrons, active_orbitals)
singles, doubles = qml.qchem.excitations(electrons, qubits)
init_params = np.random.uniform(0, 1, size=(len(singles) + len(doubles),), requires_grad=True)
coords = init_coords
params = init_params
hf_state = qml.qchem.hf_state(electrons, qubits)
singles, doubles = qml.qchem.excitations(electrons, qubits)
dev = qml.device("default.qubit", wires=qubits)
@qml.qnode(dev)
def ansatz(params, obs):
qml.templates.AllSinglesDoubles(params, range(qubits), hf_state, singles, doubles)
return qml.expval(obs)
def cost(params, coords):
H = water_hamiltonian(coords)
return ansatz(params, H)
def grad_coords(params, coords):
grad_h = finite_diff(water_hamiltonian, coords)
grad = [ansatz(params, obs) for obs in grad_h]
return np.array(grad)
opt_params = qml.AdamOptimizer(0.05)
opt_coords = qml.AdamOptimizer(0.1)
for i in range(500):
coords.requires_grad = False
params.requires_grad = True
params, _ = opt_params.step(cost, params, coords)
coords.requires_grad = True
params.requires_grad = False
_, coords = opt_coords.step(cost, params, coords, grad_fn=grad_coords)
print("cost", cost(params, coords))
return coords
optimize_coordinates()
Hope this helps!
