Drawing the circuit in pennylane

PennyLane Help
1

I am trying to draw the circuit in pennylane for the kernel code: https://github.com/thubregtsen/qhack/blob/master/paper/noiseless_MNIST.py
I am able to draw the circuit but not able to understand much

uob8y
uob8y1920×961 73.2 KB

. Please guide me how to understand this circuit? Also, is there any better way to draw the circuit?

from keras.datasets import mnist
from keras.datasets import mnist

# import numpy as np
import matplotlib as mpl
import matplotlib.pyplot as plt

import pennylane as qml
from pennylane import numpy as np


## device and circuit characteristics
width = 4
depth = 8

dev = qml.device("default.qubit", wires=width)
wires = list(range(width))
print("The value of wires= ", wires)

# Our Ansatz
def layer(x1, params, wires, i0=0, inc=1):
    """Building block of the embedding Ansatz"""
    i = i0
    
    for j, wire in enumerate(wires):
       
        qml.Hadamard(wires=[wire])
        
        qml.RZ(x1[i % len(x1)], wires=[wire])
        i += inc
        qml.RY(params[0, j], wires=[wire])
        
    qml.broadcast(unitary=qml.CRZ, pattern="ring", wires=wires, parameters=params[1])


# @qml.template # type: ignore
def ansatz(x1, params, wires):   
    """The embedding Ansatz"""
    for j, layer_params in enumerate(params):
        layer(x1, layer_params, wires, i0=j * len(wires))
        
        


# init the embedding kernel
@qml.qnode(dev)
def kernel(x1, x2, params):
    ansatz(x1, params, wires)
    qml.adjoint(ansatz)(x2, params, wires)                      # type: ignore
    return qml.expval(qml.Projector([0]*width, wires=wires))


x1 = np.array([0.5, 0.25])

x2 = np.array([0.28571429, 0.53571429])

params = np.array([[[5.27502056, 1.47427201, 4.93076496, 2.46091687],
  [5.31970254, 3.72079515, 5.23318151, 3.6141986 ]],

 [[4.56873422, 4.71736634, 3.74526732, 4.03018843],
  [2.50768043, 6.05648213, 0.15691596, 4.6339621 ]],

 [[3.13801229, 4.06608748, 3.67299568, 0.56942335],
  [3.7249661,  0.21089213, 1.81957038, 5.11239481]],

 [[0.23371406, 0.38418188, 1.57283641, 2.1281051 ],
  [4.82044232, 2.29357766, 2.28309303, 1.36251918]],

 [[3.81316379, 1.72378489, 5.33909833, 1.99670676],
  [0.95241059, 4.91302177, 5.11174524, 0.50581628]],

 [[2.93171839, 4.72425322, 4.13123708, 2.1082139 ],
  [3.44454441, 3.57751799, 4.82533803, 3.01520779]],

 [[4.82192352, 4.9103957,  5.62978572, 2.72576366],
  [2.52244612, 2.06081289, 6.28092847, 4.55329168]],

 [[5.67932649, 1.43286835, 6.13243934, 3.72696941],
  [5.50192702, 1.69022441, 3.61985605, 4.94172026]]])


# print(qml.draw(kernel(x1, x2, params), charset="unicode")) # type: ignore

fig, ax = qml.draw_mpl(kernel)(x1, x2, params)

#   print(qml.draw(circuit_qnode)(X_train_dist_zero_not_zero[0], params))

plt.show()
2

Hello @Manu_Chaudhary,

The qml.broadcast(unitary=qml.CRZ, pattern=“ring”, wires=wires, parameters=params[1]) appears to be crowding the H, RZ, RY.

Function draw_mpl contains useful information. qml.draw_mpl — PennyLane 0.31.1 documentation

You may have to adjust the scaling of fig, ax by setting fig.set_size_inches(#, #) and adjusting the #s, code by OpenAI, Best regards.

3

Thank you the response. I am new to pennylane. Can you make changes in the above line so that the circuit is displayed correctly.

4

Hello @Manu_Chaudhary,

I was able to get a circuit print out running your specific code in a colab notebook; no fig.set needed. Be sure to check the depth of the circuit to see if this is what you want as the pattern changes. Perhaps run in a different IDE, or other settings.

download
download11519×519 109 KB

1 Like
5

Thank you @kevinkawchak for this great great help. The circuit displays correctly when I am running the code on the notebook.

2 Likes