gene.py

'''
This file is used to generate a quantum circuit from a gene

author : Yu-Cheng Chung
email  : ycchung@ntnu.edu.tw
date   : 2023 12 oct

dependencies:
    qiskit
    numpy
    
'''

import qiskit as qk
import numpy as np


class Gene_Circuit(object):
    '''
    Gene Circuit

    Methods:
        generate_circuit_from_gene: generate a quantum circuit with num_qubit qubits
        circuit_0 ~ circuit_10: quantum circuit connect with gene

    '''
    def __init__(self,gene,num_qubit) -> None:
        '''
        Args:
            gene: a array with shape (num_qubit, length_gene) with element called G_ij
            num_qubit: number of qubits
        
        object:
            self.gene: a array with shape (num_qubit, length_gene) with element called G_ij
            self.num_qubit: number of qubits
            self.circuit: a quantum circuit with num_qubit qubits generated from gene
            self.draw: draw the circuit
        '''
        self.gene = gene
        self.num_qubit = num_qubit
        self.circuit = qk.transpile(self.generate_circuit_from_gene())
        self.draw = self.circuit.draw
        self.num_parameters = self.circuit.num_parameters
        self.depth = self.circuit.depth

    def bind_parameters(self, theta : list|np.ndarray) -> qk.QuantumCircuit:
        '''
        bind parameter to the circuit
        Args:
            theta: a list of theta
        Returns:
            bind_circuit: a quantum circuit with parameter binded
        '''
        binded_circuit = self.circuit.bind_parameters({self.circuit.parameters[i]:theta[i] for i in range(len(theta))})
        return binded_circuit

    def generate_circuit_from_gene(self)->qk.QuantumCircuit:
        '''
        Generate a quantum circuit with num_qubit qubits
        Args:
            self.gene: a array with shape (num_qubit, length_gene) with element called G_ij
            self.num_qubit: number of qubits
        Returns:
            circuit: a quantum circuit

        The circuit is generated by the following rule:
        ------------------------------------------------
        | G_00 | G_01 | G_02 | G_03 | G_04 | ...| G_0m |
        ------------------------------------------------
        | G_10 | G_11 | G_12 | G_13 | G_14 | ...| G_1m |
        ------------------------------------------------
        ...
        ------------------------------------------------
        | G_n0 | G_n1 | G_n2 | G_n3 | G_n4 | ...| G_nm |
        ------------------------------------------------

        G_ij is the j-th gate of the i-th qubit
        G_ij will look like tuple(gate,cotrol)
        gate is the index of the gate in gene_gates
        control is the index of the control qubit
        (if the gate have no control, will ignore the control value)

        '''
        
        #gene_gates is a list of gates use to generate the circuit
        gene_gates = ['empty','empty','empty','rx','ry','rz','cx']
        theta_index = 0
        circuit = qk.QuantumCircuit(self.num_qubit)
        gene = self.gene
        gene = gene.transpose(1,0,2)
        temp_last_gate = ['empty' for i in range(self.num_qubit)]
        count_rgate = [0 for i in range(self.num_qubit)]
        for j,G_nj in enumerate(gene):
            for i,G_ij in enumerate(G_nj):
                gate = gene_gates[G_ij[0]]
                if gate == temp_last_gate[i] and gate in ['rx','ry','rz','h','x','sx']:
                    continue
                control = G_ij[1]
                if control >= self.num_qubit:
                    control = control % self.num_qubit
                if gate == 'empty':
                    continue
                elif gate in ['rx','ry','rz']:
                    if count_rgate[i]>=3:
                        continue
                    getattr(circuit,gate)(qk.circuit.Parameter(f'theta_{theta_index}'),i)
                    temp_last_gate[i] = gate
                    count_rgate[i]+=1
                    theta_index+=1
                elif gate == 'cx':
                    if control == i:
                        continue
                    else:
                        circuit.cx(control,i)
                        temp_last_gate[control] = 'empty'
                        temp_last_gate[i] = gate
                        count_rgate[control] = 0
                        count_rgate[i] = 0
                elif gate in ['h','x','sx']:
                    getattr(circuit,gate)(i)
                    temp_last_gate[i] = gate
                    count_rgate[i] = 0
        # circuit.draw("mpl",filename='test3.png')
        return circuit
    
if __name__ == '__main__':
    '''
    test gene:
        np.random.seed(776)
        gene = np.random.randint(0,7,size=(4,8,2))


    the circuit would look like:

                             ┌───┐     ┌─────────────┐     ┌───┐     ┌─────────────┐┌─────────────┐
    q_0: ────────────────────┤ X ├─────┤ Rx(theta_3) ├─────┤ X ├─────┤ Ry(theta_7) ├┤ Rz(theta_8) ├
         ┌─────────────┐     └─┬─┘     ├─────────────┤     └─┬─┘     ├─────────────┤└────┬───┬────┘
    q_1: ┤ Ry(theta_0) ├───────┼───────┤ Rz(theta_4) ├───────■───────┤ Rz(theta_5) ├─────┤ X ├─────
         └─────────────┘       │       └─────────────┘               └─────────────┘     └─┬─┘     
    q_2: ──────────────────────■───────────────────────────────────────────────────────────■───────
                        ┌─────────────┐┌─────────────┐┌─────────────┐                              
    q_3: ───────────────┤ Rz(theta_1) ├┤ Rx(theta_2) ├┤ Rz(theta_6) ├──────────────────────────────
                        └─────────────┘└─────────────┘└─────────────┘           
                        
    '''
    np.random.seed(776)
    gene = np.random.randint(0,7,size=(4,8,2))
    num_qubit = 4
    gene_circuit = Gene_Circuit(gene,num_qubit)
    print(gene)
    print(gene_circuit.draw())
    print(gene_circuit.num_parameters)
    print(gene_circuit.depth())