This guides provides the minimal info for external users to use the EURIQA-blue API correctly.

EURIQA provides the API interface which submits a list of circuits and collect the list of probabiilty histograms: prob_vector = run_on_EURIQA(circuits, num_shots). More inforamtion returns is possible but needs to coordinate with the experiment operator in advance.

For the purpose of code development, one can use the following example. Please refer to the euriqa_interface.py for the docstrings of run_on_EURIQA for any input/output format issue.

import qiskit
import numpy as np
from euriqa_interface import run_on_EURIQA


qc0 = qiskit.QuantumCircuit(2)
qc0.rxx(np.pi/2,0,1)

qc1 = qc0.copy()
qc1.rx(np.pi/2,0)

qasm_circuits = [c.qasm() for c in  [qc0, qc1]]


# Do necessary preprocessing including basis transformation to generate list of circuits in OpenQASM2.0 string format.
# For local test (returns exact simulation prob result):
probs_local = run_on_EURIQA(qasm_circuits,num_shots=100, run_simulation_local=True)
# For actual experiment (this shows what it looks like to experimentalists, api calls to the actual machine is only accessable to EURIQA internal computers):
# probs_exp = run_on_EURIQA(qasm_circuits,num_shots=100)

print(probs_local)
# Then do post-processing using probs

Expected output:

[array([0.5, 0. , 0. , 0.5]), array([0.25, 0.25, 0.25, 0.25])]

!!! It is highly recommended provide version info on key modules or requiremnets.txt to best reproduce the working enviroment.

Euriqa runs on an older version of qiskit might not recognize the qasm string generated from newer version of qiskit even if both of them are OpenQASM2.0. In order to test if your qasm string works on the eurqia system, please test your qasm strings with correct qiskit version. The following is recommended procedure.

Below is the example for generating and saving the qasm string.

qc0.remove_final_measurements()
with open(path+'/qc0.qasm','w') as f:
    f.write(qc0.qasm())

Depending on your qiksit version, you can generate exact probability histogram using abs(qiskit.quantum_info.Statevector(cq).data)**2 or abs(qiskit.quantum_info.Statevector.from_instruction(cq))**2 for newer and older qiskit version repectively. Then you can save the result locally for later comparison.

pip install qiskit-terra==0.16.1

so that the version result of qiskit.__qiskit_version__ is: {'qiskit-terra': '0.16.1', 'qiskit-aer': None, 'qiskit-ignis': None, 'qiskit-ibmq-provider': None, 'qiskit-aqua': None, 'qiskit': None}. Also the numpy version is 0.18.1.

Comparing the probability histogram generated with your version of qiskit and older version (0.18.1) of qiskit to make sure they result in the same probability histogram.