I am making my classification project in quantum machine learning, currently trying to run PegasosQSVC in backend! The code runs normally in my environment however when I connect to backend using provider.get_backend() from QiskitRuntimeService() and try to run it on backend it shows different errors, the first error is that I can't run my code on a simulator when I chose a simulator backend, the other one is the error I show in this question. My code is as follows :

# importing feature map 
from qiskit.circuit.library import ZFeatureMap 

# define feature map to be used in the algorithm 
featuremap = ZFeatureMap(feature_dimension=num_qubits, reps = 3) 

# transpile the featuremap to make compatible with the backend topology 

from qiskit.compiler.transpiler import transpile
transpiled_featuremap = transpile(circuits=featuremap, backend=backend) 

# set the session in which we will run the circuits 
from qiskit_ibm_runtime import Session, Sampler

with Session(service= service, backend=backend) as session:
    sampler = Sampler(session=session, backend=backend) 

# set the fidelity parameter 
from qiskit_algorithms.state_fidelities import ComputeUncompute

fidelity = ComputeUncompute(sampler=sampler) 

# setting the kernel 
from qiskit_machine_learning.kernels import FidelityQuantumKernel 

qkernel = FidelityQuantumKernel(feature_map=transpiled_featuremap, fidelity=fidelity)  

# fitting the classifier 
from qiskit_machine_learning.algorithms import PegasosQSVC 
pegasos = PegasosQSVC(quantum_kernel=qkernel, C = C, num_steps=tau)


pegasos_score = pegasos.score(test_features,test_labels) 

print(f"Pegasosqsvc classification test score: {pegasos_score*100}")

When I run this code I get the following error:

The instruction sxdg on qubits (1,) is not supported by the target system. Circuits that do not match the target hardware definition are no longer supported after March 4, 2024. See the transpilation documentation (https://docs.quantum.ibm.com/transpile) for instructions to transform circuits and the primitive examples (https://docs.quantum.ibm.com/run/primitives-examples) to see this coupled with operator transformations. 

What can I do to solve this issue?


1 Answer 1


ComputeUncomputecomposes the circuit with its inverse() and this ends up not conforming to the ISA requirement. There is an issue in Qiskit Algorithms specifically about that as well a more general issue around Qiskit Algorithms and ISA circuits. With the code as-is I do not know a solution - if you consider modifying some code then a workaround would be to take a local copy of ComputeUncompute and modify it to transpile the circuit it uses (ie the circuit1 + circuit2.inverse()) for some given backend.

It should however still run on local Aer simulator if you use the Sampler from Aer.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.