Skip to content

Building a noise model for simulating a Qiskit quantum circuit in the presence of errors.

License

Notifications You must be signed in to change notification settings

Aditya-Ramachandran/Quantum-Noise-Models

Repository files navigation

forthebadge made-with-python

Python 3.6 GitHub license

Quantum-Noise-Models

Building a noise model for simulating a Qiskit quantum circuit in the presence of errors. We know that the simulator always imitates an ideal Quantum Computer so when we run a circuit on it, it won't show any noise (Quantum Noise) and always gives us the exact expected result with any moise. But the quantum computers we have today are way too noisy and are not at all ideal. So I made this so when you run your circuit on a simulator, an approximate NoiseModel can be generated automatically from the properties of real device backends from the IBMQ provider using the from_backend() method. Then I also coded a custom noise model using some pre-defined quantum error functions in the NoiseModel class (the Depolarizing Noise Model) and it we deliberately add errors in it and then plot the final results in the form of a histogram.

Softwares and Libraries used

  • Jupyter Notebook
  • Matplotlib
  • Qiskit
    1. qiskit-terra
    2. qiskit-aer
    3. qiskit-ibmq-provider

About

Building a noise model for simulating a Qiskit quantum circuit in the presence of errors.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published