Rick0317
commented
1 year ago The explanation of the code is divided into three parts.
1, Structure of the code in tequila/src/hamiltonian/custom_jw_transform.py
First, it is not implemented in the form of a class because the instance of it will not be used other than the transformation; plus, no inheritance relation was used. Second, the only non-private function is the "custom_jw_transform()" function. Other functions work as helper functions. By doing so, we can easily point out the flaw in the code.
2, What can users do
Users can perform their custom Jordan Wigner transformation in any of the following ways.
- Only specifying the fermion indices
- Specifying the fermion indices and a list of qubits
- Specifying the fermion indices and a list of qubits and qubit maps (qubit maps are the specification of linking some fermion indices and some qubits ) For fermion indices, users can either use a string format or a list of strings.
3, Test cases
Test cases were created for 20 cases with three types of inputs for each case. Plus, whether the output of the custom Jordan Wigner transformation possesses the desired eigenvalue property is tested.
Custom Jordan Wigner transformation is implemented in tequila/src/hamiltonian/custom_jw_transform.py file. Test cases for custom_jw_transform() are shown in tequila/tests/test_custom_jw_transform.py file.
Other specifications and comments are written down below