Remove discussion of addresses from API and docs and replace with "CI strings" #54
Conversation
|
@caleb-johnson , should we also change the name of variables? |
Yes, I believe I got them all. Please let me know if you see any I missed |
qiskit_addon_sqd/fermion.py
Outdated
| @@ -193,20 +192,20 @@ def optimize_orbitals( | |||
| """ | |||
| if isinstance(bitstring_matrix, tuple): | |||
| warnings.warn( | |||
| "Passing a length-2 tuple of sorted addresses to define the subspace is deprecated. Users " | |||
| "Passing a length-2 tuple of base-10 determinants to define the subspace is deprecated. Users " | |||
There was a problem hiding this comment.
I think simply "determinants" is more accurate than "base-10 determinants." They are simply integers, without reference to a "base."
There was a problem hiding this comment.
Updated the warning message
qiskit_addon_sqd/fermion.py
Outdated
| bitstring_matrix: np.ndarray, open_shell: bool = False | ||
| ) -> tuple[np.ndarray, np.ndarray]: | ||
| """ | ||
| Convert bitstrings (rows) in a ``bitstring_matrix`` into base-10 integer representations of determinants. |
There was a problem hiding this comment.
They're just integers, not "base-10" integers.
There was a problem hiding this comment.
I started thinking that integers can be specified as base-2, which is a pretty natural way to specify these configurations as well. So I thought being explicit wouldn't hurt. I guess saying "integer" in a software context implies base-10
There was a problem hiding this comment.
The integers can be specified in base-2, or any other base. That's my point. The base 10 is not special here, nor in a general software context.
There was a problem hiding this comment.
Removed all base-10 mentions in fermion module
There was a problem hiding this comment.
Well, on further thought, base 2 is special here, because the integer representation of a determinant is obtained by converting its string representation to an integer, in base 2.
|
I have a separate general comment, which is that I think calling integers "base-10 integers" is confusing. They're just integers. Not sure if we want to fix that here or in a separate PR. |
Ya I commented on this above. A base-2 integer is also an integer, and it's a natural structure in this context, so I wanted to be explicit. If you think it just adds confusion or is unnecessary. I've removed all of that from the |
Alright as we discussed offline, there is no such thing as a "base-2 integer" or a "base-10 integer," only "integers." I'm fine with fixing the terminology in a separate PR. I've opened #57. |
The term "address", as used by PySCF, was being incorrectly used throughout the API and docs. Let's revert that discussion and use the correct/consistent-with-pyscf term, "CI strings".