Split QFI into QFI and QGT (Qiskit#9085)

* wip skelton * add lin_comb_qgt * reflected comments Co-authored-by: ElePT <57907331+ElePT@users.noreply.github.com> Co-authored-by: Julien Gacon <jul@zurich.ibm.com> * added unittests * fix a copyright year * fix * fix doc * added qfi as an entry point * fix lint * fix lint * updated qfi * reflected comments Co-authored-by: Julien Gacon <jul@zurich.ibm.com> * fix doc * add qfi_result into init.py * reflected comments Co-authored-by: ElePT <57907331+ElePT@users.noreply.github.com> Co-authored-by: ElePT <57907331+ElePT@users.noreply.github.com> Co-authored-by: Julien Gacon <jul@zurich.ibm.com> Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
ElePT · Jan 17, 2023 · 4210fbd · 4210fbd
1 parent 2ffa230
commit 4210fbd
Show file tree

Hide file tree

Showing 13 changed files with 1,245 additions and 780 deletions.
diff --git a/qiskit/algorithms/gradients/__init__.py b/qiskit/algorithms/gradients/__init__.py
@@ -1,6 +1,6 @@
 # This code is part of Qiskit.
 #
-# (C) Copyright IBM 2022.
+# (C) Copyright IBM 2022, 2023
 #
 # This code is licensed under the Apache License, Version 2.0. You may
 # obtain a copy of this license in the LICENSE.txt file in the root directory
@@ -43,13 +43,14 @@
    ParamShiftSamplerGradient
    SPSASamplerGradient
 
-QFI
-===
+Quantum Geometric Tensor
+========================
 .. autosummary::
    :toctree: ../stubs/
 
-   BaseQFI
-   LinCombQFI
+   BaseQGT
+   LinCombQGT
+   QFI
 
 Results
 =======
@@ -59,39 +60,44 @@
 
    EstimatorGradientResult
    QFIResult
+   QGTResult
    SamplerGradientResult
 """
 
 from .base_estimator_gradient import BaseEstimatorGradient
-from .base_qfi import BaseQFI
+from .base_qgt import BaseQGT
 from .base_sampler_gradient import BaseSamplerGradient
 from .estimator_gradient_result import EstimatorGradientResult
 from .finite_diff_estimator_gradient import FiniteDiffEstimatorGradient
 from .finite_diff_sampler_gradient import FiniteDiffSamplerGradient
 from .lin_comb_estimator_gradient import DerivativeType, LinCombEstimatorGradient
-from .lin_comb_qfi import LinCombQFI
+from .lin_comb_qgt import LinCombQGT
 from .lin_comb_sampler_gradient import LinCombSamplerGradient
 from .param_shift_estimator_gradient import ParamShiftEstimatorGradient
 from .param_shift_sampler_gradient import ParamShiftSamplerGradient
+from .qfi import QFI
 from .qfi_result import QFIResult
+from .qgt_result import QGTResult
 from .sampler_gradient_result import SamplerGradientResult
 from .spsa_estimator_gradient import SPSAEstimatorGradient
 from .spsa_sampler_gradient import SPSASamplerGradient
 
 __all__ = [
     "BaseEstimatorGradient",
-    "BaseQFI",
+    "BaseQGT",
     "BaseSamplerGradient",
     "DerivativeType",
     "EstimatorGradientResult",
     "FiniteDiffEstimatorGradient",
     "FiniteDiffSamplerGradient",
     "LinCombEstimatorGradient",
-    "LinCombQFI",
+    "LinCombQGT",
     "LinCombSamplerGradient",
     "ParamShiftEstimatorGradient",
     "ParamShiftSamplerGradient",
+    "QFI",
     "QFIResult",
+    "QGTResult",
     "SamplerGradientResult",
     "SPSAEstimatorGradient",
     "SPSASamplerGradient",

diff --git a/qiskit/algorithms/gradients/base_estimator_gradient.py b/qiskit/algorithms/gradients/base_estimator_gradient.py
@@ -1,6 +1,6 @@
 # This code is part of Qiskit.
 #
-# (C) Copyright IBM 2022.
+# (C) Copyright IBM 2022, 2023
 #
 # This code is licensed under the Apache License, Version 2.0. You may
 # obtain a copy of this license in the LICENSE.txt file in the root directory
@@ -152,7 +152,7 @@ def _preprocess(
         Args:
             circuits: The list of quantum circuits to compute the gradients.
             parameter_values: The list of parameter values to be bound to the circuit.
-            parameters: The sequence of parameters to calculate only the gradients of the specified
+            parameter_sets: The sequence of parameters to calculate only the gradients of the specified
                 parameters.
             supported_gates: The supported gates used to transpile the circuit.
 
@@ -182,32 +182,32 @@ def _postprocess(
         results: EstimatorGradientResult,
         circuits: Sequence[QuantumCircuit],
         parameter_values: Sequence[Sequence[float]],
-        parameter_sets: Sequence[set[Parameter] | None],
+        parameter_sets: Sequence[set[Parameter]],
     ) -> EstimatorGradientResult:
-        """Postprocess the gradient. This computes the gradient of the original circuit from the
-        gradient of the gradient circuit by using the chain rule.
+        """Postprocess the gradients. This method computes the gradient of the original circuits
+        by applying the chain rule to the gradient of the circuits with unique parameters.
 
         Args:
-            results: The results of the gradient of the gradient circuits.
-            circuits: The list of quantum circuits to compute the gradients.
-            parameter_values: The list of parameter values to be bound to the circuit.
-            parameters: The sequence of parameters to calculate only the gradients of the specified
-                parameters.
+            results: The computed gradients for the circuits with unique parameters.
+            circuits: The list of original circuits submitted for gradient computation.
+            parameter_values: The list of parameter values to be bound to the circuits.
+            parameter_sets: An optional subset of parameters with respect to which the gradients should
+                be calculated.
 
         Returns:
-            The results of the gradient of the original circuits.
+            The gradients of the original circuits.
         """
         gradients, metadata = [], []
         for idx, (circuit, parameter_values_, parameter_set) in enumerate(
             zip(circuits, parameter_values, parameter_sets)
         ):
-            unique_gradient = np.zeros(len(parameter_set))
+            gradient = np.zeros(len(parameter_set))
             if (
                 "derivative_type" in results.metadata[idx]
                 and results.metadata[idx]["derivative_type"] == DerivativeType.COMPLEX
             ):
                 # If the derivative type is complex, cast the gradient to complex.
-                unique_gradient = unique_gradient.astype("complex")
+                gradient = gradient.astype("complex")
             gradient_circuit = self._gradient_circuit_cache[_circuit_key(circuit)]
             g_parameter_set = _make_gradient_parameter_set(gradient_circuit, parameter_set)
             # Make a map from the gradient parameter to the respective index in the gradient.
@@ -233,11 +233,11 @@ def _postprocess(
                         bound_coeff = coeff
                     # The original gradient is a sum of the gradients of the parameters in the
                     # gradient circuit multiplied by the coefficients.
-                    unique_gradient[i] += (
+                    gradient[i] += (
                         float(bound_coeff)
                         * results.gradients[idx][g_parameter_indices[g_parameter]]
                     )
-            gradients.append(unique_gradient)
+            gradients.append(gradient)
             metadata.append([{"parameters": parameter_indices}])
         return EstimatorGradientResult(
             gradients=gradients, metadata=metadata, options=results.options
@@ -256,10 +256,9 @@ def _validate_arguments(
             circuits: The list of quantum circuits to compute the gradients.
             observables: The list of observables.
             parameter_values: The list of parameter values to be bound to the circuit.
-            parameters: The Sequence of Sequence of Parameters to calculate only the gradients of
-                the specified parameters. Each Sequence of Parameters corresponds to a circuit in
-                ``circuits``. Defaults to None, which means that the gradients of all parameters in
-                each circuit are calculated.
+            parameter_sets: The Sequence of parameter sets to calculate only the gradients of
+                the specified parameters. Each set of parameters corresponds to a circuit in
+                ``circuits``.
 
         Raises:
             ValueError: Invalid arguments are given.

diff --git a/qiskit/algorithms/gradients/base_qfi.py b/qiskit/algorithms/gradients/base_qfi.py