refactor adapted to pull request comments

src/qibotn/backends/__init__.py

@@ -1,5 +1,50 @@
-import importlib.metadata as im
+from typing import Union
 
-from qibotn.backends import MetaBackend
+from qibo.config import raise_error
 
-__version__ = im.version(__package__)
+from qibotn.backends.abstract import QibotnBackend
+from qibotn.backends.cutensornet import CuTensorNet  # pylint: disable=E0401
+from qibotn.backends.quimb import QuimbBackend  # pylint: disable=E0401
+
+PLATFORMS = ("cutensornet", "qutensornet", "qmatchatea")
+
+
+class MetaBackend:
+    """Meta-backend class which takes care of loading the qibotn backends."""
+
+    @staticmethod
+    def load(platform: str, runcard: dict = None) -> QibotnBackend:
+        """Loads the backend.
+
+        Args:
+            platform (str): Name of the backend to load: either `cutensornet` or `qutensornet`.
+            runcard (dict): Dictionary containing the simulation settings.
+        Returns:
+            qibo.backends.abstract.Backend: The loaded backend.
+        """
+
+        if platform == "cutensornet":  # pragma: no cover
+            return CuTensorNet(runcard)
+        elif platform == "quimb":  # pragma: no cover
+            return QuimbBackend(runcard)
+        elif platform == "qmatchatea":  # pragma: no cover
+            from qibotn.backends.qmatchatea import QMatchaTeaBackend
+
+            return QMatchaTeaBackend()
+        else:
+            raise_error(
+                NotImplementedError,
+                f"Unsupported platform {platform}, please pick one in {PLATFORMS}",
+            )
+
+    def list_available(self) -> dict:
+        """Lists all the available qibotn backends."""
+        available_backends = {}
+        for platform in PLATFORMS:
+            try:
+                MetaBackend.load(platform=platform)
+                available = True
+            except:
+                available = False
+            available_backends[platform] = available
+        return available_backends

src/qibotn/backends/quimb.py

@@ -24,6 +24,7 @@ class QuimbBackend(QibotnBackend, NumpyBackend):
         self,
         ansatz: str = "MPS",
         max_bond_dimension: int = 10,
+        n_most_frequent_states: int = 100,
     ):
         """
         Configure tensor network simulation.
@@ -40,6 +41,7 @@ class QuimbBackend(QibotnBackend, NumpyBackend):
         """
         self.ansatz = ansatz
         self.max_bond_dimension = max_bond_dimension
+        self.n_most_frequent_states = n_most_frequent_states
 
     def setup_backend_specifics(self, qimb_backend="numpy"):
         """Setup backend specifics.
@@ -69,6 +71,8 @@ class QuimbBackend(QibotnBackend, NumpyBackend):
                 The number of shots for sampling the circuit. If None, no sampling is performed, and the full statevector is used.
             return_array : bool, optional
                 If True, returns the statevector as a dense array. Default is False.
+            n_most_frequent_states : int, optional
+                The number of most frequent computational basis states to return. Default is 100.
             **prob_kwargs : dict, optional
                 Additional keyword arguments for probability computation (currently unused).
 
@@ -109,9 +113,7 @@ class QuimbBackend(QibotnBackend, NumpyBackend):
         )
 
         frequencies = Counter(circ_quimb.sample(nshots)) if nshots is not None else None
-        main_frequencies = {
-            state: count for state, count in frequencies.most_common(n=100)
-        }
+        main_frequencies = {state: count for state, count in frequencies.most_common(self.n_most_frequent_states)}
         computational_states = [state for state in main_frequencies.keys()]
         amplitudes = {
             state: circ_quimb.amplitude(state) for state in computational_states