Tutorial for the ODE QNN Regressor

docs/examples/example_qnn_ode_solver.nblink

@@ -0,0 +1,6 @@
+{
+    "path": "../../examples/tutorials/qnn_ode_solver.ipynb",
+    "extra-media": [
+        "../../examples/tutorials/images"
+    ]
+}

docs/examples/examples_index.rst

@@ -10,4 +10,5 @@ Examples
     example_kernel_digit_classification
     example_kernel_grid_search
     example_quantum_bayesian_optimization
-    example_qnn_backend_mitigation
+    example_qnn_backend_mitigation
+    example_qnn_ode_solver

examples/tutorials/quantum_bayesian_optimization.ipynb

@@ -34,7 +34,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "The follwing example will adapt the structre of the examples made in [scikit-optimize](https://scikit-optimize.github.io/stable/auto_examples/bayesian-optimization.html).\n",
+    "The follwing example will adapt the structure of the examples made in [scikit-optimize](https://scikit-optimize.github.io/stable/auto_examples/bayesian-optimization.html).\n",
     "\n",
     "We start by defining the Bayesian optimization base class."
    ]

pyproject.toml

@@ -93,4 +93,5 @@ extension-pkg-allow-list = [
 addopts = [
     "--import-mode=importlib",
     "--ignore=examples/tutorials/qnn_backend_mitigation.ipynb",
+    "--ignore=examples/tutorials/qnn_ode_solver.ipynb",
 ]

src/squlearn/encoding_circuit/circuit_library/kyriienko_nonlinear_encoding_circuit.py

@@ -198,6 +198,7 @@ def variational_gate_block(
             )
             QC.add_layer(layer, num_layers=self.num_encoding_layers)
 
+            QC = QC.get_circuit(features, [])
         elif self.encoding_style in ("chebyshev_tower", "chebyshev_sparse"):
             QC = QuantumCircuit(self.num_qubits)
             for layer in range(self.num_encoding_layers):
@@ -223,7 +224,9 @@ def variational_gate_block(
             QC = variational_gate_block(QC, self.num_variational_layers, self.num_qubits)[0]
         elif self.variational_arrangement == "ABA":
             if self.num_qubits % self.block_width != 0:
-                raise ValueError("Block width must be a divisor of the number of qubits.")
+                raise ValueError(
+                    f"block_width =  {self.block_width} must be a divisor of the num_qubits = {self.num_qubits}."
+                )
 
             number_of_blocks = int(np.ceil(self.num_qubits / self.block_width))  # vertical blocks
             shifting_factor = np.floor(self.block_width / 2)