Fix additional ruff compliance issues

- Fixed abstract method issue in BaseOptimizer.reset method
- Fixed line length violations in test files and demo code
- Reduced from 18 to 10 remaining ruff violations
- All core functionality preserved and tested

Author: shretadas

neural_network/optimizers/base_optimizer.py

@@ -77,7 +77,6 @@ def reset(self) -> None:
         or when you want to clear any accumulated state (like momentum).
         Default implementation does nothing, but optimizers with state should override.
         """
-        ...
 
     def __str__(self) -> str:
         """String representation of the optimizer."""

neural_network/optimizers/momentum_sgd.py

@@ -118,7 +118,9 @@ def _check_shapes_and_get_velocity(
             parameters: float | list[float | list[float]],
             gradients: float | list[float | list[float]],
             velocity_values: float | list[float | list[float]]
-        ) -> tuple[float | list[float | list[float]], float | list[float | list[float]]]:
+        ) -> tuple[
+            float | list[float | list[float]], float | list[float | list[float]]
+        ]:
             # Handle scalar case
             if isinstance(parameters, (int, float)):
                 if not isinstance(gradients, (int, float)):
@@ -130,7 +132,9 @@ def _check_shapes_and_get_velocity(
                     velocity_values = 0.0
 
                 # Update velocity: v = β * v + (1-β) * g
-                new_velocity = self.momentum * velocity_values + (1 - self.momentum) * gradients
+                new_velocity = (
+                    self.momentum * velocity_values + (1 - self.momentum) * gradients
+                )
                 # Update parameter: θ = θ - alpha * v
                 new_param = parameters - self.learning_rate * new_velocity
 
@@ -271,7 +275,8 @@ def __str__(self) -> str:
                 f"  SGD:      f = {f_sgd:.6f}, x = ({x_sgd[0]:6.3f}, {x_sgd[1]:6.3f})"
             )
             print(
-                f"  Momentum: f = {f_momentum:.6f}, x = ({x_momentum[0]:6.3f}, {x_momentum[1]:6.3f})"
+                f"  Momentum: f = {f_momentum:.6f}, x = "
+                f"({x_momentum[0]:6.3f}, {x_momentum[1]:6.3f})"
             )
 
     print("\\nFinal comparison:")
@@ -280,5 +285,6 @@ def __str__(self) -> str:
     print(f"SGD final loss: {f_final_sgd:.6f}")
     print(f"Momentum final loss: {f_final_momentum:.6f}")
     print(
-        f"Improvement with momentum: {((f_final_sgd - f_final_momentum) / f_final_sgd * 100):.1f}%"
+        f"Improvement with momentum: "
+        f"{((f_final_sgd - f_final_momentum) / f_final_sgd * 100):.1f}%"
     )

neural_network/optimizers/nag.py

@@ -132,7 +132,9 @@ def _nag_update_recursive(
                     velocity = 0.0
 
                 # Update velocity: v = β * v + (1-β) * g
-                new_velocity = self.momentum * velocity + (1 - self.momentum) * gradients
+                new_velocity = (
+                    self.momentum * velocity + (1 - self.momentum) * gradients
+                )
 
                 # NAG update: θ = θ - alpha * (β * v + (1-β) * g)
                 nesterov_update = (

neural_network/optimizers/test_optimizers.py

@@ -201,7 +201,8 @@ def rosenbrock_grad(x: float, y: float) -> list[float]:
         loss = rosenbrock(x, y)
         distance_to_optimum = math.sqrt((x - 1) ** 2 + (y - 1) ** 2)
         print(
-            f"{name:12s}: loss = {loss:8.3f}, pos = ({x:6.3f}, {y:6.3f}), dist = {distance_to_optimum:.4f}"
+            f"{name:12s}: loss = {loss:8.3f}, pos = ({x:6.3f}, {y:6.3f}), "
+            f"dist = {distance_to_optimum:.4f}"
         )
 
         if loss < best_loss:
@@ -255,7 +256,8 @@ def convergence_analysis() -> None:
         final_x = positions[name][0]
         if steps is not None:
             print(
-                f"{name:12s}: converged in {steps:2d} steps (final |x| = {abs(final_x):.6f})"
+                f"{name:12s}: converged in {steps:2d} steps "
+                f"(final |x| = {abs(final_x):.6f})"
             )
         else:
             print(f"{name:12s}: did not converge (final |x| = {abs(final_x):.6f})")