evolib 0.2.0b2.dev1__tar.gz → 0.2.0b2.dev2__tar.gz

{evolib-0.2.0b2.dev1/evolib.egg-info → evolib-0.2.0b2.dev2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: evolib
-Version: 0.2.0b2.dev1
+Version: 0.2.0b2.dev2
 Summary: A modular framework for evolutionary strategies and neuroevolution.
 Author-email: EvoLib <evolib@dismail.de>
 License: MIT License

{evolib-0.2.0b2.dev1 → evolib-0.2.0b2.dev2}/evolib/core/population.py

@@ -442,8 +442,16 @@ class Pop:
             if indiv.age < self.max_indiv_age
         ]
 
+        if not survivors:
+            survivors = [self.best()]
+            print(
+                f"[Warning] All individuals exceeded max_age={self.max_indiv_age}. "
+                "Keeping best individual to prevent population collapse."
+            )
+
         removed_count = len(self.indivs) - len(survivors)
-        self.indivs = survivors
+        if removed_count > 0:
+            self.indivs = survivors
 
         return removed_count
 
@@ -743,7 +751,7 @@ class Pop:
                 on_generation_start(self)
 
             self.run_one_generation(strategy=strategy)
-            self.print_status(verbosity=1 if verbosity >= 1 else 0)
+            self.print_status(verbosity)
 
             if on_generation_end:
                 on_generation_end(self)

{evolib-0.2.0b2.dev1 → evolib-0.2.0b2.dev2}/evolib/operators/strategy.py

@@ -44,6 +44,9 @@ def evolve_mu_plus_lambda(pop: "Pop") -> None:
     if not pop.indivs:
         raise ValueError("Population is empty.")
 
+    # Age all current individuals
+    pop.age_indivs()
+
     # Update mutation/crossover parameters
     pop.update_parameters()
 
@@ -64,6 +67,9 @@ def evolve_mu_plus_lambda(pop: "Pop") -> None:
     # Select the best individuals
     replace_mu_lambda(pop, combined)
 
+    # Remove individuals that exceed max_age
+    pop.remove_old_indivs()
+
     pop.update_statistics()
 
 
@@ -79,6 +85,9 @@ def evolve_mu_comma_lambda(pop: "Pop") -> None:
     if not pop.indivs:
         raise ValueError("Population is empty.")
 
+    # Age all current individuals
+    pop.age_indivs()
+
     pop.evaluate_fitness()
 
     # Update mutation/crossover parameters
@@ -138,6 +147,9 @@ def evolve_steady_state(pop: "Pop") -> None:
             "to your YAML configuration."
         )
 
+    # Age all current individuals
+    pop.age_indivs()
+
     # Select parents (configurable)
     if (
         pop.config.selection is not None
@@ -166,6 +178,9 @@ def evolve_steady_state(pop: "Pop") -> None:
     # Replace worst individuals (excluding elites)
     replace_steady_state(pop, offspring, num_replace=pop.lambda_)
 
+    # Remove individuals that exceed max_age
+    pop.remove_old_indivs()
+
     # Update population statistics (including aging, diversity, fitness metrics)
     pop.update_statistics()
 
@@ -192,6 +207,9 @@ def evolve_flexible(pop: "Pop") -> None:
     if pop._replacement_fn is None:
         raise ValueError("Replacement function not configured.")
 
+    # Age all current individuals
+    pop.age_indivs()
+
     # Fitness
     pop.evaluate_fitness()
 
@@ -223,5 +241,8 @@ def evolve_flexible(pop: "Pop") -> None:
     # Replacement (via configured strategy)
     pop._replacement_fn(pop, offspring)
 
+    # Remove individuals that exceed max_age
+    pop.remove_old_indivs()
+
     # Update statistics
     pop.update_statistics()

{evolib-0.2.0b2.dev1 → evolib-0.2.0b2.dev2}/evolib/utils/plotting.py

@@ -277,6 +277,10 @@ def plot_approximation(
     true_label: str = "Target",
     pred_marker: str | None = None,
     true_marker: str | None = None,
+    pred_lw: int = 2,
+    true_lw: int = 2,
+    pred_ls: str = "--",
+    true_ls: str = "-",
     x_vals: Sequence[float] | np.ndarray | None = None,
     y_limits: tuple[float, float] | None = None,
     dpi: int = 150,
@@ -377,9 +381,23 @@ def plot_approximation(
     ax.set_xlabel(xlabel)
     ax.set_ylabel(ylabel)
 
-    ax.plot(x_vals, y_true, color="black", label=true_label, marker=true_marker, lw=2)
     ax.plot(
-        x_vals, y_pred, color="red", label=pred_label, marker=pred_marker, lw=2, ls="--"
+        x_vals,
+        y_true,
+        color="black",
+        label=true_label,
+        marker=true_marker,
+        lw=true_lw,
+        ls=true_ls,
+    )
+    ax.plot(
+        x_vals,
+        y_pred,
+        color="red",
+        label=pred_label,
+        marker=pred_marker,
+        lw=pred_lw,
+        ls=pred_ls,
     )
 
     if support_points is not None:

{evolib-0.2.0b2.dev1 → evolib-0.2.0b2.dev2/evolib.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: evolib
-Version: 0.2.0b2.dev1
+Version: 0.2.0b2.dev2
 Summary: A modular framework for evolutionary strategies and neuroevolution.
 Author-email: EvoLib <evolib@dismail.de>
 License: MIT License

{evolib-0.2.0b2.dev1 → evolib-0.2.0b2.dev2}/evolib.egg-info/SOURCES.txt

@@ -445,6 +445,7 @@ examples/06_netvector/03_netvector_gain_and_bias.py
 examples/07_evonet/01_sine_approximation.py
 examples/07_evonet/02_image_approximation.py
 examples/07_evonet/03_structural_xor.py
+examples/07_evonet/04_recurrent_bitseq.py
 examples/07_evonet/resize_all_images.py
 tests/test_benchmarks.py
 tests/test_config_loader.py

evolib-0.2.0b2.dev2/examples/07_evonet/04_recurrent_bitseq.py

@@ -0,0 +1,63 @@
+from evolib import Population, Individual, FitnessFunction, plot_approximation
+from evolib.representation.evonet import EvoNet
+import matplotlib.pyplot as plt
+import numpy as np
+
+FRAME_FOLDER = "04_frames_point"
+
+input_seq = [0, 1, 1, 0, 1, 0, 0, 0, 1, 1, 1] * 4
+target_seq = [1, 1, 0, 1, 0, 0, 0, 1, 1, 1, 0] * 4
+warmup_steps = 11
+
+
+def fitness_bitseq(indiv: Individual) -> float:
+    """
+    Evaluate how well the EvoNet predicts the next bit in a fixed sequence.
+    """
+    net: EvoNet = indiv.para["brain"].net
+    net.reset(full=True)
+
+    total_error = 0.0
+    for t in range(len(input_seq)):
+        output = net.calc([input_seq[t]])[0]
+        if t >= warmup_steps:
+            error = output - target_seq[t]
+            total_error += error**2
+    
+    return total_error / (len(input_seq) - warmup_steps)
+
+
+def save_plot(pop: Population):
+    """
+    Plot network predictions vs target values over time.
+    """
+    best = pop.best()
+    y_preds = [best.para["brain"].net.calc([bit])[0] for bit in input_seq]
+
+    plot_approximation(
+        y_preds,
+        target_seq,
+        title=f"Bit Prediction over Time (gen={pop.generation_num}, "
+        f"MSE={best.fitness:.4f})",
+        pred_label="Prediction",
+        show=False,
+        show_grid=False,
+        save_path=f"{FRAME_FOLDER}/gen_{pop.generation_num:03d}.png",
+        y_limits=(-0.2, 1.2),
+        true_marker="o",
+        pred_marker="o",
+        true_lw=0,
+        pred_lw=1,
+        true_ls="",
+        pred_ls="",
+    )
+
+    print(pop.mu)
+
+
+if __name__ == "__main__":
+    pop = Population("configs/04_recurrent_bitseq.yaml", 
+                     fitness_function = fitness_bitseq)
+    pop.run(verbosity=1, on_generation_end=save_plot)
+#    pop.run(verbosity=2)
+

{evolib-0.2.0b2.dev1 → evolib-0.2.0b2.dev2}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "evolib"
-version = "0.2.0b2.dev1"
+version = "0.2.0b2.dev2"
 description = "A modular framework for evolutionary strategies and neuroevolution."
 authors = [
     { name = "EvoLib", email = "evolib@dismail.de" }