evolib 0.2.0b2.dev7__tar.gz → 0.2.0b3.dev2__tar.gz

{evolib-0.2.0b2.dev7/evolib.egg-info → evolib-0.2.0b3.dev2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: evolib
-Version: 0.2.0b2.dev7
+Version: 0.2.0b3.dev2
 Summary: A modular framework for evolutionary strategies and neuroevolution.
 Author-email: EvoLib <evolib@dismail.de>
 License: MIT License
@@ -45,6 +45,10 @@ Requires-Dist: matplotlib>=3.10
 Provides-Extra: dev
 Requires-Dist: mypy; extra == "dev"
 Requires-Dist: types-PyYAML; extra == "dev"
+Requires-Dist: ray; extra == "dev"
+Requires-Dist: gymnasium; extra == "dev"
+Requires-Dist: imageio; extra == "dev"
+Requires-Dist: Box2D; extra == "dev"
 Provides-Extra: docs
 Requires-Dist: sphinx; extra == "docs"
 Requires-Dist: sphinx-rtd-theme; extra == "docs"
@@ -76,6 +80,7 @@ EvoLib is a lightweight and transparent framework for evolutionary computation,
 - **Modularity**: mutation, selection, crossover, and parameter representations can be freely combined.  
 - **Educational value**: examples and a clean API make it practical for illustrating evolutionary concepts.  
 - **Neuroevolution support**: structural mutations (adding/removing neurons and connections) and evolvable networks via EvoNet.  
+- **Gymnasium integration**: run standard RL benchmarks (e.g. CartPole, LunarLander) via a simple wrapper.
 - **Parallel evaluation (optional)**: basic support for [Ray](https://www.ray.io/) to speed up fitness evaluations.  
 - **Type-checked**: PEP8 compliant, and consistent code style.  
 
@@ -206,6 +211,27 @@ It can be applied to:
 
 ---
 
+### Gymnasium Integration
+
+EvoLib provides a lightweight wrapper for [Gymnasium](https://gymnasium.farama.org/) environments.
+This allows you to evaluate evolutionary agents directly on well-known RL benchmarks such as **CartPole**, **LunarLander**, or **Pendulum**.
+
+- **Headless evaluation**: returns total episode reward as fitness.
+- **Visualization**: render episodes and save them as GIFs.
+- **Discrete & continuous action spaces** are both supported.
+
+👉 Example: [`examples/01_cartpole.py`](examples/01_cartpole.py)
+
+```python
+from evolibs.gym_wrapper import GymEnv
+
+env = GymEnv("CartPole-v1", max_steps=500)
+fitness = env.evaluate(indiv)         # run one episode
+gif = env.visualize(indiv, gen=10)    # render & save as GIF
+```
+
+---
+
 ## Preview: Pygame Integration
 
 Early prototypes demonstrate how evolutionary algorithms can evolve both neural networks and sensor properties such as number, range, and orientation for agents in 2D worlds built with pygame. This illustrates how networks and sensors co-adapt to dynamic environments with collisions and feedback.
@@ -253,6 +279,7 @@ For deeper exploration, see the [full examples directory](examples/)
 - [X] Neuroevolution
 - [X] Topological Evolution (neurons, edges)
 - [X] Ray Support for Parallel Evaluation (early prototypes)
+- [X] OpenAI Gymnasium / Gym Wrapper
 - [ ] Advanced Visualization
 - [ ] Game Environment Integration (pygame, PettingZoo - early prototypes)
 

{evolib-0.2.0b2.dev7 → evolib-0.2.0b3.dev2}/README.md

@@ -20,6 +20,7 @@ EvoLib is a lightweight and transparent framework for evolutionary computation,
 - **Modularity**: mutation, selection, crossover, and parameter representations can be freely combined.  
 - **Educational value**: examples and a clean API make it practical for illustrating evolutionary concepts.  
 - **Neuroevolution support**: structural mutations (adding/removing neurons and connections) and evolvable networks via EvoNet.  
+- **Gymnasium integration**: run standard RL benchmarks (e.g. CartPole, LunarLander) via a simple wrapper.
 - **Parallel evaluation (optional)**: basic support for [Ray](https://www.ray.io/) to speed up fitness evaluations.  
 - **Type-checked**: PEP8 compliant, and consistent code style.  
 
@@ -150,6 +151,27 @@ It can be applied to:
 
 ---
 
+### Gymnasium Integration
+
+EvoLib provides a lightweight wrapper for [Gymnasium](https://gymnasium.farama.org/) environments.
+This allows you to evaluate evolutionary agents directly on well-known RL benchmarks such as **CartPole**, **LunarLander**, or **Pendulum**.
+
+- **Headless evaluation**: returns total episode reward as fitness.
+- **Visualization**: render episodes and save them as GIFs.
+- **Discrete & continuous action spaces** are both supported.
+
+👉 Example: [`examples/01_cartpole.py`](examples/01_cartpole.py)
+
+```python
+from evolibs.gym_wrapper import GymEnv
+
+env = GymEnv("CartPole-v1", max_steps=500)
+fitness = env.evaluate(indiv)         # run one episode
+gif = env.visualize(indiv, gen=10)    # render & save as GIF
+```
+
+---
+
 ## Preview: Pygame Integration
 
 Early prototypes demonstrate how evolutionary algorithms can evolve both neural networks and sensor properties such as number, range, and orientation for agents in 2D worlds built with pygame. This illustrates how networks and sensors co-adapt to dynamic environments with collisions and feedback.
@@ -197,6 +219,7 @@ For deeper exploration, see the [full examples directory](examples/)
 - [X] Neuroevolution
 - [X] Topological Evolution (neurons, edges)
 - [X] Ray Support for Parallel Evaluation (early prototypes)
+- [X] OpenAI Gymnasium / Gym Wrapper
 - [ ] Advanced Visualization
 - [ ] Game Environment Integration (pygame, PettingZoo - early prototypes)
 

{evolib-0.2.0b2.dev7 → evolib-0.2.0b3.dev2}/docs/config_guide.md

@@ -155,6 +155,21 @@ modules:
 
 ---
 
+## D) Parallel Evaluation (optional)
+
+For expensive problems, EvoLib can evaluate individuals in parallel using [Ray](https://www.ray.io/).
+
+```yaml
+parallel:
+  backend: ray         # backend: none | ray
+  num_cpus: 2          # number of logical CPUs (only used in local mode)
+  address: auto        # "auto" = local Ray; or "ray://host:10001" for remote
+```
+
+If omitted, EvoLib runs in single-threaded mode.
+
+---
+
 ## Further examples
 
 For complete, runnable examples including fitness definitions and visualization,

{evolib-0.2.0b2.dev7 → evolib-0.2.0b3.dev2}/docs/config_parameter.md

@@ -17,6 +17,26 @@ The parameters are grouped into **global parameters**, **evolution strategy sett
 | `num_elites`          | int  | 0       | Number of top individuals copied unchanged into the next generation. |
 | `max_indiv_age`       | int  | 0       | Maximum age of individuals (0 = no age limit).                       |
 
+
+## Parallelization Settings
+
+Optional parameters to enable parallel evaluation of individuals.
+
+| Parameter   | Type | Default | Explanation                                                                 |
+| ----------- | ---- | ------- | --------------------------------------------------------------------------- |
+| `backend`   | str  | none    | Parallel backend (`ray` or `none`).                                          |
+| `num_cpus`  | int  | 1       | Number of logical CPUs Ray may use for evaluation.                          |
+| `address`   | str  | auto    | `"auto"` = local Ray; or `ray://host:port` for connecting to a remote Ray cluster. |
+
+Example:
+
+```yaml
+parallel:
+  backend: ray
+  num_cpus: 4
+  address: auto
+```
+
 ---
 
 ## Stopping Criteria

{evolib-0.2.0b2.dev7 → evolib-0.2.0b3.dev2}/docs/index.md

@@ -20,6 +20,7 @@ EvoLib is a lightweight and transparent framework for evolutionary computation,
 - **Modularity**: mutation, selection, crossover, and parameter representations can be freely combined.  
 - **Educational value**: examples and a clean API make it practical for illustrating evolutionary concepts.  
 - **Neuroevolution support**: structural mutations (adding/removing neurons and connections) and evolvable networks via EvoNet.  
+- **Parallel evaluation (optional)**: basic support for [Ray](https://www.ray.io/) to speed up fitness evaluations.  
 - **Type-checked**: PEP8 compliant, and consistent code style.  
 
 
@@ -190,7 +191,7 @@ For deeper exploration, see the [full examples directory](examples/)
 - [X] Composite Parameters (multi-module individuals)
 - [X] Neuroevolution
 - [X] Topological Evolution (neurons, edges)
-- [ ] Co-Evolution & Speciation Support
+- [X] Ray Support for Parallel Evaluation (early prototypes)
 - [ ] Advanced Visualization
 - [ ] Game Environment Integration (pygame, PettingZoo - early prototypes)
 - [ ] Ray Support for Parallel Evaluation (early prototypes)

{evolib-0.2.0b2.dev7 → evolib-0.2.0b3.dev2}/evolib/api.py

@@ -1,5 +1,6 @@
 from evolib.core.individual import Indiv as Individual
 from evolib.core.population import Pop as Population
+from evolib.envs.gym_wrapper import GymEnv
 from evolib.interfaces.types import FitnessFunction
 from evolib.io.checkpoint import (
     load_best_indiv,
@@ -100,6 +101,7 @@ __all__ = [
     "load_best_indiv",
     "resume_or_create",
     "FitnessFunction",
+    "GymEnv",
 ]
 
 Pop = Population

{evolib-0.2.0b2.dev7 → evolib-0.2.0b3.dev2}/evolib/config/base_component_config.py

@@ -10,7 +10,7 @@ into the respective Para* representations and operator modules.
 
 from typing import Literal, Optional, Union
 
-from pydantic import BaseModel, Field, model_validator
+from pydantic import BaseModel, ConfigDict, Field, model_validator
 
 from evolib.interfaces.enums import (
     CrossoverOperator,
@@ -59,6 +59,8 @@ class MutationConfig(BaseModel):
     helpers.
     """
 
+    model_config = ConfigDict(extra="forbid")
+
     strategy: MutationStrategy = Field(..., description="Mutation strategy to use.")
 
     # Generic / commonly used parameters
@@ -133,11 +135,15 @@ class MutationConfig(BaseModel):
         # Strategy-specific light requirements
         if self.strategy.name == "CONSTANT":
             if self.strength is None:
-                raise ValueError("CONSTANT strategy requires 'strength'.")
+                raise ValueError("CONSTANT requires 'strength'.")
+            if self.init_strength is not None or self.init_probability is not None:
+                raise ValueError("CONSTANT must not define init_* fields.")
 
         if self.strategy.name == "EXPONENTIAL_DECAY":
             if self.init_strength is None:
                 raise ValueError("EXPONENTIAL_DECAY requires 'init_strength'.")
+            if self.strength is not None:
+                raise ValueError("EXPONENTIAL_DECAY must not define 'strength'.")
 
         if self.strategy.name == "ADAPTIVE_GLOBAL":
             if self.init_strength is None or self.init_probability is None:
@@ -168,6 +174,8 @@ class StructuralMutationConfig(BaseModel):
     max_edges: Optional[int] = None
     recurrent: Optional[Literal["none", "direct", "local", "all"]] = "none"
 
+    model_config = ConfigDict(extra="forbid")
+
     @model_validator(mode="after")
     def _check_ranges(self) -> "StructuralMutationConfig":
         for name in [
@@ -194,6 +202,9 @@ class StructuralMutationConfig(BaseModel):
 
 
 class ActivationMutationConfig(BaseModel):
+
+    model_config = ConfigDict(extra="forbid")
+
     probability: float = Field(
         ..., description="Per-neuron mutation probability in [0,1]."
     )
@@ -230,6 +241,8 @@ class EvoNetMutationConfig(MutationConfig):
         - structural: MutationConfig for structural choices (optional)
     """
 
+    model_config = ConfigDict(extra="forbid")
+
     biases: Optional[MutationConfig] = Field(
         default=None, description="Optional override for bias mutation."
     )
@@ -265,6 +278,8 @@ class CrossoverConfig(BaseModel):
         - ...others can be added as needed
     """
 
+    model_config = ConfigDict(extra="forbid")
+
     strategy: CrossoverStrategy = Field(..., description="Crossover strategy to use.")
     operator: Optional[CrossoverOperator] = Field(
         default=None, description="Concrete crossover operator (if applicable)."

{evolib-0.2.0b2.dev7 → evolib-0.2.0b3.dev2}/evolib/config/evonet_component_config.py

@@ -14,7 +14,7 @@ config resolution.
 from typing import Literal, Optional, Tuple, Union
 
 from evonet.activation import ACTIVATIONS
-from pydantic import BaseModel, Field, field_validator, validator
+from pydantic import BaseModel, ConfigDict, Field, field_validator, validator
 from pydantic_core import core_schema
 
 from evolib.config.base_component_config import (
@@ -47,6 +47,8 @@ class EvoNetComponentConfig(BaseModel):
               strength: 0.05
     """
 
+    model_config = ConfigDict(extra="forbid")
+
     # Module type is fixed to "evonet"
     type: Literal["evonet"] = "evonet"
 

{evolib-0.2.0b2.dev7 → evolib-0.2.0b3.dev2}/evolib/config/schema.py

@@ -1,7 +1,7 @@
 # SPDX-License-Identifier: MIT
 from typing import Any, Dict, Optional
 
-from pydantic import BaseModel, Field, model_validator
+from pydantic import BaseModel, ConfigDict, Field, model_validator
 
 from evolib.config.component_registry import get_component_config_class
 from evolib.interfaces.enums import (
@@ -19,6 +19,8 @@ class EvolutionConfig(BaseModel):
     Para* representations and operator modules.
     """
 
+    model_config = ConfigDict(extra="forbid")
+
     strategy: EvolutionStrategy = Field(
         ..., description="High-level evolution strategy (e.g. (mu_plus_lambda)."
     )
@@ -32,6 +34,8 @@ class SelectionConfig(BaseModel):
     semantics are implemented in the selection registry.
     """
 
+    model_config = ConfigDict(extra="forbid")
+
     strategy: SelectionStrategy = Field(
         ..., description="Parent selection strategy (e.g. tournament, ranking)."
     )
@@ -56,6 +60,8 @@ class ReplacementConfig(BaseModel):
     Concrete behavior is implemented in the replacement registry.
     """
 
+    model_config = ConfigDict(extra="forbid")
+
     strategy: ReplacementStrategy = Field(
         ..., description="Survivor selection strategy (e.g. replace_worst, anneal)."
     )
@@ -70,6 +76,8 @@ class ReplacementConfig(BaseModel):
 class StoppingCriteria(BaseModel):
     """Optional stopping criteria for early stopping."""
 
+    model_config = ConfigDict(extra="forbid")
+
     target_fitness: Optional[float] = Field(
         None,
         description="Stop once fitness is below (or above if maximize) this value.",
@@ -96,6 +104,8 @@ class ParallelConfig(BaseModel):
     only Ray is supported).
     """
 
+    model_config = ConfigDict(extra="forbid")
+
     backend: str = Field(
         default="none",
         description="Parallel backend to use: 'none' (default) or 'ray'.",
@@ -124,6 +134,8 @@ class FullConfig(BaseModel):
     VectorComponentConfig, EvoNetComponentConfig).
     """
 
+    model_config = ConfigDict(extra="forbid")
+
     # Global run parameters
     parent_pool_size: int = Field(
         ..., description="Number of parents retained in each generation."
@@ -199,3 +211,20 @@ class FullConfig(BaseModel):
 
         data["modules"] = resolved
         return data
+
+    @model_validator(mode="after")
+    def _check_consistency(self) -> "FullConfig":
+        """Global sanity checks for the main configuration."""
+        if self.parent_pool_size <= 0:
+            raise ValueError("parent_pool_size must be > 0")
+        if self.offspring_pool_size <= 0:
+            raise ValueError("offspring_pool_size must be > 0")
+        if self.num_elites < 0:
+            raise ValueError("num_elites must be >= 0")
+        if self.num_elites > self.parent_pool_size:
+            raise ValueError("num_elites cannot exceed parent_pool_size")
+        if self.max_generations <= 0:
+            raise ValueError("max_generations must be > 0")
+        if self.max_indiv_age < 0:
+            raise ValueError("max_indiv_age must be >= 0")
+        return self

{evolib-0.2.0b2.dev7 → evolib-0.2.0b3.dev2}/evolib/config/vector_component_config.py

@@ -1,7 +1,7 @@
 # SPDX-License-Identifier: MIT
 from typing import Any, Literal, Optional, Tuple, Union
 
-from pydantic import BaseModel, Field, field_validator, model_validator
+from pydantic import BaseModel, ConfigDict, Field, field_validator, model_validator
 
 from evolib.config.base_component_config import CrossoverConfig, MutationConfig
 from evolib.interfaces.enums import RepresentationType
@@ -29,6 +29,8 @@ class VectorComponentConfig(BaseModel):
               strength: 0.05
     """
 
+    model_config = ConfigDict(extra="forbid")
+
     # Fixed module type: "vector"
     type: RepresentationType = Field(
         default=RepresentationType.VECTOR,

evolib-0.2.0b3.dev2/evolib/envs/gym_wrapper.py

@@ -0,0 +1,159 @@
+# SPDX-License-Identifier: MIT
+from __future__ import annotations
+
+from typing import TYPE_CHECKING, Any, Union, cast
+
+try:
+    import gymnasium as gym
+except ImportError as e:
+    raise ImportError(
+        "Gymnasium is required for GymEnv. Please install with `pip install gymnasium`."
+    ) from e
+
+try:
+    import imageio
+except ImportError as e:
+    raise ImportError(
+        "imageio is required for GymEnv visualization. "
+        "Please install with `pip install imageio`."
+    ) from e
+
+import numpy as np
+
+if TYPE_CHECKING:
+    from evolib import Individual
+
+
+class GymEnv:
+    """Thin wrapper to run OpenAI Gymnasium environments with EvoLib Individuals."""
+
+    def __init__(self, env_name: str, max_steps: int = 500, **env_kwargs: Any):
+        """
+        Initialize a Gym environment.
+
+        Args:
+            env_name: Name of the Gymnasium environment, e.g. "FrozenLake-v1".
+            max_steps: Maximum number of steps per episode.
+            **env_kwargs: Extra arguments passed directly to gym.make(),
+                          e.g. is_slippery=False for FrozenLake.
+        """
+        self.env_name = env_name
+        self.max_steps = max_steps
+        self.env_kwargs = env_kwargs
+        # headless env (no Render)
+        self.env = gym.make(env_name, **env_kwargs)
+
+    def evaluate(
+        self,
+        indiv: Individual,
+        module: str = "brain",
+        episodes: int = 1,
+    ) -> float:
+        """
+        Run one or multiple episodes headless and return average total reward.
+
+        Args:
+            indiv: Individual whose network acts in the environment.
+            module: Which module in para to use for decision making.
+            episodes: How many episodes to average over (default: 1).
+
+        Returns:
+            Average total reward across all episodes.
+        """
+        total_reward = 0.0
+
+        for _ in range(episodes):
+            obs, _ = self.env.reset()
+            ep_reward = 0.0
+
+            for _ in range(self.max_steps):
+                if isinstance(obs, np.ndarray):
+                    obs_list = obs.tolist()
+                elif np.isscalar(obs):
+                    obs_list = [float(cast(float, obs))]
+                else:
+                    obs_list = list(obs)
+
+                action = indiv.para[module].net.calc(obs_list)
+
+                # Discrete Action-Spaces --> argmax
+                if hasattr(self.env.action_space, "n"):
+                    action = int(np.argmax(action))
+                else:
+                    action = np.array(action, dtype=np.float32)
+
+                obs, reward, terminated, truncated, _ = self.env.step(action)
+                ep_reward += float(reward)
+
+                if terminated or truncated:
+                    break
+
+            total_reward += ep_reward
+
+        return total_reward / episodes
+
+    def visualize(
+        self,
+        indiv: Individual,
+        gen: int,
+        filename: str | None = None,
+        fps: int = 30,
+        module: str = "brain",
+    ) -> str:
+        """
+        Render an episode with the given individual and save as GIF using imageio.
+
+        Args:
+            indiv: Individual to visualize.
+            gen: Generation number (used in default filename).
+            filename: Optional filename for output GIF.
+            fps: Frames per second for GIF.
+            module: Which module in para to use for decision making.
+
+        Returns:
+            Path to saved GIF.
+        """
+
+        env = gym.make(
+            self.env_name,
+            render_mode="rgb_array",
+            max_episode_steps=self.max_steps,
+            **self.env_kwargs,
+        )
+        obs, _ = env.reset()
+
+        RenderFrame = Union[np.ndarray, list[np.ndarray], None]
+        frames: list[np.ndarray] = []
+
+        for _ in range(self.max_steps):
+            if isinstance(obs, np.ndarray):
+                obs_list = obs.tolist()
+            elif np.isscalar(obs):
+                obs_list = [float(cast(float, obs))]
+            else:
+                obs_list = list(obs)
+
+            action = indiv.para[module].net.calc(obs_list)
+
+            if hasattr(env.action_space, "n"):
+                action = int(np.argmax(action))
+            else:
+                action = np.array(action, dtype=np.float32)
+
+            obs, reward, terminated, truncated, _ = env.step(action)
+
+            frame: RenderFrame = env.render()
+            if isinstance(frame, np.ndarray):
+                frames.append(frame)
+
+            if terminated or truncated:
+                break
+
+        env.close()
+
+        if filename is None:
+            filename = f"{self.env_name}_gen{gen:04d}.gif"
+
+        imageio.mimsave(filename, cast(list[Any], frames), fps=fps)
+
+        return filename

{evolib-0.2.0b2.dev7 → evolib-0.2.0b3.dev2}/evolib/io/checkpoint.py

@@ -66,7 +66,7 @@ def save_checkpoint(pop: Pop, *, run_name: str = "default") -> None:
 
 def resume_from_checkpoint(
     run_name: str = "default",
-    fitness_fn: Optional[FitnessFunction] = None,
+    fitness_function: Optional[FitnessFunction] = None,
     silent_fail: bool = True,
 ) -> Optional[Pop]:
     """
@@ -74,7 +74,7 @@ def resume_from_checkpoint(
 
     Args:
     run_name (str): Identifier of the checkpoint file.
-    fitness_fn (callable, optional): Fitness function to assign.
+    fitness_function (callable, optional): Fitness function to assign.
     silent_fail (bool): If True, return None instead of raising FileNotFoundError.
 
 
@@ -89,8 +89,8 @@ def resume_from_checkpoint(
 
     pop = load_population_pickle(path)
 
-    if fitness_fn:
-        pop.set_fitness_function(fitness_fn)
+    if fitness_function:
+        pop.set_fitness_function(fitness_function)
 
     # Restore initializer
     pop.para_initializer = build_composite_initializer(pop.config)
@@ -99,14 +99,14 @@ def resume_from_checkpoint(
 
 
 def resume_or_create(
-    config_path: str, fitness_fn: FitnessFunction, run_name: str = "default"
+    config_path: str, fitness_function: FitnessFunction, run_name: str = "default"
 ) -> Pop:
     """
     Try to resume a saved run, otherwise initialize a new population.
 
     Args:
     config_path (str): Path to the YAML configuration.
-    fitness_fn (callable): Fitness function for individuals.
+    fitness_function (callable): Fitness function for individuals.
     run_name (str): Optional name for checkpoint file.
 
 
@@ -115,7 +115,7 @@ def resume_or_create(
     """
 
     pop = resume_from_checkpoint(
-        run_name=run_name, fitness_fn=fitness_fn, silent_fail=True
+        run_name=run_name, fitness_function=fitness_function, silent_fail=True
     )
 
     if pop is not None:
@@ -126,7 +126,7 @@ def resume_or_create(
         return pop
 
     print(f"[Create] New population for run_name={run_name}")
-    return Pop(config_path=config_path, fitness_function=fitness_fn)
+    return Pop(config_path=config_path, fitness_function=fitness_function)
 
 
 def save_best_indiv(pop: Pop, *, run_name: str = "default") -> None:

{evolib-0.2.0b2.dev7 → evolib-0.2.0b3.dev2}/evolib/representation/evonet.py

@@ -392,7 +392,7 @@ class EvoNet(ParaBase):
     def print_status(self) -> None:
         print(f"[EvoNet] : {self.net} ")
 
-    def print_graph(
+    def plot(
         self,
         name: str,
         engine: str = "neato",
@@ -412,7 +412,7 @@ class EvoNet(ParaBase):
             thickness_on (bool): Adjust edge thickness by weight.
             fillcolors_on (bool): Fill nodes with colors by type.
         """
-        self.net.print_graph(
+        self.net.plot(
             name=name,
             engine=engine,
             labels_on=labels_on,

{evolib-0.2.0b2.dev7 → evolib-0.2.0b3.dev2}/evolib/utils/benchmarks.py

@@ -1,7 +1,7 @@
 # SPDX-License-Identifier: MIT
 """Common mathematical benchmark functions for optimization tasks."""
 
-from typing import Literal, Sequence
+from typing import Sequence
 
 import numpy as np
 from numpy.random import default_rng
@@ -10,9 +10,7 @@ from numpy.random import default_rng
 def generate_timeseries(
     length: int,
     normalize: bool = True,
-    pattern: Literal[
-        "default", "trend_switch", "parabolic", "zigzag", "shock"
-    ] = "default",
+    pattern: str = "default",
     seed: int | None = None,
 ) -> np.ndarray:
     """
@@ -33,6 +31,9 @@ def generate_timeseries(
         np.ndarray: The generated time series.
     """
 
+    if pattern not in {"default", "trend_switch", "parabolic", "zigzag", "shock"}:
+        raise ValueError(f"Unknown pattern: {pattern}")
+
     if seed is not None:
         rng = default_rng(seed)
     else:

{evolib-0.2.0b2.dev7 → evolib-0.2.0b3.dev2}/evolib/utils/plotting.py

@@ -513,7 +513,7 @@ def save_combined_net_plot(
     Saves a combined image of network structure and approximation plot.
 
     Args:
-        net: EvoNet instance with .print_graph().
+        net: EvoNet instance with .plot().
         X (np.ndarray): Input values (for plotting).
         Y_true (np.ndarray): Ground truth values (e.g. sin(x)).
         Y_pred (np.ndarray): Network prediction values.
@@ -523,7 +523,7 @@ def save_combined_net_plot(
     """
     with tempfile.TemporaryDirectory() as tmpdir:
         graph_path = os.path.join(tmpdir, "net")
-        net.print_graph(graph_path, fillcolors_on=True, thickness_on=True)
+        net.plot(graph_path, fillcolors_on=True, thickness_on=True)
         img_net = Image.open(graph_path + ".png")
 
         # Create approximation plot