Fix islands implementation

CONTRIBUTING.md

@@ -29,7 +29,7 @@ black openevolve tests examples
 
 ## Pull Request Process
 
-1. Create a new branch for your feature or bugfix: `git checkout -b feature/your-feature-name`
+1. Create a new branch for your feature or bugfix: `git checkout -b feat-your-feature-name`
 2. Make your changes
 3. Add tests for your changes
 4. Run the tests to make sure everything passes: `python -m unittest discover tests`

configs/README.md

@@ -0,0 +1,73 @@
+# OpenEvolve Configuration Files
+
+This directory contains configuration files for OpenEvolve with examples for different use cases.
+
+## Configuration Files
+
+### `default_config.yaml`
+The main configuration file containing all available options with sensible defaults. This file includes:
+- Complete documentation for all configuration parameters
+- Default values for all settings
+- **Island-based evolution parameters** for proper evolutionary diversity
+
+Use this file as a template for your own configurations.
+
+### `island_config_example.yaml`
+A practical example configuration demonstrating proper island-based evolution setup. Shows:
+- Recommended island settings for most use cases
+- Balanced migration parameters
+- Complete working configuration
+
+### `island_examples.yaml`
+Multiple example configurations for different scenarios:
+- **Maximum Diversity**: Many islands, frequent migration
+- **Focused Exploration**: Few islands, rare migration  
+- **Balanced Approach**: Default recommended settings
+- **Quick Exploration**: Small-scale rapid testing
+- **Large-Scale Evolution**: Complex optimization runs
+
+Includes guidelines for choosing parameters based on your problem characteristics.
+
+## Island-Based Evolution Parameters
+
+The key new parameters for proper evolutionary diversity are:
+
+```yaml
+database:
+  num_islands: 5                      # Number of separate populations
+  migration_interval: 50              # Migrate every N generations  
+  migration_rate: 0.1                 # Fraction of top programs to migrate
+```
+
+### Parameter Guidelines
+
+- **num_islands**: 3-10 for most problems (more = more diversity)
+- **migration_interval**: 25-100 generations (higher = more independence)
+- **migration_rate**: 0.05-0.2 (5%-20%, higher = faster knowledge sharing)
+
+### When to Use What
+
+- **Complex problems** → More islands, less frequent migration
+- **Simple problems** → Fewer islands, more frequent migration
+- **Long runs** → More islands to maintain diversity
+- **Short runs** → Fewer islands for faster convergence
+
+## Usage
+
+Copy any of these files as a starting point for your configuration:
+
+```bash
+cp configs/default_config.yaml my_config.yaml
+# Edit my_config.yaml for your specific needs
+```
+
+Then use with OpenEvolve:
+
+```python
+from openevolve import OpenEvolve
+evolve = OpenEvolve(
+    initial_program_path="program.py",
+    evaluation_file="evaluator.py", 
+    config_path="my_config.yaml"
+)
+```

configs/default_config.yaml

@@ -55,9 +55,7 @@ prompt:
       - "I suggest the following improvements:"
       - "We can enhance this code by:"
 
-  # Meta-prompting (experimental)
-  use_meta_prompting: false           # Use LLM to generate parts of the prompt
-  meta_prompt_weight: 0.1             # Weight for meta-prompting influence
+  # Note: meta-prompting features are not yet implemented
 
 # Database configuration
 database:
@@ -68,13 +66,19 @@ database:
   # Evolutionary parameters
   population_size: 1000               # Maximum number of programs to keep in memory
   archive_size: 100                   # Size of elite archive
-  num_islands: 5                      # Number of islands for island model
+  num_islands: 5                      # Number of islands for island model (separate populations)
+
+  # Island-based evolution parameters
+  # Islands provide diversity by maintaining separate populations that evolve independently.
+  # Migration periodically shares the best solutions between adjacent islands.
+  migration_interval: 50              # Migrate between islands every N generations
+  migration_rate: 0.1                 # Fraction of top programs to migrate (0.1 = 10%)
 
   # Selection parameters
   elite_selection_ratio: 0.1          # Ratio of elite programs to select
   exploration_ratio: 0.2              # Ratio of exploration vs exploitation
   exploitation_ratio: 0.7             # Ratio of exploitation vs random selection
-  diversity_metric: "edit_distance"   # Diversity metric (edit_distance, feature_based)
+  # Note: diversity_metric is fixed to "edit_distance" (feature_based not implemented)
 
   # Feature map dimensions for MAP-Elites
   feature_dimensions:                 # Dimensions for MAP-Elites feature map
@@ -88,9 +92,7 @@ evaluator:
   timeout: 300                        # Maximum evaluation time in seconds
   max_retries: 3                      # Maximum number of retries for evaluation
 
-  # Resource limits
-  memory_limit_mb: null               # Memory limit for evaluation (null = no limit)
-  cpu_limit: null                     # CPU limit for evaluation (null = no limit)
+  # Note: resource limits (memory_limit_mb, cpu_limit) are not yet implemented
 
   # Evaluation strategies
   cascade_evaluation: true            # Use cascade evaluation to filter bad solutions early
@@ -101,7 +103,7 @@ evaluator:
 
   # Parallel evaluation
   parallel_evaluations: 4             # Number of parallel evaluations
-  distributed: false                  # Use distributed evaluation
+  # Note: distributed evaluation is not yet implemented
 
   # LLM-based feedback (experimental)
   use_llm_feedback: false             # Use LLM to evaluate code quality

configs/island_config_example.yaml

@@ -0,0 +1,55 @@
+# OpenEvolve Island-Based Evolution Configuration
+# This configuration demonstrates the proper use of island-based evolution
+
+# General settings
+max_iterations: 1000
+checkpoint_interval: 100
+log_level: "INFO"
+
+# LLM configuration
+llm:
+  primary_model: "gemini-2.0-flash-lite"
+  primary_model_weight: 0.8
+  secondary_model: "gemini-2.0-flash"
+  secondary_model_weight: 0.2
+  temperature: 0.7
+  top_p: 0.95
+  max_tokens: 4096
+
+# Database configuration with proper island settings
+database:
+  population_size: 500
+  archive_size: 100
+
+  # Island-based evolution settings
+  num_islands: 5                    # Number of separate populations
+  migration_interval: 50            # Migrate every 50 generations
+  migration_rate: 0.1               # Migrate 10% of top programs
+
+  # Selection parameters
+  elite_selection_ratio: 0.1
+  exploration_ratio: 0.3
+  exploitation_ratio: 0.7
+  # Note: diversity_metric fixed to "edit_distance"
+
+  # Feature map dimensions for MAP-Elites
+  feature_dimensions: ["score", "complexity"]
+  feature_bins: 10
+
+# Prompt configuration
+prompt:
+  num_top_programs: 3
+  num_diverse_programs: 2
+  use_template_stochasticity: true
+
+# Evaluator configuration
+evaluator:
+  timeout: 300
+  max_retries: 3
+  cascade_evaluation: true
+  parallel_evaluations: 4
+
+# Evolution settings
+diff_based_evolution: true
+allow_full_rewrites: false
+max_code_length: 10000

configs/island_examples.yaml

@@ -0,0 +1,80 @@
+# OpenEvolve Island-Based Evolution Configuration Examples
+# Different configurations for various use cases
+
+# Configuration for Maximum Diversity (Many Islands, Frequent Migration)
+# Use this when you want to explore the search space thoroughly
+# Good for: Complex problems, avoiding local optima, long runs
+max_diversity:
+  database:
+    num_islands: 10                   # More islands = more diversity
+    migration_interval: 25            # More frequent migration
+    migration_rate: 0.2               # Higher migration rate
+    population_size: 1000
+    archive_size: 200
+
+# Configuration for Focused Exploration (Few Islands, Rare Migration)  
+# Use this when you want deeper exploration within each island
+# Good for: Problems with clear structure, shorter runs
+focused_exploration:
+  database:
+    num_islands: 3                    # Fewer islands = deeper exploration
+    migration_interval: 100           # Less frequent migration
+    migration_rate: 0.05              # Lower migration rate
+    population_size: 500
+    archive_size: 50
+
+# Configuration for Balanced Approach (Default Settings)
+# Use this as a starting point for most problems
+# Good for: General use, medium-length runs
+balanced:
+  database:
+    num_islands: 5                    # Balanced number of islands
+    migration_interval: 50            # Moderate migration frequency
+    migration_rate: 0.1               # Moderate migration rate
+    population_size: 1000
+    archive_size: 100
+
+# Configuration for Quick Exploration (Small Scale)
+# Use this for rapid prototyping and testing
+# Good for: Small problems, quick experiments
+quick_exploration:
+  database:
+    num_islands: 3
+    migration_interval: 20
+    migration_rate: 0.15
+    population_size: 200
+    archive_size: 30
+
+# Configuration for Large-Scale Evolution (High Performance)
+# Use this for complex problems requiring extensive search
+# Good for: Complex optimization, long evolutionary runs
+large_scale:
+  database:
+    num_islands: 15                   # Many islands for parallel exploration
+    migration_interval: 75            # Balanced migration timing
+    migration_rate: 0.08              # Conservative migration rate
+    population_size: 2000             # Large populations
+    archive_size: 300
+
+# Guidelines for choosing parameters:
+#
+# num_islands:
+#   - More islands = more diversity, slower convergence
+#   - Fewer islands = faster convergence, risk of premature convergence
+#   - Recommended: 3-10 for most problems
+#
+# migration_interval:
+#   - Lower values = more frequent knowledge sharing
+#   - Higher values = more independent evolution
+#   - Recommended: 25-100 generations
+#
+# migration_rate:
+#   - Higher values = faster knowledge propagation
+#   - Lower values = preserve island diversity longer
+#   - Recommended: 0.05-0.2 (5%-20%)
+#
+# Rule of thumb:
+#   - Complex problems → More islands, less frequent migration
+#   - Simple problems → Fewer islands, more frequent migration
+#   - Long runs → More islands to maintain diversity
+#   - Short runs → Fewer islands for faster convergence

openevolve/config.py

@@ -80,6 +80,13 @@ class DatabaseConfig:
     feature_dimensions: List[str] = field(default_factory=lambda: ["score", "complexity"])
     feature_bins: int = 10
 
+    # Migration parameters for island-based evolution
+    migration_interval: int = 50  # Migrate every N generations
+    migration_rate: float = 0.1  # Fraction of population to migrate
+
+    # Random seed for reproducible sampling
+    random_seed: Optional[int] = None
+
 
 @dataclass
 class EvaluatorConfig:
@@ -188,8 +195,9 @@ def to_dict(self) -> Dict[str, Any]:
                 "num_diverse_programs": self.prompt.num_diverse_programs,
                 "use_template_stochasticity": self.prompt.use_template_stochasticity,
                 "template_variations": self.prompt.template_variations,
-                "use_meta_prompting": self.prompt.use_meta_prompting,
-                "meta_prompt_weight": self.prompt.meta_prompt_weight,
+                # Note: meta-prompting features not implemented
+                # "use_meta_prompting": self.prompt.use_meta_prompting,
+                # "meta_prompt_weight": self.prompt.meta_prompt_weight,
             },
             "database": {
                 "db_path": self.database.db_path,
@@ -200,19 +208,25 @@ def to_dict(self) -> Dict[str, Any]:
                 "elite_selection_ratio": self.database.elite_selection_ratio,
                 "exploration_ratio": self.database.exploration_ratio,
                 "exploitation_ratio": self.database.exploitation_ratio,
-                "diversity_metric": self.database.diversity_metric,
+                # Note: diversity_metric fixed to "edit_distance"
+                # "diversity_metric": self.database.diversity_metric,
                 "feature_dimensions": self.database.feature_dimensions,
                 "feature_bins": self.database.feature_bins,
+                "migration_interval": self.database.migration_interval,
+                "migration_rate": self.database.migration_rate,
+                "random_seed": self.database.random_seed,
             },
             "evaluator": {
                 "timeout": self.evaluator.timeout,
                 "max_retries": self.evaluator.max_retries,
-                "memory_limit_mb": self.evaluator.memory_limit_mb,
-                "cpu_limit": self.evaluator.cpu_limit,
+                # Note: resource limits not implemented
+                # "memory_limit_mb": self.evaluator.memory_limit_mb,
+                # "cpu_limit": self.evaluator.cpu_limit,
                 "cascade_evaluation": self.evaluator.cascade_evaluation,
                 "cascade_thresholds": self.evaluator.cascade_thresholds,
                 "parallel_evaluations": self.evaluator.parallel_evaluations,
-                "distributed": self.evaluator.distributed,
+                # Note: distributed evaluation not implemented
+                # "distributed": self.evaluator.distributed,
                 "use_llm_feedback": self.evaluator.use_llm_feedback,
                 "llm_feedback_weight": self.evaluator.llm_feedback_weight,
             },