phase 2 changes

codegithubka · codegithubka · commit e5fd91d03be4 · 2026-01-26T12:01:15.000+01:00
diff --git a/docs/HPC_GUIDE.md b/docs/HPC_GUIDE.md
@@ -47,7 +47,7 @@ watch -n 10 "ls -1 results_JOBID  | wc -l"
 
 # Fetching the results once the job is done
 
-scp -r kanagnostopoul@snellius.surf.nl:~/CSS_Project/results/phase1_JOBID/ ./results/
+scp -r kanagnostopoul@snellius.surf.nl:~/CSS_Project/results/phase1_18680757/ ./results/
 ```
 
 The jobscript template can be found in ```run_analysis.sh``` (default rome paritition).
diff --git a/models/config.py b/models/config.py
@@ -83,8 +83,7 @@ class Config:
     critical_prey_death: float = 0.04 # FIXME; Change after obtaining results
     
     # Prey parameter sweep (Phase 1)
-    prey_birth_range: Tuple[float, float] = (0.10, 0.35)
-    prey_death_range: Tuple[float, float] = (0.0, 0.20)
+    prey_death_range: Tuple[float, float] = (0.0, 0.2)
     n_prey_birth: int = 15   # FIXME: Decide number of grid points along prey axes
     n_prey_death: int = 20
     
@@ -105,7 +104,7 @@ class Config:
     # Evo
     with_evolution: bool = False
     evolve_sd: float = 0.10
-    evolve_min: float = 0.001
+    evolve_min: float = 0.0
     evolve_max: float = 0.10
     
     # Sensitivity: mutation strength values to test
@@ -216,7 +215,7 @@ def estimate_runtime(self, n_cores: int = 32) -> str:
     # Evolution settings
     with_evolution=True,
     evolve_sd=0.01,             # Smaller mutation rate for smoother convergence
-    evolve_min=0.001,
+    evolve_min=0.0,
     evolve_max=0.20,            # Allow full range
     
     collect_pcf=False,
diff --git a/scripts/experiments.py b/scripts/experiments.py
@@ -188,12 +188,11 @@ def run_single_simulation(
     warmup_steps = cfg.get_warmup_steps(grid_size)
     measurement_steps = cfg.get_measurement_steps(grid_size)
     
+    
     # Warmup phase
     for _ in range(warmup_steps):
         model.update()
         
-        
-    # FIXME: Remove the two lines below after phase 2.
     if with_evolution:
         model._evolution_stopped = True
         
@@ -219,7 +218,30 @@ def run_single_simulation(
         if with_evolution:
             stats = get_evolved_stats(model, "prey_death")
             evolved_means.append(stats["mean"])
-            evolved_stds.append(stats["std"])  
+            evolved_stds.append(stats["std"])
+        
+        # Cluster analysis (at end of measurement)
+        if step == measurement_steps - 1:
+            prey_survived = prey_pops[-1] > min_count
+            pred_survived = pred_pops[-1] > (min_count // 4)
+            
+            if prey_survived:
+                prey_stats = get_cluster_stats_fast(model.grid, 1)
+                cluster_sizes_prey = prey_stats['sizes'].tolist()
+                largest_fractions_prey.append(prey_stats['largest_fraction'])
+            
+            if pred_survived:
+                pred_stats = get_cluster_stats_fast(model.grid, 2)
+                cluster_sizes_pred = pred_stats['sizes'].tolist()
+                largest_fractions_pred.append(pred_stats['largest_fraction'])
+            
+            # PCF requires both
+            if compute_pcf and prey_survived and pred_survived:
+                max_dist = min(grid_size / 2, cfg.pcf_max_distance)
+                pcf_data = compute_all_pcfs_fast(model.grid, max_dist, cfg.pcf_n_bins)
+                pcf_samples['prey_prey'].append(pcf_data['prey_prey'])
+                pcf_samples['pred_pred'].append(pcf_data['pred_pred'])
+                pcf_samples['prey_pred'].append(pcf_data['prey_pred'])  
                 
     # Compile results
     result = {
@@ -375,45 +397,36 @@ def run_phase2(cfg: Config, output_dir: Path, logger: logging.Logger) -> List[Di
     """
     Phase 2: Self-organization analysis.
     
-    SOC Hypothesis: When prey_death is allowed to evolve, the population
-    should self-organize toward the critical point (~0.095-0.105 from Phase 1)
-    regardless of initial prey_death value.
+    SOC Hypothesis: Prey evolve toward critical critical point regardless of initial conditions.
     
-    Test: Start evolution from different initial prey_death values.
-    If SOC holds, all runs converge to similar final evolved_prey_death.
+    NOTE: Test is currently start evo from different intial prey_death values (?)
+    If SOC holds, then all runs converge to the same final prey_death near critical point.
+    
+    FIXME: This run script needs to be adjusted
     """
     from joblib import Parallel, delayed
     
     warmup_numba_kernels(cfg.grid_size, directed_hunting=cfg.directed_hunting)
     
-    # Vary initial prey_death across subcritical to supercritical range
-    # Based on Phase 1: critical point ~0.095-0.105
-    initial_prey_deaths = np.array([0.02, 0.04, 0.06, 0.08, 0.10, 0.12])
+    # Test at multiple prey_birth values
+    prey_births = cfg.get_prey_births()
+    
+    # Vary intial prey_death
+    initial_prey_deaths = np.linspace(cfg.prey_death_range[0], cfg.prey_death_range[1], 20)
     
     jobs = []
-    for initial_pd in initial_prey_deaths:
-        for rep in range(cfg.n_replicates):
-            params = {"initial_pd": initial_pd, "phase": 2}
-            seed = generate_unique_seed(params, rep)
-            
-            # Fixed prey_birth, varying initial prey_death, evolution ON
-            jobs.append((
-                cfg.prey_birth,       # prey_birth (fixed at 0.2)
-                initial_pd,           # prey_death (initial value, will evolve)
-                cfg.predator_birth,   # predator_birth (fixed)
-                cfg.predator_death,   # predator_death (fixed)
-                cfg.grid_size,
-                seed,
-                cfg,
-                True                  # with_evolution=True
-            ))
-            
+    for pb in prey_births:
+        for initial_pd in initial_prey_deaths:
+            for rep in range(cfg.n_replicates):
+                params = {"pb": pb, "initial_pd": initial_pd, "phase": 2}
+                seed = generate_unique_seed(params, rep)
+                jobs.append((pb, initial_pd, cfg.predator_birth, cfg.predator_death,
+                            cfg.grid_size, seed, cfg, True))
     
     logger.info(f"Phase 2: {len(jobs):,} simulations")
-    logger.info(f"  Initial prey_death values: {initial_prey_deaths.tolist()}")
-    logger.info(f"  Replicates per condition: {cfg.n_replicates}")
-    logger.info(f"  Evolution: ENABLED (sd={cfg.evolve_sd})")
-    logger.info(f"  Expected convergence: ~0.095-0.105 (critical point)")
+    logger.info(f"  prey_birth values: {len(prey_births)}")
+    logger.info(f"  initial prey_death values: {len(initial_prey_deaths)}")
+    logger.info(f"  Replicates: {cfg.n_replicates}")
     
     output_jsonl = output_dir / "phase2_results.jsonl"
     all_results = []
@@ -429,13 +442,9 @@ def run_phase2(cfg: Config, output_dir: Path, logger: logging.Logger) -> List[Di
     
     meta = {
         "phase": 2,
-        "description": "Self-organization toward criticality (SOC test)",
+        "description": "Self-organization toward criticality",
         "n_sims": len(all_results),
         "initial_prey_deaths": initial_prey_deaths.tolist(),
-        "prey_birth": cfg.prey_birth,
-        "evolve_sd": cfg.evolve_sd,
-        "measurement_steps": cfg.measurement_steps,
-        "expected_convergence": "~0.095-0.105",
         "timestamp": time.strftime("%Y-%m-%d %H:%M:%S"),
     }
     with open(output_dir / "phase2_metadata.json", "w") as f:
