Create predator-prey class

All that needs to be done is add the evolution of death rates to get the MVP

Author: StormKoper

CA.py

@@ -134,3 +134,107 @@ def run(self, steps: int) -> None:
 		for _ in range(steps):
 			self.update()
 
+
+class PP(CA):
+	"""Predator-Prey cellular automaton.
+
+	States: 0 = empty, 1 = prey, 2 = predator
+
+	Expected params keys (all values in [0,1]):
+	- "prey_death": probability a prey dies each step
+	- "predator_death": probability a predator dies each step
+	- "reproduction": per-neighbor prey reproduction probability
+	- "consumption": per-neighbor predation probability
+
+	Defaults are provided for any missing keys and user-provided values
+	are preserved. Any unknown keys in `params` will raise a ValueError.
+	"""
+
+	def __init__(
+		self,
+		rows: int,
+		cols: int,
+		densities: Tuple[float, ...],
+		neighborhood: str,
+		params: Dict[str, object],
+		cell_params: Dict[str, object],
+		seed: Optional[int] = None,
+	) -> None:
+		# initialize base CA
+		super().__init__(rows, cols, densities, neighborhood, params, cell_params, seed)
+
+		# Enforce predator-prey has exactly two species
+		assert self.n_species == 2, "PP model requires exactly two species (prey=1, predator=2)"
+
+		# Allowed parameter keys and defaults
+		_allowed = {
+			"prey_death": 0.02,
+			"predator_death": 0.05,
+			"reproduction": 0.2,
+			"consumption": 0.5,
+		}
+
+		# Check for unknown user-specified keys (in the params dict provided by user)
+		user_keys = set(self.params.keys())
+		unknown = user_keys.difference(_allowed.keys())
+		if len(unknown) > 0:
+			raise ValueError(f"Unknown parameter keys for PP: {sorted(list(unknown))}")
+
+		# Fill defaults for missing keys without overriding user-specified values
+		for k, v in _allowed.items():
+			if k not in self.params:
+				self.params[k] = v
+
+		# Validate parameter ranges
+		for k in _allowed.keys():
+			val = self.params[k]
+			if not isinstance(val, (int, float)):
+				raise TypeError(f"Parameter '{k}' must be numeric")
+			if not (0.0 <= float(val) <= 1.0):
+				raise ValueError(f"Parameter '{k}' must be between 0 and 1 (got {val})")
+
+	def update(self) -> None:
+		"""One update step for predator-prey dynamics.
+
+		Uses a copy of the current grid to evaluate rules so newly changed
+		cells do not immediately influence other rules in the same step.
+		"""
+		# copy of the grid to base all decisions on
+		old = self.grid.copy()
+
+		# neighbor counts for each species (index 0 -> prey, 1 -> predator)
+		counts = self.count_neighbors()
+		prey_neighbors = counts[0]
+		pred_neighbors = counts[1] if self.n_species >= 2 else np.zeros_like(self.grid)
+
+		rows, cols = self.grid.shape
+		# Reproduction into empty cells from neighboring prey
+		empty_mask = old == 0
+		# probability that at least one neighboring prey reproduces into the cell
+		birth_param = float(self.params["reproduction"])
+		birth_prob = 1.0 - np.power(1.0 - birth_param, prey_neighbors)
+		rand = self.generator.random(size=(rows, cols))
+		birth_cells = empty_mask & (prey_neighbors > 0) & (rand < birth_prob)
+		self.grid[birth_cells] = 1
+
+		# Predation: prey replaced by predator due to neighboring predators
+		prey_mask = old == 1
+		cons_param = float(self.params["consumption"])
+		cons_prob = 1.0 - np.power(1.0 - cons_param, pred_neighbors)
+		rand = self.generator.random(size=(rows, cols))
+		predation_cells = prey_mask & (pred_neighbors > 0) & (rand < cons_prob)
+		self.grid[predation_cells] = 2
+
+		# Deaths: use the copied `old` grid so newly-occupied cells are not killed immediately
+		# Prey death
+		prey_death_p = float(self.params["prey_death"])
+		rand = self.generator.random(size=(rows, cols))
+		prey_death_cells = (old == 1) & (rand < prey_death_p)
+		self.grid[prey_death_cells] = 0
+
+		# Predator death
+		pred_death_p = float(self.params["predator_death"])
+		rand = self.generator.random(size=(rows, cols))
+		pred_death_cells = (old == 2) & (rand < pred_death_p)
+		self.grid[pred_death_cells] = 0
+

prompts.md

@@ -157,4 +157,115 @@ class CA:
 Removed a few unnecessary lines.
 
 ## Add update logic
-Now the basic predator-prey update rules need to be defined. This is done
+Now the basic predator-prey update rules need to be defined. This is done by creating a predator-prey class inheriting the CA class and overriding the update function.
+
+### prompt
+Create a predator-prey class named PP which inherits the CA class. Create default values for the following parameters: prey_death, predator_death, reproduction, consumption. Make sure not to override the user-set values for these parameters. Also raise an error if there are any other user-defined parameters and check whether the set parameters are valid (between 0 and 1).
+
+Next, override the update function with the following functionality. First, create a copy of the current grid to be used for all the rules. Then, use the inherited count_neighbors function. For each empty cell with at least one neighboring prey, populate it with a prey (state 1) with a random chance. Use generator.random() < 1 - (1 - birth)**prey_neighbors to simulate the possibility for each neighboring prey to reproduce into the cell. For each prey cell in the copied grid with at least one neighboring predator, replace it with a predator with a random chance. use the same logic for this. Afterwards, handle the prey and predator deaths stochastically based on the relevant parameter. Use the copied grid for this to ensure newly occupied cells dont instantly become empty.
+
+### output
+class PP(CA):
+	"""Predator-Prey cellular automaton.
+
+	States: 0 = empty, 1 = prey, 2 = predator
+
+	Expected params keys (all values in [0,1]):
+	- "prey_death": probability a prey dies each step
+	- "predator_death": probability a predator dies each step
+	- "reproduction": per-neighbor prey reproduction probability
+	- "consumption": per-neighbor predation probability
+
+	Defaults are provided for any missing keys and user-provided values
+	are preserved. Any unknown keys in `params` will raise a ValueError.
+	"""
+
+	def __init__(
+		self,
+		rows: int,
+		cols: int,
+		densities: Tuple[float, ...],
+		neighborhood: str,
+		params: Dict[str, object],
+		cell_params: Dict[str, object],
+		seed: Optional[int] = None,
+	) -> None:
+		# initialize base CA
+		super().__init__(rows, cols, densities, neighborhood, params, cell_params, seed)
+
+		# Enforce predator-prey has exactly two species
+		assert self.n_species == 2, "PP model requires exactly two species (prey=1, predator=2)"
+
+		# Allowed parameter keys and defaults
+		_allowed = {
+			"prey_death": 0.02,
+			"predator_death": 0.05,
+			"reproduction": 0.2,
+			"consumption": 0.5,
+		}
+
+		# Check for unknown user-specified keys (in the params dict provided by user)
+		user_keys = set(self.params.keys())
+		unknown = user_keys.difference(_allowed.keys())
+		if len(unknown) > 0:
+			raise ValueError(f"Unknown parameter keys for PP: {sorted(list(unknown))}")
+
+		# Fill defaults for missing keys without overriding user-specified values
+		for k, v in _allowed.items():
+			if k not in self.params:
+				self.params[k] = v
+
+		# Validate parameter ranges
+		for k in _allowed.keys():
+			val = self.params[k]
+			if not isinstance(val, (int, float)):
+				raise TypeError(f"Parameter '{k}' must be numeric")
+			if not (0.0 <= float(val) <= 1.0):
+				raise ValueError(f"Parameter '{k}' must be between 0 and 1 (got {val})")
+
+	def update(self) -> None:
+		"""One update step for predator-prey dynamics.
+
+		Uses a copy of the current grid to evaluate rules so newly changed
+		cells do not immediately influence other rules in the same step.
+		"""
+		# copy of the grid to base all decisions on
+		old = self.grid.copy()
+
+		# neighbor counts for each species (index 0 -> prey, 1 -> predator)
+		counts = self.count_neighbors()
+		prey_neighbors = counts[0]
+		pred_neighbors = counts[1] if self.n_species >= 2 else np.zeros_like(self.grid)
+
+		rows, cols = self.grid.shape
+		# Reproduction into empty cells from neighboring prey
+		empty_mask = old == 0
+		# probability that at least one neighboring prey reproduces into the cell
+		birth_param = float(self.params["reproduction"])
+		birth_prob = 1.0 - np.power(1.0 - birth_param, prey_neighbors)
+		rand = self.generator.random(size=(rows, cols))
+		birth_cells = empty_mask & (prey_neighbors > 0) & (rand < birth_prob)
+		self.grid[birth_cells] = 1
+
+		# Predation: prey replaced by predator due to neighboring predators
+		prey_mask = old == 1
+		cons_param = float(self.params["consumption"])
+		cons_prob = 1.0 - np.power(1.0 - cons_param, pred_neighbors)
+		rand = self.generator.random(size=(rows, cols))
+		predation_cells = prey_mask & (pred_neighbors > 0) & (rand < cons_prob)
+		self.grid[predation_cells] = 2
+
+		# Deaths: use the copied `old` grid so newly-occupied cells are not killed immediately
+		# Prey death
+		prey_death_p = float(self.params["prey_death"])
+		rand = self.generator.random(size=(rows, cols))
+		prey_death_cells = (old == 1) & (rand < prey_death_p)
+		self.grid[prey_death_cells] = 0
+
+		# Predator death
+		pred_death_p = float(self.params["predator_death"])
+		rand = self.generator.random(size=(rows, cols))
+		pred_death_cells = (old == 2) & (rand < pred_death_p)
+		self.grid[pred_death_cells] = 0
+
+### changes

test.ipynb

@@ -0,0 +1,48 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "id": "fcec2b3a",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "766\n",
+      "765\n"
+     ]
+    }
+   ],
+   "source": [
+    "import numpy as np\n",
+    "x = np.random.rand(1000, 5)\n",
+    "print(sum(x.min(axis=1) < 0.25))\n",
+    "y = np.random.rand(1000)\n",
+    "print(sum(y < 1 - (1 - 0.25)**5))"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "base",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.13.9"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}