merging from private repo

.gitignore

@@ -1 +1 @@
-src/marl-qd-code
+src/pyenv-marl-qd/*

references/README.md

@@ -31,6 +31,8 @@ The core topics are Evolutionaty Algorithms and Quality Diversity on Multi-Agent
 - [Novelty Seeking Multiagent Evolutionary Reinforcement Learning](https://dl.acm.org/doi/pdf/10.1145/3583131.3590428), Ayhan Alp Aydeniz, Robert Loftin, Kagan Tumer.
 - [Quality-diversity in dissimilarity spaces](https://dl.acm.org/doi/pdf/10.1145/3583131.3590409), Steve Huntsman.
 - [Bayesian Quality Diversity Search with Interactive Illumination](https://dl.acm.org/doi/pdf/10.1145/3583131.3590486), Paul Kent, Juergen Branke.
+- [Quality-Diversity Optimization with MAP-Elites and Surrogate Models](https://dl.acm.org/doi/pdf/10.1145/3377930.3389813), Antoine Cully, Jeff Clune, Danesh Tarapore, Jean-Baptiste Mouret.
+- [Interpretable pipelines with evolutionarily optimized modules for reinforcement learning tasks with visual inputs](https://arxiv.org/pdf/2202.04943.pdf)
 
 ## Environments
 ### Farama

src/QD_MARL/agents.py

@@ -0,0 +1,72 @@
+class Agent:
+    def __init__(self, name, squad, set_, tree, manual_policy, to_optimize):
+        self._name = name
+        self._squad = squad
+        self._set = set_
+        self._tree = tree.deep_copy() if tree is not None else None
+        self._manual_policy = manual_policy
+        self._to_optimize = to_optimize
+        self._score = []
+
+    def get_name(self):
+        return self._name
+
+    def get_squad(self):
+        return self._squad
+
+    def get_set(self):
+        return self._set
+
+    def to_optimize(self):
+        return self._to_optimize
+
+    def get_tree(self):
+        return self._tree.deep_copy()
+
+    def get_output(self, observation):
+        if self._to_optimize:
+            return self._tree.get_output(observation)
+        else:
+            return self._manual_policy.get_output(observation)
+
+    def set_reward(self, reward):
+        self._tree.set_reward(reward)
+        self._score[-1] = reward
+
+    def get_score_statistics(self, params):
+        score_values = [score_dict[key] for score_dict in self._score for key in score_dict]
+        return getattr(np, f"{params['type']}")(a=score_values, **params['params'])#Can't compare dicts with >
+
+    def new_episode(self):
+        self._score.append(0)
+
+    def has_policy(self):
+        return not self._manual_policy is None
+
+    def __str__(self):
+        return f"Name: {self._name}; Squad: {self._squad}; Set: {self._set}; Optimize: {str(self._to_optimize)}"
+
+class CoachAgent(Agent):
+    def __init__(self, name, squad, set_, tree, manual_policy, to_optimize):
+        super().__init__(name, squad, set_, tree, manual_policy, to_optimize)
+
+    def get_output(self, observation):
+        return super().get_output(observation)
+
+    def set_reward(self, reward):
+        return super().set_reward(reward)
+
+    def get_score_statistics(self, params):
+        return super().get_score_statistics(params)
+
+    def new_episode(self):
+        return super().new_episode()
+
+    def has_policy(self):
+        return super().has_policy()
+
+    def __str__(self):
+        return super().__str__()
+
+    def select_team(self):
+        pass

src/QD_MARL/algorithms/__init__.py

@@ -0,0 +1,13 @@
+from .common import *
+from .continuous_optimization import *
+from .genetic_algorithm import *
+from .genetic_programming import *
+from .grammatical_evolution import *
+from .individuals import *
+from .map_elites import *
+from .map_elites_ge import *
+from .map_elites_Pyribs import *
+from .mapElitesCMA_pyRibs import *
+from .mapElitesCMA  import *
+from .mapElitesCMA_pyRibs_GE import *
+

src/QD_MARL/algorithms/common.py

@@ -0,0 +1,49 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+"""
+    src.common
+    ~~~~~~~~~~
+
+    This module contains common utilities for optimizers
+
+    :copyright: (c) 2021 by Leonardo Lucio Custode.
+    :license: MIT, see LICENSE for more details.
+"""
+
+
+class OptMetaClass(type):
+    _registry = {}
+
+    def __new__(meta, name, bases, class_dict):
+        cls = type.__new__(meta, name, bases, class_dict)
+        OptMetaClass._registry[cls.__name__] = cls
+        return cls
+
+    @staticmethod
+    def get(class_name):
+        """
+        Retrieves the class associated to the string
+
+        :class_name: The name of the class
+        :returns: A class
+        """
+        return OptMetaClass._registry[class_name]
+
+
+class ContinuousOptimizationMetaClass(type):
+    _registry = {}
+
+    def __init__(cls, clsname, bases, methods):
+        super().__init__(clsname, bases, methods)
+        ContinuousOptimizationMetaClass._registry[cls.__name__] = cls
+
+    @staticmethod
+    def get(class_name):
+        """
+        Retrieves the class associated to the string
+
+        :class_name: The name of the class
+        :returns: A class
+        """
+        return ContinuousOptimizationMetaClass._registry[class_name]
+