Add the new feature of customized initial population

tests/test_custom_iniPop.py

@@ -0,0 +1,17 @@
+from tpot import TPOTClassifier
+from sklearn.datasets import load_digits
+from sklearn.model_selection import train_test_split
+
+digits = load_digits()
+X_train, X_test, y_train, y_test = train_test_split(digits.data, digits.target,
+                                                    train_size=0.75, test_size=0.25, random_state=42)
+
+individual_str1 = 'MultinomialNB(input_matrix, MultinomialNB__alpha=0.1, MultinomialNB__fit_prior=True)'
+individual_str2 = 'GaussianNB(DecisionTreeClassifier(input_matrix, DecisionTreeClassifier__criterion=entropy, DecisionTreeClassifier__max_depth=4, DecisionTreeClassifier__min_samples_leaf=17, DecisionTreeClassifier__min_samples_split=13))'
+individual_str3 = 'GaussianNB(SelectFwe(CombineDFs(input_matrix, ZeroCount(input_matrix))))'
+
+est = TPOTClassifier(generations=3, population_size=5, verbosity=2, random_state=42, config_dict=None,
+                     customized_initial_population=[individual_str1, individual_str2, individual_str3],
+                      )
+est.fit(X_train, y_train)
+print(est.score(X_test, y_test))

tpot/base.py

@@ -128,6 +128,7 @@ def __init__(
         verbosity=0,
         disable_update_check=False,
         log_file=None,
+        customized_initial_population=None,
     ):
         """Set up the genetic programming algorithm for pipeline optimization.
 
@@ -310,6 +311,7 @@ def __init__(
         self.disable_update_check = disable_update_check
         self.random_state = random_state
         self.log_file = log_file
+        self.customized_initial_population = customized_initial_population
 
     def _setup_template(self, template):
         self.template = template
@@ -557,6 +559,9 @@ def _setup_toolbox(self):
         self._toolbox.register(
             "population", tools.initRepeat, list, self._toolbox.individual
         )
+        self._toolbox.register(
+            "customized_population", self._initPopulation_customized, customized_initial_population = self.customized_initial_population
+        )
         self._toolbox.register("compile", self._compile_to_sklearn)
         self._toolbox.register("select", tools.selNSGA2)
         self._toolbox.register("mate", self._mate_operator)
@@ -764,7 +769,10 @@ def fit(self, features, target, sample_weight=None, groups=None):
 
         # assign population, self._pop can only be not None if warm_start is enabled
         if not self._pop:
-            self._pop = self._toolbox.population(n=self.population_size)
+            if not self.customized_initial_population:
+                self._pop = self._toolbox.population(n=self.population_size) # generate initial population by default
+            else:
+                self._pop = self._toolbox.customized_population(customized_initial_population=self.customized_initial_population) # generate initial population by custom
 
         def pareto_eq(ind1, ind2):
             """Determine whether two individuals are equal on the Pareto front.
@@ -2026,6 +2034,21 @@ def _generate(self, pset, min_, max_, condition, type_=None):
                     stack.append((depth + 1, arg))
         return expr
 
+    def _initPopulation_customized(self, customized_initial_population):
+        iniPop = [] # a list of <class 'deap.creator.Individual'> pipelines
+        for individual_str in customized_initial_population:
+            individual = creator.Individual.from_string(individual_str, self._pset) # converting individual_str to individual
+            iniPop.append(individual)
+        "check if #customized initial pipelines <= #population"
+        if len(iniPop) <= self.population_size:
+            for _ in range(self.population_size - len(iniPop)):
+                individual_rand = self._toolbox.individual()
+                iniPop.append(individual_rand)
+            print(len(customized_initial_population), "customized pipelines +", self.population_size - len(customized_initial_population), "randomized pipelines as initial population.")
+        else:
+            raise Exception("the number of customized initial pipelines > the number of population size!")
+        return iniPop
+
 
     @property
     def classes_(self):