Removing feature extractor.

Author: alucantonio

examples/ex1.yaml

@@ -68,7 +68,6 @@ def __init__(self,
                  error_metric: Callable | None = None,
                  predict_func: Callable | None = None,
                  common_data: Dict | None = None,
-                 feature_extractors: List = [],
                  toolbox: base.Toolbox = None,
                  individualCreator: gp.PrimitiveTree = None,
                  NINDIVIDUALS: int = 10,
@@ -183,7 +182,7 @@ def __init__(self,
                 parallel = Parallel(n_jobs=num_jobs, return_as="generator")
         else:
             parallel = None
-        self.__register_map(feature_extractors, parallel)
+        self.__register_map(parallel)
 
         self.plot_initialized = False
         self.fig_id = 0
@@ -423,17 +422,14 @@ def __register_predict_func(self):
         self.toolbox.register("evaluate_test_sols",
                               self.predict_func, **args_predict_func)
 
-    def __register_map(self, individ_feature_extractors: List[Callable] | None = None,
-                       parallel=None):
+    def __register_map(self, parallel=None):
         def mapper(f, individuals, toolbox):
             if self.parallel_lib == "ray":
                 fitnesses = []*len(individuals)
                 toolbox_ref = ray.put(toolbox)
                 for i in range(0, len(individuals), self.batch_size):
                     individuals_batch = individuals[i:i+self.batch_size]
-                    fitnesses.append(
-                        f(individuals_batch, individ_feature_extractors,
-                          toolbox_ref))
+                    fitnesses.append(f(individuals_batch, toolbox_ref))
                 fitnesses = list(chain(*ray.get(fitnesses)))
             # elif self.parallel_lib == "joblib":
             #     fitnesses = list(parallel((delayed(f)(*args)

examples/ex1.ipynb renamed to examples/example_poisson.ipynb

@@ -32,7 +32,7 @@ def eval_MSE_sol(individual, true_data):
 
 
 @ray.remote
-def predict(individuals_str, feature_extractors, toolbox, true_data):
+def predict(individuals_str, toolbox, true_data):
 
     callables = compile_individuals(toolbox, individuals_str)
 
@@ -45,7 +45,7 @@ def predict(individuals_str, feature_extractors, toolbox, true_data):
 
 
 @ray.remote
-def score(individuals_str, feature_extractors, toolbox, true_data):
+def score(individuals_str, toolbox, true_data):
 
     callables = compile_individuals(toolbox, individuals_str)
 
@@ -58,7 +58,7 @@ def score(individuals_str, feature_extractors, toolbox, true_data):
 
 
 @ray.remote
-def fitness(individuals_str, feature_extractors, toolbox, true_data):
+def fitness(individuals_str, toolbox, true_data):
     callables = compile_individuals(toolbox, individuals_str)
 
     fitnesses = [None]*len(individuals_str)

src/alpine/gp/gpsymbreg.py

@@ -88,8 +88,7 @@ def obj(x, y):
 
 
 @ray.remote
-def predict(individuals_str: list[str], individ_feature_extractors: list[Callable],
-            toolbox, D: Dataset, S: SimplicialComplex,
+def predict(individuals_str: list[str], toolbox, D: Dataset, S: SimplicialComplex,
             u_0: C.CochainP0, penalty: dict) -> List:
 
     callables = compile_individuals(toolbox, individuals_str)
@@ -103,8 +102,7 @@ def predict(individuals_str: list[str], individ_feature_extractors: list[Callabl
 
 
 @ray.remote
-def score(individuals_str: list[str], individ_feature_extractors: list[Callable],
-          toolbox, D: Dataset, S: SimplicialComplex,
+def score(individuals_str: list[str], toolbox, D: Dataset, S: SimplicialComplex,
           u_0: C.CochainP0, penalty: dict) -> List:
 
     callables = compile_individuals(toolbox, individuals_str)
@@ -118,12 +116,11 @@ def score(individuals_str: list[str], individ_feature_extractors: list[Callable]
 
 
 @ray.remote
-def fitness(individuals_str: list[str], individ_feature_extractors: list[Callable],
-            toolbox, D: Dataset, S: SimplicialComplex,
+def fitness(individuals_str: list[str], toolbox, D: Dataset, S: SimplicialComplex,
             u_0: C.CochainP0, penalty: dict) -> Tuple[float, ]:
 
     callables = compile_individuals(toolbox, individuals_str)
-    indlen = get_features_batch(individ_feature_extractors, individuals_str)
+    indlen = get_features_batch([len], individuals_str)
 
     fitnesses = [None]*len(individuals_str)
     for i, ind in enumerate(callables):
@@ -181,8 +178,8 @@ def test_poisson1d(set_test_dir, yamlfile):
         pset=pset, fitness=fitness.remote,
         error_metric=score.remote, predict_func=predict.remote,
         config_file_data=config_file_data, print_log=True,
-        common_data=common_params, feature_extractors=[len],
-        seed=seed_str, plot_history=False, save_best_individual=True,
+        common_data=common_params, seed=seed_str,
+        plot_history=False, save_best_individual=True,
         save_train_fit_history=True, output_path="./")
 
     train_data = Dataset("D", X_train, y_train)