version 0.6.1

examples/scripts/binary_classifier.py

@@ -20,11 +20,11 @@
 
 automl = AutoML(
     #results_path="AutoML_30",
-    #algorithms=["Neural Network"],
-    total_time_limit=200,
+    algorithms=["Random Forest"],
+    total_time_limit=20,
     explain_level=0,
     # validation={"validation_type": "split"},
-    mode="Compete",
+    mode="Explain",
     # validation={"validation_type": "split"}
     validation={
         "validation_type": "kfold",

examples/scripts/multi_class_classifier.py

@@ -19,7 +19,7 @@
 
 automl = AutoML(
     # results_path="AutoML_41",
-    algorithms=["Xgboost"],
+    algorithms=["Random Forest"],
     # algorithms=["Neural Network"],
     #    "Linear",
     #    "Xgboost",

setup.py

@@ -10,7 +10,7 @@
 
 setup(
     name='mljar-supervised',
-    version='0.6.0',
+    version='0.6.1',
     description='Automates Machine Learning Pipeline with Feature Engineering and Hyper-Parameters Tuning',
     long_description=long_description,
     long_description_content_type="text/markdown",

supervised/__init__.py

@@ -1,3 +1,3 @@
-__version__ = "0.6.0"
+__version__ = "0.6.1"
 
 from supervised.automl import AutoML

supervised/automl.py

@@ -943,7 +943,9 @@ def fit(self, X_train, y_train, X_validation=None, y_validation=None):
                     self._algorithms,
                 )
 
-            self._time_ctrl.log_time("prepare_data", "prep_0", "prepare_data", time.time()-self._start_time)
+            self._time_ctrl.log_time(
+                "prepare_data", "prepare_data", "prepare_data", time.time() - self._start_time
+            )
 
             for step in steps:
                 self._fit_level = step

supervised/tuner/time_controller.py

@@ -97,12 +97,12 @@ def enough_time_for_step(self, fit_level):
 
             time_should_use = self.time_should_use(fit_level)
 
-            #print("not_so_random should use", time_should_use)
-            #print(total_time_spend)
-            #print(
+            # print("not_so_random should use", time_should_use)
+            # print(total_time_spend)
+            # print(
             #    self.step_spend("simple_algorithms")
             #    + self.step_spend("default_algorithms")
-            #)
+            # )
 
             if total_time_spend > time_should_use + self.step_spend(
                 "simple_algorithms"
@@ -164,7 +164,7 @@ def enough_time(self, model_type, step):
         time_left = self._total_time_limit - total_time_spend
         # no time left, do not train any more models, sorry ...
         if time_left < 0:
-            #print("No time left", time_left)
+            # print("No time left", time_left)
             return False
 
         # there is still time and model_type was not tested yet
@@ -175,7 +175,7 @@ def enough_time(self, model_type, step):
         # check the fit level type
         # we dont want to spend too much time on one step
         if not self.enough_time_for_step(step):
-            #print("Not enough time for step", step)
+            # print("Not enough time for step", step)
             return False
 
         # check if there is enough time for model to train
@@ -233,8 +233,8 @@ def log_time(self, model_name, model_type, fit_level, train_time):
                 "train_time": train_time,
             }
         ]
-        #print(pd.DataFrame(self._spend))
-        #print("Already spend", self.already_spend())
+        # print(pd.DataFrame(self._spend))
+        # print("Already spend", self.already_spend())
 
     def step_spend(self, step):
         return np.sum([s["train_time"] for s in self._spend if s["fit_level"] == step])