remove xgboost

docs/history.rst

@@ -71,4 +71,8 @@ History
 
 0.5.1 (2017-08-25)
 ------------------
-* improvement in verbose mode for reader (display target quantiles for regression)
+* improvement in verbose mode for reader (display target quantiles for regression)
+
+0.6.0 (2019-04-26)
+------------------
+* remove xgboost installation

python-package/README.rst

@@ -7,7 +7,7 @@ Compatibilities
 ---------------
 
 * *It is compatible with:* **Python 2.7 - 3.6**. & **64-bit version only** (32-bit python is not supported)
-* *Operating system:* **Linux**. (MacOS & Windows very soon...)
+* *Operating system:* **Linux**. (EXPERIMENTAL for MacOS & Windows)
 
 
 Preparation 
@@ -26,16 +26,18 @@ First, make sure you have `setuptools <https://pypi.python.org/pypi/setuptools>`
 .. code-block:: console
 
     $ pip install cmake
-    
-    
+
+
+If you get any errors during preparation, please refer to `LightGBM's installation guide <https://github.com/Microsoft/LightGBM/tree/master/python-package#lightgbm-python-package>`__
+
 
 Installation
 ------------
 
 Install from pip 
 ~~~~~~~~~~~~~~~~
 
-MLBox is now available on **PyPI**, so you only need to run the following command:
+MLBox is available on **PyPI**, so you only need to run the following command:
 
 .. code-block:: console
 

python-package/mlbox/__init__.py

@@ -2,7 +2,7 @@
 
 __author__ = """Axel ARONIO DE ROMBLAY"""
 __email__ = 'axelderomblay@gmail.com'
-__version__ = '0.5.3'
+__version__ = '0.6.0'
 
 from .preprocessing import *
 from .encoding import *

python-package/mlbox/model/supervised/classification/classifier.py

@@ -9,20 +9,9 @@
 import pandas as pd
 from sklearn.ensemble import (AdaBoostClassifier, BaggingClassifier,
                               ExtraTreesClassifier, RandomForestClassifier)
-from xgboost import XGBClassifier
 from sklearn.linear_model import LogisticRegression
 from sklearn.tree import DecisionTreeClassifier
-
-global lgbm_installed
-
-try:
-    from lightgbm import LGBMClassifier
-    lgbm_installed = True
-except Exception:
-    warnings.warn(
-        "Package lightgbm is not installed. Model LightGBM will be replaced by"
-        "XGBoost")
-    lgbm_installed = False
+from lightgbm import LGBMClassifier
 
 
 class Classifier():
@@ -31,10 +20,9 @@ class Classifier():
 
     Parameters
     ----------
-    strategy : str, default = "LightGBM" if installed else "XGBoost"
+    strategy : str, default = "LightGBM"
         The choice for the classifier.
-        Available strategies = "LightGBM" (if installed), "XGBoost",
-        "RandomForest", "ExtraTrees", "Tree", "Bagging", "AdaBoost" or "Linear"
+        Available strategies = "LightGBM", "RandomForest", "ExtraTrees", "Tree", "Bagging", "AdaBoost" or "Linear"
 
     **params : default = None
         Parameters of the corresponding classifier.
@@ -46,10 +34,7 @@ def __init__(self, **params):
         if ("strategy" in params):
             self.__strategy = params["strategy"]
         else:
-            if (lgbm_installed):
-                self.__strategy = "LightGBM"
-            else:
-                self.__strategy = "XGBoost"
+            self.__strategy = "LightGBM"
 
         self.__classif_params = {}
 
@@ -111,31 +96,10 @@ def __set_classifier(self, strategy):
                 n_estimators=400, max_depth=10, max_features='sqrt',
                 bootstrap=True, n_jobs=-1, random_state=0)
 
-        elif(strategy == 'XGBoost'):
-            self.__classifier = XGBClassifier(n_estimators=500, max_depth=6,
-                                              learning_rate=0.05,
-                                              colsample_bytree=0.8,
-                                              colsample_bylevel=1.,
-                                              subsample=0.9,
-                                              nthread=-1, seed=0)
-
         elif(strategy == "LightGBM"):
-            if(lgbm_installed):
-                self.__classifier = LGBMClassifier(
-                    n_estimators=500, learning_rate=0.05,
-                    colsample_bytree=0.8, subsample=0.9, nthread=-1, seed=0)
-            else:
-                warnings.warn(
-                    "Package lightgbm is not installed. Model LightGBM will be"
-                    "replaced by XGBoost")
-                self.__strategy = "XGBoost"
-                self.__classifier = XGBClassifier(n_estimators=500,
-                                                  max_depth=6,
-                                                  learning_rate=0.05,
-                                                  colsample_bytree=0.8,
-                                                  colsample_bylevel=1.,
-                                                  subsample=0.9, nthread=-1,
-                                                  seed=0)
+            self.__classifier = LGBMClassifier(
+                n_estimators=500, learning_rate=0.05,
+                colsample_bytree=0.8, subsample=0.9, nthread=-1, seed=0)
 
         elif(strategy == 'ExtraTrees'):
             self.__classifier = ExtraTreesClassifier(
@@ -170,8 +134,8 @@ def __set_classifier(self, strategy):
 
         else:
             raise ValueError(
-                "Strategy invalid. Please choose between 'LightGBM' "
-                "(if installed), 'XGBoost', 'RandomForest', 'ExtraTrees', "
+                "Strategy invalid. Please choose between 'LightGBM'"
+                ", 'RandomForest', 'ExtraTrees', "
                 "'Tree', 'Bagging', 'AdaBoost' or 'Linear'")
 
 
@@ -231,8 +195,7 @@ def feature_importances(self):
                 for i, col in enumerate(self.__col):
                     importance[col] = f[i]
 
-            elif (self.get_params()["strategy"] in ["LightGBM", "XGBoost",
-                                                    "RandomForest",
+            elif (self.get_params()["strategy"] in ["LightGBM", "RandomForest",
                                                     "ExtraTrees", "Tree"]):
 
                 importance = {}
@@ -248,7 +211,7 @@ def feature_importances(self):
                 norm = self.get_estimator().estimator_weights_.sum()
 
                 try:
-                    # XGB, RF, ET, Tree and AdaBoost
+                    # LGB, RF, ET, Tree and AdaBoost
                     f = sum(weight * est.feature_importances_
                         for weight, est in zip(self.get_estimator().estimator_weights_, self.get_estimator().estimators_)) / norm  # noqa
 
@@ -270,7 +233,7 @@ def feature_importances(self):
                     d = {}
 
                     try:
-                        # XGB, RF, ET, Tree and AdaBoost
+                        # LGB, RF, ET, Tree and AdaBoost
                         f = b.feature_importances_
                     except:  # noqa
                         # Linear

python-package/mlbox/model/supervised/classification/stacking_classifier.py

@@ -22,7 +22,7 @@ class StackingClassifier():
 
     Parameters
     ----------
-    base_estimators : list, default = [Classifier(strategy="XGBoost"), Classifier(strategy="RandomForest"),Classifier(strategy="ExtraTrees")]
+    base_estimators : list, default = [Classifier(strategy="LightGBM"), Classifier(strategy="RandomForest"),Classifier(strategy="ExtraTrees")]
         List of estimators to fit in the first level using a cross validation.
 
     level_estimator : object, default = LogisticRegression()
@@ -46,7 +46,7 @@ class StackingClassifier():
     """
 
     def __init__(self,
-                 base_estimators=[Classifier(strategy="XGBoost"),
+                 base_estimators=[Classifier(strategy="LightGBM"),
                                   Classifier(strategy="RandomForest"),
                                   Classifier(strategy="ExtraTrees")],
                  level_estimator=LogisticRegression(n_jobs=-1),

python-package/mlbox/model/supervised/regression/regressor.py

@@ -9,20 +9,9 @@
 import pandas as pd
 from sklearn.ensemble import (AdaBoostRegressor, BaggingRegressor,
                               ExtraTreesRegressor, RandomForestRegressor)
-from xgboost import XGBRegressor
 from sklearn.linear_model import Ridge
 from sklearn.tree import DecisionTreeRegressor
-
-global lgbm_installed
-
-try:
-    from lightgbm import LGBMRegressor
-    lgbm_installed = True
-except Exception:
-    warnings.warn(
-        "Package lightgbm is not installed. Model LightGBM will be replaced"
-        "by XGBoost")
-    lgbm_installed = False
+from lightgbm import LGBMRegressor
 
 
 class Regressor():
@@ -32,10 +21,9 @@ class Regressor():
 
     Parameters
     ----------
-    strategy : str, default = "LightGBM" if installed else "XGBoost"
+    strategy : str, default = "LightGBM"
         The choice for the regressor.
-        Available strategies = "LightGBM" (if installed), "XGBoost",
-        "RandomForest", "ExtraTrees", "Tree", "Bagging", "AdaBoost" or "Linear"
+        Available strategies = "LightGBM", "RandomForest", "ExtraTrees", "Tree", "Bagging", "AdaBoost" or "Linear"
 
     **params : default = None
         Parameters of the corresponding regressor.
@@ -47,10 +35,7 @@ def __init__(self, **params):
         if ("strategy" in params):
             self.__strategy = params["strategy"]
         else:
-            if (lgbm_installed):
-                self.__strategy = "LightGBM"
-            else:
-                self.__strategy = "XGBoost"
+            self.__strategy = "LightGBM"
 
         self.__regress_params = {}
 
@@ -112,26 +97,10 @@ def __set_regressor(self, strategy):
                 n_estimators=400, max_depth=10, max_features='sqrt',
                 bootstrap=True, n_jobs=-1, random_state=0)
 
-        elif(strategy == 'XGBoost'):
-            self.__regressor = XGBRegressor(
-                n_estimators=500, max_depth=6, learning_rate=0.05,
-                colsample_bytree=0.8, colsample_bylevel=1., subsample=0.9,
-                nthread=-1, seed=0)
-
         elif(strategy == "LightGBM"):
-            if(lgbm_installed):
-                self.__regressor = LGBMRegressor(
-                    n_estimators=500, learning_rate=0.05,
-                    colsample_bytree=0.8, subsample=0.9, nthread=-1, seed=0)
-            else:
-                warnings.warn(
-                    "Package lightgbm is not installed. Model LightGBM will be"
-                    "replaced by XGBoost")
-                self.__strategy = "XGBoost"
-                self.__regressor = XGBRegressor(
-                    n_estimators=500, max_depth=6, learning_rate=0.05,
-                    colsample_bytree=0.8, colsample_bylevel=1.,
-                    subsample=0.9, nthread=-1, seed=0)
+            self.__regressor = LGBMRegressor(
+                n_estimators=500, learning_rate=0.05,
+                colsample_bytree=0.8, subsample=0.9, nthread=-1, seed=0)
 
         elif(strategy == 'ExtraTrees'):
             self.__regressor = ExtraTreesRegressor(
@@ -163,8 +132,8 @@ def __set_regressor(self, strategy):
 
         else:
             raise ValueError(
-                "Strategy invalid. Please choose between 'LightGBM' "
-                "(if installed), 'XGBoost', 'RandomForest', 'ExtraTrees', "
+                "Strategy invalid. Please choose between 'LightGBM'"
+                ", 'RandomForest', 'ExtraTrees', "
                 "'Tree', 'Bagging', 'AdaBoost' or 'Linear'")
 
 
@@ -224,8 +193,7 @@ def feature_importances(self):
                 for i, col in enumerate(self.__col):
                     importance[col] = f[i]
 
-            elif (self.get_params()["strategy"] in ["LightGBM", "XGBoost",
-                                                    "RandomForest",
+            elif (self.get_params()["strategy"] in ["LightGBM", "RandomForest",
                                                     "ExtraTrees", "Tree"]):
 
                 importance = {}
@@ -241,7 +209,7 @@ def feature_importances(self):
                 norm = self.get_estimator().estimator_weights_.sum()
 
                 try:
-                    # XGB, RF, ET, Tree and AdaBoost
+                    # LGB, RF, ET, Tree and AdaBoost
                     # TODO: Refactor this part
                     f = sum(weight * est.feature_importances_ for weight, est in zip(self.get_estimator().estimator_weights_, self.get_estimator().estimators_)) / norm  # noqa
 
@@ -261,7 +229,7 @@ def feature_importances(self):
                     d = {}
 
                     try:
-                        # XGB, RF, ET, Tree and AdaBoost
+                        # LGB, RF, ET, Tree and AdaBoost
                         f = b.feature_importances_
                     except Exception:
                         f = np.abs(b.coef_)  # Linear

python-package/mlbox/model/supervised/regression/stacking_regressor.py

@@ -23,7 +23,7 @@ class StackingRegressor():
 
     Parameters
     ----------
-    base_estimators : list, default = [Regressor(strategy="XGBoost"), Regressor(strategy="RandomForest"), Regressor(strategy="ExtraTrees")]
+    base_estimators : list, default = [Regressor(strategy="LightGBM"), Regressor(strategy="RandomForest"), Regressor(strategy="ExtraTrees")]
         List of estimators to fit in the first level using a cross validation.
 
     level_estimator : object, default = LinearRegression()
@@ -43,7 +43,7 @@ class StackingRegressor():
         Verbose mode.
     """
 
-    def __init__(self, base_estimators=[Regressor(strategy="XGBoost"),
+    def __init__(self, base_estimators=[Regressor(strategy="LightGBM"),
                                         Regressor(strategy="RandomForest"),
                                         Regressor(strategy="ExtraTrees")],
                  level_estimator=LinearRegression(), n_folds=5,

python-package/requirements.txt

@@ -1,11 +1,9 @@
-numpy==1.13.0
-matplotlib==2.0.2
+numpy>=1.13.0
+matplotlib>=2.0.2
 hyperopt==0.1
-Keras==2.0.4
-pandas==0.20.3
-joblib==0.11
+Keras==2.1.2
+pandas==0.21.0
 scikit-learn==0.19.0
-Theano==0.9.0
-xgboost==0.6a2
-lightgbm==2.0.2
+Theano==1.0.1
+lightgbm==2.0.11
 networkx==1.11

python-package/setup.py

@@ -4,20 +4,11 @@
 import pip
 from setuptools import setup
 from setuptools.command.install import install
+from mlbox.__init__ import __version__
+
+with open('requirements.txt', 'rt') as fh:
+    requirements = fh.read().splitlines()
 
-requirements = [
-    "numpy==1.13.0",
-    "matplotlib==2.0.2",
-    "hyperopt==0.1",
-    "Keras==2.0.4",
-    "pandas==0.20.3",
-    "joblib==0.11",
-    "scikit-learn==0.19.0",
-    "Theano==0.9.0",
-    "xgboost==0.6a2",
-    "lightgbm==2.0.2", 
-    "networkx==1.11"
-]
 
 class OverrideInstallCommand(install):
     def run(self):
@@ -45,7 +36,7 @@ def run(self):
 
 setup(
     name='mlbox',
-    version='0.5.3',
+    version=__version__,
     description="A powerful Automated Machine Learning python library.",
     long_description=readme,
     author="Axel ARONIO DE ROMBLAY",