add secondary setup for extensions

.github/workflows/python-publish.yml

@@ -30,8 +30,8 @@ jobs:
       - name: Run examples 
         run: pip install .&&find examples -maxdepth 2 -name "*.py" -exec  python3 {} \;
 
-      - name: Publish to PyPI                
-        uses: pypa/gh-action-pypi-publish@release/v1
-        with:
-          password: ${{ secrets.PYPI_GLOBAL_MLSAUCE }}
-          repository-url: https://upload.pypi.org/legacy/          
+      # - name: Publish to PyPI                
+      #   uses: pypa/gh-action-pypi-publish@release/v1
+      #   with:
+      #     password: ${{ secrets.PYPI_GLOBAL_MLSAUCE }}
+      #     repository-url: https://upload.pypi.org/legacy/          

mlsauce/adaopt/_adaopt.py

@@ -8,11 +8,10 @@
 from tqdm import tqdm
 from ..utils import subsample
 from ..utils import cluster 
-
-try:
+try: 
     from . import _adaoptc as adaoptc
 except ImportError:
-    pass
+    import _adaoptc as adaoptc
 
 
 class AdaOpt(BaseEstimator, ClassifierMixin):

mlsauce/adaopt/setup2.py

@@ -0,0 +1,8 @@
+import os 
+from distutils.core import setup
+from Cython.Build import cythonize
+
+dir_path = os.path.dirname(os.path.realpath(__file__))
+
+setup(ext_modules=cythonize(os.path.join(dir_path, "_adaoptc.pyx"), 
+                            compiler_directives={'language_level' : "3"}))

mlsauce/booster/_booster_classifier.py

@@ -5,7 +5,10 @@
 from sklearn.base import BaseEstimator
 from sklearn.base import ClassifierMixin
 from sklearn.preprocessing import PolynomialFeatures
-from . import _boosterc as boosterc
+try: 
+    from . import _boosterc as boosterc
+except ImportError:
+    import _boosterc as boosterc
 from ..utils import cluster 
 
 
@@ -26,6 +29,10 @@ class LSBoostClassifier(BaseEstimator, ClassifierMixin):
         reg_lambda: float
             L2 regularization parameter for successive errors in the optimizer
             (at training time).
+        
+        alpha: float
+            compromise between L1 and L2 regularization (must be in [0, 1]), 
+            for `solver` == 'enet'.
 
         row_sample: float
             percentage of rows chosen from the training set.
@@ -53,7 +60,8 @@ class LSBoostClassifier(BaseEstimator, ClassifierMixin):
             type of backend; must be in ('cpu', 'gpu', 'tpu')
 
         solver: str
-            type of 'weak' learner; currently in ('ridge', 'lasso')
+            type of 'weak' learner; currently in ('ridge', 'lasso', 'enet'). 
+            'enet' is a combination of 'ridge' and 'lasso' called Elastic Net.
 
         activation: str
             activation function: currently 'relu', 'relu6', 'sigmoid', 'tanh'
@@ -78,6 +86,7 @@ def __init__(
         learning_rate=0.1,
         n_hidden_features=5,
         reg_lambda=0.1,
+        alpha=0.5,
         row_sample=1,
         col_sample=1,
         dropout=0,
@@ -113,7 +122,8 @@ def __init__(
         assert solver in (
             "ridge",
             "lasso",
-        ), "`solver` must be in ('ridge', 'lasso')"
+            "enet",
+        ), "`solver` must be in ('ridge', 'lasso', 'enet')"
 
         sys_platform = platform.system()
 
@@ -127,6 +137,8 @@ def __init__(
         self.learning_rate = learning_rate
         self.n_hidden_features = n_hidden_features
         self.reg_lambda = reg_lambda
+        assert (alpha >= 0 and alpha <= 1), "`alpha` must be in [0, 1]"
+        self.alpha = alpha
         self.row_sample = row_sample
         self.col_sample = col_sample
         self.dropout = dropout
@@ -194,6 +206,7 @@ def fit(self, X, y, **kwargs):
                 learning_rate=self.learning_rate,
                 n_hidden_features=self.n_hidden_features,
                 reg_lambda=self.reg_lambda,
+                alpha=self.alpha,
                 row_sample=self.row_sample,
                 col_sample=self.col_sample,
                 dropout=self.dropout,
@@ -213,6 +226,7 @@ def fit(self, X, y, **kwargs):
                 learning_rate=self.learning_rate,
                 n_hidden_features=self.n_hidden_features,
                 reg_lambda=self.reg_lambda,
+                alpha=self.alpha,
                 row_sample=self.row_sample,
                 col_sample=self.col_sample,
                 dropout=self.dropout,

mlsauce/booster/_booster_regressor.py

@@ -26,6 +26,10 @@ class LSBoostRegressor(BaseEstimator, RegressorMixin):
         reg_lambda: float
             L2 regularization parameter for successive errors in the optimizer
             (at training time).
+        
+        alpha: float
+            compromise between L1 and L2 regularization (must be in [0, 1]), 
+            for `solver` == 'enet'
 
         row_sample: float
             percentage of rows chosen from the training set.
@@ -88,6 +92,7 @@ def __init__(
         learning_rate=0.1,
         n_hidden_features=5,
         reg_lambda=0.1,
+        alpha=0.5,
         row_sample=1,
         col_sample=1,
         dropout=0,
@@ -140,6 +145,8 @@ def __init__(
         self.learning_rate = learning_rate
         self.n_hidden_features = n_hidden_features
         self.reg_lambda = reg_lambda
+        assert (alpha >= 0 and alpha <= 1), "`alpha` must be in [0, 1]"
+        self.alpha = alpha
         self.row_sample = row_sample
         self.col_sample = col_sample
         self.dropout = dropout
@@ -210,6 +217,7 @@ def fit(self, X, y, **kwargs):
                 learning_rate=self.learning_rate,
                 n_hidden_features=self.n_hidden_features,
                 reg_lambda=self.reg_lambda,
+                alpha=self.alpha,
                 row_sample=self.row_sample,
                 col_sample=self.col_sample,
                 dropout=self.dropout,
@@ -229,6 +237,7 @@ def fit(self, X, y, **kwargs):
                 learning_rate=self.learning_rate,
                 n_hidden_features=self.n_hidden_features,
                 reg_lambda=self.reg_lambda,
+                alpha=self.alpha,
                 row_sample=self.row_sample,
                 col_sample=self.col_sample,
                 dropout=self.dropout,