bump -> v0.14.0

Author: thierrymoudiki

CHANGES.md

@@ -1,3 +1,7 @@
+# version 0.14.0
+
+- add polynomial (interactions) features to `LSBoostRegressor` and `LSBoostClassifier`
+
 # version 0.13.1
 
 - add clustering to `LSBoostRegressor`, `LSBoostClassifier`, and `AdaOpt`

examples/lsboost_classifier.py

@@ -246,6 +246,20 @@
 start = time()
 print(obj.score(X_test, y_test))
 print(time()-start)
+print(f"loss: {np.log(obj.obj['loss'])}")
+
+obj = ms.LSBoostClassifier(solver="lasso", 
+                           n_clusters=3, 
+                           clustering_method="gmm",
+                           degree=2)
+print(obj.get_params())
+start = time()
+obj.fit(X_train, y_train)
+print(time()-start)
+start = time()
+print(obj.score(X_test, y_test))
+print(time()-start)
+print(f"loss: {np.log(obj.obj['loss'])}")
 
 # MORE DATA NEEDED # MORE DATA NEEDED # MORE DATA NEEDED
 # obj = ms.LSBoostClassifier(backend="gpu", solver="lasso")

examples/lsboost_regressor_pi.py

@@ -121,6 +121,21 @@
 print(f"splitconformal kde coverage 2: {np.mean((preds.upper >= y_test)*(preds.lower <= y_test))}")
 
 
+obj = ms.LSBoostRegressor(col_sample=0.9, row_sample=0.9,
+                          replications=50, 
+                          type_pi="kde",
+                          degree=2)
+print(obj.get_params())
+start = time()
+obj.fit(X_train, y_train)
+print(f"Elapsed: {time()-start}")
+start = time()
+preds = obj.predict(X_test, return_pi=True, 
+                    method="splitconformal")
+print(time()-start)
+print(f"splitconformal kde coverage 2: {np.mean((preds.upper >= y_test)*(preds.lower <= y_test))}")
+
+
 
 # lasso
 

mlsauce.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: mlsauce
-Version: 0.13.0
+Version: 0.13.1
 Summary: Miscellaneous Statistical/Machine Learning tools
 Maintainer: T. Moudiki
 Maintainer-email: thierry.moudiki@gmail.com

mlsauce/booster/_booster_classifier.py

@@ -7,6 +7,7 @@
 import numpy as np
 import platform
 import warnings
+from sklearn.preprocessing import PolynomialFeatures
 from sklearn.base import BaseEstimator
 from sklearn.base import ClassifierMixin
 
@@ -70,6 +71,9 @@ class LSBoostClassifier(BaseEstimator, ClassifierMixin):
 
         cluster_scaling: str
             scaling method for clustering: currently 'standard', 'robust', 'minmax'
+                
+        degree: int
+            degree of features interactions to include in the model
 
     """
 
@@ -92,6 +96,7 @@ def __init__(
         n_clusters=0,
         clustering_method="kmeans",
         cluster_scaling="standard",
+        degree=0
     ):
         if n_clusters > 0:
             assert clustering_method in (
@@ -142,6 +147,8 @@ def __init__(
         self.clustering_method = clustering_method
         self.cluster_scaling = cluster_scaling
         self.scaler_, self.label_encoder_, self.clusterer_ = None, None, None
+        self.degree = degree
+        self.poly_ = None 
 
     def fit(self, X, y, **kwargs):
         """Fit Booster (classifier) to training data (X, y)
@@ -162,6 +169,10 @@ def fit(self, X, y, **kwargs):
             self: object.
         """
 
+        if self.degree > 1:
+            self.poly_ = PolynomialFeatures(degree=self.degree, interaction_only=True)
+            X = self.poly_.fit_transform(X.copy())[:,1:]
+
         if self.n_clusters > 0:
             clustered_X, self.scaler_, self.label_encoder_, self.clusterer_ = (
                 cluster(
@@ -232,6 +243,9 @@ def predict_proba(self, X, **kwargs):
 
             probability estimates for test data: {array-like}
         """
+        if self.degree > 0:            
+            X = self.poly_.transform(X.copy())[:,1:]
+
         if self.n_clusters > 0:
             X = np.column_stack(
                 (

mlsauce/booster/_booster_regressor.py

@@ -9,6 +9,7 @@
 import warnings
 from sklearn.base import BaseEstimator
 from sklearn.base import RegressorMixin
+from sklearn.preprocessing import PolynomialFeatures
 from . import _boosterc as boosterc
 from ..predictioninterval import PredictionInterval
 from ..utils import cluster
@@ -81,6 +82,9 @@ class LSBoostRegressor(BaseEstimator, RegressorMixin):
 
         cluster_scaling: str
             scaling method for clustering: currently 'standard', 'robust', 'minmax'
+        
+        degree: int
+            degree of features interactions to include in the model
 
     """
 
@@ -106,6 +110,7 @@ def __init__(
         n_clusters=0,
         clustering_method="kmeans",
         cluster_scaling="standard",
+        degree=0
     ):
         if n_clusters > 0:
             assert clustering_method in (
@@ -158,7 +163,9 @@ def __init__(
         self.n_clusters = n_clusters
         self.clustering_method = clustering_method
         self.cluster_scaling = cluster_scaling
-        self.scaler_, self.label_encoder_, self.clusterer_ = None, None, None
+        self.scaler_, self.label_encoder_, self.clusterer_ = None, None, None        
+        self.degree = degree
+        self.poly_ = None
 
     def fit(self, X, y, **kwargs):
         """Fit Booster (regressor) to training data (X, y)
@@ -179,6 +186,10 @@ def fit(self, X, y, **kwargs):
             self: object.
         """
 
+        if self.degree > 1:
+            self.poly_ = PolynomialFeatures(degree=self.degree, interaction_only=True)
+            X = self.poly_.fit_transform(X.copy())[:,1:]
+
         if self.n_clusters > 0:
             clustered_X, self.scaler_, self.label_encoder_, self.clusterer_ = (
                 cluster(
@@ -242,6 +253,8 @@ def predict(self, X, level=95, method=None, **kwargs):
 
             probability estimates for test data: {array-like}
         """
+        if self.degree > 0:                        
+            X = self.poly_.transform(X.copy())[:,1:]
 
         if self.n_clusters > 0:
             X = np.column_stack(

setup.py

@@ -38,7 +38,7 @@
 MAINTAINER_EMAIL = 'thierry.moudiki@gmail.com'
 LICENSE = 'BSD3 Clause Clear'
 
-__version__ = '0.13.1'
+__version__ = '0.14.0'
 
 VERSION = __version__