Merge pull request #72 from Techtonique/thierrymoudiki-patch-1

Create elasticNet2Regressor.py

Author: thierrymoudiki

examples/custombackprop.py

@@ -0,0 +1,39 @@
+import nnetsauce as ns 
+import numpy as np 
+from sklearn.metrics import root_mean_squared_error
+from sklearn.model_selection import train_test_split
+from sklearn.datasets import load_diabetes, fetch_california_housing
+from sklearn.linear_model import Ridge
+from time import time 
+
+load_datasets = [load_diabetes(), fetch_california_housing()]
+
+datasets_names = ["diabetes", "housing"]
+
+for i, data in enumerate(load_datasets):
+    
+    X = data.data
+    y= data.target
+
+    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = .2, random_state = 13)
+
+    regr = ns.CustomBackPropRegressor(base_model=Ridge(), 
+                                      type_grad="finitediff")
+
+    start = time()
+    regr.fit(X_train, y_train)
+    preds = regr.predict(X_test)
+    print("Elapsed: ", time()-start)
+
+    print(f"RMSE for {datasets_names[i]} : {root_mean_squared_error(preds, y_test)}")
+
+    regr = ns.CustomBackPropRegressor(base_model=Ridge(), 
+                                      type_grad="finitediff", 
+                                      type_loss="quantile")
+
+    start = time()
+    regr.fit(X_train, y_train)
+    preds = regr.predict(X_test)
+    print("Elapsed: ", time()-start)
+
+    print(f"RMSE for {datasets_names[i]} : {root_mean_squared_error(preds, y_test)}")

examples/elasticnet2regressor.py

@@ -0,0 +1,35 @@
+import nnetsauce as ns 
+import numpy as np 
+from sklearn.metrics import root_mean_squared_error
+from sklearn.model_selection import train_test_split
+from sklearn.datasets import load_diabetes, fetch_california_housing
+from time import time 
+
+load_datasets = [load_diabetes(), fetch_california_housing()]
+
+datasets_names = ["diabetes", "housing"]
+
+for i, data in enumerate(load_datasets):
+    
+    X = data.data
+    y= data.target
+
+    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = .2, random_state = 13)
+
+    regr = ns.ElasticNet2Regressor(solver="lbfgs")
+
+    start = time()
+    regr.fit(X_train, y_train)
+    preds = regr.predict(X_test)
+    print("Elapsed: ", time()-start)
+
+    print(f"RMSE for {datasets_names[i]} : {root_mean_squared_error(preds, y_test)}")
+
+    regr = ns.ElasticNet2Regressor(solver="cd")
+
+    start = time()
+    regr.fit(X_train, y_train)
+    preds = regr.predict(X_test)
+    print("Elapsed: ", time()-start)
+
+    print(f"RMSE for {datasets_names[i]} : {root_mean_squared_error(preds, y_test)}")

nnetsauce/__init__.py

@@ -3,6 +3,7 @@
 from .boosting.adaBoostClassifier import AdaBoostClassifier
 from .custom.customClassifier import CustomClassifier
 from .custom.customRegressor import CustomRegressor
+from .custom.customBackpropRegressor import CustomBackPropRegressor
 from .datasets import Downloader
 from .deep.deepClassifier import DeepClassifier
 from .deep.deepRegressor import DeepRegressor
@@ -32,6 +33,7 @@
 from .ridge2.ridge2Classifier import Ridge2Classifier
 from .ridge2.ridge2Regressor import Ridge2Regressor
 from .ridge2.ridge2MultitaskClassifier import Ridge2MultitaskClassifier
+from .ridge2.elasticNet2Regressor import ElasticNet2Regressor
 from .rvfl.bayesianrvflRegressor import BayesianRVFLRegressor
 from .rvfl.bayesianrvfl2Regressor import BayesianRVFL2Regressor
 from .sampling import SubSampler
@@ -47,10 +49,12 @@
     "ClassicalMTS",
     "CustomClassifier",
     "CustomRegressor",
+    "CustomBackPropRegressor",
     "DeepClassifier",
     "DeepRegressor",
     "DeepMTS",
     "Downloader",
+    "ElasticNet2Regressor",
     "GLMClassifier",
     "GLMRegressor",
     "KernelRidge",

nnetsauce/base/base.py

@@ -352,12 +352,6 @@ def create_layer(self, scaled_X, W=None):
                         seed=self.seed,
                     )
 
-                # self.W_ = hash_sim[self.nodes_sim](
-                #             n_dims=n_features,
-                #             n_points=self.n_hidden_features,
-                #             seed=self.seed,
-                #         )
-
                 assert (
                     scaled_X.shape[1] == self.W_.shape[0]
                 ), "check dimensions of covariates X and matrix W"

nnetsauce/custom/__init__.py

@@ -1,5 +1,7 @@
 from .custom import Custom
 from .customClassifier import CustomClassifier
 from .customRegressor import CustomRegressor
+from .customBackpropRegressor import CustomBackPropRegressor
 
-__all__ = ["Custom", "CustomClassifier", "CustomRegressor"]
+__all__ = ["Custom", "CustomClassifier", 
+           "CustomRegressor", "CustomBackPropRegressor"]