commit-live-students · prodigy803 · Dec 17, 2018 · Dec 17, 2018 · Dec 17, 2018 · Dec 17, 2018
diff --git a/__pycache__/__init__.cpython-36.pyc b/__pycache__/__init__.cpython-36.pyc
diff --git a/q01_my_decision_regressor/__pycache__/__init__.cpython-36.pyc b/q01_my_decision_regressor/__pycache__/__init__.cpython-36.pyc
diff --git a/q01_my_decision_regressor/__pycache__/build.cpython-36.pyc b/q01_my_decision_regressor/__pycache__/build.cpython-36.pyc
diff --git a/q01_my_decision_regressor/build.py b/q01_my_decision_regressor/build.py
@@ -1,17 +1,29 @@
+# %load q01_my_decision_regressor/build.py
 # default imports
 from sklearn.model_selection import GridSearchCV
 from sklearn.tree import DecisionTreeRegressor
 from sklearn.metrics import r2_score
 from sklearn.model_selection import train_test_split
 import pandas as pd
 
-data = pd.read_csv("./data/house_pricing.csv")
+data = pd.read_csv('./data/house_pricing.csv')
 X = data.iloc[:, :-1]
 y = data.iloc[:, -1]
 X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=9)
 
-param_grid = {"max_depth": [2, 3, 5, 6, 8, 10, 15, 20, 30, 50],
-              "max_leaf_nodes": [2, 3, 4, 5, 10, 15, 20],
-              "max_features": [4, 8, 20, 25]}
+param_grid = {'max_depth': [2, 3, 5, 6, 8, 10, 15, 20, 30, 50],
+              'max_leaf_nodes': [2, 3, 4, 5, 10, 15, 20],
+              'max_features': [4, 8, 20, 25]}
 
 # Write your solution here :
+def my_decision_regressor(X_train,X_test,y_train,y_test,param_grid):
+    dt_regressor = DecisionTreeRegressor(random_state=9)
+    grid_search = GridSearchCV(dt_regressor,param_grid=param_grid,cv=5)
+
+    grid_search.fit(X_train,y_train)
+    predictions = grid_search.predict(X_test)
+
+    return r2_score(y_test,predictions),grid_search.best_params_
+
+
+
diff --git a/q01_my_decision_regressor/tests/__pycache__/__init__.cpython-36.pyc b/q01_my_decision_regressor/tests/__pycache__/__init__.cpython-36.pyc
diff --git a/q01_my_decision_regressor/tests/__pycache__/test_q01_my_decision_regressor.cpython-36.pyc b/q01_my_decision_regressor/tests/__pycache__/test_q01_my_decision_regressor.cpython-36.pyc
diff --git a/q02_decision_regressor_plot/__pycache__/__init__.cpython-36.pyc b/q02_decision_regressor_plot/__pycache__/__init__.cpython-36.pyc
diff --git a/q02_decision_regressor_plot/__pycache__/build.cpython-36.pyc b/q02_decision_regressor_plot/__pycache__/build.cpython-36.pyc
diff --git a/q02_decision_regressor_plot/build.py b/q02_decision_regressor_plot/build.py
@@ -1,3 +1,4 @@
+# %load q02_decision_regressor_plot/build.py
 # default imports
 from sklearn.model_selection import train_test_split
 from sklearn.tree import DecisionTreeRegressor
@@ -6,12 +7,35 @@
 import matplotlib.pyplot as plt
 import numpy as np
 plt.switch_backend('agg')
-
-data = pd.read_csv("./data/house_pricing.csv")
+data = pd.read_csv('./data/house_pricing.csv')
 X = data.iloc[:, :-1]
 y = data.iloc[:, -1]
 X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=9)
 
 depth_list = [2, 8, 10, 15, 20, 25, 30, 35, 45, 50, 80]
 
 # Write your solution here :
+def decision_regressor_plot(X_train,X_test,y_train,y_test,depth_list):
+
+    errors_test = []
+    errors_train = []
+    for i in range(len(depth_list)):
+        dt_Reg = DecisionTreeRegressor(max_depth=depth_list[i],random_state=9)
+        dt_Reg.fit(X_train,y_train)
+
+        preds_train = dt_Reg.predict(X_train)
+        preds_test = dt_Reg.predict(X_test)
+
+        errors_train.append(mean_squared_error(y_train,preds_train))
+        errors_test.append(mean_squared_error(y_test,preds_test))
+
+    plt.plot(depth_list,errors_train,label='Train Error')
+    plt.plot(depth_list,errors_test,label='Test Error')
+    plt.legend()
+    plt.ylabel('Mean Squared Error')
+    plt.xlabel('Max Depth')
+    plt.show()
+decision_regressor_plot(X_train,X_test,y_train,y_test,depth_list)
+
+
+
diff --git a/q02_decision_regressor_plot/tests/__pycache__/__init__.cpython-36.pyc b/q02_decision_regressor_plot/tests/__pycache__/__init__.cpython-36.pyc
diff --git a/...decision_regressor_plot/tests/__pycache__/test_q02_decision_regressor_plot.cpython-36.pyc b/...decision_regressor_plot/tests/__pycache__/test_q02_decision_regressor_plot.cpython-36.pyc
diff --git a/q03_my_decision_classifier/__pycache__/__init__.cpython-36.pyc b/q03_my_decision_classifier/__pycache__/__init__.cpython-36.pyc
diff --git a/q03_my_decision_classifier/__pycache__/build.cpython-36.pyc b/q03_my_decision_classifier/__pycache__/build.cpython-36.pyc
diff --git a/q03_my_decision_classifier/build.py b/q03_my_decision_classifier/build.py
@@ -1,3 +1,4 @@
+# %load q03_my_decision_classifier/build.py
 # default imports
 from sklearn.model_selection import RandomizedSearchCV
 from sklearn.tree import DecisionTreeClassifier
@@ -6,16 +7,25 @@
 import pandas as pd
 import numpy as np
 
-data = pd.read_csv("./data/loan_prediction.csv")
+data = pd.read_csv('./data/loan_prediction.csv')
 np.random.seed(9)
 X = data.iloc[:, :-1]
 y = data.iloc[:, -1]
 X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=9)
 
-param_grid = {"max_depth": [8, 10, 15, 20],
-              "max_leaf_nodes": [2, 5, 9, 15, 20],
-              "max_features": [1, 2, 3, 5]}
+param_grid = {'max_depth': [8, 10, 15, 20],
+              'max_leaf_nodes': [2, 5, 9, 15, 20],
+              'max_features': [1, 2, 3, 5]}
 
 
 # Write your solution here :
+def my_decision_classifier(X_train,X_test,y_train,y_test,param_grid,n_iter_search=10):
+    dt_reg = DecisionTreeClassifier(random_state=9)
+    random_cv = RandomizedSearchCV(dt_reg,param_distributions=param_grid,n_iter=n_iter_search)
+
+    random_cv.fit(X_train,y_train)
+    return accuracy_score(y_test,random_cv.predict(X_test)),random_cv.best_params_
+
+
+
 
diff --git a/q03_my_decision_classifier/tests/__pycache__/__init__.cpython-36.pyc b/q03_my_decision_classifier/tests/__pycache__/__init__.cpython-36.pyc
diff --git a/q03_my_decision_classifier/tests/__pycache__/test_q03_my_decision_classifier.cpython-36.pyc b/q03_my_decision_classifier/tests/__pycache__/test_q03_my_decision_classifier.cpython-36.pyc
diff --git a/q04_decision_classifier_plot/__pycache__/__init__.cpython-36.pyc b/q04_decision_classifier_plot/__pycache__/__init__.cpython-36.pyc
diff --git a/q04_decision_classifier_plot/__pycache__/build.cpython-36.pyc b/q04_decision_classifier_plot/__pycache__/build.cpython-36.pyc
diff --git a/q04_decision_classifier_plot/build.py b/q04_decision_classifier_plot/build.py
@@ -1,3 +1,4 @@
+# %load q04_decision_classifier_plot/build.py
 # default imports
 from sklearn.model_selection import RandomizedSearchCV
 from sklearn.tree import DecisionTreeClassifier
@@ -8,7 +9,7 @@
 import numpy as np
 plt.switch_backend('agg')
 
-data = pd.read_csv("./data/loan_prediction.csv")
+data = pd.read_csv('./data/loan_prediction.csv')
 np.random.seed(9)
 X = data.iloc[:, :-1]
 y = data.iloc[:, -1]
@@ -18,3 +19,22 @@
 
 
 # Write your solution here :
+def decision_classifier_plot(X_train,X_test,y_train,y_test,depth_list):
+    error_train = []
+    error_test = []
+    for i in range(len(depth_list)):
+        dt_classifier = DecisionTreeClassifier(max_depth=depth_list[i],random_state=9)
+        dt_classifier.fit(X_train,y_train)
+        predict_train = dt_classifier.predict(X_train)
+        predict_test = dt_classifier.predict(X_test)
+
+        error_train.append(accuracy_score(y_train,predict_train))
+
+        error_test.append(accuracy_score(y_test,predict_test))
+
+
+    plt.plot(depth_list,error_test)
+    plt.plot(depth_list,error_train)
+    plt.show()
+
+
diff --git a/q04_decision_classifier_plot/tests/__pycache__/__init__.cpython-36.pyc b/q04_decision_classifier_plot/tests/__pycache__/__init__.cpython-36.pyc
diff --git a/...cision_classifier_plot/tests/__pycache__/test_q04_decision_classifier_plot.cpython-36.pyc b/...cision_classifier_plot/tests/__pycache__/test_q04_decision_classifier_plot.cpython-36.pyc