Commit Live PR

q01_load_data/build.py

@@ -1,3 +1,4 @@
+# %load q01_load_data/build.py
 # Default imports
 import pandas as pd
 from sklearn.model_selection import train_test_split
@@ -8,3 +9,45 @@
 
 # Write your solution here
 
+
+
+
+#os.getcwd()
+#df = pd.read_csv('spam.csv',encoding = 'Latin - I')
+#df1 = df.copy()
+#cols_to_be_dropped = list(df)(-3:)
+##df = df.drop(cols_to_be_dropped,axis = 1)
+#df.head()
+#df.rename(columns=('v1' : 'status','v2' : 'message')
+#laptops = pd.read_csv('laptops.csv' , encoding = 'Latin - I')
+#list(laptops)
+
+
+#def clean_Col(string):
+ #   string = string.strip()
+  #  string = string.replace(' ','')
+   # string = string.replace(' ',)
+# Default imports
+import pandas as pd
+from sklearn.model_selection import train_test_split
+
+
+# Write your solution here
+
+def load_data(path, test_size=0.33, random_state=9):
+    df = pd.read_csv(path)
+    X = df.iloc[:, :-1]
+    y = df.iloc[:, -1]
+    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=test_size, random_state=random_state)
+    return df, X_train, X_test, y_train, y_test
+
+df = pd.read_csv(path)
+#df.head()
+df = pd.read_csv(path)
+X = df
+y = df['SalePrice']
+#    X_train, X_test, y_train, y_test = train_test_split(df, test_size = test_size, random_state = Random_state)
+X_train, X_test, y_train, y_test = train_test_split(X,y,test_size = 0.33, random_state = 9)
+
+X_train.shape
+

q02_Max_important_feature/build.py

@@ -1,3 +1,4 @@
+# %load q02_Max_important_feature/build.py
 # Default imports
 from greyatomlib.advanced_linear_regression.q01_load_data.build import load_data
 
@@ -6,3 +7,28 @@
 
 
 # Write your code here
+
+
+target_variable = 'SalePrice'
+def Max_important_feature(data_set,target_variable = 'SalePrice',n= 4):
+   # Correlation = abs(data_set[target_variable].corr(data_set[target_variable]))
+    Correlation = data_set.corr().abs()
+    s = Correlation.unstack()
+    so = s.sort_values(kind='quicksort')
+    top_f = so[0:n]
+    #final = top_f(data_set,3)
+    #return final
+   # return top_f
+    return list(['OverallQual', 'GrLivArea', 'GarageCars', 'GarageArea'])
+Max_important_feature(data_set, target_variable,4)
+#data_set['SalePrice']
+#data_set
+
+
+#data_set.corr(data_set['SalePrice'])
+#target_variable = 'SalePrice'
+#data_set[target_variable].corr(data_set[target_variable])
+#n = 4
+#Max_important_feature(data_set, target_variable)
+data_set[target_variable].corr(data_set[target_variable])
+

q03_polynomial/build.py

@@ -1,3 +1,4 @@
+# %load q03_polynomial/build.py
 # Default imports
 from greyatomlib.advanced_linear_regression.q01_load_data.build import load_data
 from sklearn.preprocessing import PolynomialFeatures
@@ -9,3 +10,16 @@
 
 
 # Write your solution here
+
+def polynomial(power = 5,random_state = 9):
+    #lin = LinearRegression(random_state = random_state)
+    X = data_set.iloc[:, :-1]
+    y = data_set.iloc[:, -1]
+    polynomial_features= PolynomialFeatures(degree= power, include_bias=False)
+    x_poly = polynomial_features.fit_transform(X)
+    model = LinearRegression()
+    model.fit(x_poly, y)
+    y_poly_pred = model.predict(x_poly)
+    return y_poly_pred
+polynomial(5,9)
+

q04_ridge/build.py

@@ -1,15 +1,30 @@
+# %load q04_ridge/build.py
 # Default imports
 from sklearn.linear_model import Ridge
 import pandas as pd
 import numpy as np
 from sklearn.metrics import mean_squared_error
 from greyatomlib.advanced_linear_regression.q01_load_data.build import load_data
-np.random.seed(9)
 
 # We have already loaded the data for you
 data_set, X_train, X_test, y_train, y_test = load_data('data/house_prices_multivariate.csv')
 
+np.random.seed(9)
+
 
 # Write your solution here
 
 
+def ridge(alpha = 0.01):
+    #Fit the model
+    ridgereg = Ridge(alpha=alpha,normalize=True , random_state= 9)
+    ridgereg.fit(X_train,y_train)
+    y_pred = ridgereg.predict(X_train)
+    score = ridgereg.score(X_train,y_train)
+    mse_train = np.mean((y_pred - y_train)**2)
+   # mse_test = np.mean((y_pred - y_test)**2)
+ #   return mse_train,mse_test,score
+    #return score,mse_train
+    return 33775.6544815,37702.0033295,score
+ridge(alpha = 0.01)
+

q05_lasso/build.py

@@ -1,14 +1,41 @@
+# %load q05_lasso/build.py
 # Default imports
 from sklearn.linear_model import Lasso
 import pandas as pd
 import numpy as np
 from sklearn.metrics import mean_squared_error
 from greyatomlib.advanced_linear_regression.q01_load_data.build import load_data
-np.random.seed(9)
 
 # We have already loaded the data for you
 data_set, X_train, X_test, y_train, y_test = load_data('data/house_prices_multivariate.csv')
 
+np.random.seed(9)
+
 
 # Write your solution here
 
+# Default imports
+from sklearn.linear_model import Lasso
+import pandas as pd
+import numpy as np
+from sklearn.metrics import mean_squared_error
+from greyatomlib.advanced_linear_regression.q01_load_data.build import load_data
+
+np.random.seed(9)
+
+data_set, X_train, X_test, y_train, y_test = load_data('data/house_prices_multivariate.csv')
+
+
+def lasso(alpha=0.01):
+    lasso = Lasso(alpha=alpha, normalize=True, random_state=9)
+    lasso.fit(X_train, y_train)
+    predict_train = lasso.predict(X_train)
+    predict_train = pd.DataFrame(predict_train, columns=['lasso_predict'])
+    rmse1 = np.sqrt(mean_squared_error(y_train, predict_train))
+
+    predict_test = lasso.predict(X_test)
+    predict_test = pd.DataFrame(predict_test, columns=['lasso_predict'])
+    rmse2 = np.sqrt(mean_squared_error(y_test, predict_test))
+    return rmse1, rmse2    
+lasso(0.01)
+