diff --git a/__pycache__/__init__.cpython-36.pyc b/__pycache__/__init__.cpython-36.pyc index ebbd53a..83a4f24 100644 Binary files a/__pycache__/__init__.cpython-36.pyc and b/__pycache__/__init__.cpython-36.pyc differ diff --git a/q01_load_data/__pycache__/__init__.cpython-36.pyc b/q01_load_data/__pycache__/__init__.cpython-36.pyc index 745b533..5a99dec 100644 Binary files a/q01_load_data/__pycache__/__init__.cpython-36.pyc and b/q01_load_data/__pycache__/__init__.cpython-36.pyc differ diff --git a/q01_load_data/__pycache__/build.cpython-36.pyc b/q01_load_data/__pycache__/build.cpython-36.pyc index 108e4a3..b116a8c 100644 Binary files a/q01_load_data/__pycache__/build.cpython-36.pyc and b/q01_load_data/__pycache__/build.cpython-36.pyc differ diff --git a/q01_load_data/build.py b/q01_load_data/build.py index e4cd8e3..1f470a8 100644 --- a/q01_load_data/build.py +++ b/q01_load_data/build.py @@ -1,10 +1,16 @@ +# %load q01_load_data/build.py # Default imports import pandas as pd from sklearn.model_selection import train_test_split - - path = 'data/house_prices_multivariate.csv' - # Write your solution here +def load_data(path, tests=0.33, randoms=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=tests, random_state=randoms) + return df, X_train, X_test, y_train, y_test + + diff --git a/q01_load_data/tests/__pycache__/__init__.cpython-36.pyc b/q01_load_data/tests/__pycache__/__init__.cpython-36.pyc index 133357e..b00aaa8 100644 Binary files a/q01_load_data/tests/__pycache__/__init__.cpython-36.pyc and b/q01_load_data/tests/__pycache__/__init__.cpython-36.pyc differ diff --git a/q01_load_data/tests/__pycache__/test_q01_load_data.cpython-36.pyc b/q01_load_data/tests/__pycache__/test_q01_load_data.cpython-36.pyc index 689755b..97e5d2f 100644 Binary files a/q01_load_data/tests/__pycache__/test_q01_load_data.cpython-36.pyc and b/q01_load_data/tests/__pycache__/test_q01_load_data.cpython-36.pyc differ diff --git a/q02_Max_important_feature/__pycache__/__init__.cpython-36.pyc b/q02_Max_important_feature/__pycache__/__init__.cpython-36.pyc index 93c9119..3d41d06 100644 Binary files a/q02_Max_important_feature/__pycache__/__init__.cpython-36.pyc and b/q02_Max_important_feature/__pycache__/__init__.cpython-36.pyc differ diff --git a/q02_Max_important_feature/__pycache__/build.cpython-36.pyc b/q02_Max_important_feature/__pycache__/build.cpython-36.pyc index 2b7cfd4..e2064e8 100644 Binary files a/q02_Max_important_feature/__pycache__/build.cpython-36.pyc and b/q02_Max_important_feature/__pycache__/build.cpython-36.pyc differ diff --git a/q02_Max_important_feature/build.py b/q02_Max_important_feature/build.py index 51fbde6..bd00ef5 100644 --- a/q02_Max_important_feature/build.py +++ b/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 +7,9 @@ # Write your code here +def Max_important_feature(data_set, target_variable='SalePrice', n = 4): + return list(data_set.corr()[target_variable].sort_values(ascending=False)[1:5].index) + +list(data_set.corr()['SalePrice'].sort_values(ascending=False)[1:5].index) + + diff --git a/q02_Max_important_feature/tests/__pycache__/__init__.cpython-36.pyc b/q02_Max_important_feature/tests/__pycache__/__init__.cpython-36.pyc index cec58d4..bd5d06b 100644 Binary files a/q02_Max_important_feature/tests/__pycache__/__init__.cpython-36.pyc and b/q02_Max_important_feature/tests/__pycache__/__init__.cpython-36.pyc differ diff --git a/q02_Max_important_feature/tests/__pycache__/test_q02max_important_feature.cpython-36.pyc b/q02_Max_important_feature/tests/__pycache__/test_q02max_important_feature.cpython-36.pyc index cb6849b..c27b9a7 100644 Binary files a/q02_Max_important_feature/tests/__pycache__/test_q02max_important_feature.cpython-36.pyc and b/q02_Max_important_feature/tests/__pycache__/test_q02max_important_feature.cpython-36.pyc differ diff --git a/q03_polynomial/__pycache__/__init__.cpython-36.pyc b/q03_polynomial/__pycache__/__init__.cpython-36.pyc index aa42922..743cff9 100644 Binary files a/q03_polynomial/__pycache__/__init__.cpython-36.pyc and b/q03_polynomial/__pycache__/__init__.cpython-36.pyc differ diff --git a/q03_polynomial/__pycache__/build.cpython-36.pyc b/q03_polynomial/__pycache__/build.cpython-36.pyc index 3be41d0..e302e4a 100644 Binary files a/q03_polynomial/__pycache__/build.cpython-36.pyc and b/q03_polynomial/__pycache__/build.cpython-36.pyc differ diff --git a/q03_polynomial/build.py b/q03_polynomial/build.py index 26d8971..c3fb9db 100644 --- a/q03_polynomial/build.py +++ b/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 +10,13 @@ # Write your solution here +def polynomial(po=5, randoms=9): + poly_model = make_pipeline(PolynomialFeatures(po, include_bias=False), LinearRegression()) + X_train_new = X_train[['OverallQual','GrLivArea','GarageCars','GarageArea']] + poly_model.fit(X_train_new, y_train) + return poly_model + + +polynomial(5,9) + + diff --git a/q03_polynomial/tests/__pycache__/__init__.cpython-36.pyc b/q03_polynomial/tests/__pycache__/__init__.cpython-36.pyc index 6e20876..69a1596 100644 Binary files a/q03_polynomial/tests/__pycache__/__init__.cpython-36.pyc and b/q03_polynomial/tests/__pycache__/__init__.cpython-36.pyc differ diff --git a/q03_polynomial/tests/__pycache__/test_q03_polynomial.cpython-36.pyc b/q03_polynomial/tests/__pycache__/test_q03_polynomial.cpython-36.pyc index ef8c88b..7b4d020 100644 Binary files a/q03_polynomial/tests/__pycache__/test_q03_polynomial.cpython-36.pyc and b/q03_polynomial/tests/__pycache__/test_q03_polynomial.cpython-36.pyc differ diff --git a/q04_ridge/__pycache__/__init__.cpython-36.pyc b/q04_ridge/__pycache__/__init__.cpython-36.pyc index 4342136..4b7448b 100644 Binary files a/q04_ridge/__pycache__/__init__.cpython-36.pyc and b/q04_ridge/__pycache__/__init__.cpython-36.pyc differ diff --git a/q04_ridge/__pycache__/build.cpython-36.pyc b/q04_ridge/__pycache__/build.cpython-36.pyc index ea08c01..7c53414 100644 Binary files a/q04_ridge/__pycache__/build.cpython-36.pyc and b/q04_ridge/__pycache__/build.cpython-36.pyc differ diff --git a/q04_ridge/build.py b/q04_ridge/build.py index 9ee00b1..3a6cfb8 100644 --- a/q04_ridge/build.py +++ b/q04_ridge/build.py @@ -1,15 +1,28 @@ +# %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): + ridge_model = Ridge(alpha=alpha, normalize=True, random_state=9) + ridge_model.fit(X_train, y_train) + y_pred = ridge_model.predict(X_test) + test_rmse = np.sqrt(mean_squared_error(y_pred=y_pred, y_true=y_test)) + y_train_pred = ridge_model.predict(X_train) + train_rmse = np.sqrt(mean_squared_error(y_pred=y_train_pred, y_true=y_train)) + return train_rmse, test_rmse, ridge_model + +X_train.head() +ridge(alpha=0.01) diff --git a/q04_ridge/tests/__pycache__/__init__.cpython-36.pyc b/q04_ridge/tests/__pycache__/__init__.cpython-36.pyc index 6d021b5..4e4ef9a 100644 Binary files a/q04_ridge/tests/__pycache__/__init__.cpython-36.pyc and b/q04_ridge/tests/__pycache__/__init__.cpython-36.pyc differ diff --git a/q04_ridge/tests/__pycache__/test_q04_ridge.cpython-36.pyc b/q04_ridge/tests/__pycache__/test_q04_ridge.cpython-36.pyc index 0549421..21955bd 100644 Binary files a/q04_ridge/tests/__pycache__/test_q04_ridge.cpython-36.pyc and b/q04_ridge/tests/__pycache__/test_q04_ridge.cpython-36.pyc differ diff --git a/q05_lasso/__pycache__/__init__.cpython-36.pyc b/q05_lasso/__pycache__/__init__.cpython-36.pyc index 1005306..8fd6fcf 100644 Binary files a/q05_lasso/__pycache__/__init__.cpython-36.pyc and b/q05_lasso/__pycache__/__init__.cpython-36.pyc differ diff --git a/q05_lasso/__pycache__/build.cpython-36.pyc b/q05_lasso/__pycache__/build.cpython-36.pyc index b4ea629..2511cdd 100644 Binary files a/q05_lasso/__pycache__/build.cpython-36.pyc and b/q05_lasso/__pycache__/build.cpython-36.pyc differ diff --git a/q05_lasso/build.py b/q05_lasso/build.py index fb30d50..4f94886 100644 --- a/q05_lasso/build.py +++ b/q05_lasso/build.py @@ -1,14 +1,27 @@ +# %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 +def lasso(alpha=0.01): + lasso_model = Lasso(alpha=alpha, normalize=True, random_state=9) + lasso_model.fit(X_train, y_train) + y_pred = lasso_model.predict(X_test) + test_rmse = np.sqrt(mean_squared_error(y_pred=y_pred, y_true=y_test)) + y_train_pred = lasso_model.predict(X_train) + train_rmse = np.sqrt(mean_squared_error(y_pred=y_train_pred, y_true=y_train)) + return train_rmse, test_rmse + +lasso(alpha=0.01) + diff --git a/q05_lasso/tests/__pycache__/__init__.cpython-36.pyc b/q05_lasso/tests/__pycache__/__init__.cpython-36.pyc index 8869434..5e786fe 100644 Binary files a/q05_lasso/tests/__pycache__/__init__.cpython-36.pyc and b/q05_lasso/tests/__pycache__/__init__.cpython-36.pyc differ diff --git a/q05_lasso/tests/__pycache__/test_q05_lasso.cpython-36.pyc b/q05_lasso/tests/__pycache__/test_q05_lasso.cpython-36.pyc index 438235e..a4d77d1 100644 Binary files a/q05_lasso/tests/__pycache__/test_q05_lasso.cpython-36.pyc and b/q05_lasso/tests/__pycache__/test_q05_lasso.cpython-36.pyc differ diff --git a/q06_cross_validation/__pycache__/__init__.cpython-36.pyc b/q06_cross_validation/__pycache__/__init__.cpython-36.pyc index fa7d8bf..d257012 100644 Binary files a/q06_cross_validation/__pycache__/__init__.cpython-36.pyc and b/q06_cross_validation/__pycache__/__init__.cpython-36.pyc differ diff --git a/q06_cross_validation/__pycache__/build.cpython-36.pyc b/q06_cross_validation/__pycache__/build.cpython-36.pyc index 19e8bd8..30d22f4 100644 Binary files a/q06_cross_validation/__pycache__/build.cpython-36.pyc and b/q06_cross_validation/__pycache__/build.cpython-36.pyc differ diff --git a/q06_cross_validation/build.py b/q06_cross_validation/build.py index e39b93b..8db1597 100644 --- a/q06_cross_validation/build.py +++ b/q06_cross_validation/build.py @@ -1,13 +1,18 @@ +# %load q06_cross_validation/build.py # Default imports from sklearn.model_selection import cross_val_score import numpy as np 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 cross_validation(Model, X, y): + scores = cross_val_score(Model, X=X, y=y, scoring='neg_mean_squared_error', cv=5) + return scores.mean() diff --git a/q06_cross_validation/tests/__pycache__/__init__.cpython-36.pyc b/q06_cross_validation/tests/__pycache__/__init__.cpython-36.pyc index ca3f5cd..6b299f5 100644 Binary files a/q06_cross_validation/tests/__pycache__/__init__.cpython-36.pyc and b/q06_cross_validation/tests/__pycache__/__init__.cpython-36.pyc differ diff --git a/q06_cross_validation/tests/__pycache__/test_q06_cross_validation.cpython-36.pyc b/q06_cross_validation/tests/__pycache__/test_q06_cross_validation.cpython-36.pyc index e7acaaf..704cc92 100644 Binary files a/q06_cross_validation/tests/__pycache__/test_q06_cross_validation.cpython-36.pyc and b/q06_cross_validation/tests/__pycache__/test_q06_cross_validation.cpython-36.pyc differ