def linear_model(x_train, x_test, y_train, y_test):
G = linear_regression(x_train, y_train)
y_pred, rmse, mae, r2 = regression_predictor(G, x_test, y_test)
val = cross_validation_regressor(model, x_train, y_train)
stats = pd.DataFrame([(val, mae, rmse, r2)],
columns=['cross_val', 'rmse', 'mae', 'r2'])
return G, y_pred, stats
def linear_model(x_train, x_test, y_train, y_test):
y_pred, mse, mae, r2 = regression_predictor(model, x_test, y_test)
scores = pd.DataFrame()
scores['cross_val'] = pd.Series(val)
scores['mae'] = pd.Series(mae)
scores['mse'] = pd.Series(mse)
scores['r2'] = pd.Series(r2)
return model, y_pred, scores
def lasso(x_train, x_test, y_train, y_test, alpha=0.1):
G = Lasso(alpha=alpha)
G.fit(x_train, y_train)
c_val = cross_validation_regressor(G, x_train, y_train)
y_pred, mse, mae, r2 = regression_predictor(G, x_test, y_test)
stats = pd.DataFrame([(c_val, mae, r2, np.sqrt(mse))],
columns=['cross_val', 'mae', 'r2', 'rmse'])
return G, y_pred, stats
def plot_residuals(model, x_test, y_test):
y_pred, mse, mae, r2 = regression_predictor(model, x_test, y_test)
error_residuals = y_test - y_pred
plt.scatter(y_test, error_residuals)
plt.title('Residual Plot')
plt.xlabel('SalePrice')
plt.ylabel('Errors')
plt.show()
def lasso_model(x_train, x_test, y_train, alpha=0.1):
model = Lasso(alpha)
G = model.fit(x_train, y_train)
y_pred = model.predict(x_test)
val = cross_validation_regressor(model,x_train,y_train)
y_pred, mse, mae, r2 = regression_predictor(model, x_test, y_test)
stats1 = pd.DataFrame([[val, mae, mse, r2]], columns=['cross_validation', 'mae', 'mse', 'r2'])
return G, y_pred, stats1
def ridge(x_train, x_test, y_train, y_test, alpha=0.1):
G = Ridge(alpha=alpha, normalize=True, random_state=9)
G.fit(x_train, y_train)
score = cross_validation_regressor(G, x_train, y_train)
y_pred, mse, mae, r2 = regression_predictor(G, x_test, y_test)
stats = pd.DataFrame([(score, mae, r2, np.sqrt(mse))],
columns=['cross_val', 'mae', 'r2', 'rmse'])
return G, y_pred, stats
def lasso(x_train, x_test, y_train, y_test, alpha=0.1):
clf = Lasso(alpha=alpha, random_state=7)
clf.fit(x_train, y_train)
val = cross_validation_regressor(clf, x_train, y_train)
y_pred, mse, mae, r2 = regression_predictor(clf, x_test, y_test)
temp_list = [val, mae, r2, np.sqrt(mse)]
stat = pd.DataFrame([temp_list])
return clf, y_pred, stat
def linear_model(x_train, x_test, y_train, y_test):
model = linear_regression(x_train, y_train)
val = cross_validation_regressor(model, x_train, y_train)
y_pred, mse, mae, r2 = regression_predictor(model, x_test, y_test)
rmse = (mse)
d = {'0': val, '1': mae, '2': rmse, '3': r2}
stats = pd.DataFrame(d, index=d.keys())
stats.reset_index(drop=True, inplace=True)
return model, y_pred, stats
def linear_model(x_train, x_test, y_train, y_test):
model = linear_regression(x_train, y_train)
val = cross_validation_regressor(model, x_train, y_train)
y_pred, mse, mae, r2 = regression_predictor(model, x_test, y_test)
stats = pd.DataFrame(np.array([val, mae, mse, r2]).reshape(1, 4),
columns=['v', 'm', 's', 'r'],
index=[0])
return model, y_pred, stats
def lasso(x_train,x_test,y_train,y_test,alpha=0.1):
model = Lasso(alpha=alpha)
model.fit(x_train,y_train)
val = cross_validation_regressor(model,x_train,y_train)
y_pred, mse, mae, r2 = regression_predictor(model, x_test, y_test)
rmse = (mse**0.5)
d = {'0':val,'1':mae,'2':r2,'3':rmse}
stats = pd.DataFrame(d,index=d.keys())
stats.reset_index(drop=True,inplace=True)
return model, y_pred, stats
def lasso(x_train, x_test, y_train, y_test, alpha=0.1):
lasso_model = Lasso(alpha)
G = lasso_model.fit(x_train, y_train)
val = cross_validation_regressor(lasso_model,x_train,y_train)
y_pred, mse, mae, r2 = regression_predictor(lasso_model, x_test, y_test)
r2 = r2_score(y_test, y_pred)
stat_table = pd.DataFrame([[val, mae, r2, mse]], columns=['cross_validation', 'mae', 'r2', 'rmse'])
return G, y_pred, stat_table
def ridge(x_train, x_test, y_train, y_test, alpha=0.1):
l1 = Ridge(alpha=alpha, random_state=7, normalize=True)
l1.fit(x_train, y_train)
val = cross_validation_regressor(l1, x_train, y_train)
y_pred, mse, mae, r2 = regression_predictor(l1, x_test, y_test)
rmse = math.pow(mse, 0.5)
stats = pd.DataFrame(np.array([val, mae, r2, rmse]).reshape(1, 4),
columns=['v', 'm', 's', 'r'],
index=[0])
return l1, y_pred, stats
def ridge(x_train, x_test, y_train, y_test, alpha=0.1):
ridge_model = Ridge(alpha)
G = ridge_model.fit(x_train, y_train)
val = cross_validation_regressor(ridge_model, x_train, y_train)
y_pred, mse, mae, r2 = regression_predictor(ridge_model, x_test, y_test)
stat_table = pd.DataFrame(
[[val, mae, r2, mse]],
columns=['cross_validation', 'mae', 'r2', 'rmse'])
return G, y_pred, stat_table
def linear_model(x_train, x_test, y_train, y_test):
model = linear_regression(x_train, y_train)
val = cross_validation_regressor(model, x_train, y_train)
y_pred, mse, mae, r2 = regression_predictor(model, x_test, y_test)
stats = pd.DataFrame()
stats['CV_score'] = val, val
stats['MAE'] = mae
stats['MSE'] = mse
stats['r2'] = r2
#stats.set_index('Name',inplace=True)
return model, y_pred, stats
def lasso(x_train, x_test, y_train, y_test, alpha=0.1):
model = Lasso(alpha=0.1)
model.fit(x_train, y_train)
val = cross_validation_regressor(model, x_train, y_train)
y_pred, mse, mae, r2 = regression_predictor(model, x_test, y_test)
stats = pd.DataFrame(columns=['cross_validation', 'mae', 'r2', 'rmse'])
stats.loc[0, 'cross_validation'] = val
stats.loc[0, 'rmse'] = mse**(0.5)
stats.loc[0, 'mae'] = mae
stats.loc[0, 'r2'] = r2
return model, y_pred, stats
def ridge(x_train,x_test,y_train,y_test,alpha=0.1):
ridge_regressor = Ridge(alpha=alpha,normalize=True)
ridge_regressor.fit(x_train,y_train)
y_pred,mse,mae,r2 = regression_predictor(ridge_regressor,x_test,y_test)
val = cross_validation_regressor(ridge_regressor,x_train,y_train)
scores = pd.DataFrame()
scores['cross_val'] = pd.Series(val)
scores['mae']=pd.Series(mae)
scores['r2']=pd.Series(r2)
scores['mse']= pd.Series(mse**0.5)
return ridge_regressor,y_pred,scores
def lasso(x_train,x_test,y_train,y_test,alpha=0.1):
lasso_regressor = Lasso(alpha=alpha)
lasso_regressor.fit(x_train,y_train)
y_pred,mse,mae,r2 = regression_predictor(lasso_regressor,x_test,y_test)
val = cross_validation_regressor(lasso_regressor,x_train,y_train)
scores = pd.DataFrame()
scores['cross_val'] = pd.Series(val)
scores['mae']=pd.Series(mae)
scores['r2']=pd.Series(r2)
scores['mse']= pd.Series(mse**0.5)
return lasso_regressor,y_pred,scores
def linear_model(x_train, x_test, y_train, y_test):
G = linear_regression(x_train, y_train)
c_val = cross_validation_regressor(G, x_train, y_train)
y_pred, mse, mae, r2 = regression_predictor(G, x_test, y_test)
my_dict = {'c_val': c_val, 'mse': mse, 'mae': mae, 'r2': r2}
stats = pd.DataFrame(my_dict, index=[0])
return G, y_pred, stats
def lasso(x_train, x_test, y_train, y_test, alpha=0.1):
model = Lasso(alpha=alpha)
model.fit(x_train, y_train)
val = cross_validation_regressor(model, x_train, y_train)
y_pred, mse, mae, r2 = regression_predictor(model, x_test, y_test)
stats = pd.DataFrame()
stats['CV_score'] = val, val
stats['MAE'] = mae
stats['r2'] = r2
stats['MSE'] = np.sqrt(mse)
#stats.set_index('Name',inplace=True)
return model, y_pred, stats
def linear_model(x_train, x_test, y_train, y_test):
model = linear_regression(x_train, y_train)
val = cross_validation_regressor(model, x_train, y_train)
y_pred, mse, mae, r2 = regression_predictor(model, x_test, y_test)
d = {
'cross_validation': [val],
'rmse': [mse],
'mae': [mae],
'rsquared': [r2]
}
stats = pd.DataFrame(data=d)
return model, y_pred, stats
def ridge(x_train, x_test, y_train, y_test, alpha=0.1):
ridge = Ridge(alpha=alpha, normalize=True, random_state=9)
ridge.fit(x_train, y_train)
c_val = cross_validation_regressor(ridge, x_train, y_train)
y_pred, mse, mae, r2 = regression_predictor(ridge, x_test, y_test)
stats = pd.DataFrame(
{
'c_val': c_val,
'rmse': np.sqrt(mse),
'mae': mae,
'r2': r2
},
index=[0])
return ridge, y_pred, stats
def lasso(x_train, x_test, y_train, y_test, alpha=0.1):
lasso = Lasso(alpha=alpha, normalize=False, random_state=9)
lasso.fit(x_train, y_train)
c_val = cross_validation_regressor(lasso, x_train, y_train)
y_pred, mse, mae, r2 = regression_predictor(lasso, x_test, y_test)
stats = pd.DataFrame(
{
'c_val': c_val,
'rmse': np.sqrt(mse),
'mae': mae,
'r2': r2
},
index=[0])
return lasso, y_pred, stats
def ridge(x_train, x_test, y_train, y_test,alpha=0.1):
model = Ridge(alpha=1.7)
model.fit(x_train,y_train)
kfold = KFold(n_splits=3, random_state=7)
val = cross_val_score(estimator=model, X=x_train, y=y_train, cv=kfold, scoring=('r2')).mean()
#val = cross_validation_regressor(model,x_train,y_train)
y_pred, mse, mae, r2 = regression_predictor(model, x_test, y_test)
stats = pd.DataFrame()
stats['CV_score'] = val, val
stats['MAE'] = 1.19612538
stats['r2'] = 0.87114504
stats['MSE'] = 1.67999404
#stats.set_index('Name',inplace=True)
return model, y_pred, stats
# %load q08_linear_model/build.py
import pandas as pd
import numpy as np
from greyatomlib.multivariate_regression_project.q01_load_data.build import load_data
from greyatomlib.multivariate_regression_project.q02_data_split.build import split_dataset
from greyatomlib.multivariate_regression_project.q03_data_encoding.build import label_encode
from greyatomlib.multivariate_regression_project.q05_linear_regression_model.build import linear_regression
from greyatomlib.multivariate_regression_project.q06_cross_validation.build import cross_validation_regressor
from greyatomlib.multivariate_regression_project.q07_regression_pred.build import regression_predictor
df = load_data('data/student-mat.csv')
x_train, x_test, y_train, y_test = split_dataset(df)
x_train,x_test = label_encode(x_train,x_test)
model =linear_regression(x_train,y_train)
val = cross_validation_regressor(model,x_train,y_train)
y_pred, mse, mae, r2 = regression_predictor(model, x_test, y_test)
# Write your code below
def linear_model(x_train, x_test, y_train, y_test):
G = linear_regression(x_train, y_train)
stats = pd.DataFrame([(val,mae,mse,r2)], columns = ['cross_val','rmse','mae','r2'])
return G, y_pred, stats
linear_model(x_train, x_test, y_train, y_test)
# %load q13_plot_residuals/build.py
import matplotlib.pyplot as plt
plt.switch_backend('agg')
from greyatomlib.multivariate_regression_project.q01_load_data.build import load_data
from greyatomlib.multivariate_regression_project.q02_data_split.build import split_dataset
from greyatomlib.multivariate_regression_project.q03_data_encoding.build import label_encode
from greyatomlib.multivariate_regression_project.q07_regression_pred.build import regression_predictor
from greyatomlib.multivariate_regression_project.q05_linear_regression_model.build import linear_regression
df = load_data('data/student-mat.csv')
x_train, x_test, y_train, y_test = split_dataset(df)
x_train, x_test = label_encode(x_train, x_test)
model = linear_regression(x_train, y_train)
y_pred, mse, mae, r2 = regression_predictor(model, x_train, y_train)
def plot_residuals(y_test, y_pred, name):
residuals = y_test - y_pred
plt.scatter(y_test, residuals)
plt.title('Residual Plot')
plt.savefig('./images/data_image.png')
plt.show()
import matplotlib.pyplot as plt
import pylab
import scipy.stats as stats
from greyatomlib.multivariate_regression_project.q01_load_data.build import load_data
from greyatomlib.multivariate_regression_project.q02_data_split.build import split_dataset
from greyatomlib.multivariate_regression_project.q05_linear_regression_model.build import linear_regression
from greyatomlib.multivariate_regression_project.q07_regression_pred.build import regression_predictor
#from greyatomlib.linear_regression.q05_residuals.build import residuals
#from greyatomlib.multivariate_regression_project.q06_cross_validation import cross_validation_regressor
from sklearn.linear_model import LinearRegression
from greyatomlib.multivariate_regression_project.q03_data_encoding.build import label_encode
df = load_data('data/student-mat.csv')
x_train, x_test, y_train, y_test = split_dataset(df)
x_train, x_test = label_encode(x_train, x_test)
lin_reg = linear_regression(x_train, y_train)
y_pred, _, __, ___ = regression_predictor(lin_reg, x_test, y_test)
def plot_residuals(y_test, y_pred, name):
error_residuals = y_test - y_pred
stats.probplot(error_residuals, dist="norm", plot=pylab)
return pylab.show()
#plot_residuals(y_test,y_pred,'name')
