1dold.py

import pandas as pd
from sklearn.linear_model import LinearRegression
from sklearn.preprocessing import PolynomialFeatures
# Referenced Aula and the code Documentation
df_full = pd.read_csv('Trips_Full_Data.csv') # I changed what these are called because i kept getting confused.
df_distance = pd.read_csv('Trips_by_Distance.csv')


df_full = df_full.dropna() # Using pandas so this is slightly different from the way I did it in 1a
df_full = df_full.fillna(0)

df_full['Week'] = df_full['Week of Date'].str.extract(r'(\d+)', expand=False).astype(int) #https://www.geeksforgeeks.org/python-pandas-series-str-extract/
df_full_week32 = df_full[df_full['Week'] == 32] # Need to group by week32

merged_df = pd.merge(df_full_week32, df_distance, on='Level', how='inner') # merging the datasets based on level, all are 'national'
# https://www.w3schools.com/python/pandas/ref_df_merge.asp

# People vs Distance: 


X = merged_df[['Trips 25-100 Miles']]
y = merged_df['Number of Trips 10-25']  

#Linear model
model = LinearRegression()
model.fit(X, y)
print("Linear model:")
r_sq = model.score(X, y)
print("Coefficient of determination (R^2):", r_sq)
print("Intercept:", model.intercept_)
print("Coefficients:", model.coef_)
y_pred = model.predict(X)
print("Predicted response:\n", y_pred)

# There r value is 0 so I don't think the relationship is linear. I am going to use polynomial now. 

poly = PolynomialFeatures(degree=2)
X_poly = poly.fit_transform(X)
poly_model = LinearRegression()
poly_model.fit(X_poly, y)
print("Polynomial model:")
y_pred = poly_model.predict(X_poly)
print("Predicted response:\n", y_pred)
print("Intercept:", poly_model.intercept_)
print("Coefficients:", poly_model.coef_)
r_sq = poly_model.score(X_poly, y)
print("Coefficient of determination (R^2):", r_sq)


# Linear model:
# ('Coefficient of determination (R^2):', 0.0)T
# ('Intercept:', 179005342.22197562)
# ('Coefficients:', array([-4.52300271e-16]))
# ('Predicted response:\n', array([1.79005342e+08, 1.79005342e+08, 1.79005342e+08, ...,
#        1.79005342e+08, 1.79005342e+08, 1.79005342e+08]))

# Polynomial model:
# ('Predicted response:\n', array([1.79005342e+08, 1.79005342e+08, 1.79005342e+08, ...,
#        1.79005342e+08, 1.79005342e+08, 1.79005342e+08]))
# ('Intercept:', 179005345.47188395)
# ('Coefficients:', array([ 0.00000000e+00, -7.46003515e-08,  4.26695402e-16]))
# ('Coefficient of determination (R^2):', -2.220446049250313e-16) 

from sklearn.model_selection import train_test_split
from sklearn.metrics import r2_score, mean_squared_error

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
poly = PolynomialFeatures(degree=2)
X_train_poly = poly.fit_transform(X_train)
X_test_poly = poly.transform(X_test)
# Initialize polynomial regression model
poly_model = LinearRegression()
poly_model.fit(X_train_poly, y_train)
y_pred = poly_model.predict(X_test_poly)
# Evaluate the model
r_squared = r2_score(y_test, y_pred)
mse = mean_squared_error(y_test, y_pred)

print("Trained model R value (R^2):", r_squared)
print("Mean squared error:", mse)