COSC 130 - Project 01

import pandas as pd
import matplotlib.pyplot as plt

df = pd.read_table(filepath_or_buffer='auto_data.txt', sep='\t')
weight = list(df.wt)
mpg = list(df.mpg)
ln_mpg = list(df.ln_mpg)

print('Length of list weight : ' + str(len(weight)))
print('Length of list mpg : ' + str(len(mpg)))
print('Length of list ln_mpg : ' + str(len(ln_mpg)))

print("{:>6}{:>8}{:>10}".format("Weight", "MPG", "LN_MPG"))
print("{:-^24}".format(""))
for i in range(0,10):
    print("{:>6}{:>8}{:>10}".format(weight[i], mpg[i], ln_mpg[i]))

plt.figure(figsize=[12,4])
plt.subplot(1,2,1)
plt.scatter(weight, mpg, c='skyblue', edgecolor='k')
plt.xlabel('Weight (in Pounds)')
plt.ylabel('MPG')
plt.title('Plot of MPG against Weight')
plt.subplot(1,2,2)
plt.scatter(weight, ln_mpg, c='skyblue', edgecolor='k')
plt.xlabel('Weight (in Pounds)')
plt.ylabel('Natural Logarithm of MPG')
plt.title('Plot of Log-MPG against Weight')
plt.show()

x_train = (weight[:300])
x_test = (weight[300:])
y_train = (ln_mpg[:300])
y_test = (ln_mpg[300:])
mpg_train = (mpg[:300])
mpg_test = (mpg[300:])
n_train = (len(x_train))
n_test = (len(x_test))

print('Training Set Size:' + ' ' + str(n_train))
print('Test Set Size:' + ' ' + str(n_test))

plt.figure(figsize=[12,4])
plt.subplot(1,2,1)
plt.scatter(x_train, y_train, c='skyblue', edgecolor='k')
plt.xlabel('Weight (in Pounds)')
plt.ylabel('Natural Logarithm of MPG')
plt.title('Training Set')
plt.subplot(1,2,2)
plt.scatter(x_test, y_test, c='skyblue', edgecolor='k')
plt.xlabel('Weight (in Pounds)')
plt.ylabel('Natural Logarithm of MPG')
plt.title('Test Set')
plt.show()

mean_x = sum(x_train)/len(x_train)
mean_y = sum(y_train)/len(y_train)

print('Mean of X = {:.2f}'.format(mean_x))
print('Mean of Y = {:.4f}'.format(mean_y))

Sxx = sum([((x - mean_x) ** 2) for x in x_train])
Syy = sum([((x - mean_y) ** 2) for x in y_train])

print('Sxx =' + ' ' + str(round(Sxx,2)))
print('Syy =' + ' ' + str(round(Syy,4)))

var_x = sum([((x - mean_x) ** 2) for x in x_train])/len(x_train)
var_y = sum([((x - mean_y) ** 2) for x in y_train])/len(y_train)

print('Variance of X =' + ' ' + str(round(var_x,2)))
print('Variance of Y =' + ' ' + str(round(var_y,4)))

Sxy = sum([((x-mean_x)*(y-mean_y)) for x, y in zip(x_train, y_train)])

print("Sxy =", round(Sxy,2))

beta_1 = Sxy / Sxx
beta_0 = mean_y - beta_1 * mean_x

print("beta_0 =", round(beta_0, 4))
print("beta_1 =", round(beta_1, 8))

y_vals = [beta_0 + beta_1 * 1500, beta_0 + beta_1 * 5500]
plt.figure(figsize=[12,4])
plt.subplot(1,2,1)
plt.scatter(x_train, y_train, c='skyblue', edgecolor='k')
plt.plot([1500,5500], y_vals, c='crimson', lw=3)
plt.xlabel('Weight (in Pounds)')
plt.ylabel('Natural Logarithm of MPG')
plt.title('Training Set')
plt.subplot(1,2,2)
plt.scatter(x_test, y_test, c='skyblue', edgecolor='k')
plt.plot([1500,5500], y_vals, c='crimson', lw=3)
plt.xlabel('Weight (in Pounds)')
plt.ylabel('Natural Logarithm of MPG')
plt.title('Test Set')
plt.show()

pred_y_train = [beta_0 + beta_1 * x for x in x_train]

error_y_train = [y_train[i] - pred_y_train[i] for i in range(len(y_train))]

print(f"{'True y':>6} {'Pred y':>10} {'Error':>10}")
print("-" * 30)
for i in range(10):
    print(f"{y_train[i]:>.4f} {pred_y_train[i]:>10.4f} {error_y_train[i]:>10.4f}")

sse_train = 0
for i in range(len(y_train)):
    sse_train += error_y_train[i]**2
sse_train = round(sse_train, 4)
print("Training SSE = ", sse_train)

y_train_mean = sum(y_train)/len(y_train)
r2 = 1 - (sse_train/sum([(y_train[i] - y_train_mean)**2 for i in range(len(y_train))]))
r2 = round(r2, 4)
print("Training r-squared = ", r2)

pred_y_test = [beta_0 + beta_1 * x for x in x_test]

error_y_test = [y_test[i] - pred_y_test[i] for i in range(len(y_test))]

print(f"{'True y':>6} {'Pred y':>10} {'Error':>10}")
print("-" * 30)
for i in range(10):
    print(f"{y_test[i]:>6.4f} {pred_y_test[i]:>10.4f} {error_y_test[i]:>10.4f}")

sse_test = 0
for i in range(len(y_test)):
    sse_test += error_y_test[i]**2
sse_test = round(sse_test, 4)
print("Test SSE =", sse_test)

y_test_mean = sum(y_test)/len(y_test)
r2 = 1 - (sse_test/sum([(y_test[i] - y_test_mean)**2 for i in range(len(y_test))]))
r2 = round(r2, 4)
print("Test r-Squared = ", r2)

plt.figure(figsize=[8,4])
plt.scatter(x_test, y_test, c='skyblue', edgecolor='k')
plt.plot([1500,5250], [beta_0 + beta_1 * 1500, beta_0 + beta_1 * 5250],
         c='crimson', lw=3)
for i in range(n_test):
    plt.plot([x_test[i], x_test[i]], [pred_y_test[i], y_test[i]],
             c='coral', zorder=0) 
plt.show()

e = 2.718281828
pred_mpg_test = [e ** y for y in pred_y_test]
for i in pred_y_test:
    pred_mpg_test.append(e**i)
print(pred_mpg_test)

error_mpg_test = [mpg_test[i] - pred_mpg_test [i] for i in range (len(mpg_test))]

print ("True MPG    Pred MPG    Error")
print("-" * 29)
for i in range (len (mpg_test)):
    if i == 10:
        break
    print ("{:8.1f}{:12.1f}{:9.1f}".format (mpg_test [i], pred_mpg_test[i], error_mpg_test[i]))