import numpy as np
import matplotlib.pyplot as plt
import a_linear_regression.linear_utils as linear
X = []
Y = []
# Read csv and populate arrays
for line in open('csv/data_1d.csv'):
x, y = line.split(',')
X.append(float(x))
Y.append(float(y))
# Converting regular arrays into numpy arrays
X = np.array(X)
Y = np.array(Y)
# Visualize the input data
plt.scatter(X, Y)
# plt.show()
Y_hat = linear.calculate_y_hat(X, Y)
# Plot with predictions
plt.scatter(X, Y)
plt.plot(X, Y_hat)
plt.show()
# Calculate R squared
print(linear.calculate_r_squared(Y, Y_hat))
def get_r_squared(X, Y):
w = linear.calculate_W(X, Y)
Y_hat = np.dot(X, w)
return linear.calculate_r_squared(Y, Y_hat)
X = []
Y = []
for line in open('csv/data_2d.csv'):
x1, x2, y = line.split(',')
X.append([float(x1), float(x2), BIAS])
Y.append(float(y))
X = np.array(X)
Y = np.array(Y)
print('Shape of X is ' + str(X.shape))
print('Shape of Y is ' + str(Y.shape))
# Plot in 3D coordinate space
# TODO: Code example didn't work...need to investigate
w = linear.calculate_W(X, Y)
print('Shape of w is ' + str(w.shape))
Y_hat = np.dot(X, w)
print('Shape of Y Hat is ' + str(Y_hat.shape))
r_squared = linear.calculate_r_squared(Y, Y_hat)
residual = Y - Y_hat
some = 1 - residual.dot(residual)
print('r_squared is ' + str(r_squared))