def estimate_beta(x, y):
beta_initial = [random.random() for x_i in x[0]]
return minimize_stochastic(squared_error,
squared_error_gradient,
x, y,
beta_initial,
0.001)
def estimate_beta_ridge(x, y, alpha):
"""use gradient descent to fit a ridge regression
with penalty alpha"""
beta_initial = [random.random() for x_i in x[0]]
return minimize_stochastic(
partial(squared_error_ridge, alpha=alpha),
partial(squared_error_ridge_gradient, alpha=alpha), x, y, beta_initial,
0.001)
def estimate_beta_ridge(x, y, alpha):
"""use gradient descent to fit a ridge regression
with penalty alpha"""
beta_initial = [random.random() for x_i in x[0]]
return minimize_stochastic(partial(squared_error_ridge, alpha=alpha),
partial(squared_error_ridge_gradient,
alpha=alpha),
x, y,
beta_initial,
0.001)
def estimate_beta(x, y):
beta_initial = [random.random() for x_i in x[0]]
return minimize_stochastic(squared_error, squared_error_gradient, x, y,
beta_initial, 0.001)
