regression1_linear.py

import numpy as np
import matplotlib.pyplot as plt

# 学習データを読み込む
train = np.loadtxt('click.csv', delimiter=',', dtype='int', skiprows=1)
train_x = train[:,0]
train_y = train[:,1]

# 標準化
mu = train_x.mean()
sigma = train_x.std()
def standardize(x):
    return(x-mu)/sigma

train_z = standardize(train_x)

# パラメータを初期化
theta0 = np.random.rand()
theta1 = np.random.rand()

print(theta0)
print(theta1)


# 予測関数
def f(x):
    return theta0 + theta1 * x

# 目的関数（y：実際の目的関数　fx：パラメータを仮置きした目的関数）：誤差
def E(x, y):
    return 0.5 * np.sum (y-f(x) **2)

# 学習率
ETA = 1e-3

# 誤差の差分
diff = 1

# 更新回数
count = 0

# 誤差の差分が0.01以下になるまでパラメータ更新を繰り返す
error = E(train_z, train_y)
while diff > 1e-2:
    tmp_theta0 = theta0 - ETA * np.sum((f(train_z)- train_y))
    tmp_theta1 = theta1 - ETA * np.sum((f(train_z)- train_y) * train_z)

    # パラメータ更新
    theta0 = tmp_theta0
    theta1 = tmp_theta1

    # 前回誤差との差分計算
    current_error = E(train_z, train_y)
    diff = error - current_error
    error = current_error

    count += 1
    log = "{}回目：theta0 = {:.3f},theta1 = {:.3f}, 差分 = {:.4f}"
    print(log.format(count,theta0,theta1,diff))

# プロットして確認
x = np.linspace(-3, 3, 100)
plt.plot(train_z, train_y, 'o')
plt.plot(x, f(x))
plt.show()