p6 adagrad.py

import numpy as np
import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D
# 本代码是一个最简单的线形回归问题，优化函数为adagrad
rate = 0.2 # learning rate
def da(y,y_p,x):
    return (y-y_p)*(-x)

def db(y,y_p):
    return (y-y_p)*(-1)
def calc_loss(a,b,x,y):
    tmp = y - (a * x + b)
    tmp = tmp ** 2  # 对矩阵内的每一个元素平方
    SSE = sum(tmp) / (2 * len(x))
    return SSE
def draw_hill(x,y):
    a = np.linspace(-20,20,100)
    print(a)
    b = np.linspace(-20,20,100)
    x = np.array(x)
    y = np.array(y)

    allSSE = np.zeros(shape=(len(a),len(b)))
    for ai in range(0,len(a)):
        for bi in range(0,len(b)):
            a0 = a[ai]
            b0 = b[bi]
            SSE = calc_loss(a=a0,b=b0,x=x,y=y)
            allSSE[ai][bi] = SSE

    a,b = np.meshgrid(a, b)

    return [a,b,allSSE]
#  模拟数据
x = [30	,35,37,	59,	70,	76,	88,	100]
y = [1100,	1423,	1377,	1800,	2304,	2588,	3495,	4839]

# 数据归一化
x_max = max(x)
x_min = min(x)
y_max = max(y)
y_min = min(y)

for i in range(0,len(x)):
    x[i] = (x[i] - x_min)/(x_max - x_min)
    y[i] = (y[i] - y_min)/(y_max - y_min)

[ha,hb,hallSSE] = draw_hill(x,y)
hallSSE = hallSSE.T# 重要，将所有的losses做一个转置。原因是矩阵是以左上角至右下角顺序排列元素，而绘图是以左下角为原点。
# 初始化a,b值
a = 10.0
b = -20.0
fig = plt.figure(1, figsize=(12, 8))
fig.suptitle('learning rate: %.2f method: adagrad'%(rate), fontsize=15)
# 绘制图1的曲面
ax = fig.add_subplot(2, 2, 1, projection='3d')
ax.set_top_view()
ax.plot_surface(ha, hb, hallSSE, rstride=2, cstride=2, cmap='rainbow')

# 绘制图2的等高线图
plt.subplot(2,2,2)
ta = np.linspace(-20, 20, 100)
tb = np.linspace(-20, 20, 100)
plt.contourf(ha,hb,hallSSE,15,alpha=0.5,cmap=plt.cm.hot)
C = plt.contour(ha,hb,hallSSE,15,colors='black')
plt.clabel(C,inline=True)
plt.xlabel('a')
plt.ylabel('b')

plt.ion() # iteration on

all_loss = []
all_step = []
last_a = a
last_b = b
n = np.array([0,0])
epsilon = 1e-8
for step in range(1,500):
    loss = 0
    all_da = 0
    all_db = 0
    for i in range(0,len(x)):
        y_p = a*x[i] + b
        loss = loss + (y[i] - y_p)*(y[i] - y_p)/2
        all_da = all_da + da(y[i],y_p,x[i])
        all_db = all_db + db(y[i],y_p)
    #loss_ = calc_loss(a = a,b=b,x=np.array(x),y=np.array(y))
    loss = loss/len(x)

    # 绘制图1中的loss点
    ax.scatter(a, b, loss, color='black')
    # 绘制图2中的loss点
    plt.subplot(2,2,2)
    plt.scatter(a,b,s=5,color='blue')
    plt.plot([last_a,a],[last_b,b],color='aqua')
    # 绘制图3中的回归直线
    plt.subplot(2, 2, 3)
    plt.plot(x, y)
    plt.plot(x, y, 'o')
    x_ = np.linspace(0, 1, 2)
    y_draw = a * x_ + b
    plt.plot(x_, y_draw)
    # 绘制图4的loss更新曲线
    all_loss.append(loss)
    all_step.append(step)
    plt.subplot(2,2,4)
    plt.plot(all_step,all_loss,color='orange')
    plt.xlabel("step")
    plt.ylabel("loss")

    # print('a = %.3f,b = %.3f' % (a,b))
    last_a = a
    last_b = b

    #-- 参数更新
    n[0] = n[0]+np.square(all_da)
    n[1] = n[1]+np.square(all_db)
    rate_new = rate/(np.sqrt(n + epsilon))
    print('rate_new a:',rate_new[0],' b:',rate_new[1])
    a = a - (rate/(np.sqrt(n[0] + epsilon)))*all_da
    b = b - (rate/(np.sqrt(n[1] + epsilon)))*all_db

    if step%1 == 0:
        print("step: ", step, " loss: ", loss)
        plt.show()
        plt.pause(0.01)
plt.show()
plt.pause(99999999999)