lesson-05-Logistic-Regression

import torch as tt
import torch.nn as nn
import matplotlib.pyplot as plt
import numpy as np
tt.manual_seed(10)

<torch._C.Generator at 0x6858f10>

机器学习模型训练步骤

迭代训练

• 数据
• 模型
• loss_fn
• optimizer

# step 1/5 生成数据
sample_nums = 100
mean_value = 1.7
bias = 1
n_data = tt.ones(sample_nums, 2)
x0 = tt.normal(mean_value * n_data, 1) + bias
# class 0 data shape(100,2)
y0 = tt.zeros(sample_nums)                         
# 类别0 标签 shape=(100, 1)
x1 = tt.normal(-mean_value * n_data, 1) + bias     
# 类别1 数据 shape=(100, 2)
y1 = tt.ones(sample_nums)                          
# 类别1 标签 shape=(100, 1)

train_x = tt.cat((x0, x1), 0)
train_y = tt.cat((y0, y1), 0)

# step 2/5 选择模型
class LR(nn.Module):
    def __init__(self):
        super(LR, self).__init__()
        self.features = nn.Linear(2,1)# y = f(wx + b)
        self.sig = nn.Sigmoid()# f(x) = 1/(1+exp(-x))
    def forward(self, x):
        x = self.features(x)
        x = self.sig(x)
        return x
    
lr_net = LR()

# step 3/5 选择损失函数
loss_fn = nn.BCELoss()

# step 4/5 选择优化器
lr = 0.01 # learning rate
optimizer = tt.optim.SGD(lr_net.parameters(), lr = lr, momentum = 0.9)

# step 5/5 模型训练
epoch = 1000

for iteration in range(epoch):
    # forward
    y_pred = lr_net(train_x)
    
    # calculate loss
    loss = loss_fn(y_pred.squeeze(), train_y)
    
    # backward
    loss.backward()
    
    # update parameters
    optimizer.step()
    
    # clear gradient
    optimizer.zero_grad()
    
    # plot
    
    if iteration % 50 ==0:
        
        mask = y_pred.ge(0.5).float().squeeze()
        # classify with threshold == 0.5
        correct = (mask == train_y).sum()
        # calculate correct number
        acc = correct.item() / train_y.size(0)
        # calculate accuracy
        
        plt.scatter(x0.data.numpy()[:, 0], x0.data.numpy()[:, 1], c='r', label='class 0')
        plt.scatter(x1.data.numpy()[:, 0], x1.data.numpy()[:, 1], c='b', label='class 1')        
        

        w0, w1 = lr_net.features.weight[0]
        w0, w1 = float(w0.item()), float(w1.item())
        plot_b = float(lr_net.features.bias[0].item())
        plot_x = np.arange(-6, 6, 0.1)
        plot_y = (-w0 * plot_x - plot_b) / w1
        
        plt.xlim(-5, 7)
        plt.ylim(-7, 7)
        plt.plot(plot_x, plot_y)

        plt.text(-5, 5, 'Loss=%.4f' % loss.data.numpy(), fontdict={'size': 20, 'color': 'red'})
        plt.title("Iteration: {}\nw0:{:.2f} w1:{:.2f} b: {:.2f} accuracy:{:.2%}".format(iteration, w0, w1, plot_b, acc))
        plt.legend()

        plt.show()
        plt.pause(0.5)

        if acc > 0.99:
            break