06. Logistic Regression

import torch
import numpy as np
import matplotlib.pyplot as plt

x_data=torch.Tensor([[1.], [2.], [3.]])
y_data=torch.Tensor([[0.], [0.], [1.]])

class LogisticReg(torch.nn.Module):
    def __init__(self):
        super(LogisticReg,self).__init__()
        self.linear=torch.nn.Linear(1,1)
    def forward(self,x):
        return torch.sigmoid(self.linear(x))
model = LogisticReg()


cirter=torch.nn.BCELoss()
optimizer=torch.optim.SGD(model.parameters(),lr=0.1)


E=500
for epoch in range(E):
    y_pred=model(x_data)
    loss=cirter(y_pred,y_data)
    print("epoch{},loss={},w={},b={}".format(epoch,loss.item(),model.linear.weight.item(),model.linear.bias.data.item()))

    optimizer.zero_grad()
    loss.backward()
    optimizer.step()

x_test=np.linspace(0,10,200)
x_t=torch.Tensor(x_test).view(200,1)
y_t=model(x_t)
y=y_t.data.numpy()
plt.plot(x_test,y)
plt.show()