import torch import torch.nn.functional as F import numpy as np import
matplotlib.pyplot as plt x_data = torch.Tensor([[1.0], [2.0], [3.0]]) y_data =
torch.Tensor([[0], [0], [1]]) # 现在的输入y为分类，故只为0或1 #
----------------------------------------------------------------准备数据集 class
LogisticRegressionModel(torch.nn.Module): def __init__(self):
super(LogisticRegressionModel, self).__init__() self.linear =
torch.nn.Linear(1, 1) # 先用linear进行线性变换，输入维度为1，输出维度为1 #
sigmoid函数中没有参数，故不需要在构造函数中初始化，因为没有参数可供训练，直接调用即可 def forward(self, x): y_pred =
F.torch.sigmoid(self.linear(x)) # 再用sigmoid函数处理线性变换后的结果，作为最后的输出 return y_pred
model = LogisticRegressionModel() #
--------------------------------------------------------------设计模型 criterion =
torch.nn.BCELoss(reduction='sum') # BCE损失函数 optimizer =
torch.optim.SGD(model.parameters(), lr=0.01) #
---------------------------------------------------------------构造损失函数和优化器 for
epoch in range(1000): y_pred = model(x_data) loss = criterion(y_pred, y_data)
optimizer.step() #
--------------------------------------------------------------训练周期 x =
np.linspace(0, 10, 200) x_t = torch.Tensor(x).view((200, 1)) y_t = model(x_t) y
= y_t.data.numpy() plt.plot(x, y) plt.plot([0, 10], [0.5, 0.5], c='r')
plt.xlabel('Hours') plt.ylabel('Probability of Pass') plt.grid() plt.show()

GitHub

Gitee