出自up主 刘二大人 06.逻辑斯蒂回归_哔哩哔哩_bilibili
BCE二类交叉熵(Binary Cross-entropy)
附上代码:
import torch.nn.functional as F
import torch
import matplotlib.pyplot as plt
import numpy as np
x_data = torch.Tensor([[1.0],[2.0],[3.0]])
y_data = torch.Tensor([[0.0],[0.0],[1.0]])
class LogisticRegressionModel(torch.nn.Module):
def __init__(self):
super(LogisticRegressionModel, self).__init__()
self.liner = torch.nn.Linear(1, 1)
def forward(self, x):
y_pred = F.sigmoid(self.liner(x))
return y_pred
model = LogisticRegressionModel()
criterion = torch.nn.BCELoss(size_average=False)
optimizer = torch.optim.SGD(model.parameters(),lr=0.01)
for epoch in range(1000):
y_pred = model(x_data)
loss = criterion(y_pred, y_data)
print(epoch, loss.item())
optimizer.zero_grad()
loss.backward()
optimizer.step()
x_test = torch.Tensor([4.0])
y_test = model(x_test)
print("y_pred = ", y_test)
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("possiblie of pass")
plt.show()
解释:
- x = np.linspace(0, 10, 200):这行代码使用NumPy的linspace函数在0到10之间生成了200个等间隔的数值,然后将这些数值赋值给变量x。这意味着x是一个包含200个元素的NumPy数组,这些元素在0到10之间均匀分布。
- x_t = torch.Tensor(x).view((200, 1)):这行代码首先将NumPy数组x转换为PyTorch张量,然后使用view函数将其形状改变为(200, 1)。这意味着x_t是一个包含200行和1列的二维张量,其元素与x中的元素相同。
- 当你调用plt.grid()时,它会在图形的背景上添加水平和垂直的灰色网格线