Pytorch处理多维特征的输入——up主：刘二大人《PyTorch深度学习实践》

b站up主：刘二大人《PyTorch深度学习实践》
教程： https://www.bilibili.com/video/BV1Y7411d7Ys?p=6&vd_source=715b347a0d6cb8aa3822e5a102f366fe
$$\begin{gathered} 三层模型:torch.nn.Linear \\ 激活函数：ReLU+sigmoid \\ 交叉熵损失函数：nn.BCELoss \\ 优化器：optim.Adam \\ 数据集：diabetes.csv.gz(在b站课程的下载里有) \end{gathered}$$

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

#数据集
xy = np.loadtxt('/content/drive/MyDrive/pytorch_lab/data/diabetes.csv.gz', delimiter=',', dtype = np.float32)

x_data = torch.from_numpy(xy[:,:-1]) #取除了最后一列的所有数据
y_data = torch.from_numpy(xy[:,[-1]]) #取最后一列

epoch_p = []
loss_p = []

#设计模型
class Model(torch.nn.Module):
  def __init__(self):
    super(Model, self).__init__()
    self.linear1 = torch.nn.Linear(8, 6) #输入feature为8维（8列 N*8），输出为6维（6列 N*6）
    self.linear2 = torch.nn.Linear(6, 4)
    self.linear3 = torch.nn.Linear(4, 1)
    self.activate = torch.nn.ReLU() #ReLU为层结构，需要添加在Model中才能使用，torch.nn.functional.relu为函数可以直接使用

  def forward(self, x):
    x = self.activate(self.linear1(x))
    x = self.activate(self.linear2(x))
    y_pred = torch.sigmoid(self.linear3(x))
    return y_pred

model = Model()

#构造损失函数和优化器
criterion = torch.nn.BCELoss() #交叉熵损失函数，计算概率损失
optimizer = torch.optim.Adam(model.parameters(), lr = 0.01)

#train
for epoch in range(1000):
  y_pred = model(x_data)
  loss = criterion(y_pred, y_data)
  loss_p.append(loss.item())
  epoch_p.append(epoch)
  print("第{}个Epoch，loss = {:.6f}".format(epoch, loss))

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



plt.figure()
plt.plot(epoch_p, loss_p, c = 'b')
plt.xlabel('Epoch')
plt.ylabel('loss')
plt.show()