import numpy as np
import torch
# 糖尿病预测研判
# 1. 处理数据
xy = np.loadtxt('diabetes.csv.gz', delimiter=',', dtype=np.float32)
x_data = torch.from_numpy(xy[:, :-1])
y_data = torch.from_numpy(xy[:, [-1]])
# 2. 建立模型
# 2. 设计模型 继承自torch.nn.Module
class Model(torch.nn.Module):
def __init__(self):
super(Model, self).__init__()
self.linear1 = torch.nn.Linear(8, 6)
self.linear2 = torch.nn.Linear(6, 4)
self.linear3 = torch.nn.Linear(4, 1)
self.sigmoid = torch.nn.Sigmoid()
# 覆盖父类方法
def forward(self, x):
"""预测结果 计算loss"""
x = self.sigmoid(self.linear1(x))
x = self.sigmoid(self.linear2(x))
x = self.sigmoid(self.linear3(x))
return x
model = Model()
# 3.0
# 损失函数
criterion = torch.nn.BCELoss(size_average=False)
# 优化器
optimizer = torch.optim.SGD(model.parameters(), lr=0.01)
# 4. 训练周期
for epoch in range(1000000):
# 调用forward函数
y_pred = model(x_data)
# 计算损失 loss为一个标量
loss = criterion(y_pred, y_data)
# print(epoch, loss)
# 清空本次计算数据 梯度清零
optimizer.zero_grad()
# 反向传播
loss.backward()
# 更新参数
optimizer.step()
print(epoch, loss.item())