【PyTorch基础教程7】多维特征input（学不会来打我啊）

# -*- coding: utf-8 -*-

"""

Created on Mon Oct 18 10:18:24 2021

@author: 86493

"""

import torch

import torch.nn as nn

import numpy as np

import matplotlib.pyplot as plt

# 这里的type不用double，特斯拉GPU才double

xy = np.loadtxt('diabetes.csv',

                delimiter = ' ',

                dtype = np.float32)

# 最后一列不要

x_data = torch.from_numpy(xy[: , : -1])

# [-1]则拿出来的是一个矩阵，去了中括号则拿出向量

y_data = torch.from_numpy(xy[:, [-1]])

losslst = []

class Model(nn.Module):

    def __init__(self):

        super(Model, self).__init__()

        self.linear1 = nn.Linear(9, 6)

        self.linear2 = nn.Linear(6, 4)

        self.linear3 = nn.Linear(4, 1)         

        # 外汇跟单gendan5.com上次logistic是调用nn.functional的Sigmoid

        self.sigmoid = nn.Sigmoid()

        # 这个也是继承Module,没有参数,比上次写法不容易出错

    def forward(self, x):

        x = self.sigmoid(self.linear1(x))

        x = self.sigmoid(self.linear2(x))

        x = self.sigmoid(self.linear3(x))

        return x

model = Model()

# 使用交叉熵作损失函数

criterion = nn.BCELoss(size_average = False)

optimizer = torch.optim.SGD(model.parameters(),

                            lr = 0.01)

# 训练，下面没有用mini-batch，后面讲dataloader再说

for epoch in range(10):

    y_predict = model(x_data)

    loss = criterion(y_predict, y_data)

    # 打印loss对象会自动调用__str__

    print(epoch, loss.item())

    losslst.append(loss.item())

    # 梯度清零后反向传播

    optimizer.zero_grad()

    loss.backward()

    # 更新权重

    optimizer.step()

# 画图

plt.plot(range(10), losslst)

plt.ylabel('Loss')

plt.xlabel('epoch')

plt.show()