基于pytorch的神经网络训练多标记数据

from sklearn.datasets import make_classification
import torch
from torch.autograd import Variable
from torch import nn
from sklearn.utils import shuffle
import numpy as np


X, y1 = make_classification(n_samples=10, n_features=100, n_informative=30, n_classes=3, random_state=1)
y2 = shuffle(y1, random_state=1)  # 分类结果随机排序
y3 = shuffle(y1, random_state=2)  # 分类结果随机排序
Y = np.vstack((y1, y2, y3)).T  # 多种分类结果组合成

from sklearn.preprocessing import OneHotEncoder
enc = OneHotEncoder()
a = enc.fit_transform(Y).toarray()
print(a)




X_train = Variable(torch.Tensor(X).float())
Y_train = Variable(torch.Tensor(a).float())

#X_test = torch.Tensor(X_test).float()

class simpleNet(nn.Module):
    def __init__(self, in_dim, n_hidden_1, n_hidden_2, out_dim):
        super(simpleNet, self).__init__()
        self.layer1 = nn.Sequential(nn.Linear(in_dim, n_hidden_1), nn.BatchNorm1d(n_hidden_1), nn.ReLU(True))
        self.layer2 = nn.Sequential(nn.Linear(n_hidden_1, n_hidden_2), nn.BatchNorm1d(n_hidden_2), nn.Sigmoid())
        self.layer3 = nn.Sequential(nn.Linear(n_hidden_2, out_dim))
    def forward(self, x):
        x = self.layer1(x)
        x = self.layer2(x)
        x = self.layer3(x)
        return x

model = simpleNet(100, 64, 32, 9)
loss_func = nn.MultiLabelSoftMarginLoss()
opt = torch.optim.Adam(model.parameters(), lr=0.1)

loss_count = []

for epoch in range(1):
    for i in range(200):
        batch_x = X_train            # torch.Size([128, 1, 28, 28])
        batch_y = Y_train            # torch.Size([128])

        out = model(batch_x)              # torch.Size([128,10])
        # 获取损失
        loss = loss_func(out, batch_y)
        # 使用优化器优化损失
        opt.zero_grad()   # 清空上一步残余更新参数值
        loss.backward()   # 误差反向传播，计算参数更新值
        opt.step()        # 将参数更新值施加到net的parmeters上

        loss_count.append(loss)
        print('{}:\t'.format(i), loss.item())
        torch.save(model, r'C:\Users\Administrator\PycharmProjects\data\cnn')

print(model(X_train))