划分数据1

import math
import numpy as np
import torch
from torch import nn
import matplotlib.pyplot as plt
import torchvision.datasets as datasets
from torch.utils import data
from torchvision import transforms
from torch.utils.data import Dataset
import torchvision


train_data=datasets.CIFAR10(root="../data", train=True, transform=transforms.ToTensor(), download=True)
test_data=datasets.CIFAR10(root="../data", train=False, transform=transforms.ToTensor(), download=True)
t0=[ i for i, x in enumerate(train_data.targets) if x == 0]
t1=[ i for i, x in enumerate(train_data.targets) if x == 1]
t2=[ i for i, x in enumerate(train_data.targets) if x == 2]
t3=[ i for i, x in enumerate(train_data.targets) if x == 3]
t4=[ i for i, x in enumerate(train_data.targets) if x == 4]
t5=[ i for i, x in enumerate(train_data.targets) if x == 5]
t6=[ i for i, x in enumerate(train_data.targets) if x == 6]
t7=[ i for i, x in enumerate(train_data.targets) if x == 7]
t8=[ i for i, x in enumerate(train_data.targets) if x == 8]
t9=[ i for i, x in enumerate(train_data.targets) if x == 9]


label0=[]
for i in t0:
    label0.append(train_data.targets[i])
label1=[]
for i in t1:
    label1.append(train_data.targets[i])
label2=[]
for i in t2:
    label2.append(train_data.targets[i])
label3=[]
for i in t3:
    label3.append(train_data.targets[i])
label4=[]
for i in t4:
    label4.append(train_data.targets[i])
label5=[]
for i in t5:
    label5.append(train_data.targets[i])
label6=[]
for i in t6:
    label6.append(train_data.targets[i])
label7=[]
for i in t7:
    label7.append(train_data.targets[i])
label8=[]
for i in t8:
    label8.append(train_data.targets[i])
label9=[]
for i in t9:
    label9.append(train_data.targets[i])

data0=np.array([])
for i in t0:
    data0=np.append(data0,train_data.data[i])
data1=np.array([])
for i in t1:
    data1=np.append(data1,train_data.data[i])
data2=np.array([])
for i in t2:
    data2=np.append(data2,train_data.data[i])
data3=np.array([])
for i in t3:
    data3=np.append(data3,train_data.data[i])
data4=np.array([])
for i in t4:
    data4=np.append(data4,train_data.data[i])
data5=np.array([])
for i in t5:
    data5=np.append(data5,train_data.data[i])
data6=np.array([])
for i in t6:
    data6=np.append(data6,train_data.data[i])
data7=np.array([])
for i in t7:
    data7=np.append(data7,train_data.data[i])
data8=np.array([])
for i in t8:
    data8=np.append(data8,train_data.data[i])
data9=np.array([])
for i in t9:
    data9=np.append(data9,train_data.data[i])

data0=data0.reshape(len(label0),32,32,3)
data1=data1.reshape(len(label1),32,32,3)
data2=data2.reshape(len(label2),32,32,3)
data3=data3.reshape(len(label3),32,32,3)
data4=data4.reshape(len(label4),32,32,3)
data5=data5.reshape(len(label5),32,32,3)
data6=data6.reshape(len(label6),32,32,3)
data7=data7.reshape(len(label7),32,32,3)
data8=data8.reshape(len(label8),32,32,3)
data9=data9.reshape(len(label9),32,32,3)

class DatasetXY(Dataset):
    def __init__(self, x, y):
        self._x = x
        self._y = y
        self._len = len(x)

    def __getitem__(self, item):  # 每次循环的时候返回的值
        return self._x[item], self._y[item]

    def __len__(self):
        return self._len

dataset0= DatasetXY(data0,label0)
dataset1= DatasetXY(data1,label1)
dataset2= DatasetXY(data2,label2)
dataset3= DatasetXY(data3,label3)
dataset4= DatasetXY(data4,label4)
dataset5= DatasetXY(data5,label5)
dataset6= DatasetXY(data6,label6)
dataset7= DatasetXY(data7,label7)
dataset8= DatasetXY(data8,label8)
dataset9= DatasetXY(data9,label9)

train1_iter=data.DataLoader(dataset0,batch_size=32,num_workers=0)