import torch
import torch.utils.data as data
import torch.nn as nn
import torch.optim as optim
import torch.nn.functional as F
import matplotlib.pyplot as plt
import numpy as np
from torch.utils.data.dataset import T_co
# 定义获取数据集的类
class TestDataset(data.Dataset):
def __init__(self):
self.Data = np.array([[1, 2], [3, 4], [2, 1], [3, 4], [4, 5]])
self.label = np.array([0, 1, 0, 1, 2])
def __getitem__(self, index):
# 将numpy转换为Tensor
txt = torch.from_numpy(self.Data[index])
label = torch.tensor(self.label[index])
return txt, label
def __len__(self):
return len(self.Data)
# 获取数据集中的数据
test = TestDataset()
# print(test[2]) # 相当于调用__getitem__(2)
# print(test.__len__())
'''
结果:
(tensor([2, 1]), tensor(0))
5
以上数据以tuple返回,每次只返回一个样本。实际上,Dateset只负责数据的抽取,调
用一次__getitem__只返回一个样本。如果希望批量处理(batch),还要同时进行shuffle和
并行加速等操作,可选择DataLoader。
'''
test_loader = data.DataLoader(dataset=test, batch_size=2, shuffle=True)
for step, train_data in enumerate(test_loader):
print('step:{}'.format(step))
x, y = train_data
print(f'x:{x}')
print(f'y:{y}')
'''
结果:
step:0
x:tensor([[4, 5],
[2, 1]])
y:tensor([2, 0])
step:1
x:tensor([[1, 2],
[3, 4]])
y:tensor([0, 1])
step:2
x:tensor([[3, 4]])
y:tensor([1])
'''
torch.utils.data简介
最新推荐文章于 2024-04-18 12:51:30 发布