获取pytorch数据集的子集

Question

我有一个网络,我想在一些数据集上训练(例如,说CIFAR10).我可以通过创建数据加载器对象

trainset = torchvision.datasets.CIFAR10(root='./data', train=True,
                                        download=True, transform=transform)
trainloader = torch.utils.data.DataLoader(trainset, batch_size=4,
                                          shuffle=True, num_workers=2)

我的问题如下:假设我想进行几次不同的训练迭代.假设我首先想要在奇数位置的所有图像上训练网络,然后在偶数位置的所有图像上训练网络,依此类推.为此,我需要能够访问这些图像.不幸的是,它似乎trainset不允许这种访问.也就是说,尝试做trainset[:1000]或更多一般trainset[mask]会抛出错误.

我可以做

trainset.train_data=trainset.train_data[mask]
trainset.train_labels=trainset.train_labels[mask]

然后

trainloader = torch.utils.data.DataLoader(trainset, batch_size=4,
                                              shuffle=True, num_workers=2)

但是,这将迫使我在每次迭代中创建完整数据集的新副本(因为我已经更改,trainset.train_data所以我需要重新定义trainset).有没有办法避免它？

理想情况下,我希望有一些"等同"的东西

trainloader = torch.utils.data.DataLoader(trainset[mask], batch_size=4,
                                              shuffle=True, num_workers=2)

Answer 1

torch.utils.data.Subset更容易，支持shuffle，并且不需要编写自己的采样器：

import torchvision
import torch

trainset = torchvision.datasets.CIFAR10(root='./data', train=True,
                                        download=True, transform=None)

evens = list(range(0, len(trainset), 2))
odds = list(range(1, len(trainset), 2))
trainset_1 = torch.utils.data.Subset(trainset, evens)
trainset_2 = torch.utils.data.Subset(trainset, odds)

trainloader_1 = torch.utils.data.DataLoader(trainset_1, batch_size=4,
                                            shuffle=True, num_workers=2)
trainloader_2 = torch.utils.data.DataLoader(trainset_2, batch_size=4,
                                            shuffle=True, num_workers=2)

Answer 2

您可以为数据集加载器定义自定义采样器,避免重新创建数据集(只需为每个不同的采样创建一个新的加载器).

class YourSampler(Sampler):
    def __init__(self, mask):
        self.mask = mask

    def __iter__(self):
        return (self.indices[i] for i in torch.nonzero(self.mask))

    def __len__(self):
        return len(self.mask)

trainset = torchvision.datasets.CIFAR10(root='./data', train=True,
                                        download=True, transform=transform)

sampler1 = YourSampler(your_mask)
sampler2 = YourSampler(your_other_mask)
trainloader_sampler1 = torch.utils.data.DataLoader(trainset, batch_size=4,
                                          sampler = sampler1, shuffle=False, num_workers=2)
trainloader_sampler2 = torch.utils.data.DataLoader(trainset, batch_size=4,
                                          sampler = sampler2, shuffle=False, num_workers=2)

PS:你可以在这里找到更多信息:http://pytorch.org/docs/master/_modules/torch/utils/data/sampler.html#Sampler