Differences with torch.utils.data.RandomSampler

View Source On Gitee

torch.utils.data.RandomSampler

class torch.utils.data.RandomSampler(data_source, replacement=False, num_samples=None, generator=None)

For more information, see torch.utils.data.RandomSampler.

mindspore.dataset.RandomSampler

class mindspore.dataset.RandomSampler(replacement=False, num_samples=None)

For more information, see mindspore.dataset.RandomSampler.

Differences

PyTorch: Samples elements randomly, random generator can be set manually.

MindSpore: Samples elements randomly, random generator is not supported.

Categories

Subcategories

PyTorch

MindSpore

Difference

Parameter

Parameter1

data_source

-

Dataset object to be sampled. MindSpore does not need this parameter.

Parameter2

replacement

replacement

-

Parameter3

num_samples

num_samples

-

Parameter4

generator

-

Specifies sampling logic. MindSpore uses global random sampling.

Code Example

import torch
from torch.utils.data import RandomSampler

torch.manual_seed(1)

class MyMapDataset(torch.utils.data.Dataset):
    def __init__(self):
        super(MyMapDataset).__init__()
        self.data = [i for i in range(4)]
    def __getitem__(self, index):
        return self.data[index]
    def __len__(self):
        return len(self.data)

ds = MyMapDataset()
sampler = RandomSampler(ds, num_samples=2, replacement=True)
dataloader = torch.utils.data.DataLoader(ds, sampler=sampler)

for data in dataloader:
    print(data)
# Out:
# tensor([2])
# tensor([0])
import mindspore as ms
from mindspore.dataset import RandomSampler

ms.dataset.config.set_seed(3)

class MyMapDataset():
    def __init__(self):
        super(MyMapDataset).__init__()
        self.data = [i for i in range(4)]
    def __getitem__(self, index):
        return self.data[index]
    def __len__(self):
        return len(self.data)

ds = MyMapDataset()
sampler = RandomSampler(num_samples=2, replacement=True)
dataloader = ms.dataset.GeneratorDataset(ds, column_names=["data"], sampler=sampler)

for data in dataloader:
    print(data)
# Out:
# [Tensor(shape=[], dtype=Int64, value= 2)]
# [Tensor(shape=[], dtype=Int64, value= 0)]