import torch from torch.utils.data import Sampler class RepeatedSampler(Sampler): def __init__(self, data_source, num_repeats): """ Arguments: - data_source (Dataset): The dataset to sample from. - num_repeats (int): The number of times each image should be repeated. """ self.data_source = data_source self.num_repeats = num_repeats def __len__(self): # The length of the sampler is the number of samples multiplied by the number of repeats return len(self.data_source) * self.num_repeats def __iter__(self): # Repeat each index 'num_repeats' times in place without shuffling indices = torch.arange(len(self.data_source)).repeat_interleave(self.num_repeats) return iter(indices.tolist())