Stacked MNIST Dataset

The Stacked MNIST dataset is derived from the standard MNIST dataset with an increased number of discrete modes. 240,000 RGB images in the size of 32×32 are synthesized by stacking three random digit images from MNIST along the color channel, resulting in 1,000 explicit modes in a uniform distribution corresponding to the number of possible triples of digits.

堆叠 MNIST 数据集源于标准的 MNIST 数据集，并增加了离散模态的数量。通过沿颜色通道堆叠 MNIST 中的三个随机数字图像，生成了 240,000 张 32×32 大小的 RGB 图像，从而在均匀分布中产生了 1,000 种显式模态，这些模态对应于可能的数字三元组的数量。