cookie-datasets
The cookie-datasets library provides various DataFrame readers for Apache Spark.
Supported Datasets
IRIS
MNIST
CIFAR
A few variants are supported:
CIFAR-10 (Binary)
CIFAR-100 (Binary)
Labeled Faces in the Wild (LFW)
Coming Soon - 0.2
Data Formats
A few of the datasets are based on common formats. In those cases, the reader is available.
LIBSVM
Example data: www.cs.ie.ntu.edu.tw
Image Directory Format
Coming Soon - 0.2