import sklearn.datasets
MLCOMP_DIR = r"~/my/data/"
data = sklearn.datasets.load_mlcomp("20news-18828", mlcomp_root=MLCOMP_DIR)

ValueError: Could not find dataset with metadata line: name: 20news-18828

The dataset used in this example is the 20 newsgroups dataset and should be
 downloaded from the http://mlcomp.org (free registration required):

  http://mlcomp.org/datasets/379