ucimlr 0.3.0

Easy access to datasets from the UCI Machine Learning Repository

An API to the UCI Machine Learning Repository

This is a python package to enable easy access to datasets in the UCI Machine Learning Repository. Note that this is not an official API. Any usage of datasets should be cited according to instructions in the UCI Machine Learning Repository.

The project is at an early alpha stage, so suggestion for changes or additions are very welcome.

Link to documentation.

Basic usage

from ucimlr import regression_datasets
abalone = regression_datasets.Abalone('dataset_folder')
print(abalone.type_)
>>> regression
print(len(abalone))
>>> 3341

Independent and dependent variables are accessed as numpy arrays:

print(abalone.x.shape)
>>> (3341, 10)
print(abalone.y.shape)
>>> (3341, 1)

Or by element access:

x, y = dataset[0]