adult-dataset

A PyTorch dataset wrapper for the Adult (Census Income) dataset. Adult is a popular dataset in machine learning fairness research.

This package provides the adult.Adult class: atorch.utils.data.Datasets loading and, optionally, downloading the Adult dataset. It can be used like the MNIST dataset in torchvision.

Beyond adult.Adult, this package also provides adult.AdultRaw, which works just as adult.Adult, but does not standardize the features in the dataset and does not apply one-hot encoding.

Installation

pip install adult-dataset

Basic Usage

from adult import Adult

# load (if necessary, download) the Adult training dataset 
train_set = Adult(root="datasets", download=True)
# load the test set
test_set = Adult(root="datasets", train=False, download=True)

inputs, target = train_set[0]  # retrieve the first sample of the training set

# iterate over the training set
for inputs, target in iter(train_set):
    ...  # Do something with a single sample

# use a PyTorch data loader
from torch.utils.data import DataLoader

loader = DataLoader(test_set, batch_size=32, shuffle=True)
for epoch in range(100):
    for inputs, targets in iter(loader):
        ...  # Do something with a batch of samples

Advanced Usage

Turn off status messages while downloading the dataset:

Adult(root=..., output_fn=None)

Use the logging module for logging status messages while downloading the dataset instead of placing the status messages on sys.stdout.

import logging

Adult(root=..., output_fn=logging.info)