Google I/O is a wrap! Catch up on TensorFlow sessions View sessions

hillstrom

  • Description:

This dataset contains 64,000 customers who last purchased within twelve months. The customers were involved in an e-mail test.

  1. 1/3 were randomly chosen to receive an e-mail campaign featuring Mens merchandise.
  2. 1/3 were randomly chosen to receive an e-mail campaign featuring Womens merchandise.
  3. 1/3 were randomly chosen to not receive an e-mail campaign.

During a period of two weeks following the e-mail campaign, results were tracked. The task is to tell the world if the Mens or Womens e-mail campaign was successful.

Split Examples
'train' 64,000
  • Feature structure:
FeaturesDict({
    'channel': Text(shape=(), dtype=tf.string),
    'conversion': tf.int64,
    'history': tf.float32,
    'history_segment': Text(shape=(), dtype=tf.string),
    'mens': tf.int64,
    'newbie': tf.int64,
    'recency': tf.int64,
    'segment': Text(shape=(), dtype=tf.string),
    'spend': tf.float32,
    'visit': tf.int64,
    'womens': tf.int64,
    'zip_code': Text(shape=(), dtype=tf.string),
})
  • Feature documentation:
Feature Class Shape Dtype Description
FeaturesDict
channel Text tf.string
conversion Tensor tf.int64
history Tensor tf.float32
history_segment Text tf.string
mens Tensor tf.int64
newbie Tensor tf.int64
recency Tensor tf.int64
segment Text tf.string
spend Tensor tf.float32
visit Tensor tf.int64
womens Tensor tf.int64
zip_code Text tf.string
  • Supervised keys (See as_supervised doc): ({'channel': 'channel', 'history': 'history', 'mens': 'mens', 'newbie': 'newbie', 'recency': 'recency', 'segment': 'segment', 'womens': 'womens', 'zip_code': 'zip_code'}, 'visit')

  • Figure (tfds.show_examples): Not supported.

  • Examples (tfds.as_dataframe):

  • Citation:
@article{entryhillstrom,
  title={Hillstrom’s MineThatData Email Analytics Challenge},
  author={ENTRY, WINNING}
}