Source code for libuplift.datasets.Lazada
"""Lazada E-comerce dataset.
See https://github.com/kailiang-zhong/DESCN/tree/main/data/Lazada_dataset/ for details.
"""
import numpy as np
from .base import _fetch_remote_csv
from .base import RemoteFileMetadata
ARCHIVE_TRAIN = RemoteFileMetadata(
filename="Lazada_train.csv.gz",
url=('https://github.com/jszymon/uplift_sklearn_data/'
'releases/download/Lazada/Lazada_trainset.csv.gz'),
checksum=('5a46ec368f9e8397267e818447ea7f3d'
'239a671731ca98c0ae9f17c19ab5a469'))
ARCHIVE_TEST = RemoteFileMetadata(
filename="Lazada_test.csv.gz",
url=('https://github.com/jszymon/uplift_sklearn_data/'
'releases/download/Lazada/Lazada_testset.csv.gz'),
checksum=('a6d8832a71e8f0b6d4e8858d91ce7de9'
'd8e0f2f58aa7208c78b1e0f7d6ba9c94'))
[docs]
def fetch_Lazada(version="test", data_home=None, download_if_missing=True,
random_state=None, shuffle=False,
categ_as_strings=False, return_X_y=False,
as_frame=False):
"""Load the Lazada e-comerce dataset.
Download it if necessary.
There is a training and test set available. The test dataset
comes from a randomized experiment, in the training set assignment
is biased.
Details of the dataset can be found in the following paper:
https://arxiv.org/pdf/2207.09920.
The license is available `here <https://github.com/jszymon/uplift_sklearn_data/releases/download/Lazada/Lazada_LICENSE>`__
Parameters
----------
version : string, optional
Specify whether to return training ('train') or testing
('test') dataset. Test dataset comes from a randomized
experiment, in the training set assignment is biased.
data_home : string, optional
Specify another download and cache folder for the datasets. By default
all scikit-learn data is stored in '~/scikit_learn_data' subfolders.
download_if_missing : boolean, default=True
If False, raise a IOError if the data is not locally available
instead of trying to download the data from the source site.
random_state : int, RandomState instance or None (default)
Determines random number generation for dataset shuffling. Pass an int
for reproducible output across multiple function calls.
shuffle : bool, default=False
Whether to shuffle dataset.
categ_as_strings : bool, default=False
Whether to return categorical variables as strings.
return_X_y : boolean, default=False.
If True, returns ``(data.data, data.target)`` instead of a Bunch
object.
as_frame : boolean, default=False
If True features are returned as pandas DataFrame. If False
features are returned as object or float array. Float array
is returned if all features are floats.
Returns
-------
dataset : dict-like object with the following attributes:
dataset.data : numpy array
Each row corresponds to the features in the dataset.
dataset.target_visit : numpy array
Each value is 1 if website visit occurred 0 otherwise.
dataset.target_conversion : numpy array
Each value is 1 if purchase occurred 0 otherwise.
dataset.target_spend : numpy array
Each value corresponds to the amount of money spent.
dataset.DESCR : string
Description of the Hillstrom dataset.
(data, target_conversion, target_visit, target_exposure) : tuple if
``return_X_y`` is True
"""
if version == "train":
ARCHIVE = ARCHIVE_TRAIN
file_name = "Lazada_train"
elif version == "test":
ARCHIVE = ARCHIVE_TEST
file_name = "Lazada_test"
else:
raise ValueError(f"Wrong version ({version}) of Lazada dataset"
"requested.\nValid choices are `train' or `test'")
# dictionaries
treatment_values = ['control', 'treated']
categ_values = dict()
# attribute descriptions
treatment_descr = [("treatment", np.int32, "is_treat")]
target_descr = [("target", np.int32, "label")]
feature_descr = [(f"f{i}", float) for i in range(83)]
ret = _fetch_remote_csv(ARCHIVE, file_name,
feature_attrs=feature_descr,
treatment_attrs=treatment_descr,
target_attrs=target_descr,
categ_as_strings=categ_as_strings,
return_X_y=return_X_y, as_frame=as_frame,
download_if_missing=download_if_missing,
random_state=random_state, shuffle=shuffle,
total_attrs=85, all_num=True
)
if not return_X_y:
ret.descr = __doc__
return ret