Skip to main content

Leave one out encoding of categorical features

Project description

leave-one-out-encoder

Leave one out coding for categorical features

See the source for this project here: https://github.com/welfare520/leave-one-out-encoder.

Getting Started

Installing

$ pip install loo_encoder

Example

Fit encoder according to X and y, and then transform it.

from loo_encoder.encoder import LeaveOneOutEncoder
import pandas as pd
import numpy as np


enc = LeaveOneOutEncoder(cols=['gender', 'country'], handle_unknown='impute', sigma=0.02, random_state=42)

X = pd.DataFrame(
    {
        "gender": ["male", "male", "female", "male"],
        "country": ["Germany", "USA", "USA", "UK"],
        "clicks": [10, 33, 47, 21]
    }
)

y = np.array([1, 2, 3, 4])

df_train = enc.fit_transform(X=X, y=y, sample_weight=X['clicks'])

Perform the transformation to new categorical data.

X_val = pd.DataFrame(
    {
        "gender": ["unknown", "male", "female", "male"],
        "country": ["Germany", "USA", "Germany", "Japan"]
    }
)

df_test = enc.transform(X=X_val)

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

loo_encoder-0.0.7.tar.gz (3.4 kB view hashes)

Uploaded Source

Built Distribution

loo_encoder-0.0.7-py3-none-any.whl (3.8 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page