A library for multiple imputaion of censored data
Project description
CensorFix
Please look at the website https://censor-fix.readthedocs.io/en/latest/
A library for multiple imputation of censored data.
This software performs multiple imputation on censored data using the probabilistic programming langauge (ppl) stan. For 1d imputation the library can fits a distribution to data to create imputations.
For missing data with many features the library can model the censored values using the other features in a round robin fashion
How to use:
The data needs to be in a pandas dataframe and the columns that need imputing must be specified. The data is assumed to have a normal distribution unless stated otherwise but there are options for the data to be uniform,t distributed or exponential. To make the data more normal you can try using sklearn.reprocessing.power_transform to make the data normal.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for censor_fix-0.0.2-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 548c33ab206e868b1a96422e755a63816d888abeef29a43af5a525500ad376d1 |
|
MD5 | 7d142a084508c14964729bf78c90bad0 |
|
BLAKE2b-256 | 7351610400f8df20f840dda886cc48f9d67a3b7b19eb6655790d3132323564f1 |