A library of recommender systems metrics for big data
Project description
recmetrics-pyspark: recommender systems metrics for big data
recmetrics-pyspark obtains the most relevant internal metrics for items recommendations from pySpark DataFrames. It efficiently handles huge amounts of data. Most routines are adapted from the recmetrics library which works with pandas DataFrames.
DISCLAIMER: recmetrics-pyspark is not affiliated nor endorsed by recmetrics or its authors. Some routines have been adapted from recmetrics to work with pySpark DataFrames and/or to handle bigger datasets. Therefore, some chunks of code have been copied verbatim, and functions and parameters names have been kept the same (as much as possible) for better usability.
Furthermore, if you are dealing with small datasets, we recommend to use the recmetrics library (https://github.com/statisticianinstilettos/recmetrics) instead, as it most efficiently handles smaller datasets.
Where to get it
The source code is currently hosted on GitHub at: https://github.com/camiloakv/recmetrics-pyspark
Binary installers for the latest released version are available at the Python Package Index (PyPI).
pip install recmetrics-pyspark
Available metrics as of version 0.0.1:
long_tail_plot
coverage
- Novelty:
novelty_refac
A small refactoring of recmetrics' implementation.novelty_pandas
Similar implementation to novelty_refac but using pandas DataFrames as inputsnovelty
pySpark implementation
personalization
intra_list_similarities
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 recmetrics_pyspark-0.0.1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 944a90aaa83a88e14c85355ea3cb0412da24f1ada01d44b19ac303f2a29de843 |
|
MD5 | 5952c952522bf612bdff4c88f206c6d8 |
|
BLAKE2b-256 | bc2ed26654d686f3307f34ad8bbfa1919f8d05132a66707554f362544b71438d |