sports-features
A library for processing sports features over a dataframe containing sports data.
Dependencies 🌐
Python 3.11.6:
- openskill
- pandas
- feature-engine
- tqdm
- scikit-learn
- geopy
- numpy
- pytest-is-running
- joblib
- timeseries-features
- textfeats
- scipy
- image-features
- requests-cache
Raison D'être 💭
sportsfeatures aims to process features relevant to predicting aspects of sporting games.
Architecture 📐
sportsfeatures is a functional library, meaning that each phase of feature extraction gets put through a different function until the final output. It contains some caching when the processing is heavy (such as skill processing). The features its computes are as follows:
- Process the player and teams skill levels using OpenSkill. This is an ELO like rating system giving a probability of win and loss.
- Compute the offensive efficiency of each team/player.
- Compute the time series values of the numeric features for each team/player over the various windows provided. This includes lag, count, sum, mean, median, var, std, min, max, skew, kurt, sem, rank.
- Compute the datetime features for any datetime columns.
- Remove the lookahead features.
Installation 📥
This is a python package hosted on pypi, so to install simply run the following command:
pip install sportsfeatures
or install using this local repository:
python setup.py install --old-and-unmanageable
Usage example 👀
The use of sportsfeatures is entirely through code due to it being a library. It attempts to hide most of its complexity from the user, so it only has a few functions of relevance in its outward API.
Generating Features
To generate features:
import datetime
import pandas as pd
from sportsfeatures.process import process
from sportsfeatures.identifier import Identifier
from sportsfeatures.entity_type import EntityType
df = ... # Your sports data
identifiers = [
Identifier(EntityType.TEAM, "teams/0/id", ["teams/0/kicks"], "teams/0"),
Identifier(EntityType.TEAM, "teams/1/id", ["teams/1/kicks"], "teams/1"),
]
df = process(df, identifiers, [datetime.timedelta(days=365), None], "dt")
This will produce a dataframe that contains the new sports related features.
License 📝
The project is available under the MIT License.
1 maintainer
Metadata
1 maintainer
Release files for sportsfeatures 0.0.120
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | |
|---|---|---|---|---|
| sportsfeatures-0.0.120.tar.gz | 25.4 kB | Without provenance | Details |
Release files / sportsfeatures-0.0.120.tar.gz
| Download URL | sportsfeatures-0.0.120.tar.gz |
|---|---|
| Size | 25.4 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
2a11dba1c47fde1a4f6e1951daa06a6c5b5e6cb35e403399805d103bf598d6c9
|
|
BLAKE2b-256 checksum How to use checksums |
1851f011a8fabdb204355ecf98d5e144ebf177118da49a5827048372a98648b6
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.11.6
|
Security for sportsfeatures 0.0.120
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
