Statistics tools for ML models and deployment
Project description
- Author:
ZeD@UChicago <zed.uchicago.edu>
- Description:
Tools for ML statistics
- Documentation:
- Example:
https://github.com/zeroknowledgediscovery/zedstat/blob/master/examples/example1.ipynb
Usage:
from zedstat import zedstat
zt=zedstat.processRoc(df=pd.read_csv('roc.csv'),
order=3,
total_samples=100000,
positive_samples=100,
alpha=0.01,
prevalence=.002)
zt.smooth(STEP=0.001)
zt.allmeasures(interpolate=True)
zt.usample(precision=3)
zt.getBounds()
print(zt.auc())
# find the high precision and high sensitivity operating points
zt.operating_zone(LRminus=.65)
rf0,txt0=zt.interpret(fpr=zt._operating_zone.fpr.values[0],number_of_positives=10)
rf1,txt1=zt.interpret(fpr=zt._operating_zone.fpr.values[1],number_of_positives=10)
display(zt._operating_zone)
print('high precision operation:\n','\n '.join(txt0))
print('high recall operation:\n','\n '.join(txt1))
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
zedstat-0.0.132.tar.gz
(80.2 kB
view hashes)
Built Distribution
zedstat-0.0.132-py3-none-any.whl
(168.7 kB
view hashes)
Close
Hashes for zedstat-0.0.132-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 72d171597ed6542e61e725587ef1968bd25af560946d70e791a71fe18dcd85a1 |
|
MD5 | 203b39f0e6a50251fe2de57d9482d227 |
|
BLAKE2b-256 | d288d7f05174498517eece938dfa15fe1e0393b76f84c4dea89bd843716b3759 |