Automate machine learning classification task report for Pak Zuherman
Project description
Classification Reportzr
Automate machine learning classification task report for Pak Zuherman
Install
pip install -U classification-reportzr
Test
pytest -v
Usage
Setting-up the experiment
from sklearn import datasets
from sklearn.svm import SVC
from classification_reportzr.reporterzr import Reporterzr
iris = datasets.load_iris()
samples, labels = iris.data[:-1], iris.target[:-1]
svc_kwargs = {'C':100.0, 'gamma':0.001}
svc_reporter = Reporterzr(SVC, svc_kwargs)
Run The Experiment
# `test_sizes` defaults to [0.1, ..., 0.9]
# `rep` defaults to 10
svc_reporter.run_experiment(samples, labels, test_sizes=[0.1, 0.2, 0.3], rep=7)
Get Accuracy Report
print(svc_reporter.report())
prints
Split: 10.0% test - 90.0% train
mean stdev accuracies
train 0.973429 0.006758 [0.978, 0.985, 0.963, 0.97, 0.978, 0.97, 0.97]
test 0.961714 0.033156 [0.933, 1.0, 1.0, 1.0, 0.933, 0.933, 0.933]
Split: 20.0% test - 80.0% train
mean stdev accuracies
train 0.965143 0.010343 [0.958, 0.966, 0.966, 0.983, 0.958, 0.95, 0.975]
test 0.952429 0.039326 [1.0, 0.967, 1.0, 0.9, 0.9, 0.967, 0.933]
Split: 30.0% test - 70.0% train
mean stdev accuracies
train 0.976571 0.009897 [0.971, 0.962, 0.99, 0.971, 0.99, 0.971, 0.981]
test 0.942857 0.026368 [0.933, 0.978, 0.911, 0.978, 0.911, 0.933, 0.956]
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
Close
Hashes for classification-reportzr-0.0.1b3.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | 884aa28ceea0d0804f394557bd19a3e356080c344a271d541ac18740d6d91ad1 |
|
MD5 | c480dbd2b254fb5c3721dd78a9abb312 |
|
BLAKE2b-256 | 5a146862cf9901c24011aecb26f9398c5e351ce5d6a56ee23c774d4b397944cc |
Close
Hashes for classification_reportzr-0.0.1b3-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 141580153339d745a0af12b15048110f16e40a35e9679c0f83216e4466dfbb13 |
|
MD5 | a50cce016877ec7577543e5c4fa37611 |
|
BLAKE2b-256 | 6c59170154ae91f4e9b050d713b7b9e54022830bfca89b430d4867f2be817581 |