Y-Scramble: a package for y-randomization validation
Project description
Y-Scramble
Y-Scramble is a simple python package to perform y-randomization validation
of machine learning models. It can be used for classification and regression tasks
and accepts models following the scikit-learn
inteface, and the user may use all scorers available at scikit-learn
(accuracy
, recall
, precision
).
Installing
Y-Scramble can be installed from PyPI using the following command:
$ pip install y-scamble
Usage
from y_scramble import Scrambler
from sklearn.tree import DecisionTreeClassifier
X, y = load_iris(return_X_y=True)
model = DecisionTreeClassifier()
scrambler = Scrambler(model=model, iterations=1000)
scores, zscores, pvalues, significances = scrambler.validate(
X, y,
scoring="accuracy",
cross_val_score_aggregator="mean",
pvalue_threshold=0.01
)
The scramble
object returns the scores, z-scores, p-values and the significancy
information for the model trained (base_model
) using the default dataset and for different randomized versions as well (scrambled_models
). These results are stores in numpy arrays, where the position of index 0 represents the base_model
and the others the scrambled_models
.
The score of the base_model
is stored in scores[0]
, and i's p-values is stored in
pvalues[0]
. If this p-value is significant, the value of significances[0]
will be
True
, indicating that base_model
shows a significantly better result when comparing to the randomized models. Following the same logic, scores[1]
to scores[1000]
, for example, will store the score values for the randomized model 1
and 1000
, respectively.
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distributions
Built Distribution
File details
Details for the file y_scramble-0.0.8-py3-none-any.whl
.
File metadata
- Download URL: y_scramble-0.0.8-py3-none-any.whl
- Upload date:
- Size: 3.7 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/3.8.0 pkginfo/1.8.2 readme-renderer/32.0 requests/2.27.1 requests-toolbelt/0.9.1 urllib3/1.26.8 tqdm/4.62.3 importlib-metadata/3.10.1 keyring/23.5.0 rfc3986/2.0.0 colorama/0.4.4 CPython/3.8.12
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 048a7a59a44f21b3481bef7ad8faf876d5b4f2eeea002eab1271accb54adb436 |
|
MD5 | 94b82a5a2c6e60f368fdb983cf0a3419 |
|
BLAKE2b-256 | c5c9e826d8fec2377d2825dc09815b9feb13bb94dff4385a1b5ac74c4c380113 |