Pinebioml

Latest version: v1.2.5

Safety actively analyzes 714973 Python packages for vulnerabilities to keep your Python projects secure.

Scan your dependencies

Page 1 of 5

1.2.5

description:
1. rollback: threshold picking for binary classification.

1.2.4

description:
1. change the backend of Lasso_selection from LassoRegression to LassoLars. there is still some numerical issue when penalty is low.
a. 50x faster than Lasso bisection
b. coeficient path now is available in .Plotting function
2. lasso bisection 30% faster.
3. Multi processing support for DT_selection.
4. Early stopping for XGBoost, Lightgbm and CatBoost.
5. Pine summary.
6. threshold tuner: Binary classification's threshold will be chosen by auc.
7. Model's search space now can be obtained by .detail()

On Going:
0. A reliable tutorial. (including mac)
1. Pine monitor, progress bar and report. experiment setting visualization.

ToDo:
1. barutoSHAP (baruto, shap, barutoshap)
2. using pretty, beautiful, good-looking, precise packages:
a. pca
b. The only OPLS da reliable(compare to others), alive, python implement
https://github.com/Omicometrics/pypls?tab=readme-ov-file
c. https://www.omicsanalyst.ca/docs/Gallery.xhtml
3. fairness learning (mljar, fairlearn)
4. cv std
n. add parameter dict(json or yaml-like)

1.2.3

description:
1. report utils - classification_summary and regression_summary:
a. revise the scores from str to pd.DataFrame.
b. now we can export the scores
2. tuner now will flush the data after tuned. for safety and privacy concerns.
3. Pine: add evaluation_ncv api to the final cv score evaluation.

1.2.2

description:
1. Tuner:
a. Formalize tuner tuning range.
b. Adding tuner.detail() to access optuna tuning range.
c. optuna tuning range changes.
2. update README.md.

1.2.1

description:
1. remove pacmap from requirment. (C++ 14 build tool dependancy)

1.2.0

description:
1. Document update
2. Add reference: now you can call tuner.reference() to list major reference of that tuner.
3. Add optuna optimize history plot


bug:
1. feature bagging:
- contradicts to log domain in selection method (the negative output of pca)
- the variance of feature will change during bagging (variance matters)
2. Lasso logistic regression in grid search need to to modified. In temporary using binary search.
3. confusion matrix no title

On Going:
0. A reliable tutorial. (including mac) / Document reference
1. Pine monitor, progress bar and report. experiment setting visualization.

ToDo:
0. threshold tuner (or like jadbio)
1. search space dict
2. report and diagnose, Pine pipeline graph.
3. using pretty, beautiful, good-looking, precise packages:
a. pca
b. The only OPLS da reliable(compare to others), alive, python implement
https://github.com/Omicometrics/pypls?tab=readme-ov-file
c. https://www.omicsanalyst.ca/docs/Gallery.xhtml
4. barutoSHAP (baruto, shap, barutoshap)
5. fairness learning (mljar, fairlearn)
6. cv std
n. multi objective optimize for optuna
n. add parameter dict(json or yaml-like)
n. interactive interface or GUI (maybe nicegui/ plotly)

Page 1 of 5

© 2025 Safety CLI Cybersecurity Inc. All Rights Reserved.