xgboost-label-encoding
xgboost-label-encoding provides small sklearn-style wrappers around
xgboost.XGBClassifier for classification workflows where the target labels are
strings or other non-numeric values.
XGBoost trains on numeric class labels. This package encodes y during fit,
trains the underlying XGBoost classifier, and decodes predictions back to the
original labels. It is intended to be used as a drop-in estimator in places where
manually applying sklearn.preprocessing.LabelEncoder to the target would be
awkward.
Installation
pip install xgboost_label_encoding
The package requires Python 3.8+ and installs against xgboost<2.
For local development:
pip install -r requirements_dev.txt
pip install -e .
make test
Usage
Use XGBoostClassifierWithLabelEncoding in place of xgboost.XGBClassifier:
from xgboost_label_encoding import XGBoostClassifierWithLabelEncoding
clf = XGBoostClassifierWithLabelEncoding(
n_estimators=100,
class_weight="balanced",
)
clf.fit(X_train, y_train) # y_train may contain labels like "Healthy" or "HIV"
labels = clf.predict(X_test)
probabilities = clf.predict_proba(X_test)
classes = clf.classes_
Most XGBoost classifier parameters are passed through unchanged. The wrapper adds these project-specific options:
class_weight: passed tosklearn.utils.class_weight.compute_sample_weight; ifsample_weightis also supplied, the two weights are multiplied.fail_if_nothing_learned: defaults toTrue; raisesValueErrorafter fitting if all feature importances are zero.
Cross-Validated Fitting
XGBoostClassifierWithLabelEncodingWithCV combines label encoding with
cross-validation over XGBoost parameters:
from sklearn.model_selection import StratifiedKFold
from xgboost_label_encoding import XGBoostClassifierWithLabelEncodingWithCV
cv = StratifiedKFold(n_splits=5, shuffle=True, random_state=42)
clf = XGBoostClassifierWithLabelEncodingWithCV(
cv=cv,
max_num_trees=200,
early_stopping_patience=10,
class_weight="balanced",
)
clf.fit(X_train, y_train)
During fit, the CV wrapper:
- builds a small default grid of
learning_rateandmin_child_weightvalues unlessparam_gridis provided; - runs
xgboost.cvwith early stopping for each parameter set; - selects the best parameter set and number of boosting rounds;
- fits the final classifier on the full training data.
If the provided CV splitter accepts a groups argument, groups can be passed
to fit.
Behavior And Limitations
- Training data must contain at least two classes.
predictreturns original labels, not encoded integers.predict_probareturns one probability column per class inclf.classes_.- For pandas DataFrame inputs, feature names containing
[,], or<are renamed internally before reaching XGBoost.feature_names_in_still exposes the original feature names, and the same renaming is applied duringpredictandpredict_proba. XGBoostCVis also available as a standalone helper for numeric-label XGBoost classification with CV-selected hyperparameters and tree count.
Development
Useful local commands:
make test
make lint
make docs
make dist
Changelog
0.0.1
- First release on PyPI.
Metadata
Release files for xgboost-label-encoding 0.0.7
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 | Actions |
|---|---|---|---|---|
| xgboost_label_encoding-0.0.7.tar.gz | 19.8 kB | Without provenance | Details for xgboost_label_encoding-0.0.7.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| xgboost_label_encoding-0.0.7-py2.py3-none-any.whl | Python 2, Python 3 | none | any | Details for xgboost_label_encoding-0.0.7-py2.py3-none-any.whl |
Total release size: 30.2 kB
Release files / xgboost_label_encoding-0.0.7.tar.gz
| Download URL | xgboost_label_encoding-0.0.7.tar.gz |
|---|---|
| Size | 19.8 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
29bb4cc17dbc26ce12a644eae3fd2686bce5d8ba06eb22c80c743ed7fe31393c
|
|
BLAKE2b-256 checksum How to use checksums |
199ec063be874cb0447225cacb21836da620df5a4223beb2d1f221d80e4295de
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.9.25
|
Release files / xgboost_label_encoding-0.0.7-py2.py3-none-any.whl
| Download URL | xgboost_label_encoding-0.0.7-py2.py3-none-any.whl |
|---|---|
| Size | 10.4 kB |
| Tags | Python 2 Python 3 |
|
SHA-256 checksum How to use checksums |
4df30a411e15cdf9091d57a4542112330792d27e4a15d097caed705091f441f3
|
|
BLAKE2b-256 checksum How to use checksums |
cbf7f65347700925b2e69fa6c420d35e98f4cb9e3cdc74659803616f8719dc45
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.9.25
|
Security for xgboost-label-encoding 0.0.7
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.
To learn more about protecting yourself from malicious actors, visit our package security documentation.