PrismBoost
PrismBoost is a gradient-boosting classifier/regressor that uses SEFR oblique splits at internal nodes (hyperplane splits instead of axis-aligned thresholds), with an optional fast C++ backend.
from prismboost import PrismBoostClassifier
from sklearn.datasets import load_breast_cancer
from sklearn.model_selection import train_test_split
X, y = load_breast_cancer(return_X_y=True)
X_train, X_test, y_train, y_test = train_test_split(
X, y, test_size=0.2, random_state=0, stratify=y
)
clf = PrismBoostClassifier(random_state=0) # capacity parameters adapt to the data
clf.fit(X_train, y_train)
print(clf.score(X_test, y_test))
print(clf.auto_config_) # what was chosen for this training set
Adaptive defaults
Every capacity parameter defaults to "auto" and is resolved from the training-set shape at fit time, the way CatBoost adapts its learning rate to dataset size. The rules were calibrated on the per-dataset Optuna optima of the 121 PMLB benchmark datasets, so an untuned model starts near a sensible configuration instead of one fixed point that under-fits large data and over-fits small data.
| Parameter | "auto" rule |
|---|---|
n_estimators |
200 |
learning_rate |
0.05 below 500 rows, else 0.1 (keeps learning_rate * n_estimators near the tuned optimum) |
max_depth |
4 below 500 rows, else 6 |
min_samples_leaf |
sqrt(n_samples) / 2, clipped to [5, 30] |
min_samples_split |
2 * min_samples_leaf |
subsample |
0.8, or 1.0 below 100 rows |
split_mode |
hybrid up to 50 features, else hybrid_sampled |
Only the shape of X is used, never y, so folds of equal size resolve identically and no label information leaks. Passing an explicit value disables adaptation for that parameter alone:
clf = PrismBoostClassifier(max_depth=3, random_state=0) # depth fixed, the rest still auto
clf.fit(X_train, y_train)
clf.max_depth_, clf.n_estimators_ # (3, 200)
Class imbalance is deliberately left alone (class_weight=None), matching XGBoost and CatBoost defaults. To reproduce pre-0.2 behaviour, pass the old values explicitly: n_estimators=100, learning_rate=0.1, max_depth=3, min_samples_leaf=10, min_samples_split=2, subsample=1.0, split_mode="hybrid_sampled".
Leaf regularization (reg_lambda)
reg_lambda is the L2 penalty on leaf weights, the same knob as XGBoost's reg_lambda. It enters
the Newton step as sum(w r) / (sum(w h) + reg_lambda) and the split gain as G^2 / (H + reg_lambda).
It defaults to 0.0, so results from earlier versions are unchanged unless you set it. Raising it
matters most on imbalanced classification: a nearly pure leaf has h = p (1 - p) close to zero,
so the unregularized Newton step is large and the accumulated scores can saturate the softmax. On a
3-class problem at a 2/10/88 class split, 200 rounds:
reg_lambda |
test log loss |
|---|---|
| 0.0 | 0.399 |
| 1.0 | 0.233 |
| 5.0 | 0.207 |
| 20.0 | 0.188 |
(Predicting the class prior scores 0.437 on the same split.) Accuracy-style metrics are much less sensitive to this than log loss is, which is why the PMLB study above, scored on macro-F1 and ROC-AUC, did not surface it.
Measured directly: over 33 PMLB datasets with Optuna tuning, reg_lambda is neutral on ROC-AUC
(13 wins, 16 losses, median difference 0.0000). It acts on calibrated probabilities, so tune it
when you score with a proper scoring rule such as log loss, or when classes are imbalanced, and
expect little from it on ranking metrics.
Early stopping (eval_set, early_stopping_rounds)
Pass validation data to fit and set early_stopping_rounds to stop once the validation loss (log
loss for classifiers, squared error for the regressor) has not improved for that many stages. The
ensemble is truncated to the best stage, so n_estimators becomes an upper bound:
clf = PrismBoostClassifier(n_estimators=1600, early_stopping_rounds=50)
clf.fit(X_train, y_train, eval_set=(X_val, y_val))
clf.best_iteration_ # stages kept
clf.validation_loss_ # validation loss after each fitted stage
Refitting with n_estimators=clf.best_iteration_ and the same random_state reproduces the
early-stopped model exactly. Passing eval_set without early_stopping_rounds only records
validation_loss_. On a 4,000-row binary problem with a 1,600-stage cap, early stopping kept 134
stages, cut fit time from 27 s to 3 s, and lowered test log loss from 0.418 to 0.328.
Time budget (fit(time_limit=...))
fit takes an optional wall-clock budget in seconds. The boosting loop checks the elapsed time
after each stage and stops, keeping the stages fitted so far, so a caller with a deadline gets a
usable model instead of an overrun:
clf = PrismBoostClassifier(n_estimators=4000, split_mode="hybrid", learning_rate=0.02)
clf.fit(X_train, y_train, time_limit=60) # seconds
It composes with early stopping: pass eval_set as well and the fit ends at whichever comes
first, keeping the best stage. On a 8,000 x 40 table with a 4,000-stage cap, an unbudgeted fit
took 326 s; the same fit with time_limit=3 took 3.1 s and kept 38 stages.
Checked between stages, so a single stage on a very large table can still overshoot: this bounds
the loop rather than guaranteeing a hard deadline. None (the default) means no budget.
Why oblique boosting?
Axis-aligned GBDTs approximate curved boundaries with staircases. PrismBoost fits linear (oblique) splits, so decision surfaces on non-linear problems are typically smoother.
Predicted-probability surfaces on moons: PrismBoost (left) vs XGBoost (right).
Decision boundaries on six synthetic 2D datasets (rows) across classifiers (columns). Lower surface roughness S is smoother.
Benchmark highlights (PMLB)
Evaluated on 121 Penn Machine Learning Benchmark classification datasets against strong baselines (CatBoost, LightGBM, LightGBM-linear, XGBoost, SPORF, Random Forest, Logistic Regression). Hyperparameters are tuned with Optuna; scores are repeated stratified CV.
Median scores (higher is better for F1 / ROC-AUC; lower is better for inference latency):
| Model | Median macro-F1 | Median ROC-AUC | Median inference (ms/row) |
|---|---|---|---|
| PrismBoost | 0.903 | 0.976 | 0.056 |
| CatBoost | 0.892 | 0.976 | 0.184 |
| XGBoost | 0.875 | 0.970 | 0.615 |
| LightGBM-linear | 0.875 | 0.970 | 0.567 |
| LightGBM | 0.870 | 0.972 | 0.550 |
| Random Forest | 0.862 | 0.964 | 5.140 |
| SPORF | 0.856 | 0.970 | 10.058 |
| Logistic Regression | 0.822 | 0.942 | 0.053 |
Average ranks (1 = best; Friedman tests significant for macro-F1 and ROC-AUC):
| Model | Macro-F1 rank | ROC-AUC rank |
|---|---|---|
| CatBoost | 3.33 | 3.48 |
| PrismBoost | 3.88 | 4.26 |
| LightGBM-linear | 3.99 | 4.26 |
| LightGBM | 4.10 | 4.10 |
| XGBoost | 4.31 | 4.24 |
| Logistic Regression | 5.31 | 5.66 |
| Random Forest | 5.48 | 5.31 |
| SPORF | 5.61 | 4.70 |
Nemenyi critical-difference diagrams (α = 0.05). Models connected by a bar are not significantly different.
On these data, PrismBoost is competitive with modern GBDTs on accuracy while remaining among the fastest at inference (second only to logistic regression; fastest non-linear model by median latency).
Install
pip install prismboost
Requires Python 3.10–3.13. A C++17 compiler and CMake are used when building the optional native extension (included for common platforms via wheels / sdist build).
From source (editable / development):
pip install -e ".[dev]"
If the C++ extension fails to build, the package still works via the pure-Python backend.
Extras:
pip install "prismboost[examples]" # Optuna for the tuning example
pip install "prismboost[dev]" # pytest, ruff
Public API
| Name | Description |
|---|---|
PrismBoostClassifier |
Classifier (sklearn-compatible) |
PrismBoostRegressor |
Regressor |
SEFR |
Linear weak learner used inside oblique splits |
auto_boosting_config(n_samples, n_features) |
The "auto" default rules, callable for inspection |
Legacy names
The project was previously called SEFRBoost. Those names are aliases of the
classes above — the same objects, so isinstance checks and old pickles keep
working — and are kept for backwards compatibility:
| Legacy name | Now |
|---|---|
SEFRBoostClassifier / SEFRBoostRegressor |
PrismBoostClassifier / PrismBoostRegressor |
SEFRGradientBoostingClassifier / SEFRGradientBoostingRegressor |
PrismBoostClassifier / PrismBoostRegressor |
prismboost.sefr_gbdt, prismboost.sefr_boost |
prismboost.prism_boost |
SEFR itself is not legacy: it is the linear model that produces each oblique
split, and it keeps its name.
Features
- Oblique tree splits from SEFR (closed-form linear separator)
- Binary and multiclass classification; regression
- Newton (second-order) split gain and leaf values, as in XGBoost and CatBoost;
second_order=Falseselects the first-order variant, where the per-sample Hessian is replaced by 1 so the gain becomes variance reduction and leaves hold the mean residual - Optional C++ backend for faster fit/predict
- sklearn estimator API (
fit,predict,predict_proba, pipelines, pickling) - Works with Optuna / GridSearchCV / RandomizedSearchCV
Examples
pip install "prismboost[examples]"
python examples/quickstart.py
python examples/optuna_tuning.py --n-trials 20
examples/quickstart.py— minimal fit / scoreexamples/optuna_tuning.py— Optuna CV search over trees, depth, learning rate,split_mode, and scaler
Docs mirror: docs/optuna_tuning.rst.
Tests
pip install -e ".[dev]"
pytest -q
License
This project is licensed under the MIT License.
Third-party note: prismboost._utils includes code derived from wnb under the BSD 3-Clause License.
Citation
If you use PrismBoost in academic work, please cite the accompanying paper (to be updated on publication).
Authors
- Hamidreza Keshavarz
- Reza Rawassizadeh
1 maintainer
Metadata
1 maintainer
Release files for prismboost 0.5.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| prismboost-0.5.0.tar.gz | 758.4 kB | Details |
Built distributions (wheels)
Total release size: 16.8 MB
Release files / prismboost-0.5.0.tar.gz
| Download URL | prismboost-0.5.0.tar.gz |
|---|---|
| Size | 758.4 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
f8d3d36bc342cc3887c715e3153c65f83af97428b52d6dc3e98d381acae3958b
|
|
BLAKE2b-256 checksum How to use checksums |
dc23f4e882707fb915ffa3789e80cbe3aa2412b3a0e0f9e9360b61c714d701d7
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.8.0
|
Release files / prismboost-0.5.0-cp313-cp313-win_arm64.whl
| Download URL | prismboost-0.5.0-cp313-cp313-win_arm64.whl |
|---|---|
| Size | 545.5 kB |
| Tags | CPython 3.13 Windows ARM64 |
|
SHA-256 checksum How to use checksums |
e2dd934787c7d9d741ef710bcc04d57c417637759bff9d79a69da0800f99ba78
|
|
BLAKE2b-256 checksum How to use checksums |
2b68985272dcd29eb0e3d9fcf31629d2490afb37a1784da7777e1d6813445569
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp313-cp313-win_amd64.whl
| Download URL | prismboost-0.5.0-cp313-cp313-win_amd64.whl |
|---|---|
| Size | 379.3 kB |
| Tags | CPython 3.13 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
6346bc419f54a4537a9da57b791670c19f24d765a6fae5bad850d8a62d5700d8
|
|
BLAKE2b-256 checksum How to use checksums |
9644234f19ab97b3e4bcf5b465f81b85e782ad5104b480586e7149b04fd9ac42
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp313-cp313-musllinux_1_2_x86_64.whl
| Download URL | prismboost-0.5.0-cp313-cp313-musllinux_1_2_x86_64.whl |
|---|---|
| Size | 1.2 MB |
| Tags | CPython 3.13 Linux musl 1.2+ x86-64 |
|
SHA-256 checksum How to use checksums |
810f878a90fb948ce98979bfb8dd0b26783047cd08d6f815837121e0eeb92ed9
|
|
BLAKE2b-256 checksum How to use checksums |
88303304837bf829e42dc3b5bbb5a9638ddb830ef8b60079e3d86592e8620d48
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp313-cp313-musllinux_1_2_aarch64.whl
| Download URL | prismboost-0.5.0-cp313-cp313-musllinux_1_2_aarch64.whl |
|---|---|
| Size | 1.2 MB |
| Tags | CPython 3.13 Linux musl 1.2+ ARM64 |
|
SHA-256 checksum How to use checksums |
0068f0963a9c4b2e7a509f5256157af08844e9049dad4a1482380fbceeeb0b88
|
|
BLAKE2b-256 checksum How to use checksums |
b7206db648ded7d89f8b17c6ea7ecb5615efbd61cce82b0764d7bd38ad239a29
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl
| Download URL | prismboost-0.5.0-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl |
|---|---|
| Size | 208.3 kB |
| Tags | CPython 3.13 Linux glibc 2.27+ x86-64 Linux glibc 2.28+ x86-64 |
|
SHA-256 checksum How to use checksums |
670e0a305a0ff435bec5c2ac9cb7d1b9678deb10bdc989fd9402bea07bea05e7
|
|
BLAKE2b-256 checksum How to use checksums |
ff2c82858773d10af8d7447c5ac0e2a28cd7735784c7266d01c9c86239c1855b
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp313-cp313-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl
| Download URL | prismboost-0.5.0-cp313-cp313-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl |
|---|---|
| Size | 189.5 kB |
| Tags | CPython 3.13 Linux glibc 2.26+ ARM64 Linux glibc 2.28+ ARM64 |
|
SHA-256 checksum How to use checksums |
2e8dd0537026b5e13ca9f134d22f31bff23788b94a8325e79a7257cb1ce0b611
|
|
BLAKE2b-256 checksum How to use checksums |
c63bc615bfcad54d647ba60081be4feef90e9dc13a40a798cbabf59ae939b826
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp313-cp313-macosx_11_0_arm64.whl
| Download URL | prismboost-0.5.0-cp313-cp313-macosx_11_0_arm64.whl |
|---|---|
| Size | 171.8 kB |
| Tags | CPython 3.13 macOS 11.0+ ARM64 |
|
SHA-256 checksum How to use checksums |
0db38711b433809966dbfef112481a447826ad5da82c753c3fb3fdf4669ad121
|
|
BLAKE2b-256 checksum How to use checksums |
e6b336d9e19d2a44066402650c23873b6834d004c525d5862677ccd8bf30eccf
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp313-cp313-macosx_10_13_x86_64.whl
| Download URL | prismboost-0.5.0-cp313-cp313-macosx_10_13_x86_64.whl |
|---|---|
| Size | 182.4 kB |
| Tags | CPython 3.13 macOS 10.13+ x86-64 |
|
SHA-256 checksum How to use checksums |
8dc8307b5c92f5940b2bacde12257326050390c4d31a44e89148cfcd1b403b88
|
|
BLAKE2b-256 checksum How to use checksums |
cef2ffa4bf5c15cc8aef8641be17523d4dc98a4212f6a4be6a5d72b48293e2fe
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp312-cp312-win_arm64.whl
| Download URL | prismboost-0.5.0-cp312-cp312-win_arm64.whl |
|---|---|
| Size | 545.5 kB |
| Tags | CPython 3.12 Windows ARM64 |
|
SHA-256 checksum How to use checksums |
33dd7587b3434af06c9887defb436f5b14ee9108ce1e5b60876f846376dcfbb3
|
|
BLAKE2b-256 checksum How to use checksums |
07627c977d75cff6028e920916787bf05fb42d6fed427c188dd367295903b814
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp312-cp312-win_amd64.whl
| Download URL | prismboost-0.5.0-cp312-cp312-win_amd64.whl |
|---|---|
| Size | 379.3 kB |
| Tags | CPython 3.12 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
378cf164c63faadab77bb2a2ae26f362acffad7e78f6c8edef8b7cee7d7e0611
|
|
BLAKE2b-256 checksum How to use checksums |
4d701073e4f2dee2583693551c7830c9649e4478b1d57a1d766d2f2d05ab1eb6
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp312-cp312-musllinux_1_2_x86_64.whl
| Download URL | prismboost-0.5.0-cp312-cp312-musllinux_1_2_x86_64.whl |
|---|---|
| Size | 1.2 MB |
| Tags | CPython 3.12 Linux musl 1.2+ x86-64 |
|
SHA-256 checksum How to use checksums |
b95605113942cc4c12d37e1281268b06eb984f261ad1f384c6efbca823dd1601
|
|
BLAKE2b-256 checksum How to use checksums |
cb5d027e6a8c232e8e29204ecf41eaee563288b10845bf2ff93429aeead31815
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp312-cp312-musllinux_1_2_aarch64.whl
| Download URL | prismboost-0.5.0-cp312-cp312-musllinux_1_2_aarch64.whl |
|---|---|
| Size | 1.2 MB |
| Tags | CPython 3.12 Linux musl 1.2+ ARM64 |
|
SHA-256 checksum How to use checksums |
8f26d3cfee5e0e56f9fd2598f2a188a87e23e1dd599b8ad210e17b1531945bda
|
|
BLAKE2b-256 checksum How to use checksums |
7f6494894a2b6f5b12ad687e8edac18396aa00f49a2fcfd4315917623e485049
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl
| Download URL | prismboost-0.5.0-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl |
|---|---|
| Size | 208.3 kB |
| Tags | CPython 3.12 Linux glibc 2.27+ x86-64 Linux glibc 2.28+ x86-64 |
|
SHA-256 checksum How to use checksums |
b516652b15382ba8b9fe3144498dbc6d81ac2c772e756ac9dcdceca5b4bdcb08
|
|
BLAKE2b-256 checksum How to use checksums |
5b13d8419b1cef37a938ad720f0bebcaaf1ccb21fb1c3ca74284fc339625c603
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp312-cp312-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl
| Download URL | prismboost-0.5.0-cp312-cp312-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl |
|---|---|
| Size | 189.5 kB |
| Tags | CPython 3.12 Linux glibc 2.26+ ARM64 Linux glibc 2.28+ ARM64 |
|
SHA-256 checksum How to use checksums |
688da7bd0a0ade4edffa979ea26dd092507ad68298d9615a3779bd54ddf6864b
|
|
BLAKE2b-256 checksum How to use checksums |
077f44f89abaf699e18b6f0a5edbb92057f44a299b4abcd28e015ef09296fa89
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp312-cp312-macosx_26_0_arm64.whl
| Download URL | prismboost-0.5.0-cp312-cp312-macosx_26_0_arm64.whl |
|---|---|
| Size | 170.4 kB |
| Tags | CPython 3.12 macOS 26.0+ ARM64 |
|
SHA-256 checksum How to use checksums |
0b560630012b6427bd22e60f6adadadce0a129f9ba6e53fea1acaf739cc92cc8
|
|
BLAKE2b-256 checksum How to use checksums |
65df4ee0fc287fb77b315cf78b0d122a32c1d605e8936fd1a862461f32e21e2f
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.8.0
|
Release files / prismboost-0.5.0-cp312-cp312-macosx_11_0_arm64.whl
| Download URL | prismboost-0.5.0-cp312-cp312-macosx_11_0_arm64.whl |
|---|---|
| Size | 171.7 kB |
| Tags | CPython 3.12 macOS 11.0+ ARM64 |
|
SHA-256 checksum How to use checksums |
60d73af5a7272ec22adf24de8a6ecfcd7c6d6f5bf49b9bcc2bf9337e99c165e4
|
|
BLAKE2b-256 checksum How to use checksums |
c816094c2a9ce1bc50ea0506973c2d8007bf9601721ae2f155d8efe3cbce4c1f
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp312-cp312-macosx_10_13_x86_64.whl
| Download URL | prismboost-0.5.0-cp312-cp312-macosx_10_13_x86_64.whl |
|---|---|
| Size | 182.3 kB |
| Tags | CPython 3.12 macOS 10.13+ x86-64 |
|
SHA-256 checksum How to use checksums |
f453792e2201670f0c93fad9ca588bbde39cffbe1e9d5a48b1a3282ff149c7d3
|
|
BLAKE2b-256 checksum How to use checksums |
5888f71572753480aad6db74e30e02bf62e1ab20baae9bbfe20e8d6c35573c8d
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp311-cp311-win_arm64.whl
| Download URL | prismboost-0.5.0-cp311-cp311-win_arm64.whl |
|---|---|
| Size | 544.3 kB |
| Tags | CPython 3.11 Windows ARM64 |
|
SHA-256 checksum How to use checksums |
8f0d2575a8018cf2c9c0c06615c199870c614913f8d9c61e433b864853d24dc1
|
|
BLAKE2b-256 checksum How to use checksums |
cc06c1302f99fc4778c203d7cc50b2e1a8ff7ffab350c88b5324820220b8785b
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp311-cp311-win_amd64.whl
| Download URL | prismboost-0.5.0-cp311-cp311-win_amd64.whl |
|---|---|
| Size | 375.7 kB |
| Tags | CPython 3.11 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
9d010692ba9bfe8fd026008fbcb9d02c3f99a02163ad28aba944ffe89317c39f
|
|
BLAKE2b-256 checksum How to use checksums |
2b3e3614a5a3f16ad3952ecf61c90daf15dd77cba20b3e720a4408be1e22a12d
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp311-cp311-musllinux_1_2_x86_64.whl
| Download URL | prismboost-0.5.0-cp311-cp311-musllinux_1_2_x86_64.whl |
|---|---|
| Size | 1.2 MB |
| Tags | CPython 3.11 Linux musl 1.2+ x86-64 |
|
SHA-256 checksum How to use checksums |
0b0e2fc0252a74748f704f9bfac6d15e8ad4874c1dd118b9244a6419d576ce53
|
|
BLAKE2b-256 checksum How to use checksums |
afd81e567a59395ee4d4c55d774345a1327644f8fe83ed03fc39cac7819ed4fe
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp311-cp311-musllinux_1_2_aarch64.whl
| Download URL | prismboost-0.5.0-cp311-cp311-musllinux_1_2_aarch64.whl |
|---|---|
| Size | 1.2 MB |
| Tags | CPython 3.11 Linux musl 1.2+ ARM64 |
|
SHA-256 checksum How to use checksums |
25864a9f9f6941eaca07e1d4205d2a021f37c154ee2fe12ae80324d2a6c7fe89
|
|
BLAKE2b-256 checksum How to use checksums |
3002b98c5a5e7b8c4cfa8cb37ac2592571f78d6bb4137b6b7f5e99e0d01d4803
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl
| Download URL | prismboost-0.5.0-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl |
|---|---|
| Size | 207.6 kB |
| Tags | CPython 3.11 Linux glibc 2.27+ x86-64 Linux glibc 2.28+ x86-64 |
|
SHA-256 checksum How to use checksums |
e76ba7003b1e5a963f9fced3fd50a7c404e6aa379382e55a45cd842b7a61c37a
|
|
BLAKE2b-256 checksum How to use checksums |
f55b798caf6abce6da9cb6103847102ae6e91e0f0657e725c7abadaee65aeee8
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp311-cp311-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl
| Download URL | prismboost-0.5.0-cp311-cp311-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl |
|---|---|
| Size | 187.9 kB |
| Tags | CPython 3.11 Linux glibc 2.26+ ARM64 Linux glibc 2.28+ ARM64 |
|
SHA-256 checksum How to use checksums |
941d6adbadd270b3f2b424cd39c51e409a204042f777e79d8600b8cf005a5515
|
|
BLAKE2b-256 checksum How to use checksums |
fb3e853a1a498da3899dd919aa77e72c33c84f7f7724502cf571db9062908798
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp311-cp311-macosx_11_0_arm64.whl
| Download URL | prismboost-0.5.0-cp311-cp311-macosx_11_0_arm64.whl |
|---|---|
| Size | 170.8 kB |
| Tags | CPython 3.11 macOS 11.0+ ARM64 |
|
SHA-256 checksum How to use checksums |
cc1eee9b68d16a8d06c403e1c17424111be8b1de19f208323f95fc15453dc49c
|
|
BLAKE2b-256 checksum How to use checksums |
aa4245fdc0dc4fd1df693439ad9e6198d06c936ef32f092ab917e966b0bbf552
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp311-cp311-macosx_10_9_x86_64.whl
| Download URL | prismboost-0.5.0-cp311-cp311-macosx_10_9_x86_64.whl |
|---|---|
| Size | 180.2 kB |
| Tags | CPython 3.11 macOS 10.9+ x86-64 |
|
SHA-256 checksum How to use checksums |
1495bc521a70688c97051656cdadce3127b194b673e31b759218066b3095cbd5
|
|
BLAKE2b-256 checksum How to use checksums |
2e77131350bb4b820504e1b282442373de9b4e6484682e0052bed7027e42995b
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp310-cp310-win_amd64.whl
| Download URL | prismboost-0.5.0-cp310-cp310-win_amd64.whl |
|---|---|
| Size | 374.9 kB |
| Tags | CPython 3.10 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
8670c45be9a35d3cdd64a3b0272510d88a395486a8748bf201999f977e610307
|
|
BLAKE2b-256 checksum How to use checksums |
bf7ac6c2c9a6e8f9c7ff1d097980d40a681c82fe01e61e52ec3a20af6c3d2548
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp310-cp310-musllinux_1_2_x86_64.whl
| Download URL | prismboost-0.5.0-cp310-cp310-musllinux_1_2_x86_64.whl |
|---|---|
| Size | 1.2 MB |
| Tags | CPython 3.10 Linux musl 1.2+ x86-64 |
|
SHA-256 checksum How to use checksums |
501a4678833e1126d7fc0ffab5b0cbb7488aafef4a87138e3733839e137aa82a
|
|
BLAKE2b-256 checksum How to use checksums |
c5ff8170407762a98017bc1a49c7ec89d9cbec3047cb74a664504365c5a1deb5
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp310-cp310-musllinux_1_2_aarch64.whl
| Download URL | prismboost-0.5.0-cp310-cp310-musllinux_1_2_aarch64.whl |
|---|---|
| Size | 1.2 MB |
| Tags | CPython 3.10 Linux musl 1.2+ ARM64 |
|
SHA-256 checksum How to use checksums |
2d1e1330a45072cf25cb0d5f2de817e976bab5627efe3c50693cfbe7c0a23fa5
|
|
BLAKE2b-256 checksum How to use checksums |
b8eda46e229a28d8b2ccf7a0a5e8ea40859d64688b14fe5f5de8b9bde27e911d
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl
| Download URL | prismboost-0.5.0-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl |
|---|---|
| Size | 207.7 kB |
| Tags | CPython 3.10 Linux glibc 2.27+ x86-64 Linux glibc 2.28+ x86-64 |
|
SHA-256 checksum How to use checksums |
837c57d102919e3ce07f129972bd7fbb04ad5a7a742f59ec13ffc22aab1d5f1b
|
|
BLAKE2b-256 checksum How to use checksums |
b75f7ef7be29ee2c9832ea0f8705cce7ae4f316cdabe7227d516e2d01f98a060
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp310-cp310-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl
| Download URL | prismboost-0.5.0-cp310-cp310-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl |
|---|---|
| Size | 187.1 kB |
| Tags | CPython 3.10 Linux glibc 2.26+ ARM64 Linux glibc 2.28+ ARM64 |
|
SHA-256 checksum How to use checksums |
e3d3fbdeb711cd3d0c401051e509ade759a162893b8c4672fefa37d20e6f7940
|
|
BLAKE2b-256 checksum How to use checksums |
601b6b3b097bb1f535858bad6d67f9a0c6d35f0c1ece05fea383bd12ab88c7ae
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp310-cp310-macosx_11_0_arm64.whl
| Download URL | prismboost-0.5.0-cp310-cp310-macosx_11_0_arm64.whl |
|---|---|
| Size | 169.8 kB |
| Tags | CPython 3.10 macOS 11.0+ ARM64 |
|
SHA-256 checksum How to use checksums |
fa0ee8ce4f05d5cb98290529669c3119ad5cecf51d8ffe77427edf799614789c
|
|
BLAKE2b-256 checksum How to use checksums |
538332bb963ab186039f689093358d2d38a716188752018364d52ff54d704ed4
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency logRelease files / prismboost-0.5.0-cp310-cp310-macosx_10_9_x86_64.whl
| Download URL | prismboost-0.5.0-cp310-cp310-macosx_10_9_x86_64.whl |
|---|---|
| Size | 178.9 kB |
| Tags | CPython 3.10 macOS 10.9+ x86-64 |
|
SHA-256 checksum How to use checksums |
075d91571c7bd745fe1e8b34e5a144501c9803ccd092859b393a33f93841bb3c
|
|
BLAKE2b-256 checksum How to use checksums |
00b6371b3c79efbb8ca225aa1e7a9f55e766e5ea46f057fdce770e3fd473d255
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 7, 2026.
Transparency log
