Skip to main content

CTBoost

CTBoost is an alpha gradient-boosting library built around conditional-inference trees. At each node it selects a feature with a conditional statistical test before optimizing that feature's split point. A native C++17 core provides CPU training, optional CUDA acceleration, and Python/scikit-learn interfaces.

Documentation · Getting started · Benchmarks · Compatibility

Version 0.1.59 adds optional loss backtracking, coupled multiclass leaf fitting, and joint multiclass feature tests. It also reduces grouped-test allocation overhead and corrects feature-test versus cut-search profiling. Existing learning defaults and the statistical feature-selection-before-cut contract remain.

Highlights

  • Regression, classification, query-group ranking, survival, multi-output, multilabel, and callable-objective training.
  • Low-level Pool/train APIs plus familiar scikit-learn estimators.
  • NumPy, pandas, SciPy sparse, Arrow, and Polars input, with fitted categorical, text, and embedding preprocessing.
  • Native CPU and CUDA training, plus Dask, Ray, and Spark barrier adapters; Spark retains an explicit collect-to-driver fallback for small jobs.
  • Snapshots, warm starts, staged prediction, callbacks, model selection, and deterministic inference manifests.
  • Exact empirical interventional TreeSHAP and JSON, Python, C++, C ABI, ONNX, pickle, and prepared-feature R/JVM inference choices.

Install

python -m pip install -U ctboost
python -m pip install -U "ctboost[sklearn,dataframes]"  # optional integrations

See the GPU installation guide for the released wheel, driver, architecture, and platform matrix. Inspect an installation with:

python -c "import ctboost; print(ctboost.build_info())"

Quick start

import numpy as np
from ctboost import CTBoostClassifier

X = np.array(
    [
        [0.2, 1.0],
        [0.8, 0.1],
        [0.1, 0.9],
        [0.9, 0.2],
        [0.3, 0.7],
        [0.7, 0.3],
    ],
    dtype=np.float32,
)
y = np.array([0, 1, 0, 1, 0, 1])

model = CTBoostClassifier(
    iterations=100,
    learning_rate=0.05,
    max_depth=4,
    random_seed=42,
)
model.fit(X, y)

probability = model.predict_proba(X)
labels = model.predict(X)

The getting-started guide covers validation sets, early stopping, categorical data, grouped feature tests, and the low-level API.

Compact multiclass vector leaves

CTBoost 0.1.58 includes compact multiclass vector leaves introduced in the 0.1.57 source tag. Use CTBoostClassifier(multi_strategy="multi_output_tree") to store one shared CPU tree per multiclass boosting round, with one score per class in each leaf. It preserves CTBoost's conditional-inference feature tests and split selection; the default scalar-tree layout remains unchanged.

See the vector-leaf guide for supported workflows, artifact compatibility, and a reproducible comparison.

Evidence and project status

For the new CPU learning controls, see safeguarded leaves and joint multiclass tests. The full solver and joint test support 3–32 classes and remain opt-in.

CTBoost is an alpha project. Its API and model formats are tested extensively, but it does not yet have the independent production history of CatBoost or XGBoost.

The latest published TabArena artifacts measure CTBoost 0.1.58 on all 51 TabArena-v0.1 Lite datasets, using the default plus 25 frozen HPO configurations, one outer split (r0f0), and eight-fold bagging. All 1,326 parent results and 10,608 child fits completed, with no imputed CTBoost tasks.

Evaluation Lite Elo
Default 1161.9
Tuned 1262.8
Tuned + ensemble 1296.9

These local Lite scores use an 87-row comparison roster; they are not an official leaderboard entry or a TabArena-Full result. Kaggle used 4 CPUs and 28 GB RAM, so its timings are not directly comparable to the canonical benchmark. The benchmark history retains the earlier 0.1.56 default-only result and explains the different rosters.

The final-source 0.1.55 pre-registered grouped-statistic panel completed 294/294 isolated fits and 42/42 exact control checks. Grouped-8 recorded nine wins, no ties, and three losses with an observed 5.63% median primary-loss improvement (task-bootstrap 95% interval: -1.23% to +13.64%), but its 1.1708 median paired fit-time ratio exceeded the frozen 1.15 ceiling. Because the promotion gates were conjunctive, grouped-8 did not advance and the conditional TabArena scout was not run. The quadratic feature test remains the default; grouped testing is opt-in.

Read the benchmark status and split-statistics research ledger for protocols, limitations, and machine-readable evidence. See the 0.1.59 release notes for the new learning controls and their validation. The TabArena scores above belong to 0.1.58; 0.1.59 has not yet been evaluated on that benchmark.

Documentation

Topic Guide
Installation and first model Getting started
Training, objectives, callbacks, and wrappers Training workflows
GPU wheels and runtime requirements GPU installation
Data, streaming, and schema metadata Data input
Categorical, text, and embeddings Feature preprocessing
SHAP, influence, and diagnostics Explainability
Dask, Ray, and Spark Distributed training
Persistence, exports, and CLI Deployment
Prepared-feature R/JVM inference Portable inference
Python symbols and signatures API reference
Source builds and tests Development

Development

python -m pip install -e ".[dev]"
pytest tests

See the development guide for native CMake builds, repository layout, and documentation checks.

Methodology and license

CTBoost draws on the conditional-inference framework of Hothorn, Hornik, and Zeileis (2006) and the modular partykit work of Hothorn and Zeileis (2015). The research ledger records the precise literature-to-implementation boundary.

Apache-2.0 licensed. See the license text.

Release files for ctboost 0.1.59

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for ctboost 0.1.59
File Size Uploaded
ctboost-0.1.59.tar.gz 790.3 kB Details

Built distributions (wheels)

Table of built distributions (wheels) for ctboost 0.1.59
File
ctboost-0.1.59-cp314-cp314-win_amd64.whl CPython 3.14 CPython 3.14 Windows x86-64 Details
ctboost-0.1.59-cp314-cp314-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl CPython 3.14 CPython 3.14 Linux glibc 2.28+ x86-64, Linux glibc 2.27+ x86-64 Details
ctboost-0.1.59-cp314-cp314-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl CPython 3.14 CPython 3.14 Linux glibc 2.27+ ARM64, Linux glibc 2.28+ ARM64 Details
ctboost-0.1.59-cp314-cp314-macosx_10_15_x86_64.whl CPython 3.14 CPython 3.14 macOS 10.15+ x86-64 Details
ctboost-0.1.59-cp313-cp313-win_amd64.whl CPython 3.13 CPython 3.13 Windows x86-64 Details
ctboost-0.1.59-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl CPython 3.13 CPython 3.13 Linux glibc 2.28+ x86-64, Linux glibc 2.27+ x86-64 Details
ctboost-0.1.59-cp313-cp313-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl CPython 3.13 CPython 3.13 Linux glibc 2.27+ ARM64, Linux glibc 2.28+ ARM64 Details
ctboost-0.1.59-cp313-cp313-macosx_10_15_x86_64.whl CPython 3.13 CPython 3.13 macOS 10.15+ x86-64 Details
ctboost-0.1.59-cp312-cp312-win_amd64.whl CPython 3.12 CPython 3.12 Windows x86-64 Details
ctboost-0.1.59-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl CPython 3.12 CPython 3.12 Linux glibc 2.28+ x86-64, Linux glibc 2.27+ x86-64 Details
ctboost-0.1.59-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl CPython 3.12 CPython 3.12 Linux glibc 2.28+ ARM64, Linux glibc 2.27+ ARM64 Details
ctboost-0.1.59-cp312-cp312-macosx_10_15_x86_64.whl CPython 3.12 CPython 3.12 macOS 10.15+ x86-64 Details
ctboost-0.1.59-cp311-cp311-win_amd64.whl CPython 3.11 CPython 3.11 Windows x86-64 Details
ctboost-0.1.59-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl CPython 3.11 CPython 3.11 Linux glibc 2.28+ x86-64, Linux glibc 2.27+ x86-64 Details
ctboost-0.1.59-cp311-cp311-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl CPython 3.11 CPython 3.11 Linux glibc 2.28+ ARM64, Linux glibc 2.27+ ARM64 Details
ctboost-0.1.59-cp311-cp311-macosx_10_15_x86_64.whl CPython 3.11 CPython 3.11 macOS 10.15+ x86-64 Details
ctboost-0.1.59-cp310-cp310-win_amd64.whl CPython 3.10 CPython 3.10 Windows x86-64 Details
ctboost-0.1.59-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl CPython 3.10 CPython 3.10 Linux glibc 2.27+ x86-64, Linux glibc 2.28+ x86-64 Details
ctboost-0.1.59-cp310-cp310-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl CPython 3.10 CPython 3.10 Linux glibc 2.28+ ARM64, Linux glibc 2.27+ ARM64 Details
ctboost-0.1.59-cp310-cp310-macosx_10_15_x86_64.whl CPython 3.10 CPython 3.10 macOS 10.15+ x86-64 Details
ctboost-0.1.59-cp39-cp39-win_amd64.whl CPython 3.9 CPython 3.9 Windows x86-64 Details
ctboost-0.1.59-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl CPython 3.9 CPython 3.9 Linux glibc 2.27+ x86-64, Linux glibc 2.28+ x86-64 Details
ctboost-0.1.59-cp39-cp39-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl CPython 3.9 CPython 3.9 Linux glibc 2.27+ ARM64, Linux glibc 2.28+ ARM64 Details
ctboost-0.1.59-cp38-cp38-win_amd64.whl CPython 3.8 CPython 3.8 Windows x86-64 Details
ctboost-0.1.59-cp38-cp38-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl CPython 3.8 CPython 3.8 Linux glibc 2.28+ x86-64, Linux glibc 2.27+ x86-64 Details
ctboost-0.1.59-cp38-cp38-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl CPython 3.8 CPython 3.8 Linux glibc 2.28+ ARM64, Linux glibc 2.27+ ARM64 Details

Total release size: 58.1 MB

Release files / ctboost-0.1.59.tar.gz

Download URL ctboost-0.1.59.tar.gz
Size 790.3 kB
Tags Source
SHA-256 checksum
How to use checksums
1cb082a415405e1d1875976000dd1c3f618a276e74daa66459f7125218b278fe
BLAKE2b-256 checksum
How to use checksums
540c05f7049ff955b7df255a3847364f810aea2eb4722ee8c7dedbd490c219e7
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 Sep 6, 2026.

Transparency log

Release files / ctboost-0.1.59-cp314-cp314-win_amd64.whl

Download URL ctboost-0.1.59-cp314-cp314-win_amd64.whl
Size 3.7 MB
Tags CPython 3.14 Windows x86-64
SHA-256 checksum
How to use checksums
25e6c8310bee01c933dd95dcf77ee28971e5d2fc54a20e1e21b5e7e795e5eb41
BLAKE2b-256 checksum
How to use checksums
ccdb1e79c3b0d54eeb3504b1d02e7df6e5f5720241e362d76ca7f73697f763b7
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 Sep 6, 2026.

Transparency log

Release files / ctboost-0.1.59-cp314-cp314-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl

Download URL ctboost-0.1.59-cp314-cp314-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl
Size 3.9 MB
Tags CPython 3.14 Linux glibc 2.27+ x86-64 Linux glibc 2.28+ x86-64
SHA-256 checksum
How to use checksums
b52ff9527593c29738ff024981a773b3e1bc05559d9c972d5b7a130d0bcc46f7
BLAKE2b-256 checksum
How to use checksums
630d4651ed513bd7f84eb0ce9852c710d83d0c9a39f0b4f50ff2d122c2e035be
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 Sep 6, 2026.

Transparency log

Release files / ctboost-0.1.59-cp314-cp314-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl

Download URL ctboost-0.1.59-cp314-cp314-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl
Size 1.3 MB
Tags CPython 3.14 Linux glibc 2.27+ ARM64 Linux glibc 2.28+ ARM64
SHA-256 checksum
How to use checksums
641110aea38359af3a99e8460e61594f9698107e3f102389b6dc3fba77c68dfe
BLAKE2b-256 checksum
How to use checksums
913e32726ba1bd59f4eb18a73dbb3e7566fcb4c9e59665a1683f6bb8b8f14676
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 Sep 6, 2026.

Transparency log

Release files / ctboost-0.1.59-cp314-cp314-macosx_10_15_x86_64.whl

Download URL ctboost-0.1.59-cp314-cp314-macosx_10_15_x86_64.whl
Size 1.1 MB
Tags CPython 3.14 macOS 10.15+ x86-64
SHA-256 checksum
How to use checksums
dac2062cc8470fe41e3c0247c17457a291e9e8a7b83134afced302b05e5f3f47
BLAKE2b-256 checksum
How to use checksums
e96ecf7fced90b50be52a524b498a8895cfa08dcaab42ae9a7b768b0585a1c74
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 Sep 6, 2026.

Transparency log

Release files / ctboost-0.1.59-cp313-cp313-win_amd64.whl

Download URL ctboost-0.1.59-cp313-cp313-win_amd64.whl
Size 3.7 MB
Tags CPython 3.13 Windows x86-64
SHA-256 checksum
How to use checksums
13b1fc5742858c9fc41385940939241ec392648f17192209f19514a55ca6fc31
BLAKE2b-256 checksum
How to use checksums
c107bfd7b39c8844c6916f52bc0d3a3de9cf42aeb76a9b27b09c1491a725cad9
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 Sep 6, 2026.

Transparency log

Release files / ctboost-0.1.59-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl

Download URL ctboost-0.1.59-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl
Size 3.9 MB
Tags CPython 3.13 Linux glibc 2.27+ x86-64 Linux glibc 2.28+ x86-64
SHA-256 checksum
How to use checksums
02c6a320b34470379ff9ee0b79f37ad46602618a49fb33d7a386ac67bd9221f2
BLAKE2b-256 checksum
How to use checksums
edde278a895023d4ae62686ffe9ed4aea5551fd71eb9e877be0c1c473bf6c277
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 Sep 6, 2026.

Transparency log

Release files / ctboost-0.1.59-cp313-cp313-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl

Download URL ctboost-0.1.59-cp313-cp313-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl
Size 1.3 MB
Tags CPython 3.13 Linux glibc 2.27+ ARM64 Linux glibc 2.28+ ARM64
SHA-256 checksum
How to use checksums
0d516bbaa14354a0456f5272a2de485a0d82634adae872cb44c10508893129b4
BLAKE2b-256 checksum
How to use checksums
198640d12f2ab930b9928847579d894cc03b52d69b2cb52a267b3fcf9baa82d3
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 Sep 6, 2026.

Transparency log

Release files / ctboost-0.1.59-cp313-cp313-macosx_10_15_x86_64.whl

Download URL ctboost-0.1.59-cp313-cp313-macosx_10_15_x86_64.whl
Size 1.1 MB
Tags CPython 3.13 macOS 10.15+ x86-64
SHA-256 checksum
How to use checksums
d64b9788e1e14b387d991a944dedc9e19ff8e653558fe6cb3752b7aa88b98ba8
BLAKE2b-256 checksum
How to use checksums
4f5094dc22b32d7f9a4d0e52ee6433b528a87cd4e9704c8cfe68ed7529f71a25
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 Sep 6, 2026.

Transparency log

Release files / ctboost-0.1.59-cp312-cp312-win_amd64.whl

Download URL ctboost-0.1.59-cp312-cp312-win_amd64.whl
Size 3.7 MB
Tags CPython 3.12 Windows x86-64
SHA-256 checksum
How to use checksums
5faca874e14d4c2bf5a41b0b80116a709fbe476bdd99a9d75d4c66203154c55b
BLAKE2b-256 checksum
How to use checksums
fd4d89eed49d7c23014b588b4b0c465c45488c02391ffb6cc288c0470bf46e06
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 Sep 6, 2026.

Transparency log

Release files / ctboost-0.1.59-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl

Download URL ctboost-0.1.59-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl
Size 3.9 MB
Tags CPython 3.12 Linux glibc 2.27+ x86-64 Linux glibc 2.28+ x86-64
SHA-256 checksum
How to use checksums
0aef0ef5ecd0307e4b51f11daba6ab65b99280f6242ff89c976acda9989612ed
BLAKE2b-256 checksum
How to use checksums
0a332968112dee49d769969dfeb019abe72ec7897058df785cc2f06d7a84ae10
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 Sep 6, 2026.

Transparency log

Release files / ctboost-0.1.59-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl

Download URL ctboost-0.1.59-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl
Size 1.3 MB
Tags CPython 3.12 Linux glibc 2.27+ ARM64 Linux glibc 2.28+ ARM64
SHA-256 checksum
How to use checksums
68e3404e28353e369a0ac1acd7cbc07c17d6a1829127e1fe4d45bcf62bd93add
BLAKE2b-256 checksum
How to use checksums
071378a642ce983c5ba3eef9e5693634f0e635516678605f5b76acacd90c2e4c
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 Sep 6, 2026.

Transparency log

Release files / ctboost-0.1.59-cp312-cp312-macosx_10_15_x86_64.whl

Download URL ctboost-0.1.59-cp312-cp312-macosx_10_15_x86_64.whl
Size 1.1 MB
Tags CPython 3.12 macOS 10.15+ x86-64
SHA-256 checksum
How to use checksums
eef28a65d24b63366cf5c4b0622eeaacf36ce00114a9564fffae3ddf805401d1
BLAKE2b-256 checksum
How to use checksums
26aab1a75741f9f3314356bd62b17fc6f88ecb347d3101127085e6b1b019fd05
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 Sep 6, 2026.

Transparency log

Release files / ctboost-0.1.59-cp311-cp311-win_amd64.whl

Download URL ctboost-0.1.59-cp311-cp311-win_amd64.whl
Size 3.7 MB
Tags CPython 3.11 Windows x86-64
SHA-256 checksum
How to use checksums
538c348d3745a1f3b7d279682a95cc9c5b6384eae5d57f9cceb15acd3d180748
BLAKE2b-256 checksum
How to use checksums
3392a430d686cd218515cf6a4d218d78bede3e66a1dddc4831e29a7916d02283
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 Sep 6, 2026.

Transparency log

Release files / ctboost-0.1.59-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl

Download URL ctboost-0.1.59-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl
Size 3.8 MB
Tags CPython 3.11 Linux glibc 2.27+ x86-64 Linux glibc 2.28+ x86-64
SHA-256 checksum
How to use checksums
37b6c146fa3cc4943185ee25ba4915c15a98eb35d8762fde842664495474f445
BLAKE2b-256 checksum
How to use checksums
1d4786dcb033b15fcc8dfa4b65465803d6132432261a9e47b7b6ec65ffff4e9a
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 Sep 6, 2026.

Transparency log

Release files / ctboost-0.1.59-cp311-cp311-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl

Download URL ctboost-0.1.59-cp311-cp311-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl
Size 1.3 MB
Tags CPython 3.11 Linux glibc 2.27+ ARM64 Linux glibc 2.28+ ARM64
SHA-256 checksum
How to use checksums
4044a537320d3e3617353be6b88b07c7c31b97bc9c04b7f13fd7cfd4c7f6f503
BLAKE2b-256 checksum
How to use checksums
c43c08218204b2cf03a5c6e5a2ed0f35fd723ad153ef93a82ae70e0f601919dc
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 Sep 6, 2026.

Transparency log

Release files / ctboost-0.1.59-cp311-cp311-macosx_10_15_x86_64.whl

Download URL ctboost-0.1.59-cp311-cp311-macosx_10_15_x86_64.whl
Size 1.1 MB
Tags CPython 3.11 macOS 10.15+ x86-64
SHA-256 checksum
How to use checksums
b6de46435c90acd085e4a4e4db4e2813f545385447d70e1aa136692d081da6d6
BLAKE2b-256 checksum
How to use checksums
ac5bd23330b9005ec32e69cb6be97c1473a25aad3a8e7874f40bb53a3bd5b7fe
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 Sep 6, 2026.

Transparency log

Release files / ctboost-0.1.59-cp310-cp310-win_amd64.whl

Download URL ctboost-0.1.59-cp310-cp310-win_amd64.whl
Size 3.7 MB
Tags CPython 3.10 Windows x86-64
SHA-256 checksum
How to use checksums
cc7111a62b57084dfcd0633e74f5a31ed15e795c5c6e6fa91a38d1c52a01ff74
BLAKE2b-256 checksum
How to use checksums
ba238a05379b32f066a8528ee5d658191999590329d6596ccdda91e944b946de
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 Sep 6, 2026.

Transparency log

Release files / ctboost-0.1.59-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl

Download URL ctboost-0.1.59-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl
Size 3.8 MB
Tags CPython 3.10 Linux glibc 2.27+ x86-64 Linux glibc 2.28+ x86-64
SHA-256 checksum
How to use checksums
66656eb1a4968533f001dfe800c1fefc2132ca8b0ffe143b0b1a97991d3c29a4
BLAKE2b-256 checksum
How to use checksums
711ff9c53846948e76bc31f0cc2fa5781813c51c740047a51ff6a21f11cdea04
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 Sep 6, 2026.

Transparency log

Release files / ctboost-0.1.59-cp310-cp310-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl

Download URL ctboost-0.1.59-cp310-cp310-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl
Size 1.3 MB
Tags CPython 3.10 Linux glibc 2.27+ ARM64 Linux glibc 2.28+ ARM64
SHA-256 checksum
How to use checksums
b6248065b229b08d4d420fc602c4bf89cae688791cd8811c881313a74dde0ac9
BLAKE2b-256 checksum
How to use checksums
856a029abf9bf02a632790795a8399b549fd84566a2d800877d744338ccb3064
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 Sep 6, 2026.

Transparency log

Release files / ctboost-0.1.59-cp310-cp310-macosx_10_15_x86_64.whl

Download URL ctboost-0.1.59-cp310-cp310-macosx_10_15_x86_64.whl
Size 1.1 MB
Tags CPython 3.10 macOS 10.15+ x86-64
SHA-256 checksum
How to use checksums
0ed99bc13a76ec1a85d90faf1db4809c79b70dab2fd3ba532cdf33c86754702c
BLAKE2b-256 checksum
How to use checksums
38442ad2f054118f571bf03d3b3dd8f2f8fc2fba54052f8dd25e84f39537b22f
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 Sep 6, 2026.

Transparency log

Release files / ctboost-0.1.59-cp39-cp39-win_amd64.whl

Download URL ctboost-0.1.59-cp39-cp39-win_amd64.whl
Size 1.0 MB
Tags CPython 3.9 Windows x86-64
SHA-256 checksum
How to use checksums
95e78e4420ac0b59219a3d5a33c25418741e00d0e7c2d56a6c11d4cd66881762
BLAKE2b-256 checksum
How to use checksums
01d9617ce8f82eb06b3ed315ed7720bdaeb4535982aba8bc27053a4f08ab3896
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 Sep 6, 2026.

Transparency log

Release files / ctboost-0.1.59-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl

Download URL ctboost-0.1.59-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl
Size 1.4 MB
Tags CPython 3.9 Linux glibc 2.27+ x86-64 Linux glibc 2.28+ x86-64
SHA-256 checksum
How to use checksums
7e02b23aebf5654cefa5123611b13bc92974bbd2cf186a2ce6cb4af78cfcf226
BLAKE2b-256 checksum
How to use checksums
25d1a8fb03298e014865dfc3885e75ded4c0f3a98484886ead78ca26cf3ff8ff
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 Sep 6, 2026.

Transparency log

Release files / ctboost-0.1.59-cp39-cp39-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl

Download URL ctboost-0.1.59-cp39-cp39-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl
Size 1.3 MB
Tags CPython 3.9 Linux glibc 2.27+ ARM64 Linux glibc 2.28+ ARM64
SHA-256 checksum
How to use checksums
34852263d7eeefb623a6e36e175977f380806f4b53e83b348bdedf02dd10b23a
BLAKE2b-256 checksum
How to use checksums
e7e3a770b6e3db4213782b351f1bd346666a1788bc30f50a39f0434fb329cdb5
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 Sep 6, 2026.

Transparency log

Release files / ctboost-0.1.59-cp38-cp38-win_amd64.whl

Download URL ctboost-0.1.59-cp38-cp38-win_amd64.whl
Size 1.0 MB
Tags CPython 3.8 Windows x86-64
SHA-256 checksum
How to use checksums
774c1fc1961422caf157a174ce4bae290c23e678b7dd67a14f731417a956798e
BLAKE2b-256 checksum
How to use checksums
2567cff4d5ebc18ffdb4243349cc694bf11b2e9c6eb1fd9b6da9af4f5d6921d6
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 Sep 6, 2026.

Transparency log

Release files / ctboost-0.1.59-cp38-cp38-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl

Download URL ctboost-0.1.59-cp38-cp38-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl
Size 1.4 MB
Tags CPython 3.8 Linux glibc 2.27+ x86-64 Linux glibc 2.28+ x86-64
SHA-256 checksum
How to use checksums
8db0454f2004b8af94ea3ad065e82b1f51a93d2149c8bbdf836d61da82c68daf
BLAKE2b-256 checksum
How to use checksums
27cf47755707fb2a8955dbb80e550916887aa5d6e0234b36d143b98c7f0d3132
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 Sep 6, 2026.

Transparency log

Release files / ctboost-0.1.59-cp38-cp38-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl

Download URL ctboost-0.1.59-cp38-cp38-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl
Size 1.3 MB
Tags CPython 3.8 Linux glibc 2.27+ ARM64 Linux glibc 2.28+ ARM64
SHA-256 checksum
How to use checksums
b24c90e10c24f16919077a9cb6e6c40581e789bfa7d9bd064f8f8bd209ae5874
BLAKE2b-256 checksum
How to use checksums
235abcfe2c51c0335150a521cead067b603753624c140f53ce93672db5410c2b
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 Sep 6, 2026.

Transparency log
Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page