SGTLearn

sgt visualization

sgtlearn is a Python package for learning Shape Generalized Trees (SGTs).

🌳 Shape Generalized Trees (SGTs): A class of decision trees where each node applies a learnable, axis-aligned shape function to one or two logical features for non-linear and interpretable splits.
👁 Interpretability: Each node's shape function can be visualized directly.
⚡ ShapeCART Algorithm: An efficient induction method for learning SGTs from data.
🔀 Extensions:
- Shape²GT (S²GT): Bivariate shape functions for richer splits.
- SGT_K: Multi-way branching generalization.
- Shape²CART & ShapeCART_K: Algorithms for learning S²GTs and SGT_Ks.

[!NOTE] This codebase is an efficient implementation of the algorithms in "Empowering Decision Trees via Shape Function Branching." See the ROADMAP for implementation status and the canonical research code for the paper's original implementation.

Installation

pip install sgtlearn

Wheels are published for CPython 3.11–3.14 on Linux, macOS, and Windows (x86_64 + arm64); no compiler is needed for a binary install. To build from source instead, see Developer Setup.

Quick Start

import matplotlib.pyplot as plt
from sklearn.model_selection import train_test_split
from sgtlearn import SGTClassifier, plot_tree, make_plus

X, y = make_plus(n_samples=1500, grid=3, margin=0.07, random_state=42)

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

model = SGTClassifier(max_depth=4, random_state=42)
model.fit(X_train, y_train)

plot_tree(model, X=X_train)
plt.show()

Read the full docs here: https://sgtlearn.readthedocs.io/en/latest/index.html

Outer growth now always selects the best available regularized split and strictly respects max_leaf_nodes, including multiway splits. Its impurity improvement uses total sample weight and the mean across target outputs. min_impurity_decrease, branching_penalty (new, default 0.0), and pairwise_penalty subtract constant costs in these units; existing positive penalties may need retuning. Inner CART and TAO retain their separate settings. See outer-growth semantics for the scoring formula.

For SGTRegressor and RandomSGForestRegressor, MAE (criterion="absolute_error" or "mae") leaves coordinate descent disabled by default. Each top-level fit emits one UserWarning, including parallel forest fits. Set the environment variable SGTLEARN_MAE_CD=1 before fitting to enable it and omit the warning. The native flag also accepts exactly true, TRUE, or yes; other values keep CD disabled. This does not change TAO settings or affect classification/MSE.

Developer Setup

Use a project-local virtual environment (.venv) so Python, pytest, and scikit-learn stay isolated and reproducible. Pick one of the paths below (uv is recommended). All require Python ≥ 3.11.

Path 1 — `uv` (recommended)

uv provisions a hermetic CPython and resolves the dev extras in one step:

uv sync --all-extras
source .venv/bin/activate   # Windows: .venv\Scripts\activate

Path 2 — `pip` + `venv` (editable)

python3 -m venv .venv
source .venv/bin/activate   # Windows: .venv\Scripts\activate
pip install -U pip
pip install -e ".[dev]"

The editable install builds the C++ extensions via scikit-build-core and installs the sgtlearn package plus native modules into .venv.

Path 3 — `pip` non-editable (into the active environment)

pip install .
pip install ".[dev]"   # dev extras (pytest, scikit-learn) only if needed

Anaconda users: Do not bootstrap the venv from an Anaconda Python. Anaconda ships a libstdc++.so.6 that lags the symbol versions produced by recent system compilers (gcc ≥ 13), so the install succeeds but import sgtlearn fails with ImportError: GLIBCXX_3.4.NN not found. Use a non-Anaconda Python — e.g. uv venv --python 3.12 .venv (downloads a hermetic CPython), pyenv, or your distro's python3.

Build Workflow (scikit-build + CMake)

pip install . drives this build path:

pyproject.toml selects scikit_build_core.build as the backend.
CMake is configured from cpp/CMakeLists.txt.
Each file in cpp/bindings/*.cpp becomes one pybind11 module target.
After each module is built, pybind11-stubgen generates a matching .pyi.
The .pyi is generated and installed in the same location as the module .so.

C++ Folder Conventions

cpp/include/sgtlearn/: public headers for the core C++ API.
cpp/src/: internal C++ implementation for the core library.
cpp/bindings/: pybind11 binding entrypoints; one .cpp file maps to one Python extension module.
cpp/tests/: C++ unit tests consumed by the cpp_tests executable target.

CMake Targets

sgtlearn_core (static library): shared C++ logic used by Python modules and tests.
<module_name> (pybind11 module, one per file in cpp/bindings/): compiled extension modules installed into the package.
cpp_tests (Catch2 executable): optional C++ test target, controlled by:
- -DSGTLEARN_BUILD_TESTS=ON (build C++ tests)
- -DSGTLEARN_BUILD_TESTS=OFF (default for pip install; the CMake option itself defaults to ON, but pyproject.toml overrides this so wheels don't ship test binaries)

Overriding CMake options from `pip`

Example (build C++ tests for one install):

pip install . --config-settings=cmake.args="-DSGTLEARN_BUILD_TESTS=ON"

License

MIT License - see LICENSE for details.

Contributing

Contributions are welcome. Please feel free to submit a pull request.

Citation

If you use this package in your research, please cite:

@article{upadhya2026empowering,
  title={Empowering Decision Trees via Shape Function Branching},
  author={Upadhya, Nakul and Cohen, Eldan},
  journal={Advances in Neural Information Processing Systems},
  volume={38},
  pages={122263--122308},
  year={2026}
}

Additionally, check out our other works on our lab website.

License
- OSI Approved :: MIT License
Operating System
- OS Independent
Programming Language
- Python :: 3

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

sgtlearn-0.3.1.tar.gz (994.5 kB view details)

Uploaded Sep 11, 2026 Source

Built Distributions

If you're not sure about the file name format, learn more about wheel file names.

The dropdown lists show the available interpreters, ABIs, and platforms. Enable javascript to be able to filter the list of wheel files.

sgtlearn-0.3.1-cp314-cp314-win_arm64.whl (1.8 MB view details)

Uploaded Sep 11, 2026 CPython 3.14Windows ARM64

sgtlearn-0.3.1-cp314-cp314-win_amd64.whl (1.9 MB view details)

Uploaded Sep 11, 2026 CPython 3.14Windows x86-64

sgtlearn-0.3.1-cp314-cp314-win32.whl (1.8 MB view details)

Uploaded Sep 11, 2026 CPython 3.14Windows x86

sgtlearn-0.3.1-cp314-cp314-musllinux_1_2_x86_64.whl (3.3 MB view details)

Uploaded Sep 11, 2026 CPython 3.14musllinux: musl 1.2+ x86-64

sgtlearn-0.3.1-cp314-cp314-musllinux_1_2_aarch64.whl (3.2 MB view details)

Uploaded Sep 11, 2026 CPython 3.14musllinux: musl 1.2+ ARM64

sgtlearn-0.3.1-cp314-cp314-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (2.3 MB view details)

Uploaded Sep 11, 2026 CPython 3.14manylinux: glibc 2.27+ x86-64manylinux: glibc 2.28+ x86-64

sgtlearn-0.3.1-cp314-cp314-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl (2.2 MB view details)

Uploaded Sep 11, 2026 CPython 3.14manylinux: glibc 2.26+ ARM64manylinux: glibc 2.28+ ARM64

sgtlearn-0.3.1-cp314-cp314-macosx_11_0_arm64.whl (1.9 MB view details)

Uploaded Sep 11, 2026 CPython 3.14macOS 11.0+ ARM64

sgtlearn-0.3.1-cp314-cp314-macosx_10_15_x86_64.whl (2.0 MB view details)

Uploaded Sep 11, 2026 CPython 3.14macOS 10.15+ x86-64

sgtlearn-0.3.1-cp313-cp313-win_arm64.whl (1.8 MB view details)

Uploaded Sep 11, 2026 CPython 3.13Windows ARM64

sgtlearn-0.3.1-cp313-cp313-win_amd64.whl (1.9 MB view details)

Uploaded Sep 11, 2026 CPython 3.13Windows x86-64

sgtlearn-0.3.1-cp313-cp313-win32.whl (1.8 MB view details)

Uploaded Sep 11, 2026 CPython 3.13Windows x86

sgtlearn-0.3.1-cp313-cp313-musllinux_1_2_x86_64.whl (3.3 MB view details)

Uploaded Sep 11, 2026 CPython 3.13musllinux: musl 1.2+ x86-64

sgtlearn-0.3.1-cp313-cp313-musllinux_1_2_aarch64.whl (3.2 MB view details)

Uploaded Sep 11, 2026 CPython 3.13musllinux: musl 1.2+ ARM64

sgtlearn-0.3.1-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (2.3 MB view details)

Uploaded Sep 11, 2026 CPython 3.13manylinux: glibc 2.27+ x86-64manylinux: glibc 2.28+ x86-64

sgtlearn-0.3.1-cp313-cp313-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl (2.2 MB view details)

Uploaded Sep 11, 2026 CPython 3.13manylinux: glibc 2.26+ ARM64manylinux: glibc 2.28+ ARM64

sgtlearn-0.3.1-cp313-cp313-macosx_11_0_arm64.whl (1.9 MB view details)

Uploaded Sep 11, 2026 CPython 3.13macOS 11.0+ ARM64

sgtlearn-0.3.1-cp313-cp313-macosx_10_13_x86_64.whl (2.0 MB view details)

Uploaded Sep 11, 2026 CPython 3.13macOS 10.13+ x86-64

sgtlearn-0.3.1-cp312-cp312-win_arm64.whl (1.8 MB view details)

Uploaded Sep 11, 2026 CPython 3.12Windows ARM64

sgtlearn-0.3.1-cp312-cp312-win_amd64.whl (1.9 MB view details)

Uploaded Sep 11, 2026 CPython 3.12Windows x86-64

sgtlearn-0.3.1-cp312-cp312-win32.whl (1.8 MB view details)

Uploaded Sep 11, 2026 CPython 3.12Windows x86

sgtlearn-0.3.1-cp312-cp312-musllinux_1_2_x86_64.whl (3.3 MB view details)

Uploaded Sep 11, 2026 CPython 3.12musllinux: musl 1.2+ x86-64

sgtlearn-0.3.1-cp312-cp312-musllinux_1_2_aarch64.whl (3.2 MB view details)

Uploaded Sep 11, 2026 CPython 3.12musllinux: musl 1.2+ ARM64

sgtlearn-0.3.1-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (2.3 MB view details)

Uploaded Sep 11, 2026 CPython 3.12manylinux: glibc 2.27+ x86-64manylinux: glibc 2.28+ x86-64

sgtlearn-0.3.1-cp312-cp312-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl (2.2 MB view details)

Uploaded Sep 11, 2026 CPython 3.12manylinux: glibc 2.26+ ARM64manylinux: glibc 2.28+ ARM64

sgtlearn-0.3.1-cp312-cp312-macosx_11_0_arm64.whl (1.9 MB view details)

Uploaded Sep 11, 2026 CPython 3.12macOS 11.0+ ARM64

sgtlearn-0.3.1-cp312-cp312-macosx_10_13_x86_64.whl (2.0 MB view details)

Uploaded Sep 11, 2026 CPython 3.12macOS 10.13+ x86-64

sgtlearn-0.3.1-cp311-cp311-win_arm64.whl (1.8 MB view details)

Uploaded Sep 11, 2026 CPython 3.11Windows ARM64

sgtlearn-0.3.1-cp311-cp311-win_amd64.whl (1.9 MB view details)

Uploaded Sep 11, 2026 CPython 3.11Windows x86-64

sgtlearn-0.3.1-cp311-cp311-win32.whl (1.8 MB view details)

Uploaded Sep 11, 2026 CPython 3.11Windows x86

sgtlearn-0.3.1-cp311-cp311-musllinux_1_2_x86_64.whl (3.3 MB view details)

Uploaded Sep 11, 2026 CPython 3.11musllinux: musl 1.2+ x86-64

sgtlearn-0.3.1-cp311-cp311-musllinux_1_2_aarch64.whl (3.2 MB view details)

Uploaded Sep 11, 2026 CPython 3.11musllinux: musl 1.2+ ARM64

sgtlearn-0.3.1-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (2.3 MB view details)

Uploaded Sep 11, 2026 CPython 3.11manylinux: glibc 2.27+ x86-64manylinux: glibc 2.28+ x86-64

sgtlearn-0.3.1-cp311-cp311-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl (2.2 MB view details)

Uploaded Sep 11, 2026 CPython 3.11manylinux: glibc 2.26+ ARM64manylinux: glibc 2.28+ ARM64

sgtlearn-0.3.1-cp311-cp311-macosx_11_0_arm64.whl (1.9 MB view details)

Uploaded Sep 11, 2026 CPython 3.11macOS 11.0+ ARM64

sgtlearn-0.3.1-cp311-cp311-macosx_10_9_x86_64.whl (2.0 MB view details)

Uploaded Sep 11, 2026 CPython 3.11macOS 10.9+ x86-64

File details

Details for the file sgtlearn-0.3.1.tar.gz.

File metadata

Download URL: sgtlearn-0.3.1.tar.gz
Upload date: Sep 11, 2026
Size: 994.5 kB
Tags: Source
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for sgtlearn-0.3.1.tar.gz
Algorithm	Hash digest
SHA256	`35d2c8264d4080eef2e7ece65a330840e5bd2a7779e1a83e803ced9ce18e28cd`
MD5	`7ad1eda984a7b165c5235291efe1b8fc`
BLAKE2b-256	`ad71a3fdf1432d73d359c7008a7f16e7fa13b35209255a6b8d90b58d5d383482`

Algorithm	Hash digest
SHA256	`643c1ee4b9da9e4911277b49d6a25783347149af6f87f3241528c2557bbb3fe6`
MD5	`5c411b229c0f3f82f3e498ecf99e21f8`
BLAKE2b-256	`3abcf4e2cea2790e5d3a7b28cbbe55e357a5dbb66fc2e8ffa401999874a8175b`

Algorithm	Hash digest
SHA256	`a5b48b330fe02bd2edad2bdb3832c6ed4cfa0c3db9f51a1c12fc0712dd2cd987`
MD5	`77c3cfd0362b93a9f720cdfa92f61f58`
BLAKE2b-256	`d1693eb4ac7b38c59180da94176ef8cd8f9139f37121f02411e30c841bf72cc1`

Algorithm	Hash digest
SHA256	`1ad39e7008c6472e624612b01c56f4880b36d12982f5778565aff895f0a2d6bb`
MD5	`0c75882c23c2d82a1de40902fba3c6f4`
BLAKE2b-256	`b18e4a4dc557f54d666e31fd69570b25ca7698d5f73d56db53a4cb0c0e92f360`

Algorithm	Hash digest
SHA256	`df34430d3901c32720cb8a2dc2b87745a865492f66498285044a70d92ab3c6fe`
MD5	`95c471505a7d729965b23a72e08ea7a6`
BLAKE2b-256	`781b4ec70a549ae47f287efc510f79b85792d9eeff587b870dd3b8327a41786f`

Algorithm	Hash digest
SHA256	`7dacab20ba4e12a010ff63055648a25599430723f835e29d49294f528ec74ac8`
MD5	`1d841ab256e7f0e3c4f25589277a7f22`
BLAKE2b-256	`5471ad3ea2a21eb83d69c9dd69dba2e6fc7880133873fa4052ad112a3b312076`

Algorithm	Hash digest
SHA256	`84c611bc6d8eef29f7a6052c39c98c89607efd47199abeede07d762e6ab77e67`
MD5	`880db0ba009ab150e76f09803e894287`
BLAKE2b-256	`0234582f5cc8f11783694ee7efdf088f30d70422233e67c9ae661c761f99d6c3`

Algorithm	Hash digest
SHA256	`df9a84aa072c90c0d7df31c64a08288ed177f4da0bcaf68acbde34afd3b5e72e`
MD5	`05ad0bc6dce8949bdff53758df086a34`
BLAKE2b-256	`1b11fbb202fcea15445d6e95e18558d99533504232fb99bbf9317cc2c9f16bab`

Algorithm	Hash digest
SHA256	`16782fee5840b4dab0f67dd3e63fac8d0fb7f74cc00bd3ee154fd20cf1ff396f`
MD5	`d03419defd1989b9f10391259ed0630a`
BLAKE2b-256	`debbc31531b6871eaef221f72984982e6f113164c50986958b8ec62bfa789dd4`

Algorithm	Hash digest
SHA256	`79c7e2ff89613065461c36eaff56b18994a1d7b017ffaa147b36d5e39b5aa562`
MD5	`e76644498780cb19509d667a57806e29`
BLAKE2b-256	`7d24341efeffc8e49f92022bfda9203b07811993cdf2c4587a8c8b62730732c5`

Algorithm	Hash digest
SHA256	`9c4c71e0af7c1ba1df8f35032feb89ccbc2593685d51e433ad9578afd0c28216`
MD5	`d9200c2d008d44339e8b9d700f6262d8`
BLAKE2b-256	`86375449a22fcb63f9e1c0ec2c9b35c0d5d1f733e7829dbb855f8be2150098d5`

Algorithm	Hash digest
SHA256	`6c1a196cc49d71047ff2ce2c043e0fda09217ff0ed3ff8d6788cc36b1c176218`
MD5	`8925c2de5174909d72e2da3eed6bfc44`
BLAKE2b-256	`0ed2e3cc97ca1896b29d4b76176963a04e4c35828e3c26ff876b890fe30a41e5`

Algorithm	Hash digest
SHA256	`38dbea93b89d0839ff60d43b69255b77e23fb71e1661826835fd33726d6dfb41`
MD5	`0fd7f7046789c42bdf0cbe1b639ac969`
BLAKE2b-256	`204eb8a87a06bf5e793305f4e9de08732b5f81fb9c83aff7b1895638a1925d57`

Algorithm	Hash digest
SHA256	`7c0e169804d850d7efc48b32f4ffb6ce0ad0bb696032820b146d4737674cad0f`
MD5	`946e8c5f3d5be70474c47131faea862c`
BLAKE2b-256	`12ee864ea6caadfe995b461263b0f8916bfbc08806c7aadf9fa0f10167931e1f`

Algorithm	Hash digest
SHA256	`9839dc69b3bd586a12fc7d230a914cb1357d265d57dcfb7c7ec773b591a1aed4`
MD5	`b1c4f495102d036b5beb818d35d44d21`
BLAKE2b-256	`24d84254a4048cea669d871d1c189840efaaacb8c2be2636496189a7c50c639a`

Algorithm	Hash digest
SHA256	`cc1249dfd01b33484fc35b88c5cf22714448fbf6817b7bbcd038e488f6927d7a`
MD5	`616ba91d22b0d6a997ad1f44ede3a767`
BLAKE2b-256	`aa6de34bfea979a0514b24e94f839655223a41e82b4afa595e963b7561feaba5`

Algorithm	Hash digest
SHA256	`c301c806ebf67e069d41260f462e2c3cffa1bf541d571b34b0a636de41392e0e`
MD5	`52c79f3ed4a2de27737517007b506afa`
BLAKE2b-256	`46ea4e4a99c08bc14241f36e6aeb1dc3e94da65f14f086be28cecfe69005b0a0`

Algorithm	Hash digest
SHA256	`cbda37be93a0db0e1b868a21357c6e246682491add378fbf396457275ebce9de`
MD5	`c464b07db21f6d46660be9cfe5560a96`
BLAKE2b-256	`8656e195c8127cf327501a2bffdd7aa5d157049def8f51b5d966f12ac86dce0a`

Algorithm	Hash digest
SHA256	`2d665ed770c2032fde701348dd63048551c6428d8416c355614759a4987d4044`
MD5	`694f6a5c314b14948e9612e3b94d1731`
BLAKE2b-256	`0dc6834480af66f7d9c905c2214617b65cfaafc0523afabd6f7bc79ca60b45a7`

Algorithm	Hash digest
SHA256	`709cd41e8c9512dc6033c13b7bc655ecd39ee8281bd6628e2781cb34bf2e6970`
MD5	`01a7020945deffb7cc10487e9f2418fd`
BLAKE2b-256	`f353c95d30f05c0c8d4bf921d076ec29ba2b7df61dbc6610f997acbaf8eb5bb4`

Algorithm	Hash digest
SHA256	`80fa4b48a02b45027f8a10dc2d6335a5f42578e2d7ecfd29cc6a803eb1858c42`
MD5	`5adb75a5a78b94868be1a338ac139bb1`
BLAKE2b-256	`dd231d33a8eda66f1f049922aa0e6e75cee14beaf6f07643128715f57ba61e1d`

sgtlearn 0.3.1

SGTLearn

Installation

Quick Start

Developer Setup

Path 1 — uv (recommended)

Path 2 — pip + venv (editable)

Path 3 — pip non-editable (into the active environment)

Build Workflow (scikit-build + CMake)

C++ Folder Conventions

CMake Targets

Overriding CMake options from pip

License

Contributing

Citation

Key dates

2 maintainers

Credits

License

Requires

Provides Extra

Classifiers

Download files

Source Distribution

Built Distributions

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Path 1 — `uv` (recommended)

Path 2 — `pip` + `venv` (editable)

Path 3 — `pip` non-editable (into the active environment)

Overriding CMake options from `pip`

Algorithm	Hash digest
SHA256	`08a794ab345442fef49f2c6e1057615c629a4a324174742ed9e742a84d96f316`
MD5	`96e00bca0d7b51e6eb93d9d6ae2ee105`
BLAKE2b-256	`2c77de82291f1c3eb3ce210af366010cf9c87b5f8a04f585be3ada54c502ee5b`

Algorithm	Hash digest
SHA256	`ff3918ad8c1d2943fa2ce572f1d22be6d622dbbca26bcee43323712a13d9af09`
MD5	`0e7bdec637a84003b177532f3fa64734`
BLAKE2b-256	`4a8859c86fcbadcfb444905fad382aed2300ac277f3b10e902c573a6b84170ad`

Algorithm	Hash digest
SHA256	`eb398bab538afb11596024bdffc04a260e047841074878ce1e229cb8cfc324d4`
MD5	`68e84c0d4fde633d1095c08130d90c5f`
BLAKE2b-256	`37c400f612f13296cae950c03056fafa7d97b366ba6840b13c78db072ae97a7f`

Algorithm	Hash digest
SHA256	`3ef2ff0d22f773831e2048c379daceef415f0f485fe224f4bdaf5a4902038abb`
MD5	`9adb2bdcdfa1710ed5a6986ea7ef50f6`
BLAKE2b-256	`de6bd1088f31a9319cc7511937b82acc8bc0c1f7d647650271c0fb0520d04448`

Algorithm	Hash digest
SHA256	`4c3639b84136a41054766f3097174f213d8c2b3518b1cdc6a1407601f3b5f2b5`
MD5	`abfeaa66a03b469454a15ce10dee0fc1`
BLAKE2b-256	`b92dfc6b6e1a5f187f4c6a4093ce745c75ecfac9b58631ecebeee33eb71f4b48`

Algorithm	Hash digest
SHA256	`b987edc6e6030dd52a2864c2ee7347bfedc9c97ccaf8a5c1414dd1295c117bd7`
MD5	`e3dd2287d9b0927b9b6067cada21fadc`
BLAKE2b-256	`65de15188fc744bfb359e95b8029729651116efa9b6a8c627d0097eee9f1c25d`

Algorithm	Hash digest
SHA256	`b6533e0f3f7bcc6cd9dda54eea112ca69bdc976e13423853c87cb148943e5922`
MD5	`ee892818c5072138b3567805acce629d`
BLAKE2b-256	`a4c5643161b5a09b8a9333a6d391013093e52a997f9b96d3d72724fa37472233`

Algorithm	Hash digest
SHA256	`2931544d1167f1116ddb7e1da158002fee3375058257646db8e49157a7afea5c`
MD5	`bba13b3d3e8d192e80a7f74e2bb23acd`
BLAKE2b-256	`8f01ac4f98d1d2a5c9c7a195956f226e4e646ee49879dc0b7888a9294777496f`

Algorithm	Hash digest
SHA256	`db3b3a0f4b6c3fac619f032ff50e0cfaf4d4f684fd21771b512c12250954616b`
MD5	`9994879634b22684432ea08b14ec1b97`
BLAKE2b-256	`dd076a39593cf73f7d9587cfe491e878857d38ca510650b5cd614fe7ba8951c1`

Algorithm	Hash digest
SHA256	`57e34df6e5e667f0ad6e57bbefa7f1de1a0ee2f823a2b929faf2bd5be860c6a1`
MD5	`efbfaa209ba6847bc1814c63a42cd99e`
BLAKE2b-256	`32e313db4f1065a8f0f07253229a28740a70f93fb76637539f52266c9bcee806`