Hummingbird compiles trained ML models into tensor computation for faster inference.
MIT License
Bot releases are hidden (Show)
model.load(location, override_flag=True)
. This is to prevent any potential exploit of the underlying pickle
module, so that users can verify the output checksum for model integrity, or set the override_flag
if the model is from a trusted source.Thank you to our newest contributor @giriprasad51 for his 3 PRs, and thank you to @mshr-h for all of the effort on apache-tvm to v0.15.0 and the ongoing support!
Full Changelog: https://github.com/microsoft/hummingbird/compare/v0.4.10...v0.4.11
Published by ksaur 10 months ago
This maintenance release updates to Python 3.11 (for all but TVM), adds support for the current versions (as of Jan 2024) of LGBM and ONNX related libraries.
We are still working on XGB==2.0 support and TVM==0.12.
Thanks to @mshr-h for the ongoing support!
Full Changelog: https://github.com/microsoft/hummingbird/compare/v0.4.9...v0.4.10
Published by ksaur over 1 year ago
Full Changelog: https://github.com/microsoft/hummingbird/compare/v0.4.8...v0.4.9
Published by ksaur over 1 year ago
sklearn.mixture.BayesianGaussianMixture
by @mshr-h in https://github.com/microsoft/hummingbird/pull/677
Full Changelog: https://github.com/microsoft/hummingbird/compare/v0.4.7...v0.4.8
Published by ksaur almost 2 years ago
Full Changelog: https://github.com/microsoft/hummingbird/compare/v0.4.6...v0.4.7
Published by ksaur almost 2 years ago
action/cache
to v3 by @mshr-h in https://github.com/microsoft/hummingbird/pull/619
Full Changelog: https://github.com/microsoft/hummingbird/compare/v0.4.5...v0.4.6
Published by ksaur about 2 years ago
onnxmltools>=1.6.0,<=1.11.0
instead of onnxmltools>=1.6.0
by @mshr-h in https://github.com/microsoft/hummingbird/pull/592
n_features_in_
instead of n_features_
by @mshr-h in https://github.com/microsoft/hummingbird/pull/604
python -m pip
instead of the pip
executable by @mshr-h in https://github.com/microsoft/hummingbird/pull/605
Extra special thanks to @mshr-h for the work maintaining our version dependencies and pipeline!
Full Changelog: https://github.com/microsoft/hummingbird/compare/v0.4.4...v0.4.5
Published by ksaur over 2 years ago
This minor release includes bug fixes for performance and external dependency updates.
Full Changelog: https://github.com/microsoft/hummingbird/compare/v0.4.3...v0.4.4
Published by ksaur over 2 years ago
This minor release includes bug fixes and external dependency updates.
Full Changelog: https://github.com/microsoft/hummingbird/compare/v0.4.2...v0.4.3
Published by ksaur almost 3 years ago
This minor release includes a new operator, some improvements, and some fixes due to external dependency updates.
scikit-learn==1.0.x
(#545) and the current version of scipy (scipy==1.7.x
) (#552)torch==1.10.x
on Linux/Mac that causes the tests to hang (for static length dimensions) and fail (for dynamic dimensions). This will be fixed in a subsequent release.Thanks to @akshjain83 and @bibhabasumohapatra for their contributions!
Published by ksaur about 3 years ago
This patch release includes a new operator and some documentation fixes.
Thanks to @marsupialtail, @ananiask8, and @parulnith for their contributions!
Published by ksaur over 3 years ago
This release includes integration with Prophet, bug/usability fixes, and versioning fixes.
We are excited to announce trend prediction support for Prophet! See our notebook for examples.
Published by ksaur over 3 years ago
This patch release includes several improvements related to load/store of a model:
pickle
instead of dill
for model load/save to enable the use of Spark broadcast to share the model (#498)It seems the most recent release of libomp
broke the pipeline for MacOS python3.6 and 3.7. We fixed this by pinning to an older version of libomp
(#500). While not directly related to Hummingbird, this information may be useful to MacOS users with older versions of python wanting to use the lightgbm
Hummingbird converters.
Thanks to @dbanda for the feedback and testing on our load/store.
Published by ksaur over 3 years ago
This release includes many new operators, version upgrades, and minor bug fixes.
1.19.4
to <=1.20.*
torch
to 1.8.1
Thanks to our first-time contributor @rathijit for fixes to the README.
Published by ksaur over 3 years ago
This release includes several bug fixes, version upgrades, and minor feature upgrades.
Also included are fixes to test flakiness and error messages.
Thanks to our first-time contributors @qin-xiong and @sleepy-owl
Published by ksaur almost 4 years ago
Published by ksaur almost 4 years ago
save
/load
features for models (#399)Thank you to @scnakandala and @masahi for their ongoing contributions!
Published by ksaur almost 4 years ago
In this release, we added PySpark.ML support, which will open new doors for collaboration in the Spark space! (#310)
So far, we support Bucketizer, VectorAssembler, and LogisticRegressionModel. We look forward to adding more operators soon!
This release also adds support for Pandas Dataframes both at conversion time and inference time (#300).
Thank you to @scnakandala for SparkML and the additional scikit-learn operators!
Thank you to @vumichien for contributing the README diagrams.
Published by ksaur about 4 years ago
This release adds a huge batch of new operators to scikit-learn, including pipeline support! It also includes bug fixes.
Pipeline
's including FeatureUnion
and ColumnTransformer
(#251)optimized_execution
to torchscript backend (#276)lbfgs
option (#261)A big thank you to our contributors: @scnakandala, @zhanjiezhu
Extra big thanks to @scnakandala for all of the scikit-learn converters!
Published by ksaur about 4 years ago
This release adds TorchScript as a backend, removes the problematic auto-installation of pytorch, improves syntax for the ONNX converter, adds notebook enhancements, and adds 3rd party library version upgrades. Hummingbird now provides the same Sklearn API across backends for executing inference.
This release adds TorchScript as a backend (#225).
Users can convert models with:
hummingbird.ml.convert(model, "torchscript", X)
After several reports from users across multiple platforms (in terms of both OS and underlying hardware), we changed the Hummingbird installer to require users to first install pytorch before installing Hummingbird (#246). This allows users to select the right pytorch version for a specific platform and should simplify the installation process and remove issues caused by having the wrong pytorch version installed.
For ONNX, we changed the API to have a more seamless experience, allowing users to interact with ONNX models in a consistent way with other models in Hummingbird (#231).
Instead of the user having to instanciate the ONNX runtime session:
- session = ort.InferenceSession(onnx_model.SerializeToString())
- onnx_pred = session.run(output_names, inputs)
The user can now just call predict
, predict_proba
, transform
, etc. as with other Hummingbird conversions.
+ onnx_pred = onnx_model.predict(X)
11
(#214)1.6.0
, except for with Python 3.5 it remains at 1.5.1
for compatibility reasons (#213)Thanks to @KranthiGV for the updated LGBM ONNX notebook example