Scoring engine for PMML models implemented using python
Switch branches/tags
Nothing to show
Clone or download
Pull request Compare This branch is 4 commits behind maxkferg:master.
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
scoring-engine
tests
.gitignore
AUTHORS.txt
LICENSE.txt
README.md
requires.txt
setup.py

README.md

PMML Scoring Engine

Scoring engine for PMML models implemented using python. The PMML scoring engine exposes predictive machine learning models as REST endpoints. Clients can send new observations to the scoring engine in the JSON file format. The scoring engine returns a JSON response containing the new scores

Binary Dependencies

sudo apt install python-dev
sudo apt install python-pip
sudo apt install libxml2-dev libxslt1-dev # XML dependency
sudo apt install libblas-dev liblapack-dev libatlas-base-dev gfortran # Numpy dependency
sudo apt install lib32z1-dev #lxml dependency
pip install lxml
pip install Cython

Installation

pmml-scoring-engine requires the development version of ScikitLearn for the GaussianProcessRegressor class. Install the development version from Github:

git clone https://github.com/scikit-learn/scikit-learn.git
cd scikit-learn
python setup.py build
sudo python setup.py install

Install dependencies with setuptools

sudo python setup.py install

Usage

git clone https://github.com/maxkferg/pmml-scoring-engine.git
cp some-pmml-file.pmml pmml-scoring-engine/scoring-engine/examples/pmml
cd pmml-scoring-engine/scoring-engine
python runserver.py

Client

r = HTTP.get('/pmml/some-pmml-filename.pmml',{xnew:[1,4,5,3,5,7,8,4,3,6,7,1]})
r.response # -> {mu:1.45324344,sd:3.2214342}

Contributors