Fractalize NLP

A group of Modules encompassing common components that are used in all Fractalize NLP products and deployments

string_generators

Module that enables encoding of structured domain knowledge into a conditional processing graph. The conditional processing graph is used to generate random samples of unstructured strings which represent natural language invocations of the underlying domain knowledge

Building the package

To build a deployable binary including source files execute:

$ python3 setup.py sdist bdist_wheel

This will produce fractalize_nlp/dist/fractalize_nlp--py3.7.egg

If planning to publish the package, update the version number in setup.py first.

Publishing the package

Use PyPi to distribute the package as follows:

$ twine upload dist/fractalize-nlp-<build version>

Note:

1. Make sure the package has been built with the correct version number.
2. Make sure all the code changes have been merged before publishing the package.

Testing the package

To run the test suite, run either:

$ pytest 

or

$ coverage run -m pytest

To generate the coverage report, the test suite must be invoked via coverage first. Then, run:

$ coverage report -m <module_name>/<relevant_script>.py

To generate the full coverage report, run:

$ coverage report -m

License

MIT © fractal-dataminds