Classify text in contracts and research papers for the San Diego Tax Payers Association and the Charter Cities Institute
The Jupyter notebooks in this depends on a set of Python packages that are included in the environments
specified in the requirements directory. The conda-minimal.yaml file has the unversioned package
names, and the other .yaml files are exports of environments for specific platforms. The platform specifications
will provide locked, repeatable versioning, but can fail if your platform is configured differently. In those
cases, the conda-minimal.yaml should work.
To install these files, you will need to install Anaconda, but if you are skilled with the command line, you might prefer miniconda. With Anaconda, you can run the Navigator and import one of the environment files. Or, from the command line you can run:
conda env create -n text_class -f conda-minimal.yaml
Or, for the Mac OS X environment:
conda env create -n text_class -f conda-minimal.yaml
Datasets for the project are under the text_classification tag in the SDRDL data repository