Further Documentation and Links¶

There are a few more resources beyond this website, including those included with the package and those of the dependencies. For links and references, see the appropriate sections below.

Package Docstring¶

All public modules, classes, functions, and methods have docstrings. To view the rendered docstrings as documentation, see the “Full Documentation” section. To see how to access such the resources in python, see the “Package Information and Resources” section of the documentation.

Installation Documentation¶

For installation, please see the “Installation Guide” page for details. For documentation on pip and conda themselves, links are provided below.

pip¶

pip documentation: https://pip.pypa.io/en/stable/.

conda¶

General conda documentation: https://docs.conda.io/projects/conda/en/latest/index.html

Managing conda environments: https://docs.conda.io/projects/conda/en/latest/user-guide/tasks/manage-environments.html

conda channels: https://docs.conda.io/projects/conda/en/latest/user-guide/concepts/channels.html

Dependency Documentation¶

Keras Models¶

For furthur documentation on keras models, refer to tensorflow’s offical documentation through https://www.tensorflow.org/api_docs/python/tf/keras.

Gensim Dictionaries¶

For furthur documentation on gensim dictionaries, refer to gensim’s offical documentation through https://radimrehurek.com/gensim/corpora/dictionary.html.

Numpy Arrays¶

Numpy arrays are used in some under-the-hood processing. For documentation on numpy arrays in general, see https://numpy.org/devdocs/reference/generated/numpy.array.html.

For ndarray, which is the return type of the preprocess() method of the Predictor class, see https://numpy.org/doc/stable/reference/generated/numpy.ndarray.html.

NLTK¶

The Predictor class uses NLTK’s tokenization algorithms through its sent_tokenize() and word_tokenize(). More detailed documentation can be found here: https://www.nltk.org/api/nltk.tokenize.html