Construct a one-word-at-a-time story with an AI agent.
You'll need NLTK v.2.0.4 (https://pypi.python.org/pypi/nltk/2.0.4)
and
language-check (pip install 3to2, pip install language-check).
More at https://pypi.python.org/pypi/language-check
Command line usage:
python main.py [-c CORPUS] [-cc CORPUS_CATEGORY] [-n N] [-e EST] [-g CHECK_GRAMMAR] [-r RECORD] [-d RECORD_DIRECTORY]
-c CORPUS : A string identifier for the corpus used to train the
language model. Treebank acts as the default corpus.
-cc CORPUS_CATEGORY : A string specifying which section of CORPUS
to use as the training corpus. News is the default category.
-n N : An integer specifying the order of the language model.
N = 3 by default.
-e EST : An integer specifying the smoothing operator. Use 1 for
the Lidstone smoothing technique, 2 for the ELE one.
The Maximum Likelihood Estimator will be used as the default
smoothing operator.
-g GRAMMAR : An integer specifying the use of grammar checking.
Use 0 to disable. Enabled by default.
-r RECORD : An integer specifying whether to record the story
or not. Use 1 to enable. 0 By default.
-d RECORD_DIRECTORY : A string specifying which directory to
choose for storing the story. "stories" by default