markbrownsword / puncpy Goto Github PK

# Install source from Github
mkdir PuncPy  
cd PuncPy  
git clone <https://github.com/markbrownsword/punc-py.git> .  

# Create Python 3.5 environment
conda create --name python35 --file requirements.txt

# Activate environment
conda activate python35

# Initial Setup for NLTK (in Python Console)
>>> import nltk
>>> nltk.download('punkt')
>>> nltk.download('averaged_perceptron_tagger')

# Run PuncPy (in Python Console)
>>> from puncpy.tagger import Tagger
>>> from puncpy.common import process_tagged_text
>>> tagger = Tagger('This is my own invention')
>>> result = tagger.tag_text()
>>> process_tagged_text(result)
[('This', 'DT'), ('is', 'VBZ'), ('my', 'PRP$'), ('own', 'JJ'), ('invention', 'NN')]

>>> exit()

# Test
python -m unittest discover -v

# Create Requirements File
conda list --export > requirements.txt

# Deactivate environment
conda deactivate