Use (ana)conda environments to keep everything nice, consistent, and cleanly separated.
The provided environment.yml file can be used to setup the necessary conda environment.
In the project directory via the terminal/command line:
conda create env
This will set up the development environment (after saying 'yes' to installation), which can be activated with:
activate lith_nlp
on Windows
and
source activate lith_nlp
on *nix-like systems
python -m ipykernel install --user --name lith-nlp --display-name "Lith NLP"
This project requires the following nltk data to be downloaded
import nltk
nltk.download('punkt')
nltk.download('stopwords')
nltk.download('wordnet')
nltk.download('averaged_perceptron_tagger')
Add the target directory path as the second parameter if necessary:
import nltk
nltk.download('stopwords', 'C:/a_directory/path')