Tutorial for applying machine learning to text data within healthcare
These instructions will get you a copy of the notebook up and running on your local machine.
The easiest way to get up and running is to install Anaconda, which provides python, numerous scientific libraries as well as jupyter. Once Anaconda is installed, navigate to directory for this repository and run jupyter via the command line (using the example directory /Projects/medical-text-nlp/):
cd /Projects/medical-text-nlp/
jupyter notebookYou can also launch jupyter through the Anaconda Navigator.
The repository requires the following libraries to be installed via the command-line or within a conda terminal
conda install scikit-learn
conda install pandas
conda install -c anaconda nltk
conda install -c conda-forge wordcloudOnce the jupyter session is running navigate to the notebook tutorial.ipynb
For enhanced replicability a Docker environment is also provided with the repo, along with a requirements.txt file specifying the Python libraries required. In order to set up the project within this framework run through the following steps
Install Docker Desktop
Build the container by running the following command
docker build --pull --rm -f "Dockerfile.dockerfile" -t medicaltextnlp:latest "."
Run docker
docker run --rm -d -p 8888:8888/tcp medicaltextnlp:latest
Navigate to the following in a web browser
http://localhost:8888/
You will need an access token which can be found in the terminal logs. For more information read the Jupyter Docker Stacks docs.
This project is licensed under the MIT License - see the LICENSE.md file for details