radx-destiller

radx-destiller summarizes key results from publications generated by the NIH RADx-rad initiative.

[WORK IN PROGRESS]

---

Setup

---

Prerequisites: Miniconda3 (light-weight, preferred) or Anaconda3 and Mamba (faster than Conda)

• Install Miniconda3
• Update an existing miniconda3 installation: conda update conda
• Install Mamba: conda install mamba -n base -c conda-forge
• Install Git (if not installed): conda install git -n base -c anaconda

---

1. Clone this Repository

git clone https://github.com/radxrad/radx-destiller
cd radx-destiller

2. Create a Conda environment

The file environment.yml specifies the Python version and all required dependencies.

mamba env create -f environment.yml

3. Setup OpenAI and/or SDSC LLM API Keys

cp env_template .env

Then edit the .env file and set your API keys

Run a Notebook

---

1. Activate the conda environment

conda activate radx-destiller

2. Start jupyter lab and navigate to the notebooks directory

jupyter lab

3. When you are finished, deactivate the conda environment

conda deactivate