This repository consists of the modules necessary for the PDF2QA API endpoints.

• requirements.txt contains the all modules required to run the API.
• environment.yml contains the environment file required to set up a conda enviorment.
• app/api/pdf2qaconverter.py contains the utility functions used by the API in the form of a class (All converter related functions can go here).
• app/api/pdf2qa-api.py contains the API endpoint used to upload the PDF file (All API related functions can go here).
• app/api/openai_key.env contains the OpenAI key stored as an environment variable. Create a .env file with the same name and store the key in the directory app/api.

  OPENAI_API_KEY=your_key_goes_here
  

• Clone the repository.
• Install Python by following the official guide.
• Set up a virtual environment

  • Once you have Python installed, it is a good practice to create a virtual python environment. Virtual environments provide a clean working space for your Python packages to be installed so that you do not have conflicts with other libraries you install for other projects.
    To create a virtual environment, Python supplies a built in venv module which provides the basic functionality needed for the virtual environment setup. Running the command below will create a virtual environment named "pdf2qa-api-env" inside the current folder you have selected in your terminal / command line:

    python -m venv pdf2qa-api-env
    

    Once you've created the virtual environment, you neeed to activate it. On widnows, run:

    pdf2qa-api-env\Scripts\activate
    

    On Unix or MacOS, run:

    source pdf2qa-api-env/bin/activate
    

    Once, the environment is activated, run the following command to install libraries from the requirements.txt file:

    pip install -r requirements.txt
    

  • If using a conda environment, create a conda environment with the following command:

    conda env create -f environment.yml
    

    If the above command throws any error, please create the environment first using the command:

    conda create -n pdf2qa-api python=3.9.17
    

    Activate the conda environment by running the command:

    conda activate pdf2qa-api
    

    Then install the requirements file by running the command:

    pip install -r requirements.txt
    

• Run the API on local server using uvicorn with the following command in the app/api/ directory:

  uvicorn pdf2qa-api:app --reload
  

• To test the API endpoint, you can use tool like Postman.