word2med-backend

Usage

API documentation can be found here

Postman

A postman collection with usage examples for each endpoint is exported to Word2Med.postman_collection.json.

To use, simply import this file within postman.

Development Configuration

Follow these steps to run the word2med backend locally

Setup

Clone this repository and initialize a virtual environment in the root directory using the tool of your choice. (e.g. virtualenv). Note that the virtual environment must use version python3.10 or greater.

git clone https://github.com/ECE493Group8/word2med-backend.git
cd word2med-backend
virtualenv venv

Activate your virtual environment

. venv/bin/activate

Install dependencies from requirements.txt

pip install -r requirements.txt

Note: If you have issues in this step, ensure you have the latest pip version by running the following with you venv activated:

curl -sS https://bootstrap.pypa.io/get-pip.py | python3.10

Define model path

The backend expects a path to a trained Word2Med model, which it will query and use to form responses. See the model repository for more info on training a Word2Med model.

The model path is defined in the models.json file. An example of such a file could be:

{
  "model_1": "path/to/my.model",
  "model_2": "another/path/to/my/other.model"
}

The key is the ID of the model. This must match the key of the model ID in the frontend's src/constants/models.js file. The value is the path to the model file.

Running Backend

make run

Generating Documentation

make docs