Zenon conversational agent aims to help doctors caregivers, and patients. Doctors may employ Zenon to handle assessment tasks, extract results from questionnaires and give an overview of the sentiment of the patient. On the other hand, caregivers and patients may use Zenon for frequently asked questions, ask for quick help over an issue, and a report for their progress.
Make sure you have installed Python 3.8 to your system.
- Create a new virtual environment (it is recommended to use pip based virtual environmnet instead of conda environments)
- After you create your environment, activate it and run the following command. Make sure to use the specific versions of these packages to avoid dependencies issues.
pip install rasa==2.8.3 rasa-sdk==2.8.1 sanic==20.12.3
- Open a terminal, activate the new virtual environment, and run the command `rasa train -d .\domain\
- Open a terminal activate the conda environment then run the command
rasa run actions
- Open a new terminal activate the conda environment and run
rasa run
- Open a new terminal and run
ngrok http 5005
then copy the "https://" url and add it to the application.
We have created multiple "domain", "rules" and "nlu" files to be faster and easier to make changes. In the domain folder, we've created a dedicated file for every questionnaire. In addition, we've created a dedicated file for relevant questions for all cases.