Skip to content

sa-y-an/retro

Repository files navigation

Stress Detection

Do you know 77 percent of people experience stress that affects their physical health. 73 percent of people have stress that impacts their mental health. If left unchecked this can have severe consequences.

What if you could monitor your stress from the comfort of your home ?

Stress detection using physiological data.

  • I have trained a deep learning model on the multimodal WESAD Dataset. ( Details in the train_dir folder )
  • Then I deployed the model to a webapp using Django.
  • Thus the users upload their phsyiological data to the Webapp, and django leverages the classifier model to classify their data between
    1. Stress
    2. Amusement
    3. Neutral

The Architecture is shown below -

Tech Stack Used

  • Django, Html/CSS
  • Tensorflow, numpy, pandas, sklearn, etc.

Setup

  • After cloning the repo, create a virtual-environment
  • Create a config.json file corrosponding to dummy.json in the stress_detection folder.
pip install -r requirements.txt
python local.py migrate
python local.py runserver

PS : I use local.py in local environment to maintain a smooth flow, refer to my stackoverflow answer here - https://stackoverflow.com/questions/68766668/django-best-practice-for-running-switching-dev-debug-product-mode/68766902#68766902 PPS: replace manage.py with local.py to get away from heroku