Skip to content

OldBonhart/MedRadService

Repository files navigation

MedRadService

MedRadService

On the verge of defeating the tyranny of biology


MNedRadService


Introductions

This is an example of the tandem of PyTorch, Django and Heroku.
This application was written as a practice.All you have to do is upload a chest x-ray, and then you will get a Grad-cam and prediction with probabilities.
If you are interested in creating any interface for other people to interact with your ML-models, then this repository can be an example and starting point for this.

About prediction model

  • This is Resnet18, trained on a dataset on NIH Chest X-ray Datasetfrom The predictive model of the service has a minimal configuration due to the limitations of the Heroku free server.

Notes on Heroku

1.Your apps configuration:

  • host: smtp.gmail.com
  • port: 587 or 465 (587 for tls, 465 for ssl)
  • protocol: tls or ssl
  • user: YOUR_USERNAME @ gmail.com
  • password: YOUR_PASSWORD
  1. The given Gmail account settings:
  • If you've turned on 2-Step Verification for your account, you might need to enter an App password.
  • Without 2-Step Verification:
    1. Allow less secure apps access to your account.
    2. Visit http://www.google.com/accounts/DisplayUnlockCaptcha and sign in with your Gmail username and password.