I am trying to deploy my ML app to streamlit Share. But after i enter my input which is a image i get error:-
Log File is Upload files for free - logs-akshat231-x-ray-app-main-ml_app_new.py-2022-02-23T08_11_42.044Z.txt - ufile.io
My requirements.txt is:-
My Packages.txt is:-
Here is github link for the repo: - GitHub - akshat231/X-Ray-App
Your app seems to be installing both opencv-python and opencv-python-headless. I think this is the cause of the error.
opencv-python seems to be a prerequisite of some other library (my guess is keras_ocr).
Although I don’t know how to fix it, I am just posting it here so that it may help others in helping you.
The problem is now solved, apparently, Streamlit doesn’t allow any kind of algorithm to be put in the code. I was trying to refine images by calling “Keras Pipeline” which is not allowed. I think you can only call the model and not any explicit ML algorithm. Once i removed that part of the overall code. It starts working good.
Thanks for the reply, though.
This isn’t quite your issue. Streamlit doesn’t prohibit any specific code from running (other than crypto mining, perhaps). The issue is that your community tier instance doesn’t have enough resources to run your ML pipeline. That’s a function of it being a free tier, not that Streamlit doesn’t run Keras code.