First of all, thank you for creating such an amazing product, I’ve loved every moment of using Streamlit so far.
I recently built my first app which provides a front end to my machine learning which replicated Airbnb’s amenity detection.
It works great locally (on my MacBook Pro 16-inch), however when I deploy to Google’s App Engine, I often run into errors such as disconnecting from server or images display 0 when they’re uploaded.
I’m very new to using Google’s App Engine, so my guess is it’s because the resources I’m using might be too small to power what I’m trying to do?
- 2 cpu’s
- 2.3gb RAM
- 20gb of disk space
For context, the ML model I’m using is a fine-tuned Detectron2 model which is about 440mb (not ideal).
Just thought I’d post here to say thank you and if anyone has a little more experience with deployments who may be able to help.
In the meantime, I’m going to try upgrading my App Engine instance and seeing what happens.
Strange error message:
Image displaying 0 on iPhone:
Image displaying 0 on iPad:
It should be noted, the app has worked several times on the desktop with no hiccups but does sometimes also result in needing to be refreshed.
My full app code on GitHub: https://github.com/mrdbourke/airbnb-amenity-detection/tree/master/app