My clud app have permanert bug with medium (~3MB) JPEG image not smalll

I created an app that uses ultralitics YOLO for object detection, but when I use the small image (15KB) the app works very well:

[14:14:48] ๐Ÿ Python dependencies were installed from /mount/src/bxapp/requirements.txt using uv.

Check if streamlit is installed

Streamlit is already installed

[14:14:50] ๐Ÿ“ฆ Processed dependencies!

0: 256x256 1 bsb, 86.6ms

Speed: 9.3ms preprocess, 86.6ms inference, 1.6ms postprocess per image at shape (1, 3, 256, 256)

But if I use a medium JPEG image (3MB), the bugs are permanent in the app. The error is:

WARNING โš ๏ธ imgsz=[6411, 8020] must be multiple of max stride 32, updating to [6432, 8032]

/app/scripts/run-streamlit.sh: line 9:   211 Killed                  sudo -E -u appuser /home/adminuser/venv/bin/streamlit "$@"
[14:30:38] โ—๏ธ Streamlit server consistently failed status checks
[14:30:38] โ—๏ธ Please fix the errors, push an update to the git repo, or reboot the app.

When I run my app locally, no error like this is reported, but in the Streamlit Cloud, I have this.
Please, any help with it?

You are probably exhausting the resources allocated to your app.

This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.