Run time error for huggingface space

Hi, I’m trying to deploy a gradio image recognition app on huggingface space. After the app starts building, it works, but after a while, there is a run time error, and the error message in the “see logs” says,
“Running on local URL: http://localhost:7860/
*** Failed to connect to ec2.gradio.app:22: [Errno 110] Connection timed out”
I have no idea why it is the case and why it is running on local url instead of the public space. The app only works if I restart the space or edit a file.
One thing to add though is that I have a deep learning model (a large file, between 95-100 MB, ending in .pkl) under the “files and versions section” which I uploaded from my laptop. Could this be the reason why the app fails after some time? How do I make the app run permanently?
My url is: https://hf.space/gradioiframe/ruhwang2001/wonder-app/+/api/predict/

1 Like

cc @cbensimon

Hey there! This is because you set share=True in interface.launch. You can remove that parameter.

Hi! The same for me! When removing the parameter, the output is not displayed!