Basically the issue is in the title. After a few second of running, i get this error:
It works on my local computer. If I change the code, so it runs during the build (the build after a push), no error, runs perfectly.
At this point, out of ideas.
(Visdif - a Hugging Face Space by englert)
Space seems to be running fine now
It’s still not working. I copied the minimal code to reproduce the error. It works on local, and works in other repos (Video To Subs - a Hugging Face Space by abnerh). I have no clue why it doesn’t work here.
import gradio as gr
def predict(input_file, downsample_size):
return input_file
demo = gr.Interface(
enable_queue=True,
title=“Frame selection by visual difference”,
description=“”,
fn=predict,
inputs=[gr.inputs.Video(label=“Upload Video File”),
gr.inputs.Number(label=“Downsample size”)],
outputs=gr.outputs.File(label=“Zip”),
)
demo.launch()
I’m having exactly the same issue in all my three outputs.
The issue does not appear locally because it is linked to the queue that is enabled by default when running in Spaces.
Here is the associated issue to fix this bug: Using `gr.File` as an input fails with large files when the queue is activated · Issue #2700 · gradio-app/gradio · GitHub.
Fix is included in Gradio release 3.11.
