Generate videos snippets for interpretation: nothing happens

[slolignier]

Hi, I'm trying to classify mouse behaviors using DLC pose estimations. It went fine all the way to the generation of gifs. The app displays "Creating gifs from mp4s..." and nothing happens. I can't see new files being created, no message in the console or in firefox (streamlit). I've waited for 3 hours. The computer is quite recent (CPU i9 13900, GPU RTX 4090, win11 with conda)
Any idea of what could cause the issue? Should I just wait more? Everything went quite fast from this point. Even the classifier was trained in a few seconds, which was surprising to me (but I do have a 451 Mb _randomforest.sav file created).
In case it is useful, I've copied/pasted all steps in a workflow file attached.
Many thanks for your help!
WORKFLOW.docx

[slolignier]

Hi, sorry no, I haven't. I'm now looking at other options like labgym, deepethogram, a-soid or even try to code something myself. Le 8 mai 2024 20:06, Sam Centanni ***@***.***> a écrit : Have you solved this issue? I'm having the same problem. —Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you authored the thread.Message ID: ***@***.***>

[samcentanni]

Not sure if this was the same issue for you, but I figured it out for me at least. If it helps anyone else- this issue happened because after you select "no" to the "have frames been extracted?" question and extract frames, you have to go back and select "yes" before selecting generating a video snippet.