Open rileybolen opened 1 month ago
I forgot to mention, I also tried to use the Serverless Inference API, through the model card widget, and locally on my computer, and both of these methods gave me the same error as above.
Hi @rileybolen, would it be ok to make this model public and retry? https://huggingface.co/rileybol/autotrain-1hkeo-o33ms I want to confirm if it is a problem related to permissions or not. Also having it public would help me reproduce the error. If it's too confidential, let me know and we will figure out something.
Hi @rileybolen, would it be ok to make this model public and retry? https://huggingface.co/rileybol/autotrain-1hkeo-o33ms I want to confirm if it is a problem related to permissions or not. Also having it public would help me reproduce the error. If it's too confidential, let me know and we will figure out something.
Hi, sorry I missed this last week. I have now made that model public for your debugging. I think the issue is related specifically to DETR/ResNet, because I tried training a new model with hustvl/yolos-tiny
instead of facebook/detr-resnet-101
and that one works well for me, so I don't really need to solve this issue anymore, but still let me know if you need any help solving it!
Looks like making it public also made the InferenceAPI work properly. Just tried running the "Palace" example for instance and it worked on serverless API.
Describe the bug
I have set up a basic HF Space from an AutoTrain object detection model. The model is based on
facebook/detr-resnet-101
. The space builds and loads properly, but when i submit an image for inference it fails with the error I have added to the logs.This is my
app.py
file, and I have not modified or added any other files.Reproduction
No response
Logs
System info