Hi,
I’m a student at Centrale Marseille working on a research project about video deepfake detection. I would like to use your model strictly for inference, without running the full training pipeline.
However, I noticed that the repository expects a settings.yaml file, which seems to be generated only after training. Since I’m not training the model myself, this file is not available, and the inference scripts fail because of it.
Could you clarify whether there is a default settings.yaml for inference, or provide guidance on how to run inference without having to execute the training process?
Thanks a lot for your work on this project.
Best regards,
Alejandro
Hi,
I’m a student at Centrale Marseille working on a research project about video deepfake detection. I would like to use your model strictly for inference, without running the full training pipeline.
However, I noticed that the repository expects a settings.yaml file, which seems to be generated only after training. Since I’m not training the model myself, this file is not available, and the inference scripts fail because of it.
Could you clarify whether there is a default settings.yaml for inference, or provide guidance on how to run inference without having to execute the training process?
Thanks a lot for your work on this project.
Best regards,
Alejandro