Hi developers,
I would like to ask a question about the 4-stage sleep staging pre-trained model in SleepKit.
Is this pre-trained four-stage sleep staging model able to directly detect and predict on other external or custom sleep datasets (such as self-collected PSG/PPG/accelerometer sleep data, or other public sleep datasets not included in the official default list)?
If it supports external datasets:
Do I need to follow a fixed data format, label mapping rule, or specific preprocessing pipeline?
Are there any official guidelines or examples for adapting custom datasets to the 4-stage staging model?
If it cannot be used directly, what is the recommended way to fine-tune or adapt the model for new unseen sleep datasets?
Thanks a lot for your guidance!
Hi developers,
I would like to ask a question about the 4-stage sleep staging pre-trained model in SleepKit.
Is this pre-trained four-stage sleep staging model able to directly detect and predict on other external or custom sleep datasets (such as self-collected PSG/PPG/accelerometer sleep data, or other public sleep datasets not included in the official default list)?
If it supports external datasets:
Do I need to follow a fixed data format, label mapping rule, or specific preprocessing pipeline?
Are there any official guidelines or examples for adapting custom datasets to the 4-stage staging model?
If it cannot be used directly, what is the recommended way to fine-tune or adapt the model for new unseen sleep datasets?
Thanks a lot for your guidance!