Data preprocessing

[christinazavou]

Hello, thanks a lot for sharing your code and data:)

In the paper you mention:

"To generate clean EEG, pure EOG and pure EMG, we firstly preprocessed the data. Then segmented them into 2-second segments. Afterwards, we re-scaleed the segments to the same variance."

I guess this refers to this where both clean and noisy eeg signals are devided with the std of the noisy eeg. Why do you use the same variance? Do you assume it will help the network?

Christina

[RunKZhang]

Hello, Christina. I also have the same question as you? Do you know the reason now?