I use the Brainflow library to perform a wavelet transformation on the OpenBCI's data. I noticed that after the wavelet transformation the data always increases.
What is the reason for this? Is there a way to keep the shape the same?
I would give the data to a neural network after the wavelet transformation. The batch size is 375. the input data is always divisible by 375 (22500/375=60), but after the wavelet transformation it is no longer divisible.
I have tested different data but the shape always changed.
Example: input shape before transformation: (16, 22500) / (16, 7125) / (16, 22125) after Wavelet transformation: (16, 22525) / (16, 7151) / (16, 22151)