search

NahidAfzali / SincNet-R

This is the implementation of the SincNet-based classifier, SincNet-R, which consists of three convolutional layers, and three deep neural network (DNN) layers for emotional EEG signals.
1 stars 0 forks source link

Is SincNet_R model architecture is same with model mentioned in other paper? #1

Open tiensu opened 1 year ago

tiensu commented 1 year ago

Hi you! I am study following paper: Zeng, H., Wu, Z., Zhang, J., Yang, C., Zhang, H., Dai, G., & Kong, W. (2019). EEG Emotion Classification Using an Improved SincNet-Based Deep Learning Model. Brain Sciences, 9(11). https://doi.org/10.3390/brainsci9110326 I am wondering your SincNet_R model is same with the SincNet_R model mentioned in above paper? In the paper, for example, in DNN section, there are three Dense (Fully Connected) layers:

image

while there are four Dense layers in your code: image

Please help me clarify that! In case of they are different, could you give me original paper which defines the model in this repository! Thank you very much!

NahidAfzali commented 1 year ago

Hi, this is the original paper: https://www.mdpi.com/2076-3425/9/11/326

tiensu commented 1 year ago

Thank @NahidAfzali ! The paper you mentioned is the same as the one I am reading. But as I was asking you about the difference between number of Dense layers. Could you help me clarify that?

Sorry about the long time to reply. I just spent the traditional holiday in my country.