Skip to content
More
Share
Explore
Meetings
Link to page
EEGNet:
Oversample classic, SMOTE, custom temporal aware oversampling:
Using class weight, normal split and no sampling
precision recall f1-score support
stop 0.33 0.04 0.07 20491
forward 0.50 0.97 0.66 34046
reverse 0.00 0.00 0.00 13835
accuracy 0.49 68372
macro avg 0.28 0.34 0.24 68372
weighted avg 0.35 0.49 0.35 68372
LSTM: Using class weight, normal split and classic oversampling
precision recall f1-score support
stop 0.33 0.04 0.07 20491
forward 0.50 0.97 0.66 34046
reverse 0.00 0.00 0.00 13835
accuracy 0.49 68372
macro avg 0.28 0.34 0.24 68372
weighted avg 0.35 0.49 0.35 68372
Dynamical Graph Convolutional Neural Networks (DGCNN):
precision recall f1-score support
stop 0.30 0.91 0.45 20491
forward 0.36 0.06 0.11 34046
reverse 1.00 0.00 0.00 13835
accuracy 0.30 68372
macro avg 0.55 0.32 0.18 68372
weighted avg 0.47 0.30 0.19 68372
Want to print your doc?
This is not the way.
This is not the way.
Try clicking the ⋯ next to your doc name or using a keyboard shortcut (
CtrlP
) instead.