Unofficial implementation of "EEGformer : A transformer–based brain activity classification method using EEG signal" (Wan et al.) in pytorch.
Simple test for binary classification was held using dataset from harunshimanto/epileptic-seizure-recognition.
>>> b 1 -> loss : 0.02516809105873108
>>> b 2 -> loss : 0.02175748720765114
>>> b 3 -> loss : 0.021974051371216774
>>> b 4 -> loss : 0.025791224092245102
>>> b 5 -> loss : 0.04797504469752312
>>> b 6 -> loss : 0.023308182135224342
>>> b 7 -> loss : 0.013465813361108303
>>> b 8 -> loss : 0.01862935908138752
>>> b 9 -> loss : 0.04344731196761131
>>> b 10 -> loss : 0.025865040719509125
>>> b 11 -> loss : 0.043744731694459915
>>> b 12 -> loss : 0.015098555013537407
>>> b 13 -> loss : 0.019317815080285072
>>> b 14 -> loss : 0.02460692636668682
acc = 0.979
sen = 0.945
spe = 0.9875
>>> epoch 30 -> tp : 189, fp : 10, tn : 790, fn : 11
- Option for Dropouts
- Interpolation