Open dparkSonata opened 8 months ago
Hi thanks for pointing this out. Yes, you are right that this is an error. Will fix in the next version after running some tests.
For now, it should only effect two layers, so I don't think change in the results would be significant even after the fix.
The MLP class takes an argument
last_layer_act
and assigns it toself.activation
in the__init__()
method, butself.activation
is not referenced after initialization in either theself.network
PyTorch Module orself.forward()
.https://github.com/snap-stanford/GEARS/blob/df09d7ae34e90f5ef25afa389daf7c5c589e710d/gears/model.py#L29