Open lscheinkman opened 5 years ago
@subutai Please review. The main change in this PR is the switch from BatchNorm to LayerNorm on LinearSDR module. See https://arxiv.org/abs/1607.06450
BatchNorm
LayerNorm
LinearSDR
This doesn’t affect any of our existing sparse networks right?
If affects LinearSDR. We can now use batch_size=1
However we cannot use LayerNorm on CNNSDR2d. It is only good for Linear and RNN networks
CNNSDR2d
@subutai Please review. The main change in this PR is the switch from
BatchNorm
toLayerNorm
onLinearSDR
module. See https://arxiv.org/abs/1607.06450