0.7.6

make it so one can plug performer into t2tvit

0.7.5

make it so one can plug performer into t2tvit

0.7.4

no need to do projection to combine attention heads for T2Ts initial …

0.7.3

add DistillableT2TViT

0.7.2

remove patch size from T2TViT

0.7.1

make sure distillation still works

0.7.0

add token to token ViT

0.6.8

fix for ambiguity in broadcasting mask

0.6.7

patch for scaling factor, thanks to @urkax

0.6.5

allow for DistillableVit to still run predictions