Closed suyuzhang closed 2 years ago
Hi @shariqfarooq123, Thanks for your work! I have a question. Have you tried to use the CNN to replace the transformer?
We tried using global pool + MLP to estimate the adaptive bins (both at the bottleneck as well as the final decoder layer) but that didn't work as well as the transformer.
Hi @shariqfarooq123, Thanks for your work! I have a question. Have you tried to use the CNN to replace the transformer?