Hi,
I cleaned a little bit the code of the stack,
and I re-ran the comparison between all the networks.
It seems that from time to time the jump of accuracy to 100 % does not occurs
or takes more time than in this example for both the Neural Turing machine and
the Neural stack
EderSantana
commented
8 years ago
Hi, I cleaned a little bit the code of the stack, and I re-ran the comparison between all the networks.
It seems that from time to time the jump of accuracy to 100 % does not occurs or takes more time than in this example for both the Neural Turing machine and the Neural stack