Mixture of Experts #1
Description
The network you wrote looks more like Mixture of Experts.
α 1 (W1 ∗x)+...+α n (Wn ∗x)
Using n times conv operations.
Has the author ever tried CondConv: (α1 W1 + ... + αn Wn ) ∗ x.
{{ message }}
The network you wrote looks more like Mixture of Experts.
α 1 (W1 ∗x)+...+α n (Wn ∗x)
Using n times conv operations.
Has the author ever tried CondConv: (α1 W1 + ... + αn Wn ) ∗ x.