CondConv: Conditionally Parameterized Convolutions for Efficient Inference

Brandon YangGabriel BenderQuoc V. LeJiquan Ngiam

   Papers with code   Abstract  PDF

Convolutional layers are one of the basic building blocks of modern deep neural networks. One fundamental assumption is that convolutional kernels should be shared for all examples in a dataset... (read more)

Benchmarked Models

RANK
MODEL
REPO
CODE RESULT
PAPER RESULT
ε-REPRODUCED
BUILD
1
EfficientNet-CondConv-B1 8 experts
79.3%
--
2
EfficientNet-CondConv-B0 8 experts
77.9%
--
3
EfficientNet-CondConv-B0 4 experts
77.3%
--