MixConv: Mixed Depthwise Convolutional Kernels

Open paper
Benchmark Model Rank Results
image-classification-on-imagenetMixNet-L#838Top 1 Accuracy: 78.9%Number of params: 7.3MGFLOPs: 0.565
image-classification-on-imagenetMixNet-M#931Top 1 Accuracy: 77%Number of params: 5.0MGFLOPs: 0.360
image-classification-on-imagenetMixNet-S#973Top 1 Accuracy: 75.8%Number of params: 4.1MGFLOPs: 0.256

Rank counts only results with a code link.