What is the need of fusing layers in MobileNetv2?

quantization
1

I have seen the static quantization tutorial, where the layers are fused before itself. I got good result with fused layers but if I don’t fuse the layers, My accuracy is very poor.

What is the effect of layer fusion?

Please do help me with this.

1 Like
2

layer fusion is going to fuse Conv+BN into a Conv module or Conv + BN + ReLU into a ConvRelu module. this does not change numerics itself. Without fusion conv, bn and relu will be quantized independently, that might be the reason why the accuracy drops.

3

But, what is the drawback of quantizing convolution, batchnorm, relu operations independently?

4

quantizing them independently will have worse performance, and also may suffer from bigger accuracy loss.

1 Like