About the sequence of the last 1x1 conv and avgpooling

[Alloman]

In your paper the last 1x1 conv is behind the avgpooling, but in this implementation the last 1x1 conv is before the avgpooling. Is this a new trick or a mistake？

[ggjy]

Hi, these two implementions can achieve similar results. The fc-bn-swish-avgpool can be a little bit better (~0.1% top-1).