Convolution layer (CONV) The convolution layer (CONV) works by using filters that perform convolution functions as it is actually scanning the input $I$ with regard to its dimensions. Its hyperparameters incorporate the filter size $File$ and stride $S$. The ensuing output $O$ is called feature map or activation map. https://financefeeds.com/sonic-labs-introduces-innovative-points-program-to-drive-defi-growth-and-user-rewards/