The output on the convolutional layer is often handed with the ReLU activation perform to bring non-linearity to your model. It will take the attribute map and replaces all the detrimental values with zero. Zero-padding—It allows us to regulate the spatial size of your output volume by padding zeros https://financefeeds.com/bitcoin-records-new-all-time-high-of-106000-best-cryptos-to-buy-now/