Open Katherine-Cao opened 2 years ago
Hello,I would like to ask if you train the Cifar10 data set through two layers of full connection. After I changed the network framework to 3 layers of full connection, the network models of gp and hp are not trained when running the RGB three-dimensional data set.
Hello, I would like to ask how the step from the formula (4) to the formula (6)in the paper is derived, and how is e obtained