There seems to be a user need for DCLS methods where position learning is restricted to a subset of dimensions chosen from the N kernel dimensions. We'll develop a generic class called DclsN_Md (N for the convolution dimension and M for the number of N learnable position dimensions). We'll start by implementing the Dcls3_1d method.
There seems to be a user need for DCLS methods where position learning is restricted to a subset of dimensions chosen from the N kernel dimensions. We'll develop a generic class called DclsN_Md (N for the convolution dimension and M for the number of N learnable position dimensions). We'll start by implementing the Dcls3_1d method.