Author(s): Yann LeCun

a neural net class with one convolutional layer and two fully connected layers. The main purpose of this class is to make replicable 2-hidden layer fully connected networks. (new net-cff ini inj ki0 kj0 tbl0 f1thick outthick prm)
[CLASS] (packages/gblearn2/net-cff.lsh)

makes a new net-cff module. ini inj : expected max size of input for preallocation of internal states ki0 kj0 : kernel size for first convolutional layer a standard fully-connected network can be obtained when ini = ki0 and inj = kj0 . tbl0 : table of connections between input anf feature maps for first layer f1thickk : number of hidden units in second hidden layer. outthick : number of outputs. prm an idx1-ddparam in which the parameters will be allocated.