This procedure is similar to the Hebbian Learning Rule with a learning rate of 1. Additionally the bias of all input and output neurons is set with the parameters p1 and p2 which have to be provided in field1 and field2. Please note that the Hebb, ClippHebb, HopFixAct and PseudoInv initialization functions are actually learning functions. The reason why those functions are called initialization functions, is the fact that there is no true training because all weights will be calculated directly. In case the values of the parameters p1 and p2 are 1 and -1 the bias of the input and output neurons will be set to ld (n) and ld (k). Where n is the number of input neurons and k is the number of output neurons. These settings are also the default settings for p1 and p2. In any other case the p1 and p2 represent the bias of the input and output neurons without any modification.