Setup.Neural.Network.A3C: Setup a Feed-Forward Neural Network for the...

In MartinKies/USLR: Reinforcement Learning with R

Description

Returns a Feed-Forward Neural Network based on keras. Important: At the moment this does only support discrete action spaces. For the continous case use Setup.Neural.Network.A3C.LSTM

Usage

Setup.Neural.Network.A3C(model.par, game.object)

Arguments

model.par	A list with parameters to set up the Network e.g. as given by Get.Def.Par.Neural.Network.A3C.
game.par	A list with parameters of the game.

MartinKies/USLR documentation built on Nov. 10, 2019, 5:24 a.m.