def __init__(self, ndim_obs, n_actions, hidden_sizes=(50, 50, 50)):
self.pi = policies.SoftmaxPolicy(model=links.MLP(
ndim_obs, n_actions, hidden_sizes, nonlinearity=F.tanh))
self.v = links.MLP(ndim_obs,
1,
hidden_sizes=hidden_sizes,
nonlinearity=F.tanh)
super().__init__(self.pi, self.v)
def __init__(self, gpu=False):
self.q_func = policies.SoftmaxPolicy(model=QFunction())
self.v_func = VFunction()
self.common = SharedFunctionSFCNN() #SharedFunctionCNN()
if gpu:
self.q_func.to_gpu(0)
self.v_func.to_gpu(0)
self.common.to_gpu(0)
#super(A3CFFSoftmax,self).__init__(self.common,self.q_func, self.v_func)
super().__init__(self.common, self.q_func, self.v_func)
def __init__(self, ndim_obs, n_actions, hidden_sizes=(64, 64)):
self.pi = policies.SoftmaxPolicy(
model=links.MLP(ndim_obs, n_actions, hidden_sizes))
self.v = links.MLP(ndim_obs, 1, hidden_sizes=hidden_sizes)
super().__init__(self.pi, self.v)
