Executor¶

class Executor¶: An executor is a realization of a symbolic architecture defined by a SymbolicNode. The actual forward and backward computation specified by the network architecture can be carried out with an executor.

bind(sym, ctx, args; args_grad=Dict(), aux_states=Dict(), grad_req=GRAD_WRITE)¶

Create an Executor by binding a SymbolicNode to concrete NDArray.

Parameters:

sym (SymbolicNode) – the network architecture describing the computation graph.
ctx (Context) – the context on which the computation should run.
args – either a list of NDArray or a dictionary of name-array pairs. Concrete arrays for all the inputs in the network architecture. The inputs typically include network parameters (weights, bias, filters, etc.), data and labels. See list_arguments() and infer_shape().
args_grad – TODO
aux_states –
grad_req –