Get dictionary representation of argument arrrays.
The dictionary that maps name of arguments to NDArrays.
Get dictionary representation of auxiliary states arrays.
The dictionary that maps name of auxiliary states to NDArrays.
Do backward pass to get the gradient of arguments.
Gradient on the outputs to be propagated back. This parameter is only needed when bind is called on outputs that are not a loss function.
Off-Heap Bytes Allocated for this object
Copy parameters from arg_params, aux_params into executor's internal array.
: dict of name to NDArray of arguments
: dict of name to NDArray of auxiliary states.
Whether allow extra parameters that are not needed by symbol If this is True, no error will be thrown when arg_params or aux_params contain extra parameters that is not needed by the executor.
Get a debug string about internal execution plan.
Debug string of the executor.
Calculate the outputs specified by the binded symbol.
whether this forward is for evaluation purpose.
Additional specification of input arguments.
Get dictionary representation of gradient arrays.
The dictionary that maps name of arguments to gradient arrays.
native Address associated with this object
Function Pointer to the NativeDeAllocator of nativeAddress
Call NativeResource.register to get the reference
Register this object for PhantomReference tracking and in ResourceScope if used inside ResourceScope.
NativeResourceRef that tracks reachability of this object using PhantomReference
Return a new executor with the same symbol and shared memory, but different input/output shapes.
Whether to allow changing the shape of unspecified arguments.
Whether to allow allocating new ndarrays that's larger than the original.
Map of string to Shape.
executor A new executor that shares memory with this.
Takes a string and an NDArrayHandle.