mxfusion.components.functions.gluon_func_eval¶
Members¶
-
class
mxfusion.components.functions.gluon_func_eval.
GluonFunctionEvaluation
(function_wrapper, input_variables, block_variables, output_variables, broadcastable=False)¶ Bases:
mxfusion.components.functions.function_evaluation.FunctionEvaluation
The evaluation of a function that is a wrapper of a MXNet Gluon block.
Parameters: - function_wrapper (MXFusion.functions.MXFusionGluonFunction) – the MXFusion wrapper of the MXNet Gluon block that the function evaluation is associated with.
- input_variables ({variable name : Variable}) – the input arguments to the function.
- block_variables ({variable name : Variable}) – the parameters in the Gluon block.
- nOutputs (int) – the number of outputs of the function.
- broadcastable – Whether the function supports broadcasting with the additional dimension for samples.
Type: boolean
-
block_variables
¶
-
replicate_self
(attribute_map=None)¶ This functions is a copy constructor for the object. In order to perform copy construction we first call
__new__()
on the class which creates a blank object. We then initialize that object using the method’s standard init procedures, and do any extra copying of attributes.Replicates this Factor, using new inputs, outputs, and a new uuid. Used during model replication to functionally replicate a factor into a new graph.
Parameters:
-
eval
(F, variables, always_return_tuple=False)¶ The method handling the execution of the function. The inputs arguments of the function are fetched from the variables argument according to their UUIDs.
Parameters: - F – the MXNet computation mode (mxnet.symbol or mxnet.ndarray)
- variables ({str(UUID): MXNet NDArray or MXNet Symbol}) – the set of MXNet arrays that holds the values of variables at runtime.
- always_return_tuple – Whether return a tuple even if there is
only one variables in outputs. :type always_return_tuple: boolean :returns: the return value of the function :rtypes: MXNet NDArray or MXNet Symbol