mxnet.executor.Executor¶

class mxnet.executor.Executor(handle, symbol, ctx, grad_req, group2ctx)[source]¶

Executor is the object providing efficient symbolic graph execution and optimization.

Examples

>>> # typical approach to create an executor is to bind symbol
>>> a = mx.sym.Variable('a')
>>> b = mx.sym.Variable('b')
>>> c = 2 * a + b
>>> texec = c.bind(mx.cpu(), {'a': mx.nd.array([1,2]), 'b':mx.nd.array([2,3])})

__init__(handle, symbol, ctx, grad_req, group2ctx)[source]¶

Constructor, used Symbol.bind and Symbol.simple_bind instead.

Parameters: handle (ExecutorHandle) – ExecutorHandle generated by calling bind.

See also

Symbol.bind(): to create executor.

Methods

`__init__`(handle, symbol, ctx, grad_req, …)	Constructor, used Symbol.bind and Symbol.simple_bind instead.
`backward`([out_grads, is_train])	Do backward pass to get the gradient of arguments.
`copy_params_from`(arg_params[, aux_params, …])	Copy parameters from arg_params, aux_params into executor’s internal array.
`debug_str`()	Get a debug string about internal execution plan.
`forward`([is_train])	Calculate the outputs specified by the bound symbol.
`reshape`([partial_shaping, allow_up_sizing])	Return a new executor with the same symbol and shared memory, but different input/output shapes.
`set_monitor_callback`(callback)	Install callback for monitor.

Attributes

`arg_dict`	Get dictionary representation of argument arrrays.
`aux_dict`	Get dictionary representation of auxiliary states arrays.
`grad_dict`	Get dictionary representation of gradient arrays.
`output_dict`	Get dictionary representation of output arrays.