Program

class paddle.static. Program

Create Python Program. Program is an abstraction of model structure, divided into computational graphs and weights. The Program has a main block that stores the computational graphs.

A set of Program usually contains startup program and main program. A startup program is set to contain some initial work, eg. initialize the Parameter, and the main program will contain the network structure and vars for train.

A set of Program can be used for test or train, in train program , Paddle will contain all content to build a train network, in test program Paddle will prune some content which is irrelevant to test, eg. backward ops and vars.

Notes:

we have default_startup_program and default_main_program by default, a pair of them will shared the parameters. The default_startup_program only run once to initialize parameters, default_main_program run in every mini batch and adjust the weights.

Returns

An empty Program.

Return type

Program

Examples

>>> import paddle
>>> import paddle.static as static

>>> paddle.enable_static()

>>> main_program = static.Program()
>>> startup_program = static.Program()
>>> with static.program_guard(main_program=main_program, startup_program=startup_program):
...    x = static.data(name="x", shape=[-1, 784], dtype='float32')
...    y = static.data(name="y", shape=[-1, 1], dtype='int32')
...    z = static.nn.fc(name="fc", x=x, size=10, activation="relu")
>>> print("main program is: {}".format(main_program))
>>> print("start up program is: {}".format(startup_program))

clone ( *args, **kwargs )

clone¶

Overloaded function.

1. clone(self: paddle.base.libpaddle.pir.Program) -> object

2. clone(self: paddle.base.libpaddle.pir.Program, arg0: pir::IrMapping) -> object

copy_to_block ( self: paddle.base.libpaddle.pir.Program, arg0: pir::IrMapping, arg1: pir::Block ) → None

copy_to_block¶

get_all_parameter_values ( self: paddle.base.libpaddle.pir.Program ) → dict[str, pir::Value]

get_all_parameter_values¶

get_output_value_by_name ( self: paddle.base.libpaddle.pir.Program, arg0: str ) → pir::Value

get_output_value_by_name¶

get_parameter_value_by_name ( self: paddle.base.libpaddle.pir.Program, arg0: str ) → pir::Value

get_parameter_value_by_name¶

get_value_by_op_id ( self: paddle.base.libpaddle.pir.Program, arg0: object ) → list[pir::Value]

get_value_by_op_id¶

global_block ( *args, **kwargs )

global_block¶

Overloaded function.

1. global_block(self: paddle.base.libpaddle.pir.Program) -> pir::Block

2. global_block(self: paddle.base.libpaddle.pir.Program) -> pir::Block

global_seed ( self: paddle.base.libpaddle.pir.Program, arg0: int ) → None

global_seed¶

list_vars ( self: paddle.base.libpaddle.pir.Program ) → list[pir::Value]

list_vars¶

num_ops ( self: paddle.base.libpaddle.pir.Program ) → int

num_ops¶

parameters_num ( self: paddle.base.libpaddle.pir.Program ) → int

parameters_num¶

set_is_test_attr ( self: paddle.base.libpaddle.pir.Program ) → None

set_is_test_attr¶

set_parameters_from ( self: paddle.base.libpaddle.pir.Program, arg0: paddle.base.libpaddle.pir.Program ) → None

set_parameters_from¶

set_state_dict ( self: paddle.base.libpaddle.pir.Program, state_dict: dict[str, paddle.base.libpaddle.DenseTensor], scope: paddle.base.libpaddle._Scope, copy_tensor: bool = False ) → None

set_state_dict¶