Represents a neural network using a Graph
that contains mostly PipeOpModule
s.
Arguments
- graph
- shapes_in
(named
integer
)
Shape info of tensors that go intograph
. Names must begraph$input$name
, possibly in different order.- output_map
(
character
)
Which ofgraph
's outputs to use. Must be a subset ofgraph$output$name
.- list_output
(
logical(1)
)
Whether output should be a list of tensors. IfFALSE
(default), thenlength(output_map)
must be 1.
Fields
graph
::Graph
The graph (consisting primarily ofPipeOpModule
s) that is wrapped by the network.input_map
::character()
The names of the input arguments of the network.shapes_in
::list()
The shapes of the input tensors of the network.output_map
::character()
Which output elements of the graph are returned by the$forward()
method.list_output
::logical(1)
Whether the output is a list of tensors.module_list
::nn_module_list
The list of modules in the network.list_output
::logical(1)
Whether the output is a list of tensors.
See also
Other Graph Network:
ModelDescriptor()
,
TorchIngressToken()
,
mlr_learners_torch_model
,
mlr_pipeops_module
,
mlr_pipeops_torch
,
mlr_pipeops_torch_ingress
,
mlr_pipeops_torch_ingress_categ
,
mlr_pipeops_torch_ingress_ltnsr
,
mlr_pipeops_torch_ingress_num
,
model_descriptor_to_learner()
,
model_descriptor_to_module()
,
model_descriptor_union()
Examples
graph = mlr3pipelines::Graph$new()
graph$add_pipeop(po("module_1", module = nn_linear(10, 20)), clone = FALSE)
graph$add_pipeop(po("module_2", module = nn_relu()), clone = FALSE)
graph$add_pipeop(po("module_3", module = nn_linear(20, 1)), clone = FALSE)
graph$add_edge("module_1", "module_2")
graph$add_edge("module_2", "module_3")
network = nn_graph(graph, shapes_in = list(module_1.input = c(NA, 10)))
x = torch_randn(16, 10)
network(module_1.input = x)
#> torch_tensor
#> -0.3387
#> 0.4314
#> -0.0250
#> 0.0625
#> 0.0520
#> 0.1413
#> 0.3711
#> 0.0675
#> 0.0952
#> 0.0680
#> -0.4348
#> 0.3879
#> 0.2678
#> -0.0356
#> -0.0562
#> 0.3363
#> [ CPUFloatType{16,1} ][ grad_fn = <AddmmBackward0> ]