Caffe2 - C++ API
A deep learning, cross platform ML framework
Public Types | Public Member Functions | Static Public Member Functions | Protected Member Functions | Protected Attributes
torch::autograd::Engine Struct Reference
Inheritance diagram for torch::autograd::Engine:
torch::autograd::python::PythonEngine

Public Types

using ready_queue_type = std::deque< std::pair< std::shared_ptr< Function >, InputBuffer >>
 
using dependencies_type = std::unordered_map< Function *, int >
 

Public Member Functions

virtual variable_list execute (const edge_list &roots, const variable_list &inputs, bool keep_graph, bool create_graph, const edge_list &outputs={})
 
virtual std::unique_ptr< AnomalyMetadatamake_anomaly_metadata ()
 
void queue_callback (std::function< void()> callback)
 
bool is_checkpoint_valid ()
 

Static Public Member Functions

static Engineget_default_engine ()
 Returns a reference to a static Engine instance.
 

Protected Member Functions

void compute_dependencies (Function *root, GraphTask &task)
 
void evaluate_function (FunctionTask &task)
 
ReadyQueueready_queue (at::Device device)
 
ReadyQueueready_queue_by_index (int device_index)
 
void start_threads ()
 
virtual void thread_init (int device)
 
virtual void thread_main (GraphTask *graph_task)
 
virtual void thread_on_exception (FunctionTask &task, std::exception &e)
 

Protected Attributes

std::once_flag start_threads_flag
 
std::vector< std::shared_ptr< ReadyQueue > > ready_queues
 
std::vector< std::function< void()> > final_callbacks
 
std::mutex post_callbacks_lock
 

Detailed Description

Definition at line 28 of file engine.h.


The documentation for this struct was generated from the following files: