FetchDataFlowGraph (Fetching Output Data by Index)

Product	Supported
Atlas A3 training product/Atlas A3 inference product	√
Atlas A2 training product/Atlas A2 inference product	√
Atlas 200I/500 A2 inference product	x
Atlas inference product	x
Atlas training product	x

Fetches output data of a graph by index.

Status FetchDataFlowGraph(uint32_t graph_id, const std::vector<uint32_t> &indexes, std::vector<Tensor> &outputs, DataFlowInfo &info, int32_t timeout)

Parameter	Input/Output	Description
graph_id	Input	ID of the graph to be executed.
indexes	Input	Indexes of outputs. Values must be unique and in the range [0, N – 1], where N is the number of outputs. One or more indexes can be specified. Example: {0, 2}, indicating that the first and third outputs are fetched.
outputs	Output	Output tensors of the computational graph. You do not need to allocate the memory manually. After the execution is complete, GE allocates and initializes the memory.
info	Output	Output data flow flag. For details, see DataFlowInfo Data Types.
timeout	Input	Data fetch timeout, in milliseconds. The value -1 means no timeout limit.

Function execution result. The following table lists the results.

Parameter	Type	Description
-	Status	SUCCESS: successful data fetch. FAILED: failed data fetch. For details about error codes of ACL APIs, see Runtime APIs > Data Types and Operation APIs > aclError. For details about other error codes, see UDF Error Codes.

None

Parent topic: Graph Running APIs