pub struct RunGraphRequest {
pub session_handle: String,
pub create_worker_session_called: bool,
pub graph_handle: String,
pub step_id: i64,
pub exec_opts: Option<ExecutorOpts>,
pub send: Vec<NamedTensorProto>,
pub recv_key: Vec<String>,
pub is_partial: bool,
pub is_last_partial_run: bool,
pub store_errors_in_response_body: bool,
pub request_id: i64,
}
Fields§
§session_handle: String
session_handle is the master-generated unique id for this session. If session_handle is non-empty, it must be the same as used when registering the graph. If it is empty, a single global namespace is used to search for the graph_handle.
create_worker_session_called: bool
Set to true if CreateWorkerSession
was called for session_handle
.
graph_handle: String
REQUIRED: graph_handle must be returned by a RegisterGraph call to the same WorkerService.
step_id: i64
A unique ID to distinguish different runs of the same graph.
The master generates a global unique step_id
to distinguish
different runs of the graph computation. Subgraphs communicate
(e.g., send/recv ops) with each other using step_id
to
distinguish tensors generated by different runs.
exec_opts: Option<ExecutorOpts>
Options for this step.
send: Vec<NamedTensorProto>
Runs the graph.
Sends the tensors in “send” into the graph before the run and
fetches the keys into RunGraphResponse.recv
after the run.
recv_key: Vec<String>
§is_partial: bool
True if the RunGraphRequest is a partial run request.
is_last_partial_run: bool
True if this is the last partial run request in a sequence of requests.
store_errors_in_response_body: bool
If true then some errors, e.g., execution errors that have long error messages, may return an OK RunGraphResponse with the actual error saved in the status_code/status_error_message fields of the response body. This is a workaround since the RPC subsystem may truncate long metadata messages.
request_id: i64
Unique identifier for this request. Every RunGraphRequest must have a unique request_id, and retried RunGraphRequests must have the same request_id. If request_id is zero, retry detection is disabled.
Retried RunGraphRequests are problematic because they may issue a RecvTensor that will have no corresponding sender and will wait forever. Workers use request_ids to reject retried RunGraph requests instead of waiting forever.
Trait Implementations§
Source§impl Clone for RunGraphRequest
impl Clone for RunGraphRequest
Source§fn clone(&self) -> RunGraphRequest
fn clone(&self) -> RunGraphRequest
1.0.0 · Source§fn clone_from(&mut self, source: &Self)
fn clone_from(&mut self, source: &Self)
source
. Read moreSource§impl Debug for RunGraphRequest
impl Debug for RunGraphRequest
Source§impl Default for RunGraphRequest
impl Default for RunGraphRequest
Source§impl Message for RunGraphRequest
impl Message for RunGraphRequest
Source§fn encoded_len(&self) -> usize
fn encoded_len(&self) -> usize
Source§fn encode<B>(&self, buf: &mut B) -> Result<(), EncodeError>
fn encode<B>(&self, buf: &mut B) -> Result<(), EncodeError>
Source§fn encode_length_delimited<B>(&self, buf: &mut B) -> Result<(), EncodeError>
fn encode_length_delimited<B>(&self, buf: &mut B) -> Result<(), EncodeError>
Source§fn decode<B>(buf: B) -> Result<Self, DecodeError>
fn decode<B>(buf: B) -> Result<Self, DecodeError>
Source§fn decode_length_delimited<B>(buf: B) -> Result<Self, DecodeError>
fn decode_length_delimited<B>(buf: B) -> Result<Self, DecodeError>
Source§fn merge<B>(&mut self, buf: B) -> Result<(), DecodeError>
fn merge<B>(&mut self, buf: B) -> Result<(), DecodeError>
self
. Read moreSource§fn merge_length_delimited<B>(&mut self, buf: B) -> Result<(), DecodeError>
fn merge_length_delimited<B>(&mut self, buf: B) -> Result<(), DecodeError>
self
.