Package org.platanios.tensorflow.proto
Interface RunOptions.ExperimentalOrBuilder
-
- All Superinterfaces:
com.google.protobuf.MessageLiteOrBuilder
,com.google.protobuf.MessageOrBuilder
- All Known Implementing Classes:
RunOptions.Experimental
,RunOptions.Experimental.Builder
- Enclosing class:
- RunOptions
public static interface RunOptions.ExperimentalOrBuilder extends com.google.protobuf.MessageOrBuilder
-
-
Method Summary
All Methods Instance Methods Abstract Methods Modifier and Type Method Description long
getCollectiveGraphKey()
If non-zero, declares that this graph is going to use collective ops and must synchronize step_ids with any other graph with this same group_key value (in a distributed computation where tasks run disjoint graphs).RunOptions.Experimental.RunHandlerPoolOptions
getRunHandlerPoolOptions()
.org.platanios.tensorflow.proto.RunOptions.Experimental.RunHandlerPoolOptions run_handler_pool_options = 3;
RunOptions.Experimental.RunHandlerPoolOptionsOrBuilder
getRunHandlerPoolOptionsOrBuilder()
.org.platanios.tensorflow.proto.RunOptions.Experimental.RunHandlerPoolOptions run_handler_pool_options = 3;
boolean
getUseRunHandlerPool()
If true, then operations (using the inter-op pool) across all session::run() calls will be centrally scheduled, optimizing for (median and tail) latency.boolean
hasRunHandlerPoolOptions()
.org.platanios.tensorflow.proto.RunOptions.Experimental.RunHandlerPoolOptions run_handler_pool_options = 3;
-
Methods inherited from interface com.google.protobuf.MessageOrBuilder
findInitializationErrors, getAllFields, getDefaultInstanceForType, getDescriptorForType, getField, getInitializationErrorString, getOneofFieldDescriptor, getRepeatedField, getRepeatedFieldCount, getUnknownFields, hasField, hasOneof
-
-
-
-
Method Detail
-
getCollectiveGraphKey
long getCollectiveGraphKey()
If non-zero, declares that this graph is going to use collective ops and must synchronize step_ids with any other graph with this same group_key value (in a distributed computation where tasks run disjoint graphs).
int64 collective_graph_key = 1;
- Returns:
- The collectiveGraphKey.
-
getUseRunHandlerPool
boolean getUseRunHandlerPool()
If true, then operations (using the inter-op pool) across all session::run() calls will be centrally scheduled, optimizing for (median and tail) latency. Consider using this option for CPU-bound workloads like inference.
bool use_run_handler_pool = 2;
- Returns:
- The useRunHandlerPool.
-
hasRunHandlerPoolOptions
boolean hasRunHandlerPoolOptions()
.org.platanios.tensorflow.proto.RunOptions.Experimental.RunHandlerPoolOptions run_handler_pool_options = 3;
- Returns:
- Whether the runHandlerPoolOptions field is set.
-
getRunHandlerPoolOptions
RunOptions.Experimental.RunHandlerPoolOptions getRunHandlerPoolOptions()
.org.platanios.tensorflow.proto.RunOptions.Experimental.RunHandlerPoolOptions run_handler_pool_options = 3;
- Returns:
- The runHandlerPoolOptions.
-
getRunHandlerPoolOptionsOrBuilder
RunOptions.Experimental.RunHandlerPoolOptionsOrBuilder getRunHandlerPoolOptionsOrBuilder()
.org.platanios.tensorflow.proto.RunOptions.Experimental.RunHandlerPoolOptions run_handler_pool_options = 3;
-
-