RunMetadata.Builder

public static final class RunMetadata.Builder 

 Metadata output (i.e., non-Tensor) for a single Run() call.
Protobuf type tensorflow.RunMetadata

Public Methods

RunMetadata.Builder
addAllFunctionGraphs(Iterable<? extends RunMetadata.FunctionGraphs> values) 
 This is only populated for graphs that are run as functions in TensorFlow
 V2.
RunMetadata.Builder
addAllPartitionGraphs(Iterable<? extends GraphDef> values) 
 Graphs of the partitions executed by executors.
RunMetadata.Builder
addFunctionGraphs(int index, RunMetadata.FunctionGraphs value) 
 This is only populated for graphs that are run as functions in TensorFlow
 V2.
RunMetadata.Builder
addFunctionGraphs(int index, RunMetadata.FunctionGraphs.Builder builderForValue) 
 This is only populated for graphs that are run as functions in TensorFlow
 V2.
RunMetadata.Builder
addFunctionGraphs(RunMetadata.FunctionGraphs value) 
 This is only populated for graphs that are run as functions in TensorFlow
 V2.
RunMetadata.Builder
addFunctionGraphs(RunMetadata.FunctionGraphs.Builder builderForValue) 
 This is only populated for graphs that are run as functions in TensorFlow
 V2.
RunMetadata.FunctionGraphs.Builder
addFunctionGraphsBuilder(int index) 
 This is only populated for graphs that are run as functions in TensorFlow
 V2.
RunMetadata.FunctionGraphs.Builder
addFunctionGraphsBuilder() 
 This is only populated for graphs that are run as functions in TensorFlow
 V2.
RunMetadata.Builder
addPartitionGraphs(int index, GraphDef value) 
 Graphs of the partitions executed by executors.
RunMetadata.Builder
addPartitionGraphs(GraphDef.Builder builderForValue) 
 Graphs of the partitions executed by executors.
RunMetadata.Builder
addPartitionGraphs(GraphDef value) 
 Graphs of the partitions executed by executors.
RunMetadata.Builder
addPartitionGraphs(int index, GraphDef.Builder builderForValue) 
 Graphs of the partitions executed by executors.
GraphDef.Builder
addPartitionGraphsBuilder(int index) 
 Graphs of the partitions executed by executors.
GraphDef.Builder
addPartitionGraphsBuilder() 
 Graphs of the partitions executed by executors.
RunMetadata.Builder
addRepeatedField(com.google.protobuf.Descriptors.FieldDescriptor field, Object value)
RunMetadata
build()
RunMetadata
buildPartial()
RunMetadata.Builder
clear()
RunMetadata.Builder
clearCostGraph() 
 The cost graph for the computation defined by the run call.
RunMetadata.Builder
clearField(com.google.protobuf.Descriptors.FieldDescriptor field)
RunMetadata.Builder
clearFunctionGraphs() 
 This is only populated for graphs that are run as functions in TensorFlow
 V2.
RunMetadata.Builder
clearOneof(com.google.protobuf.Descriptors.OneofDescriptor oneof)
RunMetadata.Builder
clearPartitionGraphs() 
 Graphs of the partitions executed by executors.
RunMetadata.Builder
clearStepStats() 
 Statistics traced for this step.
RunMetadata.Builder
clone()
CostGraphDef
getCostGraph() 
 The cost graph for the computation defined by the run call.
CostGraphDef.Builder
getCostGraphBuilder() 
 The cost graph for the computation defined by the run call.
CostGraphDefOrBuilder
getCostGraphOrBuilder() 
 The cost graph for the computation defined by the run call.
RunMetadata
getDefaultInstanceForType()
final static com.google.protobuf.Descriptors.Descriptor
getDescriptor()
com.google.protobuf.Descriptors.Descriptor
getDescriptorForType()
RunMetadata.FunctionGraphs
getFunctionGraphs(int index) 
 This is only populated for graphs that are run as functions in TensorFlow
 V2.
RunMetadata.FunctionGraphs.Builder
getFunctionGraphsBuilder(int index) 
 This is only populated for graphs that are run as functions in TensorFlow
 V2.
List<RunMetadata.FunctionGraphs.Builder>
getFunctionGraphsBuilderList() 
 This is only populated for graphs that are run as functions in TensorFlow
 V2.
int
getFunctionGraphsCount() 
 This is only populated for graphs that are run as functions in TensorFlow
 V2.
List<RunMetadata.FunctionGraphs>
getFunctionGraphsList() 
 This is only populated for graphs that are run as functions in TensorFlow
 V2.
RunMetadata.FunctionGraphsOrBuilder
getFunctionGraphsOrBuilder(int index) 
 This is only populated for graphs that are run as functions in TensorFlow
 V2.
List<? extends RunMetadata.FunctionGraphsOrBuilder>
getFunctionGraphsOrBuilderList() 
 This is only populated for graphs that are run as functions in TensorFlow
 V2.
GraphDef
getPartitionGraphs(int index) 
 Graphs of the partitions executed by executors.
GraphDef.Builder
getPartitionGraphsBuilder(int index) 
 Graphs of the partitions executed by executors.
List<GraphDef.Builder>
getPartitionGraphsBuilderList() 
 Graphs of the partitions executed by executors.
int
getPartitionGraphsCount() 
 Graphs of the partitions executed by executors.
List<GraphDef>
getPartitionGraphsList() 
 Graphs of the partitions executed by executors.
GraphDefOrBuilder
getPartitionGraphsOrBuilder(int index) 
 Graphs of the partitions executed by executors.
List<? extends GraphDefOrBuilder>
getPartitionGraphsOrBuilderList() 
 Graphs of the partitions executed by executors.
StepStats
getStepStats() 
 Statistics traced for this step.
StepStats.Builder
getStepStatsBuilder() 
 Statistics traced for this step.
StepStatsOrBuilder
getStepStatsOrBuilder() 
 Statistics traced for this step.
boolean
hasCostGraph() 
 The cost graph for the computation defined by the run call.
boolean
hasStepStats() 
 Statistics traced for this step.
final boolean
isInitialized()
RunMetadata.Builder
mergeCostGraph(CostGraphDef value) 
 The cost graph for the computation defined by the run call.
RunMetadata.Builder
mergeFrom(com.google.protobuf.Message other)
RunMetadata.Builder
mergeFrom(com.google.protobuf.CodedInputStream input, com.google.protobuf.ExtensionRegistryLite extensionRegistry)
RunMetadata.Builder
mergeStepStats(StepStats value) 
 Statistics traced for this step.
final RunMetadata.Builder
mergeUnknownFields(com.google.protobuf.UnknownFieldSet unknownFields)
RunMetadata.Builder
removeFunctionGraphs(int index) 
 This is only populated for graphs that are run as functions in TensorFlow
 V2.
RunMetadata.Builder
removePartitionGraphs(int index) 
 Graphs of the partitions executed by executors.
RunMetadata.Builder
setCostGraph(CostGraphDef value) 
 The cost graph for the computation defined by the run call.
RunMetadata.Builder
setCostGraph(CostGraphDef.Builder builderForValue) 
 The cost graph for the computation defined by the run call.
RunMetadata.Builder
setField(com.google.protobuf.Descriptors.FieldDescriptor field, Object value)
RunMetadata.Builder
setFunctionGraphs(int index, RunMetadata.FunctionGraphs value) 
 This is only populated for graphs that are run as functions in TensorFlow
 V2.
RunMetadata.Builder
setFunctionGraphs(int index, RunMetadata.FunctionGraphs.Builder builderForValue) 
 This is only populated for graphs that are run as functions in TensorFlow
 V2.
RunMetadata.Builder
setPartitionGraphs(int index, GraphDef.Builder builderForValue) 
 Graphs of the partitions executed by executors.
RunMetadata.Builder
setPartitionGraphs(int index, GraphDef value) 
 Graphs of the partitions executed by executors.
RunMetadata.Builder
setRepeatedField(com.google.protobuf.Descriptors.FieldDescriptor field, int index, Object value)
RunMetadata.Builder
setStepStats(StepStats.Builder builderForValue) 
 Statistics traced for this step.
RunMetadata.Builder
setStepStats(StepStats value) 
 Statistics traced for this step.
final RunMetadata.Builder
setUnknownFields(com.google.protobuf.UnknownFieldSet unknownFields)

Inherited Methods

Public Methods

public RunMetadata.Builder addAllFunctionGraphs (Iterable<? extends RunMetadata.FunctionGraphs> values) 

 This is only populated for graphs that are run as functions in TensorFlow
 V2. There will be an entry below for each function that is traced.
 The main use cases of the post_optimization_graph and the partition_graphs
 is to give the caller insight into the graphs that were actually run by the
 runtime. Additional information (such as those in step_stats) will match
 these graphs.
 We also include the pre_optimization_graph since it is usually easier to
 read, and is helpful in situations where the caller wants to get a high
 level idea of what the built graph looks like (since the various graph
 optimization passes might change the structure of the graph significantly).
repeated .tensorflow.RunMetadata.FunctionGraphs function_graphs = 4;

public RunMetadata.Builder addAllPartitionGraphs (Iterable<? extends GraphDef> values) 

 Graphs of the partitions executed by executors.
repeated .tensorflow.GraphDef partition_graphs = 3;

public RunMetadata.Builder addFunctionGraphs (int index, RunMetadata.FunctionGraphs value) 

 This is only populated for graphs that are run as functions in TensorFlow
 V2. There will be an entry below for each function that is traced.
 The main use cases of the post_optimization_graph and the partition_graphs
 is to give the caller insight into the graphs that were actually run by the
 runtime. Additional information (such as those in step_stats) will match
 these graphs.
 We also include the pre_optimization_graph since it is usually easier to
 read, and is helpful in situations where the caller wants to get a high
 level idea of what the built graph looks like (since the various graph
 optimization passes might change the structure of the graph significantly).
repeated .tensorflow.RunMetadata.FunctionGraphs function_graphs = 4;

public RunMetadata.Builder addFunctionGraphs (int index, RunMetadata.FunctionGraphs.Builder builderForValue) 

 This is only populated for graphs that are run as functions in TensorFlow
 V2. There will be an entry below for each function that is traced.
 The main use cases of the post_optimization_graph and the partition_graphs
 is to give the caller insight into the graphs that were actually run by the
 runtime. Additional information (such as those in step_stats) will match
 these graphs.
 We also include the pre_optimization_graph since it is usually easier to
 read, and is helpful in situations where the caller wants to get a high
 level idea of what the built graph looks like (since the various graph
 optimization passes might change the structure of the graph significantly).
repeated .tensorflow.RunMetadata.FunctionGraphs function_graphs = 4;

public RunMetadata.Builder addFunctionGraphs (RunMetadata.FunctionGraphs value) 

 This is only populated for graphs that are run as functions in TensorFlow
 V2. There will be an entry below for each function that is traced.
 The main use cases of the post_optimization_graph and the partition_graphs
 is to give the caller insight into the graphs that were actually run by the
 runtime. Additional information (such as those in step_stats) will match
 these graphs.
 We also include the pre_optimization_graph since it is usually easier to
 read, and is helpful in situations where the caller wants to get a high
 level idea of what the built graph looks like (since the various graph
 optimization passes might change the structure of the graph significantly).
repeated .tensorflow.RunMetadata.FunctionGraphs function_graphs = 4;

public RunMetadata.Builder addFunctionGraphs (RunMetadata.FunctionGraphs.Builder builderForValue) 

 This is only populated for graphs that are run as functions in TensorFlow
 V2. There will be an entry below for each function that is traced.
 The main use cases of the post_optimization_graph and the partition_graphs
 is to give the caller insight into the graphs that were actually run by the
 runtime. Additional information (such as those in step_stats) will match
 these graphs.
 We also include the pre_optimization_graph since it is usually easier to
 read, and is helpful in situations where the caller wants to get a high
 level idea of what the built graph looks like (since the various graph
 optimization passes might change the structure of the graph significantly).
repeated .tensorflow.RunMetadata.FunctionGraphs function_graphs = 4;

public RunMetadata.FunctionGraphs.Builder addFunctionGraphsBuilder (int index) 

 This is only populated for graphs that are run as functions in TensorFlow
 V2. There will be an entry below for each function that is traced.
 The main use cases of the post_optimization_graph and the partition_graphs
 is to give the caller insight into the graphs that were actually run by the
 runtime. Additional information (such as those in step_stats) will match
 these graphs.
 We also include the pre_optimization_graph since it is usually easier to
 read, and is helpful in situations where the caller wants to get a high
 level idea of what the built graph looks like (since the various graph
 optimization passes might change the structure of the graph significantly).
repeated .tensorflow.RunMetadata.FunctionGraphs function_graphs = 4;

public RunMetadata.FunctionGraphs.Builder addFunctionGraphsBuilder () 

 This is only populated for graphs that are run as functions in TensorFlow
 V2. There will be an entry below for each function that is traced.
 The main use cases of the post_optimization_graph and the partition_graphs
 is to give the caller insight into the graphs that were actually run by the
 runtime. Additional information (such as those in step_stats) will match
 these graphs.
 We also include the pre_optimization_graph since it is usually easier to
 read, and is helpful in situations where the caller wants to get a high
 level idea of what the built graph looks like (since the various graph
 optimization passes might change the structure of the graph significantly).
repeated .tensorflow.RunMetadata.FunctionGraphs function_graphs = 4;

public RunMetadata.Builder addPartitionGraphs (int index, GraphDef value) 

 Graphs of the partitions executed by executors.
repeated .tensorflow.GraphDef partition_graphs = 3;

public RunMetadata.Builder addPartitionGraphs (GraphDef.Builder builderForValue) 

 Graphs of the partitions executed by executors.
repeated .tensorflow.GraphDef partition_graphs = 3;

public RunMetadata.Builder addPartitionGraphs (GraphDef value) 

 Graphs of the partitions executed by executors.
repeated .tensorflow.GraphDef partition_graphs = 3;

public RunMetadata.Builder addPartitionGraphs (int index, GraphDef.Builder builderForValue) 

 Graphs of the partitions executed by executors.
repeated .tensorflow.GraphDef partition_graphs = 3;

public GraphDef.Builder addPartitionGraphsBuilder (int index) 

 Graphs of the partitions executed by executors.
repeated .tensorflow.GraphDef partition_graphs = 3;

public GraphDef.Builder addPartitionGraphsBuilder () 

 Graphs of the partitions executed by executors.
repeated .tensorflow.GraphDef partition_graphs = 3;

public RunMetadata.Builder addRepeatedField (com.google.protobuf.Descriptors.FieldDescriptor field, Object value)

public RunMetadata build ()

public RunMetadata buildPartial ()

public RunMetadata.Builder clear ()

public RunMetadata.Builder clearCostGraph () 

 The cost graph for the computation defined by the run call.
.tensorflow.CostGraphDef cost_graph = 2;

public RunMetadata.Builder clearField (com.google.protobuf.Descriptors.FieldDescriptor field)

public RunMetadata.Builder clearFunctionGraphs () 

 This is only populated for graphs that are run as functions in TensorFlow
 V2. There will be an entry below for each function that is traced.
 The main use cases of the post_optimization_graph and the partition_graphs
 is to give the caller insight into the graphs that were actually run by the
 runtime. Additional information (such as those in step_stats) will match
 these graphs.
 We also include the pre_optimization_graph since it is usually easier to
 read, and is helpful in situations where the caller wants to get a high
 level idea of what the built graph looks like (since the various graph
 optimization passes might change the structure of the graph significantly).
repeated .tensorflow.RunMetadata.FunctionGraphs function_graphs = 4;

public RunMetadata.Builder clearOneof (com.google.protobuf.Descriptors.OneofDescriptor oneof)

public RunMetadata.Builder clearPartitionGraphs () 

 Graphs of the partitions executed by executors.
repeated .tensorflow.GraphDef partition_graphs = 3;

public RunMetadata.Builder clearStepStats () 

 Statistics traced for this step. Populated if tracing is turned on via the
 "RunOptions" proto.
 EXPERIMENTAL: The format and set of events may change in future versions.
.tensorflow.StepStats step_stats = 1;

public RunMetadata.Builder clone ()

public CostGraphDef getCostGraph () 

 The cost graph for the computation defined by the run call.
.tensorflow.CostGraphDef cost_graph = 2;

public CostGraphDef.Builder getCostGraphBuilder () 

 The cost graph for the computation defined by the run call.
.tensorflow.CostGraphDef cost_graph = 2;

public CostGraphDefOrBuilder getCostGraphOrBuilder () 

 The cost graph for the computation defined by the run call.
.tensorflow.CostGraphDef cost_graph = 2;

public RunMetadata getDefaultInstanceForType ()

public static final com.google.protobuf.Descriptors.Descriptor getDescriptor ()

public com.google.protobuf.Descriptors.Descriptor getDescriptorForType ()

public RunMetadata.FunctionGraphs getFunctionGraphs (int index) 

 This is only populated for graphs that are run as functions in TensorFlow
 V2. There will be an entry below for each function that is traced.
 The main use cases of the post_optimization_graph and the partition_graphs
 is to give the caller insight into the graphs that were actually run by the
 runtime. Additional information (such as those in step_stats) will match
 these graphs.
 We also include the pre_optimization_graph since it is usually easier to
 read, and is helpful in situations where the caller wants to get a high
 level idea of what the built graph looks like (since the various graph
 optimization passes might change the structure of the graph significantly).
repeated .tensorflow.RunMetadata.FunctionGraphs function_graphs = 4;

public RunMetadata.FunctionGraphs.Builder getFunctionGraphsBuilder (int index) 

 This is only populated for graphs that are run as functions in TensorFlow
 V2. There will be an entry below for each function that is traced.
 The main use cases of the post_optimization_graph and the partition_graphs
 is to give the caller insight into the graphs that were actually run by the
 runtime. Additional information (such as those in step_stats) will match
 these graphs.
 We also include the pre_optimization_graph since it is usually easier to
 read, and is helpful in situations where the caller wants to get a high
 level idea of what the built graph looks like (since the various graph
 optimization passes might change the structure of the graph significantly).
repeated .tensorflow.RunMetadata.FunctionGraphs function_graphs = 4;

public List<RunMetadata.FunctionGraphs.Builder> getFunctionGraphsBuilderList () 

 This is only populated for graphs that are run as functions in TensorFlow
 V2. There will be an entry below for each function that is traced.
 The main use cases of the post_optimization_graph and the partition_graphs
 is to give the caller insight into the graphs that were actually run by the
 runtime. Additional information (such as those in step_stats) will match
 these graphs.
 We also include the pre_optimization_graph since it is usually easier to
 read, and is helpful in situations where the caller wants to get a high
 level idea of what the built graph looks like (since the various graph
 optimization passes might change the structure of the graph significantly).
repeated .tensorflow.RunMetadata.FunctionGraphs function_graphs = 4;

public int getFunctionGraphsCount () 

 This is only populated for graphs that are run as functions in TensorFlow
 V2. There will be an entry below for each function that is traced.
 The main use cases of the post_optimization_graph and the partition_graphs
 is to give the caller insight into the graphs that were actually run by the
 runtime. Additional information (such as those in step_stats) will match
 these graphs.
 We also include the pre_optimization_graph since it is usually easier to
 read, and is helpful in situations where the caller wants to get a high
 level idea of what the built graph looks like (since the various graph
 optimization passes might change the structure of the graph significantly).
repeated .tensorflow.RunMetadata.FunctionGraphs function_graphs = 4;

public List<RunMetadata.FunctionGraphs> getFunctionGraphsList () 

 This is only populated for graphs that are run as functions in TensorFlow
 V2. There will be an entry below for each function that is traced.
 The main use cases of the post_optimization_graph and the partition_graphs
 is to give the caller insight into the graphs that were actually run by the
 runtime. Additional information (such as those in step_stats) will match
 these graphs.
 We also include the pre_optimization_graph since it is usually easier to
 read, and is helpful in situations where the caller wants to get a high
 level idea of what the built graph looks like (since the various graph
 optimization passes might change the structure of the graph significantly).
repeated .tensorflow.RunMetadata.FunctionGraphs function_graphs = 4;

public RunMetadata.FunctionGraphsOrBuilder getFunctionGraphsOrBuilder (int index) 

 This is only populated for graphs that are run as functions in TensorFlow
 V2. There will be an entry below for each function that is traced.
 The main use cases of the post_optimization_graph and the partition_graphs
 is to give the caller insight into the graphs that were actually run by the
 runtime. Additional information (such as those in step_stats) will match
 these graphs.
 We also include the pre_optimization_graph since it is usually easier to
 read, and is helpful in situations where the caller wants to get a high
 level idea of what the built graph looks like (since the various graph
 optimization passes might change the structure of the graph significantly).
repeated .tensorflow.RunMetadata.FunctionGraphs function_graphs = 4;

public List<? extends RunMetadata.FunctionGraphsOrBuilder> getFunctionGraphsOrBuilderList () 

 This is only populated for graphs that are run as functions in TensorFlow
 V2. There will be an entry below for each function that is traced.
 The main use cases of the post_optimization_graph and the partition_graphs
 is to give the caller insight into the graphs that were actually run by the
 runtime. Additional information (such as those in step_stats) will match
 these graphs.
 We also include the pre_optimization_graph since it is usually easier to
 read, and is helpful in situations where the caller wants to get a high
 level idea of what the built graph looks like (since the various graph
 optimization passes might change the structure of the graph significantly).
repeated .tensorflow.RunMetadata.FunctionGraphs function_graphs = 4;

public GraphDef getPartitionGraphs (int index) 

 Graphs of the partitions executed by executors.
repeated .tensorflow.GraphDef partition_graphs = 3;

public GraphDef.Builder getPartitionGraphsBuilder (int index) 

 Graphs of the partitions executed by executors.
repeated .tensorflow.GraphDef partition_graphs = 3;

public List<GraphDef.Builder> getPartitionGraphsBuilderList () 

 Graphs of the partitions executed by executors.
repeated .tensorflow.GraphDef partition_graphs = 3;

public int getPartitionGraphsCount () 

 Graphs of the partitions executed by executors.
repeated .tensorflow.GraphDef partition_graphs = 3;

public List<GraphDef> getPartitionGraphsList () 

 Graphs of the partitions executed by executors.
repeated .tensorflow.GraphDef partition_graphs = 3;

public GraphDefOrBuilder getPartitionGraphsOrBuilder (int index) 

 Graphs of the partitions executed by executors.
repeated .tensorflow.GraphDef partition_graphs = 3;

public List<? extends GraphDefOrBuilder> getPartitionGraphsOrBuilderList () 

 Graphs of the partitions executed by executors.
repeated .tensorflow.GraphDef partition_graphs = 3;

public StepStats getStepStats () 

 Statistics traced for this step. Populated if tracing is turned on via the
 "RunOptions" proto.
 EXPERIMENTAL: The format and set of events may change in future versions.
.tensorflow.StepStats step_stats = 1;

public StepStats.Builder getStepStatsBuilder () 

 Statistics traced for this step. Populated if tracing is turned on via the
 "RunOptions" proto.
 EXPERIMENTAL: The format and set of events may change in future versions.
.tensorflow.StepStats step_stats = 1;

public StepStatsOrBuilder getStepStatsOrBuilder () 

 Statistics traced for this step. Populated if tracing is turned on via the
 "RunOptions" proto.
 EXPERIMENTAL: The format and set of events may change in future versions.
.tensorflow.StepStats step_stats = 1;

public boolean hasCostGraph () 

 The cost graph for the computation defined by the run call.
.tensorflow.CostGraphDef cost_graph = 2;

public boolean hasStepStats () 

 Statistics traced for this step. Populated if tracing is turned on via the
 "RunOptions" proto.
 EXPERIMENTAL: The format and set of events may change in future versions.
.tensorflow.StepStats step_stats = 1;

public final boolean isInitialized ()

public RunMetadata.Builder mergeCostGraph (CostGraphDef value) 

 The cost graph for the computation defined by the run call.
.tensorflow.CostGraphDef cost_graph = 2;

public RunMetadata.Builder mergeFrom (com.google.protobuf.Message other)

public RunMetadata.Builder mergeFrom (com.google.protobuf.CodedInputStream input, com.google.protobuf.ExtensionRegistryLite extensionRegistry)

Throws
IOException

public RunMetadata.Builder mergeStepStats (StepStats value) 

 Statistics traced for this step. Populated if tracing is turned on via the
 "RunOptions" proto.
 EXPERIMENTAL: The format and set of events may change in future versions.
.tensorflow.StepStats step_stats = 1;

public final RunMetadata.Builder mergeUnknownFields (com.google.protobuf.UnknownFieldSet unknownFields)

public RunMetadata.Builder removeFunctionGraphs (int index) 

 This is only populated for graphs that are run as functions in TensorFlow
 V2. There will be an entry below for each function that is traced.
 The main use cases of the post_optimization_graph and the partition_graphs
 is to give the caller insight into the graphs that were actually run by the
 runtime. Additional information (such as those in step_stats) will match
 these graphs.
 We also include the pre_optimization_graph since it is usually easier to
 read, and is helpful in situations where the caller wants to get a high
 level idea of what the built graph looks like (since the various graph
 optimization passes might change the structure of the graph significantly).
repeated .tensorflow.RunMetadata.FunctionGraphs function_graphs = 4;

public RunMetadata.Builder removePartitionGraphs (int index) 

 Graphs of the partitions executed by executors.
repeated .tensorflow.GraphDef partition_graphs = 3;

public RunMetadata.Builder setCostGraph (CostGraphDef value) 

 The cost graph for the computation defined by the run call.
.tensorflow.CostGraphDef cost_graph = 2;

public RunMetadata.Builder setCostGraph (CostGraphDef.Builder builderForValue) 

 The cost graph for the computation defined by the run call.
.tensorflow.CostGraphDef cost_graph = 2;

public RunMetadata.Builder setField (com.google.protobuf.Descriptors.FieldDescriptor field, Object value)

public RunMetadata.Builder setFunctionGraphs (int index, RunMetadata.FunctionGraphs value) 

 This is only populated for graphs that are run as functions in TensorFlow
 V2. There will be an entry below for each function that is traced.
 The main use cases of the post_optimization_graph and the partition_graphs
 is to give the caller insight into the graphs that were actually run by the
 runtime. Additional information (such as those in step_stats) will match
 these graphs.
 We also include the pre_optimization_graph since it is usually easier to
 read, and is helpful in situations where the caller wants to get a high
 level idea of what the built graph looks like (since the various graph
 optimization passes might change the structure of the graph significantly).
repeated .tensorflow.RunMetadata.FunctionGraphs function_graphs = 4;

public RunMetadata.Builder setFunctionGraphs (int index, RunMetadata.FunctionGraphs.Builder builderForValue) 

 This is only populated for graphs that are run as functions in TensorFlow
 V2. There will be an entry below for each function that is traced.
 The main use cases of the post_optimization_graph and the partition_graphs
 is to give the caller insight into the graphs that were actually run by the
 runtime. Additional information (such as those in step_stats) will match
 these graphs.
 We also include the pre_optimization_graph since it is usually easier to
 read, and is helpful in situations where the caller wants to get a high
 level idea of what the built graph looks like (since the various graph
 optimization passes might change the structure of the graph significantly).
repeated .tensorflow.RunMetadata.FunctionGraphs function_graphs = 4;

public RunMetadata.Builder setPartitionGraphs (int index, GraphDef.Builder builderForValue) 

 Graphs of the partitions executed by executors.
repeated .tensorflow.GraphDef partition_graphs = 3;

public RunMetadata.Builder setPartitionGraphs (int index, GraphDef value) 

 Graphs of the partitions executed by executors.
repeated .tensorflow.GraphDef partition_graphs = 3;

public RunMetadata.Builder setRepeatedField (com.google.protobuf.Descriptors.FieldDescriptor field, int index, Object value)

public RunMetadata.Builder setStepStats (StepStats.Builder builderForValue) 

 Statistics traced for this step. Populated if tracing is turned on via the
 "RunOptions" proto.
 EXPERIMENTAL: The format and set of events may change in future versions.
.tensorflow.StepStats step_stats = 1;

public RunMetadata.Builder setStepStats (StepStats value) 

 Statistics traced for this step. Populated if tracing is turned on via the
 "RunOptions" proto.
 EXPERIMENTAL: The format and set of events may change in future versions.
.tensorflow.StepStats step_stats = 1;

public final RunMetadata.Builder setUnknownFields (com.google.protobuf.UnknownFieldSet unknownFields)