Package org.platanios.tensorflow.proto
Class ConfigProto.Experimental.Builder
- java.lang.Object
-
- com.google.protobuf.AbstractMessageLite.Builder
-
- com.google.protobuf.AbstractMessage.Builder<BuilderType>
-
- com.google.protobuf.GeneratedMessageV3.Builder<ConfigProto.Experimental.Builder>
-
- org.platanios.tensorflow.proto.ConfigProto.Experimental.Builder
-
- All Implemented Interfaces:
com.google.protobuf.Message.Builder
,com.google.protobuf.MessageLite.Builder
,com.google.protobuf.MessageLiteOrBuilder
,com.google.protobuf.MessageOrBuilder
,java.lang.Cloneable
,ConfigProto.ExperimentalOrBuilder
- Enclosing class:
- ConfigProto.Experimental
public static final class ConfigProto.Experimental.Builder extends com.google.protobuf.GeneratedMessageV3.Builder<ConfigProto.Experimental.Builder> implements ConfigProto.ExperimentalOrBuilder
Everything inside Experimental is subject to change and is not subject to API stability guarantees in https://www.tensorflow.org/guide/version_compat.
Protobuf typeorg.platanios.tensorflow.proto.ConfigProto.Experimental
-
-
Method Summary
All Methods Static Methods Instance Methods Concrete Methods Modifier and Type Method Description ConfigProto.Experimental.Builder
addRepeatedField(com.google.protobuf.Descriptors.FieldDescriptor field, java.lang.Object value)
ConfigProto.Experimental
build()
ConfigProto.Experimental
buildPartial()
ConfigProto.Experimental.Builder
clear()
ConfigProto.Experimental.Builder
clearCollectiveDeterministicSequentialExecution()
If true, make collective op execution order sequential and deterministic for potentially concurrent collective instances.ConfigProto.Experimental.Builder
clearCollectiveGroupLeader()
Task name for group resolution.ConfigProto.Experimental.Builder
clearCollectiveNccl()
If true, use NCCL for CollectiveOps.ConfigProto.Experimental.Builder
clearDisableOutputPartitionGraphs()
If true, the session will not store an additional copy of the graph for each subgraph.ConfigProto.Experimental.Builder
clearDisableThreadSpinning()
If using a direct session, disable spinning while waiting for work in the thread pool.ConfigProto.Experimental.Builder
clearEnableMlirBridge()
Whether to enable the MLIR-based TF->XLA bridge.ConfigProto.Experimental.Builder
clearEnableMlirGraphOptimization()
Whether to enable the MLIR-based Graph optimizations.ConfigProto.Experimental.Builder
clearExecutorType()
Which executor to use, the default executor will be used if it is an empty string or "DEFAULT"ConfigProto.Experimental.Builder
clearField(com.google.protobuf.Descriptors.FieldDescriptor field)
ConfigProto.Experimental.Builder
clearOneof(com.google.protobuf.Descriptors.OneofDescriptor oneof)
ConfigProto.Experimental.Builder
clearOptimizeForStaticGraph()
If true, the session may treat the graph as being static for optimization purposes.ConfigProto.Experimental.Builder
clearRecvBufMaxChunk()
Guidance to formatting of large RecvBuf fields for transfer.ConfigProto.Experimental.Builder
clearSessionMetadata()
Metadata about the session.ConfigProto.Experimental.Builder
clearShareClusterDevicesInSession()
This was promoted to a non-experimental API.ConfigProto.Experimental.Builder
clearShareSessionStateInClusterspecPropagation()
In the following, session state means the value of a variable, elements in a hash table, or any other resource, accessible by worker sessions held by a TF server.ConfigProto.Experimental.Builder
clearUseNumaAffinity()
If true, and supported by the platform, the runtime will attempt to use NUMA affinity where applicable.ConfigProto.Experimental.Builder
clearXlaFusionAutotunerThresh()
Minimum number of batches run through the XLA graph before XLA fusion autotuner is enabled.ConfigProto.Experimental.Builder
clone()
boolean
getCollectiveDeterministicSequentialExecution()
If true, make collective op execution order sequential and deterministic for potentially concurrent collective instances.java.lang.String
getCollectiveGroupLeader()
Task name for group resolution.com.google.protobuf.ByteString
getCollectiveGroupLeaderBytes()
Task name for group resolution.boolean
getCollectiveNccl()
If true, use NCCL for CollectiveOps.ConfigProto.Experimental
getDefaultInstanceForType()
static com.google.protobuf.Descriptors.Descriptor
getDescriptor()
com.google.protobuf.Descriptors.Descriptor
getDescriptorForType()
boolean
getDisableOutputPartitionGraphs()
If true, the session will not store an additional copy of the graph for each subgraph.boolean
getDisableThreadSpinning()
If using a direct session, disable spinning while waiting for work in the thread pool.boolean
getEnableMlirBridge()
Whether to enable the MLIR-based TF->XLA bridge.boolean
getEnableMlirGraphOptimization()
Whether to enable the MLIR-based Graph optimizations.java.lang.String
getExecutorType()
Which executor to use, the default executor will be used if it is an empty string or "DEFAULT"com.google.protobuf.ByteString
getExecutorTypeBytes()
Which executor to use, the default executor will be used if it is an empty string or "DEFAULT"boolean
getOptimizeForStaticGraph()
If true, the session may treat the graph as being static for optimization purposes.int
getRecvBufMaxChunk()
Guidance to formatting of large RecvBuf fields for transfer.SessionMetadata
getSessionMetadata()
Metadata about the session.SessionMetadata.Builder
getSessionMetadataBuilder()
Metadata about the session.SessionMetadataOrBuilder
getSessionMetadataOrBuilder()
Metadata about the session.boolean
getShareClusterDevicesInSession()
This was promoted to a non-experimental API.boolean
getShareSessionStateInClusterspecPropagation()
In the following, session state means the value of a variable, elements in a hash table, or any other resource, accessible by worker sessions held by a TF server.boolean
getUseNumaAffinity()
If true, and supported by the platform, the runtime will attempt to use NUMA affinity where applicable.long
getXlaFusionAutotunerThresh()
Minimum number of batches run through the XLA graph before XLA fusion autotuner is enabled.boolean
hasSessionMetadata()
Metadata about the session.protected com.google.protobuf.GeneratedMessageV3.FieldAccessorTable
internalGetFieldAccessorTable()
boolean
isInitialized()
ConfigProto.Experimental.Builder
mergeFrom(com.google.protobuf.CodedInputStream input, com.google.protobuf.ExtensionRegistryLite extensionRegistry)
ConfigProto.Experimental.Builder
mergeFrom(com.google.protobuf.Message other)
ConfigProto.Experimental.Builder
mergeFrom(ConfigProto.Experimental other)
ConfigProto.Experimental.Builder
mergeSessionMetadata(SessionMetadata value)
Metadata about the session.ConfigProto.Experimental.Builder
mergeUnknownFields(com.google.protobuf.UnknownFieldSet unknownFields)
ConfigProto.Experimental.Builder
setCollectiveDeterministicSequentialExecution(boolean value)
If true, make collective op execution order sequential and deterministic for potentially concurrent collective instances.ConfigProto.Experimental.Builder
setCollectiveGroupLeader(java.lang.String value)
Task name for group resolution.ConfigProto.Experimental.Builder
setCollectiveGroupLeaderBytes(com.google.protobuf.ByteString value)
Task name for group resolution.ConfigProto.Experimental.Builder
setCollectiveNccl(boolean value)
If true, use NCCL for CollectiveOps.ConfigProto.Experimental.Builder
setDisableOutputPartitionGraphs(boolean value)
If true, the session will not store an additional copy of the graph for each subgraph.ConfigProto.Experimental.Builder
setDisableThreadSpinning(boolean value)
If using a direct session, disable spinning while waiting for work in the thread pool.ConfigProto.Experimental.Builder
setEnableMlirBridge(boolean value)
Whether to enable the MLIR-based TF->XLA bridge.ConfigProto.Experimental.Builder
setEnableMlirGraphOptimization(boolean value)
Whether to enable the MLIR-based Graph optimizations.ConfigProto.Experimental.Builder
setExecutorType(java.lang.String value)
Which executor to use, the default executor will be used if it is an empty string or "DEFAULT"ConfigProto.Experimental.Builder
setExecutorTypeBytes(com.google.protobuf.ByteString value)
Which executor to use, the default executor will be used if it is an empty string or "DEFAULT"ConfigProto.Experimental.Builder
setField(com.google.protobuf.Descriptors.FieldDescriptor field, java.lang.Object value)
ConfigProto.Experimental.Builder
setOptimizeForStaticGraph(boolean value)
If true, the session may treat the graph as being static for optimization purposes.ConfigProto.Experimental.Builder
setRecvBufMaxChunk(int value)
Guidance to formatting of large RecvBuf fields for transfer.ConfigProto.Experimental.Builder
setRepeatedField(com.google.protobuf.Descriptors.FieldDescriptor field, int index, java.lang.Object value)
ConfigProto.Experimental.Builder
setSessionMetadata(SessionMetadata value)
Metadata about the session.ConfigProto.Experimental.Builder
setSessionMetadata(SessionMetadata.Builder builderForValue)
Metadata about the session.ConfigProto.Experimental.Builder
setShareClusterDevicesInSession(boolean value)
This was promoted to a non-experimental API.ConfigProto.Experimental.Builder
setShareSessionStateInClusterspecPropagation(boolean value)
In the following, session state means the value of a variable, elements in a hash table, or any other resource, accessible by worker sessions held by a TF server.ConfigProto.Experimental.Builder
setUnknownFields(com.google.protobuf.UnknownFieldSet unknownFields)
ConfigProto.Experimental.Builder
setUseNumaAffinity(boolean value)
If true, and supported by the platform, the runtime will attempt to use NUMA affinity where applicable.ConfigProto.Experimental.Builder
setXlaFusionAutotunerThresh(long value)
Minimum number of batches run through the XLA graph before XLA fusion autotuner is enabled.-
Methods inherited from class com.google.protobuf.GeneratedMessageV3.Builder
getAllFields, getField, getFieldBuilder, getOneofFieldDescriptor, getParentForChildren, getRepeatedField, getRepeatedFieldBuilder, getRepeatedFieldCount, getUnknownFields, hasField, hasOneof, internalGetMapField, internalGetMutableMapField, isClean, markClean, newBuilderForField, onBuilt, onChanged, setUnknownFieldsProto3
-
Methods inherited from class com.google.protobuf.AbstractMessage.Builder
findInitializationErrors, getInitializationErrorString, internalMergeFrom, mergeDelimitedFrom, mergeDelimitedFrom, mergeFrom, mergeFrom, mergeFrom, mergeFrom, mergeFrom, mergeFrom, mergeFrom, mergeFrom, mergeFrom, newUninitializedMessageException, toString
-
Methods inherited from class com.google.protobuf.AbstractMessageLite.Builder
addAll, addAll, mergeFrom, newUninitializedMessageException
-
-
-
-
Method Detail
-
getDescriptor
public static final com.google.protobuf.Descriptors.Descriptor getDescriptor()
-
internalGetFieldAccessorTable
protected com.google.protobuf.GeneratedMessageV3.FieldAccessorTable internalGetFieldAccessorTable()
- Specified by:
internalGetFieldAccessorTable
in classcom.google.protobuf.GeneratedMessageV3.Builder<ConfigProto.Experimental.Builder>
-
clear
public ConfigProto.Experimental.Builder clear()
- Specified by:
clear
in interfacecom.google.protobuf.Message.Builder
- Specified by:
clear
in interfacecom.google.protobuf.MessageLite.Builder
- Overrides:
clear
in classcom.google.protobuf.GeneratedMessageV3.Builder<ConfigProto.Experimental.Builder>
-
getDescriptorForType
public com.google.protobuf.Descriptors.Descriptor getDescriptorForType()
- Specified by:
getDescriptorForType
in interfacecom.google.protobuf.Message.Builder
- Specified by:
getDescriptorForType
in interfacecom.google.protobuf.MessageOrBuilder
- Overrides:
getDescriptorForType
in classcom.google.protobuf.GeneratedMessageV3.Builder<ConfigProto.Experimental.Builder>
-
getDefaultInstanceForType
public ConfigProto.Experimental getDefaultInstanceForType()
- Specified by:
getDefaultInstanceForType
in interfacecom.google.protobuf.MessageLiteOrBuilder
- Specified by:
getDefaultInstanceForType
in interfacecom.google.protobuf.MessageOrBuilder
-
build
public ConfigProto.Experimental build()
- Specified by:
build
in interfacecom.google.protobuf.Message.Builder
- Specified by:
build
in interfacecom.google.protobuf.MessageLite.Builder
-
buildPartial
public ConfigProto.Experimental buildPartial()
- Specified by:
buildPartial
in interfacecom.google.protobuf.Message.Builder
- Specified by:
buildPartial
in interfacecom.google.protobuf.MessageLite.Builder
-
clone
public ConfigProto.Experimental.Builder clone()
- Specified by:
clone
in interfacecom.google.protobuf.Message.Builder
- Specified by:
clone
in interfacecom.google.protobuf.MessageLite.Builder
- Overrides:
clone
in classcom.google.protobuf.GeneratedMessageV3.Builder<ConfigProto.Experimental.Builder>
-
setField
public ConfigProto.Experimental.Builder setField(com.google.protobuf.Descriptors.FieldDescriptor field, java.lang.Object value)
- Specified by:
setField
in interfacecom.google.protobuf.Message.Builder
- Overrides:
setField
in classcom.google.protobuf.GeneratedMessageV3.Builder<ConfigProto.Experimental.Builder>
-
clearField
public ConfigProto.Experimental.Builder clearField(com.google.protobuf.Descriptors.FieldDescriptor field)
- Specified by:
clearField
in interfacecom.google.protobuf.Message.Builder
- Overrides:
clearField
in classcom.google.protobuf.GeneratedMessageV3.Builder<ConfigProto.Experimental.Builder>
-
clearOneof
public ConfigProto.Experimental.Builder clearOneof(com.google.protobuf.Descriptors.OneofDescriptor oneof)
- Specified by:
clearOneof
in interfacecom.google.protobuf.Message.Builder
- Overrides:
clearOneof
in classcom.google.protobuf.GeneratedMessageV3.Builder<ConfigProto.Experimental.Builder>
-
setRepeatedField
public ConfigProto.Experimental.Builder setRepeatedField(com.google.protobuf.Descriptors.FieldDescriptor field, int index, java.lang.Object value)
- Specified by:
setRepeatedField
in interfacecom.google.protobuf.Message.Builder
- Overrides:
setRepeatedField
in classcom.google.protobuf.GeneratedMessageV3.Builder<ConfigProto.Experimental.Builder>
-
addRepeatedField
public ConfigProto.Experimental.Builder addRepeatedField(com.google.protobuf.Descriptors.FieldDescriptor field, java.lang.Object value)
- Specified by:
addRepeatedField
in interfacecom.google.protobuf.Message.Builder
- Overrides:
addRepeatedField
in classcom.google.protobuf.GeneratedMessageV3.Builder<ConfigProto.Experimental.Builder>
-
mergeFrom
public ConfigProto.Experimental.Builder mergeFrom(com.google.protobuf.Message other)
- Specified by:
mergeFrom
in interfacecom.google.protobuf.Message.Builder
- Overrides:
mergeFrom
in classcom.google.protobuf.AbstractMessage.Builder<ConfigProto.Experimental.Builder>
-
mergeFrom
public ConfigProto.Experimental.Builder mergeFrom(ConfigProto.Experimental other)
-
isInitialized
public final boolean isInitialized()
- Specified by:
isInitialized
in interfacecom.google.protobuf.MessageLiteOrBuilder
- Overrides:
isInitialized
in classcom.google.protobuf.GeneratedMessageV3.Builder<ConfigProto.Experimental.Builder>
-
mergeFrom
public ConfigProto.Experimental.Builder mergeFrom(com.google.protobuf.CodedInputStream input, com.google.protobuf.ExtensionRegistryLite extensionRegistry) throws java.io.IOException
- Specified by:
mergeFrom
in interfacecom.google.protobuf.Message.Builder
- Specified by:
mergeFrom
in interfacecom.google.protobuf.MessageLite.Builder
- Overrides:
mergeFrom
in classcom.google.protobuf.AbstractMessage.Builder<ConfigProto.Experimental.Builder>
- Throws:
java.io.IOException
-
getCollectiveGroupLeader
public java.lang.String getCollectiveGroupLeader()
Task name for group resolution.
string collective_group_leader = 1;
- Specified by:
getCollectiveGroupLeader
in interfaceConfigProto.ExperimentalOrBuilder
- Returns:
- The collectiveGroupLeader.
-
getCollectiveGroupLeaderBytes
public com.google.protobuf.ByteString getCollectiveGroupLeaderBytes()
Task name for group resolution.
string collective_group_leader = 1;
- Specified by:
getCollectiveGroupLeaderBytes
in interfaceConfigProto.ExperimentalOrBuilder
- Returns:
- The bytes for collectiveGroupLeader.
-
setCollectiveGroupLeader
public ConfigProto.Experimental.Builder setCollectiveGroupLeader(java.lang.String value)
Task name for group resolution.
string collective_group_leader = 1;
- Parameters:
value
- The collectiveGroupLeader to set.- Returns:
- This builder for chaining.
-
clearCollectiveGroupLeader
public ConfigProto.Experimental.Builder clearCollectiveGroupLeader()
Task name for group resolution.
string collective_group_leader = 1;
- Returns:
- This builder for chaining.
-
setCollectiveGroupLeaderBytes
public ConfigProto.Experimental.Builder setCollectiveGroupLeaderBytes(com.google.protobuf.ByteString value)
Task name for group resolution.
string collective_group_leader = 1;
- Parameters:
value
- The bytes for collectiveGroupLeader to set.- Returns:
- This builder for chaining.
-
getExecutorType
public java.lang.String getExecutorType()
Which executor to use, the default executor will be used if it is an empty string or "DEFAULT"
string executor_type = 3;
- Specified by:
getExecutorType
in interfaceConfigProto.ExperimentalOrBuilder
- Returns:
- The executorType.
-
getExecutorTypeBytes
public com.google.protobuf.ByteString getExecutorTypeBytes()
Which executor to use, the default executor will be used if it is an empty string or "DEFAULT"
string executor_type = 3;
- Specified by:
getExecutorTypeBytes
in interfaceConfigProto.ExperimentalOrBuilder
- Returns:
- The bytes for executorType.
-
setExecutorType
public ConfigProto.Experimental.Builder setExecutorType(java.lang.String value)
Which executor to use, the default executor will be used if it is an empty string or "DEFAULT"
string executor_type = 3;
- Parameters:
value
- The executorType to set.- Returns:
- This builder for chaining.
-
clearExecutorType
public ConfigProto.Experimental.Builder clearExecutorType()
Which executor to use, the default executor will be used if it is an empty string or "DEFAULT"
string executor_type = 3;
- Returns:
- This builder for chaining.
-
setExecutorTypeBytes
public ConfigProto.Experimental.Builder setExecutorTypeBytes(com.google.protobuf.ByteString value)
Which executor to use, the default executor will be used if it is an empty string or "DEFAULT"
string executor_type = 3;
- Parameters:
value
- The bytes for executorType to set.- Returns:
- This builder for chaining.
-
getRecvBufMaxChunk
public int getRecvBufMaxChunk()
Guidance to formatting of large RecvBuf fields for transfer. Any positive value sets the max chunk size. 0 defaults to 4096. Any negative value indicates no max, i.e. one chunk only.
int32 recv_buf_max_chunk = 4;
- Specified by:
getRecvBufMaxChunk
in interfaceConfigProto.ExperimentalOrBuilder
- Returns:
- The recvBufMaxChunk.
-
setRecvBufMaxChunk
public ConfigProto.Experimental.Builder setRecvBufMaxChunk(int value)
Guidance to formatting of large RecvBuf fields for transfer. Any positive value sets the max chunk size. 0 defaults to 4096. Any negative value indicates no max, i.e. one chunk only.
int32 recv_buf_max_chunk = 4;
- Parameters:
value
- The recvBufMaxChunk to set.- Returns:
- This builder for chaining.
-
clearRecvBufMaxChunk
public ConfigProto.Experimental.Builder clearRecvBufMaxChunk()
Guidance to formatting of large RecvBuf fields for transfer. Any positive value sets the max chunk size. 0 defaults to 4096. Any negative value indicates no max, i.e. one chunk only.
int32 recv_buf_max_chunk = 4;
- Returns:
- This builder for chaining.
-
getUseNumaAffinity
public boolean getUseNumaAffinity()
If true, and supported by the platform, the runtime will attempt to use NUMA affinity where applicable. One consequence will be the existence of as many CPU devices as there are available NUMA nodes.
bool use_numa_affinity = 5;
- Specified by:
getUseNumaAffinity
in interfaceConfigProto.ExperimentalOrBuilder
- Returns:
- The useNumaAffinity.
-
setUseNumaAffinity
public ConfigProto.Experimental.Builder setUseNumaAffinity(boolean value)
If true, and supported by the platform, the runtime will attempt to use NUMA affinity where applicable. One consequence will be the existence of as many CPU devices as there are available NUMA nodes.
bool use_numa_affinity = 5;
- Parameters:
value
- The useNumaAffinity to set.- Returns:
- This builder for chaining.
-
clearUseNumaAffinity
public ConfigProto.Experimental.Builder clearUseNumaAffinity()
If true, and supported by the platform, the runtime will attempt to use NUMA affinity where applicable. One consequence will be the existence of as many CPU devices as there are available NUMA nodes.
bool use_numa_affinity = 5;
- Returns:
- This builder for chaining.
-
getCollectiveDeterministicSequentialExecution
public boolean getCollectiveDeterministicSequentialExecution()
If true, make collective op execution order sequential and deterministic for potentially concurrent collective instances.
bool collective_deterministic_sequential_execution = 6;
- Specified by:
getCollectiveDeterministicSequentialExecution
in interfaceConfigProto.ExperimentalOrBuilder
- Returns:
- The collectiveDeterministicSequentialExecution.
-
setCollectiveDeterministicSequentialExecution
public ConfigProto.Experimental.Builder setCollectiveDeterministicSequentialExecution(boolean value)
If true, make collective op execution order sequential and deterministic for potentially concurrent collective instances.
bool collective_deterministic_sequential_execution = 6;
- Parameters:
value
- The collectiveDeterministicSequentialExecution to set.- Returns:
- This builder for chaining.
-
clearCollectiveDeterministicSequentialExecution
public ConfigProto.Experimental.Builder clearCollectiveDeterministicSequentialExecution()
If true, make collective op execution order sequential and deterministic for potentially concurrent collective instances.
bool collective_deterministic_sequential_execution = 6;
- Returns:
- This builder for chaining.
-
getCollectiveNccl
public boolean getCollectiveNccl()
If true, use NCCL for CollectiveOps. This feature is highly experimental.
bool collective_nccl = 7;
- Specified by:
getCollectiveNccl
in interfaceConfigProto.ExperimentalOrBuilder
- Returns:
- The collectiveNccl.
-
setCollectiveNccl
public ConfigProto.Experimental.Builder setCollectiveNccl(boolean value)
If true, use NCCL for CollectiveOps. This feature is highly experimental.
bool collective_nccl = 7;
- Parameters:
value
- The collectiveNccl to set.- Returns:
- This builder for chaining.
-
clearCollectiveNccl
public ConfigProto.Experimental.Builder clearCollectiveNccl()
If true, use NCCL for CollectiveOps. This feature is highly experimental.
bool collective_nccl = 7;
- Returns:
- This builder for chaining.
-
getShareSessionStateInClusterspecPropagation
public boolean getShareSessionStateInClusterspecPropagation()
In the following, session state means the value of a variable, elements in a hash table, or any other resource, accessible by worker sessions held by a TF server. When ClusterSpec propagation is enabled, the value of isolate_session_state is ignored when deciding whether to share session states in a TF server (for backwards compatibility reasons). - If share_session_state_in_clusterspec_propagation is true, the session states are shared. - If share_session_state_in_clusterspec_propagation is false, session states are isolated. When clusterspec propagation is not used, the value of share_session_state_in_clusterspec_propagation is ignored when deciding whether to share session states in a TF server. - If isolate_session_state is true, session states are isolated. - If isolate_session_state is false, session states are shared. TODO(b/129330037): Add a single API that consistently treats isolate_session_state and ClusterSpec propagation.
bool share_session_state_in_clusterspec_propagation = 8;
- Specified by:
getShareSessionStateInClusterspecPropagation
in interfaceConfigProto.ExperimentalOrBuilder
- Returns:
- The shareSessionStateInClusterspecPropagation.
-
setShareSessionStateInClusterspecPropagation
public ConfigProto.Experimental.Builder setShareSessionStateInClusterspecPropagation(boolean value)
In the following, session state means the value of a variable, elements in a hash table, or any other resource, accessible by worker sessions held by a TF server. When ClusterSpec propagation is enabled, the value of isolate_session_state is ignored when deciding whether to share session states in a TF server (for backwards compatibility reasons). - If share_session_state_in_clusterspec_propagation is true, the session states are shared. - If share_session_state_in_clusterspec_propagation is false, session states are isolated. When clusterspec propagation is not used, the value of share_session_state_in_clusterspec_propagation is ignored when deciding whether to share session states in a TF server. - If isolate_session_state is true, session states are isolated. - If isolate_session_state is false, session states are shared. TODO(b/129330037): Add a single API that consistently treats isolate_session_state and ClusterSpec propagation.
bool share_session_state_in_clusterspec_propagation = 8;
- Parameters:
value
- The shareSessionStateInClusterspecPropagation to set.- Returns:
- This builder for chaining.
-
clearShareSessionStateInClusterspecPropagation
public ConfigProto.Experimental.Builder clearShareSessionStateInClusterspecPropagation()
In the following, session state means the value of a variable, elements in a hash table, or any other resource, accessible by worker sessions held by a TF server. When ClusterSpec propagation is enabled, the value of isolate_session_state is ignored when deciding whether to share session states in a TF server (for backwards compatibility reasons). - If share_session_state_in_clusterspec_propagation is true, the session states are shared. - If share_session_state_in_clusterspec_propagation is false, session states are isolated. When clusterspec propagation is not used, the value of share_session_state_in_clusterspec_propagation is ignored when deciding whether to share session states in a TF server. - If isolate_session_state is true, session states are isolated. - If isolate_session_state is false, session states are shared. TODO(b/129330037): Add a single API that consistently treats isolate_session_state and ClusterSpec propagation.
bool share_session_state_in_clusterspec_propagation = 8;
- Returns:
- This builder for chaining.
-
getDisableThreadSpinning
public boolean getDisableThreadSpinning()
If using a direct session, disable spinning while waiting for work in the thread pool. This may result in higher latency for completing ops, but in the case where there is a lot of spinning may result in lower CPU usage.
bool disable_thread_spinning = 9;
- Specified by:
getDisableThreadSpinning
in interfaceConfigProto.ExperimentalOrBuilder
- Returns:
- The disableThreadSpinning.
-
setDisableThreadSpinning
public ConfigProto.Experimental.Builder setDisableThreadSpinning(boolean value)
If using a direct session, disable spinning while waiting for work in the thread pool. This may result in higher latency for completing ops, but in the case where there is a lot of spinning may result in lower CPU usage.
bool disable_thread_spinning = 9;
- Parameters:
value
- The disableThreadSpinning to set.- Returns:
- This builder for chaining.
-
clearDisableThreadSpinning
public ConfigProto.Experimental.Builder clearDisableThreadSpinning()
If using a direct session, disable spinning while waiting for work in the thread pool. This may result in higher latency for completing ops, but in the case where there is a lot of spinning may result in lower CPU usage.
bool disable_thread_spinning = 9;
- Returns:
- This builder for chaining.
-
getShareClusterDevicesInSession
public boolean getShareClusterDevicesInSession()
This was promoted to a non-experimental API. Please use ConfigProto.share_cluster_devices_in_session instead.
bool share_cluster_devices_in_session = 10;
- Specified by:
getShareClusterDevicesInSession
in interfaceConfigProto.ExperimentalOrBuilder
- Returns:
- The shareClusterDevicesInSession.
-
setShareClusterDevicesInSession
public ConfigProto.Experimental.Builder setShareClusterDevicesInSession(boolean value)
This was promoted to a non-experimental API. Please use ConfigProto.share_cluster_devices_in_session instead.
bool share_cluster_devices_in_session = 10;
- Parameters:
value
- The shareClusterDevicesInSession to set.- Returns:
- This builder for chaining.
-
clearShareClusterDevicesInSession
public ConfigProto.Experimental.Builder clearShareClusterDevicesInSession()
This was promoted to a non-experimental API. Please use ConfigProto.share_cluster_devices_in_session instead.
bool share_cluster_devices_in_session = 10;
- Returns:
- This builder for chaining.
-
hasSessionMetadata
public boolean hasSessionMetadata()
Metadata about the session. If set, this can be used by the runtime and the Ops for debugging, monitoring, etc. NOTE: This is currently used and propagated only by the direct session.
.org.platanios.tensorflow.proto.SessionMetadata session_metadata = 11;
- Specified by:
hasSessionMetadata
in interfaceConfigProto.ExperimentalOrBuilder
- Returns:
- Whether the sessionMetadata field is set.
-
getSessionMetadata
public SessionMetadata getSessionMetadata()
Metadata about the session. If set, this can be used by the runtime and the Ops for debugging, monitoring, etc. NOTE: This is currently used and propagated only by the direct session.
.org.platanios.tensorflow.proto.SessionMetadata session_metadata = 11;
- Specified by:
getSessionMetadata
in interfaceConfigProto.ExperimentalOrBuilder
- Returns:
- The sessionMetadata.
-
setSessionMetadata
public ConfigProto.Experimental.Builder setSessionMetadata(SessionMetadata value)
Metadata about the session. If set, this can be used by the runtime and the Ops for debugging, monitoring, etc. NOTE: This is currently used and propagated only by the direct session.
.org.platanios.tensorflow.proto.SessionMetadata session_metadata = 11;
-
setSessionMetadata
public ConfigProto.Experimental.Builder setSessionMetadata(SessionMetadata.Builder builderForValue)
Metadata about the session. If set, this can be used by the runtime and the Ops for debugging, monitoring, etc. NOTE: This is currently used and propagated only by the direct session.
.org.platanios.tensorflow.proto.SessionMetadata session_metadata = 11;
-
mergeSessionMetadata
public ConfigProto.Experimental.Builder mergeSessionMetadata(SessionMetadata value)
Metadata about the session. If set, this can be used by the runtime and the Ops for debugging, monitoring, etc. NOTE: This is currently used and propagated only by the direct session.
.org.platanios.tensorflow.proto.SessionMetadata session_metadata = 11;
-
clearSessionMetadata
public ConfigProto.Experimental.Builder clearSessionMetadata()
Metadata about the session. If set, this can be used by the runtime and the Ops for debugging, monitoring, etc. NOTE: This is currently used and propagated only by the direct session.
.org.platanios.tensorflow.proto.SessionMetadata session_metadata = 11;
-
getSessionMetadataBuilder
public SessionMetadata.Builder getSessionMetadataBuilder()
Metadata about the session. If set, this can be used by the runtime and the Ops for debugging, monitoring, etc. NOTE: This is currently used and propagated only by the direct session.
.org.platanios.tensorflow.proto.SessionMetadata session_metadata = 11;
-
getSessionMetadataOrBuilder
public SessionMetadataOrBuilder getSessionMetadataOrBuilder()
Metadata about the session. If set, this can be used by the runtime and the Ops for debugging, monitoring, etc. NOTE: This is currently used and propagated only by the direct session.
.org.platanios.tensorflow.proto.SessionMetadata session_metadata = 11;
- Specified by:
getSessionMetadataOrBuilder
in interfaceConfigProto.ExperimentalOrBuilder
-
getOptimizeForStaticGraph
public boolean getOptimizeForStaticGraph()
If true, the session may treat the graph as being static for optimization purposes. If this option is set to true when a session is created, the full GraphDef must be passed in a single call to Session::Create(), and Session::Extend() may not be supported.
bool optimize_for_static_graph = 12;
- Specified by:
getOptimizeForStaticGraph
in interfaceConfigProto.ExperimentalOrBuilder
- Returns:
- The optimizeForStaticGraph.
-
setOptimizeForStaticGraph
public ConfigProto.Experimental.Builder setOptimizeForStaticGraph(boolean value)
If true, the session may treat the graph as being static for optimization purposes. If this option is set to true when a session is created, the full GraphDef must be passed in a single call to Session::Create(), and Session::Extend() may not be supported.
bool optimize_for_static_graph = 12;
- Parameters:
value
- The optimizeForStaticGraph to set.- Returns:
- This builder for chaining.
-
clearOptimizeForStaticGraph
public ConfigProto.Experimental.Builder clearOptimizeForStaticGraph()
If true, the session may treat the graph as being static for optimization purposes. If this option is set to true when a session is created, the full GraphDef must be passed in a single call to Session::Create(), and Session::Extend() may not be supported.
bool optimize_for_static_graph = 12;
- Returns:
- This builder for chaining.
-
getEnableMlirBridge
public boolean getEnableMlirBridge()
Whether to enable the MLIR-based TF->XLA bridge. This is a replacement to the existing bridge, and not ready for production usage yet. If this option is set to true when a session is created, MLIR is used to perform the set of graph transformations to put the graph in a form that can be executed with delegation of some computations to an accelerator. This builds on the model of XLA where a subset of the graph is encapsulated and attached to a "compile" operation, whose result is fed to an "execute" operation. The kernel for these operations is responsible to lower the encapsulated graph to a particular device.
bool enable_mlir_bridge = 13;
- Specified by:
getEnableMlirBridge
in interfaceConfigProto.ExperimentalOrBuilder
- Returns:
- The enableMlirBridge.
-
setEnableMlirBridge
public ConfigProto.Experimental.Builder setEnableMlirBridge(boolean value)
Whether to enable the MLIR-based TF->XLA bridge. This is a replacement to the existing bridge, and not ready for production usage yet. If this option is set to true when a session is created, MLIR is used to perform the set of graph transformations to put the graph in a form that can be executed with delegation of some computations to an accelerator. This builds on the model of XLA where a subset of the graph is encapsulated and attached to a "compile" operation, whose result is fed to an "execute" operation. The kernel for these operations is responsible to lower the encapsulated graph to a particular device.
bool enable_mlir_bridge = 13;
- Parameters:
value
- The enableMlirBridge to set.- Returns:
- This builder for chaining.
-
clearEnableMlirBridge
public ConfigProto.Experimental.Builder clearEnableMlirBridge()
Whether to enable the MLIR-based TF->XLA bridge. This is a replacement to the existing bridge, and not ready for production usage yet. If this option is set to true when a session is created, MLIR is used to perform the set of graph transformations to put the graph in a form that can be executed with delegation of some computations to an accelerator. This builds on the model of XLA where a subset of the graph is encapsulated and attached to a "compile" operation, whose result is fed to an "execute" operation. The kernel for these operations is responsible to lower the encapsulated graph to a particular device.
bool enable_mlir_bridge = 13;
- Returns:
- This builder for chaining.
-
getEnableMlirGraphOptimization
public boolean getEnableMlirGraphOptimization()
Whether to enable the MLIR-based Graph optimizations. This will become a part of standard Tensorflow graph optimization pipeline, currently this is only used for gradual migration and testing new passes that are replacing existing optimizations in Grappler.
bool enable_mlir_graph_optimization = 16;
- Specified by:
getEnableMlirGraphOptimization
in interfaceConfigProto.ExperimentalOrBuilder
- Returns:
- The enableMlirGraphOptimization.
-
setEnableMlirGraphOptimization
public ConfigProto.Experimental.Builder setEnableMlirGraphOptimization(boolean value)
Whether to enable the MLIR-based Graph optimizations. This will become a part of standard Tensorflow graph optimization pipeline, currently this is only used for gradual migration and testing new passes that are replacing existing optimizations in Grappler.
bool enable_mlir_graph_optimization = 16;
- Parameters:
value
- The enableMlirGraphOptimization to set.- Returns:
- This builder for chaining.
-
clearEnableMlirGraphOptimization
public ConfigProto.Experimental.Builder clearEnableMlirGraphOptimization()
Whether to enable the MLIR-based Graph optimizations. This will become a part of standard Tensorflow graph optimization pipeline, currently this is only used for gradual migration and testing new passes that are replacing existing optimizations in Grappler.
bool enable_mlir_graph_optimization = 16;
- Returns:
- This builder for chaining.
-
getDisableOutputPartitionGraphs
public boolean getDisableOutputPartitionGraphs()
If true, the session will not store an additional copy of the graph for each subgraph. If this option is set to true when a session is created, the `RunOptions.output_partition_graphs` options must not be set.
bool disable_output_partition_graphs = 14;
- Specified by:
getDisableOutputPartitionGraphs
in interfaceConfigProto.ExperimentalOrBuilder
- Returns:
- The disableOutputPartitionGraphs.
-
setDisableOutputPartitionGraphs
public ConfigProto.Experimental.Builder setDisableOutputPartitionGraphs(boolean value)
If true, the session will not store an additional copy of the graph for each subgraph. If this option is set to true when a session is created, the `RunOptions.output_partition_graphs` options must not be set.
bool disable_output_partition_graphs = 14;
- Parameters:
value
- The disableOutputPartitionGraphs to set.- Returns:
- This builder for chaining.
-
clearDisableOutputPartitionGraphs
public ConfigProto.Experimental.Builder clearDisableOutputPartitionGraphs()
If true, the session will not store an additional copy of the graph for each subgraph. If this option is set to true when a session is created, the `RunOptions.output_partition_graphs` options must not be set.
bool disable_output_partition_graphs = 14;
- Returns:
- This builder for chaining.
-
getXlaFusionAutotunerThresh
public long getXlaFusionAutotunerThresh()
Minimum number of batches run through the XLA graph before XLA fusion autotuner is enabled. Default value of zero disables the autotuner. The XLA fusion autotuner can improve performance by executing a heuristic search on the compiler parameters.
int64 xla_fusion_autotuner_thresh = 15;
- Specified by:
getXlaFusionAutotunerThresh
in interfaceConfigProto.ExperimentalOrBuilder
- Returns:
- The xlaFusionAutotunerThresh.
-
setXlaFusionAutotunerThresh
public ConfigProto.Experimental.Builder setXlaFusionAutotunerThresh(long value)
Minimum number of batches run through the XLA graph before XLA fusion autotuner is enabled. Default value of zero disables the autotuner. The XLA fusion autotuner can improve performance by executing a heuristic search on the compiler parameters.
int64 xla_fusion_autotuner_thresh = 15;
- Parameters:
value
- The xlaFusionAutotunerThresh to set.- Returns:
- This builder for chaining.
-
clearXlaFusionAutotunerThresh
public ConfigProto.Experimental.Builder clearXlaFusionAutotunerThresh()
Minimum number of batches run through the XLA graph before XLA fusion autotuner is enabled. Default value of zero disables the autotuner. The XLA fusion autotuner can improve performance by executing a heuristic search on the compiler parameters.
int64 xla_fusion_autotuner_thresh = 15;
- Returns:
- This builder for chaining.
-
setUnknownFields
public final ConfigProto.Experimental.Builder setUnknownFields(com.google.protobuf.UnknownFieldSet unknownFields)
- Specified by:
setUnknownFields
in interfacecom.google.protobuf.Message.Builder
- Overrides:
setUnknownFields
in classcom.google.protobuf.GeneratedMessageV3.Builder<ConfigProto.Experimental.Builder>
-
mergeUnknownFields
public final ConfigProto.Experimental.Builder mergeUnknownFields(com.google.protobuf.UnknownFieldSet unknownFields)
- Specified by:
mergeUnknownFields
in interfacecom.google.protobuf.Message.Builder
- Overrides:
mergeUnknownFields
in classcom.google.protobuf.GeneratedMessageV3.Builder<ConfigProto.Experimental.Builder>
-
-