الفئة النهائية العامة الثابتة ConfigProto.Builder
Session configuration parameters. The system picks appropriate values for fields that are not set.
tensorflow.ConfigProto الأساليب العامة
| ConfigProto.Builder | addAllDeviceFilters (قيم Iterable<String>) When any filters are present sessions will ignore all devices which do not match the filters. |
| ConfigProto.Builder | addAllSessionInterOpThreadPool (Iterable<؟ يمتد ThreadPoolOptionProto > القيم) This option is experimental - it may be replaced with a different mechanism in the future. |
| ConfigProto.Builder | addDeviceFilters (قيمة السلسلة) When any filters are present sessions will ignore all devices which do not match the filters. |
| ConfigProto.Builder | addDeviceFiltersBytes (قيمة com.google.protobuf.ByteString) When any filters are present sessions will ignore all devices which do not match the filters. |
| ConfigProto.Builder | addRepeatedField (حقل com.google.protobuf.Descriptors.FieldDescriptor، قيمة الكائن) |
| ConfigProto.Builder | addSessionInterOpThreadPool (قيمة ThreadPoolOptionProto ) This option is experimental - it may be replaced with a different mechanism in the future. |
| ConfigProto.Builder | addSessionInterOpThreadPool (مؤشر int، قيمة ThreadPoolOptionProto ) This option is experimental - it may be replaced with a different mechanism in the future. |
| ConfigProto.Builder | addSessionInterOpThreadPool ( ThreadPoolOptionProto.Builder builderForValue) This option is experimental - it may be replaced with a different mechanism in the future. |
| ConfigProto.Builder | addSessionInterOpThreadPool (مؤشر int، ThreadPoolOptionProto.Builder builderForValue) This option is experimental - it may be replaced with a different mechanism in the future. |
| ThreadPoolOptionProto.Builder | addSessionInterOpThreadPoolBuilder (فهرس كثافة العمليات) This option is experimental - it may be replaced with a different mechanism in the future. |
| ThreadPoolOptionProto.Builder | addSessionInterOpThreadPoolBuilder () This option is experimental - it may be replaced with a different mechanism in the future. |
| ConfigProto | يبني () |
| ConfigProto | بناء جزئي () |
| ConfigProto.Builder | واضح () |
| ConfigProto.Builder | ClearAllowSoftPlacement () Whether soft placement is allowed. |
| ConfigProto.Builder | كلير كلستر ديف () Optional list of all workers to use in this session. |
| ConfigProto.Builder | |
| ConfigProto.Builder | مرشحات كلير ديفيس () When any filters are present sessions will ignore all devices which do not match the filters. |
| ConfigProto.Builder | واضحتجريبي () .tensorflow.ConfigProto.Experimental experimental = 16; |
| ConfigProto.Builder | ClearField (حقل com.google.protobuf.Descriptors.FieldDescriptor) |
| ConfigProto.Builder | خيارات ClearGpu () Options that apply to all GPUs. |
| ConfigProto.Builder | خيارات ClearGraph () Options that apply to all graphs. |
| ConfigProto.Builder | ClearInterOpParallelismThreads () Nodes that perform blocking operations are enqueued on a pool of inter_op_parallelism_threads available in each process. |
| ConfigProto.Builder | ClearIntraOpParallelismThreads () The execution of an individual op (for some op types) can be parallelized on a pool of intra_op_parallelism_threads. |
| ConfigProto.Builder | ClearIsolateSessionState () If true, any resources such as Variables used in the session will not be shared with other sessions. |
| ConfigProto.Builder | ClearLogDevicePlacement () Whether device placements should be logged. |
| ConfigProto.Builder | ClearOneof (com.google.protobuf.Descriptors.OneofDescriptor oneof) |
| ConfigProto.Builder | ClearOperationTimeoutInMs () Global timeout for all blocking operations in this session. |
| ConfigProto.Builder | كليربلاسيمينتبيريود () Assignment of Nodes to Devices is recomputed every placement_period steps until the system warms up (at which point the recomputation typically slows down automatically). |
| ConfigProto.Builder | كليررببوكتيونس () Options that apply when this session uses the distributed runtime. |
| ConfigProto.Builder | ClearSessionInterOpThreadPool () This option is experimental - it may be replaced with a different mechanism in the future. |
| ConfigProto.Builder | ClearShareClusterDevicesInSession () When true, WorkerSessions are created with device attributes from the full cluster. |
| ConfigProto.Builder | ClearUsePerSessionThreads () If true, use a new set of threads for this session rather than the global pool of threads. |
| ConfigProto.Builder | استنساخ () |
| منطقية | يحتوي علىDeviceCount (مفتاح السلسلة) Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. |
| منطقية | الحصول علىAllowSoftPlacement () Whether soft placement is allowed. |
| ClusterDef | الحصول على ClusterDef () Optional list of all workers to use in this session. |
| ClusterDef.Builder | الحصول على ClusterDefBuilder () Optional list of all workers to use in this session. |
| ClusterDefOrBuilder | الحصول على ClusterDefOrBuilder () Optional list of all workers to use in this session. |
| ConfigProto | |
| النهائي الثابت com.google.protobuf.Descriptors.Descriptor | |
| com.google.protobuf.Descriptors.Descriptor | |
| خريطة<سلسلة، عدد صحيح> | getDeviceCount () استخدم getDeviceCountMap() بدلاً من ذلك. |
| كثافة العمليات | getDeviceCountCount () Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. |
| خريطة<سلسلة، عدد صحيح> | getDeviceCountMap () Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. |
| كثافة العمليات | getDeviceCountOrDefault (مفتاح السلسلة، int defaultValue) Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. |
| كثافة العمليات | getDeviceCountOrThrow (مفتاح السلسلة) Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. |
| خيط | getDeviceFilters (فهرس كثافة العمليات) When any filters are present sessions will ignore all devices which do not match the filters. |
| com.google.protobuf.ByteString | getDeviceFiltersBytes (فهرس كثافة العمليات) When any filters are present sessions will ignore all devices which do not match the filters. |
| كثافة العمليات | getDeviceFiltersCount () When any filters are present sessions will ignore all devices which do not match the filters. |
| com.google.protobuf.ProtocolStringList | getDeviceFiltersList () When any filters are present sessions will ignore all devices which do not match the filters. |
| ConfigProto.Experimental | الحصول التجريبي () .tensorflow.ConfigProto.Experimental experimental = 16; |
| ConfigProto.Experimental.Builder | الحصول على التجريبية () .tensorflow.ConfigProto.Experimental experimental = 16; |
| ConfigProto.ExperimentalOrBuilder | الحصول على التجريبية أو البناء () .tensorflow.ConfigProto.Experimental experimental = 16; |
| خيارات GPU | خيارات getGpu () Options that apply to all GPUs. |
| GPUOptions.Builder | الحصول علىGpuOptionsBuilder () Options that apply to all GPUs. |
| GPUOptionsOrBuilder | getGpuOptionsOrBuilder () Options that apply to all GPUs. |
| خيارات الرسم البياني | خيارات الرسم البياني () Options that apply to all graphs. |
| GraphOptions.Builder | الحصول على GraphOptionsBuilder () Options that apply to all graphs. |
| GraphOptionsOrBuilder | الحصول على GraphOptionsOrBuilder () Options that apply to all graphs. |
| كثافة العمليات | getInterOpParallelismThreads () Nodes that perform blocking operations are enqueued on a pool of inter_op_parallelism_threads available in each process. |
| كثافة العمليات | getIntraOpParallelismThreads () The execution of an individual op (for some op types) can be parallelized on a pool of intra_op_parallelism_threads. |
| منطقية | getIsolateSessionState () If true, any resources such as Variables used in the session will not be shared with other sessions. |
| منطقية | getLogDevicePlacement () Whether device placements should be logged. |
| خريطة<سلسلة، عدد صحيح> | getMutableDeviceCount () استخدم أدوات الوصول البديلة للطفرات بدلاً من ذلك. |
| طويل | الحصول علىOperationTimeoutInMs () Global timeout for all blocking operations in this session. |
| كثافة العمليات | الحصول على فترة التنسيب () Assignment of Nodes to Devices is recomputed every placement_period steps until the system warms up (at which point the recomputation typically slows down automatically). |
| خيارات RPC | خيارات الحصول على Rpc () Options that apply when this session uses the distributed runtime. |
| RPCOptions.Builder | الحصول علىRpcOptionsBuilder () Options that apply when this session uses the distributed runtime. |
| RPCOptionsOrBuilder | getRpcOptionsOrBuilder () Options that apply when this session uses the distributed runtime. |
| ThreadPoolOptionProto | getSessionInterOpThreadPool (فهرس كثافة العمليات) This option is experimental - it may be replaced with a different mechanism in the future. |
| ThreadPoolOptionProto.Builder | getSessionInterOpThreadPoolBuilder (فهرس كثافة العمليات) This option is experimental - it may be replaced with a different mechanism in the future. |
| القائمة< ThreadPoolOptionProto.Builder > | getSessionInterOpThreadPoolBuilderList () This option is experimental - it may be replaced with a different mechanism in the future. |
| كثافة العمليات | getSessionInterOpThreadPoolCount () This option is experimental - it may be replaced with a different mechanism in the future. |
| القائمة< ThreadPoolOptionProto > | getSessionInterOpThreadPoolList () This option is experimental - it may be replaced with a different mechanism in the future. |
| ThreadPoolOptionProtoOrBuilder | getSessionInterOpThreadPoolOrBuilder (فهرس كثافة العمليات) This option is experimental - it may be replaced with a different mechanism in the future. |
| القائمة<؟ يمتد ThreadPoolOptionProtoOrBuilder > | getSessionInterOpThreadPoolOrBuilderList () This option is experimental - it may be replaced with a different mechanism in the future. |
| منطقية | getShareClusterDevicesInSession () When true, WorkerSessions are created with device attributes from the full cluster. |
| منطقية | getUsePerSessionThreads () If true, use a new set of threads for this session rather than the global pool of threads. |
| منطقية | hasClusterDef () Optional list of all workers to use in this session. |
| منطقية | تجريبي () .tensorflow.ConfigProto.Experimental experimental = 16; |
| منطقية | hasGpuOptions () Options that apply to all GPUs. |
| منطقية | hasGraphOptions () Options that apply to all graphs. |
| منطقية | hasRpcOptions () Options that apply when this session uses the distributed runtime. |
| منطقية نهائية | تمت التهيئة () |
| ConfigProto.Builder | |
| ConfigProto.Builder | |
| ConfigProto.Builder | دمج من (com.google.protobuf.Message أخرى) |
| ConfigProto.Builder | دمج من (com.google.protobuf.CodedInputStream input، com.google.protobuf.ExtensionRegistryLite ExtensionRegistry) |
| ConfigProto.Builder | |
| ConfigProto.Builder | |
| ConfigProto.Builder | mergeRpcOptions (قيمة RPCOptions ) Options that apply when this session uses the distributed runtime. |
| النهائي ConfigProto.Builder | دمجUnknownFields (com.google.protobuf.UnknownFieldSet UnknownFields) |
| ConfigProto.Builder | putAllDeviceCount (قيم Map<String, Integer>) Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. |
| ConfigProto.Builder | putDeviceCount (مفتاح السلسلة، قيمة int) Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. |
| ConfigProto.Builder | RemoveDeviceCount (مفتاح السلسلة) Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. |
| ConfigProto.Builder | إزالةSessionInterOpThreadPool (فهرس كثافة العمليات) This option is experimental - it may be replaced with a different mechanism in the future. |
| ConfigProto.Builder | setAllowSoftPlacement (قيمة منطقية) Whether soft placement is allowed. |
| ConfigProto.Builder | setClusterDef ( ClusterDef.Builder builderForValue) Optional list of all workers to use in this session. |
| ConfigProto.Builder | |
| ConfigProto.Builder | setDeviceFilters (مؤشر int، قيمة السلسلة) When any filters are present sessions will ignore all devices which do not match the filters. |
| ConfigProto.Builder | setExperimental (قيمة ConfigProto.Experimental ) .tensorflow.ConfigProto.Experimental experimental = 16; |
| ConfigProto.Builder | setExperimental ( ConfigProto.Experimental.Builder builderForValue) .tensorflow.ConfigProto.Experimental experimental = 16; |
| ConfigProto.Builder | setField (حقل com.google.protobuf.Descriptors.FieldDescriptor، قيمة الكائن) |
| ConfigProto.Builder | |
| ConfigProto.Builder | |
| ConfigProto.Builder | |
| ConfigProto.Builder | |
| ConfigProto.Builder | setInterOpParallelismThreads (قيمة int) Nodes that perform blocking operations are enqueued on a pool of inter_op_parallelism_threads available in each process. |
| ConfigProto.Builder | setIntraOpParallelismThreads (قيمة int) The execution of an individual op (for some op types) can be parallelized on a pool of intra_op_parallelism_threads. |
| ConfigProto.Builder | setIsolateSessionState (قيمة منطقية) If true, any resources such as Variables used in the session will not be shared with other sessions. |
| ConfigProto.Builder | setLogDevicePlacement (قيمة منطقية) Whether device placements should be logged. |
| ConfigProto.Builder | setOperationTimeoutInMs (قيمة طويلة) Global timeout for all blocking operations in this session. |
| ConfigProto.Builder | setPlacementPeriod (قيمة int) Assignment of Nodes to Devices is recomputed every placement_period steps until the system warms up (at which point the recomputation typically slows down automatically). |
| ConfigProto.Builder | setRepeatedField (حقل com.google.protobuf.Descriptors.FieldDescriptor، فهرس int، قيمة الكائن) |
| ConfigProto.Builder | |
| ConfigProto.Builder | setRpcOptions ( RPCOptions.Builder builderForValue) Options that apply when this session uses the distributed runtime. |
| ConfigProto.Builder | setSessionInterOpThreadPool (مؤشر int، قيمة ThreadPoolOptionProto ) This option is experimental - it may be replaced with a different mechanism in the future. |
| ConfigProto.Builder | setSessionInterOpThreadPool (فهرس كثافة العمليات، ThreadPoolOptionProto.Builder builderForValue) This option is experimental - it may be replaced with a different mechanism in the future. |
| ConfigProto.Builder | setShareClusterDevicesInSession (قيمة منطقية) When true, WorkerSessions are created with device attributes from the full cluster. |
| النهائي ConfigProto.Builder | setUnknownFields (com.google.protobuf.UnknownFieldSet UnknownFields) |
| ConfigProto.Builder | setUsePerSessionThreads (قيمة منطقية) If true, use a new set of threads for this session rather than the global pool of threads. |
الطرق الموروثة
الأساليب العامة
public ConfigProto.Builder addAllDeviceFilters (قيم Iterable<String>)
When any filters are present sessions will ignore all devices which do not match the filters. Each filter can be partially specified, e.g. "/job:ps" "/job:worker/replica:3", etc.
repeated string device_filters = 4; public ConfigProto.Builder addAllSessionInterOpThreadPool (Iterable<؟ يمتد ThreadPoolOptionProto > القيم)
This option is experimental - it may be replaced with a different mechanism in the future. Configures session thread pools. If this is configured, then RunOptions for a Run call can select the thread pool to use. The intended use is for when some session invocations need to run in a background pool limited to a small number of threads: - For example, a session may be configured to have one large pool (for regular compute) and one small pool (for periodic, low priority work); using the small pool is currently the mechanism for limiting the inter-op parallelism of the low priority work. Note that it does not limit the parallelism of work spawned by a single op kernel implementation. - Using this setting is normally not needed in training, but may help some serving use cases. - It is also generally recommended to set the global_name field of this proto, to avoid creating multiple large pools. It is typically better to run the non-low-priority work, even across sessions, in a single large pool.
repeated .tensorflow.ThreadPoolOptionProto session_inter_op_thread_pool = 12; public ConfigProto.Builder addDeviceFilters (قيمة السلسلة)
When any filters are present sessions will ignore all devices which do not match the filters. Each filter can be partially specified, e.g. "/job:ps" "/job:worker/replica:3", etc.
repeated string device_filters = 4; public ConfigProto.Builder addDeviceFiltersBytes (قيمة com.google.protobuf.ByteString)
When any filters are present sessions will ignore all devices which do not match the filters. Each filter can be partially specified, e.g. "/job:ps" "/job:worker/replica:3", etc.
repeated string device_filters = 4; public ConfigProto.Builder addRepeatedField (حقل com.google.protobuf.Descriptors.FieldDescriptor، قيمة الكائن)
public ConfigProto.Builder addSessionInterOpThreadPool (قيمة ThreadPoolOptionProto )
This option is experimental - it may be replaced with a different mechanism in the future. Configures session thread pools. If this is configured, then RunOptions for a Run call can select the thread pool to use. The intended use is for when some session invocations need to run in a background pool limited to a small number of threads: - For example, a session may be configured to have one large pool (for regular compute) and one small pool (for periodic, low priority work); using the small pool is currently the mechanism for limiting the inter-op parallelism of the low priority work. Note that it does not limit the parallelism of work spawned by a single op kernel implementation. - Using this setting is normally not needed in training, but may help some serving use cases. - It is also generally recommended to set the global_name field of this proto, to avoid creating multiple large pools. It is typically better to run the non-low-priority work, even across sessions, in a single large pool.
repeated .tensorflow.ThreadPoolOptionProto session_inter_op_thread_pool = 12; public ConfigProto.Builder addSessionInterOpThreadPool (مؤشر int، قيمة ThreadPoolOptionProto )
This option is experimental - it may be replaced with a different mechanism in the future. Configures session thread pools. If this is configured, then RunOptions for a Run call can select the thread pool to use. The intended use is for when some session invocations need to run in a background pool limited to a small number of threads: - For example, a session may be configured to have one large pool (for regular compute) and one small pool (for periodic, low priority work); using the small pool is currently the mechanism for limiting the inter-op parallelism of the low priority work. Note that it does not limit the parallelism of work spawned by a single op kernel implementation. - Using this setting is normally not needed in training, but may help some serving use cases. - It is also generally recommended to set the global_name field of this proto, to avoid creating multiple large pools. It is typically better to run the non-low-priority work, even across sessions, in a single large pool.
repeated .tensorflow.ThreadPoolOptionProto session_inter_op_thread_pool = 12; public ConfigProto.Builder addSessionInterOpThreadPool ( ThreadPoolOptionProto.Builder builderForValue)
This option is experimental - it may be replaced with a different mechanism in the future. Configures session thread pools. If this is configured, then RunOptions for a Run call can select the thread pool to use. The intended use is for when some session invocations need to run in a background pool limited to a small number of threads: - For example, a session may be configured to have one large pool (for regular compute) and one small pool (for periodic, low priority work); using the small pool is currently the mechanism for limiting the inter-op parallelism of the low priority work. Note that it does not limit the parallelism of work spawned by a single op kernel implementation. - Using this setting is normally not needed in training, but may help some serving use cases. - It is also generally recommended to set the global_name field of this proto, to avoid creating multiple large pools. It is typically better to run the non-low-priority work, even across sessions, in a single large pool.
repeated .tensorflow.ThreadPoolOptionProto session_inter_op_thread_pool = 12; public ConfigProto.Builder addSessionInterOpThreadPool (مؤشر int، ThreadPoolOptionProto.Builder builderForValue)
This option is experimental - it may be replaced with a different mechanism in the future. Configures session thread pools. If this is configured, then RunOptions for a Run call can select the thread pool to use. The intended use is for when some session invocations need to run in a background pool limited to a small number of threads: - For example, a session may be configured to have one large pool (for regular compute) and one small pool (for periodic, low priority work); using the small pool is currently the mechanism for limiting the inter-op parallelism of the low priority work. Note that it does not limit the parallelism of work spawned by a single op kernel implementation. - Using this setting is normally not needed in training, but may help some serving use cases. - It is also generally recommended to set the global_name field of this proto, to avoid creating multiple large pools. It is typically better to run the non-low-priority work, even across sessions, in a single large pool.
repeated .tensorflow.ThreadPoolOptionProto session_inter_op_thread_pool = 12; ThreadPoolOptionProto.Builder العام addSessionInterOpThreadPoolBuilder (فهرس كثافة العمليات)
This option is experimental - it may be replaced with a different mechanism in the future. Configures session thread pools. If this is configured, then RunOptions for a Run call can select the thread pool to use. The intended use is for when some session invocations need to run in a background pool limited to a small number of threads: - For example, a session may be configured to have one large pool (for regular compute) and one small pool (for periodic, low priority work); using the small pool is currently the mechanism for limiting the inter-op parallelism of the low priority work. Note that it does not limit the parallelism of work spawned by a single op kernel implementation. - Using this setting is normally not needed in training, but may help some serving use cases. - It is also generally recommended to set the global_name field of this proto, to avoid creating multiple large pools. It is typically better to run the non-low-priority work, even across sessions, in a single large pool.
repeated .tensorflow.ThreadPoolOptionProto session_inter_op_thread_pool = 12; ThreadPoolOptionProto.Builder العام addSessionInterOpThreadPoolBuilder ()
This option is experimental - it may be replaced with a different mechanism in the future. Configures session thread pools. If this is configured, then RunOptions for a Run call can select the thread pool to use. The intended use is for when some session invocations need to run in a background pool limited to a small number of threads: - For example, a session may be configured to have one large pool (for regular compute) and one small pool (for periodic, low priority work); using the small pool is currently the mechanism for limiting the inter-op parallelism of the low priority work. Note that it does not limit the parallelism of work spawned by a single op kernel implementation. - Using this setting is normally not needed in training, but may help some serving use cases. - It is also generally recommended to set the global_name field of this proto, to avoid creating multiple large pools. It is typically better to run the non-low-priority work, even across sessions, in a single large pool.
repeated .tensorflow.ThreadPoolOptionProto session_inter_op_thread_pool = 12; ConfigProto.Builder العام ClearAllowSoftPlacement ()
Whether soft placement is allowed. If allow_soft_placement is true, an op will be placed on CPU if 1. there's no GPU implementation for the OP or 2. no GPU devices are known or registered or 3. need to co-locate with reftype input(s) which are from CPU.
bool allow_soft_placement = 7; ConfigProto.Builder العام ClearClusterDef ()
Optional list of all workers to use in this session.
.tensorflow.ClusterDef cluster_def = 14; ConfigProto.Builder العامة ClearDeviceFilters ()
When any filters are present sessions will ignore all devices which do not match the filters. Each filter can be partially specified, e.g. "/job:ps" "/job:worker/replica:3", etc.
repeated string device_filters = 4; ConfigProto.Builder العام ClearExperimental ()
.tensorflow.ConfigProto.Experimental experimental = 16;
ConfigProto.Builder العام ClearGpuOptions ()
Options that apply to all GPUs.
.tensorflow.GPUOptions gpu_options = 6; ConfigProto.Builder العامة ClearGraphOptions ()
Options that apply to all graphs.
.tensorflow.GraphOptions graph_options = 10; ConfigProto.Builder العام ClearInterOpParallelismThreads ()
Nodes that perform blocking operations are enqueued on a pool of inter_op_parallelism_threads available in each process. 0 means the system picks an appropriate number. Negative means all operations are performed in caller's thread. Note that the first Session created in the process sets the number of threads for all future sessions unless use_per_session_threads is true or session_inter_op_thread_pool is configured.
int32 inter_op_parallelism_threads = 5; ConfigProto.Builder العام ClearIntraOpParallelismThreads ()
The execution of an individual op (for some op types) can be
parallelized on a pool of intra_op_parallelism_threads.
0 means the system picks an appropriate number.
If you create an ordinary session, e.g., from Python or C++,
then there is exactly one intra op thread pool per process.
The first session created determines the number of threads in this pool.
All subsequent sessions reuse/share this one global pool.
There are notable exceptions to the default behavior describe above:
1. There is an environment variable for overriding this thread pool,
named TF_OVERRIDE_GLOBAL_THREADPOOL.
2. When connecting to a server, such as a remote `tf.train.Server`
instance, then this option will be ignored altogether.
int32 intra_op_parallelism_threads = 2; ConfigProto.Builder العام ClearIsolateSessionState ()
If true, any resources such as Variables used in the session will not be shared with other sessions. However, when clusterspec propagation is enabled, this field is ignored and sessions are always isolated.
bool isolate_session_state = 15; ConfigProto.Builder العام ClearLogDevicePlacement ()
Whether device placements should be logged.
bool log_device_placement = 8; ConfigProto.Builder العام ClearOperationTimeoutInMs ()
Global timeout for all blocking operations in this session. If non-zero, and not overridden on a per-operation basis, this value will be used as the deadline for all blocking operations.
int64 operation_timeout_in_ms = 11; ConfigProto.Builder العام ClearPlacementPeriod ()
Assignment of Nodes to Devices is recomputed every placement_period steps until the system warms up (at which point the recomputation typically slows down automatically).
int32 placement_period = 3; ConfigProto.Builder العام ClearRpcOptions ()
Options that apply when this session uses the distributed runtime.
.tensorflow.RPCOptions rpc_options = 13; ConfigProto.Builder العام ClearSessionInterOpThreadPool ()
This option is experimental - it may be replaced with a different mechanism in the future. Configures session thread pools. If this is configured, then RunOptions for a Run call can select the thread pool to use. The intended use is for when some session invocations need to run in a background pool limited to a small number of threads: - For example, a session may be configured to have one large pool (for regular compute) and one small pool (for periodic, low priority work); using the small pool is currently the mechanism for limiting the inter-op parallelism of the low priority work. Note that it does not limit the parallelism of work spawned by a single op kernel implementation. - Using this setting is normally not needed in training, but may help some serving use cases. - It is also generally recommended to set the global_name field of this proto, to avoid creating multiple large pools. It is typically better to run the non-low-priority work, even across sessions, in a single large pool.
repeated .tensorflow.ThreadPoolOptionProto session_inter_op_thread_pool = 12; ConfigProto.Builder العام ClearShareClusterDevicesInSession ()
When true, WorkerSessions are created with device attributes from the full cluster. This is helpful when a worker wants to partition a graph (for example during a PartitionedCallOp).
bool share_cluster_devices_in_session = 17; ConfigProto.Builder العام ClearUsePerSessionThreads ()
If true, use a new set of threads for this session rather than the global pool of threads. Only supported by direct sessions. If false, use the global threads created by the first session, or the per-session thread pools configured by session_inter_op_thread_pool. This option is deprecated. The same effect can be achieved by setting session_inter_op_thread_pool to have one element, whose num_threads equals inter_op_parallelism_threads.
bool use_per_session_threads = 9; المنطقية العامة تحتوي علىDeviceCount (مفتاح السلسلة)
Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. If a particular device type is not found in the map, the system picks an appropriate number.
map<string, int32> device_count = 1; getAllowSoftPlacement () المنطقية العامة
Whether soft placement is allowed. If allow_soft_placement is true, an op will be placed on CPU if 1. there's no GPU implementation for the OP or 2. no GPU devices are known or registered or 3. need to co-locate with reftype input(s) which are from CPU.
bool allow_soft_placement = 7; ClusterDef العامة getClusterDef ()
Optional list of all workers to use in this session.
.tensorflow.ClusterDef cluster_def = 14; ClusterDef.Builder العامة getClusterDefBuilder ()
Optional list of all workers to use in this session.
.tensorflow.ClusterDef cluster_def = 14; getClusterDefOrBuilder العامة getClusterDefOrBuilder ()
Optional list of all workers to use in this session.
.tensorflow.ClusterDef cluster_def = 14; النهائي العام الثابت com.google.protobuf.Descriptors.Descriptor getDescriptor ()
com.google.protobuf.Descriptors.Descriptor getDescriptorForType () العام
int public getDeviceCountCount ()
Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. If a particular device type is not found in the map, the system picks an appropriate number.
map<string, int32> device_count = 1; الخريطة العامة<String, Integer> getDeviceCountMap ()
Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. If a particular device type is not found in the map, the system picks an appropriate number.
map<string, int32> device_count = 1; int public getDeviceCountOrDefault (مفتاح السلسلة، int defaultValue)
Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. If a particular device type is not found in the map, the system picks an appropriate number.
map<string, int32> device_count = 1; int public getDeviceCountOrThrow (مفتاح السلسلة)
Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. If a particular device type is not found in the map, the system picks an appropriate number.
map<string, int32> device_count = 1; سلسلة getDeviceFilters العامة (مؤشر int)
When any filters are present sessions will ignore all devices which do not match the filters. Each filter can be partially specified, e.g. "/job:ps" "/job:worker/replica:3", etc.
repeated string device_filters = 4; com.google.protobuf.ByteString getDeviceFiltersBytes العام (مؤشر int)
When any filters are present sessions will ignore all devices which do not match the filters. Each filter can be partially specified, e.g. "/job:ps" "/job:worker/replica:3", etc.
repeated string device_filters = 4; int public getDeviceFiltersCount ()
When any filters are present sessions will ignore all devices which do not match the filters. Each filter can be partially specified, e.g. "/job:ps" "/job:worker/replica:3", etc.
repeated string device_filters = 4; com.google.protobuf.ProtocolStringList العامة getDeviceFiltersList ()
When any filters are present sessions will ignore all devices which do not match the filters. Each filter can be partially specified, e.g. "/job:ps" "/job:worker/replica:3", etc.
repeated string device_filters = 4; ConfigProto.Experimental.Builder العامة getExperimentalBuilder ()
.tensorflow.ConfigProto.Experimental experimental = 16;
ConfigProto.ExperimentalOrBuilder العامة getExperimentalOrBuilder ()
.tensorflow.ConfigProto.Experimental experimental = 16;
خيارات GPU العامة getGpuOptions ()
Options that apply to all GPUs.
.tensorflow.GPUOptions gpu_options = 6; GPUOptions.Builder العام getGpuOptionsBuilder ()
Options that apply to all GPUs.
.tensorflow.GPUOptions gpu_options = 6; GPUOptionsOrBuilder العام getGpuOptionsOrBuilder ()
Options that apply to all GPUs.
.tensorflow.GPUOptions gpu_options = 6; خيارات GraphOptions العامة getGraphOptions ()
Options that apply to all graphs.
.tensorflow.GraphOptions graph_options = 10; GraphOptions.Builder العامة getGraphOptionsBuilder ()
Options that apply to all graphs.
.tensorflow.GraphOptions graph_options = 10; GraphOptionsOrBuilder العام getGraphOptionsOrBuilder ()
Options that apply to all graphs.
.tensorflow.GraphOptions graph_options = 10; int public getInterOpParallelismThreads ()
Nodes that perform blocking operations are enqueued on a pool of inter_op_parallelism_threads available in each process. 0 means the system picks an appropriate number. Negative means all operations are performed in caller's thread. Note that the first Session created in the process sets the number of threads for all future sessions unless use_per_session_threads is true or session_inter_op_thread_pool is configured.
int32 inter_op_parallelism_threads = 5; int العام getIntraOpParallelismThreads ()
The execution of an individual op (for some op types) can be
parallelized on a pool of intra_op_parallelism_threads.
0 means the system picks an appropriate number.
If you create an ordinary session, e.g., from Python or C++,
then there is exactly one intra op thread pool per process.
The first session created determines the number of threads in this pool.
All subsequent sessions reuse/share this one global pool.
There are notable exceptions to the default behavior describe above:
1. There is an environment variable for overriding this thread pool,
named TF_OVERRIDE_GLOBAL_THREADPOOL.
2. When connecting to a server, such as a remote `tf.train.Server`
instance, then this option will be ignored altogether.
int32 intra_op_parallelism_threads = 2; getIsolateSessionState () المنطقية العامة
If true, any resources such as Variables used in the session will not be shared with other sessions. However, when clusterspec propagation is enabled, this field is ignored and sessions are always isolated.
bool isolate_session_state = 15; getLogDevicePlacement () المنطقية العامة
Whether device placements should be logged.
bool log_device_placement = 8; الخريطة العامة<String, Integer> getMutableDeviceCount ()
استخدم أدوات الوصول البديلة للطفرات بدلاً من ذلك.
getOperationTimeoutInMs العامة الطويلة ()
Global timeout for all blocking operations in this session. If non-zero, and not overridden on a per-operation basis, this value will be used as the deadline for all blocking operations.
int64 operation_timeout_in_ms = 11; int public getPlacementPeriod ()
Assignment of Nodes to Devices is recomputed every placement_period steps until the system warms up (at which point the recomputation typically slows down automatically).
int32 placement_period = 3; RPCOptions العامة getRpcOptions ()
Options that apply when this session uses the distributed runtime.
.tensorflow.RPCOptions rpc_options = 13; RPCOptions.Builder العامة getRpcOptionsBuilder ()
Options that apply when this session uses the distributed runtime.
.tensorflow.RPCOptions rpc_options = 13; RPCOptionsOrBuilder العام getRpcOptionsOrBuilder ()
Options that apply when this session uses the distributed runtime.
.tensorflow.RPCOptions rpc_options = 13; ThreadPoolOptionProto public getSessionInterOpThreadPool (مؤشر int)
This option is experimental - it may be replaced with a different mechanism in the future. Configures session thread pools. If this is configured, then RunOptions for a Run call can select the thread pool to use. The intended use is for when some session invocations need to run in a background pool limited to a small number of threads: - For example, a session may be configured to have one large pool (for regular compute) and one small pool (for periodic, low priority work); using the small pool is currently the mechanism for limiting the inter-op parallelism of the low priority work. Note that it does not limit the parallelism of work spawned by a single op kernel implementation. - Using this setting is normally not needed in training, but may help some serving use cases. - It is also generally recommended to set the global_name field of this proto, to avoid creating multiple large pools. It is typically better to run the non-low-priority work, even across sessions, in a single large pool.
repeated .tensorflow.ThreadPoolOptionProto session_inter_op_thread_pool = 12; ThreadPoolOptionProto.Builder العام getSessionInterOpThreadPoolBuilder (فهرس كثافة العمليات)
This option is experimental - it may be replaced with a different mechanism in the future. Configures session thread pools. If this is configured, then RunOptions for a Run call can select the thread pool to use. The intended use is for when some session invocations need to run in a background pool limited to a small number of threads: - For example, a session may be configured to have one large pool (for regular compute) and one small pool (for periodic, low priority work); using the small pool is currently the mechanism for limiting the inter-op parallelism of the low priority work. Note that it does not limit the parallelism of work spawned by a single op kernel implementation. - Using this setting is normally not needed in training, but may help some serving use cases. - It is also generally recommended to set the global_name field of this proto, to avoid creating multiple large pools. It is typically better to run the non-low-priority work, even across sessions, in a single large pool.
repeated .tensorflow.ThreadPoolOptionProto session_inter_op_thread_pool = 12; القائمة العامة< ThreadPoolOptionProto.Builder > getSessionInterOpThreadPoolBuilderList ()
This option is experimental - it may be replaced with a different mechanism in the future. Configures session thread pools. If this is configured, then RunOptions for a Run call can select the thread pool to use. The intended use is for when some session invocations need to run in a background pool limited to a small number of threads: - For example, a session may be configured to have one large pool (for regular compute) and one small pool (for periodic, low priority work); using the small pool is currently the mechanism for limiting the inter-op parallelism of the low priority work. Note that it does not limit the parallelism of work spawned by a single op kernel implementation. - Using this setting is normally not needed in training, but may help some serving use cases. - It is also generally recommended to set the global_name field of this proto, to avoid creating multiple large pools. It is typically better to run the non-low-priority work, even across sessions, in a single large pool.
repeated .tensorflow.ThreadPoolOptionProto session_inter_op_thread_pool = 12; int public getSessionInterOpThreadPoolCount ()
This option is experimental - it may be replaced with a different mechanism in the future. Configures session thread pools. If this is configured, then RunOptions for a Run call can select the thread pool to use. The intended use is for when some session invocations need to run in a background pool limited to a small number of threads: - For example, a session may be configured to have one large pool (for regular compute) and one small pool (for periodic, low priority work); using the small pool is currently the mechanism for limiting the inter-op parallelism of the low priority work. Note that it does not limit the parallelism of work spawned by a single op kernel implementation. - Using this setting is normally not needed in training, but may help some serving use cases. - It is also generally recommended to set the global_name field of this proto, to avoid creating multiple large pools. It is typically better to run the non-low-priority work, even across sessions, in a single large pool.
repeated .tensorflow.ThreadPoolOptionProto session_inter_op_thread_pool = 12; القائمة العامة< ThreadPoolOptionProto > getSessionInterOpThreadPoolList ()
This option is experimental - it may be replaced with a different mechanism in the future. Configures session thread pools. If this is configured, then RunOptions for a Run call can select the thread pool to use. The intended use is for when some session invocations need to run in a background pool limited to a small number of threads: - For example, a session may be configured to have one large pool (for regular compute) and one small pool (for periodic, low priority work); using the small pool is currently the mechanism for limiting the inter-op parallelism of the low priority work. Note that it does not limit the parallelism of work spawned by a single op kernel implementation. - Using this setting is normally not needed in training, but may help some serving use cases. - It is also generally recommended to set the global_name field of this proto, to avoid creating multiple large pools. It is typically better to run the non-low-priority work, even across sessions, in a single large pool.
repeated .tensorflow.ThreadPoolOptionProto session_inter_op_thread_pool = 12; ThreadPoolOptionProtoOrBuilder العام getSessionInterOpThreadPoolOrBuilder (فهرس كثافة العمليات)
This option is experimental - it may be replaced with a different mechanism in the future. Configures session thread pools. If this is configured, then RunOptions for a Run call can select the thread pool to use. The intended use is for when some session invocations need to run in a background pool limited to a small number of threads: - For example, a session may be configured to have one large pool (for regular compute) and one small pool (for periodic, low priority work); using the small pool is currently the mechanism for limiting the inter-op parallelism of the low priority work. Note that it does not limit the parallelism of work spawned by a single op kernel implementation. - Using this setting is normally not needed in training, but may help some serving use cases. - It is also generally recommended to set the global_name field of this proto, to avoid creating multiple large pools. It is typically better to run the non-low-priority work, even across sessions, in a single large pool.
repeated .tensorflow.ThreadPoolOptionProto session_inter_op_thread_pool = 12; القائمة العامة <؟ يمتد ThreadPoolOptionProtoOrBuilder > getSessionInterOpThreadPoolOrBuilderList ()
This option is experimental - it may be replaced with a different mechanism in the future. Configures session thread pools. If this is configured, then RunOptions for a Run call can select the thread pool to use. The intended use is for when some session invocations need to run in a background pool limited to a small number of threads: - For example, a session may be configured to have one large pool (for regular compute) and one small pool (for periodic, low priority work); using the small pool is currently the mechanism for limiting the inter-op parallelism of the low priority work. Note that it does not limit the parallelism of work spawned by a single op kernel implementation. - Using this setting is normally not needed in training, but may help some serving use cases. - It is also generally recommended to set the global_name field of this proto, to avoid creating multiple large pools. It is typically better to run the non-low-priority work, even across sessions, in a single large pool.
repeated .tensorflow.ThreadPoolOptionProto session_inter_op_thread_pool = 12; getShareClusterDevicesInSession () المنطقية العامة
When true, WorkerSessions are created with device attributes from the full cluster. This is helpful when a worker wants to partition a graph (for example during a PartitionedCallOp).
bool share_cluster_devices_in_session = 17; getUsePerSessionThreads المنطقية العامة ()
If true, use a new set of threads for this session rather than the global pool of threads. Only supported by direct sessions. If false, use the global threads created by the first session, or the per-session thread pools configured by session_inter_op_thread_pool. This option is deprecated. The same effect can be achieved by setting session_inter_op_thread_pool to have one element, whose num_threads equals inter_op_parallelism_threads.
bool use_per_session_threads = 9; المنطق المنطقي العام hasClusterDef ()
Optional list of all workers to use in this session.
.tensorflow.ClusterDef cluster_def = 14; القيمة المنطقية العامة تجريبية ()
.tensorflow.ConfigProto.Experimental experimental = 16;
المنطق المنطقي العام hasGpuOptions ()
Options that apply to all GPUs.
.tensorflow.GPUOptions gpu_options = 6; المنطق المنطقي العام hasGraphOptions ()
Options that apply to all graphs.
.tensorflow.GraphOptions graph_options = 10; hasRpcOptions () المنطقية العامة
Options that apply when this session uses the distributed runtime.
.tensorflow.RPCOptions rpc_options = 13; تمت تهيئة القيمة المنطقية النهائية العامة ()
ConfigProto.Builder العام mergeClusterDef (قيمة ClusterDef )
Optional list of all workers to use in this session.
.tensorflow.ClusterDef cluster_def = 14; دمج ConfigProto.Builder العام (قيمة ConfigProto.Experimental )
.tensorflow.ConfigProto.Experimental experimental = 16;
عام ConfigProto.Builder mergeFrom (com.google.protobuf.CodedInputStream input, com.google.protobuf.ExtensionRegistryLite ExtensionRegistry)
رميات
| IOEException |
|---|
عام ConfigProto.Builder mergeGpuOptions (قيمة GPUOptions )
Options that apply to all GPUs.
.tensorflow.GPUOptions gpu_options = 6; عام ConfigProto.Builder mergeGraphOptions (قيمة GraphOptions )
Options that apply to all graphs.
.tensorflow.GraphOptions graph_options = 10; عام ConfigProto.Builder mergeRpcOptions (قيمة RPCOptions )
Options that apply when this session uses the distributed runtime.
.tensorflow.RPCOptions rpc_options = 13; النهائي العام ConfigProto.Builder mergeUnknownFields (com.google.protobuf.UnknownFieldSetUnknownFields)
public ConfigProto.Builder putAllDeviceCount (قيم Map<String, Integer>)
Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. If a particular device type is not found in the map, the system picks an appropriate number.
map<string, int32> device_count = 1; ConfigProto.Builder العام putDeviceCount (مفتاح السلسلة، قيمة int)
Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. If a particular device type is not found in the map, the system picks an appropriate number.
map<string, int32> device_count = 1; إزالة ConfigProto.Builder العامة (مفتاح السلسلة)
Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. If a particular device type is not found in the map, the system picks an appropriate number.
map<string, int32> device_count = 1; ConfigProto.Builder العام RemoveSessionInterOpThreadPool (فهرس كثافة العمليات)
This option is experimental - it may be replaced with a different mechanism in the future. Configures session thread pools. If this is configured, then RunOptions for a Run call can select the thread pool to use. The intended use is for when some session invocations need to run in a background pool limited to a small number of threads: - For example, a session may be configured to have one large pool (for regular compute) and one small pool (for periodic, low priority work); using the small pool is currently the mechanism for limiting the inter-op parallelism of the low priority work. Note that it does not limit the parallelism of work spawned by a single op kernel implementation. - Using this setting is normally not needed in training, but may help some serving use cases. - It is also generally recommended to set the global_name field of this proto, to avoid creating multiple large pools. It is typically better to run the non-low-priority work, even across sessions, in a single large pool.
repeated .tensorflow.ThreadPoolOptionProto session_inter_op_thread_pool = 12; مجموعة ConfigProto.Builder العامة AllowSoftPlacement (قيمة منطقية)
Whether soft placement is allowed. If allow_soft_placement is true, an op will be placed on CPU if 1. there's no GPU implementation for the OP or 2. no GPU devices are known or registered or 3. need to co-locate with reftype input(s) which are from CPU.
bool allow_soft_placement = 7; مجموعة ConfigProto.Builder العامة ClusterDef ( ClusterDef.Builder builderForValue)
Optional list of all workers to use in this session.
.tensorflow.ClusterDef cluster_def = 14; مجموعة ConfigProto.Builder العامة ClusterDef (قيمة ClusterDef )
Optional list of all workers to use in this session.
.tensorflow.ClusterDef cluster_def = 14; مجموعة ConfigProto.Builder العامة (مؤشر int، قيمة السلسلة)
When any filters are present sessions will ignore all devices which do not match the filters. Each filter can be partially specified, e.g. "/job:ps" "/job:worker/replica:3", etc.
repeated string device_filters = 4; مجموعة ConfigProto.Builder العامة التجريبية (قيمة ConfigProto.Experimental )
.tensorflow.ConfigProto.Experimental experimental = 16;
مجموعة ConfigProto.Builder العامة التجريبية ( ConfigProto.Experimental.Builder builderForValue)
.tensorflow.ConfigProto.Experimental experimental = 16;
public ConfigProto.Builder setField (حقل com.google.protobuf.Descriptors.FieldDescriptor، قيمة الكائن)
مجموعة ConfigProto.Builder العامة GpuOptions ( GPUOptions.Builder builderForValue)
Options that apply to all GPUs.
.tensorflow.GPUOptions gpu_options = 6; مجموعة ConfigProto.Builder العامة GpuOptions (قيمة GPUOptions )
Options that apply to all GPUs.
.tensorflow.GPUOptions gpu_options = 6; مجموعة ConfigProto.Builder العامة GraphOptions ( GraphOptions.Builder builderForValue)
Options that apply to all graphs.
.tensorflow.GraphOptions graph_options = 10; مجموعة ConfigProto.Builder العامة GraphOptions (قيمة GraphOptions )
Options that apply to all graphs.
.tensorflow.GraphOptions graph_options = 10; مجموعة ConfigProto.Builder العامة InterOpParallelismThreads (قيمة int)
Nodes that perform blocking operations are enqueued on a pool of inter_op_parallelism_threads available in each process. 0 means the system picks an appropriate number. Negative means all operations are performed in caller's thread. Note that the first Session created in the process sets the number of threads for all future sessions unless use_per_session_threads is true or session_inter_op_thread_pool is configured.
int32 inter_op_parallelism_threads = 5; مجموعة ConfigProto.Builder العامة IntraOpParallelismThreads (قيمة int)
The execution of an individual op (for some op types) can be
parallelized on a pool of intra_op_parallelism_threads.
0 means the system picks an appropriate number.
If you create an ordinary session, e.g., from Python or C++,
then there is exactly one intra op thread pool per process.
The first session created determines the number of threads in this pool.
All subsequent sessions reuse/share this one global pool.
There are notable exceptions to the default behavior describe above:
1. There is an environment variable for overriding this thread pool,
named TF_OVERRIDE_GLOBAL_THREADPOOL.
2. When connecting to a server, such as a remote `tf.train.Server`
instance, then this option will be ignored altogether.
int32 intra_op_parallelism_threads = 2; مجموعة ConfigProto.Builder العامة IsolateSessionState (قيمة منطقية)
If true, any resources such as Variables used in the session will not be shared with other sessions. However, when clusterspec propagation is enabled, this field is ignored and sessions are always isolated.
bool isolate_session_state = 15; مجموعة ConfigProto.Builder العامة ، LogDevicePlacement (قيمة منطقية)
Whether device placements should be logged.
bool log_device_placement = 8; ConfigProto.Builder العام setOperationTimeoutInMs (قيمة طويلة)
Global timeout for all blocking operations in this session. If non-zero, and not overridden on a per-operation basis, this value will be used as the deadline for all blocking operations.
int64 operation_timeout_in_ms = 11; مجموعة ConfigProto.Builder العامة PlacementPeriod (قيمة int)
Assignment of Nodes to Devices is recomputed every placement_period steps until the system warms up (at which point the recomputation typically slows down automatically).
int32 placement_period = 3; ConfigProto.Builder العام setRepeatedField (حقل com.google.protobuf.Descriptors.FieldDescriptor، مؤشر int، قيمة الكائن)
مجموعة ConfigProto.Builder العامة (قيمة RPCOptions )
Options that apply when this session uses the distributed runtime.
.tensorflow.RPCOptions rpc_options = 13; مجموعة ConfigProto.Builder العامة ( RPCOptions.Builder builderForValue)
Options that apply when this session uses the distributed runtime.
.tensorflow.RPCOptions rpc_options = 13; مجموعة ConfigProto.Builder العامة SessionInterOpThreadPool (مؤشر int، قيمة ThreadPoolOptionProto )
This option is experimental - it may be replaced with a different mechanism in the future. Configures session thread pools. If this is configured, then RunOptions for a Run call can select the thread pool to use. The intended use is for when some session invocations need to run in a background pool limited to a small number of threads: - For example, a session may be configured to have one large pool (for regular compute) and one small pool (for periodic, low priority work); using the small pool is currently the mechanism for limiting the inter-op parallelism of the low priority work. Note that it does not limit the parallelism of work spawned by a single op kernel implementation. - Using this setting is normally not needed in training, but may help some serving use cases. - It is also generally recommended to set the global_name field of this proto, to avoid creating multiple large pools. It is typically better to run the non-low-priority work, even across sessions, in a single large pool.
repeated .tensorflow.ThreadPoolOptionProto session_inter_op_thread_pool = 12; مجموعة ConfigProto.Builder العامة SessionInterOpThreadPool (مؤشر int، ThreadPoolOptionProto.Builder builderForValue)
This option is experimental - it may be replaced with a different mechanism in the future. Configures session thread pools. If this is configured, then RunOptions for a Run call can select the thread pool to use. The intended use is for when some session invocations need to run in a background pool limited to a small number of threads: - For example, a session may be configured to have one large pool (for regular compute) and one small pool (for periodic, low priority work); using the small pool is currently the mechanism for limiting the inter-op parallelism of the low priority work. Note that it does not limit the parallelism of work spawned by a single op kernel implementation. - Using this setting is normally not needed in training, but may help some serving use cases. - It is also generally recommended to set the global_name field of this proto, to avoid creating multiple large pools. It is typically better to run the non-low-priority work, even across sessions, in a single large pool.
repeated .tensorflow.ThreadPoolOptionProto session_inter_op_thread_pool = 12; مجموعة ConfigProto.Builder العامة ShareClusterDevicesInSession (قيمة منطقية)
When true, WorkerSessions are created with device attributes from the full cluster. This is helpful when a worker wants to partition a graph (for example during a PartitionedCallOp).
bool share_cluster_devices_in_session = 17; النهائي العام ConfigProto.Builder setUnknownFields (com.google.protobuf.UnknownFieldSetUnknownFields)
مجموعة ConfigProto.Builder العامة UsePerSessionThreads (قيمة منطقية)
If true, use a new set of threads for this session rather than the global pool of threads. Only supported by direct sessions. If false, use the global threads created by the first session, or the per-session thread pools configured by session_inter_op_thread_pool. This option is deprecated. The same effect can be achieved by setting session_inter_op_thread_pool to have one element, whose num_threads equals inter_op_parallelism_threads.
bool use_per_session_threads = 9;