ConfigProto.Builder คลาสสุดท้ายแบบคงที่สาธารณะ
Session configuration parameters. The system picks appropriate values for fields that are not set.Protobuf ประเภท
tensorflow.ConfigProto วิธีการสาธารณะ
| ConfigProto.Builder | addAllDeviceFilters (ค่า Iterable<String>) When any filters are present sessions will ignore all devices which do not match the filters. |
| ConfigProto.Builder | addAllSessionInterOpThreadPool (Iterable<? ขยาย ThreadPoolOptionProto > ค่า) This option is experimental - it may be replaced with a different mechanism in the future. |
| ConfigProto.Builder | addDeviceFilters (ค่าสตริง) When any filters are present sessions will ignore all devices which do not match the filters. |
| ConfigProto.Builder | addDeviceFiltersBytes (ค่า com.google.protobuf.ByteString) When any filters are present sessions will ignore all devices which do not match the filters. |
| ConfigProto.Builder | addRepeatedField (com.google.protobuf.Descriptors.FieldDescriptor ช่อง ค่าอ็อบเจ็กต์) |
| ConfigProto.Builder | addSessionInterOpThreadPool (ค่า ThreadPoolOptionProto ) This option is experimental - it may be replaced with a different mechanism in the future. |
| ConfigProto.Builder | addSessionInterOpThreadPool (ดัชนี int, ค่า 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 (ดัชนี int, ThreadPoolOptionProto.Builder builderForValue) This option is experimental - it may be replaced with a different mechanism in the future. |
| ThreadPoolOptionProto.Builder | addSessionInterOpThreadPoolBuilder (ดัชนี int) 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 | สร้าง () |
| ConfigProto | สร้างบางส่วน () |
| ConfigProto.Builder | ชัดเจน () |
| ConfigProto.Builder | clearAllowSoftตำแหน่ง () Whether soft placement is allowed. |
| ConfigProto.Builder | clearClusterDef () Optional list of all workers to use in this session. |
| ConfigProto.Builder | |
| ConfigProto.Builder | ล้างตัวกรองอุปกรณ์ () When any filters are present sessions will ignore all devices which do not match the filters. |
| ConfigProto.Builder | ชัดเจนการทดลอง () .tensorflow.ConfigProto.Experimental experimental = 16; |
| ConfigProto.Builder | clearField (ฟิลด์ com.google.protobuf.Descriptors.FieldDescriptor) |
| ConfigProto.Builder | clearGpuOptions () Options that apply to all GPUs. |
| ConfigProto.Builder | clearGraphOptions () 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 | ล้าง IsolateSessionState () 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 | ชัดเจนตำแหน่งระยะเวลา () 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 | clearRpcOptions () 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 | โคลน () |
| บูลีน | containsDeviceCount (คีย์สตริง) Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. |
| บูลีน | getAllowSoftPlacement () Whether soft placement is allowed. |
| คลัสเตอร์ดีฟ | getClusterDef () Optional list of all workers to use in this session. |
| ClusterDef.Builder | getClusterDefBuilder () Optional list of all workers to use in this session. |
| ClusterDefOrBuilder | getClusterDefOrBuilder () Optional list of all workers to use in this session. |
| ConfigProto | |
| com.google.protobuf.Descriptors.Descriptor แบบคงที่ขั้นสุดท้าย | รับคำอธิบาย () |
| com.google.protobuf.Descriptors.Descriptor | |
| แมป<สตริง จำนวนเต็ม> | รับอุปกรณ์นับ () ใช้ getDeviceCountMap() แทน |
| ภายใน | getDeviceCountCount () Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. |
| แมป<สตริง จำนวนเต็ม> | getDeviceCountMap () Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. |
| ภายใน | getDeviceCountOrDefault (คีย์สตริง, int defaultValue) Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. |
| ภายใน | getDeviceCountOrThrow (คีย์สตริง) Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. |
| สตริง | getDeviceFilters (ดัชนี int) When any filters are present sessions will ignore all devices which do not match the filters. |
| com.google.protobuf.ByteString | getDeviceFiltersBytes (ดัชนี int) When any filters are present sessions will ignore all devices which do not match the filters. |
| ภายใน | 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 ทดลอง | รับการทดลอง () .tensorflow.ConfigProto.Experimental experimental = 16; |
| ConfigProto.Experimental.Builder | รับExperimentalBuilder () .tensorflow.ConfigProto.Experimental experimental = 16; |
| ConfigProto.ExperimentalOrBuilder | รับExperimentalOrBuilder () .tensorflow.ConfigProto.Experimental experimental = 16; |
| ตัวเลือก GPU | getGpuOptions () Options that apply to all GPUs. |
| GPUOptions.ตัวสร้าง | getGpuOptionsBuilder () Options that apply to all GPUs. |
| GPUOptionsOrBuilder | getGpuOptionsOrBuilder () Options that apply to all GPUs. |
| ตัวเลือกกราฟ | getGraphOptions () Options that apply to all graphs. |
| GraphOptions.ตัวสร้าง | getGraphOptionsBuilder () Options that apply to all graphs. |
| GraphOptionsOrBuilder | getGraphOptionsOrBuilder () Options that apply to all graphs. |
| ภายใน | getInterOpParallelismThreads () Nodes that perform blocking operations are enqueued on a pool of inter_op_parallelism_threads available in each process. |
| ภายใน | getIntraOpParallelismThreads () The execution of an individual op (for some op types) can be parallelized on a pool of intra_op_parallelism_threads. |
| บูลีน | getIsolateSessionState () If true, any resources such as Variables used in the session will not be shared with other sessions. |
| บูลีน | getLogDevicePlacement () Whether device placements should be logged. |
| แมป<สตริง จำนวนเต็ม> | getMutableDeviceCount () ใช้ตัวเข้าถึงการกลายพันธุ์สำรองแทน |
| ยาว | getOperationTimeoutInMs () Global timeout for all blocking operations in this session. |
| ภายใน | 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). |
| RPCตัวเลือก | getRpcOptions () Options that apply when this session uses the distributed runtime. |
| RPCOptions.ตัวสร้าง | getRpcOptionsBuilder () Options that apply when this session uses the distributed runtime. |
| RPCOptionsOrBuilder | getRpcOptionsOrBuilder () Options that apply when this session uses the distributed runtime. |
| ThreadPoolOptionProto | getSessionInterOpThreadPool (ดัชนี int) This option is experimental - it may be replaced with a different mechanism in the future. |
| ThreadPoolOptionProto.Builder | getSessionInterOpThreadPoolBuilder (ดัชนี int) This option is experimental - it may be replaced with a different mechanism in the future. |
| รายการ < ThreadPoolOptionProto.Builder > | getSessionInterOpThreadPoolBuilderList () This option is experimental - it may be replaced with a different mechanism in the future. |
| ภายใน | getSessionInterOpThreadPoolCount () This option is experimental - it may be replaced with a different mechanism in the future. |
| รายการ < ThreadPoolOptionProto > | getSessionInterOpThreadPoolList () This option is experimental - it may be replaced with a different mechanism in the future. |
| ThreadPoolOptionProtoOrBuilder | getSessionInterOpThreadPoolOrBuilder (ดัชนี int) This option is experimental - it may be replaced with a different mechanism in the future. |
| รายการ<? ขยาย ThreadPoolOptionProtoOrBuilder > | getSessionInterOpThreadPoolOrBuilderList () This option is experimental - it may be replaced with a different mechanism in the future. |
| บูลีน | getShareClusterDevicesInSession () When true, WorkerSessions are created with device attributes from the full cluster. |
| บูลีน | getUsePerSessionThreads () If true, use a new set of threads for this session rather than the global pool of threads. |
| บูลีน | hasClusterDef () Optional list of all workers to use in this session. |
| บูลีน | มีการทดลอง () .tensorflow.ConfigProto.Experimental experimental = 16; |
| บูลีน | hasGpuOptions () Options that apply to all GPUs. |
| บูลีน | hasGraphOptions () Options that apply to all graphs. |
| บูลีน | hasRpcOptions () Options that apply when this session uses the distributed runtime. |
| บูลีนสุดท้าย | |
| ConfigProto.Builder | |
| ConfigProto.Builder | |
| ConfigProto.Builder | mergeFrom (com.google.protobuf.ข้อความ อื่น ๆ ) |
| ConfigProto.Builder | mergeFrom (com.google.protobuf.CodedInputStream อินพุต com.google.protobuf.ExtensionRegistryLite extensionRegistry) |
| ConfigProto.Builder | |
| ConfigProto.Builder | |
| ConfigProto.Builder | mergeRpcOptions (ค่า RPCOptions ) Options that apply when this session uses the distributed runtime. |
| ConfigProto.Builder สุดท้าย | mergeUnknownFields (com.google.protobuf.UnknownFieldSetknownFields) |
| ConfigProto.Builder | putAllDeviceCount (ค่าแมป <สตริง, จำนวนเต็ม>) Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. |
| ConfigProto.Builder | putDeviceCount (คีย์สตริง, ค่า int) Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. |
| ConfigProto.Builder | RemoveDeviceCount (คีย์สตริง) Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. |
| ConfigProto.Builder | RemoveSessionInterOpThreadPool (ดัชนี int) This option is experimental - it may be replaced with a different mechanism in the future. |
| ConfigProto.Builder | setAllowSoftPlacement (ค่าบูลีน) 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 (ดัชนี int, ค่าสตริง) When any filters are present sessions will ignore all devices which do not match the filters. |
| ConfigProto.Builder | |
| ConfigProto.Builder | setExperimental ( ConfigProto.Experimental.Builder builderForValue) .tensorflow.ConfigProto.Experimental experimental = 16; |
| ConfigProto.Builder | setField (ฟิลด์ com.google.protobuf.Descriptors.FieldDescriptor ค่าอ็อบเจ็กต์) |
| ConfigProto.Builder | |
| ConfigProto.Builder | |
| ConfigProto.Builder | |
| ConfigProto.Builder | |
| ConfigProto.Builder | setInterOpParallelismThreads (ค่า int) Nodes that perform blocking operations are enqueued on a pool of inter_op_parallelism_threads available in each process. |
| ConfigProto.Builder | setIntraOpParallelismThreads (ค่า 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 (ค่าบูลีน) If true, any resources such as Variables used in the session will not be shared with other sessions. |
| ConfigProto.Builder | setLogDevicePlacement (ค่าบูลีน) Whether device placements should be logged. |
| ConfigProto.Builder | setOperationTimeoutInMs (ค่ายาว) Global timeout for all blocking operations in this session. |
| ConfigProto.Builder | setPlacementPeriod (ค่า 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 (ฟิลด์ com.google.protobuf.Descriptors.FieldDescriptor, ดัชนี int, ค่าอ็อบเจ็กต์) |
| ConfigProto.Builder | |
| ConfigProto.Builder | setRpcOptions ( RPCOptions.Builder builderForValue) Options that apply when this session uses the distributed runtime. |
| ConfigProto.Builder | setSessionInterOpThreadPool (ดัชนี int, ค่า ThreadPoolOptionProto ) This option is experimental - it may be replaced with a different mechanism in the future. |
| ConfigProto.Builder | setSessionInterOpThreadPool (ดัชนี int, ThreadPoolOptionProto.Builder builderForValue) This option is experimental - it may be replaced with a different mechanism in the future. |
| ConfigProto.Builder | setShareClusterDevicesInSession (ค่าบูลีน) When true, WorkerSessions are created with device attributes from the full cluster. |
| ConfigProto.Builder สุดท้าย | setUnknownFields (com.google.protobuf.UnknownFieldSetknownFields) |
| ConfigProto.Builder | setUsePerSessionThreads (ค่าบูลีน) If true, use a new set of threads for this session rather than the global pool of threads. |
วิธีการสืบทอด
วิธีการสาธารณะ
ConfigProto.Builder สาธารณะ addAllDeviceFilters (ค่า 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; ConfigProto.Builder สาธารณะ addAllSessionInterOpThreadPool (Iterable <? ขยาย ThreadPoolOptionProto > ค่า)
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 สาธารณะ addDeviceFilters (ค่าสตริง)
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 สาธารณะ addDeviceFiltersBytes (ค่า 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; ConfigProto.Builder addRepeatedField สาธารณะ (ฟิลด์ com.google.protobuf.Descriptors.FieldDescriptor ค่าอ็อบเจ็กต์)
ConfigProto.Builder สาธารณะ addSessionInterOpThreadPool (ค่า 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; ConfigProto.Builder สาธารณะ addSessionInterOpThreadPool (ดัชนี int, ค่า 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; ConfigProto.Builder สาธารณะ 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; ConfigProto.Builder สาธารณะ addSessionInterOpThreadPool (ดัชนี 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 สาธารณะ addSessionInterOpThreadPoolBuilder (ดัชนี 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 สาธารณะ 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 สาธารณะ 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 สาธารณะ clearClusterDef ()
Optional list of all workers to use in this session.
.tensorflow.ClusterDef cluster_def = 14; ConfigProto.Builder สาธารณะ 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 สาธารณะ clearExperimental ()
.tensorflow.ConfigProto.Experimental experimental = 16;
ConfigProto.Builder สาธารณะ clearGpuOptions ()
Options that apply to all GPUs.
.tensorflow.GPUOptions gpu_options = 6; ConfigProto.Builder สาธารณะ clearGraphOptions ()
Options that apply to all graphs.
.tensorflow.GraphOptions graph_options = 10; ConfigProto.Builder สาธารณะ 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 สาธารณะ 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 สาธารณะ 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 สาธารณะ clearLogDevicePlacement ()
Whether device placements should be logged.
bool log_device_placement = 8; ConfigProto.Builder สาธารณะ 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 สาธารณะ 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 สาธารณะ clearRpcOptions ()
Options that apply when this session uses the distributed runtime.
.tensorflow.RPCOptions rpc_options = 13; ConfigProto.Builder สาธารณะ 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 สาธารณะ 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 สาธารณะ 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; บูลีนสาธารณะ ประกอบด้วยDeviceCount (คีย์สตริง)
แผนที่
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; บูลีนสาธารณะ getAllowSoftPlacement ()
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 สาธารณะ getClusterDef ()
Optional list of all workers to use in this session.
.tensorflow.ClusterDef cluster_def = 14; ClusterDef.Builder สาธารณะ getClusterDefBuilder ()
Optional list of all workers to use in this session.
.tensorflow.ClusterDef cluster_def = 14; ClusterDefOrBuilder สาธารณะ getClusterDefOrBuilder ()
Optional list of all workers to use in this session.
.tensorflow.ClusterDef cluster_def = 14; สาธารณะคงที่สุดท้าย com.google.protobuf.Descriptors.Descriptor getDescriptor ()
สาธารณะ com.google.protobuf.Descriptors.Descriptor getDescriptorForType ()
สาธารณะ int getDeviceCountCount ()
แผนที่
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; แผนที่สาธารณะ <String, Integer> getDeviceCountMap ()
แผนที่
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; สาธารณะ int getDeviceCountOrDefault (คีย์สตริง, int defaultValue)
แผนที่
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 สาธารณะ getDeviceCountOrThrow (คีย์สตริง)
แผนที่
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; สตริงสาธารณะ getDeviceFilters (ดัชนี 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; สาธารณะ com.google.protobuf.ByteString getDeviceFiltersBytes (ดัชนี 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; สาธารณะ int 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; สาธารณะ com.google.protobuf.ProtocolStringList 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 สาธารณะ ทดลอง getExperimental ()
.tensorflow.ConfigProto.Experimental experimental = 16;
ConfigProto.Experimental.Builder สาธารณะ getExperimentalBuilder ()
.tensorflow.ConfigProto.Experimental experimental = 16;
ConfigProto.ExperimentalOrBuilder สาธารณะ getExperimentalOrBuilder ()
.tensorflow.ConfigProto.Experimental experimental = 16;
GPUOptions สาธารณะ getGpuOptions ()
Options that apply to all GPUs.
.tensorflow.GPUOptions gpu_options = 6; GPUOptions.Builder สาธารณะ getGpuOptionsBuilder ()
Options that apply to all GPUs.
.tensorflow.GPUOptions gpu_options = 6; GPUOptionsOrBuilder สาธารณะ getGpuOptionsOrBuilder ()
Options that apply to all GPUs.
.tensorflow.GPUOptions gpu_options = 6; GraphOptions สาธารณะ getGraphOptions ()
Options that apply to all graphs.
.tensorflow.GraphOptions graph_options = 10; GraphOptions.Builder สาธารณะ getGraphOptionsBuilder ()
Options that apply to all graphs.
.tensorflow.GraphOptions graph_options = 10; GraphOptionsOrBuilder สาธารณะ getGraphOptionsOrBuilder ()
Options that apply to all graphs.
.tensorflow.GraphOptions graph_options = 10; int สาธารณะ 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 สาธารณะ 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; บูลีนสาธารณะ getIsolateSessionState ()
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; บูลีนสาธารณะ getLogDevicePlacement ()
Whether device placements should be logged.
bool log_device_placement = 8; แผนที่สาธารณะ <String, Integer> getMutableDeviceCount ()
ใช้ตัวเข้าถึงการกลายพันธุ์สำรองแทน
สาธารณะ getOperationTimeoutInMs ยาว ()
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; สาธารณะ int 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 สาธารณะ getRpcOptions ()
Options that apply when this session uses the distributed runtime.
.tensorflow.RPCOptions rpc_options = 13; RPCOptions.Builder สาธารณะ getRpcOptionsBuilder ()
Options that apply when this session uses the distributed runtime.
.tensorflow.RPCOptions rpc_options = 13; RPCOptionsOrBuilder สาธารณะ getRpcOptionsOrBuilder ()
Options that apply when this session uses the distributed runtime.
.tensorflow.RPCOptions rpc_options = 13; ThreadPoolOptionPro สาธารณะ เพื่อรับSessionInterOpThreadPool (ดัชนี 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 สาธารณะ getSessionInterOpThreadPoolBuilder (ดัชนี 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; รายการสาธารณะ < 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 สาธารณะ 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; รายการสาธารณะ < 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 สาธารณะ getSessionInterOpThreadPoolOrBuilder (ดัชนี 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; รายการสาธารณะ<? ขยาย 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; บูลีนสาธารณะ getShareClusterDevicesInSession ()
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; บูลีนสาธารณะ getUsePerSessionThreads ()
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; hasClusterDef บูลีนสาธารณะ ()
Optional list of all workers to use in this session.
.tensorflow.ClusterDef cluster_def = 14; บูลีนสาธารณะ มีการทดลอง ()
.tensorflow.ConfigProto.Experimental experimental = 16;
hasGpuOptions บูลีนสาธารณะ ()
Options that apply to all GPUs.
.tensorflow.GPUOptions gpu_options = 6; hasGraphOptions บูลีนสาธารณะ ()
Options that apply to all graphs.
.tensorflow.GraphOptions graph_options = 10; บูลีนสาธารณะ hasRpcOptions ()
Options that apply when this session uses the distributed runtime.
.tensorflow.RPCOptions rpc_options = 13; บูลีนสุดท้ายสาธารณะ isInitialized ()
ConfigProto.Builder สาธารณะ mergeClusterDef (ค่า ClusterDef )
Optional list of all workers to use in this session.
.tensorflow.ClusterDef cluster_def = 14; ConfigProto.Builder สาธารณะ ผสาน Experimental ( ConfigProto.Experimental มูลค่า)
.tensorflow.ConfigProto.Experimental experimental = 16;
ConfigProto.Builder สาธารณะ ผสานจาก (com.google.protobuf.CodedInputStream อินพุต com.google.protobuf.ExtensionRegistryLite extensionRegistry)
ขว้าง
| IOข้อยกเว้น |
|---|
ConfigProto.Builder สาธารณะ ผสาน GpuOptions (ค่า GPUOptions )
Options that apply to all GPUs.
.tensorflow.GPUOptions gpu_options = 6; ConfigProto.Builder สาธารณะ mergeGraphOptions (ค่า GraphOptions )
Options that apply to all graphs.
.tensorflow.GraphOptions graph_options = 10; ConfigProto.Builder สาธารณะ ผสาน RpcOptions (ค่า RPCOptions )
Options that apply when this session uses the distributed runtime.
.tensorflow.RPCOptions rpc_options = 13; ConfigProto.Builder สาธารณะขั้นสุดท้าย ผสาน UnknownFields (com.google.protobuf.UnknownFieldSetknownFields)
ConfigProto.Builder สาธารณะ putAllDeviceCount (ค่าแมป <สตริง, จำนวนเต็ม>)
แผนที่
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 สาธารณะ putDeviceCount (คีย์สตริง, ค่า int)
แผนที่
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 สาธารณะ RemoveDeviceCount (คีย์สตริง)
แผนที่
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 สาธารณะ RemoveSessionInterOpThreadPool (ดัชนี 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; ConfigProto.Builder สาธารณะ setAllowSoftPlacement (ค่าบูลีน)
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 setClusterDef สาธารณะ ( ClusterDef.Builder builderForValue)
Optional list of all workers to use in this session.
.tensorflow.ClusterDef cluster_def = 14; ConfigProto.Builder setClusterDef สาธารณะ (ค่า ClusterDef )
Optional list of all workers to use in this session.
.tensorflow.ClusterDef cluster_def = 14;ConfigProto.Builder สาธารณะ setDeviceFilters (ดัชนี int, ค่าสตริง)
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 setExperimental สาธารณะ ( ConfigProto.Experimental ค่า)
.tensorflow.ConfigProto.Experimental experimental = 16;
ConfigProto.Builder สาธารณะ setExperimental ( ConfigProto.Experimental.Builder builderForValue)
.tensorflow.ConfigProto.Experimental experimental = 16;
ConfigProto.Builder setField สาธารณะ (ฟิลด์ com.google.protobuf.Descriptors.FieldDescriptor ค่าอ็อบเจ็กต์)
ConfigProto.Builder setGpuOptions สาธารณะ ( GPUOptions.Builder builderForValue)
Options that apply to all GPUs.
.tensorflow.GPUOptions gpu_options = 6; ConfigProto.Builder setGpuOptions สาธารณะ (ค่า GPUOptions )
Options that apply to all GPUs.
.tensorflow.GPUOptions gpu_options = 6; ConfigProto.Builder setGraphOptions สาธารณะ ( GraphOptions.Builder builderForValue)
Options that apply to all graphs.
.tensorflow.GraphOptions graph_options = 10; ConfigProto.Builder setGraphOptions สาธารณะ (ค่า GraphOptions )
Options that apply to all graphs.
.tensorflow.GraphOptions graph_options = 10; ConfigProto.Builder สาธารณะ setInterOpParallelismThreads (ค่า 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; ConfigProto.Builder สาธารณะ setIntraOpParallelismThreads (ค่า 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; ConfigProto.Builder สาธารณะ setIsolateSessionState (ค่าบูลีน)
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 สาธารณะ setLogDevicePlacement (ค่าบูลีน)
Whether device placements should be logged.
bool log_device_placement = 8; ConfigProto.Builder สาธารณะ setOperationTimeoutInMs (ค่ายาว)
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 สาธารณะ setPlacementPeriod (ค่า 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 setRepeatedField สาธารณะ (ฟิลด์ com.google.protobuf.Descriptors.FieldDescriptor, ดัชนี int, ค่าอ็อบเจ็กต์)
ConfigProto.Builder setRpcOptions สาธารณะ (ค่า RPCOptions )
Options that apply when this session uses the distributed runtime.
.tensorflow.RPCOptions rpc_options = 13; ConfigProto.Builder setRpcOptions สาธารณะ ( RPCOptions.Builder builderForValue)
Options that apply when this session uses the distributed runtime.
.tensorflow.RPCOptions rpc_options = 13; ConfigProto.Builder สาธารณะ setSessionInterOpThreadPool (ดัชนี int, ค่า 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; ConfigProto.Builder สาธารณะ setSessionInterOpThreadPool (ดัชนี 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; ConfigProto.Builder สาธารณะ setShareClusterDevicesInSession (ค่าบูลีน)
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 สุดท้ายสาธารณะ setUnknownFields (com.google.protobuf.UnknownFieldSetknownFields)
ConfigProto.Builder สาธารณะ setUsePerSessionThreads (ค่าบูลีน)
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;