public interface
ConfigProtoOrBuilder
Known Indirect Subclasses |
Public Methods
abstract boolean |
containsDeviceCount(String key)
Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. |
abstract boolean |
getAllowSoftPlacement()
Whether soft placement is allowed. |
abstract ClusterDef |
getClusterDef()
Optional list of all workers to use in this session. |
abstract ClusterDefOrBuilder |
getClusterDefOrBuilder()
Optional list of all workers to use in this session. |
abstract Map<String, Integer> |
getDeviceCount()
Use
getDeviceCountMap() instead. |
abstract int |
getDeviceCountCount()
Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. |
abstract Map<String, Integer> |
getDeviceCountMap()
Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. |
abstract int |
getDeviceCountOrDefault(String key, int defaultValue)
Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. |
abstract int |
getDeviceCountOrThrow(String key)
Map from device type name (e.g., "CPU" or "GPU" ) to maximum number of devices of that type to use. |
abstract String |
getDeviceFilters(int index)
When any filters are present sessions will ignore all devices which do not match the filters. |
abstract com.google.protobuf.ByteString |
getDeviceFiltersBytes(int index)
When any filters are present sessions will ignore all devices which do not match the filters. |
abstract int |
getDeviceFiltersCount()
When any filters are present sessions will ignore all devices which do not match the filters. |
abstract List<String> |
getDeviceFiltersList()
When any filters are present sessions will ignore all devices which do not match the filters. |
abstract ConfigProto.Experimental |
getExperimental()
.tensorflow.ConfigProto.Experimental experimental = 16;
|
abstract ConfigProto.ExperimentalOrBuilder |
getExperimentalOrBuilder()
.tensorflow.ConfigProto.Experimental experimental = 16;
|
abstract GPUOptions |
getGpuOptions()
Options that apply to all GPUs. |
abstract GPUOptionsOrBuilder |
getGpuOptionsOrBuilder()
Options that apply to all GPUs. |
abstract GraphOptions |
getGraphOptions()
Options that apply to all graphs. |
abstract GraphOptionsOrBuilder |
getGraphOptionsOrBuilder()
Options that apply to all graphs. |
abstract int |
getInterOpParallelismThreads()
Nodes that perform blocking operations are enqueued on a pool of inter_op_parallelism_threads available in each process. |
abstract int |
getIntraOpParallelismThreads()
The execution of an individual op (for some op types) can be parallelized on a pool of intra_op_parallelism_threads. |
abstract boolean |
getIsolateSessionState()
If true, any resources such as Variables used in the session will not be shared with other sessions. |
abstract boolean |
getLogDevicePlacement()
Whether device placements should be logged. |
abstract long |
getOperationTimeoutInMs()
Global timeout for all blocking operations in this session. |
abstract 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). |
abstract RPCOptions |
getRpcOptions()
Options that apply when this session uses the distributed runtime. |
abstract RPCOptionsOrBuilder |
getRpcOptionsOrBuilder()
Options that apply when this session uses the distributed runtime. |
abstract ThreadPoolOptionProto |
getSessionInterOpThreadPool(int index)
This option is experimental - it may be replaced with a different mechanism in the future. |
abstract int |
getSessionInterOpThreadPoolCount()
This option is experimental - it may be replaced with a different mechanism in the future. |
abstract List<ThreadPoolOptionProto> |
getSessionInterOpThreadPoolList()
This option is experimental - it may be replaced with a different mechanism in the future. |
abstract ThreadPoolOptionProtoOrBuilder |
getSessionInterOpThreadPoolOrBuilder(int index)
This option is experimental - it may be replaced with a different mechanism in the future. |
abstract List<? extends ThreadPoolOptionProtoOrBuilder> |
getSessionInterOpThreadPoolOrBuilderList()
This option is experimental - it may be replaced with a different mechanism in the future. |
abstract boolean |
getShareClusterDevicesInSession()
When true, WorkerSessions are created with device attributes from the full cluster. |
abstract boolean |
getUsePerSessionThreads()
If true, use a new set of threads for this session rather than the global pool of threads. |
abstract boolean |
hasClusterDef()
Optional list of all workers to use in this session. |
abstract boolean |
hasExperimental()
.tensorflow.ConfigProto.Experimental experimental = 16;
|
abstract boolean |
hasGpuOptions()
Options that apply to all GPUs. |
abstract boolean |
hasGraphOptions()
Options that apply to all graphs. |
abstract boolean |
hasRpcOptions()
Options that apply when this session uses the distributed runtime. |
Public Methods
public abstract boolean containsDeviceCount (String key)
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;
public abstract boolean 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;
public abstract ClusterDef getClusterDef ()
Optional list of all workers to use in this session.
.tensorflow.ClusterDef cluster_def = 14;
public abstract ClusterDefOrBuilder getClusterDefOrBuilder ()
Optional list of all workers to use in this session.
.tensorflow.ClusterDef cluster_def = 14;
public abstract int 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;
public abstract Map<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;
public abstract int getDeviceCountOrDefault (String key, 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;
public abstract int getDeviceCountOrThrow (String key)
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;
public abstract String getDeviceFilters (int index)
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 abstract com.google.protobuf.ByteString getDeviceFiltersBytes (int index)
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 abstract 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;
public abstract List<String> 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;
public abstract ConfigProto.Experimental getExperimental ()
.tensorflow.ConfigProto.Experimental experimental = 16;
public abstract ConfigProto.ExperimentalOrBuilder getExperimentalOrBuilder ()
.tensorflow.ConfigProto.Experimental experimental = 16;
public abstract GPUOptions getGpuOptions ()
Options that apply to all GPUs.
.tensorflow.GPUOptions gpu_options = 6;
public abstract GPUOptionsOrBuilder getGpuOptionsOrBuilder ()
Options that apply to all GPUs.
.tensorflow.GPUOptions gpu_options = 6;
public abstract GraphOptions getGraphOptions ()
Options that apply to all graphs.
.tensorflow.GraphOptions graph_options = 10;
public abstract GraphOptionsOrBuilder getGraphOptionsOrBuilder ()
Options that apply to all graphs.
.tensorflow.GraphOptions graph_options = 10;
public abstract int 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;
public abstract int 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;
public abstract boolean 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;
public abstract boolean getLogDevicePlacement ()
Whether device placements should be logged.
bool log_device_placement = 8;
public abstract long getOperationTimeoutInMs ()
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;
public abstract 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;
public abstract RPCOptions getRpcOptions ()
Options that apply when this session uses the distributed runtime.
.tensorflow.RPCOptions rpc_options = 13;
public abstract RPCOptionsOrBuilder getRpcOptionsOrBuilder ()
Options that apply when this session uses the distributed runtime.
.tensorflow.RPCOptions rpc_options = 13;
public abstract ThreadPoolOptionProto getSessionInterOpThreadPool (int index)
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 abstract int 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;
public abstract List<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;
public abstract ThreadPoolOptionProtoOrBuilder getSessionInterOpThreadPoolOrBuilder (int index)
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 abstract List<? extends 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;
public abstract boolean 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;
public abstract boolean 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;
public abstract boolean hasClusterDef ()
Optional list of all workers to use in this session.
.tensorflow.ClusterDef cluster_def = 14;
public abstract boolean hasExperimental ()
.tensorflow.ConfigProto.Experimental experimental = 16;
public abstract boolean hasGpuOptions ()
Options that apply to all GPUs.
.tensorflow.GPUOptions gpu_options = 6;
public abstract boolean hasGraphOptions ()
Options that apply to all graphs.
.tensorflow.GraphOptions graph_options = 10;
public abstract boolean hasRpcOptions ()
Options that apply when this session uses the distributed runtime.
.tensorflow.RPCOptions rpc_options = 13;