GPUOptions.Experimental

публичный статический конечный класс GPUOptions.Experimental

Тип protobuf tensorflow.GPUOptions.Experimental

Вложенные классы

сорт	GPUOptions.Experimental.Builder		Тип protobuf tensorflow.GPUOptions.Experimental
сорт	GPUOptions.Experimental.VirtualDevices		Configuration for breaking down a visible GPU into multiple "virtual" devices.
интерфейс	GPUOptions.Experimental.VirtualDevicesOrBuilder

Константы

интервал	COLLECTIVE_RING_ORDER_FIELD_NUMBER
интервал	KERNEL_TRACKER_MAX_BYTES_FIELD_NUMBER
интервал	KERNEL_TRACKER_MAX_INTERVAL_FIELD_NUMBER
интервал	KERNEL_TRACKER_MAX_PENDING_FIELD_NUMBER
интервал	NUM_DEV_TO_DEV_COPY_STREAMS_FIELD_NUMBER
интервал	TIMESTAMPED_ALLOCATOR_FIELD_NUMBER
интервал	USE_UNIFIED_MEMORY_FIELD_NUMBER
интервал	VIRTUAL_DEVICES_FIELD_NUMBER

Публичные методы

логическое значение	равно (Объект obj)
Нить	getCollectiveRingOrder () If non-empty, defines a good GPU ring order on a single worker based on device interconnect.
com.google.protobuf.ByteString	getCollectiveRingOrderBytes () If non-empty, defines a good GPU ring order on a single worker based on device interconnect.
статический GPUOptions.Экспериментальный	getDefaultInstance ()
GPUOptions.Экспериментальный	getDefaultInstanceForType ()
окончательный статический com.google.protobuf.Descriptors.Descriptor	получитьдескриптор ()
интервал	getKernelTrackerMaxBytes () If kernel_tracker_max_bytes = n > 0, then a tracking event is inserted after every series of kernels allocating a sum of memory >= n.
интервал	getKernelTrackerMaxInterval () Parameters for GPUKernelTracker.
интервал	getKernelTrackerMaxPending () If kernel_tracker_max_pending > 0 then no more than this many tracking events can be outstanding at a time.
интервал	getNumDevToDevCopyStreams () If > 1, the number of device-to-device copy streams to create for each GPUDevice.
	getParserForType ()
интервал	getSerializedSize ()
логическое значение	getTimestampedAllocator () If true then extra work is done by GPUDevice and GPUBFCAllocator to keep track of when GPU memory is freed and when kernels actually complete so that we can know when a nominally free memory chunk is really not subject to pending use.
окончательный com.google.protobuf.UnknownFieldSet	getUnknownFields ()
логическое значение	getUseUnifiedMemory () If true, uses CUDA unified memory for memory allocations.
GPUOptions.Experimental.VirtualDevices	getVirtualDevices (индекс int) The multi virtual device settings.
интервал	getVirtualDevicesCount () The multi virtual device settings.
Список < GPUOptions.Experimental.VirtualDevices >	getVirtualDevicesList () The multi virtual device settings.
GPUOptions.Experimental.VirtualDevicesOrBuilder	getVirtualDevicesOrBuilder (индекс int) The multi virtual device settings.
Список<? расширяет GPUOptions.Experimental.VirtualDevicesOrBuilder >	getVirtualDevicesOrBuilderList () The multi virtual device settings.
интервал	хэш-код ()
последнее логическое значение	isInitialized ()
статический GPUOptions.Experimental.Builder	newBuilder ( GPUOptions.Экспериментальный прототип)
статический GPUOptions.Experimental.Builder	новыйстроитель ()
GPUOptions.Experimental.Builder	новыйБилдерФорТип ()
статический GPUOptions.Экспериментальный	parseDelimitedFrom (входной поток)
статический GPUOptions.Экспериментальный	parseDelimitedFrom (ввод InputStream, com.google.protobuf.ExtensionRegistryLite ExtensionRegistry)
статический GPUOptions.Экспериментальный	parseFrom (данные ByteBuffer)
статический GPUOptions.Экспериментальный	parseFrom (ввод com.google.protobuf.CodedInputStream, com.google.protobuf.ExtensionRegistryLite ExtensionRegistry)
статический GPUOptions.Экспериментальный	parseFrom (данные ByteBuffer, com.google.protobuf.ExtensionRegistryLite ExtensionRegistry)
статический GPUOptions.Экспериментальный	parseFrom (вход com.google.protobuf.CodedInputStream)
статический GPUOptions.Экспериментальный	parseFrom (данные byte[], com.google.protobuf.ExtensionRegistryLite ExtensionRegistry)
статический GPUOptions.Экспериментальный	parseFrom (данные com.google.protobuf.ByteString)
статический GPUOptions.Экспериментальный	parseFrom (ввод входного потока, com.google.protobuf.ExtensionRegistryLite ExtensionRegistry)
статический GPUOptions.Экспериментальный	parseFrom (данные com.google.protobuf.ByteString, com.google.protobuf.ExtensionRegistryLite ExtensionRegistry)
статический	парсер ()
GPUOptions.Experimental.Builder	toBuilder ()
пустота	writeTo (вывод com.google.protobuf.CodedOutputStream)

Унаследованные методы

Из класса java.lang.Object

логическое значение	равно (Объект arg0)
последний класс<?>	получитьКласс ()
интервал	хэш-код ()
окончательная пустота	уведомить ()
окончательная пустота	уведомитьВсе ()
Нить	toString ()
окончательная пустота	подождать (длинный arg0, int arg1)
окончательная пустота	подождите (длинный arg0)
окончательная пустота	ждать ()

Из интерфейса org.tensorflow.proto.framework.GPUOptions.ExperimentalOrBuilder

абстрактная строка	getCollectiveRingOrder () If non-empty, defines a good GPU ring order on a single worker based on device interconnect.
абстрактный com.google.protobuf.ByteString	getCollectiveRingOrderBytes () If non-empty, defines a good GPU ring order on a single worker based on device interconnect.
абстрактный int	getKernelTrackerMaxBytes () If kernel_tracker_max_bytes = n > 0, then a tracking event is inserted after every series of kernels allocating a sum of memory >= n.
абстрактный int	getKernelTrackerMaxInterval () Parameters for GPUKernelTracker.
абстрактный int	getKernelTrackerMaxPending () If kernel_tracker_max_pending > 0 then no more than this many tracking events can be outstanding at a time.
абстрактный int	getNumDevToDevCopyStreams () If > 1, the number of device-to-device copy streams to create for each GPUDevice.
абстрактное логическое значение	getTimestampedAllocator () If true then extra work is done by GPUDevice and GPUBFCAllocator to keep track of when GPU memory is freed and when kernels actually complete so that we can know when a nominally free memory chunk is really not subject to pending use.
абстрактное логическое значение	getUseUnifiedMemory () If true, uses CUDA unified memory for memory allocations.
абстрактный GPUOptions.Experimental.VirtualDevices	getVirtualDevices (индекс int) The multi virtual device settings.
абстрактный int	getVirtualDevicesCount () The multi virtual device settings.
абстрактный список < GPUOptions.Experimental.VirtualDevices >	getVirtualDevicesList () The multi virtual device settings.
абстрактный GPUOptions.Experimental.VirtualDevicesOrBuilder	getVirtualDevicesOrBuilder (индекс int) The multi virtual device settings.
абстрактный список<? расширяет GPUOptions.Experimental.VirtualDevicesOrBuilder >	getVirtualDevicesOrBuilderList () The multi virtual device settings.

Константы

общедоступный статический окончательный int COLLECTIVE_RING_ORDER_FIELD_NUMBER

Постоянное значение: 4

общедоступный статический окончательный int KERNEL_TRACKER_MAX_BYTES_FIELD_NUMBER

Постоянное значение: 8

общедоступный статический окончательный int KERNEL_TRACKER_MAX_INTERVAL_FIELD_NUMBER

Постоянное значение: 7

общедоступный статический окончательный int KERNEL_TRACKER_MAX_PENDING_FIELD_NUMBER

Постоянное значение: 9

общедоступный статический окончательный int NUM_DEV_TO_DEV_COPY_STREAMS_FIELD_NUMBER

Постоянное значение: 3

общедоступный статический финал int TIMESTAMPED_ALLOCATOR_FIELD_NUMBER

Постоянное значение: 5

общедоступный статический окончательный int USE_UNIFIED_MEMORY_FIELD_NUMBER

Постоянное значение: 2

общедоступный статический финал int VIRTUAL_DEVICES_FIELD_NUMBER

Постоянное значение: 1

Публичные методы

общедоступное логическое значение равно (Object obj)

публичная строка getCollectiveRingOrder ()

 If non-empty, defines a good GPU ring order on a single worker based on
 device interconnect.  This assumes that all workers have the same GPU
 topology.  Specify as a comma-separated string, e.g. "3,2,1,0,7,6,5,4".
 This ring order is used by the RingReducer implementation of
 CollectiveReduce, and serves as an override to automatic ring order
 generation in OrderTaskDeviceMap() during CollectiveParam resolution.
 

string collective_ring_order = 4;

public com.google.protobuf.ByteString getCollectiveRingOrderBytes ()

 If non-empty, defines a good GPU ring order on a single worker based on
 device interconnect.  This assumes that all workers have the same GPU
 topology.  Specify as a comma-separated string, e.g. "3,2,1,0,7,6,5,4".
 This ring order is used by the RingReducer implementation of
 CollectiveReduce, and serves as an override to automatic ring order
 generation in OrderTaskDeviceMap() during CollectiveParam resolution.
 

string collective_ring_order = 4;

общедоступный статический GPUOptions.Experimental getDefaultInstance ()

общедоступный GPUOptions.Экспериментальный getDefaultInstanceForType ()

общедоступный статический окончательный com.google.protobuf.Descriptors.Descriptor getDescriptor ()

public int getKernelTrackerMaxBytes ()

 If kernel_tracker_max_bytes = n > 0, then a tracking event is
 inserted after every series of kernels allocating a sum of
 memory >= n.  If one kernel allocates b * n bytes, then one
 event will be inserted after it, but it will count as b against
 the pending limit.
 

int32 kernel_tracker_max_bytes = 8;

public int getKernelTrackerMaxInterval ()

 Parameters for GPUKernelTracker.  By default no kernel tracking is done.
 Note that timestamped_allocator is only effective if some tracking is
 specified.
 If kernel_tracker_max_interval = n > 0, then a tracking event
 is inserted after every n kernels without an event.
 

int32 kernel_tracker_max_interval = 7;

public int getKernelTrackerMaxPending ()

 If kernel_tracker_max_pending > 0 then no more than this many
 tracking events can be outstanding at a time.  An attempt to
 launch an additional kernel will stall until an event
 completes.
 

int32 kernel_tracker_max_pending = 9;

public int getNumDevToDevCopyStreams ()

 If > 1, the number of device-to-device copy streams to create
 for each GPUDevice.  Default value is 0, which is automatically
 converted to 1.
 

int32 num_dev_to_dev_copy_streams = 3;

общественный getParserForType ()

public int getSerializedSize ()

общедоступное логическое значение getTimestampedAllocator ()

 If true then extra work is done by GPUDevice and GPUBFCAllocator to
 keep track of when GPU memory is freed and when kernels actually
 complete so that we can know when a nominally free memory chunk
 is really not subject to pending use.
 

bool timestamped_allocator = 5;

публичный финал com.google.protobuf.UnknownFieldSet getUnknownFields ()

общедоступное логическое значение getUseUnifiedMemory ()

 If true, uses CUDA unified memory for memory allocations. If
 per_process_gpu_memory_fraction option is greater than 1.0, then unified
 memory is used regardless of the value for this field. See comments for
 per_process_gpu_memory_fraction field for more details and requirements
 of the unified memory. This option is useful to oversubscribe memory if
 multiple processes are sharing a single GPU while individually using less
 than 1.0 per process memory fraction.
 

bool use_unified_memory = 2;

public GPUOptions.Experimental.VirtualDevices getVirtualDevices (индекс int)

 The multi virtual device settings. If empty (not set), it will create
 single virtual device on each visible GPU, according to the settings
 in "visible_device_list" above. Otherwise, the number of elements in the
 list must be the same as the number of visible GPUs (after
 "visible_device_list" filtering if it is set), and the string represented
 device names (e.g. /device:GPU:<id>) will refer to the virtual
 devices and have the <id> field assigned sequentially starting from 0,
 according to the order they appear in this list and the "memory_limit"
 list inside each element. For example,
   visible_device_list = "1,0"
   virtual_devices { memory_limit: 1GB memory_limit: 2GB }
   virtual_devices {}
 will create three virtual devices as:
   /device:GPU:0 -> visible GPU 1 with 1GB memory
   /device:GPU:1 -> visible GPU 1 with 2GB memory
   /device:GPU:2 -> visible GPU 0 with all available memory
 NOTE:
 1. It's invalid to set both this and "per_process_gpu_memory_fraction"
    at the same time.
 2. Currently this setting is per-process, not per-session. Using
    different settings in different sessions within same process will
    result in undefined behavior.
 

repeated .tensorflow.GPUOptions.Experimental.VirtualDevices virtual_devices = 1;

public int getVirtualDevicesCount ()

 The multi virtual device settings. If empty (not set), it will create
 single virtual device on each visible GPU, according to the settings
 in "visible_device_list" above. Otherwise, the number of elements in the
 list must be the same as the number of visible GPUs (after
 "visible_device_list" filtering if it is set), and the string represented
 device names (e.g. /device:GPU:<id>) will refer to the virtual
 devices and have the <id> field assigned sequentially starting from 0,
 according to the order they appear in this list and the "memory_limit"
 list inside each element. For example,
   visible_device_list = "1,0"
   virtual_devices { memory_limit: 1GB memory_limit: 2GB }
   virtual_devices {}
 will create three virtual devices as:
   /device:GPU:0 -> visible GPU 1 with 1GB memory
   /device:GPU:1 -> visible GPU 1 with 2GB memory
   /device:GPU:2 -> visible GPU 0 with all available memory
 NOTE:
 1. It's invalid to set both this and "per_process_gpu_memory_fraction"
    at the same time.
 2. Currently this setting is per-process, not per-session. Using
    different settings in different sessions within same process will
    result in undefined behavior.
 

repeated .tensorflow.GPUOptions.Experimental.VirtualDevices virtual_devices = 1;

общедоступный список < GPUOptions.Experimental.VirtualDevices > getVirtualDevicesList ()

 The multi virtual device settings. If empty (not set), it will create
 single virtual device on each visible GPU, according to the settings
 in "visible_device_list" above. Otherwise, the number of elements in the
 list must be the same as the number of visible GPUs (after
 "visible_device_list" filtering if it is set), and the string represented
 device names (e.g. /device:GPU:<id>) will refer to the virtual
 devices and have the <id> field assigned sequentially starting from 0,
 according to the order they appear in this list and the "memory_limit"
 list inside each element. For example,
   visible_device_list = "1,0"
   virtual_devices { memory_limit: 1GB memory_limit: 2GB }
   virtual_devices {}
 will create three virtual devices as:
   /device:GPU:0 -> visible GPU 1 with 1GB memory
   /device:GPU:1 -> visible GPU 1 with 2GB memory
   /device:GPU:2 -> visible GPU 0 with all available memory
 NOTE:
 1. It's invalid to set both this and "per_process_gpu_memory_fraction"
    at the same time.
 2. Currently this setting is per-process, not per-session. Using
    different settings in different sessions within same process will
    result in undefined behavior.
 

repeated .tensorflow.GPUOptions.Experimental.VirtualDevices virtual_devices = 1;

public GPUOptions.Experimental.VirtualDevicesOrBuilder getVirtualDevicesOrBuilder (индекс int)

 The multi virtual device settings. If empty (not set), it will create
 single virtual device on each visible GPU, according to the settings
 in "visible_device_list" above. Otherwise, the number of elements in the
 list must be the same as the number of visible GPUs (after
 "visible_device_list" filtering if it is set), and the string represented
 device names (e.g. /device:GPU:<id>) will refer to the virtual
 devices and have the <id> field assigned sequentially starting from 0,
 according to the order they appear in this list and the "memory_limit"
 list inside each element. For example,
   visible_device_list = "1,0"
   virtual_devices { memory_limit: 1GB memory_limit: 2GB }
   virtual_devices {}
 will create three virtual devices as:
   /device:GPU:0 -> visible GPU 1 with 1GB memory
   /device:GPU:1 -> visible GPU 1 with 2GB memory
   /device:GPU:2 -> visible GPU 0 with all available memory
 NOTE:
 1. It's invalid to set both this and "per_process_gpu_memory_fraction"
    at the same time.
 2. Currently this setting is per-process, not per-session. Using
    different settings in different sessions within same process will
    result in undefined behavior.
 

repeated .tensorflow.GPUOptions.Experimental.VirtualDevices virtual_devices = 1;

общедоступный список<? расширяет GPUOptions.Experimental.VirtualDevicesOrBuilder > getVirtualDevicesOrBuilderList ()

 The multi virtual device settings. If empty (not set), it will create
 single virtual device on each visible GPU, according to the settings
 in "visible_device_list" above. Otherwise, the number of elements in the
 list must be the same as the number of visible GPUs (after
 "visible_device_list" filtering if it is set), and the string represented
 device names (e.g. /device:GPU:<id>) will refer to the virtual
 devices and have the <id> field assigned sequentially starting from 0,
 according to the order they appear in this list and the "memory_limit"
 list inside each element. For example,
   visible_device_list = "1,0"
   virtual_devices { memory_limit: 1GB memory_limit: 2GB }
   virtual_devices {}
 will create three virtual devices as:
   /device:GPU:0 -> visible GPU 1 with 1GB memory
   /device:GPU:1 -> visible GPU 1 with 2GB memory
   /device:GPU:2 -> visible GPU 0 with all available memory
 NOTE:
 1. It's invalid to set both this and "per_process_gpu_memory_fraction"
    at the same time.
 2. Currently this setting is per-process, not per-session. Using
    different settings in different sessions within same process will
    result in undefined behavior.
 

repeated .tensorflow.GPUOptions.Experimental.VirtualDevices virtual_devices = 1;

public int hashCode ()

публичное окончательное логическое значение isInitialized ()

общедоступный статический GPUOptions.Experimental.Builder newBuilder (прототип GPUOptions.Experimental )

общедоступный статический GPUOptions.Experimental.Builder newBuilder ()

общедоступный GPUOptions.Experimental.Builder newBuilderForType ()

общедоступный статический GPUOptions.Experimental parseDelimitedFrom (входной поток InputStream)

Броски

Исключение IO

общедоступный статический GPUOptions.Experimental parseDelimitedFrom (входной поток InputStream, com.google.protobuf.ExtensionRegistryLite ExtensionRegistry)

Броски

Исключение IO

общедоступный статический GPUOptions.Experimental parseFrom (данные ByteBuffer)

Броски

Инвалидпротоколбуфферисключение

общедоступный статический GPUOptions.Experimental parseFrom (вход com.google.protobuf.CodedInputStream, com.google.protobuf.ExtensionRegistryLite ExtensionRegistry)

Броски

Исключение IO

public static GPUOptions.Experimental parseFrom (данные ByteBuffer, com.google.protobuf.ExtensionRegistryLite ExtensionRegistry)

Броски

Инвалидпротоколбуфферисключение

общедоступный статический GPUOptions.Experimental parseFrom (вход com.google.protobuf.CodedInputStream)

Броски

Исключение IO

общедоступный статический GPUOptions.Experimental parseFrom (данные byte[], com.google.protobuf.ExtensionRegistryLite ExtensionRegistry)

Броски

Инвалидпротоколбуфферисключение

общедоступный статический GPUOptions.Experimental parseFrom (данные com.google.protobuf.ByteString)

Броски

Инвалидпротоколбуфферисключение

общедоступный статический GPUOptions.Experimental parseFrom (входной поток InputStream, com.google.protobuf.ExtensionRegistryLite ExtensionRegistry)

Броски

Исключение IO

общедоступный статический GPUOptions.Experimental parseFrom (данные com.google.protobuf.ByteString, com.google.protobuf.ExtensionRegistryLite ExtensionRegistry)

Броски

Инвалидпротоколбуфферисключение

общественный статический парсер ()

общедоступный GPUOptions.Experimental.Builder toBuilder ()

public void writeTo (вывод com.google.protobuf.CodedOutputStream)

Броски

Исключение IO