الفئة النهائية العامة الثابتة GPUOptions.Experimental.Builder
نوع Protobuf tensorflow.GPUOptions.Experimental
الأساليب العامة
| GPUOptions.Experimental.Builder | addAllVirtualDevices (Iterable<? يمتد قيم GPUOptions.Experimental.VirtualDevices >) The multi virtual device settings. |
| GPUOptions.Experimental.Builder | addRepeatedField (حقل com.google.protobuf.Descriptors.FieldDescriptor، قيمة الكائن) |
| GPUOptions.Experimental.Builder | addVirtualDevices ( GPUOptions.Experimental.VirtualDevices.Builder builderForValue) The multi virtual device settings. |
| GPUOptions.Experimental.Builder | addVirtualDevices (مؤشر int، قيمة GPUOptions.Experimental.VirtualDevices ) The multi virtual device settings. |
| GPUOptions.Experimental.Builder | |
| GPUOptions.Experimental.Builder | addVirtualDevices (مؤشر int، GPUOptions.Experimental.VirtualDevices.Builder builderForValue) The multi virtual device settings. |
| GPUOptions.Experimental.VirtualDevices.Builder | addVirtualDevicesBuilder () The multi virtual device settings. |
| GPUOptions.Experimental.VirtualDevices.Builder | addVirtualDevicesBuilder (فهرس كثافة العمليات) The multi virtual device settings. |
| GPUOptions.التجريبية | يبني () |
| GPUOptions.التجريبية | بناء جزئي () |
| GPUOptions.Experimental.Builder | واضح () |
| GPUOptions.Experimental.Builder | ClearCollectiveRingOrder () If non-empty, defines a good GPU ring order on a single worker based on device interconnect. |
| GPUOptions.Experimental.Builder | ClearField (حقل com.google.protobuf.Descriptors.FieldDescriptor) |
| GPUOptions.Experimental.Builder | ClearKernelTrackerMaxBytes () If kernel_tracker_max_bytes = n > 0, then a tracking event is inserted after every series of kernels allocating a sum of memory >= n. |
| GPUOptions.Experimental.Builder | ClearKernelTrackerMaxInterval () Parameters for GPUKernelTracker. |
| GPUOptions.Experimental.Builder | ClearKernelTrackerMaxPending () If kernel_tracker_max_pending > 0 then no more than this many tracking events can be outstanding at a time. |
| GPUOptions.Experimental.Builder | ClearNumDevToDevCopyStreams () If > 1, the number of device-to-device copy streams to create for each GPUDevice. |
| GPUOptions.Experimental.Builder | ClearOneof (com.google.protobuf.Descriptors.OneofDescriptor oneof) |
| GPUOptions.Experimental.Builder | ClearTimestampedAllocator () 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. |
| GPUOptions.Experimental.Builder | ClearUseUnifiedMemory () If true, uses CUDA unified memory for memory allocations. |
| GPUOptions.Experimental.Builder | ClearVirtualDevices () The multi virtual device settings. |
| GPUOptions.Experimental.Builder | استنساخ () |
| خيط | الحصول علىCollectiveRingOrder () If non-empty, defines a good GPU ring order on a single worker based on device interconnect. |
| com.google.protobuf.ByteString | الحصول علىCollectiveRingOrderBytes () If non-empty, defines a good GPU ring order on a single worker based on device interconnect. |
| GPUOptions.التجريبية | |
| النهائي الثابت com.google.protobuf.Descriptors.Descriptor | |
| com.google.protobuf.Descriptors.Descriptor | |
| كثافة العمليات | 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. |
| كثافة العمليات | getKernelTrackerMaxInterval () Parameters for GPUKernelTracker. |
| كثافة العمليات | getKernelTrackerMaxPending () If kernel_tracker_max_pending > 0 then no more than this many tracking events can be outstanding at a time. |
| كثافة العمليات | getNumDevToDevCopyStreams () If > 1, the number of device-to-device copy streams to create for each GPUDevice. |
| منطقية | الحصول على TimestampedAllocator () 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. |
| منطقية | الحصول علىUseUnifiedMemory () If true, uses CUDA unified memory for memory allocations. |
| GPUOptions.Experimental.VirtualDevices | getVirtualDevices (فهرس كثافة العمليات) The multi virtual device settings. |
| GPUOptions.Experimental.VirtualDevices.Builder | getVirtualDevicesBuilder (فهرس كثافة العمليات) The multi virtual device settings. |
| القائمة< GPUOptions.Experimental.VirtualDevices.Builder > | getVirtualDevicesBuilderList () The multi virtual device settings. |
| كثافة العمليات | الحصول على VirtualDevicesCount () The multi virtual device settings. |
| القائمة< GPUOptions.Experimental.VirtualDevices > | قائمة الأجهزة الافتراضية () The multi virtual device settings. |
| GPUOptions.Experimental.VirtualDevicesOrBuilder | getVirtualDevicesOrBuilder (فهرس كثافة العمليات) The multi virtual device settings. |
| القائمة<؟ يمتد GPUOptions.Experimental.VirtualDevicesOrBuilder > | getVirtualDevicesOrBuilderList () The multi virtual device settings. |
| منطقية نهائية | تمت التهيئة () |
| GPUOptions.Experimental.Builder | دمج من (com.google.protobuf.Message أخرى) |
| GPUOptions.Experimental.Builder | دمج من (com.google.protobuf.CodedInputStream input، com.google.protobuf.ExtensionRegistryLite ExtensionRegistry) |
| GPUOptions.Experimental.Builder النهائي | دمجUnknownFields (com.google.protobuf.UnknownFieldSet UnknownFields) |
| GPUOptions.Experimental.Builder | إزالة VirtualDevices (فهرس كثافة العمليات) The multi virtual device settings. |
| GPUOptions.Experimental.Builder | setCollectiveRingOrder (قيمة السلسلة) If non-empty, defines a good GPU ring order on a single worker based on device interconnect. |
| GPUOptions.Experimental.Builder | setCollectiveRingOrderBytes (قيمة com.google.protobuf.ByteString) If non-empty, defines a good GPU ring order on a single worker based on device interconnect. |
| GPUOptions.Experimental.Builder | setField (حقل com.google.protobuf.Descriptors.FieldDescriptor، قيمة الكائن) |
| GPUOptions.Experimental.Builder | setKernelTrackerMaxBytes (قيمة int) If kernel_tracker_max_bytes = n > 0, then a tracking event is inserted after every series of kernels allocating a sum of memory >= n. |
| GPUOptions.Experimental.Builder | setKernelTrackerMaxInterval (قيمة int) Parameters for GPUKernelTracker. |
| GPUOptions.Experimental.Builder | setKernelTrackerMaxPending (قيمة int) If kernel_tracker_max_pending > 0 then no more than this many tracking events can be outstanding at a time. |
| GPUOptions.Experimental.Builder | setNumDevToDevCopyStreams (قيمة int) If > 1, the number of device-to-device copy streams to create for each GPUDevice. |
| GPUOptions.Experimental.Builder | setRepeatedField (حقل com.google.protobuf.Descriptors.FieldDescriptor، فهرس int، قيمة الكائن) |
| GPUOptions.Experimental.Builder | setTimestampedAllocator (قيمة منطقية) 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. |
| GPUOptions.Experimental.Builder النهائي | setUnknownFields (com.google.protobuf.UnknownFieldSet UnknownFields) |
| GPUOptions.Experimental.Builder | setUseUnifiedMemory (قيمة منطقية) If true, uses CUDA unified memory for memory allocations. |
| GPUOptions.Experimental.Builder | setVirtualDevices (مؤشر int، GPUOptions.Experimental.VirtualDevices.Builder builderForValue) The multi virtual device settings. |
| GPUOptions.Experimental.Builder | setVirtualDevices (مؤشر int، قيمة GPUOptions.Experimental.VirtualDevices ) The multi virtual device settings. |
الطرق الموروثة
الأساليب العامة
GPUOptions.Experimental.Builder العامة addAllVirtualDevices (Iterable<؟ يمتد GPUOptions.Experimental.VirtualDevices > القيم)
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; GPUOptions.Experimental.Builder العام addRepeatedField (حقل com.google.protobuf.Descriptors.FieldDescriptor، قيمة الكائن)
GPUOptions.Experimental.Builder العامة addVirtualDevices ( GPUOptions.Experimental.VirtualDevices.Builder builderForValue)
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; GPUOptions.Experimental.Builder العامة addVirtualDevices (مؤشر int، قيمة GPUOptions.Experimental.VirtualDevices )
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; GPUOptions.Experimental.Builder العامة addVirtualDevices (قيمة GPUOptions.Experimental.VirtualDevices )
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; GPUOptions.Experimental.Builder العامة addVirtualDevices (مؤشر int، GPUOptions.Experimental.VirtualDevices.Builder builderForValue)
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; GPUOptions.Experimental.VirtualDevices.Builder العامة addVirtualDevicesBuilder ()
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; GPUOptions.Experimental.VirtualDevices.Builder العامة addVirtualDevicesBuilder (فهرس كثافة العمليات)
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; GPUOptions.Experimental.Builder العام ClearCollectiveRingOrder ()
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; GPUOptions.Experimental.Builder ClearField العام (حقل com.google.protobuf.Descriptors.FieldDescriptor)
GPUOptions.Experimental.Builder العامة ClearKernelTrackerMaxBytes ()
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; GPUOptions.Experimental.Builder العامة ClearKernelTrackerMaxInterval ()
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; GPUOptions.Experimental.Builder العامة ClearKernelTrackerMaxPending ()
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; GPUOptions.Experimental.Builder العامة ClearNumDevToDevCopyStreams ()
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; GPUOptions.Experimental.Builder ClearOneof العام (com.google.protobuf.Descriptors.OneofDescriptor oneof)
GPUOptions.Experimental.Builder العام ، ClearTimestampedAllocator ()
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; GPUOptions.Experimental.Builder العامة ClearUseUnifiedMemory ()
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; GPUOptions.Experimental.Builder العام ClearVirtualDevices ()
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; سلسلة عامة 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; com.google.protobuf.ByteString العامة 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; النهائي العام الثابت com.google.protobuf.Descriptors.Descriptor getDescriptor ()
com.google.protobuf.Descriptors.Descriptor getDescriptorForType () العام
int العام 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; int public 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; int العام 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; int public 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; getTimestampedAllocator () المنطقي العام
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; getUseUnifiedMemory () المنطقية العامة
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; GPUOptions.Experimental.VirtualDevices العامة getVirtualDevices (مؤشر 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; GPUOptions.Experimental.VirtualDevices.Builder العامة getVirtualDevicesBuilder (فهرس كثافة العمليات)
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; القائمة العامة< GPUOptions.Experimental.VirtualDevices.Builder > getVirtualDevicesBuilderList ()
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; int public 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; القائمة العامة< 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; GPUOptions.Experimental.VirtualDevicesOrBuilder العام getVirtualDevicesOrBuilder (فهرس كثافة العمليات)
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; القائمة العامة <؟ يمتد 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; تمت تهيئة القيمة المنطقية النهائية العامة ()
GPUOptions.Experimental.Builder العام mergeFrom (com.google.protobuf.CodedInputStream input, com.google.protobuf.ExtensionRegistryLite ExtensionRegistry)
رميات
| IOEException |
|---|
GPUOptions.Experimental.Builder النهائي العام ، mergeUnknownFields (com.google.protobuf.UnknownFieldSetUnknownFields)
GPUOptions.Experimental.Builder العام يزيلVirtualDevices (مؤشر 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; GPUOptions.Experimental.Builder العامة setCollectiveRingOrder (قيمة السلسلة)
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; GPUOptions.Experimental.Builder العامة setCollectiveRingOrderBytes (قيمة com.google.protobuf.ByteString)
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; GPUOptions.Experimental.Builder العام setField (com.google.protobuf.Descriptors.FieldDescriptor الحقل، قيمة الكائن)
GPUOptions.Experimental.Builder العامة setKernelTrackerMaxBytes (قيمة int)
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; GPUOptions.Experimental.Builder العامة setKernelTrackerMaxInterval (قيمة int)
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; GPUOptions.Experimental.Builder العامة setKernelTrackerMaxPending (قيمة int)
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; مجموعة GPUOptions.Experimental.Builder العامة NumDevToDevCopyStreams (قيمة int)
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; GPUOptions.Experimental.Builder العام setRepeatedField (com.google.protobuf.Descriptors.FieldDescriptor الحقل، مؤشر int، قيمة الكائن)
GPUOptions.Experimental.Builder العامة setTimestampedAllocator (قيمة منطقية)
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; GPUOptions.Experimental.Builder setUnknownFields (com.google.protobuf.UnknownFieldSetknownFields) النهائي العام
GPUOptions.Experimental.Builder العامة setUseUnifiedMemory (قيمة منطقية)
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; مجموعة GPUOptions.Experimental.Builder العامة VirtualDevices (مؤشر int، GPUOptions.Experimental.VirtualDevices.Builder builderForValue)
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; مجموعة GPUOptions.Experimental.Builder العامة VirtualDevices (مؤشر int، قيمة GPUOptions.Experimental.VirtualDevices )
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;