الواجهة العامة GPUOptionsOrBuilder
| الفئات الفرعية غير المباشرة المعروفة |
الأساليب العامة
| سلسلة مجردة | GetAllocatorType () The type of GPU allocation strategy to use. |
| مجردة com.google.protobuf.ByteString | الحصول علىAllocatorTypeBytes () The type of GPU allocation strategy to use. |
| منطقية مجردة | الحصول على السماح بالنمو () If true, the allocator does not pre-allocate the entire specified GPU memory region, instead starting small and growing as needed. |
| مجردة طويلة | getDeferredDeletionBytes () Delay deletion of up to this many bytes to reduce the number of interactions with gpu driver code. |
| مجردة GPUOptions.Experimental | الحصول التجريبي () Everything inside experimental is subject to change and is not subject to API stability guarantees in https://www.tensorflow.org/guide/version_compat. |
| مجردة GPUOptions.ExperimentalOrBuilder | الحصول على التجريبية أو البناء () Everything inside experimental is subject to change and is not subject to API stability guarantees in https://www.tensorflow.org/guide/version_compat. |
| منطقية مجردة | الحصول علىForceGpuCompatible () Force all tensors to be gpu_compatible. |
| مجردة مزدوجة | getPerProcessGpuMemoryFraction () Fraction of the available GPU memory to allocate for each process. |
| كثافة العمليات مجردة | getPollingActiveDelayUsecs () In the event polling loop sleep this many microseconds between PollEvents calls, when the queue is not empty. |
| كثافة العمليات مجردة | getPollingInactiveDelayMsecs () This field is deprecated and ignored. |
| سلسلة مجردة | getVisibleDeviceList () A comma-separated list of GPU ids that determines the 'visible' to 'virtual' mapping of GPU devices. |
| مجردة com.google.protobuf.ByteString | getVisibleDeviceListBytes () A comma-separated list of GPU ids that determines the 'visible' to 'virtual' mapping of GPU devices. |
| منطقية مجردة | تجريبي () Everything inside experimental is subject to change and is not subject to API stability guarantees in https://www.tensorflow.org/guide/version_compat. |
الأساليب العامة
سلسلة مجردة عامة getAllocatorType ()
The type of GPU allocation strategy to use.
Allowed values:
"": The empty string (default) uses a system-chosen default
which may change over time.
"BFC": A "Best-fit with coalescing" algorithm, simplified from a
version of dlmalloc.
string allocator_type = 2; الملخص العام com.google.protobuf.ByteString getAllocatorTypeBytes ()
The type of GPU allocation strategy to use.
Allowed values:
"": The empty string (default) uses a system-chosen default
which may change over time.
"BFC": A "Best-fit with coalescing" algorithm, simplified from a
version of dlmalloc.
string allocator_type = 2; الملخص المنطقي العام getAllowGrowth ()
If true, the allocator does not pre-allocate the entire specified GPU memory region, instead starting small and growing as needed.
bool allow_growth = 4; الملخص العام الطويل getDeferredDeletionBytes ()
Delay deletion of up to this many bytes to reduce the number of interactions with gpu driver code. If 0, the system chooses a reasonable default (several MBs).
int64 deferred_deletion_bytes = 3; الملخص العام GPUOptions.getExperimental ( )
Everything inside experimental is subject to change and is not subject to API stability guarantees in https://www.tensorflow.org/guide/version_compat.
.tensorflow.GPUOptions.Experimental experimental = 9; الملخص العام GPUOptions.ExperimentalOrBuilder getExperimentalOrBuilder ()
Everything inside experimental is subject to change and is not subject to API stability guarantees in https://www.tensorflow.org/guide/version_compat.
.tensorflow.GPUOptions.Experimental experimental = 9; الملخص المنطقي العام getForceGpuCompatible ()
Force all tensors to be gpu_compatible. On a GPU-enabled TensorFlow, enabling this option forces all CPU tensors to be allocated with Cuda pinned memory. Normally, TensorFlow will infer which tensors should be allocated as the pinned memory. But in case where the inference is incomplete, this option can significantly speed up the cross-device memory copy performance as long as it fits the memory. Note that this option is not something that should be enabled by default for unknown or very large models, since all Cuda pinned memory is unpageable, having too much pinned memory might negatively impact the overall host system performance.
bool force_gpu_compatible = 8; الملخص العام المزدوج getPerProcessGpuMemoryFraction ()
Fraction of the available GPU memory to allocate for each process. 1 means to allocate all of the GPU memory, 0.5 means the process allocates up to ~50% of the available GPU memory. GPU memory is pre-allocated unless the allow_growth option is enabled. If greater than 1.0, uses CUDA unified memory to potentially oversubscribe the amount of memory available on the GPU device by using host memory as a swap space. Accessing memory not available on the device will be significantly slower as that would require memory transfer between the host and the device. Options to reduce the memory requirement should be considered before enabling this option as this may come with a negative performance impact. Oversubscription using the unified memory requires Pascal class or newer GPUs and it is currently only supported on the Linux operating system. See https://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html#um-requirements for the detailed requirements.
double per_process_gpu_memory_fraction = 1; الملخص العام int getPollingActiveDelayUsecs ()
In the event polling loop sleep this many microseconds between PollEvents calls, when the queue is not empty. If value is not set or set to 0, gets set to a non-zero default.
int32 polling_active_delay_usecs = 6; الملخص العام int getPollingInactiveDelayMsecs ()
This field is deprecated and ignored.
int32 polling_inactive_delay_msecs = 7; سلسلة مجردة عامة getVisibleDeviceList ()
A comma-separated list of GPU ids that determines the 'visible'
to 'virtual' mapping of GPU devices. For example, if TensorFlow
can see 8 GPU devices in the process, and one wanted to map
visible GPU devices 5 and 3 as "/device:GPU:0", and "/device:GPU:1",
then one would specify this field as "5,3". This field is similar in
spirit to the CUDA_VISIBLE_DEVICES environment variable, except
it applies to the visible GPU devices in the process.
NOTE:
1. The GPU driver provides the process with the visible GPUs
in an order which is not guaranteed to have any correlation to
the *physical* GPU id in the machine. This field is used for
remapping "visible" to "virtual", which means this operates only
after the process starts. Users are required to use vendor
specific mechanisms (e.g., CUDA_VISIBLE_DEVICES) to control the
physical to visible device mapping prior to invoking TensorFlow.
2. In the code, the ids in this list are also called "platform GPU id"s,
and the 'virtual' ids of GPU devices (i.e. the ids in the device
name "/device:GPU:<id>") are also called "TF GPU id"s. Please
refer to third_party/tensorflow/core/common_runtime/gpu/gpu_id.h
for more information.
string visible_device_list = 5; الملخص العام com.google.protobuf.ByteString getVisibleDeviceListBytes ()
A comma-separated list of GPU ids that determines the 'visible'
to 'virtual' mapping of GPU devices. For example, if TensorFlow
can see 8 GPU devices in the process, and one wanted to map
visible GPU devices 5 and 3 as "/device:GPU:0", and "/device:GPU:1",
then one would specify this field as "5,3". This field is similar in
spirit to the CUDA_VISIBLE_DEVICES environment variable, except
it applies to the visible GPU devices in the process.
NOTE:
1. The GPU driver provides the process with the visible GPUs
in an order which is not guaranteed to have any correlation to
the *physical* GPU id in the machine. This field is used for
remapping "visible" to "virtual", which means this operates only
after the process starts. Users are required to use vendor
specific mechanisms (e.g., CUDA_VISIBLE_DEVICES) to control the
physical to visible device mapping prior to invoking TensorFlow.
2. In the code, the ids in this list are also called "platform GPU id"s,
and the 'virtual' ids of GPU devices (i.e. the ids in the device
name "/device:GPU:<id>") are also called "TF GPU id"s. Please
refer to third_party/tensorflow/core/common_runtime/gpu/gpu_id.h
for more information.
string visible_device_list = 5; الملخص المنطقي العام التجريبي ()
Everything inside experimental is subject to change and is not subject to API stability guarantees in https://www.tensorflow.org/guide/version_compat.
.tensorflow.GPUOptions.Experimental experimental = 9;