공개 정적 최종 클래스 GPUOptions.Builder
Protobuf 유형 tensorflow.GPUOptions
공개 방법
| GPU옵션.빌더 | addRepeatedField (com.google.protobuf.Descriptors.FieldDescriptor 필드, 개체 값) |
| GPU옵션 | 짓다 () |
| GPU옵션 | 빌드부분 () |
| GPU옵션.빌더 | 분명한 () |
| GPU옵션.빌더 | 클리어 할당자 유형 () The type of GPU allocation strategy to use. |
| GPU옵션.빌더 | 클리어허용성장 () If true, the allocator does not pre-allocate the entire specified GPU memory region, instead starting small and growing as needed. |
| GPU옵션.빌더 | 클리어DeferredDeletionBytes () Delay deletion of up to this many bytes to reduce the number of interactions with gpu driver code. |
| GPU옵션.빌더 | 명확한실험적 () Everything inside experimental is subject to change and is not subject to API stability guarantees in https://www.tensorflow.org/guide/version_compat. |
| GPU옵션.빌더 | ClearField (com.google.protobuf.Descriptors.FieldDescriptor 필드) |
| GPU옵션.빌더 | 클리어포스Gpu호환 () Force all tensors to be gpu_compatible. |
| GPU옵션.빌더 | ClearOneof (com.google.protobuf.Descriptors.OneofDescriptor oneof) |
| GPU옵션.빌더 | ClearPerProcessGpuMemoryFraction () Fraction of the available GPU memory to allocate for each process. |
| GPU옵션.빌더 | ClearPollingActiveDelayUsecs () In the event polling loop sleep this many microseconds between PollEvents calls, when the queue is not empty. |
| GPU옵션.빌더 | ClearPollingInactiveDelayMsecs () This field is deprecated and ignored. |
| GPU옵션.빌더 | 클리어VisibleDeviceList () A comma-separated list of GPU ids that determines the 'visible' to 'virtual' mapping of GPU devices. |
| GPU옵션.빌더 | 클론 () |
| 끈 | getAllocatorType () The type of GPU allocation strategy to use. |
| com.google.protobuf.ByteString | getAllocatorTypeBytes () The type of GPU allocation strategy to use. |
| 부울 | getAllowGrowth () If true, the allocator does not pre-allocate the entire specified GPU memory region, instead starting small and growing as needed. |
| GPU옵션 | |
| 긴 | getDeferredDeletionBytes () Delay deletion of up to this many bytes to reduce the number of interactions with gpu driver code. |
| 최종 정적 com.google.protobuf.Descriptors.Descriptor | |
| com.google.protobuf.Descriptors.Descriptor | |
| GPUOptions.Experimental | getExperimental () Everything inside experimental is subject to change and is not subject to API stability guarantees in https://www.tensorflow.org/guide/version_compat. |
| GPUOptions.Experimental.Builder | getExperimentalBuilder () Everything inside experimental is subject to change and is not subject to API stability guarantees in https://www.tensorflow.org/guide/version_compat. |
| 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. |
| 부울 | getForceGpu호환 () Force all tensors to be gpu_compatible. |
| 더블 | getPerProcessGpuMemoryFraction () Fraction of the available GPU memory to allocate for each process. |
| 정수 | getPollingActiveDelayUsecs () In the event polling loop sleep this many microseconds between PollEvents calls, when the queue is not empty. |
| 정수 | getPollingInactiveDelayMsecs () This field is deprecated and ignored. |
| 끈 | getVisibleDeviceList () A comma-separated list of GPU ids that determines the 'visible' to 'virtual' mapping of GPU devices. |
| com.google.protobuf.ByteString | getVisibleDeviceListBytes () A comma-separated list of GPU ids that determines the 'visible' to 'virtual' mapping of GPU devices. |
| 부울 | 실험적 () Everything inside experimental is subject to change and is not subject to API stability guarantees in https://www.tensorflow.org/guide/version_compat. |
| 최종 부울 | 초기화됨 () |
| GPU옵션.빌더 | mergeExperimental ( GPUOptions.Experimental 값) Everything inside experimental is subject to change and is not subject to API stability guarantees in https://www.tensorflow.org/guide/version_compat. |
| GPU옵션.빌더 | mergeFrom (com.google.protobuf.다른 메시지 보내기) |
| GPU옵션.빌더 | mergeFrom (com.google.protobuf.CodedInputStream 입력, com.google.protobuf.ExtensionRegistryLite ExtensionRegistry) |
| 최종 GPUOptions.Builder | mergeUnknownFields (com.google.protobuf.UnknownFieldSet 알려지지 않은Fields) |
| GPU옵션.빌더 | setAllocatorType (문자열 값) The type of GPU allocation strategy to use. |
| GPU옵션.빌더 | setAllocatorTypeBytes (com.google.protobuf.ByteString 값) The type of GPU allocation strategy to use. |
| GPU옵션.빌더 | setAllowGrowth (부울 값) If true, the allocator does not pre-allocate the entire specified GPU memory region, instead starting small and growing as needed. |
| GPU옵션.빌더 | setDeferredDeletionBytes (긴 값) Delay deletion of up to this many bytes to reduce the number of interactions with gpu driver code. |
| GPU옵션.빌더 | setExperimental ( GPUOptions.Experimental.Builder builderForValue) Everything inside experimental is subject to change and is not subject to API stability guarantees in https://www.tensorflow.org/guide/version_compat. |
| GPU옵션.빌더 | setExperimental ( GPUOptions.Experimental 값) Everything inside experimental is subject to change and is not subject to API stability guarantees in https://www.tensorflow.org/guide/version_compat. |
| GPU옵션.빌더 | setField (com.google.protobuf.Descriptors.FieldDescriptor 필드, 개체 값) |
| GPU옵션.빌더 | setForceGpuCompatible (부울 값) Force all tensors to be gpu_compatible. |
| GPU옵션.빌더 | setPerProcessGpuMemoryFraction (이중 값) Fraction of the available GPU memory to allocate for each process. |
| GPU옵션.빌더 | setPollingActiveDelayUsecs (int 값) In the event polling loop sleep this many microseconds between PollEvents calls, when the queue is not empty. |
| GPU옵션.빌더 | setPollingInactiveDelayMsecs (정수 값) This field is deprecated and ignored. |
| GPU옵션.빌더 | setRepeatedField (com.google.protobuf.Descriptors.FieldDescriptor 필드, int 인덱스, 개체 값) |
| 최종 GPUOptions.Builder | setUnknownFields (com.google.protobuf.UnknownFieldSet UnknownFields) |
| GPU옵션.빌더 | setVisibleDeviceList (문자열 값) A comma-separated list of GPU ids that determines the 'visible' to 'virtual' mapping of GPU devices. |
| GPU옵션.빌더 | setVisibleDeviceListBytes (com.google.protobuf.ByteString 값) A comma-separated list of GPU ids that determines the 'visible' to 'virtual' mapping of GPU devices. |
상속된 메서드
공개 방법
공개 GPUOptions.Builder ClearAllocatorType ()
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; 공개 GPUOptions.Builder ClearAllowGrowth ()
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; 공개 GPUOptions.Builder ClearDeferredDeletionBytes ()
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; 공개 GPUOptions.Builder ClearExperimental ()
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; 공개 GPUOptions.Builder ClearForceGpuCompatible ()
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; 공개 GPUOptions.Builder ClearPerProcessGpuMemoryFraction ()
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; 공개 GPUOptions.Builder ClearPollingActiveDelayUsecs ()
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; 공개 GPUOptions.Builder ClearPollingInactiveDelayMsecs ()
This field is deprecated and ignored.
int32 polling_inactive_delay_msecs = 7; 공개 GPUOptions.Builder ClearVisibleDeviceList ()
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; 공개 문자열 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; 공개 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; 공개 부울 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; 공개 긴 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; 공개 정적 최종 com.google.protobuf.Descriptors.Descriptor getDescriptor ()
공개 com.google.protobuf.Descriptors.Descriptor getDescriptorForType ()
공개 GPUOptions.Experimental 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; 공개 GPUOptions.Experimental.Builder getExperimentalBuilder ()
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; 공개 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; 공개 부울 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; 공개 이중 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; 공개 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; 공개 int getPollingInactiveDelayMsecs ()
This field is deprecated and ignored.
int32 polling_inactive_delay_msecs = 7; 공개 문자열 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; 공개 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; 공개 부울 hasExperimental ()
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; 공개 최종 부울 isInitialized ()
공개 GPUOptions.Builder mergeExperimental ( GPUOptions.Experimental 값)
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; 공개 GPUOptions.Builder mergeFrom (com.google.protobuf.CodedInputStream 입력, com.google.protobuf.ExtensionRegistryLite ExtensionRegistry)
던지기
| IO예외 |
|---|
공용 GPUOptions.Builder setAllocatorType (문자열 값)
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; 공개 GPUOptions.Builder setAllocatorTypeBytes (com.google.protobuf.ByteString 값)
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; 공개 GPUOptions.Builder setAllowGrowth (부울 값)
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; 공용 GPUOptions.Builder setDeferredDeletionBytes (긴 값)
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; 공개 GPUOptions.Builder setExperimental ( GPUOptions.Experimental.Builder builderForValue)
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; 공개 GPUOptions.Builder setExperimental ( GPUOptions.Experimental 값)
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; 공개 GPUOptions.Builder setForceGpuCompatible (부울 값)
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; 공개 GPUOptions.Builder setPerProcessGpuMemoryFraction (이중 값)
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; 공개 GPUOptions.Builder setPollingActiveDelayUsecs (int 값)
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; 공개 GPUOptions.Builder setPollingInactiveDelayMsecs (int 값)
This field is deprecated and ignored.
int32 polling_inactive_delay_msecs = 7; 공개 GPUOptions.Builder setRepeatedField (com.google.protobuf.Descriptors.FieldDescriptor 필드, int 인덱스, 개체 값)
공개 GPUOptions.Builder setVisibleDeviceList (문자열 값)
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; 공개 GPUOptions.Builder setVisibleDeviceListBytes (com.google.protobuf.ByteString 값)
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;