सार्वजनिक स्थैतिक अंतिम वर्ग GPUOptions.Builder
प्रोटोबफ़ प्रकार tensorflow.GPUOptions
सार्वजनिक तरीके
| GPUOptions.बिल्डर | addRepeatedField (com.google.protobuf.Descriptors.FieldDescriptor फ़ील्ड, ऑब्जेक्ट मान) |
| जीपीयू विकल्प | निर्माण () |
| जीपीयू विकल्प | बिल्डआंशिक () |
| GPUOptions.बिल्डर | स्पष्ट () |
| GPUOptions.बिल्डर | क्लियरएलोकेटरटाइप () The type of GPU allocation strategy to use. |
| GPUOptions.बिल्डर | ClearAllowGrowth () If true, the allocator does not pre-allocate the entire specified GPU memory region, instead starting small and growing as needed. |
| GPUOptions.बिल्डर | ClearDeferredDeletionBytes () Delay deletion of up to this many bytes to reduce the number of interactions with gpu driver code. |
| GPUOptions.बिल्डर | स्पष्टप्रयोगात्मक () Everything inside experimental is subject to change and is not subject to API stability guarantees in https://www.tensorflow.org/guide/version_compat. |
| GPUOptions.बिल्डर | क्लियरफ़ील्ड (com.google.protobuf.Descriptors.FieldDescriptor फ़ील्ड) |
| GPUOptions.बिल्डर | क्लियरफोर्सजीपीयूसंगत () Force all tensors to be gpu_compatible. |
| GPUOptions.बिल्डर | ClearOneof (com.google.protobuf.Descriptors.OneofDescriptor oneof) |
| GPUOptions.बिल्डर | ClearPerProcessGpuMemoryFraction () Fraction of the available GPU memory to allocate for each process. |
| GPUOptions.बिल्डर | ClearPollingActiveDelayUsecs () In the event polling loop sleep this many microseconds between PollEvents calls, when the queue is not empty. |
| GPUOptions.बिल्डर | ClearPollingInactiveDelayMsecs () This field is deprecated and ignored. |
| GPUOptions.बिल्डर | क्लियरविज़िबलडिवाइसलिस्ट () A comma-separated list of GPU ids that determines the 'visible' to 'virtual' mapping of GPU devices. |
| GPUOptions.बिल्डर | क्लोन () |
| डोरी | getAllocatorType () The type of GPU allocation strategy to use. |
| com.google.protobuf.ByteString | getAllocatorTypeBytes () The type of GPU allocation strategy to use. |
| बूलियन | getAllowGrowth () If true, the allocator does not pre-allocate the entire specified GPU memory region, instead starting small and growing as needed. |
| जीपीयू विकल्प | |
| लंबा | getDeferredDeletionBytes () Delay deletion of up to this many bytes to reduce the number of interactions with gpu driver code. |
| अंतिम स्थिर com.google.protobuf.Descriptors.Descriptor | |
| com.google.protobuf.Descriptors.Descriptor | |
| GPUOptions.प्रायोगिक | प्रयोगात्मक प्राप्त करें () Everything inside experimental is subject to change and is not subject to API stability guarantees in https://www.tensorflow.org/guide/version_compat. |
| GPUOptions.प्रायोगिक.बिल्डर | 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. |
| बूलियन | getForceGpuसंगत () Force all tensors to be gpu_compatible. |
| दोहरा | getPerProcessGpuMemoryFraction () Fraction of the available GPU memory to allocate for each process. |
| int यहाँ | getPollingActiveDelayUsecs () In the event polling loop sleep this many microseconds between PollEvents calls, when the queue is not empty. |
| int यहाँ | getPollingInactiveDelayMsecs () This field is deprecated and ignored. |
| डोरी | 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. |
| बूलियन | प्रयोगात्मक है () Everything inside experimental is subject to change and is not subject to API stability guarantees in https://www.tensorflow.org/guide/version_compat. |
| अंतिम बूलियन | |
| GPUOptions.बिल्डर | मर्जप्रायोगिक ( GPUOptions.प्रायोगिक मूल्य) Everything inside experimental is subject to change and is not subject to API stability guarantees in https://www.tensorflow.org/guide/version_compat. |
| GPUOptions.बिल्डर | मर्जफ्रॉम (com.google.protobuf.Message अन्य) |
| GPUOptions.बिल्डर | मर्जफ्रॉम (com.google.protobuf.CodedInputStream इनपुट, com.google.protobuf.ExtensionRegistryLite एक्सटेंशनरजिस्ट्री) |
| अंतिम GPUOptions.Builder | मर्जअज्ञातफ़ील्ड्स (com.google.protobuf.UnknownFieldSet अज्ञातफ़ील्ड्स) |
| GPUOptions.बिल्डर | setAllocatorType (स्ट्रिंग मान) The type of GPU allocation strategy to use. |
| GPUOptions.बिल्डर | setAllocatorTypeBytes (com.google.protobuf.ByteString मान) The type of GPU allocation strategy to use. |
| GPUOptions.बिल्डर | setAllowGrowth (बूलियन मान) If true, the allocator does not pre-allocate the entire specified GPU memory region, instead starting small and growing as needed. |
| GPUOptions.बिल्डर | setDeferredDeletionBytes (लंबा मान) Delay deletion of up to this many bytes to reduce the number of interactions with gpu driver code. |
| GPUOptions.बिल्डर | सेटएक्सपेरिमेंटल ( 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. |
| GPUOptions.बिल्डर | सेटप्रायोगिक ( GPUOptions.प्रायोगिक मूल्य) Everything inside experimental is subject to change and is not subject to API stability guarantees in https://www.tensorflow.org/guide/version_compat. |
| GPUOptions.बिल्डर | सेटफ़ील्ड (com.google.protobuf.Descriptors.FieldDescriptor फ़ील्ड, ऑब्जेक्ट मान) |
| GPUOptions.बिल्डर | setForceGpuCompatible (बूलियन मान) Force all tensors to be gpu_compatible. |
| GPUOptions.बिल्डर | setPerProcessGpuMemoryFraction (दोगुना मान) Fraction of the available GPU memory to allocate for each process. |
| GPUOptions.बिल्डर | setPollingActiveDelayUsecs (int मान) In the event polling loop sleep this many microseconds between PollEvents calls, when the queue is not empty. |
| GPUOptions.बिल्डर | setPollingInactiveDelayMsecs (int मान) This field is deprecated and ignored. |
| GPUOptions.बिल्डर | setRepeatedField (com.google.protobuf.Descriptors.FieldDescriptor फ़ील्ड, इंट इंडेक्स, ऑब्जेक्ट वैल्यू) |
| अंतिम GPUOptions.Builder | अज्ञात फ़ील्ड सेट करें (com.google.protobuf. अज्ञात फ़ील्ड सेट अज्ञात फ़ील्ड) |
| GPUOptions.बिल्डर | setVisibleDeviceList (स्ट्रिंग मान) A comma-separated list of GPU ids that determines the 'visible' to 'virtual' mapping of GPU devices. |
| GPUOptions.बिल्डर | setVisibleDeviceListBytes (com.google.protobuf.ByteString मान) A comma-separated list of GPU ids that determines the 'visible' to 'virtual' mapping of GPU devices. |
विरासत में मिली विधियाँ
सार्वजनिक तरीके
सार्वजनिक GPUOptions.Builder addRepeatedField (com.google.protobuf.Descriptors.FieldDescriptor फ़ील्ड, ऑब्जेक्ट मान)
सार्वजनिक 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; सार्वजनिक 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; सार्वजनिक 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; सार्वजनिक GPUOptions.Builder स्पष्टप्रायोगिक ()
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; सार्वजनिक 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; सार्वजनिक 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; सार्वजनिक 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; सार्वजनिक GPUOptions.Builder ClearPollingInactiveDelayMsecs ()
This field is deprecated and ignored.
int32 polling_inactive_delay_msecs = 7; सार्वजनिक 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; सार्वजनिक स्ट्रिंग 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; सार्वजनिक 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; सार्वजनिक बूलियन 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; सार्वजनिक लंबे समय तक 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; सार्वजनिक स्थैतिक अंतिम com.google.protobuf.Descriptors.Descriptor getDescriptor ()
सार्वजनिक com.google.protobuf.Descriptors.Descriptor getDescriptorForType ()
सार्वजनिक GPUOptions. प्रायोगिक 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; सार्वजनिक 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; सार्वजनिक 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; सार्वजनिक बूलियन 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; सार्वजनिक डबल 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; सार्वजनिक 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; सार्वजनिक int getPollingInactiveDelayMsecs ()
This field is deprecated and ignored.
int32 polling_inactive_delay_msecs = 7; सार्वजनिक स्ट्रिंग 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; सार्वजनिक 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; सार्वजनिक बूलियन में प्रायोगिक () है
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; सार्वजनिक अंतिम बूलियन आरंभीकृत है ()
सार्वजनिक GPUOptions.Builder मर्जप्रायोगिक ( GPUOptions.प्रायोगिक मान)
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; सार्वजनिक GPUOptions.Builder mergeFrom (com.google.protobuf.CodedInputStream इनपुट, com.google.protobuf.ExtensionRegistryLite एक्सटेंशनरजिस्ट्री)
फेंकता
| आईओ अपवाद |
|---|
सार्वजनिक अंतिम GPUOptions.Builder mergeUnknownFields (com.google.protobuf.UnknownFieldSet अज्ञातफील्ड्स)
सार्वजनिक GPUOptions.Builder setAllocatorType (स्ट्रिंग मान)
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; सार्वजनिक GPUOptions.Builder setAllocatorTypeBytes (com.google.protobuf.ByteString मान)
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; सार्वजनिक GPUOptions.Builder setAllowGrowth (बूलियन मान)
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; सार्वजनिक GPUOptions.Builder setDeferredDeletionBytes (लंबा मान)
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; सार्वजनिक 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;सार्वजनिक GPUOptions.Builder setExperimental ( GPUOptions.Experimental मान)
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; सार्वजनिक GPUOptions.Builder setField (com.google.protobuf.Descriptors.FieldDescriptor फ़ील्ड, ऑब्जेक्ट मान)
सार्वजनिक GPUOptions.Builder setForceGpuCompatible (बूलियन मान)
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; सार्वजनिक GPUOptions.Builder setPerProcessGpuMemoryFraction (दोगुना मान)
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; सार्वजनिक GPUOptions.Builder setPollingActiveDelayUsecs (int मान)
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; सार्वजनिक GPUOptions.Builder setPollingInactiveDelayMsecs (int मान)
This field is deprecated and ignored.
int32 polling_inactive_delay_msecs = 7; सार्वजनिक GPUOptions.Builder setRepeatedField (com.google.protobuf.Descriptors.FieldDescriptor फ़ील्ड, int अनुक्रमणिका, ऑब्जेक्ट मान)
सार्वजनिक अंतिम GPUOptions.Builder setUnknownFields (com.google.protobuf.UnknownFieldSet अज्ञातFields)
सार्वजनिक GPUOptions.Builder setVisibleDeviceList (स्ट्रिंग मान)
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; सार्वजनिक GPUOptions.Builder setVisibleDeviceListBytes (com.google.protobuf.ByteString मान)
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;