পাবলিক স্ট্যাটিক ফাইনাল ক্লাস GPUOptions.Builder
Protobuf টাইপ tensorflow.GPUOptions
পাবলিক পদ্ধতি
| GPUOptions.Builder | addRepeatedField (com.google.protobuf.Descriptors.FieldDescriptor ক্ষেত্র, বস্তুর মান) |
| জিপিইউ অপশন | নির্মাণ () |
| জিপিইউ অপশন | বিল্ড আংশিক () |
| GPUOptions.Builder | পরিষ্কার () |
| GPUOptions.Builder | সাফ অ্যালোকেটার টাইপ () The type of GPU allocation strategy to use. |
| GPUOptions.Builder | clearAllowGrowth () If true, the allocator does not pre-allocate the entire specified GPU memory region, instead starting small and growing as needed. |
| GPUOptions.Builder | সাফ ডিফারেড ডিলিটশনবাইটস () Delay deletion of up to this many bytes to reduce the number of interactions with gpu driver code. |
| GPUOptions.Builder | পরিষ্কার পরীক্ষামূলক () Everything inside experimental is subject to change and is not subject to API stability guarantees in https://www.tensorflow.org/guide/version_compat. |
| GPUOptions.Builder | clearField (com.google.protobuf.Descriptors.FieldDescriptor ক্ষেত্র) |
| GPUOptions.Builder | clearForceGpu সামঞ্জস্যপূর্ণ () Force all tensors to be gpu_compatible. |
| GPUOptions.Builder | clearOneof (com.google.protobuf.Descriptors.OneofDescriptor oneof) |
| GPUOptions.Builder | clearPerProcessGpuMemoryFraction () Fraction of the available GPU memory to allocate for each process. |
| GPUOptions.Builder | PollingActiveDelayUsecs () In the event polling loop sleep this many microseconds between PollEvents calls, when the queue is not empty. |
| GPUOptions.Builder | ক্লিয়ারপোলিংইনঅ্যাক্টিভডেলেএমসেক্স () This field is deprecated and ignored. |
| GPUOptions.Builder | পরিষ্কার দৃশ্যমান ডিভাইস তালিকা () A comma-separated list of GPU ids that determines the 'visible' to 'virtual' mapping of GPU devices. |
| GPUOptions.Builder | ক্লোন () |
| স্ট্রিং | getAllocatorType () The type of GPU allocation strategy to use. |
| com.google.protobuf.ByteString | getAllocatorTypeBytes () The type of GPU allocation strategy to use. |
| বুলিয়ান | getAllowGrowth () If true, the allocator does not pre-allocate the entire specified GPU memory region, instead starting small and growing as needed. |
| জিপিইউ অপশন | |
| দীর্ঘ | getDeferredDeletionBytes () Delay deletion of up to this many bytes to reduce the number of interactions with gpu driver code. |
| চূড়ান্ত স্ট্যাটিক com.google.protobuf.Descriptors.Descriptor | |
| com.google.protobuf.Descriptors.Descriptor | |
| GPUOptions.পরীক্ষামূলক | পরীক্ষামূলক () 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 | গেট এক্সপেরিমেন্টাল বিল্ডার () 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. |
| বুলিয়ান | getForceGpu সামঞ্জস্যপূর্ণ () Force all tensors to be gpu_compatible. |
| দ্বিগুণ | getPerProcessGpuMemoryFraction () Fraction of the available GPU memory to allocate for each process. |
| int | getPollingActiveDelayUsecs () In the event polling loop sleep this many microseconds between PollEvents calls, when the queue is not empty. |
| int | getPollingInactiveDelayMsecs () This field is deprecated and ignored. |
| স্ট্রিং | 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. |
| বুলিয়ান | আছে পরীক্ষামূলক () Everything inside experimental is subject to change and is not subject to API stability guarantees in https://www.tensorflow.org/guide/version_compat. |
| চূড়ান্ত বুলিয়ান | |
| GPUOptions.Builder | মার্জ এক্সপেরিমেন্টাল ( GPUOptions. এক্সপেরিমেন্টাল মান) Everything inside experimental is subject to change and is not subject to API stability guarantees in https://www.tensorflow.org/guide/version_compat. |
| GPUOptions.Builder | mergeFrom (com.google.protobuf.অন্যান্য বার্তা পাঠান) |
| GPUOptions.Builder | mergeFrom (com.google.protobuf.CodedInputStream ইনপুট, com.google.protobuf.ExtensionRegistryLite এক্সটেনশন রেজিস্ট্রি) |
| চূড়ান্ত GPUOptions.Builder | একত্রিত করুন অজানাক্ষেত্র (com.google.protobuf.UnknownFieldSet অজানাক্ষেত্র) |
| GPUOptions.Builder | setAllocatorType (স্ট্রিং মান) The type of GPU allocation strategy to use. |
| GPUOptions.Builder | setAllocatorTypeBytes (com.google.protobuf.ByteString মান) The type of GPU allocation strategy to use. |
| GPUOptions.Builder | setAllowGrowth (বুলিয়ান মান) If true, the allocator does not pre-allocate the entire specified GPU memory region, instead starting small and growing as needed. |
| GPUOptions.Builder | setDeferredDeletionBytes (দীর্ঘ মান) Delay deletion of up to this many bytes to reduce the number of interactions with gpu driver code. |
| GPUOptions.Builder | পরীক্ষামূলক সেট করুন ( 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.Builder | পরীক্ষামূলক সেট করুন ( GPUOptions. পরীক্ষামূলক মান) Everything inside experimental is subject to change and is not subject to API stability guarantees in https://www.tensorflow.org/guide/version_compat. |
| GPUOptions.Builder | setField (com.google.protobuf.Descriptors.FieldDescriptor ক্ষেত্র, বস্তুর মান) |
| GPUOptions.Builder | setForceGpu কম্প্যাটিবল (বুলিয়ান মান) Force all tensors to be gpu_compatible. |
| GPUOptions.Builder | setPerProcessGpuMemoryFraction (ডবল মান) Fraction of the available GPU memory to allocate for each process. |
| GPUOptions.Builder | সেটপোলিংএক্টিভডেলেইউসেক্স (int মান) In the event polling loop sleep this many microseconds between PollEvents calls, when the queue is not empty. |
| GPUOptions.Builder | সেটপোলিংইনঅ্যাক্টিভডেলেএমসেক্স (int মান) This field is deprecated and ignored. |
| GPUOptions.Builder | setRepeatedField (com.google.protobuf.Descriptors.FieldDescriptor ক্ষেত্র, int সূচক, অবজেক্ট মান) |
| চূড়ান্ত GPUOptions.Builder | সেটUnknownFields (com.google.protobuf.UnknownFieldসেট অজানাফিল্ড) |
| GPUOptions.Builder | setVisibleDeviceList (স্ট্রিং মান) A comma-separated list of GPU ids that determines the 'visible' to 'virtual' mapping of GPU devices. |
| GPUOptions.Builder | setVisibleDeviceListBytes (com.google.protobuf.ByteString মান) A comma-separated list of GPU ids that determines the 'visible' to 'virtual' mapping of GPU devices. |
উত্তরাধিকারসূত্রে প্রাপ্ত পদ্ধতি
পাবলিক পদ্ধতি
সর্বজনীন GPUOptions.Builder addRepeatedField (com.google.protobuf.Descriptors.FieldDescriptor ক্ষেত্র, বস্তুর মান)
সর্বজনীন 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; সর্বজনীন 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; সর্বজনীন 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; সর্বজনীন 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; সর্বজনীন 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; সর্বজনীন 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; সর্বজনীন 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; সর্বজনীন GPUOptions.Builder clearPollingInactiveDelayMsecs ()
This field is deprecated and ignored.
int32 polling_inactive_delay_msecs = 7; সর্বজনীন 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; পাবলিক স্ট্রিং 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; সর্বজনীন 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; পাবলিক বুলিয়ান 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; সর্বজনীন দীর্ঘ 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; পাবলিক স্ট্যাটিক ফাইনাল com.google.protobuf.Descriptors.Descriptor getDescriptor ()
সর্বজনীন com.google.protobuf.Descriptors.Descriptor getDescriptorForType ()
সর্বজনীন GPUOptions. পরীক্ষামূলক 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; সর্বজনীন 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; সর্বজনীন 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; পাবলিক বুলিয়ান getForceGpu Compatible ()
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; পাবলিক ডাবল 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; পাবলিক 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; পাবলিক int getPollingInactiveDelayMsecs ()
This field is deprecated and ignored.
int32 polling_inactive_delay_msecs = 7; সর্বজনীন স্ট্রিং 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; সর্বজনীন 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; পাবলিক বুলিয়ান আছে এক্সপেরিমেন্টাল ()
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; সর্বজনীন চূড়ান্ত বুলিয়ান শুরু হয়েছে ()
সর্বজনীন GPUOptions.Builder mergeExperimental ( GPUOptions.পরীক্ষামূলক মান)
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; সর্বজনীন GPUOptions.Builder mergeFrom (com.google.protobuf.CodedInputStream ইনপুট, com.google.protobuf.ExtensionRegistryLite extensionRegistry)
নিক্ষেপ করে
| IO ব্যতিক্রম |
|---|
সর্বজনীন চূড়ান্ত GPUOptions.Builder mergeUnknownFields (com.google.protobuf.UnknownFieldSet unknownFields)
সর্বজনীন GPUOptions.Builder setAllocatorType (স্ট্রিং মান)
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; সর্বজনীন GPUOptions.Builder setAllocatorTypeBytes (com.google.protobuf.ByteString মান)
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; সর্বজনীন GPUOptions.Builder setAllowGrowth (বুলিয়ান মান)
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; সর্বজনীন GPUOptions.Builder setDeferredDeletionBytes (দীর্ঘ মান)
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; সর্বজনীন 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;সর্বজনীন GPUOptions.Builder setExperimental ( GPUOptions.পরীক্ষামূলক মান)
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; সর্বজনীন GPUOptions.Builder setField (com.google.protobuf.Descriptors.FieldDescriptor ক্ষেত্র, বস্তুর মান)
সর্বজনীন GPUOptions.Builder setForceGpuCompatible (বুলিয়ান মান)
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; সর্বজনীন GPUOptions.Builder setPerProcessGpuMemoryFraction (ডবল মান)
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; সর্বজনীন GPUOptions.Builder setPollingActiveDelayUsecs (int মান)
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; সর্বজনীন GPUOptions.Builder setPollingInactiveDelayMsecs (int মান)
This field is deprecated and ignored.
int32 polling_inactive_delay_msecs = 7; সর্বজনীন GPUOptions.Builder setRepeatedField (com.google.protobuf.Descriptors.FieldDescriptor ক্ষেত্র, int সূচক, অবজেক্ট মান)
সর্বজনীন চূড়ান্ত GPUOptions.Builder setUnknownFields (com.google.protobuf.UnknownFieldSet unknownFields)
সর্বজনীন GPUOptions.Builder setVisibleDeviceList (স্ট্রিং মান)
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; সর্বজনীন GPUOptions.Builder setVisibleDeviceListBytes (com.google.protobuf.ByteString মান)
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;