KernelDefOrBuilder

antarmuka publik KernelDefOrBuilder

Subkelas Tidak Langsung yang Diketahui KernelDef , KernelDef.Builder KernelDef Tipe protobuf tensorflow.KernelDef KernelDef.Builder Tipe protobuf tensorflow.KernelDef

Metode Publik

abstrak KernelDef.AttrConstraint	getConstraint (indeks int) repeated .tensorflow.KernelDef.AttrConstraint constraint = 3;
abstrak ke dalam	dapatkanConstraintCount () repeated .tensorflow.KernelDef.AttrConstraint constraint = 3;
Daftar abstrak< KernelDef.AttrConstraint >	dapatkanConstraintList () repeated .tensorflow.KernelDef.AttrConstraint constraint = 3;
abstrak KernelDef.AttrConstraintOrBuilder	getConstraintOrBuilder (indeks int) repeated .tensorflow.KernelDef.AttrConstraint constraint = 3;
Daftar abstrak<? memperluas KernelDef.AttrConstraintOrBuilder >	dapatkanConstraintOrBuilderList () repeated .tensorflow.KernelDef.AttrConstraint constraint = 3;
Tali abstrak	dapatkanDeviceType () Type of device this kernel runs on.
abstrak com.google.protobuf.ByteString	dapatkanDeviceTypeBytes () Type of device this kernel runs on.
Tali abstrak	getHostMemoryArg (indeks int) Names of the Op's input_/output_args that reside in host memory instead of device memory.
abstrak com.google.protobuf.ByteString	getHostMemoryArgBytes (indeks int) Names of the Op's input_/output_args that reside in host memory instead of device memory.
abstrak ke dalam	dapatkanHostMemoryArgCount () Names of the Op's input_/output_args that reside in host memory instead of device memory.
Daftar abstrak<String>	dapatkanHostMemoryArgList () Names of the Op's input_/output_args that reside in host memory instead of device memory.
Tali abstrak	dapatkanLabel () This allows experimental kernels to be registered for an op that won't be used unless the user specifies a "_kernel" attr with value matching this.
abstrak com.google.protobuf.ByteString	dapatkanLabelBytes () This allows experimental kernels to be registered for an op that won't be used unless the user specifies a "_kernel" attr with value matching this.
Tali abstrak	dapatkanOp () Must match the name of an Op.
abstrak com.google.protobuf.ByteString	dapatkanOpBytes () Must match the name of an Op.
abstrak ke dalam	dapatkan Prioritas () Prioritization of kernel amongst different devices.

Metode Publik

abstrak publik KernelDef.AttrConstraint getConstraint (indeks int)

repeated .tensorflow.KernelDef.AttrConstraint constraint = 3;

abstrak publik int getConstraintCount ()

repeated .tensorflow.KernelDef.AttrConstraint constraint = 3;

Daftar abstrak publik< KernelDef.AttrConstraint > getConstraintList ()

repeated .tensorflow.KernelDef.AttrConstraint constraint = 3;

abstrak publik KernelDef.AttrConstraintOrBuilder getConstraintOrBuilder (int indeks)

repeated .tensorflow.KernelDef.AttrConstraint constraint = 3;

Daftar abstrak publik<? memperluas KernelDef.AttrConstraintOrBuilder > getConstraintOrBuilderList ()

repeated .tensorflow.KernelDef.AttrConstraint constraint = 3;

String abstrak publik getDeviceType ()

 Type of device this kernel runs on.
 

string device_type = 2;

abstrak publik com.google.protobuf.ByteString getDeviceTypeBytes ()

 Type of device this kernel runs on.
 

string device_type = 2;

String abstrak publik getHostMemoryArg (int indeks)

 Names of the Op's input_/output_args that reside in host memory
 instead of device memory.
 

repeated string host_memory_arg = 4;

abstrak publik com.google.protobuf.ByteString getHostMemoryArgBytes (indeks int)

 Names of the Op's input_/output_args that reside in host memory
 instead of device memory.
 

repeated string host_memory_arg = 4;

abstrak publik int getHostMemoryArgCount ()

 Names of the Op's input_/output_args that reside in host memory
 instead of device memory.
 

repeated string host_memory_arg = 4;

Daftar abstrak publik<String> getHostMemoryArgList ()

 Names of the Op's input_/output_args that reside in host memory
 instead of device memory.
 

repeated string host_memory_arg = 4;

abstrak publik String getLabel ()

 This allows experimental kernels to be registered for an op that
 won't be used unless the user specifies a "_kernel" attr with
 value matching this.
 

string label = 5;

abstrak publik com.google.protobuf.ByteString getLabelBytes ()

 This allows experimental kernels to be registered for an op that
 won't be used unless the user specifies a "_kernel" attr with
 value matching this.
 

string label = 5;

abstrak publik String getOp ()

 Must match the name of an Op.
 

string op = 1;

abstrak publik com.google.protobuf.ByteString getOpBytes ()

 Must match the name of an Op.
 

string op = 1;

abstrak publik int getPriority ()

 Prioritization of kernel amongst different devices. By default we assume
 priority is 0. The higher the priority the better. By default (i.e. if
 this is not set), we prefer GPU kernels over CPU.
 

int32 priority = 6;