KernelDefOrBuilder

공용 인터페이스 KernelDefOrBuilder
알려진 간접 하위 클래스

공개 방법

추상 KernelDef.AttrConstraint
getConstraint (정수 인덱스)
repeated .tensorflow.KernelDef.AttrConstraint constraint = 3;
추상 정수
getConstraintCount ()
repeated .tensorflow.KernelDef.AttrConstraint constraint = 3;
추상 목록< KernelDef.AttrConstraint >
getConstraintList ()
repeated .tensorflow.KernelDef.AttrConstraint constraint = 3;
추상 KernelDef.AttrConstraintOrBuilder
getConstraintOrBuilder (정수 인덱스)
repeated .tensorflow.KernelDef.AttrConstraint constraint = 3;
추상 목록<? KernelDef.AttrConstraintOrBuilder 확장 >
getConstraintOrBuilderList ()
repeated .tensorflow.KernelDef.AttrConstraint constraint = 3;
추상 문자열
getDeviceType ()
 Type of device this kernel runs on.
추상 com.google.protobuf.ByteString
getDeviceTypeBytes ()
 Type of device this kernel runs on.
추상 문자열
getHostMemoryArg (정수 인덱스)
 Names of the Op's input_/output_args that reside in host memory
 instead of device memory.
추상 com.google.protobuf.ByteString
getHostMemoryArgBytes (정수 인덱스)
 Names of the Op's input_/output_args that reside in host memory
 instead of device memory.
추상 정수
getHostMemoryArgCount ()
 Names of the Op's input_/output_args that reside in host memory
 instead of device memory.
추상 목록<문자열>
getHostMemoryArgList ()
 Names of the Op's input_/output_args that reside in host memory
 instead of device memory.
추상 문자열
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.
추상 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.
추상 문자열
getOp ()
 Must match the name of an Op.
추상 com.google.protobuf.ByteString
getOpBytes ()
 Must match the name of an Op.
추상 정수
우선순위 ()
 Prioritization of kernel amongst different devices.

공개 방법

공개 추상 KernelDef.AttrConstraint getConstraint (int 인덱스)

repeated .tensorflow.KernelDef.AttrConstraint constraint = 3;

공개 추상 int getConstraintCount ()

repeated .tensorflow.KernelDef.AttrConstraint constraint = 3;

공개 추상 목록 < KernelDef.AttrConstraint > getConstraintList ()

repeated .tensorflow.KernelDef.AttrConstraint constraint = 3;

공개 추상 KernelDef.AttrConstraintOrBuilder getConstraintOrBuilder (int 인덱스)

repeated .tensorflow.KernelDef.AttrConstraint constraint = 3;

공개 요약 목록<? KernelDef.AttrConstraintOrBuilder > getConstraintOrBuilderList () 를 확장합니다.

repeated .tensorflow.KernelDef.AttrConstraint constraint = 3;

공개 추상 문자열 getDeviceType () 

 Type of device this kernel runs on.
string device_type = 2;

공개 추상 com.google.protobuf.ByteString getDeviceTypeBytes () 

 Type of device this kernel runs on.
string device_type = 2;

공개 추상 문자열 getHostMemoryArg (int 인덱스) 

 Names of the Op's input_/output_args that reside in host memory
 instead of device memory.
repeated string host_memory_arg = 4;

공개 추상 com.google.protobuf.ByteString getHostMemoryArgBytes (int 인덱스) 

 Names of the Op's input_/output_args that reside in host memory
 instead of device memory.
repeated string host_memory_arg = 4;

공개 추상 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;

공개 추상 List<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;

공개 추상 문자열 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;

공개 추상 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;

공개 추상 문자열 getOp () 

 Must match the name of an Op.
string op = 1;

공개 추상 com.google.protobuf.ByteString getOpBytes () 

 Must match the name of an Op.
string op = 1;

공개 추상 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;