interface publique KernelDefOrBuilder
| Sous-classes indirectes connues |
Méthodes publiques
| abstrait KernelDef.AttrConstraint | getConstraint (index int) repeated .tensorflow.KernelDef.AttrConstraint constraint = 3; |
| abstrait entier | getConstraintCount () repeated .tensorflow.KernelDef.AttrConstraint constraint = 3; |
| Liste abstraite < KernelDef.AttrConstraint > | getConstraintList () repeated .tensorflow.KernelDef.AttrConstraint constraint = 3; |
| abstrait KernelDef.AttrConstraintOrBuilder | getConstraintOrBuilder (index int) repeated .tensorflow.KernelDef.AttrConstraint constraint = 3; |
| Liste abstraite <? étend KernelDef.AttrConstraintOrBuilder > | getConstraintOrBuilderList () repeated .tensorflow.KernelDef.AttrConstraint constraint = 3; |
| chaîne abstraite | getDeviceType () Type of device this kernel runs on. |
| résumé com.google.protobuf.ByteString | getDeviceTypeBytes () Type of device this kernel runs on. |
| chaîne abstraite | getHostMemoryArg (index int) Names of the Op's input_/output_args that reside in host memory instead of device memory. |
| résumé com.google.protobuf.ByteString | getHostMemoryArgBytes (index int) Names of the Op's input_/output_args that reside in host memory instead of device memory. |
| abstrait entier | getHostMemoryArgCount () Names of the Op's input_/output_args that reside in host memory instead of device memory. |
| Liste abstraite<String> | getHostMemoryArgList () Names of the Op's input_/output_args that reside in host memory instead of device memory. |
| chaîne abstraite | 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. |
| résumé 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. |
| chaîne abstraite | obtenirOp () Must match the name of an Op. |
| résumé com.google.protobuf.ByteString | getOpBytes () Must match the name of an Op. |
| abstrait entier | getPriorité () Prioritization of kernel amongst different devices. |
Méthodes publiques
public abstrait KernelDef.AttrConstraint getConstraint (index int)
repeated .tensorflow.KernelDef.AttrConstraint constraint = 3;
public abstrait int getConstraintCount ()
repeated .tensorflow.KernelDef.AttrConstraint constraint = 3;
Liste abstraite publique < KernelDef.AttrConstraint > getConstraintList ()
repeated .tensorflow.KernelDef.AttrConstraint constraint = 3;
public abstrait KernelDef.AttrConstraintOrBuilder getConstraintOrBuilder (index int)
repeated .tensorflow.KernelDef.AttrConstraint constraint = 3;
Liste des résumés publics<? étend KernelDef.AttrConstraintOrBuilder > getConstraintOrBuilderList ()
repeated .tensorflow.KernelDef.AttrConstraint constraint = 3;
chaîne abstraite publique getDeviceType ()
Type of device this kernel runs on.
string device_type = 2; résumé public com.google.protobuf.ByteString getDeviceTypeBytes ()
Type of device this kernel runs on.
string device_type = 2; chaîne abstraite publique getHostMemoryArg (index int)
Names of the Op's input_/output_args that reside in host memory instead of device memory.
repeated string host_memory_arg = 4; résumé public com.google.protobuf.ByteString getHostMemoryArgBytes (index int)
Names of the Op's input_/output_args that reside in host memory instead of device memory.
repeated string host_memory_arg = 4; public abstrait 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; liste abstraite publique<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; chaîne abstraite publique 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; résumé public 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; chaîne abstraite publique getOp ()
Must match the name of an Op.
string op = 1; résumé public com.google.protobuf.ByteString getOpBytes ()
Must match the name of an Op.
string op = 1; public abstrait 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;