Package org.tensorflow.framework
Interface OpDefOrBuilder
-
- All Superinterfaces:
org.nd4j.shade.protobuf.MessageLiteOrBuilder,org.nd4j.shade.protobuf.MessageOrBuilder
- All Known Implementing Classes:
OpDef,OpDef.Builder
public interface OpDefOrBuilder extends org.nd4j.shade.protobuf.MessageOrBuilder
-
-
Method Summary
All Methods Instance Methods Abstract Methods Modifier and Type Method Description booleangetAllowsUninitializedInput()By default, all inputs to an Op must be initialized Tensors.OpDef.AttrDefgetAttr(int index)repeated .tensorflow.OpDef.AttrDef attr = 4;intgetAttrCount()repeated .tensorflow.OpDef.AttrDef attr = 4;List<OpDef.AttrDef>getAttrList()repeated .tensorflow.OpDef.AttrDef attr = 4;OpDef.AttrDefOrBuildergetAttrOrBuilder(int index)repeated .tensorflow.OpDef.AttrDef attr = 4;List<? extends OpDef.AttrDefOrBuilder>getAttrOrBuilderList()repeated .tensorflow.OpDef.AttrDef attr = 4;OpDeprecationgetDeprecation()Optional deprecation based on GraphDef versions.OpDeprecationOrBuildergetDeprecationOrBuilder()Optional deprecation based on GraphDef versions.StringgetDescription()Additional, longer human-readable description of what the Op does.org.nd4j.shade.protobuf.ByteStringgetDescriptionBytes()Additional, longer human-readable description of what the Op does.OpDef.ArgDefgetInputArg(int index)Description of the input(s).intgetInputArgCount()Description of the input(s).List<OpDef.ArgDef>getInputArgList()Description of the input(s).OpDef.ArgDefOrBuildergetInputArgOrBuilder(int index)Description of the input(s).List<? extends OpDef.ArgDefOrBuilder>getInputArgOrBuilderList()Description of the input(s).booleangetIsAggregate()If is_aggregate is true, then this operation accepts N >= 2 inputs and produces 1 output all of the same type.booleangetIsCommutative()True if the operation is commutative ("op(a,b) == op(b,a)" for all inputs)booleangetIsStateful()Ops are marked as stateful if their behavior depends on some state beyond their input tensors (e.g.StringgetName()Op names starting with an underscore are reserved for internal use.org.nd4j.shade.protobuf.ByteStringgetNameBytes()Op names starting with an underscore are reserved for internal use.OpDef.ArgDefgetOutputArg(int index)Description of the output(s).intgetOutputArgCount()Description of the output(s).List<OpDef.ArgDef>getOutputArgList()Description of the output(s).OpDef.ArgDefOrBuildergetOutputArgOrBuilder(int index)Description of the output(s).List<? extends OpDef.ArgDefOrBuilder>getOutputArgOrBuilderList()Description of the output(s).StringgetSummary()One-line human-readable description of what the Op does.org.nd4j.shade.protobuf.ByteStringgetSummaryBytes()One-line human-readable description of what the Op does.booleanhasDeprecation()Optional deprecation based on GraphDef versions.-
Methods inherited from interface org.nd4j.shade.protobuf.MessageOrBuilder
findInitializationErrors, getAllFields, getDefaultInstanceForType, getDescriptorForType, getField, getInitializationErrorString, getOneofFieldDescriptor, getRepeatedField, getRepeatedFieldCount, getUnknownFields, hasField, hasOneof
-
-
-
-
Method Detail
-
getName
String getName()
Op names starting with an underscore are reserved for internal use. Names should be CamelCase and match the regexp "[A-Z][a-zA-Z0-9_]*".
string name = 1;- Returns:
- The name.
-
getNameBytes
org.nd4j.shade.protobuf.ByteString getNameBytes()
Op names starting with an underscore are reserved for internal use. Names should be CamelCase and match the regexp "[A-Z][a-zA-Z0-9_]*".
string name = 1;- Returns:
- The bytes for name.
-
getInputArgList
List<OpDef.ArgDef> getInputArgList()
Description of the input(s).
repeated .tensorflow.OpDef.ArgDef input_arg = 2;
-
getInputArg
OpDef.ArgDef getInputArg(int index)
Description of the input(s).
repeated .tensorflow.OpDef.ArgDef input_arg = 2;
-
getInputArgCount
int getInputArgCount()
Description of the input(s).
repeated .tensorflow.OpDef.ArgDef input_arg = 2;
-
getInputArgOrBuilderList
List<? extends OpDef.ArgDefOrBuilder> getInputArgOrBuilderList()
Description of the input(s).
repeated .tensorflow.OpDef.ArgDef input_arg = 2;
-
getInputArgOrBuilder
OpDef.ArgDefOrBuilder getInputArgOrBuilder(int index)
Description of the input(s).
repeated .tensorflow.OpDef.ArgDef input_arg = 2;
-
getOutputArgList
List<OpDef.ArgDef> getOutputArgList()
Description of the output(s).
repeated .tensorflow.OpDef.ArgDef output_arg = 3;
-
getOutputArg
OpDef.ArgDef getOutputArg(int index)
Description of the output(s).
repeated .tensorflow.OpDef.ArgDef output_arg = 3;
-
getOutputArgCount
int getOutputArgCount()
Description of the output(s).
repeated .tensorflow.OpDef.ArgDef output_arg = 3;
-
getOutputArgOrBuilderList
List<? extends OpDef.ArgDefOrBuilder> getOutputArgOrBuilderList()
Description of the output(s).
repeated .tensorflow.OpDef.ArgDef output_arg = 3;
-
getOutputArgOrBuilder
OpDef.ArgDefOrBuilder getOutputArgOrBuilder(int index)
Description of the output(s).
repeated .tensorflow.OpDef.ArgDef output_arg = 3;
-
getAttrList
List<OpDef.AttrDef> getAttrList()
repeated .tensorflow.OpDef.AttrDef attr = 4;
-
getAttr
OpDef.AttrDef getAttr(int index)
repeated .tensorflow.OpDef.AttrDef attr = 4;
-
getAttrCount
int getAttrCount()
repeated .tensorflow.OpDef.AttrDef attr = 4;
-
getAttrOrBuilderList
List<? extends OpDef.AttrDefOrBuilder> getAttrOrBuilderList()
repeated .tensorflow.OpDef.AttrDef attr = 4;
-
getAttrOrBuilder
OpDef.AttrDefOrBuilder getAttrOrBuilder(int index)
repeated .tensorflow.OpDef.AttrDef attr = 4;
-
hasDeprecation
boolean hasDeprecation()
Optional deprecation based on GraphDef versions.
.tensorflow.OpDeprecation deprecation = 8;- Returns:
- Whether the deprecation field is set.
-
getDeprecation
OpDeprecation getDeprecation()
Optional deprecation based on GraphDef versions.
.tensorflow.OpDeprecation deprecation = 8;- Returns:
- The deprecation.
-
getDeprecationOrBuilder
OpDeprecationOrBuilder getDeprecationOrBuilder()
Optional deprecation based on GraphDef versions.
.tensorflow.OpDeprecation deprecation = 8;
-
getSummary
String getSummary()
One-line human-readable description of what the Op does.
string summary = 5;- Returns:
- The summary.
-
getSummaryBytes
org.nd4j.shade.protobuf.ByteString getSummaryBytes()
One-line human-readable description of what the Op does.
string summary = 5;- Returns:
- The bytes for summary.
-
getDescription
String getDescription()
Additional, longer human-readable description of what the Op does.
string description = 6;- Returns:
- The description.
-
getDescriptionBytes
org.nd4j.shade.protobuf.ByteString getDescriptionBytes()
Additional, longer human-readable description of what the Op does.
string description = 6;- Returns:
- The bytes for description.
-
getIsCommutative
boolean getIsCommutative()
True if the operation is commutative ("op(a,b) == op(b,a)" for all inputs)bool is_commutative = 18;- Returns:
- The isCommutative.
-
getIsAggregate
boolean getIsAggregate()
If is_aggregate is true, then this operation accepts N >= 2 inputs and produces 1 output all of the same type. Should be associative and commutative, and produce output with the same shape as the input. The optimizer may replace an aggregate op taking input from multiple devices with a tree of aggregate ops that aggregate locally within each device (and possibly within groups of nearby devices) before communicating. TODO(josh11b): Implement that optimization.
bool is_aggregate = 16;- Returns:
- The isAggregate.
-
getIsStateful
boolean getIsStateful()
Ops are marked as stateful if their behavior depends on some state beyond their input tensors (e.g. variable reading op) or if they have a side-effect (e.g. printing or asserting ops). Equivalently, stateless ops must always produce the same output for the same input and have no side-effects. By default Ops may be moved between devices. Stateful ops should either not be moved, or should only be moved if that state can also be moved (e.g. via some sort of save / restore). Stateful ops are guaranteed to never be optimized away by Common Subexpression Elimination (CSE).
bool is_stateful = 17;- Returns:
- The isStateful.
-
getAllowsUninitializedInput
boolean getAllowsUninitializedInput()
By default, all inputs to an Op must be initialized Tensors. Ops that may initialize tensors for the first time should set this field to true, to allow the Op to take an uninitialized Tensor as input.
bool allows_uninitialized_input = 19;- Returns:
- The allowsUninitializedInput.
-
-