public static final class
OptimizerOptions.Builder
Options passed to the graph optimizer
tensorflow.OptimizerOptions
Public Methods
OptimizerOptions.Builder |
addRepeatedField(com.google.protobuf.Descriptors.FieldDescriptor field, Object value)
|
OptimizerOptions |
build()
|
OptimizerOptions | |
OptimizerOptions.Builder |
clear()
|
OptimizerOptions.Builder |
clearDoCommonSubexpressionElimination()
If true, optimize the graph using common subexpression elimination. |
OptimizerOptions.Builder |
clearDoConstantFolding()
If true, perform constant folding optimization on the graph. |
OptimizerOptions.Builder |
clearDoFunctionInlining()
If true, perform function inlining on the graph. |
OptimizerOptions.Builder |
clearField(com.google.protobuf.Descriptors.FieldDescriptor field)
|
OptimizerOptions.Builder |
clearGlobalJitLevel()
.tensorflow.OptimizerOptions.GlobalJitLevel global_jit_level = 5;
|
OptimizerOptions.Builder |
clearMaxFoldedConstantInBytes()
Constant folding optimization replaces tensors whose values can be predetermined, with constant nodes. |
OptimizerOptions.Builder |
clearOneof(com.google.protobuf.Descriptors.OneofDescriptor oneof)
|
OptimizerOptions.Builder |
clearOptLevel()
Overall optimization level. |
OptimizerOptions.Builder |
clone()
|
OptimizerOptions | |
final static com.google.protobuf.Descriptors.Descriptor | |
com.google.protobuf.Descriptors.Descriptor | |
boolean |
getDoCommonSubexpressionElimination()
If true, optimize the graph using common subexpression elimination. |
boolean |
getDoConstantFolding()
If true, perform constant folding optimization on the graph. |
boolean |
getDoFunctionInlining()
If true, perform function inlining on the graph. |
OptimizerOptions.GlobalJitLevel |
getGlobalJitLevel()
.tensorflow.OptimizerOptions.GlobalJitLevel global_jit_level = 5;
|
int |
getGlobalJitLevelValue()
.tensorflow.OptimizerOptions.GlobalJitLevel global_jit_level = 5;
|
long |
getMaxFoldedConstantInBytes()
Constant folding optimization replaces tensors whose values can be predetermined, with constant nodes. |
OptimizerOptions.Level |
getOptLevel()
Overall optimization level. |
int |
getOptLevelValue()
Overall optimization level. |
final boolean | |
OptimizerOptions.Builder |
mergeFrom(com.google.protobuf.CodedInputStream input, com.google.protobuf.ExtensionRegistryLite extensionRegistry)
|
OptimizerOptions.Builder |
mergeFrom(com.google.protobuf.Message other)
|
final OptimizerOptions.Builder |
mergeUnknownFields(com.google.protobuf.UnknownFieldSet unknownFields)
|
OptimizerOptions.Builder |
setDoCommonSubexpressionElimination(boolean value)
If true, optimize the graph using common subexpression elimination. |
OptimizerOptions.Builder |
setDoConstantFolding(boolean value)
If true, perform constant folding optimization on the graph. |
OptimizerOptions.Builder |
setDoFunctionInlining(boolean value)
If true, perform function inlining on the graph. |
OptimizerOptions.Builder |
setField(com.google.protobuf.Descriptors.FieldDescriptor field, Object value)
|
OptimizerOptions.Builder |
setGlobalJitLevel(OptimizerOptions.GlobalJitLevel value)
.tensorflow.OptimizerOptions.GlobalJitLevel global_jit_level = 5;
|
OptimizerOptions.Builder |
setGlobalJitLevelValue(int value)
.tensorflow.OptimizerOptions.GlobalJitLevel global_jit_level = 5;
|
OptimizerOptions.Builder |
setMaxFoldedConstantInBytes(long value)
Constant folding optimization replaces tensors whose values can be predetermined, with constant nodes. |
OptimizerOptions.Builder | |
OptimizerOptions.Builder |
setOptLevelValue(int value)
Overall optimization level. |
OptimizerOptions.Builder |
setRepeatedField(com.google.protobuf.Descriptors.FieldDescriptor field, int index, Object value)
|
final OptimizerOptions.Builder |
setUnknownFields(com.google.protobuf.UnknownFieldSet unknownFields)
|
Inherited Methods
Public Methods
public OptimizerOptions.Builder addRepeatedField (com.google.protobuf.Descriptors.FieldDescriptor field, Object value)
public OptimizerOptions.Builder clearDoCommonSubexpressionElimination ()
If true, optimize the graph using common subexpression elimination.
bool do_common_subexpression_elimination = 1;
public OptimizerOptions.Builder clearDoConstantFolding ()
If true, perform constant folding optimization on the graph.
bool do_constant_folding = 2;
public OptimizerOptions.Builder clearDoFunctionInlining ()
If true, perform function inlining on the graph.
bool do_function_inlining = 4;
public OptimizerOptions.Builder clearGlobalJitLevel ()
.tensorflow.OptimizerOptions.GlobalJitLevel global_jit_level = 5;
public OptimizerOptions.Builder clearMaxFoldedConstantInBytes ()
Constant folding optimization replaces tensors whose values can be predetermined, with constant nodes. To avoid inserting too large constants, the size of each constant created can be limited. If this value is zero, a default limit of 10 MiB will be applied. If constant folding optimization is disabled, this value is ignored.
int64 max_folded_constant_in_bytes = 6;
public OptimizerOptions.Builder clearOptLevel ()
Overall optimization level. The actual optimizations applied will be the logical OR of the flags that this level implies and any flags already set.
.tensorflow.OptimizerOptions.Level opt_level = 3;
public static final com.google.protobuf.Descriptors.Descriptor getDescriptor ()
public com.google.protobuf.Descriptors.Descriptor getDescriptorForType ()
public boolean getDoCommonSubexpressionElimination ()
If true, optimize the graph using common subexpression elimination.
bool do_common_subexpression_elimination = 1;
public boolean getDoConstantFolding ()
If true, perform constant folding optimization on the graph.
bool do_constant_folding = 2;
public boolean getDoFunctionInlining ()
If true, perform function inlining on the graph.
bool do_function_inlining = 4;
public OptimizerOptions.GlobalJitLevel getGlobalJitLevel ()
.tensorflow.OptimizerOptions.GlobalJitLevel global_jit_level = 5;
public int getGlobalJitLevelValue ()
.tensorflow.OptimizerOptions.GlobalJitLevel global_jit_level = 5;
public long getMaxFoldedConstantInBytes ()
Constant folding optimization replaces tensors whose values can be predetermined, with constant nodes. To avoid inserting too large constants, the size of each constant created can be limited. If this value is zero, a default limit of 10 MiB will be applied. If constant folding optimization is disabled, this value is ignored.
int64 max_folded_constant_in_bytes = 6;
public OptimizerOptions.Level getOptLevel ()
Overall optimization level. The actual optimizations applied will be the logical OR of the flags that this level implies and any flags already set.
.tensorflow.OptimizerOptions.Level opt_level = 3;
public int getOptLevelValue ()
Overall optimization level. The actual optimizations applied will be the logical OR of the flags that this level implies and any flags already set.
.tensorflow.OptimizerOptions.Level opt_level = 3;
public final boolean isInitialized ()
public OptimizerOptions.Builder mergeFrom (com.google.protobuf.CodedInputStream input, com.google.protobuf.ExtensionRegistryLite extensionRegistry)
Throws
IOException |
---|
public final OptimizerOptions.Builder mergeUnknownFields (com.google.protobuf.UnknownFieldSet unknownFields)
public OptimizerOptions.Builder setDoCommonSubexpressionElimination (boolean value)
If true, optimize the graph using common subexpression elimination.
bool do_common_subexpression_elimination = 1;
public OptimizerOptions.Builder setDoConstantFolding (boolean value)
If true, perform constant folding optimization on the graph.
bool do_constant_folding = 2;
public OptimizerOptions.Builder setDoFunctionInlining (boolean value)
If true, perform function inlining on the graph.
bool do_function_inlining = 4;
public OptimizerOptions.Builder setField (com.google.protobuf.Descriptors.FieldDescriptor field, Object value)
public OptimizerOptions.Builder setGlobalJitLevel (OptimizerOptions.GlobalJitLevel value)
.tensorflow.OptimizerOptions.GlobalJitLevel global_jit_level = 5;
public OptimizerOptions.Builder setGlobalJitLevelValue (int value)
.tensorflow.OptimizerOptions.GlobalJitLevel global_jit_level = 5;
public OptimizerOptions.Builder setMaxFoldedConstantInBytes (long value)
Constant folding optimization replaces tensors whose values can be predetermined, with constant nodes. To avoid inserting too large constants, the size of each constant created can be limited. If this value is zero, a default limit of 10 MiB will be applied. If constant folding optimization is disabled, this value is ignored.
int64 max_folded_constant_in_bytes = 6;
public OptimizerOptions.Builder setOptLevel (OptimizerOptions.Level value)
Overall optimization level. The actual optimizations applied will be the logical OR of the flags that this level implies and any flags already set.
.tensorflow.OptimizerOptions.Level opt_level = 3;
public OptimizerOptions.Builder setOptLevelValue (int value)
Overall optimization level. The actual optimizations applied will be the logical OR of the flags that this level implies and any flags already set.
.tensorflow.OptimizerOptions.Level opt_level = 3;