सार्वजनिक स्थैतिक अंतिम वर्ग RewriterConfig.Builder
Graph rewriting is experimental and subject to change, not covered by any API stability guarantees.
tensorflow.RewriterConfig सार्वजनिक तरीके
| रीराइटरकॉन्फिग.बिल्डर | addAllCustomOptimizers (Iterable<? RewriterConfig.CustomGraphOptimizer > मान बढ़ाता है) list of CustomGraphOptimizers to apply. |
| रीराइटरकॉन्फिग.बिल्डर | addAllOptimizers (Iterable<String> मान) If non-empty, will use this as an alternative way to specify a list of optimizations to turn on and the order of the optimizations (replacing the meta-optimizer). |
| रीराइटरकॉन्फिग.बिल्डर | ऐडकस्टमऑप्टिमाइज़र ( RewriterConfig.CustomGraphOptimizer.Builder BuilderForValue) list of CustomGraphOptimizers to apply. |
| रीराइटरकॉन्फिग.बिल्डर | ऐडकस्टमऑप्टिमाइजर्स (इंट इंडेक्स, रीराइटरकॉन्फिग.कस्टमग्राफऑप्टिमाइजर.बिल्डर बिल्डरफॉरवैल्यू) list of CustomGraphOptimizers to apply. |
| रीराइटरकॉन्फिग.बिल्डर | ऐडकस्टमऑप्टिमाइज़र ( RewriterConfig.CustomGraph ऑप्टिमाइज़र मान) list of CustomGraphOptimizers to apply. |
| रीराइटरकॉन्फिग.बिल्डर | ऐडकस्टमऑप्टिमाइजर्स (इंट इंडेक्स, रीराइटरकॉन्फिग.कस्टमग्राफऑप्टिमाइजर वैल्यू) list of CustomGraphOptimizers to apply. |
| रीराइटरकॉन्फिग.कस्टमग्राफऑप्टिमाइज़र.बिल्डर | addCustomOptimizersBuilder (int अनुक्रमणिका) list of CustomGraphOptimizers to apply. |
| रीराइटरकॉन्फिग.कस्टमग्राफऑप्टिमाइज़र.बिल्डर | ऐडकस्टमऑप्टिमाइज़रबिल्डर () list of CustomGraphOptimizers to apply. |
| रीराइटरकॉन्फिग.बिल्डर | ऐडऑप्टिमाइज़र (स्ट्रिंग मान) If non-empty, will use this as an alternative way to specify a list of optimizations to turn on and the order of the optimizations (replacing the meta-optimizer). |
| रीराइटरकॉन्फिग.बिल्डर | ऐडऑप्टिमाइज़रबाइट्स (com.google.protobuf.ByteString मान) If non-empty, will use this as an alternative way to specify a list of optimizations to turn on and the order of the optimizations (replacing the meta-optimizer). |
| रीराइटरकॉन्फिग.बिल्डर | addRepeatedField (com.google.protobuf.Descriptors.FieldDescriptor फ़ील्ड, ऑब्जेक्ट मान) |
| रीराइटर कॉन्फिग | निर्माण () |
| रीराइटर कॉन्फिग | बिल्डआंशिक () |
| रीराइटरकॉन्फिग.बिल्डर | स्पष्ट () |
| रीराइटरकॉन्फिग.बिल्डर | स्पष्ट अंकगणित अनुकूलन () Arithmetic optimizations (default is ON) e.g. |
| रीराइटरकॉन्फिग.बिल्डर | क्लियरऑटोमिक्स्ड प्रिसिजन () Optimize data types for CUDA (default is OFF). |
| रीराइटरकॉन्फिग.बिल्डर | ClearAutoMixedPrecisionMkl () Optimize data types for MKL (default is OFF). |
| रीराइटरकॉन्फिग.बिल्डर | क्लियरऑटोपैरेलल () Configures AutoParallel optimization passes either through the meta-optimizer or when manually specified through the optimizers field. |
| रीराइटरकॉन्फिग.बिल्डर | क्लियरकॉमनसबग्राफएलिमिनेशन () Common subgraph elimination (default is ON) e.g. |
| रीराइटरकॉन्फिग.बिल्डर | क्लियरकॉन्स्टेंटफोल्डिंग () Fold constants (default is ON) Statically infer the value of tensors when possible, and materialize the result using constants. |
| रीराइटरकॉन्फिग.बिल्डर | ClearCpuLayoutConversion () CPU Conversion settings between NHCW and NCHW. |
| रीराइटरकॉन्फिग.बिल्डर | क्लियरकस्टमऑप्टिमाइज़र () list of CustomGraphOptimizers to apply. |
| रीराइटरकॉन्फिग.बिल्डर | क्लियरडिबगस्ट्रिपर () Strips debug-related nodes from the graph (off by default). |
| रीराइटरकॉन्फिग.बिल्डर | स्पष्टनिर्भरता अनुकूलन () Control dependency optimizations (default is ON). |
| रीराइटरकॉन्फिग.बिल्डर | क्लियरडिसेबलमेटाऑप्टिमाइज़र () Disable the entire meta optimizer (off by default). |
| रीराइटरकॉन्फिग.बिल्डर | ClearDisableModelPruning () If true, don't remove unnecessary ops from the graph bool disable_model_pruning = 2; |
| रीराइटरकॉन्फिग.बिल्डर | क्लियरएक्सपेरिमेंटलडिसएबलकंप्रेस्डटेन्सरऑप्टिमाइज़ेशन () Disable optimizations that assume compressed tensors. |
| रीराइटरकॉन्फिग.बिल्डर | क्लियरफ़ेलऑनऑप्टिमाइज़रत्रुटि () If true, any optimization pass failing will cause the MetaOptimizer to stop with an error. |
| रीराइटरकॉन्फिग.बिल्डर | क्लियरफ़ील्ड (com.google.protobuf.Descriptors.FieldDescriptor फ़ील्ड) |
| रीराइटरकॉन्फिग.बिल्डर | क्लियरफ़ंक्शनऑप्टिमाइज़ेशन () Function optimizations (default is ON). |
| रीराइटरकॉन्फिग.बिल्डर | स्पष्टकार्यान्वयन चयनकर्ता () Enable the swap of kernel implementations based on the device placement (default is ON). |
| रीराइटरकॉन्फिग.बिल्डर | क्लियरइंटरऑप्टिमाइज़रवेरिफायरकॉन्फिग () VerifierConfig specifying the verifiers to be run after every optimizer. |
| रीराइटरकॉन्फिग.बिल्डर | क्लियरलेआउटऑप्टिमाइज़र () Optimize tensor layouts (default is ON) e.g. |
| रीराइटरकॉन्फिग.बिल्डर | क्लियरलूपऑप्टिमाइज़ेशन () Loop optimizations (default is ON). |
| रीराइटरकॉन्फिग.बिल्डर | क्लियरमेमोरीऑप्टिमाइज़ेशन () Configures memory optimization passes through the meta-optimizer. |
| रीराइटरकॉन्फिग.बिल्डर | क्लियरमेमोरीऑप्टिमाइज़रटार्गेटनोडनेमस्कोप () A node name scope for node names which are valid outputs of recomputations. |
| रीराइटरकॉन्फिग.बिल्डर | क्लियरमेटाऑप्टिमाइज़रइटरेशन () Controls how many times we run the optimizers in meta optimizer (default is once). |
| रीराइटरकॉन्फिग.बिल्डर | क्लियरमेटाऑप्टिमाइज़रटाइमआउटएमएस () Maximum number of milliseconds to spend optimizing a single graph before timing out. |
| रीराइटरकॉन्फिग.बिल्डर | क्लियरमिनग्राफनोड्स () The minimum number of nodes in a graph to optimizer. |
| रीराइटरकॉन्फिग.बिल्डर | ClearOneof (com.google.protobuf.Descriptors.OneofDescriptor oneof) |
| रीराइटरकॉन्फिग.बिल्डर | क्लियरऑप्टिमाइजर्स () If non-empty, will use this as an alternative way to specify a list of optimizations to turn on and the order of the optimizations (replacing the meta-optimizer). |
| रीराइटरकॉन्फिग.बिल्डर | क्लियरपिनटूहोस्टऑप्टिमाइज़ेशन () Force small ops onto the CPU (default is OFF). |
| रीराइटरकॉन्फिग.बिल्डर | क्लियरपोस्टऑप्टिमाइजेशनवेरिफायरकॉन्फिग () VerifierConfig specifying the verifiers to be run at the end, after all optimizers have run. |
| रीराइटरकॉन्फिग.बिल्डर | क्लीयररीमैपिंग () Remapping (default is ON) Remap subgraphs onto more efficient implementations. |
| रीराइटरकॉन्फिग.बिल्डर | क्लियरस्कोप्डएलोकेटरऑप्टिमाइज़ेशन () Try to allocate some independent Op outputs contiguously in order to merge or eliminate downstream Ops (off by default). |
| रीराइटरकॉन्फिग.बिल्डर | ClearScopedAllocatorOpts () .tensorflow.ScopedAllocatorOptions scoped_allocator_opts = 16; |
| रीराइटरकॉन्फिग.बिल्डर | क्लियरशेपऑप्टिमाइज़ेशन () Shape optimizations (default is ON) Simplify computations made on shapes. |
| रीराइटरकॉन्फिग.बिल्डर | क्लोन () |
| रीराइटरकॉन्फिग.टॉगल | अंकगणित अनुकूलन प्राप्त करें () Arithmetic optimizations (default is ON) e.g. |
| int यहाँ | getArithmeticOptimizationValue () Arithmetic optimizations (default is ON) e.g. |
| रीराइटरकॉन्फिग.टॉगल | getAutoMixedPrecision () Optimize data types for CUDA (default is OFF). |
| रीराइटरकॉन्फिग.टॉगल | getAutoMixedPrecisionMkl () Optimize data types for MKL (default is OFF). |
| int यहाँ | getAutoMixedPrecisionMklValue () Optimize data types for MKL (default is OFF). |
| int यहाँ | getAutoMixedPrecisionValue () Optimize data types for CUDA (default is OFF). |
| ऑटोसमानान्तर विकल्प | ऑटोपैरेलल प्राप्त करें () Configures AutoParallel optimization passes either through the meta-optimizer or when manually specified through the optimizers field. |
| ऑटोपैरेललऑप्शंस.बिल्डर | getAutoParallelBuilder () Configures AutoParallel optimization passes either through the meta-optimizer or when manually specified through the optimizers field. |
| ऑटोपैरेललऑप्शनऑरबिल्डर | getAutoParallelOrBuilder () Configures AutoParallel optimization passes either through the meta-optimizer or when manually specified through the optimizers field. |
| रीराइटरकॉन्फिग.टॉगल | getCommonSubgraphElination () Common subgraph elimination (default is ON) e.g. |
| int यहाँ | getCommonSubgraphElinationValue () Common subgraph elimination (default is ON) e.g. |
| रीराइटरकॉन्फिग.टॉगल | getConstantFolding () Fold constants (default is ON) Statically infer the value of tensors when possible, and materialize the result using constants. |
| int यहाँ | getConstantFoldingValue () Fold constants (default is ON) Statically infer the value of tensors when possible, and materialize the result using constants. |
| RewriterConfig.CpuLayout | getCpuLayoutConversion () CPU Conversion settings between NHCW and NCHW. |
| int यहाँ | getCpuLayoutConversionValue () CPU Conversion settings between NHCW and NCHW. |
| रीराइटरकॉन्फिग.कस्टमग्राफऑप्टिमाइज़र | getCustomOptimizers (int अनुक्रमणिका) list of CustomGraphOptimizers to apply. |
| रीराइटरकॉन्फिग.कस्टमग्राफऑप्टिमाइज़र.बिल्डर | getCustomOptimizersBuilder (int अनुक्रमणिका) list of CustomGraphOptimizers to apply. |
| सूची< RewriterConfig.CustomGraphOptimizer.Builder > | getCustomOptimizersBuilderList () list of CustomGraphOptimizers to apply. |
| int यहाँ | getCustomOptimizersCount () list of CustomGraphOptimizers to apply. |
| सूची< RewriterConfig.CustomGraphOptimizer > | GetCustomOptimizersList () list of CustomGraphOptimizers to apply. |
| रीराइटरकॉन्फिग.कस्टमग्राफऑप्टिमाइज़रऑरबिल्डर | getCustomOptimizersOrBuilder (int अनुक्रमणिका) list of CustomGraphOptimizers to apply. |
| सूची<? RewriterConfig.CustomGraphOptimizerOrBuilder > का विस्तार करता है | getCustomOptimizersOrBuilderList () list of CustomGraphOptimizers to apply. |
| रीराइटरकॉन्फिग.टॉगल | getDebugस्ट्रिपर () Strips debug-related nodes from the graph (off by default). |
| int यहाँ | getDebugStripperValue () Strips debug-related nodes from the graph (off by default). |
| रीराइटर कॉन्फिग | |
| रीराइटरकॉन्फिग.टॉगल | निर्भरता अनुकूलन प्राप्त करें () Control dependency optimizations (default is ON). |
| int यहाँ | getDependencyOptimizationValue () Control dependency optimizations (default is ON). |
| अंतिम स्थिर com.google.protobuf.Descriptors.Descriptor | |
| com.google.protobuf.Descriptors.Descriptor | |
| बूलियन | getDisableMetaऑप्टिमाइज़र () Disable the entire meta optimizer (off by default). |
| बूलियन | getDisableModelPruning () If true, don't remove unnecessary ops from the graph bool disable_model_pruning = 2; |
| बूलियन | getExperimentalDisableCompressedTensorOptimization () Disable optimizations that assume compressed tensors. |
| बूलियन | getFairOnOptimizerErrors () If true, any optimization pass failing will cause the MetaOptimizer to stop with an error. |
| रीराइटरकॉन्फिग.टॉगल | गेटफ़ंक्शनऑप्टिमाइज़ेशन () Function optimizations (default is ON). |
| int यहाँ | getFunctionOptimizationValue () Function optimizations (default is ON). |
| रीराइटरकॉन्फिग.टॉगल | कार्यान्वयन चयनकर्ता प्राप्त करें () Enable the swap of kernel implementations based on the device placement (default is ON). |
| int यहाँ | getImplementationSelectorValue () Enable the swap of kernel implementations based on the device placement (default is ON). |
| सत्यापनकर्ताकॉन्फिग | getInterOptimizerVerifierConfig () VerifierConfig specifying the verifiers to be run after every optimizer. |
| सत्यापनकर्ताकॉन्फिग.बिल्डर | getInterOptimizerVerifierConfigBuilder () VerifierConfig specifying the verifiers to be run after every optimizer. |
| सत्यापनकर्ताकॉन्फिगऑरबिल्डर | getInterOptimizerVerifierConfigOrBuilder () VerifierConfig specifying the verifiers to be run after every optimizer. |
| रीराइटरकॉन्फिग.टॉगल | गेटलेआउटऑप्टिमाइज़र () Optimize tensor layouts (default is ON) e.g. |
| int यहाँ | getLayoutऑप्टिमाइज़रवैल्यू () Optimize tensor layouts (default is ON) e.g. |
| रीराइटरकॉन्फिग.टॉगल | getLoopऑप्टिमाइज़ेशन () Loop optimizations (default is ON). |
| int यहाँ | getLoopऑप्टिमाइज़ेशनवैल्यू () Loop optimizations (default is ON). |
| RewriterConfig.MemOptType | getMemoryअनुकूलन () Configures memory optimization passes through the meta-optimizer. |
| int यहाँ | getMemoryOptimizationValue () Configures memory optimization passes through the meta-optimizer. |
| डोरी | getMemoryOptimizerTargetNodeNameScope () A node name scope for node names which are valid outputs of recomputations. |
| com.google.protobuf.ByteString | getMemoryOptimizerTargetNodeNameScopeBytes () A node name scope for node names which are valid outputs of recomputations. |
| RewriterConfig.NumIterationsType | getMetaOptimizerIterations () Controls how many times we run the optimizers in meta optimizer (default is once). |
| int यहाँ | getMetaOptimizerIterationsValue () Controls how many times we run the optimizers in meta optimizer (default is once). |
| लंबा | getMetaOptimizerTimeoutMs () Maximum number of milliseconds to spend optimizing a single graph before timing out. |
| int यहाँ | getMinGraphNodes () The minimum number of nodes in a graph to optimizer. |
| डोरी | गेटऑप्टिमाइज़र (इंट इंडेक्स) If non-empty, will use this as an alternative way to specify a list of optimizations to turn on and the order of the optimizations (replacing the meta-optimizer). |
| com.google.protobuf.ByteString | getOptimizersBytes (int अनुक्रमणिका) If non-empty, will use this as an alternative way to specify a list of optimizations to turn on and the order of the optimizations (replacing the meta-optimizer). |
| int यहाँ | गेटऑप्टिमाइजर्सकाउंट () If non-empty, will use this as an alternative way to specify a list of optimizations to turn on and the order of the optimizations (replacing the meta-optimizer). |
| com.google.protobuf.ProtocolStringList | getऑप्टिमाइज़रलिस्ट () If non-empty, will use this as an alternative way to specify a list of optimizations to turn on and the order of the optimizations (replacing the meta-optimizer). |
| रीराइटरकॉन्फिग.टॉगल | getPinToHostऑप्टिमाइज़ेशन () Force small ops onto the CPU (default is OFF). |
| int यहाँ | getPinToHostOptimizationValue () Force small ops onto the CPU (default is OFF). |
| सत्यापनकर्ताकॉन्फिग | getPostOptimizationVerifierConfig () VerifierConfig specifying the verifiers to be run at the end, after all optimizers have run. |
| सत्यापनकर्ताकॉन्फिग.बिल्डर | getPostOptimizationVerifierConfigBuilder () VerifierConfig specifying the verifiers to be run at the end, after all optimizers have run. |
| सत्यापनकर्ताकॉन्फिगऑरबिल्डर | getPostOptimizationVerifierConfigOrBuilder () VerifierConfig specifying the verifiers to be run at the end, after all optimizers have run. |
| रीराइटरकॉन्फिग.टॉगल | getRemapping () Remapping (default is ON) Remap subgraphs onto more efficient implementations. |
| int यहाँ | getRemappingValue () Remapping (default is ON) Remap subgraphs onto more efficient implementations. |
| रीराइटरकॉन्फिग.टॉगल | getScopedAllocatorOptimization () Try to allocate some independent Op outputs contiguously in order to merge or eliminate downstream Ops (off by default). |
| int यहाँ | getScopedAllocatorOptimizationValue () Try to allocate some independent Op outputs contiguously in order to merge or eliminate downstream Ops (off by default). |
| स्कोप्डएलोकेटर विकल्प | getScopedAllocatorOpts () .tensorflow.ScopedAllocatorOptions scoped_allocator_opts = 16; |
| स्कोप्डएलोकेटरऑप्शंस.बिल्डर | getScopedAllocatorOptsBuilder () .tensorflow.ScopedAllocatorOptions scoped_allocator_opts = 16; |
| स्कोप्डएलोकेटरऑप्शंसऑरबिल्डर | getScopedAllocatorOptsOrBuilder () .tensorflow.ScopedAllocatorOptions scoped_allocator_opts = 16; |
| रीराइटरकॉन्फिग.टॉगल | getShapeअनुकूलन () Shape optimizations (default is ON) Simplify computations made on shapes. |
| int यहाँ | getShapeOptimizationValue () Shape optimizations (default is ON) Simplify computations made on shapes. |
| बूलियन | इसमेंऑटोपैरेलल () है Configures AutoParallel optimization passes either through the meta-optimizer or when manually specified through the optimizers field. |
| बूलियन | इसमेंइंटरऑप्टिमाइज़रवेरिफायरकॉन्फिग () है VerifierConfig specifying the verifiers to be run after every optimizer. |
| बूलियन | hasPostOptimizationVerifierConfig () VerifierConfig specifying the verifiers to be run at the end, after all optimizers have run. |
| बूलियन | hasScopedAllocatorOpts () .tensorflow.ScopedAllocatorOptions scoped_allocator_opts = 16; |
| अंतिम बूलियन | |
| रीराइटरकॉन्फिग.बिल्डर | मर्जऑटोपैरेलल ( ऑटोपैरेललऑप्शन मान) Configures AutoParallel optimization passes either through the meta-optimizer or when manually specified through the optimizers field. |
| रीराइटरकॉन्फिग.बिल्डर | मर्जफ्रॉम (com.google.protobuf.Message अन्य) |
| रीराइटरकॉन्फिग.बिल्डर | मर्जफ्रॉम (com.google.protobuf.CodedInputStream इनपुट, com.google.protobuf.ExtensionRegistryLite एक्सटेंशनरजिस्ट्री) |
| रीराइटरकॉन्फिग.बिल्डर | मर्जइंटरऑप्टिमाइजरवेरिफायरकॉन्फिग ( वेरिफायरकॉन्फिग मान) VerifierConfig specifying the verifiers to be run after every optimizer. |
| रीराइटरकॉन्फिग.बिल्डर | मर्जपोस्टऑप्टिमाइजेशनवेरिफायरकॉन्फिग ( वेरिफायरकॉन्फिग मान) VerifierConfig specifying the verifiers to be run at the end, after all optimizers have run. |
| रीराइटरकॉन्फिग.बिल्डर | mergeScopedAllocatorOpts ( स्कोप्डएलोकेटरऑप्शंस मान) .tensorflow.ScopedAllocatorOptions scoped_allocator_opts = 16; |
| अंतिम रीराइटरकॉन्फिग.बिल्डर | मर्जअज्ञातफ़ील्ड्स (com.google.protobuf.UnknownFieldSet अज्ञातफ़ील्ड्स) |
| रीराइटरकॉन्फिग.बिल्डर | रिमूवकस्टमऑप्टिमाइज़र (इंट इंडेक्स) list of CustomGraphOptimizers to apply. |
| रीराइटरकॉन्फिग.बिल्डर | setArithmeticOptimization ( RewriterConfig.Toggle मान) Arithmetic optimizations (default is ON) e.g. |
| रीराइटरकॉन्फिग.बिल्डर | setArithmeticOptimizationValue (int मान) Arithmetic optimizations (default is ON) e.g. |
| रीराइटरकॉन्फिग.बिल्डर | |
| रीराइटरकॉन्फिग.बिल्डर | |
| रीराइटरकॉन्फिग.बिल्डर | setAutoMixedPrecisionMklValue (int मान) Optimize data types for MKL (default is OFF). |
| रीराइटरकॉन्फिग.बिल्डर | setAutoMixedPrecisionValue (int मान) Optimize data types for CUDA (default is OFF). |
| रीराइटरकॉन्फिग.बिल्डर | सेटऑटोपैरेलल ( AutoParallelOptions.Builder BuilderForValue) Configures AutoParallel optimization passes either through the meta-optimizer or when manually specified through the optimizers field. |
| रीराइटरकॉन्फिग.बिल्डर | सेटऑटोपैरेलल ( ऑटोपैरेललऑप्शन मान) Configures AutoParallel optimization passes either through the meta-optimizer or when manually specified through the optimizers field. |
| रीराइटरकॉन्फिग.बिल्डर | setCommonSubgraphElination ( RewriterConfig.Toggle मान) Common subgraph elimination (default is ON) e.g. |
| रीराइटरकॉन्फिग.बिल्डर | setCommonSubgraphElinationValue (int मान) Common subgraph elimination (default is ON) e.g. |
| रीराइटरकॉन्फिग.बिल्डर | setConstantFolding ( RewriterConfig.Toggle मान) Fold constants (default is ON) Statically infer the value of tensors when possible, and materialize the result using constants. |
| रीराइटरकॉन्फिग.बिल्डर | setConstantFoldingValue (int मान) Fold constants (default is ON) Statically infer the value of tensors when possible, and materialize the result using constants. |
| रीराइटरकॉन्फिग.बिल्डर | setCpuLayoutConversion ( RewriterConfig.CpuLayout मान) CPU Conversion settings between NHCW and NCHW. |
| रीराइटरकॉन्फिग.बिल्डर | setCpuLayoutConversionValue (int मान) CPU Conversion settings between NHCW and NCHW. |
| रीराइटरकॉन्फिग.बिल्डर | सेटकस्टमऑप्टिमाइज़र (इंट इंडेक्स, रीराइटरकॉन्फिग.कस्टमग्राफऑप्टिमाइज़र.बिल्डर बिल्डरफॉरवैल्यू) list of CustomGraphOptimizers to apply. |
| रीराइटरकॉन्फिग.बिल्डर | सेटकस्टमऑप्टिमाइज़र (इंट इंडेक्स, रीराइटरकॉन्फिग.कस्टमग्राफऑप्टिमाइज़र मान) list of CustomGraphOptimizers to apply. |
| रीराइटरकॉन्फिग.बिल्डर | setDebugStripper ( RewriterConfig.Toggle मान) Strips debug-related nodes from the graph (off by default). |
| रीराइटरकॉन्फिग.बिल्डर | setDebugStripperValue (int मान) Strips debug-related nodes from the graph (off by default). |
| रीराइटरकॉन्फिग.बिल्डर | सेटडिपेंडेंसीऑप्टिमाइज़ेशन ( RewriterConfig.Toggle मान) Control dependency optimizations (default is ON). |
| रीराइटरकॉन्फिग.बिल्डर | सेटडिपेंडेंसीऑप्टिमाइज़ेशनवैल्यू (इंट वैल्यू) Control dependency optimizations (default is ON). |
| रीराइटरकॉन्फिग.बिल्डर | सेटडिसेबलमेटाऑप्टिमाइज़र (बूलियन मान) Disable the entire meta optimizer (off by default). |
| रीराइटरकॉन्फिग.बिल्डर | setDisableModelPruning (बूलियन मान) If true, don't remove unnecessary ops from the graph bool disable_model_pruning = 2; |
| रीराइटरकॉन्फिग.बिल्डर | setExperimentalDisableCompressedTensorOptimization (बूलियन मान) Disable optimizations that assume compressed tensors. |
| रीराइटरकॉन्फिग.बिल्डर | सेटफ़ेलऑनऑप्टिमाइज़रत्रुटि (बूलियन मान) If true, any optimization pass failing will cause the MetaOptimizer to stop with an error. |
| रीराइटरकॉन्फिग.बिल्डर | सेटफ़ील्ड (com.google.protobuf.Descriptors.FieldDescriptor फ़ील्ड, ऑब्जेक्ट मान) |
| रीराइटरकॉन्फिग.बिल्डर | |
| रीराइटरकॉन्फिग.बिल्डर | सेटफ़ंक्शनऑप्टिमाइज़ेशनवैल्यू (int मान) Function optimizations (default is ON). |
| रीराइटरकॉन्फिग.बिल्डर | setImplementationSelector ( RewriterConfig.Toggle मान) Enable the swap of kernel implementations based on the device placement (default is ON). |
| रीराइटरकॉन्फिग.बिल्डर | setImplementationSelectorValue (int मान) Enable the swap of kernel implementations based on the device placement (default is ON). |
| रीराइटरकॉन्फिग.बिल्डर | सेटइंटरऑप्टिमाइज़रवेरिफायरकॉन्फिग ( वेरिफायरकॉन्फिग.बिल्डर बिल्डरफॉरवैल्यू) VerifierConfig specifying the verifiers to be run after every optimizer. |
| रीराइटरकॉन्फिग.बिल्डर | setInterOptimizerVerifierConfig ( VerifierConfig मान) VerifierConfig specifying the verifiers to be run after every optimizer. |
| रीराइटरकॉन्फिग.बिल्डर | |
| रीराइटरकॉन्फिग.बिल्डर | सेटलेआउटऑप्टिमाइज़रवैल्यू (इंट वैल्यू) Optimize tensor layouts (default is ON) e.g. |
| रीराइटरकॉन्फिग.बिल्डर | |
| रीराइटरकॉन्फिग.बिल्डर | सेटलूपऑप्टिमाइज़ेशनवैल्यू (इंट वैल्यू) Loop optimizations (default is ON). |
| रीराइटरकॉन्फिग.बिल्डर | सेटमेमोरीऑप्टिमाइज़ेशन ( RewriterConfig.MemOptType मान) Configures memory optimization passes through the meta-optimizer. |
| रीराइटरकॉन्फिग.बिल्डर | सेटमेमोरीऑप्टिमाइज़ेशनवैल्यू (इंट वैल्यू) Configures memory optimization passes through the meta-optimizer. |
| रीराइटरकॉन्फिग.बिल्डर | setMemoryOptimizerTargetNodeNameScope (स्ट्रिंग मान) A node name scope for node names which are valid outputs of recomputations. |
| रीराइटरकॉन्फिग.बिल्डर | setMemoryOptimizerTargetNodeNameScopeBytes (com.google.protobuf.ByteString मान) A node name scope for node names which are valid outputs of recomputations. |
| रीराइटरकॉन्फिग.बिल्डर | setMetaOptimizerIterations ( RewriterConfig.NumIterationsType मान) Controls how many times we run the optimizers in meta optimizer (default is once). |
| रीराइटरकॉन्फिग.बिल्डर | setMetaOptimizerIterationsValue (int मान) Controls how many times we run the optimizers in meta optimizer (default is once). |
| रीराइटरकॉन्फिग.बिल्डर | setMetaOptimizerTimeoutMs (लंबा मान) Maximum number of milliseconds to spend optimizing a single graph before timing out. |
| रीराइटरकॉन्फिग.बिल्डर | setMinGraphNodes (int मान) The minimum number of nodes in a graph to optimizer. |
| रीराइटरकॉन्फिग.बिल्डर | सेटऑप्टिमाइज़र (इंट इंडेक्स, स्ट्रिंग वैल्यू) If non-empty, will use this as an alternative way to specify a list of optimizations to turn on and the order of the optimizations (replacing the meta-optimizer). |
| रीराइटरकॉन्फिग.बिल्डर | setPinToHostOptimization ( RewriterConfig.Toggle मान) Force small ops onto the CPU (default is OFF). |
| रीराइटरकॉन्फिग.बिल्डर | setPinToHostOptimizationValue (int मान) Force small ops onto the CPU (default is OFF). |
| रीराइटरकॉन्फिग.बिल्डर | सेटपोस्टऑप्टिमाइजेशनवेरिफायरकॉन्फिग ( वेरिफायरकॉन्फिग मान) VerifierConfig specifying the verifiers to be run at the end, after all optimizers have run. |
| रीराइटरकॉन्फिग.बिल्डर | सेटपोस्टऑप्टिमाइजेशनवेरिफायरकॉन्फिग ( वेरिफायरकॉन्फिग.बिल्डर बिल्डरफॉरवैल्यू) VerifierConfig specifying the verifiers to be run at the end, after all optimizers have run. |
| रीराइटरकॉन्फिग.बिल्डर | सेटरीमैपिंग ( RewriterConfig.Toggle मान) Remapping (default is ON) Remap subgraphs onto more efficient implementations. |
| रीराइटरकॉन्फिग.बिल्डर | setRemappingValue (int मान) Remapping (default is ON) Remap subgraphs onto more efficient implementations. |
| रीराइटरकॉन्फिग.बिल्डर | setRepeatedField (com.google.protobuf.Descriptors.FieldDescriptor फ़ील्ड, इंट इंडेक्स, ऑब्जेक्ट वैल्यू) |
| रीराइटरकॉन्फिग.बिल्डर | setScopedAllocatorOptimization ( RewriterConfig.Toggle मान) Try to allocate some independent Op outputs contiguously in order to merge or eliminate downstream Ops (off by default). |
| रीराइटरकॉन्फिग.बिल्डर | setScopedAllocatorOptimizationValue (int मान) Try to allocate some independent Op outputs contiguously in order to merge or eliminate downstream Ops (off by default). |
| रीराइटरकॉन्फिग.बिल्डर | setScopedAllocatorOpts ( ScopedAllocatorOptions मान) .tensorflow.ScopedAllocatorOptions scoped_allocator_opts = 16; |
| रीराइटरकॉन्फिग.बिल्डर | setScopedAllocatorOpts ( ScopedAllocatorOptions.Builder BuilderForValue) .tensorflow.ScopedAllocatorOptions scoped_allocator_opts = 16; |
| रीराइटरकॉन्फिग.बिल्डर | सेटशेपऑप्टिमाइज़ेशन ( RewriterConfig.Toggle मान) Shape optimizations (default is ON) Simplify computations made on shapes. |
| रीराइटरकॉन्फिग.बिल्डर | सेटशेपऑप्टिमाइज़ेशनवैल्यू (इंट वैल्यू) Shape optimizations (default is ON) Simplify computations made on shapes. |
| अंतिम रीराइटरकॉन्फिग.बिल्डर | अज्ञात फ़ील्ड सेट करें (com.google.protobuf. अज्ञात फ़ील्ड सेट अज्ञात फ़ील्ड) |
विरासत में मिले तरीके
सार्वजनिक तरीके
सार्वजनिक रीराइटरकॉन्फिग.बिल्डर ऐडऑलकस्टमऑप्टिमाइजर्स (इटरेबल<? रीराइटरकॉन्फिग.कस्टमग्राफऑप्टिमाइजर > मान बढ़ाता है)
list of CustomGraphOptimizers to apply.
repeated .tensorflow.RewriterConfig.CustomGraphOptimizer custom_optimizers = 200; सार्वजनिक RewriterConfig.Builder addAllOptimizers (Iterable<String> मान)
If non-empty, will use this as an alternative way to specify a list of
optimizations to turn on and the order of the optimizations (replacing the
meta-optimizer).
Of the RewriterConfig options, only the AutoParallel configuration options
(the auto_parallel field) apply to manually requested optimization passes
("autoparallel"). Memory optimization passes ("memory") invoked here are
not configurable (in contrast to memory optimization passes through the
meta-optimizer) and act only on manual op annotations.
Custom optimizers (see custom_optimizers) that are not part of this
schedule will be run after - in the order that they were specified.
repeated string optimizers = 100; सार्वजनिक रीराइटरकॉन्फिग.बिल्डर ऐडकस्टमऑप्टिमाइजर्स ( रीराइटरकॉन्फिग.कस्टमग्राफऑप्टिमाइजर.बिल्डर बिल्डरफॉरवैल्यू)
list of CustomGraphOptimizers to apply.
repeated .tensorflow.RewriterConfig.CustomGraphOptimizer custom_optimizers = 200; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर ऐडकस्टमऑप्टिमाइजर्स (इंट इंडेक्स, रीराइटरकॉन्फिग.कस्टमग्राफऑप्टिमाइजर.बिल्डर बिल्डरफॉरवैल्यू)
list of CustomGraphOptimizers to apply.
repeated .tensorflow.RewriterConfig.CustomGraphOptimizer custom_optimizers = 200; सार्वजनिक रिवाइटरकॉन्फ़िग.बिल्डर ऐडकस्टमऑप्टिमाइज़र ( रीराइटरकॉन्फ़िग.कस्टमग्राफऑप्टिमाइज़र मान)
list of CustomGraphOptimizers to apply.
repeated .tensorflow.RewriterConfig.CustomGraphOptimizer custom_optimizers = 200; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर ऐडकस्टमऑप्टिमाइजर्स (इंट इंडेक्स, रीराइटरकॉन्फिग.कस्टमग्राफऑप्टिमाइजर वैल्यू)
list of CustomGraphOptimizers to apply.
repeated .tensorflow.RewriterConfig.CustomGraphOptimizer custom_optimizers = 200; सार्वजनिक रिवाइटरकॉन्फ़िग.कस्टमग्राफ़ऑप्टिमाइज़र.बिल्डर ऐडकस्टमऑप्टिमाइज़रबिल्डर (इंट इंडेक्स)
list of CustomGraphOptimizers to apply.
repeated .tensorflow.RewriterConfig.CustomGraphOptimizer custom_optimizers = 200; सार्वजनिक रिवाइटरकॉन्फ़िग.कस्टमग्राफ़ऑप्टिमाइज़र.बिल्डर ऐडकस्टमऑप्टिमाइज़रबिल्डर ()
list of CustomGraphOptimizers to apply.
repeated .tensorflow.RewriterConfig.CustomGraphOptimizer custom_optimizers = 200; सार्वजनिक रिवाइटरकॉन्फ़िग.बिल्डर ऐडऑप्टिमाइज़र (स्ट्रिंग मान)
If non-empty, will use this as an alternative way to specify a list of
optimizations to turn on and the order of the optimizations (replacing the
meta-optimizer).
Of the RewriterConfig options, only the AutoParallel configuration options
(the auto_parallel field) apply to manually requested optimization passes
("autoparallel"). Memory optimization passes ("memory") invoked here are
not configurable (in contrast to memory optimization passes through the
meta-optimizer) and act only on manual op annotations.
Custom optimizers (see custom_optimizers) that are not part of this
schedule will be run after - in the order that they were specified.
repeated string optimizers = 100; सार्वजनिक रिवाइटरकॉन्फ़िग.बिल्डर ऐडऑप्टिमाइज़रबाइट्स (com.google.protobuf.ByteString मान)
If non-empty, will use this as an alternative way to specify a list of
optimizations to turn on and the order of the optimizations (replacing the
meta-optimizer).
Of the RewriterConfig options, only the AutoParallel configuration options
(the auto_parallel field) apply to manually requested optimization passes
("autoparallel"). Memory optimization passes ("memory") invoked here are
not configurable (in contrast to memory optimization passes through the
meta-optimizer) and act only on manual op annotations.
Custom optimizers (see custom_optimizers) that are not part of this
schedule will be run after - in the order that they were specified.
repeated string optimizers = 100; सार्वजनिक RewriterConfig.Builder addRepeatedField (com.google.protobuf.Descriptors.FieldDescriptor फ़ील्ड, ऑब्जेक्ट मान)
सार्वजनिक रिवाइटरकॉन्फ़िग.बिल्डर क्लियरअरिथमेटिकऑप्टिमाइज़ेशन ()
Arithmetic optimizations (default is ON) e.g. Simplify arithmetic ops; merge ops with same value (like constants).
.tensorflow.RewriterConfig.Toggle arithmetic_optimization = 7; सार्वजनिक RewriterConfig.Builder ClearAutoMixedPrecision ()
Optimize data types for CUDA (default is OFF). This will try to use float16 on GPU which is faster. Note that this can change the numerical stability of the graph and may require the use of loss scaling to maintain model convergence.
.tensorflow.RewriterConfig.Toggle auto_mixed_precision = 23; सार्वजनिक RewriterConfig.Builder ClearAutoMixedPrecisionMkl ()
Optimize data types for MKL (default is OFF). This will try to use bfloat16 on CPUs, which is faster. Note that this can change the numerical stability of the graph.
.tensorflow.RewriterConfig.Toggle auto_mixed_precision_mkl = 25; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर क्लियरऑटोपैरेलल ()
Configures AutoParallel optimization passes either through the meta-optimizer or when manually specified through the optimizers field.
.tensorflow.AutoParallelOptions auto_parallel = 5; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर क्लियरकॉमनसबग्राफएलिमिनेशन ()
Common subgraph elimination (default is ON) e.g. Simplify arithmetic ops; merge ops with same value (like constants).
.tensorflow.RewriterConfig.Toggle common_subgraph_elimination = 24; सार्वजनिक RewriterConfig.Builder ClearConstantFolding ()
Fold constants (default is ON) Statically infer the value of tensors when possible, and materialize the result using constants.
.tensorflow.RewriterConfig.Toggle constant_folding = 3; सार्वजनिक RewriterConfig.Builder ClearCpuLayoutConversion ()
CPU Conversion settings between NHCW and NCHW.
.tensorflow.RewriterConfig.CpuLayout cpu_layout_conversion = 50; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर क्लियरकस्टमऑप्टिमाइज़र ()
list of CustomGraphOptimizers to apply.
repeated .tensorflow.RewriterConfig.CustomGraphOptimizer custom_optimizers = 200; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर क्लियरडीबगस्ट्रिपर ()
Strips debug-related nodes from the graph (off by default).
.tensorflow.RewriterConfig.Toggle debug_stripper = 11; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर क्लियरडिपेंडेंसीऑप्टिमाइज़ेशन ()
Control dependency optimizations (default is ON). Remove redundant control dependencies, which may enable other optimization.
.tensorflow.RewriterConfig.Toggle dependency_optimization = 8; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर क्लियरडिसेबलमेटाऑप्टिमाइज़र ()
Disable the entire meta optimizer (off by default).
bool disable_meta_optimizer = 19; सार्वजनिक RewriterConfig.Builder ClearDisableModelPruning ()
If true, don't remove unnecessary ops from the graph
bool disable_model_pruning = 2; सार्वजनिक RewriterConfig.Builder ClearExperimentalDisableCompressedTensorOptimization ()
Disable optimizations that assume compressed tensors. Note that this flag is experimental and may be removed in the future.
bool experimental_disable_compressed_tensor_optimization = 26; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर क्लियरफ़ेलऑनऑप्टिमाइज़रइरर्स ()
If true, any optimization pass failing will cause the MetaOptimizer to stop with an error. By default - or when set to false, failing passes are skipped silently.
bool fail_on_optimizer_errors = 21; सार्वजनिक रिवाइटरकॉन्फ़िग.बिल्डर क्लियरफ़ील्ड (com.google.protobuf.Descriptors.FieldDescriptor फ़ील्ड)
सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर क्लियरफंक्शनऑप्टिमाइजेशन ()
Function optimizations (default is ON).
.tensorflow.RewriterConfig.Toggle function_optimization = 10; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर क्लियरइम्प्लीमेंटेशनसेलेक्टर ()
Enable the swap of kernel implementations based on the device placement (default is ON).
.tensorflow.RewriterConfig.Toggle implementation_selector = 22; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर क्लियरइंटरऑप्टिमाइजरवेरिफायरकॉन्फिग ()
VerifierConfig specifying the verifiers to be run after every optimizer.
.tensorflow.VerifierConfig inter_optimizer_verifier_config = 300; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर क्लियरलेआउटऑप्टिमाइज़र ()
Optimize tensor layouts (default is ON) e.g. This will try to use NCHW layout on GPU which is faster.
.tensorflow.RewriterConfig.Toggle layout_optimizer = 1; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर क्लियरलूपऑप्टिमाइजेशन ()
Loop optimizations (default is ON).
.tensorflow.RewriterConfig.Toggle loop_optimization = 9; सार्वजनिक रिवाइटरकॉन्फ़िग.बिल्डर क्लियरमेमोरीऑप्टिमाइज़ेशन ()
Configures memory optimization passes through the meta-optimizer. Has no effect on manually requested memory optimization passes in the optimizers field.
.tensorflow.RewriterConfig.MemOptType memory_optimization = 4; सार्वजनिक रिवाइटरकॉन्फ़िग.बिल्डर क्लियरमेमोरीऑप्टिमाइज़रटार्गेटनोडनेमस्कोप ()
A node name scope for node names which are valid outputs of recomputations. Inputs to nodes that match this scope may be recomputed (subject either to manual annotation of those input nodes or to manual annotation and heuristics depending on memory_optimization), but the nodes themselves will not be recomputed. This matches any sub-scopes as well, meaning the scope can appear not just as a top-level scope. For example, if the value is "gradients/", the default, it will match node name "gradients/foo", "foo/gradients/bar", but not "foo_gradients/"
string memory_optimizer_target_node_name_scope = 6; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर क्लियरमेटाऑप्टिमाइज़रइटरेशन ()
Controls how many times we run the optimizers in meta optimizer (default is once).
.tensorflow.RewriterConfig.NumIterationsType meta_optimizer_iterations = 12; सार्वजनिक रिवाइटरकॉन्फ़िग.बिल्डर क्लियरमेटाऑप्टिमाइज़रटाइमआउटएमएस ()
Maximum number of milliseconds to spend optimizing a single graph before timing out. If equal to 0 the system picks a default (currently 5 minutes). If less than 0 the optimizer will never time out.
int64 meta_optimizer_timeout_ms = 20; सार्वजनिक RewriterConfig.Builder ClearMinGraphNodes ()
The minimum number of nodes in a graph to optimizer. For smaller graphs, optimization is skipped. 0 means the system picks an appropriate number. < 0 means do not skip optimization.
int32 min_graph_nodes = 17; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर क्लियरऑप्टिमाइजर्स ()
If non-empty, will use this as an alternative way to specify a list of
optimizations to turn on and the order of the optimizations (replacing the
meta-optimizer).
Of the RewriterConfig options, only the AutoParallel configuration options
(the auto_parallel field) apply to manually requested optimization passes
("autoparallel"). Memory optimization passes ("memory") invoked here are
not configurable (in contrast to memory optimization passes through the
meta-optimizer) and act only on manual op annotations.
Custom optimizers (see custom_optimizers) that are not part of this
schedule will be run after - in the order that they were specified.
repeated string optimizers = 100; सार्वजनिक रिवाइटरकॉन्फ़िग.बिल्डर क्लियरपिनटूहोस्टऑप्टिमाइज़ेशन ()
Force small ops onto the CPU (default is OFF).
.tensorflow.RewriterConfig.Toggle pin_to_host_optimization = 18; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर क्लियरपोस्टऑप्टिमाइजेशनवेरिफायरकॉन्फिग ()
VerifierConfig specifying the verifiers to be run at the end, after all optimizers have run.
.tensorflow.VerifierConfig post_optimization_verifier_config = 301; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर क्लियररीमैपिंग ()
Remapping (default is ON) Remap subgraphs onto more efficient implementations.
.tensorflow.RewriterConfig.Toggle remapping = 14; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर क्लियरस्कोप्डएलोकेटरऑप्टिमाइज़ेशन ()
Try to allocate some independent Op outputs contiguously in order to merge or eliminate downstream Ops (off by default).
.tensorflow.RewriterConfig.Toggle scoped_allocator_optimization = 15; सार्वजनिक RewriterConfig.Builder ClearScopedAllocatorOpts ()
.tensorflow.ScopedAllocatorOptions scoped_allocator_opts = 16;
सार्वजनिक रिवाइटरकॉन्फ़िग.बिल्डर क्लियरशेपऑप्टिमाइज़ेशन ()
Shape optimizations (default is ON) Simplify computations made on shapes.
.tensorflow.RewriterConfig.Toggle shape_optimization = 13; सार्वजनिक रिवाइटरकॉन्फ़िग.टॉगल getArithmeticOptimization ()
Arithmetic optimizations (default is ON) e.g. Simplify arithmetic ops; merge ops with same value (like constants).
.tensorflow.RewriterConfig.Toggle arithmetic_optimization = 7; सार्वजनिक पूर्णांक getArithmeticOptimizationValue ()
Arithmetic optimizations (default is ON) e.g. Simplify arithmetic ops; merge ops with same value (like constants).
.tensorflow.RewriterConfig.Toggle arithmetic_optimization = 7; सार्वजनिक RewriterConfig.Toggle getAutoMixedPrecision ()
Optimize data types for CUDA (default is OFF). This will try to use float16 on GPU which is faster. Note that this can change the numerical stability of the graph and may require the use of loss scaling to maintain model convergence.
.tensorflow.RewriterConfig.Toggle auto_mixed_precision = 23; सार्वजनिक RewriterConfig.Toggle getAutoMixedPrecisionMkl ()
Optimize data types for MKL (default is OFF). This will try to use bfloat16 on CPUs, which is faster. Note that this can change the numerical stability of the graph.
.tensorflow.RewriterConfig.Toggle auto_mixed_precision_mkl = 25; सार्वजनिक int getAutoMixedPrecisionMklValue ()
Optimize data types for MKL (default is OFF). This will try to use bfloat16 on CPUs, which is faster. Note that this can change the numerical stability of the graph.
.tensorflow.RewriterConfig.Toggle auto_mixed_precision_mkl = 25; सार्वजनिक int getAutoMixedPrecisionValue ()
Optimize data types for CUDA (default is OFF). This will try to use float16 on GPU which is faster. Note that this can change the numerical stability of the graph and may require the use of loss scaling to maintain model convergence.
.tensorflow.RewriterConfig.Toggle auto_mixed_precision = 23; सार्वजनिक AutoParallelOptions getAutoParallel ()
Configures AutoParallel optimization passes either through the meta-optimizer or when manually specified through the optimizers field.
.tensorflow.AutoParallelOptions auto_parallel = 5; सार्वजनिक AutoParallelOptions.Builder getAutoParallelBuilder ()
Configures AutoParallel optimization passes either through the meta-optimizer or when manually specified through the optimizers field.
.tensorflow.AutoParallelOptions auto_parallel = 5; सार्वजनिक AutoParallelOptionsOrBuilder getAutoParallelOrBuilder ()
Configures AutoParallel optimization passes either through the meta-optimizer or when manually specified through the optimizers field.
.tensorflow.AutoParallelOptions auto_parallel = 5; सार्वजनिक RewriterConfig.Toggle getCommonSubgraphElination ()
Common subgraph elimination (default is ON) e.g. Simplify arithmetic ops; merge ops with same value (like constants).
.tensorflow.RewriterConfig.Toggle common_subgraph_elimination = 24; सार्वजनिक int getCommonSubgraphElinationValue ()
Common subgraph elimination (default is ON) e.g. Simplify arithmetic ops; merge ops with same value (like constants).
.tensorflow.RewriterConfig.Toggle common_subgraph_elimination = 24; सार्वजनिक रिवाइटरकॉन्फ़िग.टॉगल getConstantFolding ()
Fold constants (default is ON) Statically infer the value of tensors when possible, and materialize the result using constants.
.tensorflow.RewriterConfig.Toggle constant_folding = 3; सार्वजनिक int getConstantFoldingValue ()
Fold constants (default is ON) Statically infer the value of tensors when possible, and materialize the result using constants.
.tensorflow.RewriterConfig.Toggle constant_folding = 3; सार्वजनिक RewriterConfig.CpuLayout getCpuLayoutConversion ()
CPU Conversion settings between NHCW and NCHW.
.tensorflow.RewriterConfig.CpuLayout cpu_layout_conversion = 50; सार्वजनिक int getCpuLayoutConversionValue ()
CPU Conversion settings between NHCW and NCHW.
.tensorflow.RewriterConfig.CpuLayout cpu_layout_conversion = 50; सार्वजनिक रिवाइटरकॉन्फिग.कस्टमग्राफऑप्टिमाइज़र गेटकस्टमऑप्टिमाइज़र (इंट इंडेक्स)
list of CustomGraphOptimizers to apply.
repeated .tensorflow.RewriterConfig.CustomGraphOptimizer custom_optimizers = 200; सार्वजनिक रिवाइटरकॉन्फिग.कस्टमग्राफऑप्टिमाइजर.बिल्डर गेटकस्टमऑप्टिमाइजर्सबिल्डर (इंट इंडेक्स)
list of CustomGraphOptimizers to apply.
repeated .tensorflow.RewriterConfig.CustomGraphOptimizer custom_optimizers = 200; सार्वजनिक सूची
list of CustomGraphOptimizers to apply.
repeated .tensorflow.RewriterConfig.CustomGraphOptimizer custom_optimizers = 200; सार्वजनिक int getCustomOptimizersCount ()
list of CustomGraphOptimizers to apply.
repeated .tensorflow.RewriterConfig.CustomGraphOptimizer custom_optimizers = 200; सार्वजनिक सूची
list of CustomGraphOptimizers to apply.
repeated .tensorflow.RewriterConfig.CustomGraphOptimizer custom_optimizers = 200; सार्वजनिक रिवाइटरकॉन्फिग.कस्टमग्राफऑप्टिमाइज़रऑरबिल्डर गेटकस्टमऑप्टिमाइज़रऑरबिल्डर (इंट इंडेक्स)
list of CustomGraphOptimizers to apply.
repeated .tensorflow.RewriterConfig.CustomGraphOptimizer custom_optimizers = 200; सार्वजनिक सूची<? RewriterConfig.CustomGraphऑप्टिमाइज़रऑरबिल्डर > getCustomऑप्टिमाइज़रऑरबिल्डरलिस्ट () का विस्तार करता है
list of CustomGraphOptimizers to apply.
repeated .tensorflow.RewriterConfig.CustomGraphOptimizer custom_optimizers = 200; सार्वजनिक रिवाइटरकॉन्फ़िग.टॉगल getDebugStripper ()
Strips debug-related nodes from the graph (off by default).
.tensorflow.RewriterConfig.Toggle debug_stripper = 11; सार्वजनिक int getDebugStripperValue ()
Strips debug-related nodes from the graph (off by default).
.tensorflow.RewriterConfig.Toggle debug_stripper = 11; सार्वजनिक RewriterConfig.Toggle getDependencyOptimization ()
Control dependency optimizations (default is ON). Remove redundant control dependencies, which may enable other optimization.
.tensorflow.RewriterConfig.Toggle dependency_optimization = 8; सार्वजनिक int getDependencyOptimizationValue ()
Control dependency optimizations (default is ON). Remove redundant control dependencies, which may enable other optimization.
.tensorflow.RewriterConfig.Toggle dependency_optimization = 8; सार्वजनिक स्थैतिक अंतिम com.google.protobuf.Descriptors.Descriptor getDescriptor ()
सार्वजनिक com.google.protobuf.Descriptors.Descriptor getDescriptorForType ()
सार्वजनिक बूलियन getDisableMetaऑप्टिमाइज़र ()
Disable the entire meta optimizer (off by default).
bool disable_meta_optimizer = 19; सार्वजनिक बूलियन getDisableModelPruning ()
If true, don't remove unnecessary ops from the graph
bool disable_model_pruning = 2; सार्वजनिक बूलियन getExperimentalDisableCompressedTensorOptimization ()
Disable optimizations that assume compressed tensors. Note that this flag is experimental and may be removed in the future.
bool experimental_disable_compressed_tensor_optimization = 26; सार्वजनिक बूलियन getFairOnOptimizerErrors ()
If true, any optimization pass failing will cause the MetaOptimizer to stop with an error. By default - or when set to false, failing passes are skipped silently.
bool fail_on_optimizer_errors = 21; सार्वजनिक रिवाइटरकॉन्फिग.टॉगल गेटफंक्शनऑप्टिमाइजेशन ()
Function optimizations (default is ON).
.tensorflow.RewriterConfig.Toggle function_optimization = 10; सार्वजनिक int getFunctionOptimizationValue ()
Function optimizations (default is ON).
.tensorflow.RewriterConfig.Toggle function_optimization = 10; सार्वजनिक RewriterConfig.Toggle getImplementationSelector ()
Enable the swap of kernel implementations based on the device placement (default is ON).
.tensorflow.RewriterConfig.Toggle implementation_selector = 22; सार्वजनिक int getImplementationSelectorValue ()
Enable the swap of kernel implementations based on the device placement (default is ON).
.tensorflow.RewriterConfig.Toggle implementation_selector = 22; सार्वजनिक सत्यापनकर्ताकॉन्फिग getInterOptimizerVerifierConfig ()
VerifierConfig specifying the verifiers to be run after every optimizer.
.tensorflow.VerifierConfig inter_optimizer_verifier_config = 300; सार्वजनिक VerifierConfig.Builder getInterOptimizerVerifierConfigBuilder ()
VerifierConfig specifying the verifiers to be run after every optimizer.
.tensorflow.VerifierConfig inter_optimizer_verifier_config = 300; सार्वजनिक VerifierConfigOrBuilder getInterOptimizerVerifierConfigOrBuilder ()
VerifierConfig specifying the verifiers to be run after every optimizer.
.tensorflow.VerifierConfig inter_optimizer_verifier_config = 300; सार्वजनिक रिवाइटरकॉन्फिग.टॉगल गेटलेआउटऑप्टिमाइज़र ()
Optimize tensor layouts (default is ON) e.g. This will try to use NCHW layout on GPU which is faster.
.tensorflow.RewriterConfig.Toggle layout_optimizer = 1; सार्वजनिक int getLayoutOptimizerValue ()
Optimize tensor layouts (default is ON) e.g. This will try to use NCHW layout on GPU which is faster.
.tensorflow.RewriterConfig.Toggle layout_optimizer = 1; सार्वजनिक रिवाइटरकॉन्फ़िग.टॉगल getLoopऑप्टिमाइज़ेशन ()
Loop optimizations (default is ON).
.tensorflow.RewriterConfig.Toggle loop_optimization = 9; सार्वजनिक int getLoopOptimizationValue ()
Loop optimizations (default is ON).
.tensorflow.RewriterConfig.Toggle loop_optimization = 9; सार्वजनिक रिवाइटरकॉन्फ़िग.मेमऑप्टटाइप गेटमेमोरीऑप्टिमाइज़ेशन ()
Configures memory optimization passes through the meta-optimizer. Has no effect on manually requested memory optimization passes in the optimizers field.
.tensorflow.RewriterConfig.MemOptType memory_optimization = 4; सार्वजनिक int getMemoryOptimizationValue ()
Configures memory optimization passes through the meta-optimizer. Has no effect on manually requested memory optimization passes in the optimizers field.
.tensorflow.RewriterConfig.MemOptType memory_optimization = 4; सार्वजनिक स्ट्रिंग getMemoryOptimizerTargetNodeNameScope ()
A node name scope for node names which are valid outputs of recomputations. Inputs to nodes that match this scope may be recomputed (subject either to manual annotation of those input nodes or to manual annotation and heuristics depending on memory_optimization), but the nodes themselves will not be recomputed. This matches any sub-scopes as well, meaning the scope can appear not just as a top-level scope. For example, if the value is "gradients/", the default, it will match node name "gradients/foo", "foo/gradients/bar", but not "foo_gradients/"
string memory_optimizer_target_node_name_scope = 6; सार्वजनिक com.google.protobuf.ByteString getMemoryOptimizerTargetNodeNameScopeBytes ()
A node name scope for node names which are valid outputs of recomputations. Inputs to nodes that match this scope may be recomputed (subject either to manual annotation of those input nodes or to manual annotation and heuristics depending on memory_optimization), but the nodes themselves will not be recomputed. This matches any sub-scopes as well, meaning the scope can appear not just as a top-level scope. For example, if the value is "gradients/", the default, it will match node name "gradients/foo", "foo/gradients/bar", but not "foo_gradients/"
string memory_optimizer_target_node_name_scope = 6; सार्वजनिक RewriterConfig.NumIterationsType getMetaOptimizerIterations ()
Controls how many times we run the optimizers in meta optimizer (default is once).
.tensorflow.RewriterConfig.NumIterationsType meta_optimizer_iterations = 12; सार्वजनिक int getMetaOptimizerIterationsValue ()
Controls how many times we run the optimizers in meta optimizer (default is once).
.tensorflow.RewriterConfig.NumIterationsType meta_optimizer_iterations = 12; सार्वजनिक लंबे getMetaOptimizerTimeoutMs ()
Maximum number of milliseconds to spend optimizing a single graph before timing out. If equal to 0 the system picks a default (currently 5 minutes). If less than 0 the optimizer will never time out.
int64 meta_optimizer_timeout_ms = 20; सार्वजनिक int getMinGraphNodes ()
The minimum number of nodes in a graph to optimizer. For smaller graphs, optimization is skipped. 0 means the system picks an appropriate number. < 0 means do not skip optimization.
int32 min_graph_nodes = 17; सार्वजनिक स्ट्रिंग गेटऑप्टिमाइज़र (इंट इंडेक्स)
If non-empty, will use this as an alternative way to specify a list of
optimizations to turn on and the order of the optimizations (replacing the
meta-optimizer).
Of the RewriterConfig options, only the AutoParallel configuration options
(the auto_parallel field) apply to manually requested optimization passes
("autoparallel"). Memory optimization passes ("memory") invoked here are
not configurable (in contrast to memory optimization passes through the
meta-optimizer) and act only on manual op annotations.
Custom optimizers (see custom_optimizers) that are not part of this
schedule will be run after - in the order that they were specified.
repeated string optimizers = 100; सार्वजनिक com.google.protobuf.ByteString getOptimizersBytes (int अनुक्रमणिका)
If non-empty, will use this as an alternative way to specify a list of
optimizations to turn on and the order of the optimizations (replacing the
meta-optimizer).
Of the RewriterConfig options, only the AutoParallel configuration options
(the auto_parallel field) apply to manually requested optimization passes
("autoparallel"). Memory optimization passes ("memory") invoked here are
not configurable (in contrast to memory optimization passes through the
meta-optimizer) and act only on manual op annotations.
Custom optimizers (see custom_optimizers) that are not part of this
schedule will be run after - in the order that they were specified.
repeated string optimizers = 100; सार्वजनिक int getOptimizersCount ()
If non-empty, will use this as an alternative way to specify a list of
optimizations to turn on and the order of the optimizations (replacing the
meta-optimizer).
Of the RewriterConfig options, only the AutoParallel configuration options
(the auto_parallel field) apply to manually requested optimization passes
("autoparallel"). Memory optimization passes ("memory") invoked here are
not configurable (in contrast to memory optimization passes through the
meta-optimizer) and act only on manual op annotations.
Custom optimizers (see custom_optimizers) that are not part of this
schedule will be run after - in the order that they were specified.
repeated string optimizers = 100; सार्वजनिक com.google.protobuf.ProtocolStringList getOptimizersList ()
If non-empty, will use this as an alternative way to specify a list of
optimizations to turn on and the order of the optimizations (replacing the
meta-optimizer).
Of the RewriterConfig options, only the AutoParallel configuration options
(the auto_parallel field) apply to manually requested optimization passes
("autoparallel"). Memory optimization passes ("memory") invoked here are
not configurable (in contrast to memory optimization passes through the
meta-optimizer) and act only on manual op annotations.
Custom optimizers (see custom_optimizers) that are not part of this
schedule will be run after - in the order that they were specified.
repeated string optimizers = 100; सार्वजनिक रिवाइटरकॉन्फ़िग.टॉगल getPinToHostऑप्टिमाइज़ेशन ()
Force small ops onto the CPU (default is OFF).
.tensorflow.RewriterConfig.Toggle pin_to_host_optimization = 18; सार्वजनिक int getPinToHostOptimizationValue ()
Force small ops onto the CPU (default is OFF).
.tensorflow.RewriterConfig.Toggle pin_to_host_optimization = 18; सार्वजनिक सत्यापनकर्ताकॉन्फिग getPostOptimizationVerifierConfig ()
VerifierConfig specifying the verifiers to be run at the end, after all optimizers have run.
.tensorflow.VerifierConfig post_optimization_verifier_config = 301; सार्वजनिक VerifierConfig.Builder getPostOptimizationVerifierConfigBuilder ()
VerifierConfig specifying the verifiers to be run at the end, after all optimizers have run.
.tensorflow.VerifierConfig post_optimization_verifier_config = 301; सार्वजनिक VerifierConfigOrBuilder getPostOptimizationVerifierConfigOrBuilder ()
VerifierConfig specifying the verifiers to be run at the end, after all optimizers have run.
.tensorflow.VerifierConfig post_optimization_verifier_config = 301; सार्वजनिक रीराइटर कॉन्फिग. टॉगल गेट रीमैपिंग ()
Remapping (default is ON) Remap subgraphs onto more efficient implementations.
.tensorflow.RewriterConfig.Toggle remapping = 14; सार्वजनिक int getRemappingValue ()
Remapping (default is ON) Remap subgraphs onto more efficient implementations.
.tensorflow.RewriterConfig.Toggle remapping = 14; सार्वजनिक RewriterConfig.Toggle getScopedAllocatorOptimization ()
Try to allocate some independent Op outputs contiguously in order to merge or eliminate downstream Ops (off by default).
.tensorflow.RewriterConfig.Toggle scoped_allocator_optimization = 15; सार्वजनिक int getScopedAllocatorOptimizationValue ()
Try to allocate some independent Op outputs contiguously in order to merge or eliminate downstream Ops (off by default).
.tensorflow.RewriterConfig.Toggle scoped_allocator_optimization = 15; सार्वजनिक scopedAllocatorOptions getScopedAllocatorOpts ()
.tensorflow.ScopedAllocatorOptions scoped_allocator_opts = 16;
सार्वजनिक scopedAllocatorOptions.Builder getScopedAllocatorOptsBuilder ()
.tensorflow.ScopedAllocatorOptions scoped_allocator_opts = 16;
सार्वजनिक scopedAllocatorOptionsOrBuilder getScopedAllocatorOptsOrBuilder ()
.tensorflow.ScopedAllocatorOptions scoped_allocator_opts = 16;
सार्वजनिक RewriterConfig.Toggle getShapeOptimization ()
Shape optimizations (default is ON) Simplify computations made on shapes.
.tensorflow.RewriterConfig.Toggle shape_optimization = 13; सार्वजनिक int getShapeOptimizationValue ()
Shape optimizations (default is ON) Simplify computations made on shapes.
.tensorflow.RewriterConfig.Toggle shape_optimization = 13; सार्वजनिक बूलियन में ऑटोपैरेलल () है
Configures AutoParallel optimization passes either through the meta-optimizer or when manually specified through the optimizers field.
.tensorflow.AutoParallelOptions auto_parallel = 5; सार्वजनिक बूलियन में इंटरऑप्टिमाइज़रवेरिफ़ायर कॉन्फ़िग () है
VerifierConfig specifying the verifiers to be run after every optimizer.
.tensorflow.VerifierConfig inter_optimizer_verifier_config = 300; सार्वजनिक बूलियन में पोस्टऑप्टिमाइज़ेशनवेरिफ़ायरकॉन्फिग () है
VerifierConfig specifying the verifiers to be run at the end, after all optimizers have run.
.tensorflow.VerifierConfig post_optimization_verifier_config = 301; सार्वजनिक बूलियन hasScopedAllocatorOpts ()
.tensorflow.ScopedAllocatorOptions scoped_allocator_opts = 16;
सार्वजनिक अंतिम बूलियन आरंभीकृत है ()
सार्वजनिक RewriterConfig.Builder mergeAutoParallel ( AutoParallelOptions मान)
Configures AutoParallel optimization passes either through the meta-optimizer or when manually specified through the optimizers field.
.tensorflow.AutoParallelOptions auto_parallel = 5; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर मर्जफ्रॉम (com.google.protobuf.CodedInputStream इनपुट, com.google.protobuf.ExtensionRegistryLite एक्सटेंशनरजिस्ट्री)
फेंकता
| आईओएक्सेप्शन |
|---|
सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर मर्जइंटरऑप्टिमाइजरवेरिफायरकॉन्फिग ( वेरिफायरकॉन्फिग मान)
VerifierConfig specifying the verifiers to be run after every optimizer.
.tensorflow.VerifierConfig inter_optimizer_verifier_config = 300; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर मर्जपोस्टऑप्टिमाइजेशनवेरिफायरकॉन्फिग ( वेरिफायरकॉन्फिग मान)
VerifierConfig specifying the verifiers to be run at the end, after all optimizers have run.
.tensorflow.VerifierConfig post_optimization_verifier_config = 301; सार्वजनिक RewriterConfig.Builder mergeScopedAllocatorOpts ( ScopedAllocatorOptions मान)
.tensorflow.ScopedAllocatorOptions scoped_allocator_opts = 16;
सार्वजनिक अंतिम रीराइटरकॉन्फ़िग.बिल्डर मर्जअज्ञातफ़ील्ड्स (com.google.protobuf.UnknownFieldSet अज्ञातफ़ील्ड्स)
सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर रिमूवकस्टमऑप्टिमाइजर्स (इंट इंडेक्स)
list of CustomGraphOptimizers to apply.
repeated .tensorflow.RewriterConfig.CustomGraphOptimizer custom_optimizers = 200; सार्वजनिक RewriterConfig.Builder setArithmeticOptimization ( RewriterConfig.Toggle मान)
Arithmetic optimizations (default is ON) e.g. Simplify arithmetic ops; merge ops with same value (like constants).
.tensorflow.RewriterConfig.Toggle arithmetic_optimization = 7; सार्वजनिक रिवाइटरकॉन्फ़िग.बिल्डर सेटअरिथमेटिकऑप्टिमाइज़ेशनवैल्यू (इंट वैल्यू)
Arithmetic optimizations (default is ON) e.g. Simplify arithmetic ops; merge ops with same value (like constants).
.tensorflow.RewriterConfig.Toggle arithmetic_optimization = 7; सार्वजनिक RewriterConfig.Builder setAutoMixedPrecision ( RewriterConfig.Toggle मान)
Optimize data types for CUDA (default is OFF). This will try to use float16 on GPU which is faster. Note that this can change the numerical stability of the graph and may require the use of loss scaling to maintain model convergence.
.tensorflow.RewriterConfig.Toggle auto_mixed_precision = 23; सार्वजनिक RewriterConfig.Builder setAutoMixedPrecisionMkl ( RewriterConfig.Toggle मान)
Optimize data types for MKL (default is OFF). This will try to use bfloat16 on CPUs, which is faster. Note that this can change the numerical stability of the graph.
.tensorflow.RewriterConfig.Toggle auto_mixed_precision_mkl = 25; सार्वजनिक RewriterConfig.Builder setAutoMixedPrecisionMklValue (int मान)
Optimize data types for MKL (default is OFF). This will try to use bfloat16 on CPUs, which is faster. Note that this can change the numerical stability of the graph.
.tensorflow.RewriterConfig.Toggle auto_mixed_precision_mkl = 25; सार्वजनिक RewriterConfig.Builder setAutoMixedPrecisionValue (int मान)
Optimize data types for CUDA (default is OFF). This will try to use float16 on GPU which is faster. Note that this can change the numerical stability of the graph and may require the use of loss scaling to maintain model convergence.
.tensorflow.RewriterConfig.Toggle auto_mixed_precision = 23; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर सेटऑटोपैरेलल ( AutoParallelOptions.Builder BuilderForValue)
Configures AutoParallel optimization passes either through the meta-optimizer or when manually specified through the optimizers field.
.tensorflow.AutoParallelOptions auto_parallel = 5;सार्वजनिक RewriterConfig.Builder setAutoParallel ( AutoParallelOptions मान)
Configures AutoParallel optimization passes either through the meta-optimizer or when manually specified through the optimizers field.
.tensorflow.AutoParallelOptions auto_parallel = 5; सार्वजनिक RewriterConfig.Builder setCommonSubgraphElination ( RewriterConfig.Toggle मान)
Common subgraph elimination (default is ON) e.g. Simplify arithmetic ops; merge ops with same value (like constants).
.tensorflow.RewriterConfig.Toggle common_subgraph_elimination = 24; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर सेटकॉमनसबग्राफएलिमिनेशनवैल्यू (इंट वैल्यू)
Common subgraph elimination (default is ON) e.g. Simplify arithmetic ops; merge ops with same value (like constants).
.tensorflow.RewriterConfig.Toggle common_subgraph_elimination = 24; सार्वजनिक RewriterConfig.Builder setConstantFolding ( RewriterConfig.Toggle मान)
Fold constants (default is ON) Statically infer the value of tensors when possible, and materialize the result using constants.
.tensorflow.RewriterConfig.Toggle constant_folding = 3; सार्वजनिक RewriterConfig.Builder setConstantFoldingValue (int मान)
Fold constants (default is ON) Statically infer the value of tensors when possible, and materialize the result using constants.
.tensorflow.RewriterConfig.Toggle constant_folding = 3; सार्वजनिक RewriterConfig.Builder setCpuLayoutConversion ( RewriterConfig.CpuLayout मान)
CPU Conversion settings between NHCW and NCHW.
.tensorflow.RewriterConfig.CpuLayout cpu_layout_conversion = 50; सार्वजनिक RewriterConfig.Builder setCpuLayoutConversionValue (int मान)
CPU Conversion settings between NHCW and NCHW.
.tensorflow.RewriterConfig.CpuLayout cpu_layout_conversion = 50; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर सेटकस्टमऑप्टिमाइजर्स (इंट इंडेक्स, रीराइटरकॉन्फिग.कस्टमग्राफऑप्टिमाइजर.बिल्डर बिल्डरफॉरवैल्यू)
list of CustomGraphOptimizers to apply.
repeated .tensorflow.RewriterConfig.CustomGraphOptimizer custom_optimizers = 200; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर सेटकस्टमऑप्टिमाइजर्स (इंट इंडेक्स, रीराइटरकॉन्फिग.कस्टमग्राफऑप्टिमाइजर वैल्यू)
list of CustomGraphOptimizers to apply.
repeated .tensorflow.RewriterConfig.CustomGraphOptimizer custom_optimizers = 200; सार्वजनिक RewriterConfig.Builder setDebugStripper ( RewriterConfig.Toggle मान)
Strips debug-related nodes from the graph (off by default).
.tensorflow.RewriterConfig.Toggle debug_stripper = 11; सार्वजनिक RewriterConfig.Builder setDebugStripperValue (int मान)
Strips debug-related nodes from the graph (off by default).
.tensorflow.RewriterConfig.Toggle debug_stripper = 11; सार्वजनिक RewriterConfig.Builder सेटनिर्भरता अनुकूलन ( RewriterConfig.Toggle मान)
Control dependency optimizations (default is ON). Remove redundant control dependencies, which may enable other optimization.
.tensorflow.RewriterConfig.Toggle dependency_optimization = 8; सार्वजनिक रिवाइटरकॉन्फ़िग.बिल्डर सेटडिपेंडेंसीऑप्टिमाइज़ेशनवैल्यू (इंट वैल्यू)
Control dependency optimizations (default is ON). Remove redundant control dependencies, which may enable other optimization.
.tensorflow.RewriterConfig.Toggle dependency_optimization = 8; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर सेटडिसेबलमेटाऑप्टिमाइज़र (बूलियन मान)
Disable the entire meta optimizer (off by default).
bool disable_meta_optimizer = 19; सार्वजनिक RewriterConfig.Builder setDisableModelPruning (बूलियन मान)
If true, don't remove unnecessary ops from the graph
bool disable_model_pruning = 2; सार्वजनिक RewriterConfig.Builder setExperimentalDisableCompressedTensorOptimization (बूलियन मान)
Disable optimizations that assume compressed tensors. Note that this flag is experimental and may be removed in the future.
bool experimental_disable_compressed_tensor_optimization = 26; सार्वजनिक रिवाइटरकॉन्फ़िग.बिल्डर सेटफ़ेलऑनऑप्टिमाइज़रत्रुटि (बूलियन मान)
If true, any optimization pass failing will cause the MetaOptimizer to stop with an error. By default - or when set to false, failing passes are skipped silently.
bool fail_on_optimizer_errors = 21; सार्वजनिक रिवाइटरकॉन्फ़िग.बिल्डर सेटफ़ील्ड (com.google.protobuf.Descriptors.FieldDescriptor फ़ील्ड, ऑब्जेक्ट मान)
सार्वजनिक RewriterConfig.Builder सेटफंक्शनऑप्टिमाइज़ेशन ( RewriterConfig.Toggle मान)
Function optimizations (default is ON).
.tensorflow.RewriterConfig.Toggle function_optimization = 10; सार्वजनिक रिवाइटरकॉन्फ़िग.बिल्डर सेटफंक्शनऑप्टिमाइज़ेशनवैल्यू (इंट वैल्यू)
Function optimizations (default is ON).
.tensorflow.RewriterConfig.Toggle function_optimization = 10; सार्वजनिक RewriterConfig.Builder setImplementationSelector ( RewriterConfig.Toggle मान)
Enable the swap of kernel implementations based on the device placement (default is ON).
.tensorflow.RewriterConfig.Toggle implementation_selector = 22; सार्वजनिक RewriterConfig.Builder setImplementationSelectorValue (int मान)
Enable the swap of kernel implementations based on the device placement (default is ON).
.tensorflow.RewriterConfig.Toggle implementation_selector = 22; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर सेटइंटरऑप्टिमाइजरवेरिफायरकॉन्फिग ( VerifierConfig.Builder BuilderForValue)
VerifierConfig specifying the verifiers to be run after every optimizer.
.tensorflow.VerifierConfig inter_optimizer_verifier_config = 300; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर सेटइंटरऑप्टिमाइजरवेरिफायरकॉन्फिग ( वेरिफायरकॉन्फिग मान)
VerifierConfig specifying the verifiers to be run after every optimizer.
.tensorflow.VerifierConfig inter_optimizer_verifier_config = 300; सार्वजनिक RewriterConfig.Builder सेटलेआउटऑप्टिमाइज़र ( RewriterConfig.Toggle मान)
Optimize tensor layouts (default is ON) e.g. This will try to use NCHW layout on GPU which is faster.
.tensorflow.RewriterConfig.Toggle layout_optimizer = 1; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर सेटलेआउटऑप्टिमाइज़रवैल्यू (इंट वैल्यू)
Optimize tensor layouts (default is ON) e.g. This will try to use NCHW layout on GPU which is faster.
.tensorflow.RewriterConfig.Toggle layout_optimizer = 1; सार्वजनिक रिवाइटरकॉन्फ़िग.बिल्डर सेटलूपऑप्टिमाइज़ेशन ( रीराइटरकॉन्फ़िग.टॉगल मान)
Loop optimizations (default is ON).
.tensorflow.RewriterConfig.Toggle loop_optimization = 9; सार्वजनिक रिवाइटरकॉन्फ़िग.बिल्डर सेटलूपऑप्टिमाइज़ेशनवैल्यू (इंट वैल्यू)
Loop optimizations (default is ON).
.tensorflow.RewriterConfig.Toggle loop_optimization = 9; सार्वजनिक RewriterConfig.Builder सेटमेमोरीऑप्टिमाइज़ेशन ( RewriterConfig.MemOptType मान)
Configures memory optimization passes through the meta-optimizer. Has no effect on manually requested memory optimization passes in the optimizers field.
.tensorflow.RewriterConfig.MemOptType memory_optimization = 4; सार्वजनिक रिवाइटरकॉन्फ़िग.बिल्डर सेटमेमोरीऑप्टिमाइज़ेशनवैल्यू (इंट वैल्यू)
Configures memory optimization passes through the meta-optimizer. Has no effect on manually requested memory optimization passes in the optimizers field.
.tensorflow.RewriterConfig.MemOptType memory_optimization = 4; सार्वजनिक रिवाइटरकॉन्फ़िग.बिल्डर सेटमेमोरीऑप्टिमाइज़रटार्गेटनोडनेमस्कोप (स्ट्रिंग मान)
A node name scope for node names which are valid outputs of recomputations. Inputs to nodes that match this scope may be recomputed (subject either to manual annotation of those input nodes or to manual annotation and heuristics depending on memory_optimization), but the nodes themselves will not be recomputed. This matches any sub-scopes as well, meaning the scope can appear not just as a top-level scope. For example, if the value is "gradients/", the default, it will match node name "gradients/foo", "foo/gradients/bar", but not "foo_gradients/"
string memory_optimizer_target_node_name_scope = 6; सार्वजनिक रिवाइटरकॉन्फ़िग.बिल्डर सेटमेमोरीऑप्टिमाइज़रटार्गेटनोडनेमस्कोपबाइट्स (com.google.protobuf.ByteString मान)
A node name scope for node names which are valid outputs of recomputations. Inputs to nodes that match this scope may be recomputed (subject either to manual annotation of those input nodes or to manual annotation and heuristics depending on memory_optimization), but the nodes themselves will not be recomputed. This matches any sub-scopes as well, meaning the scope can appear not just as a top-level scope. For example, if the value is "gradients/", the default, it will match node name "gradients/foo", "foo/gradients/bar", but not "foo_gradients/"
string memory_optimizer_target_node_name_scope = 6; सार्वजनिक RewriterConfig.Builder setMetaOptimizerIterations ( RewriterConfig.NumIterationsType मान)
Controls how many times we run the optimizers in meta optimizer (default is once).
.tensorflow.RewriterConfig.NumIterationsType meta_optimizer_iterations = 12; सार्वजनिक रिवाइटरकॉन्फ़िग.बिल्डर सेटमेटाऑप्टिमाइज़रइटरेशन वैल्यू (इंट वैल्यू)
Controls how many times we run the optimizers in meta optimizer (default is once).
.tensorflow.RewriterConfig.NumIterationsType meta_optimizer_iterations = 12; सार्वजनिक रिवाइटरकॉन्फ़िग.बिल्डर सेटमेटाऑप्टिमाइज़रटाइमआउटएमएस (लंबा मान)
Maximum number of milliseconds to spend optimizing a single graph before timing out. If equal to 0 the system picks a default (currently 5 minutes). If less than 0 the optimizer will never time out.
int64 meta_optimizer_timeout_ms = 20; सार्वजनिक RewriterConfig.Builder setMinGraphNodes (int मान)
The minimum number of nodes in a graph to optimizer. For smaller graphs, optimization is skipped. 0 means the system picks an appropriate number. < 0 means do not skip optimization.
int32 min_graph_nodes = 17; सार्वजनिक रिवाइटरकॉन्फ़िग.बिल्डर सेटऑप्टिमाइज़र (इंट इंडेक्स, स्ट्रिंग मान)
If non-empty, will use this as an alternative way to specify a list of
optimizations to turn on and the order of the optimizations (replacing the
meta-optimizer).
Of the RewriterConfig options, only the AutoParallel configuration options
(the auto_parallel field) apply to manually requested optimization passes
("autoparallel"). Memory optimization passes ("memory") invoked here are
not configurable (in contrast to memory optimization passes through the
meta-optimizer) and act only on manual op annotations.
Custom optimizers (see custom_optimizers) that are not part of this
schedule will be run after - in the order that they were specified.
repeated string optimizers = 100; सार्वजनिक रिवाइटरकॉन्फ़िग.बिल्डर सेटपिनटूहोस्टऑप्टिमाइज़ेशन ( रीराइटरकॉन्फ़िग.टॉगल मान)
Force small ops onto the CPU (default is OFF).
.tensorflow.RewriterConfig.Toggle pin_to_host_optimization = 18; सार्वजनिक रिवाइटरकॉन्फ़िग.बिल्डर सेटपिनटूहोस्टऑप्टिमाइज़ेशनवैल्यू (इंट वैल्यू)
Force small ops onto the CPU (default is OFF).
.tensorflow.RewriterConfig.Toggle pin_to_host_optimization = 18; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर सेटपोस्टऑप्टिमाइजेशनवेरिफायरकॉन्फिग ( वेरिफायरकॉन्फिग मान)
VerifierConfig specifying the verifiers to be run at the end, after all optimizers have run.
.tensorflow.VerifierConfig post_optimization_verifier_config = 301; सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर सेटपोस्टऑप्टिमाइजेशनवेरिफायरकॉन्फिग ( VerifierConfig.Builder BuilderForValue)
VerifierConfig specifying the verifiers to be run at the end, after all optimizers have run.
.tensorflow.VerifierConfig post_optimization_verifier_config = 301; सार्वजनिक RewriterConfig.Builder setRemapping ( RewriterConfig.Toggle मान)
Remapping (default is ON) Remap subgraphs onto more efficient implementations.
.tensorflow.RewriterConfig.Toggle remapping = 14; सार्वजनिक RewriterConfig.Builder setRemappingValue (int मान)
Remapping (default is ON) Remap subgraphs onto more efficient implementations.
.tensorflow.RewriterConfig.Toggle remapping = 14; सार्वजनिक RewriterConfig.Builder setRepeatedField (com.google.protobuf.Descriptors.FieldDescriptor फ़ील्ड, int अनुक्रमणिका, ऑब्जेक्ट मान)
सार्वजनिक RewriterConfig.Builder setScopedAllocatorOptimization ( RewriterConfig.Toggle मान)
Try to allocate some independent Op outputs contiguously in order to merge or eliminate downstream Ops (off by default).
.tensorflow.RewriterConfig.Toggle scoped_allocator_optimization = 15; सार्वजनिक RewriterConfig.Builder setScopedAllocatorOptimizationValue (int मान)
Try to allocate some independent Op outputs contiguously in order to merge or eliminate downstream Ops (off by default).
.tensorflow.RewriterConfig.Toggle scoped_allocator_optimization = 15; सार्वजनिक RewriterConfig.Builder setScopedAllocatorOpts ( ScopedAllocatorOptions मान)
.tensorflow.ScopedAllocatorOptions scoped_allocator_opts = 16;
सार्वजनिक रिवाइटरकॉन्फिग.बिल्डर सेटस्कोप्डएलोकेटरऑप्ट्स ( स्कोप्डएलोकेटरऑप्शंस.बिल्डर बिल्डरफॉरवैल्यू)
.tensorflow.ScopedAllocatorOptions scoped_allocator_opts = 16;
सार्वजनिक RewriterConfig.Builder सेटशेपऑप्टिमाइज़ेशन ( RewriterConfig.Toggle मान)
Shape optimizations (default is ON) Simplify computations made on shapes.
.tensorflow.RewriterConfig.Toggle shape_optimization = 13; सार्वजनिक रिवाइटरकॉन्फ़िग.बिल्डर सेटशेपऑप्टिमाइज़ेशनवैल्यू (इंट वैल्यू)
Shape optimizations (default is ON) Simplify computations made on shapes.
.tensorflow.RewriterConfig.Toggle shape_optimization = 13;