Package onnx
Class OnnxMl.TrainingInfoProto
- java.lang.Object
-
- org.nd4j.shade.protobuf.AbstractMessageLite
-
- org.nd4j.shade.protobuf.AbstractMessage
-
- org.nd4j.shade.protobuf.GeneratedMessageV3
-
- onnx.OnnxMl.TrainingInfoProto
-
- All Implemented Interfaces:
Serializable
,OnnxMl.TrainingInfoProtoOrBuilder
,org.nd4j.shade.protobuf.Message
,org.nd4j.shade.protobuf.MessageLite
,org.nd4j.shade.protobuf.MessageLiteOrBuilder
,org.nd4j.shade.protobuf.MessageOrBuilder
- Enclosing class:
- OnnxMl
public static final class OnnxMl.TrainingInfoProto extends org.nd4j.shade.protobuf.GeneratedMessageV3 implements OnnxMl.TrainingInfoProtoOrBuilder
Training information TrainingInfoProto stores information for training a model. In particular, this defines two functionalities: an initialization-step and a training-algorithm-step. Initialization resets the model back to its original state as if no training has been performed. Training algorithm improves the model based on input data. The semantics of the initialization-step is that the initializers in ModelProto.graph and in TrainingInfoProto.algorithm are first initialized as specified by the initializers in the graph, and then updated by the "initialization_binding" in every instance in ModelProto.training_info. The field "algorithm" defines a computation graph which represents a training algorithm's step. After the execution of a TrainingInfoProto.algorithm, the initializers specified by "update_binding" may be immediately updated. If the targeted training algorithm contains consecutive update steps (such as block coordinate descent methods), the user needs to create a TrainingInfoProto for each step.
Protobuf typeonnx.TrainingInfoProto
- See Also:
- Serialized Form
-
-
Nested Class Summary
Nested Classes Modifier and Type Class Description static class
OnnxMl.TrainingInfoProto.Builder
Training information TrainingInfoProto stores information for training a model.-
Nested classes/interfaces inherited from class org.nd4j.shade.protobuf.GeneratedMessageV3
org.nd4j.shade.protobuf.GeneratedMessageV3.BuilderParent, org.nd4j.shade.protobuf.GeneratedMessageV3.ExtendableBuilder<MessageType extends org.nd4j.shade.protobuf.GeneratedMessageV3.ExtendableMessage,BuilderType extends org.nd4j.shade.protobuf.GeneratedMessageV3.ExtendableBuilder<MessageType,BuilderType>>, org.nd4j.shade.protobuf.GeneratedMessageV3.ExtendableMessage<MessageType extends org.nd4j.shade.protobuf.GeneratedMessageV3.ExtendableMessage>, org.nd4j.shade.protobuf.GeneratedMessageV3.ExtendableMessageOrBuilder<MessageType extends org.nd4j.shade.protobuf.GeneratedMessageV3.ExtendableMessage>, org.nd4j.shade.protobuf.GeneratedMessageV3.FieldAccessorTable, org.nd4j.shade.protobuf.GeneratedMessageV3.UnusedPrivateParameter
-
-
Field Summary
Fields Modifier and Type Field Description static int
ALGORITHM_FIELD_NUMBER
static int
INITIALIZATION_BINDING_FIELD_NUMBER
static int
INITIALIZATION_FIELD_NUMBER
static int
UPDATE_BINDING_FIELD_NUMBER
-
Method Summary
All Methods Static Methods Instance Methods Concrete Methods Modifier and Type Method Description boolean
equals(Object obj)
OnnxMl.GraphProto
getAlgorithm()
This field represents a training algorithm step.OnnxMl.GraphProtoOrBuilder
getAlgorithmOrBuilder()
This field represents a training algorithm step.static OnnxMl.TrainingInfoProto
getDefaultInstance()
OnnxMl.TrainingInfoProto
getDefaultInstanceForType()
static org.nd4j.shade.protobuf.Descriptors.Descriptor
getDescriptor()
OnnxMl.GraphProto
getInitialization()
This field describes a graph to compute the initial tensors upon starting the training process.OnnxMl.StringStringEntryProto
getInitializationBinding(int index)
This field specifies the bindings from the outputs of "initialization" to some initializers in "ModelProto.graph.initializer" and the "algorithm.initializer" in the same TrainingInfoProto.int
getInitializationBindingCount()
This field specifies the bindings from the outputs of "initialization" to some initializers in "ModelProto.graph.initializer" and the "algorithm.initializer" in the same TrainingInfoProto.List<OnnxMl.StringStringEntryProto>
getInitializationBindingList()
This field specifies the bindings from the outputs of "initialization" to some initializers in "ModelProto.graph.initializer" and the "algorithm.initializer" in the same TrainingInfoProto.OnnxMl.StringStringEntryProtoOrBuilder
getInitializationBindingOrBuilder(int index)
This field specifies the bindings from the outputs of "initialization" to some initializers in "ModelProto.graph.initializer" and the "algorithm.initializer" in the same TrainingInfoProto.List<? extends OnnxMl.StringStringEntryProtoOrBuilder>
getInitializationBindingOrBuilderList()
This field specifies the bindings from the outputs of "initialization" to some initializers in "ModelProto.graph.initializer" and the "algorithm.initializer" in the same TrainingInfoProto.OnnxMl.GraphProtoOrBuilder
getInitializationOrBuilder()
This field describes a graph to compute the initial tensors upon starting the training process.org.nd4j.shade.protobuf.Parser<OnnxMl.TrainingInfoProto>
getParserForType()
int
getSerializedSize()
org.nd4j.shade.protobuf.UnknownFieldSet
getUnknownFields()
OnnxMl.StringStringEntryProto
getUpdateBinding(int index)
Gradient-based training is usually an iterative procedure.int
getUpdateBindingCount()
Gradient-based training is usually an iterative procedure.List<OnnxMl.StringStringEntryProto>
getUpdateBindingList()
Gradient-based training is usually an iterative procedure.OnnxMl.StringStringEntryProtoOrBuilder
getUpdateBindingOrBuilder(int index)
Gradient-based training is usually an iterative procedure.List<? extends OnnxMl.StringStringEntryProtoOrBuilder>
getUpdateBindingOrBuilderList()
Gradient-based training is usually an iterative procedure.boolean
hasAlgorithm()
This field represents a training algorithm step.int
hashCode()
boolean
hasInitialization()
This field describes a graph to compute the initial tensors upon starting the training process.protected org.nd4j.shade.protobuf.GeneratedMessageV3.FieldAccessorTable
internalGetFieldAccessorTable()
boolean
isInitialized()
static OnnxMl.TrainingInfoProto.Builder
newBuilder()
static OnnxMl.TrainingInfoProto.Builder
newBuilder(OnnxMl.TrainingInfoProto prototype)
OnnxMl.TrainingInfoProto.Builder
newBuilderForType()
protected OnnxMl.TrainingInfoProto.Builder
newBuilderForType(org.nd4j.shade.protobuf.GeneratedMessageV3.BuilderParent parent)
protected Object
newInstance(org.nd4j.shade.protobuf.GeneratedMessageV3.UnusedPrivateParameter unused)
static OnnxMl.TrainingInfoProto
parseDelimitedFrom(InputStream input)
static OnnxMl.TrainingInfoProto
parseDelimitedFrom(InputStream input, org.nd4j.shade.protobuf.ExtensionRegistryLite extensionRegistry)
static OnnxMl.TrainingInfoProto
parseFrom(byte[] data)
static OnnxMl.TrainingInfoProto
parseFrom(byte[] data, org.nd4j.shade.protobuf.ExtensionRegistryLite extensionRegistry)
static OnnxMl.TrainingInfoProto
parseFrom(InputStream input)
static OnnxMl.TrainingInfoProto
parseFrom(InputStream input, org.nd4j.shade.protobuf.ExtensionRegistryLite extensionRegistry)
static OnnxMl.TrainingInfoProto
parseFrom(ByteBuffer data)
static OnnxMl.TrainingInfoProto
parseFrom(ByteBuffer data, org.nd4j.shade.protobuf.ExtensionRegistryLite extensionRegistry)
static OnnxMl.TrainingInfoProto
parseFrom(org.nd4j.shade.protobuf.ByteString data)
static OnnxMl.TrainingInfoProto
parseFrom(org.nd4j.shade.protobuf.ByteString data, org.nd4j.shade.protobuf.ExtensionRegistryLite extensionRegistry)
static OnnxMl.TrainingInfoProto
parseFrom(org.nd4j.shade.protobuf.CodedInputStream input)
static OnnxMl.TrainingInfoProto
parseFrom(org.nd4j.shade.protobuf.CodedInputStream input, org.nd4j.shade.protobuf.ExtensionRegistryLite extensionRegistry)
static org.nd4j.shade.protobuf.Parser<OnnxMl.TrainingInfoProto>
parser()
OnnxMl.TrainingInfoProto.Builder
toBuilder()
void
writeTo(org.nd4j.shade.protobuf.CodedOutputStream output)
-
Methods inherited from class org.nd4j.shade.protobuf.GeneratedMessageV3
canUseUnsafe, computeStringSize, computeStringSizeNoTag, emptyBooleanList, emptyDoubleList, emptyFloatList, emptyIntList, emptyLongList, getAllFields, getDescriptorForType, getField, getOneofFieldDescriptor, getRepeatedField, getRepeatedFieldCount, hasField, hasOneof, internalGetMapField, isStringEmpty, makeExtensionsImmutable, mergeFromAndMakeImmutableInternal, mutableCopy, mutableCopy, mutableCopy, mutableCopy, mutableCopy, newBooleanList, newBuilderForType, newDoubleList, newFloatList, newIntList, newLongList, parseDelimitedWithIOException, parseDelimitedWithIOException, parseUnknownField, parseUnknownFieldProto3, parseWithIOException, parseWithIOException, parseWithIOException, parseWithIOException, serializeBooleanMapTo, serializeIntegerMapTo, serializeLongMapTo, serializeStringMapTo, writeReplace, writeString, writeStringNoTag
-
Methods inherited from class org.nd4j.shade.protobuf.AbstractMessage
findInitializationErrors, getInitializationErrorString, hashBoolean, hashEnum, hashEnumList, hashFields, hashLong, toString
-
Methods inherited from class org.nd4j.shade.protobuf.AbstractMessageLite
addAll, addAll, checkByteStringIsUtf8, toByteArray, toByteString, writeDelimitedTo, writeTo
-
Methods inherited from class java.lang.Object
clone, finalize, getClass, notify, notifyAll, wait, wait, wait
-
-
-
-
Field Detail
-
INITIALIZATION_FIELD_NUMBER
public static final int INITIALIZATION_FIELD_NUMBER
- See Also:
- Constant Field Values
-
ALGORITHM_FIELD_NUMBER
public static final int ALGORITHM_FIELD_NUMBER
- See Also:
- Constant Field Values
-
INITIALIZATION_BINDING_FIELD_NUMBER
public static final int INITIALIZATION_BINDING_FIELD_NUMBER
- See Also:
- Constant Field Values
-
UPDATE_BINDING_FIELD_NUMBER
public static final int UPDATE_BINDING_FIELD_NUMBER
- See Also:
- Constant Field Values
-
-
Method Detail
-
newInstance
protected Object newInstance(org.nd4j.shade.protobuf.GeneratedMessageV3.UnusedPrivateParameter unused)
- Overrides:
newInstance
in classorg.nd4j.shade.protobuf.GeneratedMessageV3
-
getUnknownFields
public final org.nd4j.shade.protobuf.UnknownFieldSet getUnknownFields()
- Specified by:
getUnknownFields
in interfaceorg.nd4j.shade.protobuf.MessageOrBuilder
- Overrides:
getUnknownFields
in classorg.nd4j.shade.protobuf.GeneratedMessageV3
-
getDescriptor
public static final org.nd4j.shade.protobuf.Descriptors.Descriptor getDescriptor()
-
internalGetFieldAccessorTable
protected org.nd4j.shade.protobuf.GeneratedMessageV3.FieldAccessorTable internalGetFieldAccessorTable()
- Specified by:
internalGetFieldAccessorTable
in classorg.nd4j.shade.protobuf.GeneratedMessageV3
-
hasInitialization
public boolean hasInitialization()
This field describes a graph to compute the initial tensors upon starting the training process. Initialization graph has no input and can have multiple outputs. Usually, trainable tensors in neural networks are randomly initialized. To achieve that, for each tensor, the user can put a random number operator such as RandomNormal or RandomUniform in TrainingInfoProto.initialization.node and assign its random output to the specific tensor using "initialization_binding". This graph can also set the initializers in "algorithm" in the same TrainingInfoProto; a use case is resetting the number of training iteration to zero. By default, this field is an empty graph and its evaluation does not produce any output. Thus, no initializer would be changed by default.
.onnx.GraphProto initialization = 1;
- Specified by:
hasInitialization
in interfaceOnnxMl.TrainingInfoProtoOrBuilder
- Returns:
- Whether the initialization field is set.
-
getInitialization
public OnnxMl.GraphProto getInitialization()
This field describes a graph to compute the initial tensors upon starting the training process. Initialization graph has no input and can have multiple outputs. Usually, trainable tensors in neural networks are randomly initialized. To achieve that, for each tensor, the user can put a random number operator such as RandomNormal or RandomUniform in TrainingInfoProto.initialization.node and assign its random output to the specific tensor using "initialization_binding". This graph can also set the initializers in "algorithm" in the same TrainingInfoProto; a use case is resetting the number of training iteration to zero. By default, this field is an empty graph and its evaluation does not produce any output. Thus, no initializer would be changed by default.
.onnx.GraphProto initialization = 1;
- Specified by:
getInitialization
in interfaceOnnxMl.TrainingInfoProtoOrBuilder
- Returns:
- The initialization.
-
getInitializationOrBuilder
public OnnxMl.GraphProtoOrBuilder getInitializationOrBuilder()
This field describes a graph to compute the initial tensors upon starting the training process. Initialization graph has no input and can have multiple outputs. Usually, trainable tensors in neural networks are randomly initialized. To achieve that, for each tensor, the user can put a random number operator such as RandomNormal or RandomUniform in TrainingInfoProto.initialization.node and assign its random output to the specific tensor using "initialization_binding". This graph can also set the initializers in "algorithm" in the same TrainingInfoProto; a use case is resetting the number of training iteration to zero. By default, this field is an empty graph and its evaluation does not produce any output. Thus, no initializer would be changed by default.
.onnx.GraphProto initialization = 1;
- Specified by:
getInitializationOrBuilder
in interfaceOnnxMl.TrainingInfoProtoOrBuilder
-
hasAlgorithm
public boolean hasAlgorithm()
This field represents a training algorithm step. Given required inputs, it computes outputs to update initializers in its own or inference graph's initializer lists. In general, this field contains loss node, gradient node, optimizer node, increment of iteration count. An execution of the training algorithm step is performed by executing the graph obtained by combining the inference graph (namely "ModelProto.graph") and the "algorithm" graph. That is, the actual the actual input/initializer/output/node/value_info/sparse_initializer list of the training graph is the concatenation of "ModelProto.graph.input/initializer/output/node/value_info/sparse_initializer" and "algorithm.input/initializer/output/node/value_info/sparse_initializer" in that order. This combined graph must satisfy the normal ONNX conditions. Now, let's provide a visualization of graph combination for clarity. Let the inference graph (i.e., "ModelProto.graph") be tensor_a, tensor_b -> MatMul -> tensor_c -> Sigmoid -> tensor_d and the "algorithm" graph be tensor_d -> Add -> tensor_e The combination process results tensor_a, tensor_b -> MatMul -> tensor_c -> Sigmoid -> tensor_d -> Add -> tensor_e Notice that an input of a node in the "algorithm" graph may reference the output of a node in the inference graph (but not the other way round). Also, inference node cannot reference inputs of "algorithm". With these restrictions, inference graph can always be run independently without training information. By default, this field is an empty graph and its evaluation does not produce any output. Evaluating the default training step never update any initializers.
.onnx.GraphProto algorithm = 2;
- Specified by:
hasAlgorithm
in interfaceOnnxMl.TrainingInfoProtoOrBuilder
- Returns:
- Whether the algorithm field is set.
-
getAlgorithm
public OnnxMl.GraphProto getAlgorithm()
This field represents a training algorithm step. Given required inputs, it computes outputs to update initializers in its own or inference graph's initializer lists. In general, this field contains loss node, gradient node, optimizer node, increment of iteration count. An execution of the training algorithm step is performed by executing the graph obtained by combining the inference graph (namely "ModelProto.graph") and the "algorithm" graph. That is, the actual the actual input/initializer/output/node/value_info/sparse_initializer list of the training graph is the concatenation of "ModelProto.graph.input/initializer/output/node/value_info/sparse_initializer" and "algorithm.input/initializer/output/node/value_info/sparse_initializer" in that order. This combined graph must satisfy the normal ONNX conditions. Now, let's provide a visualization of graph combination for clarity. Let the inference graph (i.e., "ModelProto.graph") be tensor_a, tensor_b -> MatMul -> tensor_c -> Sigmoid -> tensor_d and the "algorithm" graph be tensor_d -> Add -> tensor_e The combination process results tensor_a, tensor_b -> MatMul -> tensor_c -> Sigmoid -> tensor_d -> Add -> tensor_e Notice that an input of a node in the "algorithm" graph may reference the output of a node in the inference graph (but not the other way round). Also, inference node cannot reference inputs of "algorithm". With these restrictions, inference graph can always be run independently without training information. By default, this field is an empty graph and its evaluation does not produce any output. Evaluating the default training step never update any initializers.
.onnx.GraphProto algorithm = 2;
- Specified by:
getAlgorithm
in interfaceOnnxMl.TrainingInfoProtoOrBuilder
- Returns:
- The algorithm.
-
getAlgorithmOrBuilder
public OnnxMl.GraphProtoOrBuilder getAlgorithmOrBuilder()
This field represents a training algorithm step. Given required inputs, it computes outputs to update initializers in its own or inference graph's initializer lists. In general, this field contains loss node, gradient node, optimizer node, increment of iteration count. An execution of the training algorithm step is performed by executing the graph obtained by combining the inference graph (namely "ModelProto.graph") and the "algorithm" graph. That is, the actual the actual input/initializer/output/node/value_info/sparse_initializer list of the training graph is the concatenation of "ModelProto.graph.input/initializer/output/node/value_info/sparse_initializer" and "algorithm.input/initializer/output/node/value_info/sparse_initializer" in that order. This combined graph must satisfy the normal ONNX conditions. Now, let's provide a visualization of graph combination for clarity. Let the inference graph (i.e., "ModelProto.graph") be tensor_a, tensor_b -> MatMul -> tensor_c -> Sigmoid -> tensor_d and the "algorithm" graph be tensor_d -> Add -> tensor_e The combination process results tensor_a, tensor_b -> MatMul -> tensor_c -> Sigmoid -> tensor_d -> Add -> tensor_e Notice that an input of a node in the "algorithm" graph may reference the output of a node in the inference graph (but not the other way round). Also, inference node cannot reference inputs of "algorithm". With these restrictions, inference graph can always be run independently without training information. By default, this field is an empty graph and its evaluation does not produce any output. Evaluating the default training step never update any initializers.
.onnx.GraphProto algorithm = 2;
- Specified by:
getAlgorithmOrBuilder
in interfaceOnnxMl.TrainingInfoProtoOrBuilder
-
getInitializationBindingList
public List<OnnxMl.StringStringEntryProto> getInitializationBindingList()
This field specifies the bindings from the outputs of "initialization" to some initializers in "ModelProto.graph.initializer" and the "algorithm.initializer" in the same TrainingInfoProto. See "update_binding" below for details. By default, this field is empty and no initializer would be changed by the execution of "initialization".
repeated .onnx.StringStringEntryProto initialization_binding = 3;
- Specified by:
getInitializationBindingList
in interfaceOnnxMl.TrainingInfoProtoOrBuilder
-
getInitializationBindingOrBuilderList
public List<? extends OnnxMl.StringStringEntryProtoOrBuilder> getInitializationBindingOrBuilderList()
This field specifies the bindings from the outputs of "initialization" to some initializers in "ModelProto.graph.initializer" and the "algorithm.initializer" in the same TrainingInfoProto. See "update_binding" below for details. By default, this field is empty and no initializer would be changed by the execution of "initialization".
repeated .onnx.StringStringEntryProto initialization_binding = 3;
- Specified by:
getInitializationBindingOrBuilderList
in interfaceOnnxMl.TrainingInfoProtoOrBuilder
-
getInitializationBindingCount
public int getInitializationBindingCount()
This field specifies the bindings from the outputs of "initialization" to some initializers in "ModelProto.graph.initializer" and the "algorithm.initializer" in the same TrainingInfoProto. See "update_binding" below for details. By default, this field is empty and no initializer would be changed by the execution of "initialization".
repeated .onnx.StringStringEntryProto initialization_binding = 3;
- Specified by:
getInitializationBindingCount
in interfaceOnnxMl.TrainingInfoProtoOrBuilder
-
getInitializationBinding
public OnnxMl.StringStringEntryProto getInitializationBinding(int index)
This field specifies the bindings from the outputs of "initialization" to some initializers in "ModelProto.graph.initializer" and the "algorithm.initializer" in the same TrainingInfoProto. See "update_binding" below for details. By default, this field is empty and no initializer would be changed by the execution of "initialization".
repeated .onnx.StringStringEntryProto initialization_binding = 3;
- Specified by:
getInitializationBinding
in interfaceOnnxMl.TrainingInfoProtoOrBuilder
-
getInitializationBindingOrBuilder
public OnnxMl.StringStringEntryProtoOrBuilder getInitializationBindingOrBuilder(int index)
This field specifies the bindings from the outputs of "initialization" to some initializers in "ModelProto.graph.initializer" and the "algorithm.initializer" in the same TrainingInfoProto. See "update_binding" below for details. By default, this field is empty and no initializer would be changed by the execution of "initialization".
repeated .onnx.StringStringEntryProto initialization_binding = 3;
- Specified by:
getInitializationBindingOrBuilder
in interfaceOnnxMl.TrainingInfoProtoOrBuilder
-
getUpdateBindingList
public List<OnnxMl.StringStringEntryProto> getUpdateBindingList()
Gradient-based training is usually an iterative procedure. In one gradient descent iteration, we apply x = x - r * g where "x" is the optimized tensor, "r" stands for learning rate, and "g" is gradient of "x" with respect to a chosen loss. To avoid adding assignments into the training graph, we split the update equation into y = x - r * g x = y The user needs to save "y = x - r * g" into TrainingInfoProto.algorithm. To tell that "y" should be assigned to "x", the field "update_binding" may contain a key-value pair of strings, "x" (key of StringStringEntryProto) and "y" (value of StringStringEntryProto). For a neural network with multiple trainable (mutable) tensors, there can be multiple key-value pairs in "update_binding". The initializers appears as keys in "update_binding" are considered mutable variables. This implies some behaviors as described below. 1. We have only unique keys in all "update_binding"s so that two variables may not have the same name. This ensures that one variable is assigned up to once. 2. The keys must appear in names of "ModelProto.graph.initializer" or "TrainingInfoProto.algorithm.initializer". 3. The values must be output names of "algorithm" or "ModelProto.graph.output". 4. Mutable variables are initialized to the value specified by the corresponding initializer, and then potentially updated by "initializer_binding"s and "update_binding"s in "TrainingInfoProto"s. This field usually contains names of trainable tensors (in ModelProto.graph), optimizer states such as momentums in advanced stochastic gradient methods (in TrainingInfoProto.graph), and number of training iterations (in TrainingInfoProto.graph). By default, this field is empty and no initializer would be changed by the execution of "algorithm".
repeated .onnx.StringStringEntryProto update_binding = 4;
- Specified by:
getUpdateBindingList
in interfaceOnnxMl.TrainingInfoProtoOrBuilder
-
getUpdateBindingOrBuilderList
public List<? extends OnnxMl.StringStringEntryProtoOrBuilder> getUpdateBindingOrBuilderList()
Gradient-based training is usually an iterative procedure. In one gradient descent iteration, we apply x = x - r * g where "x" is the optimized tensor, "r" stands for learning rate, and "g" is gradient of "x" with respect to a chosen loss. To avoid adding assignments into the training graph, we split the update equation into y = x - r * g x = y The user needs to save "y = x - r * g" into TrainingInfoProto.algorithm. To tell that "y" should be assigned to "x", the field "update_binding" may contain a key-value pair of strings, "x" (key of StringStringEntryProto) and "y" (value of StringStringEntryProto). For a neural network with multiple trainable (mutable) tensors, there can be multiple key-value pairs in "update_binding". The initializers appears as keys in "update_binding" are considered mutable variables. This implies some behaviors as described below. 1. We have only unique keys in all "update_binding"s so that two variables may not have the same name. This ensures that one variable is assigned up to once. 2. The keys must appear in names of "ModelProto.graph.initializer" or "TrainingInfoProto.algorithm.initializer". 3. The values must be output names of "algorithm" or "ModelProto.graph.output". 4. Mutable variables are initialized to the value specified by the corresponding initializer, and then potentially updated by "initializer_binding"s and "update_binding"s in "TrainingInfoProto"s. This field usually contains names of trainable tensors (in ModelProto.graph), optimizer states such as momentums in advanced stochastic gradient methods (in TrainingInfoProto.graph), and number of training iterations (in TrainingInfoProto.graph). By default, this field is empty and no initializer would be changed by the execution of "algorithm".
repeated .onnx.StringStringEntryProto update_binding = 4;
- Specified by:
getUpdateBindingOrBuilderList
in interfaceOnnxMl.TrainingInfoProtoOrBuilder
-
getUpdateBindingCount
public int getUpdateBindingCount()
Gradient-based training is usually an iterative procedure. In one gradient descent iteration, we apply x = x - r * g where "x" is the optimized tensor, "r" stands for learning rate, and "g" is gradient of "x" with respect to a chosen loss. To avoid adding assignments into the training graph, we split the update equation into y = x - r * g x = y The user needs to save "y = x - r * g" into TrainingInfoProto.algorithm. To tell that "y" should be assigned to "x", the field "update_binding" may contain a key-value pair of strings, "x" (key of StringStringEntryProto) and "y" (value of StringStringEntryProto). For a neural network with multiple trainable (mutable) tensors, there can be multiple key-value pairs in "update_binding". The initializers appears as keys in "update_binding" are considered mutable variables. This implies some behaviors as described below. 1. We have only unique keys in all "update_binding"s so that two variables may not have the same name. This ensures that one variable is assigned up to once. 2. The keys must appear in names of "ModelProto.graph.initializer" or "TrainingInfoProto.algorithm.initializer". 3. The values must be output names of "algorithm" or "ModelProto.graph.output". 4. Mutable variables are initialized to the value specified by the corresponding initializer, and then potentially updated by "initializer_binding"s and "update_binding"s in "TrainingInfoProto"s. This field usually contains names of trainable tensors (in ModelProto.graph), optimizer states such as momentums in advanced stochastic gradient methods (in TrainingInfoProto.graph), and number of training iterations (in TrainingInfoProto.graph). By default, this field is empty and no initializer would be changed by the execution of "algorithm".
repeated .onnx.StringStringEntryProto update_binding = 4;
- Specified by:
getUpdateBindingCount
in interfaceOnnxMl.TrainingInfoProtoOrBuilder
-
getUpdateBinding
public OnnxMl.StringStringEntryProto getUpdateBinding(int index)
Gradient-based training is usually an iterative procedure. In one gradient descent iteration, we apply x = x - r * g where "x" is the optimized tensor, "r" stands for learning rate, and "g" is gradient of "x" with respect to a chosen loss. To avoid adding assignments into the training graph, we split the update equation into y = x - r * g x = y The user needs to save "y = x - r * g" into TrainingInfoProto.algorithm. To tell that "y" should be assigned to "x", the field "update_binding" may contain a key-value pair of strings, "x" (key of StringStringEntryProto) and "y" (value of StringStringEntryProto). For a neural network with multiple trainable (mutable) tensors, there can be multiple key-value pairs in "update_binding". The initializers appears as keys in "update_binding" are considered mutable variables. This implies some behaviors as described below. 1. We have only unique keys in all "update_binding"s so that two variables may not have the same name. This ensures that one variable is assigned up to once. 2. The keys must appear in names of "ModelProto.graph.initializer" or "TrainingInfoProto.algorithm.initializer". 3. The values must be output names of "algorithm" or "ModelProto.graph.output". 4. Mutable variables are initialized to the value specified by the corresponding initializer, and then potentially updated by "initializer_binding"s and "update_binding"s in "TrainingInfoProto"s. This field usually contains names of trainable tensors (in ModelProto.graph), optimizer states such as momentums in advanced stochastic gradient methods (in TrainingInfoProto.graph), and number of training iterations (in TrainingInfoProto.graph). By default, this field is empty and no initializer would be changed by the execution of "algorithm".
repeated .onnx.StringStringEntryProto update_binding = 4;
- Specified by:
getUpdateBinding
in interfaceOnnxMl.TrainingInfoProtoOrBuilder
-
getUpdateBindingOrBuilder
public OnnxMl.StringStringEntryProtoOrBuilder getUpdateBindingOrBuilder(int index)
Gradient-based training is usually an iterative procedure. In one gradient descent iteration, we apply x = x - r * g where "x" is the optimized tensor, "r" stands for learning rate, and "g" is gradient of "x" with respect to a chosen loss. To avoid adding assignments into the training graph, we split the update equation into y = x - r * g x = y The user needs to save "y = x - r * g" into TrainingInfoProto.algorithm. To tell that "y" should be assigned to "x", the field "update_binding" may contain a key-value pair of strings, "x" (key of StringStringEntryProto) and "y" (value of StringStringEntryProto). For a neural network with multiple trainable (mutable) tensors, there can be multiple key-value pairs in "update_binding". The initializers appears as keys in "update_binding" are considered mutable variables. This implies some behaviors as described below. 1. We have only unique keys in all "update_binding"s so that two variables may not have the same name. This ensures that one variable is assigned up to once. 2. The keys must appear in names of "ModelProto.graph.initializer" or "TrainingInfoProto.algorithm.initializer". 3. The values must be output names of "algorithm" or "ModelProto.graph.output". 4. Mutable variables are initialized to the value specified by the corresponding initializer, and then potentially updated by "initializer_binding"s and "update_binding"s in "TrainingInfoProto"s. This field usually contains names of trainable tensors (in ModelProto.graph), optimizer states such as momentums in advanced stochastic gradient methods (in TrainingInfoProto.graph), and number of training iterations (in TrainingInfoProto.graph). By default, this field is empty and no initializer would be changed by the execution of "algorithm".
repeated .onnx.StringStringEntryProto update_binding = 4;
- Specified by:
getUpdateBindingOrBuilder
in interfaceOnnxMl.TrainingInfoProtoOrBuilder
-
isInitialized
public final boolean isInitialized()
- Specified by:
isInitialized
in interfaceorg.nd4j.shade.protobuf.MessageLiteOrBuilder
- Overrides:
isInitialized
in classorg.nd4j.shade.protobuf.GeneratedMessageV3
-
writeTo
public void writeTo(org.nd4j.shade.protobuf.CodedOutputStream output) throws IOException
- Specified by:
writeTo
in interfaceorg.nd4j.shade.protobuf.MessageLite
- Overrides:
writeTo
in classorg.nd4j.shade.protobuf.GeneratedMessageV3
- Throws:
IOException
-
getSerializedSize
public int getSerializedSize()
- Specified by:
getSerializedSize
in interfaceorg.nd4j.shade.protobuf.MessageLite
- Overrides:
getSerializedSize
in classorg.nd4j.shade.protobuf.GeneratedMessageV3
-
equals
public boolean equals(Object obj)
- Specified by:
equals
in interfaceorg.nd4j.shade.protobuf.Message
- Overrides:
equals
in classorg.nd4j.shade.protobuf.AbstractMessage
-
hashCode
public int hashCode()
- Specified by:
hashCode
in interfaceorg.nd4j.shade.protobuf.Message
- Overrides:
hashCode
in classorg.nd4j.shade.protobuf.AbstractMessage
-
parseFrom
public static OnnxMl.TrainingInfoProto parseFrom(ByteBuffer data) throws org.nd4j.shade.protobuf.InvalidProtocolBufferException
- Throws:
org.nd4j.shade.protobuf.InvalidProtocolBufferException
-
parseFrom
public static OnnxMl.TrainingInfoProto parseFrom(ByteBuffer data, org.nd4j.shade.protobuf.ExtensionRegistryLite extensionRegistry) throws org.nd4j.shade.protobuf.InvalidProtocolBufferException
- Throws:
org.nd4j.shade.protobuf.InvalidProtocolBufferException
-
parseFrom
public static OnnxMl.TrainingInfoProto parseFrom(org.nd4j.shade.protobuf.ByteString data) throws org.nd4j.shade.protobuf.InvalidProtocolBufferException
- Throws:
org.nd4j.shade.protobuf.InvalidProtocolBufferException
-
parseFrom
public static OnnxMl.TrainingInfoProto parseFrom(org.nd4j.shade.protobuf.ByteString data, org.nd4j.shade.protobuf.ExtensionRegistryLite extensionRegistry) throws org.nd4j.shade.protobuf.InvalidProtocolBufferException
- Throws:
org.nd4j.shade.protobuf.InvalidProtocolBufferException
-
parseFrom
public static OnnxMl.TrainingInfoProto parseFrom(byte[] data) throws org.nd4j.shade.protobuf.InvalidProtocolBufferException
- Throws:
org.nd4j.shade.protobuf.InvalidProtocolBufferException
-
parseFrom
public static OnnxMl.TrainingInfoProto parseFrom(byte[] data, org.nd4j.shade.protobuf.ExtensionRegistryLite extensionRegistry) throws org.nd4j.shade.protobuf.InvalidProtocolBufferException
- Throws:
org.nd4j.shade.protobuf.InvalidProtocolBufferException
-
parseFrom
public static OnnxMl.TrainingInfoProto parseFrom(InputStream input) throws IOException
- Throws:
IOException
-
parseFrom
public static OnnxMl.TrainingInfoProto parseFrom(InputStream input, org.nd4j.shade.protobuf.ExtensionRegistryLite extensionRegistry) throws IOException
- Throws:
IOException
-
parseDelimitedFrom
public static OnnxMl.TrainingInfoProto parseDelimitedFrom(InputStream input) throws IOException
- Throws:
IOException
-
parseDelimitedFrom
public static OnnxMl.TrainingInfoProto parseDelimitedFrom(InputStream input, org.nd4j.shade.protobuf.ExtensionRegistryLite extensionRegistry) throws IOException
- Throws:
IOException
-
parseFrom
public static OnnxMl.TrainingInfoProto parseFrom(org.nd4j.shade.protobuf.CodedInputStream input) throws IOException
- Throws:
IOException
-
parseFrom
public static OnnxMl.TrainingInfoProto parseFrom(org.nd4j.shade.protobuf.CodedInputStream input, org.nd4j.shade.protobuf.ExtensionRegistryLite extensionRegistry) throws IOException
- Throws:
IOException
-
newBuilderForType
public OnnxMl.TrainingInfoProto.Builder newBuilderForType()
- Specified by:
newBuilderForType
in interfaceorg.nd4j.shade.protobuf.Message
- Specified by:
newBuilderForType
in interfaceorg.nd4j.shade.protobuf.MessageLite
-
newBuilder
public static OnnxMl.TrainingInfoProto.Builder newBuilder()
-
newBuilder
public static OnnxMl.TrainingInfoProto.Builder newBuilder(OnnxMl.TrainingInfoProto prototype)
-
toBuilder
public OnnxMl.TrainingInfoProto.Builder toBuilder()
- Specified by:
toBuilder
in interfaceorg.nd4j.shade.protobuf.Message
- Specified by:
toBuilder
in interfaceorg.nd4j.shade.protobuf.MessageLite
-
newBuilderForType
protected OnnxMl.TrainingInfoProto.Builder newBuilderForType(org.nd4j.shade.protobuf.GeneratedMessageV3.BuilderParent parent)
- Specified by:
newBuilderForType
in classorg.nd4j.shade.protobuf.GeneratedMessageV3
-
getDefaultInstance
public static OnnxMl.TrainingInfoProto getDefaultInstance()
-
parser
public static org.nd4j.shade.protobuf.Parser<OnnxMl.TrainingInfoProto> parser()
-
getParserForType
public org.nd4j.shade.protobuf.Parser<OnnxMl.TrainingInfoProto> getParserForType()
- Specified by:
getParserForType
in interfaceorg.nd4j.shade.protobuf.Message
- Specified by:
getParserForType
in interfaceorg.nd4j.shade.protobuf.MessageLite
- Overrides:
getParserForType
in classorg.nd4j.shade.protobuf.GeneratedMessageV3
-
getDefaultInstanceForType
public OnnxMl.TrainingInfoProto getDefaultInstanceForType()
- Specified by:
getDefaultInstanceForType
in interfaceorg.nd4j.shade.protobuf.MessageLiteOrBuilder
- Specified by:
getDefaultInstanceForType
in interfaceorg.nd4j.shade.protobuf.MessageOrBuilder
-
-