Libraries/Microsoft.ML.UniversalModelFormat.xml
|
<?xml version="1.0"?>
<doc> <assembly> <name>Microsoft.ML.UniversalModelFormat</name> </assembly> <members> <member name="T:LotusIR.GraphReflection"> <summary>Holder for reflection information generated from graph.proto</summary> </member> <member name="P:LotusIR.GraphReflection.Descriptor"> <summary>File descriptor for graph.proto</summary> </member> <member name="T:LotusIR.Version"> <summary> To be compatible with both proto2 and proto3, we will use a version number that is not defined by the default value but an explicit enum number. </summary> </member> <member name="F:LotusIR.Version.None"> <summary> Just for proto3 compatibility </summary> </member> <member name="F:LotusIR.Version.IrVersion"> <summary> The version field is always serialized and we will use it to store the version that the graph is generated from. This helps us set up version control. We should use version as xx(major) - xx(minor) - xxxx(bugfix) and we are starting with 00000001. </summary> </member> <member name="T:LotusIR.AttributeProto"> <summary> A named attribute containing either singular float, integer, string and tensor values, or repeated float, integer, string and tensor values. An AttributeProto MUST contain the name field, and *only one* of the following content fields, effectively enforcing a C/C++ union equivalent. </summary> </member> <member name="F:LotusIR.AttributeProto.NameFieldNumber"> <summary>Field number for the "name" field.</summary> </member> <member name="P:LotusIR.AttributeProto.Name"> <summary> The name field MUST be present for this version of the IR. </summary> </member> <member name="F:LotusIR.AttributeProto.FFieldNumber"> <summary>Field number for the "f" field.</summary> </member> <member name="P:LotusIR.AttributeProto.F"> <summary> float </summary> </member> <member name="F:LotusIR.AttributeProto.IFieldNumber"> <summary>Field number for the "i" field.</summary> </member> <member name="P:LotusIR.AttributeProto.I"> <summary> int </summary> </member> <member name="F:LotusIR.AttributeProto.SFieldNumber"> <summary>Field number for the "s" field.</summary> </member> <member name="P:LotusIR.AttributeProto.S"> <summary> UTF-8 string </summary> </member> <member name="F:LotusIR.AttributeProto.TFieldNumber"> <summary>Field number for the "t" field.</summary> </member> <member name="P:LotusIR.AttributeProto.T"> <summary> tensor value </summary> </member> <member name="F:LotusIR.AttributeProto.GFieldNumber"> <summary>Field number for the "g" field.</summary> </member> <member name="P:LotusIR.AttributeProto.G"> <summary> graph </summary> </member> <member name="F:LotusIR.AttributeProto.FloatsFieldNumber"> <summary>Field number for the "floats" field.</summary> </member> <member name="P:LotusIR.AttributeProto.Floats"> <summary> list of floats </summary> </member> <member name="F:LotusIR.AttributeProto.IntsFieldNumber"> <summary>Field number for the "ints" field.</summary> </member> <member name="P:LotusIR.AttributeProto.Ints"> <summary> list of ints </summary> </member> <member name="F:LotusIR.AttributeProto.StringsFieldNumber"> <summary>Field number for the "strings" field.</summary> </member> <member name="P:LotusIR.AttributeProto.Strings"> <summary> list of UTF-8 strings </summary> </member> <member name="F:LotusIR.AttributeProto.TensorsFieldNumber"> <summary>Field number for the "tensors" field.</summary> </member> <member name="P:LotusIR.AttributeProto.Tensors"> <summary> list of tensors </summary> </member> <member name="F:LotusIR.AttributeProto.GraphsFieldNumber"> <summary>Field number for the "graphs" field.</summary> </member> <member name="P:LotusIR.AttributeProto.Graphs"> <summary> list of graph </summary> </member> <member name="F:LotusIR.AttributeProto.TypeFieldNumber"> <summary>Field number for the "type" field.</summary> </member> <member name="F:LotusIR.AttributeProto.Types_FieldNumber"> <summary>Field number for the "types" field.</summary> </member> <member name="F:LotusIR.AttributeProto.ShapeFieldNumber"> <summary>Field number for the "shape" field.</summary> </member> <member name="P:LotusIR.AttributeProto.Shape"> <summary> ISSUE:13807134,dbox: Do we ever see shape showing up as an attribute value? If so, won't it always be accompanied by a TypeProto? </summary> </member> <member name="F:LotusIR.AttributeProto.ShapesFieldNumber"> <summary>Field number for the "shapes" field.</summary> </member> <member name="T:LotusIR.ValueInfoProto"> <summary> Defines information on value, including the name, the type, and the shape of the value. </summary> </member> <member name="F:LotusIR.ValueInfoProto.NameFieldNumber"> <summary>Field number for the "name" field.</summary> </member> <member name="P:LotusIR.ValueInfoProto.Name"> <summary> This field MUST be present in this version of the IR. </summary> </member> <member name="F:LotusIR.ValueInfoProto.TypeFieldNumber"> <summary>Field number for the "type" field.</summary> </member> <member name="P:LotusIR.ValueInfoProto.Type"> <summary> This field MUST be present in this version of the IR. </summary> </member> <member name="F:LotusIR.ValueInfoProto.DocStringFieldNumber"> <summary>Field number for the "doc_string" field.</summary> </member> <member name="P:LotusIR.ValueInfoProto.DocString"> <summary> An human-readable documentation for this node in the graph. This text MAY contain Markdown markup that conforms to http://commonmark.org/. </summary> </member> <member name="T:LotusIR.NodeProto"> <summary> Defines a node in a computation graph. Each graph node is either an operator or a function call. A node that is similar to the notion of "layer" or "operator" in many deep learning frameworks. For example, it can be a node of type "Conv" that takes in an image, a filter tensor and a bias tensor, and produces the convolved output. NOTE: Control flow is defined by two built-in operators: Cond(p, true_input, false_input) takes three inputs, where p is a boolean scalar tensor, true_input is the list of inputs to the true branch of cond, and false_input is the list of inputs to the false branch of cond. The true and false branches are defined as functions that takes true_input and false_input as inputs respectively. The two functions must have the same number of outputs, and each corresponding output must have the same types, and have compatible shapes. While(vars, consts) takes two inputs, where vars are the initial values of the loop variables and consts are the values of constants used inside the loop. The loop condition and loop body are defined as functions. The functions take both vars and consts as inputs. The loop condition function returns a boolean scalar tensor. The loop body function has the form: body(vars, consts) = new_vars, where new_vars are the new values of the loop variables after one iteration so must match vars in terms of types and shapes. </summary> </member> <member name="F:LotusIR.NodeProto.InputFieldNumber"> <summary>Field number for the "input" field.</summary> </member> <member name="P:LotusIR.NodeProto.Input"> <summary> The named inputs of the node. </summary> </member> <member name="F:LotusIR.NodeProto.OutputFieldNumber"> <summary>Field number for the "output" field.</summary> </member> <member name="P:LotusIR.NodeProto.Output"> <summary> The named outputs of the node. </summary> </member> <member name="F:LotusIR.NodeProto.NameFieldNumber"> <summary>Field number for the "name" field.</summary> </member> <member name="P:LotusIR.NodeProto.Name"> <summary> The name of this node. This field is and used to uniquely identify nodes in the graph. </summary> </member> <member name="F:LotusIR.NodeProto.OpTypeFieldNumber"> <summary>Field number for the "op_type" field.</summary> </member> <member name="P:LotusIR.NodeProto.OpType"> <summary> The name of the operator/function called by the node. This field MUST be present for this version of the IR. </summary> </member> <member name="F:LotusIR.NodeProto.AttributeFieldNumber"> <summary>Field number for the "attribute" field.</summary> </member> <member name="P:LotusIR.NodeProto.Attribute"> <summary> Additional named attributes. </summary> </member> <member name="F:LotusIR.NodeProto.DocStringFieldNumber"> <summary>Field number for the "doc_string" field.</summary> </member> <member name="P:LotusIR.NodeProto.DocString"> <summary> An human-readable documentation for this node in the graph. This text MAY contain Markdown markup that conforms to http://commonmark.org/. </summary> </member> <member name="F:LotusIR.NodeProto.InputArgCountFieldNumber"> <summary>Field number for the "input_arg_count" field.</summary> </member> <member name="P:LotusIR.NodeProto.InputArgCount"> <summary> The number of inputs for each argument of the operator/function. A formal parameter of the op may take a variable number of inputs that is only known when this node is constructed. BUG:13806939,dbox: I'm assuming that this field is like input_arg_info in that a zero element/missing array implies that one needs to crawl the graph to figure out the input counts, yes? Confirm and I'll make clear. Otherwise, we need to require it to be present and accurate. </summary> </member> <member name="F:LotusIR.NodeProto.ControlInputFieldNumber"> <summary>Field number for the "control_input" field.</summary> </member> <member name="P:LotusIR.NodeProto.ControlInput"> <summary> Specify a list of named nodes that must be executed before this node. Framework may use this to give users the ability to impose additional execution orders for the operations. </summary> </member> <member name="T:LotusIR.ModelProto"> <summary> ModelProto is a top-level file/container format for bundling a ML model. The semantics of the model are described by the GraphProto that represents a parameterized computation graph against a set of named operators that are defined independently from the graph. </summary> </member> <member name="F:LotusIR.ModelProto.IrVersionFieldNumber"> <summary>Field number for the "ir_version" field.</summary> </member> <member name="P:LotusIR.ModelProto.IrVersion"> <summary> The version of the IR this model targets. See Version enum above. This field MUST be present. </summary> </member> <member name="F:LotusIR.ModelProto.ProducerNameFieldNumber"> <summary>Field number for the "producer_name" field.</summary> </member> <member name="P:LotusIR.ModelProto.ProducerName"> <summary> The name of the framework or tool used to generate this model. This field SHOULD be present to indicate which implementation/tool/framework emitted the model. </summary> </member> <member name="F:LotusIR.ModelProto.ProducerVersionFieldNumber"> <summary>Field number for the "producer_version" field.</summary> </member> <member name="P:LotusIR.ModelProto.ProducerVersion"> <summary> The version of the framework or tool used to generate this model. This field SHOULD be present to indicate which implementation/tool/framework emitted the model. </summary> </member> <member name="F:LotusIR.ModelProto.DomainFieldNumber"> <summary>Field number for the "domain" field.</summary> </member> <member name="P:LotusIR.ModelProto.Domain"> <summary> Domain name of the model. We use reverse domain names as name space indicators. For example: `com.facebook.fair` or `com.microsoft.cognitiveservices` Together with `model_version` and GraphProto.name, this forms the unique identity of the graph. </summary> </member> <member name="F:LotusIR.ModelProto.ModelVersionFieldNumber"> <summary>Field number for the "model_version" field.</summary> </member> <member name="P:LotusIR.ModelProto.ModelVersion"> <summary> The version of the graph encoded. See Version enum below. </summary> </member> <member name="F:LotusIR.ModelProto.DocStringFieldNumber"> <summary>Field number for the "doc_string" field.</summary> </member> <member name="P:LotusIR.ModelProto.DocString"> <summary> A human-readable documentation for this model. Markdown is allowed. </summary> </member> <member name="F:LotusIR.ModelProto.GraphFieldNumber"> <summary>Field number for the "graph" field.</summary> </member> <member name="P:LotusIR.ModelProto.Graph"> <summary> The parameterized graph that is evaluated to execute the model. </summary> </member> <member name="F:LotusIR.ModelProto.ModelAuthorFieldNumber"> <summary>Field number for the "model_author" field.</summary> </member> <member name="P:LotusIR.ModelProto.ModelAuthor"> <summary> The name of the author who created the graph. </summary> </member> <member name="F:LotusIR.ModelProto.ModelLicenseFieldNumber"> <summary>Field number for the "model_license" field.</summary> </member> <member name="P:LotusIR.ModelProto.ModelLicense"> <summary> licensing information concerning use or origination of the graph. This text MAY contain Markdown markup that conforms to http://commonmark.org/. </summary> </member> <member name="T:LotusIR.GraphProto"> <summary> GraphProto defines a parameterized series of nodes to form a directed acyclic graph. This is the equivalent of the "network" and "graph" in many deep learning frameworks. All the input/output tensors are explicitly named so a framework can run any subgraph of the graph by feeding and fetching the named tensors. </summary> </member> <member name="F:LotusIR.GraphProto.NodeFieldNumber"> <summary>Field number for the "node" field.</summary> </member> <member name="P:LotusIR.GraphProto.Node"> <summary> The nodes in the graph. </summary> </member> <member name="F:LotusIR.GraphProto.NameFieldNumber"> <summary>Field number for the "name" field.</summary> </member> <member name="P:LotusIR.GraphProto.Name"> <summary> The name of the graph. </summary> </member> <member name="F:LotusIR.GraphProto.InitializerFieldNumber"> <summary>Field number for the "initializer" field.</summary> </member> <member name="P:LotusIR.GraphProto.Initializer"> <summary> A list of named tensor values (constants), used to specify default values for some of the inputs of the graph. Each TensorProto entry must have a distinct name (within the list) that also appears in the input list. In an evaluation, the default value specified here is used if and only if user specifies no value for the corresponding input parameter. May be used to pass serialized parameters for networks. </summary> </member> <member name="F:LotusIR.GraphProto.DocStringFieldNumber"> <summary>Field number for the "doc_string" field.</summary> </member> <member name="P:LotusIR.GraphProto.DocString"> <summary> A human-readable documentation for this graph. Markdown is allowed. </summary> </member> <member name="F:LotusIR.GraphProto.InputFieldNumber"> <summary>Field number for the "input" field.</summary> </member> <member name="P:LotusIR.GraphProto.Input"> <summary> The inputs and outputs of the graph. </summary> </member> <member name="F:LotusIR.GraphProto.OutputFieldNumber"> <summary>Field number for the "output" field.</summary> </member> <member name="F:LotusIR.GraphProto.ValueInfoFieldNumber"> <summary>Field number for the "value_info" field.</summary> </member> <member name="P:LotusIR.GraphProto.ValueInfo"> <summary> Information for the values in the graph. The ValueInfoProto.name's must be distinct. It is for a value to appear in value_info list. </summary> </member> <member name="F:LotusIR.GraphProto.FunctionFieldNumber"> <summary>Field number for the "function" field.</summary> </member> <member name="P:LotusIR.GraphProto.Function"> <summary> The function definitions of the graph. They can only only be used (i.e., called) in this graph. Each FunctionDefProto in function MUST have a unique name. </summary> </member> <member name="F:LotusIR.GraphProto.OperatorFieldNumber"> <summary>Field number for the "operator" field.</summary> </member> <member name="P:LotusIR.GraphProto.Operator"> <summary> The externally defined operators declared by this graph. </summary> </member> <member name="F:LotusIR.GraphProto.ImportedLibrariesFieldNumber"> <summary>Field number for the "imported_libraries" field.</summary> </member> <member name="P:LotusIR.GraphProto.ImportedLibraries"> <summary> Imported libraries are referenced as a collection of strings in the form of absolute URIs or relative paths. Where such relative paths are rooted is defined by tools and runtime implementations. </summary> </member> <member name="T:LotusIR.TensorProto"> <summary> A message defined to store a tensor in its serialized format. </summary> </member> <member name="F:LotusIR.TensorProto.DimsFieldNumber"> <summary>Field number for the "dims" field.</summary> </member> <member name="P:LotusIR.TensorProto.Dims"> <summary> The shape of the tensor. </summary> </member> <member name="F:LotusIR.TensorProto.DataTypeFieldNumber"> <summary>Field number for the "data_type" field.</summary> </member> <member name="P:LotusIR.TensorProto.DataType"> <summary> The data type of the tensor. </summary> </member> <member name="F:LotusIR.TensorProto.SegmentFieldNumber"> <summary>Field number for the "segment" field.</summary> </member> <member name="F:LotusIR.TensorProto.FloatDataFieldNumber"> <summary>Field number for the "float_data" field.</summary> </member> <member name="P:LotusIR.TensorProto.FloatData"> <summary> For float and complex64 values Complex64 tensors are encoded as a single array of floats, with the real components appearing in odd numbered positions, and the corresponding imaginary component apparing in the subsequent even numbered position. (e.g., [1.0 + 2.0i, 3.0 + 4.0i] is encoded as [1.0, 2.0 ,3.0 ,4.0] When this field is present, the data_type field MUST be FLOAT or COMPLEX64. </summary> </member> <member name="F:LotusIR.TensorProto.Int32DataFieldNumber"> <summary>Field number for the "int32_data" field.</summary> </member> <member name="P:LotusIR.TensorProto.Int32Data"> <summary> For int32, uint8, int8, uint16, int16, bool, and float16 values float16 values must be bit-wise converted to an uint16_t prior to writing to the buffer. When this field is present, the data_type field MUST be INT32, INT16, INT8, UINT16, INT8, BOOL, or FLOAT32 </summary> </member> <member name="F:LotusIR.TensorProto.StringDataFieldNumber"> <summary>Field number for the "string_data" field.</summary> </member> <member name="P:LotusIR.TensorProto.StringData"> <summary> For strings. Each element of string_data is a UTF-8 encoded Unicode string. No trailing null, no leading BOM. The protobuf "string" scalar type is not used to match ML community conventions. When this field is present, the data_type field MUST be STRING </summary> </member> <member name="F:LotusIR.TensorProto.Int64DataFieldNumber"> <summary>Field number for the "int64_data" field.</summary> </member> <member name="P:LotusIR.TensorProto.Int64Data"> <summary> For int64. When this field is present, the data_type field MUST be INT64 </summary> </member> <member name="F:LotusIR.TensorProto.NameFieldNumber"> <summary>Field number for the "name" field.</summary> </member> <member name="P:LotusIR.TensorProto.Name"> <summary> ly, a name for the tensor. </summary> </member> <member name="F:LotusIR.TensorProto.RawDataFieldNumber"> <summary>Field number for the "raw_data" field.</summary> </member> <member name="P:LotusIR.TensorProto.RawData"> <summary> Serializations can either use one of the fields above, or use this raw bytes field. The only exception is the string case, where one is to store the content in the repeated bytes string_data field. When this raw_data field is used to store tensor value, elements MUST be stored in as fixed-width, little-endian order. Floating-point data types MUST be stored in IEEE 754 format. Complex64 elements must be written as two consecutive FLOAT values, real component first. Complex128 elements must be written as two consecutive DOUBLE values, real component first. Boolean type MUST be written one byte per tensor element (00000001 for true, 00000000 for false). Note: the advantage of specific field rather than the raw_data field is that in some cases (e.g. int data), protobuf does a better packing via variable length storage, and may lead to smaller binary footprint. When this field is present, the data_type field MUST NOT be STRING or UNDEFINED </summary> </member> <member name="F:LotusIR.TensorProto.DoubleDataFieldNumber"> <summary>Field number for the "double_data" field.</summary> </member> <member name="P:LotusIR.TensorProto.DoubleData"> <summary> For double Complex64 tensors are encoded as a single array of doubles, with the real components appearing in odd numbered positions, and the corresponding imaginary component apparing in the subsequent even numbered position. (e.g., [1.0 + 2.0i, 3.0 + 4.0i] is encoded as [1.0, 2.0 ,3.0 ,4.0] When this field is present, the data_type field MUST be DOUBLE or COMPLEX128 </summary> </member> <member name="F:LotusIR.TensorProto.Uint64DataFieldNumber"> <summary>Field number for the "uint64_data" field.</summary> </member> <member name="P:LotusIR.TensorProto.Uint64Data"> <summary> For uint64 and uint32 values When this field is present, the data_type field MUST be UINT32 or UINT64 </summary> </member> <member name="T:LotusIR.TensorProto.Types"> <summary>Container for nested types declared in the TensorProto message type.</summary> </member> <member name="F:LotusIR.TensorProto.Types.DataType.Float"> <summary> Basic types. </summary> </member> <member name="F:LotusIR.TensorProto.Types.DataType.Uint8"> <summary> uint8_t </summary> </member> <member name="F:LotusIR.TensorProto.Types.DataType.Int8"> <summary> int8_t </summary> </member> <member name="F:LotusIR.TensorProto.Types.DataType.Uint16"> <summary> uint16_t </summary> </member> <member name="F:LotusIR.TensorProto.Types.DataType.Int16"> <summary> int16_t </summary> </member> <member name="F:LotusIR.TensorProto.Types.DataType.Int32"> <summary> int32_t </summary> </member> <member name="F:LotusIR.TensorProto.Types.DataType.Int64"> <summary> int64_t </summary> </member> <member name="F:LotusIR.TensorProto.Types.DataType.String"> <summary> string </summary> </member> <member name="F:LotusIR.TensorProto.Types.DataType.Bool"> <summary> bool </summary> </member> <member name="F:LotusIR.TensorProto.Types.DataType.Float16"> <summary> Advanced types </summary> </member> <member name="F:LotusIR.TensorProto.Types.DataType.Complex64"> <summary> complex with float32 real and imaginary components </summary> </member> <member name="F:LotusIR.TensorProto.Types.DataType.Complex128"> <summary> complex with float64 real and imaginary components </summary> </member> <member name="T:LotusIR.TensorProto.Types.Segment"> <summary> For very large tensors, we may want to store them in chunks, in which case the following fields will specify the segment that is stored in the current TensorProto. </summary> </member> <member name="F:LotusIR.TensorProto.Types.Segment.BeginFieldNumber"> <summary>Field number for the "begin" field.</summary> </member> <member name="F:LotusIR.TensorProto.Types.Segment.EndFieldNumber"> <summary>Field number for the "end" field.</summary> </member> <member name="T:LotusIR.SparseTensorProto"> <summary> A sparse tensor must be stored as three dense tensors: 1. dims: The shape of the original dense tensor. 2. indices: A 2-D tensor specifying the indices of the nonzero elements. 3. values: A 1-D tensor containing the values of the nonzero elements. </summary> </member> <member name="F:LotusIR.SparseTensorProto.DimsFieldNumber"> <summary>Field number for the "dims" field.</summary> </member> <member name="P:LotusIR.SparseTensorProto.Dims"> <summary> The dimensions in the tensor. </summary> </member> <member name="F:LotusIR.SparseTensorProto.IndicesFieldNumber"> <summary>Field number for the "indices" field.</summary> </member> <member name="P:LotusIR.SparseTensorProto.Indices"> <summary> This field MUST be present this version of the IR. </summary> </member> <member name="F:LotusIR.SparseTensorProto.ValuesFieldNumber"> <summary>Field number for the "values" field.</summary> </member> <member name="P:LotusIR.SparseTensorProto.Values"> <summary> This field MUST be present this version of the IR. </summary> </member> <member name="T:LotusIR.TypeProto"> <summary> Define the types. </summary> </member> <member name="F:LotusIR.TypeProto.TensorTypeFieldNumber"> <summary>Field number for the "tensor_type" field.</summary> </member> <member name="P:LotusIR.TypeProto.TensorType"> <summary> The type of a tensor. </summary> </member> <member name="F:LotusIR.TypeProto.SparseTensorTypeFieldNumber"> <summary>Field number for the "sparse_tensor_type" field.</summary> </member> <member name="P:LotusIR.TypeProto.SparseTensorType"> <summary> The type of a sparse tensor. </summary> </member> <member name="F:LotusIR.TypeProto.HandleTypeFieldNumber"> <summary>Field number for the "handle_type" field.</summary> </member> <member name="P:LotusIR.TypeProto.HandleType"> <summary> The type of an opaque handle. A handle is used to represent a reference to a resource managed by the framework runtime. </summary> </member> <member name="F:LotusIR.TypeProto.TupleTypeFieldNumber"> <summary>Field number for the "tuple_type" field.</summary> </member> <member name="P:LotusIR.TypeProto.TupleType"> <summary> The type of a tuple. </summary> </member> <member name="F:LotusIR.TypeProto.SeqTypeFieldNumber"> <summary>Field number for the "seq_type" field.</summary> </member> <member name="P:LotusIR.TypeProto.SeqType"> <summary> The type of a sequence. </summary> </member> <member name="F:LotusIR.TypeProto.MapTypeFieldNumber"> <summary>Field number for the "map_type" field.</summary> </member> <member name="P:LotusIR.TypeProto.MapType"> <summary> The type of a map. </summary> </member> <member name="T:LotusIR.TypeProto.ValueOneofCase"> <summary>Enum of possible cases for the "value" oneof.</summary> </member> <member name="T:LotusIR.TypeProto.Types"> <summary>Container for nested types declared in the TypeProto message type.</summary> </member> <member name="T:LotusIR.TypeProto.Types.TensorShapeProto"> <summary> Defines a tensor shape. A dimension can be either an integer value or a symbolic variable. A symbolic variable represents an unknown dimension. </summary> </member> <member name="F:LotusIR.TypeProto.Types.TensorShapeProto.DimFieldNumber"> <summary>Field number for the "dim" field.</summary> </member> <member name="T:LotusIR.TypeProto.Types.TensorShapeProto.Types"> <summary>Container for nested types declared in the TensorShapeProto message type.</summary> </member> <member name="F:LotusIR.TypeProto.Types.TensorShapeProto.Types.Dimension.DimValueFieldNumber"> <summary>Field number for the "dim_value" field.</summary> </member> <member name="F:LotusIR.TypeProto.Types.TensorShapeProto.Types.Dimension.DimParamFieldNumber"> <summary>Field number for the "dim_param" field.</summary> </member> <member name="P:LotusIR.TypeProto.Types.TensorShapeProto.Types.Dimension.DimParam"> <summary> namespace Shape </summary> </member> <member name="T:LotusIR.TypeProto.Types.TensorShapeProto.Types.Dimension.ValueOneofCase"> <summary>Enum of possible cases for the "value" oneof.</summary> </member> <member name="F:LotusIR.TypeProto.Types.TensorTypeProto.ElemTypeFieldNumber"> <summary>Field number for the "elem_type" field.</summary> </member> <member name="P:LotusIR.TypeProto.Types.TensorTypeProto.ElemType"> <summary> This field MUST NOT have the value of UNDEFINED This field MUST be present for this version of the IR. </summary> </member> <member name="F:LotusIR.TypeProto.Types.TensorTypeProto.ShapeFieldNumber"> <summary>Field number for the "shape" field.</summary> </member> <member name="F:LotusIR.TypeProto.Types.SparseTensorTypeProto.ElemTypeFieldNumber"> <summary>Field number for the "elem_type" field.</summary> </member> <member name="P:LotusIR.TypeProto.Types.SparseTensorTypeProto.ElemType"> <summary> This field MUST NOT have the value of UNDEFINED This field MUST be present for this version of the IR. </summary> </member> <member name="F:LotusIR.TypeProto.Types.SparseTensorTypeProto.ShapeFieldNumber"> <summary>Field number for the "shape" field.</summary> </member> <member name="F:LotusIR.TypeProto.Types.TupleTypeProto.ElemTypeFieldNumber"> <summary>Field number for the "elem_type" field.</summary> </member> <member name="F:LotusIR.TypeProto.Types.SeqTypeProto.ElemTypeFieldNumber"> <summary>Field number for the "elem_type" field.</summary> </member> <member name="P:LotusIR.TypeProto.Types.SeqTypeProto.ElemType"> <summary> This field MUST be present for this version of the IR. </summary> </member> <member name="F:LotusIR.TypeProto.Types.MapTypeProto.KeyTypeFieldNumber"> <summary>Field number for the "key_type" field.</summary> </member> <member name="P:LotusIR.TypeProto.Types.MapTypeProto.KeyType"> <summary> This field MUST be present for this version of the IR. This field MUST refer to an integral type ([U]INT{8|16|32|64}) or STRING </summary> </member> <member name="F:LotusIR.TypeProto.Types.MapTypeProto.ValueTypeFieldNumber"> <summary>Field number for the "value_type" field.</summary> </member> <member name="P:LotusIR.TypeProto.Types.MapTypeProto.ValueType"> <summary> This field MUST be present for this version of the IR. This field MUST NOT refer to UNDEFINED </summary> </member> <member name="F:LotusIR.ValueProto.DenseTensorFieldNumber"> <summary>Field number for the "dense_tensor" field.</summary> </member> <member name="P:LotusIR.ValueProto.DenseTensor"> <summary> A dense tensor. </summary> </member> <member name="F:LotusIR.ValueProto.SparseTensorFieldNumber"> <summary>Field number for the "sparse_tensor" field.</summary> </member> <member name="P:LotusIR.ValueProto.SparseTensor"> <summary> A sparse tensor. </summary> </member> <member name="F:LotusIR.ValueProto.HandleFieldNumber"> <summary>Field number for the "handle" field.</summary> </member> <member name="P:LotusIR.ValueProto.Handle"> <summary> A handle. </summary> </member> <member name="F:LotusIR.ValueProto.TupleFieldNumber"> <summary>Field number for the "tuple" field.</summary> </member> <member name="P:LotusIR.ValueProto.Tuple"> <summary> A tuple. </summary> </member> <member name="F:LotusIR.ValueProto.SeqFieldNumber"> <summary>Field number for the "seq" field.</summary> </member> <member name="P:LotusIR.ValueProto.Seq"> <summary> A sequence. </summary> </member> <member name="F:LotusIR.ValueProto.MapFieldNumber"> <summary>Field number for the "map" field.</summary> </member> <member name="P:LotusIR.ValueProto.Map"> <summary> A map. </summary> </member> <member name="T:LotusIR.ValueProto.ValueOneofCase"> <summary>Enum of possible cases for the "value" oneof.</summary> </member> <member name="T:LotusIR.ValueProto.Types"> <summary>Container for nested types declared in the ValueProto message type.</summary> </member> <member name="T:LotusIR.ValueProto.Types.HandleProto"> <summary> Defines a handle in its serialized format. </summary> </member> <member name="F:LotusIR.ValueProto.Types.HandleProto.UidFieldNumber"> <summary>Field number for the "uid" field.</summary> </member> <member name="P:LotusIR.ValueProto.Types.HandleProto.Uid"> <summary> This field MUST be present this version of the IR. </summary> </member> <member name="T:LotusIR.ValueProto.Types.TupleProto"> <summary> Defines a tuple in its serialized format. </summary> </member> <member name="F:LotusIR.ValueProto.Types.TupleProto.ElemsFieldNumber"> <summary>Field number for the "elems" field.</summary> </member> <member name="T:LotusIR.ValueProto.Types.SequenceProto"> <summary> Defines a sequence in its serialized format. </summary> </member> <member name="F:LotusIR.ValueProto.Types.SequenceProto.ElemsFieldNumber"> <summary>Field number for the "elems" field.</summary> </member> <member name="T:LotusIR.ValueProto.Types.MapProto"> <summary> Defines a map in its serialized format. Maps are serialized as two single-dimensional tensors for storage efficiency. The dimensions of each tensor MUST be identical and the key at position N corresponds to the value at position N. Keys SHOULD be unique. When a given key appears multiple times, the value that corresponds last occurance of the key is the value. This is consistent with protobuf3 encoding rules for map. </summary> </member> <member name="F:LotusIR.ValueProto.Types.MapProto.KeysFieldNumber"> <summary>Field number for the "keys" field.</summary> </member> <member name="P:LotusIR.ValueProto.Types.MapProto.Keys"> <summary> This field MUST be present for this version of the IR. The data type of the tensor MUST refer to an integral type ([U]INT{8|16|32|64}) or STRING </summary> </member> <member name="F:LotusIR.ValueProto.Types.MapProto.ValuesFieldNumber"> <summary>Field number for the "values" field.</summary> </member> <member name="P:LotusIR.ValueProto.Types.MapProto.Values"> <summary> This field MUST be present for this version of the IR. </summary> </member> <member name="F:LotusIR.ParameterDeclProto.NameFieldNumber"> <summary>Field number for the "name" field.</summary> </member> <member name="F:LotusIR.ParameterDeclProto.TypeFieldNumber"> <summary>Field number for the "type" field.</summary> </member> <member name="F:LotusIR.ParameterDeclProto.DocStringFieldNumber"> <summary>Field number for the "doc_string" field.</summary> </member> <member name="P:LotusIR.ParameterDeclProto.DocString"> <summary> An human-readable documentation for this parameter. </summary> </member> <member name="T:LotusIR.FunctionDefProto"> <summary> Defines a function. </summary> </member> <member name="F:LotusIR.FunctionDefProto.NameFieldNumber"> <summary>Field number for the "name" field.</summary> </member> <member name="P:LotusIR.FunctionDefProto.Name"> <summary> The name of the function. This field MUST be present for this version of the IR. </summary> </member> <member name="F:LotusIR.FunctionDefProto.InputParamsFieldNumber"> <summary>Field number for the "input_params" field.</summary> </member> <member name="P:LotusIR.FunctionDefProto.InputParams"> <summary> The input parameters of the function. </summary> </member> <member name="F:LotusIR.FunctionDefProto.OutputParamsFieldNumber"> <summary>Field number for the "output_params" field.</summary> </member> <member name="P:LotusIR.FunctionDefProto.OutputParams"> <summary> The output parameters of the function. </summary> </member> <member name="F:LotusIR.FunctionDefProto.NodeFieldNumber"> <summary>Field number for the "node" field.</summary> </member> <member name="P:LotusIR.FunctionDefProto.Node"> <summary> The body of the function. </summary> </member> <member name="F:LotusIR.FunctionDefProto.AttrFieldNumber"> <summary>Field number for the "attr" field.</summary> </member> <member name="P:LotusIR.FunctionDefProto.Attr"> <summary> The named attributes of the function. </summary> </member> <member name="F:LotusIR.FunctionDefProto.DocStringFieldNumber"> <summary>Field number for the "doc_string" field.</summary> </member> <member name="P:LotusIR.FunctionDefProto.DocString"> <summary> An human-readable documentation for this node in the graph. This text MAY contain Markdown markup that conforms to http://commonmark.org/. </summary> </member> <member name="F:LotusIR.SignatureDeclProto.InputParamsFieldNumber"> <summary>Field number for the "input_params" field.</summary> </member> <member name="P:LotusIR.SignatureDeclProto.InputParams"> <summary> The formal input parameters to the operation or function </summary> </member> <member name="F:LotusIR.SignatureDeclProto.OutputParamsFieldNumber"> <summary>Field number for the "output_params" field.</summary> </member> <member name="P:LotusIR.SignatureDeclProto.OutputParams"> <summary> The formal output parameters to the operation or function </summary> </member> <member name="F:LotusIR.SignatureDeclProto.InputAttributesFieldNumber"> <summary>Field number for the "input_attributes" field.</summary> </member> <member name="P:LotusIR.SignatureDeclProto.InputAttributes"> <summary> The declaration of expected attributes to the operation or function </summary> </member> <member name="F:LotusIR.SignatureDeclProto.DocStringFieldNumber"> <summary>Field number for the "doc_string" field.</summary> </member> <member name="P:LotusIR.SignatureDeclProto.DocString"> <summary> An human-readable documentation for this signature. </summary> </member> <member name="F:LotusIR.OperatorDeclProto.NameFieldNumber"> <summary>Field number for the "name" field.</summary> </member> <member name="P:LotusIR.OperatorDeclProto.Name"> <summary> This field MUST be present for this version of the IR. </summary> </member> <member name="F:LotusIR.OperatorDeclProto.SignatureFieldNumber"> <summary>Field number for the "signature" field.</summary> </member> <member name="P:LotusIR.OperatorDeclProto.Signature"> <summary> This field MUST contain at least one SignatureDeclProto. This field MAY contain multiple SignatureDeclProtos, one per type signature supported by this operator. </summary> </member> <member name="F:LotusIR.OperatorDeclProto.DocStringFieldNumber"> <summary>Field number for the "doc_string" field.</summary> </member> <member name="P:LotusIR.OperatorDeclProto.DocString"> <summary> An human-readable documentation for this operator. </summary> </member> <member name="T:LotusIR.LibraryProto"> <summary> A library is a top-level format that contains the declaration of operators and the definition of functions. </summary> </member> <member name="F:LotusIR.LibraryProto.IrVersionFieldNumber"> <summary>Field number for the "ir_version" field.</summary> </member> <member name="P:LotusIR.LibraryProto.IrVersion"> <summary> The version of the IR this graph targets. See Version enum below. This field MUST be present this version of the IR. </summary> </member> <member name="F:LotusIR.LibraryProto.ProducerVersionFieldNumber"> <summary>Field number for the "producer_version" field.</summary> </member> <member name="P:LotusIR.LibraryProto.ProducerVersion"> <summary> The version of the framework runtime that generates this graph. This producer_version has the same format as ir_version. </summary> </member> <member name="F:LotusIR.LibraryProto.ProducerTagFieldNumber"> <summary>Field number for the "producer_tag" field.</summary> </member> <member name="P:LotusIR.LibraryProto.ProducerTag"> <summary> The name of the framework used to generate this graph in the form "framework_name[-tag]". Tag is and provides additional information such as `alpha` or `beta` or `rc3`. </summary> </member> <member name="F:LotusIR.LibraryProto.ModelVersionFieldNumber"> <summary>Field number for the "model_version" field.</summary> </member> <member name="P:LotusIR.LibraryProto.ModelVersion"> <summary> An version identifier used to track evolution of this library. This model_version has the same format as ir_version. </summary> </member> <member name="F:LotusIR.LibraryProto.ModelAuthorFieldNumber"> <summary>Field number for the "model_author" field.</summary> </member> <member name="P:LotusIR.LibraryProto.ModelAuthor"> <summary> The name of the author who created the library. </summary> </member> <member name="F:LotusIR.LibraryProto.ModelLicenseFieldNumber"> <summary>Field number for the "model_license" field.</summary> </member> <member name="P:LotusIR.LibraryProto.ModelLicense"> <summary> licensing information concerning use or origination of the library. </summary> </member> <member name="F:LotusIR.LibraryProto.NameFieldNumber"> <summary>Field number for the "name" field.</summary> </member> <member name="P:LotusIR.LibraryProto.Name"> <summary> The name of the library. </summary> </member> <member name="F:LotusIR.LibraryProto.DomainFieldNumber"> <summary>Field number for the "domain" field.</summary> </member> <member name="P:LotusIR.LibraryProto.Domain"> <summary> Domain of the graph. We use reverse domain names as name space indicators. For example: `com.facebook.fair` or `com.microsoft.cognitiveservices` Together with `name` and `model_version`, this forms the unique identity of the library. </summary> </member> <member name="F:LotusIR.LibraryProto.DocStringFieldNumber"> <summary>Field number for the "doc_string" field.</summary> </member> <member name="P:LotusIR.LibraryProto.DocString"> <summary> An human-readable documentation for this graph. </summary> </member> <member name="F:LotusIR.LibraryProto.OperatorFieldNumber"> <summary>Field number for the "operator" field.</summary> </member> <member name="P:LotusIR.LibraryProto.Operator"> <summary> The operators declared by this library. </summary> </member> <member name="F:LotusIR.LibraryProto.FunctionFieldNumber"> <summary>Field number for the "function" field.</summary> </member> <member name="P:LotusIR.LibraryProto.Function"> <summary> The function definitions of the library. </summary> </member> <member name="F:LotusIR.LibraryProto.ImportedLibrariesFieldNumber"> <summary>Field number for the "imported_libraries" field.</summary> </member> <member name="P:LotusIR.LibraryProto.ImportedLibraries"> <summary> Imported libraries are referenced as a collection of strings in the form of absolute URIs or relative paths. Where such relative paths are rooted is defined by tools and runtime implementations. </summary> </member> <member name="T:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.OnnxMlReflection"> <summary>Holder for reflection information generated from onnx-ml.proto3</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.OnnxMlReflection.Descriptor"> <summary>File descriptor for onnx-ml.proto3</summary> </member> <member name="T:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.Version"> <summary> To be compatible with both proto2 and proto3, we will use a version number that is not defined by the default value but an explicit enum number. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.Version.StartVersion"> <summary> proto3 requires the first enum value to be zero. We add this just to appease the compiler. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.Version.IrVersion20171010"> <summary> The version field is always serialized and we will use it to store the version that the graph is generated from. This helps us set up version control. We should use version as xx(major) - xx(minor) - xxxx(bugfix) and we are starting with 0x00000001 (0.0.1), which was the version we published on Oct 10, 2017. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.Version.IrVersion20171030"> <summary> IR_VERSION 0.0.2 published on Oct 30, 2017 - Added type discriminator to AttributeProto to support proto3 users </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.Version.IrVersion"> <summary> IR VERSION 0.0.3 published on Nov 3, 2017 - For operator versioning: - Added new message OperatorSetIdProto - Added opset_import in ModelProto - For vendor extensions, added domain in NodeProto </summary> </member> <member name="T:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto"> <summary> A named attribute containing either singular float, integer, string and tensor values, or repeated float, integer, string and tensor values. An AttributeProto MUST contain the name field, and *only one* of the following content fields, effectively enforcing a C/C++ union equivalent. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.NameFieldNumber"> <summary>Field number for the "name" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.Name"> <summary> The name field MUST be present for this version of the IR. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.DocStringFieldNumber"> <summary>Field number for the "doc_string" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.DocString"> <summary> A human-readable documentation for this attribute. Markdown is allowed. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.TypeFieldNumber"> <summary>Field number for the "type" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.Type"> <summary> The type field MUST be present for this version of the IR. For 0.0.1 versions of the IR, this field was not defined, and implementations needed to use has_field hueristics to determine which value field was in use. For IR_VERSION 0.0.2 or later, this field MUST be set and match the f|i|s|t|... field in use. This change was made to accomodate proto3 implementations. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.FFieldNumber"> <summary>Field number for the "f" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.F"> <summary> Exactly ONE of the following fields must be present for this version of the IR </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.IFieldNumber"> <summary>Field number for the "i" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.I"> <summary> int </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.SFieldNumber"> <summary>Field number for the "s" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.S"> <summary> UTF-8 string </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.TFieldNumber"> <summary>Field number for the "t" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.T"> <summary> tensor value </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.GFieldNumber"> <summary>Field number for the "g" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.G"> <summary> graph </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.FloatsFieldNumber"> <summary>Field number for the "floats" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.Floats"> <summary> list of floats </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.IntsFieldNumber"> <summary>Field number for the "ints" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.Ints"> <summary> list of ints </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.StringsFieldNumber"> <summary>Field number for the "strings" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.Strings"> <summary> list of UTF-8 strings </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.TensorsFieldNumber"> <summary>Field number for the "tensors" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.Tensors"> <summary> list of tensors </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.GraphsFieldNumber"> <summary>Field number for the "graphs" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.Graphs"> <summary> list of graph </summary> </member> <member name="T:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.Types"> <summary>Container for nested types declared in the AttributeProto message type.</summary> </member> <member name="T:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.AttributeProto.Types.AttributeType"> <summary> Note: this enum is structurally identical to the OpSchema::AttrType enum defined in schema.h. If you rev one, you likely need to rev the other. </summary> </member> <member name="T:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.ValueInfoProto"> <summary> Defines information on value, including the name, the type, and the shape of the value. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.ValueInfoProto.NameFieldNumber"> <summary>Field number for the "name" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.ValueInfoProto.Name"> <summary> This field MUST be present in this version of the IR. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.ValueInfoProto.TypeFieldNumber"> <summary>Field number for the "type" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.ValueInfoProto.Type"> <summary> This field MUST be present in this version of the IR. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.ValueInfoProto.DocStringFieldNumber"> <summary>Field number for the "doc_string" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.ValueInfoProto.DocString"> <summary> A human-readable documentation for this value. Markdown is allowed. </summary> </member> <member name="T:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.NodeProto"> <summary> NodeProto stores a node that is similar to the notion of "layer" or "operator" in many deep learning frameworks. For example, it can be a node of type "Conv" that takes in an image, a filter tensor and a bias tensor, and produces the convolved output. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.NodeProto.InputFieldNumber"> <summary>Field number for the "input" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.NodeProto.Input"> <summary> namespace Value </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.NodeProto.OutputFieldNumber"> <summary>Field number for the "output" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.NodeProto.Output"> <summary> namespace Value </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.NodeProto.NameFieldNumber"> <summary>Field number for the "name" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.NodeProto.Name"> <summary> An optional identifier for this node in a graph. This field MAY be absent in ths version of the IR. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.NodeProto.OpTypeFieldNumber"> <summary>Field number for the "op_type" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.NodeProto.OpType"> <summary> The symbolic identifier of the Operator to execute. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.NodeProto.DomainFieldNumber"> <summary>Field number for the "domain" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.NodeProto.Domain"> <summary> The domain of the OperatorSet that specifies the operator named by op_type. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.NodeProto.AttributeFieldNumber"> <summary>Field number for the "attribute" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.NodeProto.Attribute"> <summary> Additional named attributes. NOTE: Simply using ValueProto.NameValuePairProto is the most general solution. I kept AttributeProto to minimize churn on CI results. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.NodeProto.DocStringFieldNumber"> <summary>Field number for the "doc_string" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.NodeProto.DocString"> <summary> A human-readable documentation for this node. Markdown is allowed. </summary> </member> <member name="T:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.ModelProto"> <summary> ModelProto is a top-level file/container format for bundling a ML model. The semantics of the model are described by the GraphProto that represents a parameterized computation graph against a set of named operators that are defined independently from the graph. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.ModelProto.IrVersionFieldNumber"> <summary>Field number for the "ir_version" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.ModelProto.IrVersion"> <summary> The version of the IR this model targets. See Version enum above. This field MUST be present. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.ModelProto.OpsetImportFieldNumber"> <summary>Field number for the "opset_import" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.ModelProto.OpsetImport"> <summary> The OperatorSets this model relies on. All ModelProtos MUST have at least one entry that specifies which version of the ONNX OperatorSet is being imported. All nodes in the ModelProto's graph will bind against the operator with the same-domain/same-op_type operator with the HIGHEST version in the referenced operator sets. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.ModelProto.ProducerNameFieldNumber"> <summary>Field number for the "producer_name" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.ModelProto.ProducerName"> <summary> The name of the framework or tool used to generate this model. This field SHOULD be present to indicate which implementation/tool/framework emitted the model. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.ModelProto.ProducerVersionFieldNumber"> <summary>Field number for the "producer_version" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.ModelProto.ProducerVersion"> <summary> The version of the framework or tool used to generate this model. This field SHOULD be present to indicate which implementation/tool/framework emitted the model. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.ModelProto.DomainFieldNumber"> <summary>Field number for the "domain" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.ModelProto.Domain"> <summary> Domain name of the model. We use reverse domain names as name space indicators. For example: `com.facebook.fair` or `com.microsoft.cognitiveservices` Together with `model_version` and GraphProto.name, this forms the unique identity of the graph. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.ModelProto.ModelVersionFieldNumber"> <summary>Field number for the "model_version" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.ModelProto.ModelVersion"> <summary> The version of the graph encoded. See Version enum below. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.ModelProto.DocStringFieldNumber"> <summary>Field number for the "doc_string" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.ModelProto.DocString"> <summary> A human-readable documentation for this model. Markdown is allowed. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.ModelProto.GraphFieldNumber"> <summary>Field number for the "graph" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.ModelProto.Graph"> <summary> The parameterized graph that is evaluated to execute the model. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.ModelProto.MetadataPropsFieldNumber"> <summary>Field number for the "metadata_props" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.ModelProto.MetadataProps"> <summary> Named metadata values; keys should be distinct. </summary> </member> <member name="T:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.StringStringEntryProto"> <summary> StringStringEntryProto follows the pattern for cross-proto-version maps. See https://developers.google.com/protocol-buffers/docs/proto3#maps </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.StringStringEntryProto.KeyFieldNumber"> <summary>Field number for the "key" field.</summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.StringStringEntryProto.ValueFieldNumber"> <summary>Field number for the "value" field.</summary> </member> <member name="T:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.GraphProto"> <summary> GraphProto defines a parameterized series of nodes to form a directed acyclic graph. This is the equivalent of the "network" and "graph" in many deep learning frameworks. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.GraphProto.NodeFieldNumber"> <summary>Field number for the "node" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.GraphProto.Node"> <summary> The nodes in the graph. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.GraphProto.NameFieldNumber"> <summary>Field number for the "name" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.GraphProto.Name"> <summary> The name of the graph. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.GraphProto.InitializerFieldNumber"> <summary>Field number for the "initializer" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.GraphProto.Initializer"> <summary> A list of named tensor values (constants), used to specify default values for some of the inputs of the graph. Each TensorProto entry must have a distinct name (within the list) that also appears in the input list. In an evaluation, the default value specified here is used if and only if user specifies no value for the corresponding input parameter. May be used to pass serialized parameters for networks. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.GraphProto.DocStringFieldNumber"> <summary>Field number for the "doc_string" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.GraphProto.DocString"> <summary> A human-readable documentation for this graph. Markdown is allowed. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.GraphProto.InputFieldNumber"> <summary>Field number for the "input" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.GraphProto.Input"> <summary> The inputs and outputs of the graph. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.GraphProto.OutputFieldNumber"> <summary>Field number for the "output" field.</summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.GraphProto.ValueInfoFieldNumber"> <summary>Field number for the "value_info" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.GraphProto.ValueInfo"> <summary> Information for the values in the graph. The ValueInfoProto.name's must be distinct. It is optional for a value to appear in value_info list. </summary> </member> <member name="T:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto"> <summary> A message defined to store a tensor in its serialized format. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.DimsFieldNumber"> <summary>Field number for the "dims" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.Dims"> <summary> The shape of the tensor. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.DataTypeFieldNumber"> <summary>Field number for the "data_type" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.DataType"> <summary> The data type of the tensor. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.SegmentFieldNumber"> <summary>Field number for the "segment" field.</summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.FloatDataFieldNumber"> <summary>Field number for the "float_data" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.FloatData"> <summary> For float and complex64 values Complex64 tensors are encoded as a single array of floats, with the real components appearing in odd numbered positions, and the corresponding imaginary component apparing in the subsequent even numbered position. (e.g., [1.0 + 2.0i, 3.0 + 4.0i] is encoded as [1.0, 2.0 ,3.0 ,4.0] When this field is present, the data_type field MUST be FLOAT or COMPLEX64. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.Int32DataFieldNumber"> <summary>Field number for the "int32_data" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.Int32Data"> <summary> For int32, uint8, int8, uint16, int16, bool, and float16 values float16 values must be bit-wise converted to an uint16_t prior to writing to the buffer. When this field is present, the data_type field MUST be INT32, INT16, INT8, UINT16, INT8, BOOL, or FLOAT32 </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.StringDataFieldNumber"> <summary>Field number for the "string_data" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.StringData"> <summary> For strings. Each element of string_data is a UTF-8 encoded Unicode string. No trailing null, no leading BOM. The protobuf "string" scalar type is not used to match ML community conventions. When this field is present, the data_type field MUST be STRING </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.Int64DataFieldNumber"> <summary>Field number for the "int64_data" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.Int64Data"> <summary> For int64. When this field is present, the data_type field MUST be INT64 </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.NameFieldNumber"> <summary>Field number for the "name" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.Name"> <summary> Optionally, a name for the tensor. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.DocStringFieldNumber"> <summary>Field number for the "doc_string" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.DocString"> <summary> A human-readable documentation for this tensor. Markdown is allowed. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.RawDataFieldNumber"> <summary>Field number for the "raw_data" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.RawData"> <summary> Serializations can either use one of the fields above, or use this raw bytes field. The only exception is the string case, where one is required to store the content in the repeated bytes string_data field. When this raw_data field is used to store tensor value, elements MUST be stored in as fixed-width, little-endian order. Floating-point data types MUST be stored in IEEE 754 format. Complex64 elements must be written as two consecutive FLOAT values, real component first. Complex128 elements must be written as two consecutive DOUBLE values, real component first. Boolean type MUST be written one byte per tensor element (00000001 for true, 00000000 for false). Note: the advantage of specific field rather than the raw_data field is that in some cases (e.g. int data), protobuf does a better packing via variable length storage, and may lead to smaller binary footprint. When this field is present, the data_type field MUST NOT be STRING or UNDEFINED </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.DoubleDataFieldNumber"> <summary>Field number for the "double_data" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.DoubleData"> <summary> For double Complex64 tensors are encoded as a single array of doubles, with the real components appearing in odd numbered positions, and the corresponding imaginary component apparing in the subsequent even numbered position. (e.g., [1.0 + 2.0i, 3.0 + 4.0i] is encoded as [1.0, 2.0 ,3.0 ,4.0] When this field is present, the data_type field MUST be DOUBLE or COMPLEX128 </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.Uint64DataFieldNumber"> <summary>Field number for the "uint64_data" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.Uint64Data"> <summary> For uint64 and uint32 values When this field is present, the data_type field MUST be UINT32 or UINT64 </summary> </member> <member name="T:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.Types"> <summary>Container for nested types declared in the TensorProto message type.</summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.Types.DataType.Float"> <summary> Basic types. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.Types.DataType.Uint8"> <summary> uint8_t </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.Types.DataType.Int8"> <summary> int8_t </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.Types.DataType.Uint16"> <summary> uint16_t </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.Types.DataType.Int16"> <summary> int16_t </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.Types.DataType.Int32"> <summary> int32_t </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.Types.DataType.Int64"> <summary> int64_t </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.Types.DataType.String"> <summary> string </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.Types.DataType.Bool"> <summary> bool </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.Types.DataType.Float16"> <summary> Advanced types </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.Types.DataType.Complex64"> <summary> complex with float32 real and imaginary components </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.Types.DataType.Complex128"> <summary> complex with float64 real and imaginary components </summary> </member> <member name="T:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.Types.Segment"> <summary> For very large tensors, we may want to store them in chunks, in which case the following fields will specify the segment that is stored in the current TensorProto. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.Types.Segment.BeginFieldNumber"> <summary>Field number for the "begin" field.</summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorProto.Types.Segment.EndFieldNumber"> <summary>Field number for the "end" field.</summary> </member> <member name="T:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorShapeProto"> <summary> Defines a tensor shape. A dimension can be either an integer value or a symbolic variable. A symbolic variable represents an unknown dimension. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorShapeProto.DimFieldNumber"> <summary>Field number for the "dim" field.</summary> </member> <member name="T:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorShapeProto.Types"> <summary>Container for nested types declared in the TensorShapeProto message type.</summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorShapeProto.Types.Dimension.DimValueFieldNumber"> <summary>Field number for the "dim_value" field.</summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorShapeProto.Types.Dimension.DimParamFieldNumber"> <summary>Field number for the "dim_param" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorShapeProto.Types.Dimension.DimParam"> <summary> namespace Shape </summary> </member> <member name="T:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TensorShapeProto.Types.Dimension.ValueOneofCase"> <summary>Enum of possible cases for the "value" oneof.</summary> </member> <member name="T:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TypeProto"> <summary> Define the types. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TypeProto.TensorTypeFieldNumber"> <summary>Field number for the "tensor_type" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TypeProto.TensorType"> <summary> The type of a tensor. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TypeProto.SequenceTypeFieldNumber"> <summary>Field number for the "sequence_type" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TypeProto.SequenceType"> <summary> The type of a sequence. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TypeProto.MapTypeFieldNumber"> <summary>Field number for the "map_type" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TypeProto.MapType"> <summary> The type of a map. </summary> </member> <member name="T:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TypeProto.ValueOneofCase"> <summary>Enum of possible cases for the "value" oneof.</summary> </member> <member name="T:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TypeProto.Types"> <summary>Container for nested types declared in the TypeProto message type.</summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TypeProto.Types.Tensor.ElemTypeFieldNumber"> <summary>Field number for the "elem_type" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TypeProto.Types.Tensor.ElemType"> <summary> This field MUST NOT have the value of UNDEFINED This field MUST be present for this version of the IR. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TypeProto.Types.Tensor.ShapeFieldNumber"> <summary>Field number for the "shape" field.</summary> </member> <member name="T:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TypeProto.Types.Sequence"> <summary> repeated T </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TypeProto.Types.Sequence.ElemTypeFieldNumber"> <summary>Field number for the "elem_type" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TypeProto.Types.Sequence.ElemType"> <summary> The type and optional shape of each element of the sequence. This field MUST be present for this version of the IR. </summary> </member> <member name="T:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TypeProto.Types.Map"> <summary> map<K,V> </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TypeProto.Types.Map.KeyTypeFieldNumber"> <summary>Field number for the "key_type" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TypeProto.Types.Map.KeyType"> <summary> This field MUST be present for this version of the IR. This field MUST refer to an integral type ([U]INT{8|16|32|64}) or STRING </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TypeProto.Types.Map.ValueTypeFieldNumber"> <summary>Field number for the "value_type" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.TypeProto.Types.Map.ValueType"> <summary> This field MUST be present for this version of the IR. </summary> </member> <member name="T:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.OperatorSetIdProto"> <summary> OperatorSets are uniquely identified by a (domain, opset_version) pair. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.OperatorSetIdProto.DomainFieldNumber"> <summary>Field number for the "domain" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.OperatorSetIdProto.Domain"> <summary> The domain of the operator set being identified. The empty string ("") or absence of this field implies the operator set that is defined as part of the ONNX specification. This field MUST be present in this version of the IR when referring to any other operator set. </summary> </member> <member name="F:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.OperatorSetIdProto.VersionFieldNumber"> <summary>Field number for the "version" field.</summary> </member> <member name="P:Microsoft.ML.Runtime.UniversalModelFormat.Onnx.OperatorSetIdProto.Version"> <summary> The version of the operator set being identified. This field MUST be present in this version of the IR. </summary> </member> </members> </doc> |